* [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 9:29 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 02/34] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery" Matthew Brost
` (35 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
From: Tomasz Lis <tomasz.lis@intel.com>
Protect access to GGTT config as this is non-static information.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 96 ++++++++++++++++++-----
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 3 +
drivers/gpu/drm/xe/xe_sriov_vf.c | 6 ++
3 files changed, 84 insertions(+), 21 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 0461d5513487..016c867e5e2b 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -440,18 +440,21 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
+ down_write(&config->lock);
+
err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
if (unlikely(err))
- return err;
+ goto out;
err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_SIZE_KEY, &size);
if (unlikely(err))
- return err;
+ goto out;
if (config->ggtt_size && config->ggtt_size != size) {
xe_gt_sriov_err(gt, "Unexpected GGTT reassignment: %lluK != %lluK\n",
size / SZ_1K, config->ggtt_size / SZ_1K);
- return -EREMCHG;
+ err = -EREMCHG;
+ goto out;
}
xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
@@ -460,8 +463,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
config->ggtt_shift = start - (s64)config->ggtt_base;
config->ggtt_base = start;
config->ggtt_size = size;
+ err = config->ggtt_size ? 0 : -ENODATA;
- return config->ggtt_size ? 0 : -ENODATA;
+out:
+ up_write(&config->lock);
+ return err;
}
static int vf_get_lmem_info(struct xe_gt *gt)
@@ -474,22 +480,28 @@ static int vf_get_lmem_info(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
+ down_write(&config->lock);
+
err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
if (unlikely(err))
- return err;
+ goto out;
if (config->lmem_size && config->lmem_size != size) {
xe_gt_sriov_err(gt, "Unexpected LMEM reassignment: %lluM != %lluM\n",
size / SZ_1M, config->lmem_size / SZ_1M);
- return -EREMCHG;
+ err = -EREMCHG;
+ goto out;
}
string_get_size(size, 1, STRING_UNITS_2, size_str, sizeof(size_str));
xe_gt_sriov_dbg_verbose(gt, "LMEM %lluM %s\n", size / SZ_1M, size_str);
config->lmem_size = size;
+ err = config->lmem_size ? 0 : -ENODATA;
- return config->lmem_size ? 0 : -ENODATA;
+out:
+ up_write(&config->lock);
+ return err;
}
static int vf_get_submission_cfg(struct xe_gt *gt)
@@ -501,23 +513,27 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
+ down_write(&config->lock);
+
err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
if (unlikely(err))
- return err;
+ goto out;
err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_DOORBELLS_KEY, &num_dbs);
if (unlikely(err))
- return err;
+ goto out;
if (config->num_ctxs && config->num_ctxs != num_ctxs) {
xe_gt_sriov_err(gt, "Unexpected CTXs reassignment: %u != %u\n",
num_ctxs, config->num_ctxs);
- return -EREMCHG;
+ err = -EREMCHG;
+ goto out;
}
if (config->num_dbs && config->num_dbs != num_dbs) {
xe_gt_sriov_err(gt, "Unexpected DBs reassignment: %u != %u\n",
num_dbs, config->num_dbs);
- return -EREMCHG;
+ err = -EREMCHG;
+ goto out;
}
xe_gt_sriov_dbg_verbose(gt, "CTXs %u DBs %u\n", num_ctxs, num_dbs);
@@ -525,7 +541,11 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
config->num_ctxs = num_ctxs;
config->num_dbs = num_dbs;
- return config->num_ctxs ? 0 : -ENODATA;
+ err = config->num_ctxs ? 0 : -ENODATA;
+
+out:
+ up_write(&config->lock);
+ return err;
}
static void vf_cache_gmdid(struct xe_gt *gt)
@@ -579,11 +599,18 @@ int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
*/
u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
{
+ struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
+ u16 val;
+
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
- xe_gt_assert(gt, gt->sriov.vf.self_config.num_ctxs);
- return gt->sriov.vf.self_config.num_ctxs;
+ down_read(&config->lock);
+ xe_gt_assert(gt, config->num_ctxs);
+ val = config->num_ctxs;
+ up_read(&config->lock);
+
+ return val;
}
/**
@@ -596,11 +623,18 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
*/
u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
{
+ struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
+ u64 val;
+
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
- xe_gt_assert(gt, gt->sriov.vf.self_config.lmem_size);
- return gt->sriov.vf.self_config.lmem_size;
+ down_read(&config->lock);
+ xe_gt_assert(gt, config->lmem_size);
+ val = config->lmem_size;
+ up_read(&config->lock);
+
+ return val;
}
/**
@@ -613,11 +647,17 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
*/
u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
{
+ struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
+ u64 val;
+
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
- xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
- return gt->sriov.vf.self_config.ggtt_size;
+ down_read(&config->lock);
+ val = config->ggtt_size;
+ up_read(&config->lock);
+
+ return val;
}
/**
@@ -630,11 +670,18 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
*/
u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
{
+ struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
+ u64 val;
+
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
- xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
- return gt->sriov.vf.self_config.ggtt_base;
+ down_read(&config->lock);
+ xe_gt_assert(gt, config->ggtt_size);
+ val = config->ggtt_base;
+ up_read(&config->lock);
+
+ return val;
}
/**
@@ -648,11 +695,16 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
{
struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
+ s64 val;
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, xe_gt_is_main_type(gt));
- return config->ggtt_shift;
+ down_read(&config->lock);
+ val = config->ggtt_shift;
+ up_read(&config->lock);
+
+ return val;
}
static int relay_action_handshake(struct xe_gt *gt, u32 *major, u32 *minor)
@@ -1044,6 +1096,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
+ down_read(&config->lock);
drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
config->ggtt_base,
config->ggtt_base + config->ggtt_size - 1);
@@ -1060,6 +1113,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
+ up_read(&config->lock);
}
/**
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index 298dedf4b009..d95857bd789b 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -6,6 +6,7 @@
#ifndef _XE_GT_SRIOV_VF_TYPES_H_
#define _XE_GT_SRIOV_VF_TYPES_H_
+#include <linux/rwsem.h>
#include <linux/types.h>
#include "xe_uc_fw_types.h"
@@ -25,6 +26,8 @@ struct xe_gt_sriov_vf_selfconfig {
u16 num_ctxs;
/** @num_dbs: assigned number of GuC doorbells IDs. */
u16 num_dbs;
+ /** @lock: lock for protecting access to all selfconfig fields. */
+ struct rw_semaphore lock;
};
/**
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index cdd9f8e78b2a..d6e2ed9b9bbc 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -197,6 +197,12 @@ static void vf_migration_init_early(struct xe_device *xe)
*/
void xe_sriov_vf_init_early(struct xe_device *xe)
{
+ struct xe_gt *gt;
+ unsigned int id;
+
+ for_each_gt(gt, xe, id)
+ init_rwsem(>->sriov.vf.self_config.lock);
+
vf_migration_init_early(xe);
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init
2025-09-24 1:15 ` [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init Matthew Brost
@ 2025-09-24 9:29 ` Michal Wajdeczko
2025-09-24 20:23 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 9:29 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> From: Tomasz Lis <tomasz.lis@intel.com>
>
> Protect access to GGTT config as this is non-static information.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 96 ++++++++++++++++++-----
> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 3 +
> drivers/gpu/drm/xe/xe_sriov_vf.c | 6 ++
> 3 files changed, 84 insertions(+), 21 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 0461d5513487..016c867e5e2b 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -440,18 +440,21 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> + down_write(&config->lock);
do we need to protect also VF2GUC communication ..
> +
> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
> if (unlikely(err))
> - return err;
> + goto out;
>
> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_SIZE_KEY, &size);
> if (unlikely(err))
> - return err;
> + goto out;
>
.. or just RW access to the data itself from here?
> if (config->ggtt_size && config->ggtt_size != size) {
> xe_gt_sriov_err(gt, "Unexpected GGTT reassignment: %lluK != %lluK\n",
> size / SZ_1K, config->ggtt_size / SZ_1K);
> - return -EREMCHG;
> + err = -EREMCHG;
> + goto out;
> }
>
> xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
> @@ -460,8 +463,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
> config->ggtt_shift = start - (s64)config->ggtt_base;
> config->ggtt_base = start;
> config->ggtt_size = size;
> + err = config->ggtt_size ? 0 : -ENODATA;
>
> - return config->ggtt_size ? 0 : -ENODATA;
> +out:
> + up_write(&config->lock);
> + return err;
> }
>
> static int vf_get_lmem_info(struct xe_gt *gt)
> @@ -474,22 +480,28 @@ static int vf_get_lmem_info(struct xe_gt *gt)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> + down_write(&config->lock);
> +
> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
> if (unlikely(err))
> - return err;
> + goto out;
>
> if (config->lmem_size && config->lmem_size != size) {
> xe_gt_sriov_err(gt, "Unexpected LMEM reassignment: %lluM != %lluM\n",
> size / SZ_1M, config->lmem_size / SZ_1M);
> - return -EREMCHG;
> + err = -EREMCHG;
> + goto out;
> }
>
> string_get_size(size, 1, STRING_UNITS_2, size_str, sizeof(size_str));
> xe_gt_sriov_dbg_verbose(gt, "LMEM %lluM %s\n", size / SZ_1M, size_str);
>
> config->lmem_size = size;
> + err = config->lmem_size ? 0 : -ENODATA;
>
> - return config->lmem_size ? 0 : -ENODATA;
> +out:
> + up_write(&config->lock);
> + return err;
> }
>
> static int vf_get_submission_cfg(struct xe_gt *gt)
> @@ -501,23 +513,27 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> + down_write(&config->lock);
> +
> err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
> if (unlikely(err))
> - return err;
> + goto out;
>
> err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_DOORBELLS_KEY, &num_dbs);
> if (unlikely(err))
> - return err;
> + goto out;
>
> if (config->num_ctxs && config->num_ctxs != num_ctxs) {
> xe_gt_sriov_err(gt, "Unexpected CTXs reassignment: %u != %u\n",
> num_ctxs, config->num_ctxs);
> - return -EREMCHG;
> + err = -EREMCHG;
> + goto out;
> }
> if (config->num_dbs && config->num_dbs != num_dbs) {
> xe_gt_sriov_err(gt, "Unexpected DBs reassignment: %u != %u\n",
> num_dbs, config->num_dbs);
> - return -EREMCHG;
> + err = -EREMCHG;
> + goto out;
> }
>
> xe_gt_sriov_dbg_verbose(gt, "CTXs %u DBs %u\n", num_ctxs, num_dbs);
> @@ -525,7 +541,11 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
> config->num_ctxs = num_ctxs;
> config->num_dbs = num_dbs;
>
> - return config->num_ctxs ? 0 : -ENODATA;
> + err = config->num_ctxs ? 0 : -ENODATA;
> +
> +out:
> + up_write(&config->lock);
> + return err;
> }
>
> static void vf_cache_gmdid(struct xe_gt *gt)
> @@ -579,11 +599,18 @@ int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
> */
> u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
> {
> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> + u16 val;
> +
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> - xe_gt_assert(gt, gt->sriov.vf.self_config.num_ctxs);
the purpose of this assert was to make sure no part of the VF driver
is trying to get num_ctxs before we finish querying it from the GuC ..
>
> - return gt->sriov.vf.self_config.num_ctxs;
> + down_read(&config->lock);
> + xe_gt_assert(gt, config->num_ctxs);
.. instead of moving this assert in the middle of the lock ..
> + val = config->num_ctxs;
> + up_read(&config->lock);
> +
.. for code clarity you can add assert here as:
xe_gt_assert(gt, val);
> + return val;
> }
>
> /**
> @@ -596,11 +623,18 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
> */
> u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
> {
> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> + u64 val;
> +
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> - xe_gt_assert(gt, gt->sriov.vf.self_config.lmem_size);
ditto
>
> - return gt->sriov.vf.self_config.lmem_size;
> + down_read(&config->lock);
> + xe_gt_assert(gt, config->lmem_size);
> + val = config->lmem_size;
> + up_read(&config->lock);
> +
> + return val;
> }
>
> /**
> @@ -613,11 +647,17 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
> */
> u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
> {
> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> + u64 val;
> +
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> - xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
>
> - return gt->sriov.vf.self_config.ggtt_size;
> + down_read(&config->lock);
> + val = config->ggtt_size;
> + up_read(&config->lock);
> +
> + return val;
> }
>
> /**
> @@ -630,11 +670,18 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
> */
> u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
> {
> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> + u64 val;
> +
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> - xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
>
> - return gt->sriov.vf.self_config.ggtt_base;
> + down_read(&config->lock);
> + xe_gt_assert(gt, config->ggtt_size);
> + val = config->ggtt_base;
> + up_read(&config->lock);
> +
> + return val;
> }
>
> /**
> @@ -648,11 +695,16 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
> s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
> {
> struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> + s64 val;
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, xe_gt_is_main_type(gt));
>
> - return config->ggtt_shift;
> + down_read(&config->lock);
> + val = config->ggtt_shift;
> + up_read(&config->lock);
> +
> + return val;
> }
>
> static int relay_action_handshake(struct xe_gt *gt, u32 *major, u32 *minor)
> @@ -1044,6 +1096,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> + down_read(&config->lock);
> drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
> config->ggtt_base,
> config->ggtt_base + config->ggtt_size - 1);
> @@ -1060,6 +1113,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
>
> drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
> drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
> + up_read(&config->lock);
> }
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> index 298dedf4b009..d95857bd789b 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> @@ -6,6 +6,7 @@
> #ifndef _XE_GT_SRIOV_VF_TYPES_H_
> #define _XE_GT_SRIOV_VF_TYPES_H_
>
> +#include <linux/rwsem.h>
> #include <linux/types.h>
> #include "xe_uc_fw_types.h"
>
> @@ -25,6 +26,8 @@ struct xe_gt_sriov_vf_selfconfig {
> u16 num_ctxs;
> /** @num_dbs: assigned number of GuC doorbells IDs. */
> u16 num_dbs;
> + /** @lock: lock for protecting access to all selfconfig fields. */
> + struct rw_semaphore lock;
> };
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> index cdd9f8e78b2a..d6e2ed9b9bbc 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> @@ -197,6 +197,12 @@ static void vf_migration_init_early(struct xe_device *xe)
> */
> void xe_sriov_vf_init_early(struct xe_device *xe)
> {
> + struct xe_gt *gt;
> + unsigned int id;
> +
> + for_each_gt(gt, xe, id)
> + init_rwsem(>->sriov.vf.self_config.lock);
this is a wrong place - VF's per-GT data shall be initialized in
the per-GT code somewhere in xe_gt_sriov_vf.c
> +
> vf_migration_init_early(xe);
> }
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init
2025-09-24 9:29 ` Michal Wajdeczko
@ 2025-09-24 20:23 ` Matthew Brost
2025-09-30 0:42 ` Lis, Tomasz
0 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:23 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 11:29:22AM +0200, Michal Wajdeczko wrote:
>
>
I'll let Tomasz respond to most of the commennts as it is his patch but
replying to last comment.
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > From: Tomasz Lis <tomasz.lis@intel.com>
> >
> > Protect access to GGTT config as this is non-static information.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 96 ++++++++++++++++++-----
> > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 3 +
> > drivers/gpu/drm/xe/xe_sriov_vf.c | 6 ++
> > 3 files changed, 84 insertions(+), 21 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 0461d5513487..016c867e5e2b 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -440,18 +440,21 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
> >
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> >
> > + down_write(&config->lock);
>
> do we need to protect also VF2GUC communication ..
>
> > +
> > err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
> > if (unlikely(err))
> > - return err;
> > + goto out;
> >
> > err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_SIZE_KEY, &size);
> > if (unlikely(err))
> > - return err;
> > + goto out;
> >
>
> .. or just RW access to the data itself from here?
>
> > if (config->ggtt_size && config->ggtt_size != size) {
> > xe_gt_sriov_err(gt, "Unexpected GGTT reassignment: %lluK != %lluK\n",
> > size / SZ_1K, config->ggtt_size / SZ_1K);
> > - return -EREMCHG;
> > + err = -EREMCHG;
> > + goto out;
> > }
> >
> > xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
> > @@ -460,8 +463,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
> > config->ggtt_shift = start - (s64)config->ggtt_base;
> > config->ggtt_base = start;
> > config->ggtt_size = size;
> > + err = config->ggtt_size ? 0 : -ENODATA;
> >
> > - return config->ggtt_size ? 0 : -ENODATA;
> > +out:
> > + up_write(&config->lock);
> > + return err;
> > }
> >
> > static int vf_get_lmem_info(struct xe_gt *gt)
> > @@ -474,22 +480,28 @@ static int vf_get_lmem_info(struct xe_gt *gt)
> >
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> >
> > + down_write(&config->lock);
> > +
> > err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
> > if (unlikely(err))
> > - return err;
> > + goto out;
> >
> > if (config->lmem_size && config->lmem_size != size) {
> > xe_gt_sriov_err(gt, "Unexpected LMEM reassignment: %lluM != %lluM\n",
> > size / SZ_1M, config->lmem_size / SZ_1M);
> > - return -EREMCHG;
> > + err = -EREMCHG;
> > + goto out;
> > }
> >
> > string_get_size(size, 1, STRING_UNITS_2, size_str, sizeof(size_str));
> > xe_gt_sriov_dbg_verbose(gt, "LMEM %lluM %s\n", size / SZ_1M, size_str);
> >
> > config->lmem_size = size;
> > + err = config->lmem_size ? 0 : -ENODATA;
> >
> > - return config->lmem_size ? 0 : -ENODATA;
> > +out:
> > + up_write(&config->lock);
> > + return err;
> > }
> >
> > static int vf_get_submission_cfg(struct xe_gt *gt)
> > @@ -501,23 +513,27 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
> >
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> >
> > + down_write(&config->lock);
> > +
> > err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
> > if (unlikely(err))
> > - return err;
> > + goto out;
> >
> > err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_DOORBELLS_KEY, &num_dbs);
> > if (unlikely(err))
> > - return err;
> > + goto out;
> >
> > if (config->num_ctxs && config->num_ctxs != num_ctxs) {
> > xe_gt_sriov_err(gt, "Unexpected CTXs reassignment: %u != %u\n",
> > num_ctxs, config->num_ctxs);
> > - return -EREMCHG;
> > + err = -EREMCHG;
> > + goto out;
> > }
> > if (config->num_dbs && config->num_dbs != num_dbs) {
> > xe_gt_sriov_err(gt, "Unexpected DBs reassignment: %u != %u\n",
> > num_dbs, config->num_dbs);
> > - return -EREMCHG;
> > + err = -EREMCHG;
> > + goto out;
> > }
> >
> > xe_gt_sriov_dbg_verbose(gt, "CTXs %u DBs %u\n", num_ctxs, num_dbs);
> > @@ -525,7 +541,11 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
> > config->num_ctxs = num_ctxs;
> > config->num_dbs = num_dbs;
> >
> > - return config->num_ctxs ? 0 : -ENODATA;
> > + err = config->num_ctxs ? 0 : -ENODATA;
> > +
> > +out:
> > + up_write(&config->lock);
> > + return err;
> > }
> >
> > static void vf_cache_gmdid(struct xe_gt *gt)
> > @@ -579,11 +599,18 @@ int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
> > */
> > u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
> > {
> > + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > + u16 val;
> > +
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > - xe_gt_assert(gt, gt->sriov.vf.self_config.num_ctxs);
>
> the purpose of this assert was to make sure no part of the VF driver
> is trying to get num_ctxs before we finish querying it from the GuC ..
>
> >
> > - return gt->sriov.vf.self_config.num_ctxs;
> > + down_read(&config->lock);
> > + xe_gt_assert(gt, config->num_ctxs);
>
> .. instead of moving this assert in the middle of the lock ..
>
> > + val = config->num_ctxs;
> > + up_read(&config->lock);
> > +
>
> .. for code clarity you can add assert here as:
>
> xe_gt_assert(gt, val);
> > + return val;
> > }
> >
> > /**
> > @@ -596,11 +623,18 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
> > */
> > u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
> > {
> > + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > + u64 val;
> > +
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > - xe_gt_assert(gt, gt->sriov.vf.self_config.lmem_size);
>
> ditto
>
> >
> > - return gt->sriov.vf.self_config.lmem_size;
> > + down_read(&config->lock);
> > + xe_gt_assert(gt, config->lmem_size);
> > + val = config->lmem_size;
> > + up_read(&config->lock);
> > +
> > + return val;
> > }
> >
> > /**
> > @@ -613,11 +647,17 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
> > */
> > u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
> > {
> > + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > + u64 val;
> > +
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > - xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
> >
> > - return gt->sriov.vf.self_config.ggtt_size;
> > + down_read(&config->lock);
> > + val = config->ggtt_size;
> > + up_read(&config->lock);
> > +
> > + return val;
> > }
> >
> > /**
> > @@ -630,11 +670,18 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
> > */
> > u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
> > {
> > + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > + u64 val;
> > +
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > - xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
> >
> > - return gt->sriov.vf.self_config.ggtt_base;
> > + down_read(&config->lock);
> > + xe_gt_assert(gt, config->ggtt_size);
> > + val = config->ggtt_base;
> > + up_read(&config->lock);
> > +
> > + return val;
> > }
> >
> > /**
> > @@ -648,11 +695,16 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
> > s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
> > {
> > struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > + s64 val;
> >
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, xe_gt_is_main_type(gt));
> >
> > - return config->ggtt_shift;
> > + down_read(&config->lock);
> > + val = config->ggtt_shift;
> > + up_read(&config->lock);
> > +
> > + return val;
> > }
> >
> > static int relay_action_handshake(struct xe_gt *gt, u32 *major, u32 *minor)
> > @@ -1044,6 +1096,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
> >
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> >
> > + down_read(&config->lock);
> > drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
> > config->ggtt_base,
> > config->ggtt_base + config->ggtt_size - 1);
> > @@ -1060,6 +1113,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
> >
> > drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
> > drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
> > + up_read(&config->lock);
> > }
> >
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > index 298dedf4b009..d95857bd789b 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > @@ -6,6 +6,7 @@
> > #ifndef _XE_GT_SRIOV_VF_TYPES_H_
> > #define _XE_GT_SRIOV_VF_TYPES_H_
> >
> > +#include <linux/rwsem.h>
> > #include <linux/types.h>
> > #include "xe_uc_fw_types.h"
> >
> > @@ -25,6 +26,8 @@ struct xe_gt_sriov_vf_selfconfig {
> > u16 num_ctxs;
> > /** @num_dbs: assigned number of GuC doorbells IDs. */
> > u16 num_dbs;
> > + /** @lock: lock for protecting access to all selfconfig fields. */
> > + struct rw_semaphore lock;
> > };
> >
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > index cdd9f8e78b2a..d6e2ed9b9bbc 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > @@ -197,6 +197,12 @@ static void vf_migration_init_early(struct xe_device *xe)
> > */
> > void xe_sriov_vf_init_early(struct xe_device *xe)
> > {
> > + struct xe_gt *gt;
> > + unsigned int id;
> > +
> > + for_each_gt(gt, xe, id)
> > + init_rwsem(>->sriov.vf.self_config.lock);
>
> this is a wrong place - VF's per-GT data shall be initialized in
> the per-GT code somewhere in xe_gt_sriov_vf.c
>
Yes, it is moved to the proper place in [1].
Matt
[1] https://patchwork.freedesktop.org/patch/676383/?series=154627&rev=2
> > +
> > vf_migration_init_early(xe);
> > }
> >
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init
2025-09-24 20:23 ` Matthew Brost
@ 2025-09-30 0:42 ` Lis, Tomasz
0 siblings, 0 replies; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-30 0:42 UTC (permalink / raw)
To: Matthew Brost, Michal Wajdeczko; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 11291 bytes --]
On 9/24/2025 10:23 PM, Matthew Brost wrote:
> On Wed, Sep 24, 2025 at 11:29:22AM +0200, Michal Wajdeczko wrote:
>>
> I'll let Tomasz respond to most of the commennts as it is his patch but
> replying to last comment.
ok, will answer.
>> On 9/24/2025 3:15 AM, Matthew Brost wrote:
>>> From: Tomasz Lis<tomasz.lis@intel.com>
>>>
>>> Protect access to GGTT config as this is non-static information.
>>>
>>> Signed-off-by: Matthew Brost<matthew.brost@intel.com>
>>> Signed-off-by: Tomasz Lis<tomasz.lis@intel.com>
>>> ---
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 96 ++++++++++++++++++-----
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 3 +
>>> drivers/gpu/drm/xe/xe_sriov_vf.c | 6 ++
>>> 3 files changed, 84 insertions(+), 21 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> index 0461d5513487..016c867e5e2b 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> @@ -440,18 +440,21 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
>>>
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>>
>>> + down_write(&config->lock);
>> do we need to protect also VF2GUC communication ..
>>
>>> +
>>> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
>>> if (unlikely(err))
>>> - return err;
>>> + goto out;
>>>
>>> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_SIZE_KEY, &size);
>>> if (unlikely(err))
>>> - return err;
>>> + goto out;
>>>
>> .. or just RW access to the data itself from here?
The protection must start before we receive GGTT start. It would make
little sense to protect just manipulation of data which we've already
received.
The post-migration recovery must have a clear state: either we got the
GGTT already, or not. It cannot have a state "it's not written
everywhere yet, but GuC may have sent it already, who knows".
The GGTT address must be protected from before the moment GuC reads it
in order to write to VF G2H CTB.
>>
>>> if (config->ggtt_size && config->ggtt_size != size) {
>>> xe_gt_sriov_err(gt, "Unexpected GGTT reassignment: %lluK != %lluK\n",
>>> size / SZ_1K, config->ggtt_size / SZ_1K);
>>> - return -EREMCHG;
>>> + err = -EREMCHG;
>>> + goto out;
>>> }
>>>
>>> xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
>>> @@ -460,8 +463,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
>>> config->ggtt_shift = start - (s64)config->ggtt_base;
>>> config->ggtt_base = start;
>>> config->ggtt_size = size;
>>> + err = config->ggtt_size ? 0 : -ENODATA;
>>>
>>> - return config->ggtt_size ? 0 : -ENODATA;
>>> +out:
>>> + up_write(&config->lock);
>>> + return err;
>>> }
>>>
>>> static int vf_get_lmem_info(struct xe_gt *gt)
>>> @@ -474,22 +480,28 @@ static int vf_get_lmem_info(struct xe_gt *gt)
>>>
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>>
>>> + down_write(&config->lock);
>>> +
>>> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
>>> if (unlikely(err))
>>> - return err;
>>> + goto out;
>>>
>>> if (config->lmem_size && config->lmem_size != size) {
>>> xe_gt_sriov_err(gt, "Unexpected LMEM reassignment: %lluM != %lluM\n",
>>> size / SZ_1M, config->lmem_size / SZ_1M);
>>> - return -EREMCHG;
>>> + err = -EREMCHG;
>>> + goto out;
>>> }
>>>
>>> string_get_size(size, 1, STRING_UNITS_2, size_str, sizeof(size_str));
>>> xe_gt_sriov_dbg_verbose(gt, "LMEM %lluM %s\n", size / SZ_1M, size_str);
>>>
>>> config->lmem_size = size;
>>> + err = config->lmem_size ? 0 : -ENODATA;
>>>
>>> - return config->lmem_size ? 0 : -ENODATA;
>>> +out:
>>> + up_write(&config->lock);
>>> + return err;
>>> }
>>>
>>> static int vf_get_submission_cfg(struct xe_gt *gt)
>>> @@ -501,23 +513,27 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
>>>
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>>
>>> + down_write(&config->lock);
>>> +
>>> err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
>>> if (unlikely(err))
>>> - return err;
>>> + goto out;
>>>
>>> err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_DOORBELLS_KEY, &num_dbs);
>>> if (unlikely(err))
>>> - return err;
>>> + goto out;
>>>
>>> if (config->num_ctxs && config->num_ctxs != num_ctxs) {
>>> xe_gt_sriov_err(gt, "Unexpected CTXs reassignment: %u != %u\n",
>>> num_ctxs, config->num_ctxs);
>>> - return -EREMCHG;
>>> + err = -EREMCHG;
>>> + goto out;
>>> }
>>> if (config->num_dbs && config->num_dbs != num_dbs) {
>>> xe_gt_sriov_err(gt, "Unexpected DBs reassignment: %u != %u\n",
>>> num_dbs, config->num_dbs);
>>> - return -EREMCHG;
>>> + err = -EREMCHG;
>>> + goto out;
>>> }
>>>
>>> xe_gt_sriov_dbg_verbose(gt, "CTXs %u DBs %u\n", num_ctxs, num_dbs);
>>> @@ -525,7 +541,11 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
>>> config->num_ctxs = num_ctxs;
>>> config->num_dbs = num_dbs;
>>>
>>> - return config->num_ctxs ? 0 : -ENODATA;
>>> + err = config->num_ctxs ? 0 : -ENODATA;
>>> +
>>> +out:
>>> + up_write(&config->lock);
>>> + return err;
>>> }
>>>
>>> static void vf_cache_gmdid(struct xe_gt *gt)
>>> @@ -579,11 +599,18 @@ int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
>>> */
>>> u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
>>> {
>>> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
>>> + u16 val;
>>> +
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
>>> - xe_gt_assert(gt, gt->sriov.vf.self_config.num_ctxs);
>> the purpose of this assert was to make sure no part of the VF driver
>> is trying to get num_ctxs before we finish querying it from the GuC ..
We now know this will never happen _before_we_finish_querying_, as
there's a lock. Read side will wait for the write side to finish.
So unless you're worried about a situation before we _start_ querying,
it's ok to even remove it completely.
>>
>>>
>>> - return gt->sriov.vf.self_config.num_ctxs;
>>> + down_read(&config->lock);
>>> + xe_gt_assert(gt, config->num_ctxs);
>> .. instead of moving this assert in the middle of the lock ..
>>
>>> + val = config->num_ctxs;
>>> + up_read(&config->lock);
>>> +
>> .. for code clarity you can add assert here as:
>>
>> xe_gt_assert(gt, val);
sure, can do that.
>>> + return val;
>>> }
>>>
>>> /**
>>> @@ -596,11 +623,18 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
>>> */
>>> u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
>>> {
>>> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
>>> + u64 val;
>>> +
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
>>> - xe_gt_assert(gt, gt->sriov.vf.self_config.lmem_size);
>> ditto
ack, though whether this stays at all, depends on further discussion in
v3 of this series.
-Tomasz
>>
>>>
>>> - return gt->sriov.vf.self_config.lmem_size;
>>> + down_read(&config->lock);
>>> + xe_gt_assert(gt, config->lmem_size);
>>> + val = config->lmem_size;
>>> + up_read(&config->lock);
>>> +
>>> + return val;
>>> }
>>>
>>> /**
>>> @@ -613,11 +647,17 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
>>> */
>>> u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
>>> {
>>> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
>>> + u64 val;
>>> +
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
>>> - xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
>>>
>>> - return gt->sriov.vf.self_config.ggtt_size;
>>> + down_read(&config->lock);
>>> + val = config->ggtt_size;
>>> + up_read(&config->lock);
>>> +
>>> + return val;
>>> }
>>>
>>> /**
>>> @@ -630,11 +670,18 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
>>> */
>>> u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
>>> {
>>> + struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
>>> + u64 val;
>>> +
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
>>> - xe_gt_assert(gt, gt->sriov.vf.self_config.ggtt_size);
>>>
>>> - return gt->sriov.vf.self_config.ggtt_base;
>>> + down_read(&config->lock);
>>> + xe_gt_assert(gt, config->ggtt_size);
>>> + val = config->ggtt_base;
>>> + up_read(&config->lock);
>>> +
>>> + return val;
>>> }
>>>
>>> /**
>>> @@ -648,11 +695,16 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
>>> s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
>>> {
>>> struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
>>> + s64 val;
>>>
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>> xe_gt_assert(gt, xe_gt_is_main_type(gt));
>>>
>>> - return config->ggtt_shift;
>>> + down_read(&config->lock);
>>> + val = config->ggtt_shift;
>>> + up_read(&config->lock);
>>> +
>>> + return val;
>>> }
>>>
>>> static int relay_action_handshake(struct xe_gt *gt, u32 *major, u32 *minor)
>>> @@ -1044,6 +1096,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
>>>
>>> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>>
>>> + down_read(&config->lock);
>>> drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
>>> config->ggtt_base,
>>> config->ggtt_base + config->ggtt_size - 1);
>>> @@ -1060,6 +1113,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
>>>
>>> drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
>>> drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
>>> + up_read(&config->lock);
>>> }
>>>
>>> /**
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> index 298dedf4b009..d95857bd789b 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> @@ -6,6 +6,7 @@
>>> #ifndef _XE_GT_SRIOV_VF_TYPES_H_
>>> #define _XE_GT_SRIOV_VF_TYPES_H_
>>>
>>> +#include <linux/rwsem.h>
>>> #include <linux/types.h>
>>> #include "xe_uc_fw_types.h"
>>>
>>> @@ -25,6 +26,8 @@ struct xe_gt_sriov_vf_selfconfig {
>>> u16 num_ctxs;
>>> /** @num_dbs: assigned number of GuC doorbells IDs. */
>>> u16 num_dbs;
>>> + /** @lock: lock for protecting access to all selfconfig fields. */
>>> + struct rw_semaphore lock;
>>> };
>>>
>>> /**
>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
>>> index cdd9f8e78b2a..d6e2ed9b9bbc 100644
>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
>>> @@ -197,6 +197,12 @@ static void vf_migration_init_early(struct xe_device *xe)
>>> */
>>> void xe_sriov_vf_init_early(struct xe_device *xe)
>>> {
>>> + struct xe_gt *gt;
>>> + unsigned int id;
>>> +
>>> + for_each_gt(gt, xe, id)
>>> + init_rwsem(>->sriov.vf.self_config.lock);
>> this is a wrong place - VF's per-GT data shall be initialized in
>> the per-GT code somewhere in xe_gt_sriov_vf.c
>>
> Yes, it is moved to the proper place in [1].
>
> Matt
>
> [1]https://patchwork.freedesktop.org/patch/676383/?series=154627&rev=2
>
>>> +
>>> vf_migration_init_early(xe);
>>> }
>>>
[-- Attachment #2: Type: text/html, Size: 14018 bytes --]
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 02/34] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery"
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
2025-09-24 1:15 ` [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 9:32 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 03/34] Revert "drm/xe/vf: Post migration, repopulate ring area for pending request" Matthew Brost
` (34 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
This reverts commit ba180a362128cb71d16c3f0ce6645448011d2607.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/abi/guc_actions_abi.h | 8 ----
drivers/gpu/drm/xe/xe_guc_submit.c | 54 ------------------------
2 files changed, 62 deletions(-)
diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
index 31090c69dfbe..47756e4674a1 100644
--- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
@@ -196,14 +196,6 @@ enum xe_guc_register_context_multi_lrc_param_offsets {
XE_GUC_REGISTER_CONTEXT_MULTI_LRC_MSG_MIN_LEN = 11,
};
-enum xe_guc_context_wq_item_offsets {
- XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN = 0,
- XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW,
- XE_GUC_CONTEXT_WQ_EL_INFO_DATA_2_GUCCTX_RINGTAIL_FREEZEPOCS,
- XE_GUC_CONTEXT_WQ_EL_INFO_DATA_3_WI_FENCE_ID,
- XE_GUC_CONTEXT_WQ_EL_CHILD_LIST_DATA_4_RINGTAIL,
-};
-
enum xe_guc_report_status {
XE_GUC_REPORT_STATUS_UNKNOWN = 0x0,
XE_GUC_REPORT_STATUS_ACKED = 0x1,
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 53024eb5670b..3ac0950f55be 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -735,18 +735,12 @@ static void wq_item_append(struct xe_exec_queue *q)
if (wq_wait_for_space(q, wqi_size))
return;
- xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN);
wqi[i++] = FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_MULTI_LRC) |
FIELD_PREP(WQ_LEN_MASK, len_dw);
- xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW);
wqi[i++] = xe_lrc_descriptor(q->lrc[0]);
- xe_gt_assert(guc_to_gt(guc), i ==
- XE_GUC_CONTEXT_WQ_EL_INFO_DATA_2_GUCCTX_RINGTAIL_FREEZEPOCS);
wqi[i++] = FIELD_PREP(WQ_GUC_ID_MASK, q->guc->id) |
FIELD_PREP(WQ_RING_TAIL_MASK, q->lrc[0]->ring.tail / sizeof(u64));
- xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_INFO_DATA_3_WI_FENCE_ID);
wqi[i++] = 0;
- xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_CHILD_LIST_DATA_4_RINGTAIL);
for (j = 1; j < q->width; ++j) {
struct xe_lrc *lrc = q->lrc[j];
@@ -767,50 +761,6 @@ static void wq_item_append(struct xe_exec_queue *q)
parallel_write(xe, map, wq_desc.tail, q->guc->wqi_tail);
}
-static int wq_items_rebase(struct xe_exec_queue *q)
-{
- struct xe_guc *guc = exec_queue_to_guc(q);
- struct xe_device *xe = guc_to_xe(guc);
- struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]);
- int i = q->guc->wqi_head;
-
- /* the ring starts after a header struct */
- iosys_map_incr(&map, offsetof(struct guc_submit_parallel_scratch, wq[0]));
-
- while ((i % WQ_SIZE) != (q->guc->wqi_tail % WQ_SIZE)) {
- u32 len_dw, type, val;
-
- if (drm_WARN_ON_ONCE(&xe->drm, i < 0 || i > 2 * WQ_SIZE))
- break;
-
- val = xe_map_rd_ring_u32(xe, &map, i / sizeof(u32) +
- XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN,
- WQ_SIZE / sizeof(u32));
- len_dw = FIELD_GET(WQ_LEN_MASK, val);
- type = FIELD_GET(WQ_TYPE_MASK, val);
-
- if (drm_WARN_ON_ONCE(&xe->drm, len_dw >= WQ_SIZE / sizeof(u32)))
- break;
-
- if (type == WQ_TYPE_MULTI_LRC) {
- val = xe_lrc_descriptor(q->lrc[0]);
- xe_map_wr_ring_u32(xe, &map, i / sizeof(u32) +
- XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW,
- WQ_SIZE / sizeof(u32), val);
- } else if (drm_WARN_ON_ONCE(&xe->drm, type != WQ_TYPE_NOOP)) {
- break;
- }
-
- i += (len_dw + 1) * sizeof(u32);
- }
-
- if ((i % WQ_SIZE) != (q->guc->wqi_tail % WQ_SIZE)) {
- xe_gt_err(q->gt, "Exec queue fixups incomplete - wqi parse failed\n");
- return -EBADMSG;
- }
- return 0;
-}
-
#define RESUME_PENDING ~0x0ull
static void submit_exec_queue(struct xe_exec_queue *q)
{
@@ -2669,10 +2619,6 @@ int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch)
err = xe_exec_queue_contexts_hwsp_rebase(q, scratch);
if (err)
break;
- if (xe_exec_queue_is_parallel(q))
- err = wq_items_rebase(q);
- if (err)
- break;
}
mutex_unlock(&guc->submission_state.lock);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 02/34] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery"
2025-09-24 1:15 ` [PATCH v2 02/34] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery" Matthew Brost
@ 2025-09-24 9:32 ` Michal Wajdeczko
2025-09-24 20:17 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 9:32 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> This reverts commit ba180a362128cb71d16c3f0ce6645448011d2607.
we still need the "why" part here, at least something like:
"Due to change in the VF migration recovery design this code
is not needed any more."
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/abi/guc_actions_abi.h | 8 ----
> drivers/gpu/drm/xe/xe_guc_submit.c | 54 ------------------------
> 2 files changed, 62 deletions(-)
^ permalink raw reply [flat|nested] 90+ messages in thread
* Re: [PATCH v2 02/34] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery"
2025-09-24 9:32 ` Michal Wajdeczko
@ 2025-09-24 20:17 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:17 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 11:32:07AM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > This reverts commit ba180a362128cb71d16c3f0ce6645448011d2607.
>
> we still need the "why" part here, at least something like:
>
> "Due to change in the VF migration recovery design this code
> is not needed any more."
>
Lucas also mentioned this me, I'll add explaination in commit messages
for the reverts.
Matt
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/abi/guc_actions_abi.h | 8 ----
> > drivers/gpu/drm/xe/xe_guc_submit.c | 54 ------------------------
> > 2 files changed, 62 deletions(-)
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 03/34] Revert "drm/xe/vf: Post migration, repopulate ring area for pending request"
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
2025-09-24 1:15 ` [PATCH v2 01/34] drm/xe/vf: Lock querying GGTT config during driver init Matthew Brost
2025-09-24 1:15 ` [PATCH v2 02/34] Revert "drm/xe/vf: Rebase exec queue parallel commands during migration recovery" Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 04/34] Revert "drm/xe/vf: Fixup CTB send buffer messages after migration" Matthew Brost
` (33 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
This reverts commit a0dda25d24e636df5c30a9370464b7cebc709faf.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_exec_queue.c | 24 ------------------------
drivers/gpu/drm/xe/xe_exec_queue.h | 3 +--
drivers/gpu/drm/xe/xe_guc_submit.c | 24 ------------------------
drivers/gpu/drm/xe/xe_guc_submit.h | 2 --
drivers/gpu/drm/xe/xe_sriov_vf.c | 1 -
5 files changed, 1 insertion(+), 53 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 37b2b93b73d6..6bfaca424ca3 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -1123,27 +1123,3 @@ int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch)
return err;
}
-
-/**
- * xe_exec_queue_jobs_ring_restore - Re-emit ring commands of requests pending on given queue.
- * @q: the &xe_exec_queue struct instance
- */
-void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q)
-{
- struct xe_gpu_scheduler *sched = &q->guc->sched;
- struct xe_sched_job *job;
-
- /*
- * This routine is used within VF migration recovery. This means
- * using the lock here introduces a restriction: we cannot wait
- * for any GFX HW response while the lock is taken.
- */
- spin_lock(&sched->base.job_list_lock);
- list_for_each_entry(job, &sched->base.pending_list, drm.list) {
- if (xe_sched_job_is_error(job))
- continue;
-
- q->ring_ops->emit_job(job);
- }
- spin_unlock(&sched->base.job_list_lock);
-}
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index 15ec852e7f7e..8821ceb838d0 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -92,7 +92,6 @@ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q);
int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch);
-void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q);
-
struct xe_lrc *xe_exec_queue_lrc(struct xe_exec_queue *q);
+
#endif
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 3ac0950f55be..16f78376f196 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -845,30 +845,6 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
return fence;
}
-/**
- * xe_guc_jobs_ring_rebase - Re-emit ring commands of requests pending
- * on all queues under a guc.
- * @guc: the &xe_guc struct instance
- */
-void xe_guc_jobs_ring_rebase(struct xe_guc *guc)
-{
- struct xe_exec_queue *q;
- unsigned long index;
-
- /*
- * This routine is used within VF migration recovery. This means
- * using the lock here introduces a restriction: we cannot wait
- * for any GFX HW response while the lock is taken.
- */
- mutex_lock(&guc->submission_state.lock);
- xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) {
- if (exec_queue_killed_or_banned_or_wedged(q))
- continue;
- xe_exec_queue_jobs_ring_restore(q);
- }
- mutex_unlock(&guc->submission_state.lock);
-}
-
static void guc_exec_queue_free_job(struct drm_sched_job *drm_job)
{
struct xe_sched_job *job = to_xe_sched_job(drm_job);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 78c3f07e31a0..5b4a0a6fd818 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -36,8 +36,6 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg,
int xe_guc_exec_queue_reset_failure_handler(struct xe_guc *guc, u32 *msg, u32 len);
int xe_guc_error_capture_handler(struct xe_guc *guc, u32 *msg, u32 len);
-void xe_guc_jobs_ring_rebase(struct xe_guc *guc);
-
struct xe_guc_submit_exec_queue_snapshot *
xe_guc_exec_queue_snapshot_capture(struct xe_exec_queue *q);
void
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index d6e2ed9b9bbc..0581b881b628 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -340,7 +340,6 @@ static int gt_vf_post_migration_fixups(struct xe_gt *gt)
err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
if (err)
goto out;
- xe_guc_jobs_ring_rebase(>->uc.guc);
xe_guc_ct_fixup_messages_with_ggtt(>->uc.guc.ct, shift);
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 04/34] Revert "drm/xe/vf: Fixup CTB send buffer messages after migration"
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (2 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 03/34] Revert "drm/xe/vf: Post migration, repopulate ring area for pending request" Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 05/34] drm/xe: Save off position in ring in which a job was programmed Matthew Brost
` (32 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
This reverts commit cef88d1265cac7d415606af73ba58926fd3cd8b7.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_guc_ct.c | 183 -------------------------------
drivers/gpu/drm/xe/xe_guc_ct.h | 2 -
drivers/gpu/drm/xe/xe_map.h | 18 ---
drivers/gpu/drm/xe/xe_sriov_vf.c | 2 -
4 files changed, 205 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 18f6327bf552..47079ab9922c 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -25,7 +25,6 @@
#include "xe_gt_printk.h"
#include "xe_gt_sriov_pf_control.h"
#include "xe_gt_sriov_pf_monitor.h"
-#include "xe_gt_sriov_printk.h"
#include "xe_guc.h"
#include "xe_guc_log.h"
#include "xe_guc_relay.h"
@@ -93,8 +92,6 @@ struct g2h_fence {
bool done;
};
-#define make_u64(hi, lo) ((u64)((u64)(u32)(hi) << 32 | (u32)(lo)))
-
static void g2h_fence_init(struct g2h_fence *g2h_fence, u32 *response_buffer)
{
memset(g2h_fence, 0, sizeof(*g2h_fence));
@@ -1793,186 +1790,6 @@ static void g2h_worker_func(struct work_struct *w)
receive_g2h(ct);
}
-static void xe_fixup_u64_in_cmds(struct xe_device *xe, struct iosys_map *cmds,
- u32 size, u32 idx, s64 shift)
-{
- u32 hi, lo;
- u64 offset;
-
- lo = xe_map_rd_ring_u32(xe, cmds, idx, size);
- hi = xe_map_rd_ring_u32(xe, cmds, idx + 1, size);
- offset = make_u64(hi, lo);
- offset += shift;
- lo = lower_32_bits(offset);
- hi = upper_32_bits(offset);
- xe_map_wr_ring_u32(xe, cmds, idx, size, lo);
- xe_map_wr_ring_u32(xe, cmds, idx + 1, size, hi);
-}
-
-/*
- * Shift any GGTT addresses within a single message left within CTB from
- * before post-migration recovery.
- * @ct: pointer to CT struct of the target GuC
- * @cmds: iomap buffer containing CT messages
- * @head: start of the target message within the buffer
- * @len: length of the target message
- * @size: size of the commands buffer
- * @shift: the address shift to be added to each GGTT reference
- * Return: true if the message was fixed or needed no fixups, false on failure
- */
-static bool ct_fixup_ggtt_in_message(struct xe_guc_ct *ct,
- struct iosys_map *cmds, u32 head,
- u32 len, u32 size, s64 shift)
-{
- struct xe_gt *gt = ct_to_gt(ct);
- struct xe_device *xe = ct_to_xe(ct);
- u32 msg[GUC_HXG_MSG_MIN_LEN];
- u32 action, i, n;
-
- xe_gt_assert(gt, len >= GUC_HXG_MSG_MIN_LEN);
-
- msg[0] = xe_map_rd_ring_u32(xe, cmds, head, size);
- action = FIELD_GET(GUC_HXG_REQUEST_MSG_0_ACTION, msg[0]);
-
- xe_gt_sriov_dbg_verbose(gt, "fixing H2G %#x\n", action);
-
- switch (action) {
- case XE_GUC_ACTION_REGISTER_CONTEXT:
- if (len != XE_GUC_REGISTER_CONTEXT_MSG_LEN)
- goto err_len;
- xe_fixup_u64_in_cmds(xe, cmds, size, head +
- XE_GUC_REGISTER_CONTEXT_DATA_5_WQ_DESC_ADDR_LOWER,
- shift);
- xe_fixup_u64_in_cmds(xe, cmds, size, head +
- XE_GUC_REGISTER_CONTEXT_DATA_7_WQ_BUF_BASE_LOWER,
- shift);
- xe_fixup_u64_in_cmds(xe, cmds, size, head +
- XE_GUC_REGISTER_CONTEXT_DATA_10_HW_LRC_ADDR, shift);
- break;
- case XE_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC:
- if (len < XE_GUC_REGISTER_CONTEXT_MULTI_LRC_MSG_MIN_LEN)
- goto err_len;
- n = xe_map_rd_ring_u32(xe, cmds, head +
- XE_GUC_REGISTER_CONTEXT_MULTI_LRC_DATA_10_NUM_CTXS, size);
- if (len != XE_GUC_REGISTER_CONTEXT_MULTI_LRC_MSG_MIN_LEN + 2 * n)
- goto err_len;
- xe_fixup_u64_in_cmds(xe, cmds, size, head +
- XE_GUC_REGISTER_CONTEXT_MULTI_LRC_DATA_5_WQ_DESC_ADDR_LOWER,
- shift);
- xe_fixup_u64_in_cmds(xe, cmds, size, head +
- XE_GUC_REGISTER_CONTEXT_MULTI_LRC_DATA_7_WQ_BUF_BASE_LOWER,
- shift);
- for (i = 0; i < n; i++)
- xe_fixup_u64_in_cmds(xe, cmds, size, head +
- XE_GUC_REGISTER_CONTEXT_MULTI_LRC_DATA_11_HW_LRC_ADDR
- + 2 * i, shift);
- break;
- default:
- break;
- }
- return true;
-
-err_len:
- xe_gt_err(gt, "Skipped G2G %#x message fixups, unexpected length (%u)\n", action, len);
- return false;
-}
-
-/*
- * Apply fixups to the next outgoing CT message within given CTB
- * @ct: the &xe_guc_ct struct instance representing the target GuC
- * @h2g: the &guc_ctb struct instance of the target buffer
- * @shift: shift to be added to all GGTT addresses within the CTB
- * @mhead: pointer to an integer storing message start position; the
- * position is changed to next message before this function return
- * @avail: size of the area available for parsing, that is length
- * of all remaining messages stored within the CTB
- * Return: size of the area available for parsing after one message
- * has been parsed, that is length remaining from the updated mhead
- */
-static int ct_fixup_ggtt_in_buffer(struct xe_guc_ct *ct, struct guc_ctb *h2g,
- s64 shift, u32 *mhead, s32 avail)
-{
- struct xe_gt *gt = ct_to_gt(ct);
- struct xe_device *xe = ct_to_xe(ct);
- u32 msg[GUC_HXG_MSG_MIN_LEN];
- u32 size = h2g->info.size;
- u32 head = *mhead;
- u32 len;
-
- xe_gt_assert(gt, avail >= (s32)GUC_CTB_MSG_MIN_LEN);
-
- /* Read header */
- msg[0] = xe_map_rd_ring_u32(xe, &h2g->cmds, head, size);
- len = FIELD_GET(GUC_CTB_MSG_0_NUM_DWORDS, msg[0]) + GUC_CTB_MSG_MIN_LEN;
-
- if (unlikely(len > (u32)avail)) {
- xe_gt_err(gt, "H2G channel broken on read, avail=%d, len=%d, fixups skipped\n",
- avail, len);
- return 0;
- }
-
- head = (head + GUC_CTB_MSG_MIN_LEN) % size;
- if (!ct_fixup_ggtt_in_message(ct, &h2g->cmds, head, msg_len_to_hxg_len(len), size, shift))
- return 0;
- *mhead = (head + msg_len_to_hxg_len(len)) % size;
-
- return avail - len;
-}
-
-/**
- * xe_guc_ct_fixup_messages_with_ggtt - Fixup any pending H2G CTB messages
- * @ct: pointer to CT struct of the target GuC
- * @ggtt_shift: shift to be added to all GGTT addresses within the CTB
- *
- * Messages in GuC to Host CTB are owned by GuC and any fixups in them
- * are made by GuC. But content of the Host to GuC CTB is owned by the
- * KMD, so fixups to GGTT references in any pending messages need to be
- * applied here.
- * This function updates GGTT offsets in payloads of pending H2G CTB
- * messages (messages which were not consumed by GuC before the VF got
- * paused).
- */
-void xe_guc_ct_fixup_messages_with_ggtt(struct xe_guc_ct *ct, s64 ggtt_shift)
-{
- struct guc_ctb *h2g = &ct->ctbs.h2g;
- struct xe_guc *guc = ct_to_guc(ct);
- struct xe_gt *gt = guc_to_gt(guc);
- u32 head, tail, size;
- s32 avail;
-
- if (unlikely(h2g->info.broken))
- return;
-
- h2g->info.head = desc_read(ct_to_xe(ct), h2g, head);
- head = h2g->info.head;
- tail = READ_ONCE(h2g->info.tail);
- size = h2g->info.size;
-
- if (unlikely(head > size))
- goto corrupted;
-
- if (unlikely(tail >= size))
- goto corrupted;
-
- avail = tail - head;
-
- /* beware of buffer wrap case */
- if (unlikely(avail < 0))
- avail += size;
- xe_gt_dbg(gt, "available %d (%u:%u:%u)\n", avail, head, tail, size);
- xe_gt_assert(gt, avail >= 0);
-
- while (avail > 0)
- avail = ct_fixup_ggtt_in_buffer(ct, h2g, ggtt_shift, &head, avail);
-
- return;
-
-corrupted:
- xe_gt_err(gt, "Corrupted H2G descriptor head=%u tail=%u size=%u, fixups not applied\n",
- head, tail, size);
- h2g->info.broken = true;
-}
-
static struct xe_guc_ct_snapshot *guc_ct_snapshot_alloc(struct xe_guc_ct *ct, bool atomic,
bool want_ctb)
{
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
index cf41210ab30a..d6c81325a76c 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.h
+++ b/drivers/gpu/drm/xe/xe_guc_ct.h
@@ -24,8 +24,6 @@ void xe_guc_ct_snapshot_print(struct xe_guc_ct_snapshot *snapshot, struct drm_pr
void xe_guc_ct_snapshot_free(struct xe_guc_ct_snapshot *snapshot);
void xe_guc_ct_print(struct xe_guc_ct *ct, struct drm_printer *p, bool want_ctb);
-void xe_guc_ct_fixup_messages_with_ggtt(struct xe_guc_ct *ct, s64 ggtt_shift);
-
static inline bool xe_guc_ct_initialized(struct xe_guc_ct *ct)
{
return ct->state != XE_GUC_CT_STATE_NOT_INITIALIZED;
diff --git a/drivers/gpu/drm/xe/xe_map.h b/drivers/gpu/drm/xe/xe_map.h
index 8d67f6ba2d95..f62e0c8b67ab 100644
--- a/drivers/gpu/drm/xe/xe_map.h
+++ b/drivers/gpu/drm/xe/xe_map.h
@@ -78,24 +78,6 @@ static inline void xe_map_write32(struct xe_device *xe, struct iosys_map *map,
iosys_map_wr(map__, offset__, type__, val__); \
})
-#define xe_map_rd_array(xe__, map__, index__, type__) \
- xe_map_rd(xe__, map__, (index__) * sizeof(type__), type__)
-
-#define xe_map_wr_array(xe__, map__, index__, type__, val__) \
- xe_map_wr(xe__, map__, (index__) * sizeof(type__), type__, val__)
-
-#define xe_map_rd_array_u32(xe__, map__, index__) \
- xe_map_rd_array(xe__, map__, index__, u32)
-
-#define xe_map_wr_array_u32(xe__, map__, index__, val__) \
- xe_map_wr_array(xe__, map__, index__, u32, val__)
-
-#define xe_map_rd_ring_u32(xe__, map__, index__, size__) \
- xe_map_rd_array_u32(xe__, map__, (index__) % (size__))
-
-#define xe_map_wr_ring_u32(xe__, map__, index__, size__, val__) \
- xe_map_wr_array_u32(xe__, map__, (index__) % (size__), val__)
-
#define xe_map_rd_field(xe__, map__, struct_offset__, struct_type__, field__) ({ \
struct xe_device *__xe = xe__; \
xe_device_assert_mem_access(__xe); \
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 0581b881b628..da064a1e7419 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -12,7 +12,6 @@
#include "xe_gt_sriov_printk.h"
#include "xe_gt_sriov_vf.h"
#include "xe_guc.h"
-#include "xe_guc_ct.h"
#include "xe_guc_submit.h"
#include "xe_irq.h"
#include "xe_lrc.h"
@@ -340,7 +339,6 @@ static int gt_vf_post_migration_fixups(struct xe_gt *gt)
err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
if (err)
goto out;
- xe_guc_ct_fixup_messages_with_ggtt(>->uc.guc.ct, shift);
}
out:
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 05/34] drm/xe: Save off position in ring in which a job was programmed
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (3 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 04/34] Revert "drm/xe/vf: Fixup CTB send buffer messages after migration" Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 06/34] drm/xe/guc: Track pending-enable source in submission state Matthew Brost
` (31 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
VF post-migration recovery needs to modify the ring with updated GGTT
addresses for pending jobs. Save off position in ring in which a job was
programmed to facilitate.
v4:
- s/VF resume/VF post-migration recovery (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
---
drivers/gpu/drm/xe/xe_ring_ops.c | 23 +++++++++++++++++++----
drivers/gpu/drm/xe/xe_sched_job_types.h | 5 +++++
2 files changed, 24 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
index d71837773d6c..ac0c6dcffe15 100644
--- a/drivers/gpu/drm/xe/xe_ring_ops.c
+++ b/drivers/gpu/drm/xe/xe_ring_ops.c
@@ -245,12 +245,14 @@ static int emit_copy_timestamp(struct xe_lrc *lrc, u32 *dw, int i)
/* for engines that don't require any special HW handling (no EUs, no aux inval, etc) */
static void __emit_job_gen12_simple(struct xe_sched_job *job, struct xe_lrc *lrc,
- u64 batch_addr, u32 seqno)
+ u64 batch_addr, u32 *head, u32 seqno)
{
u32 dw[MAX_JOB_SIZE_DW], i = 0;
u32 ppgtt_flag = get_ppgtt_flag(job);
struct xe_gt *gt = job->q->gt;
+ *head = lrc->ring.tail;
+
i = emit_copy_timestamp(lrc, dw, i);
if (job->ring_ops_flush_tlb) {
@@ -296,7 +298,7 @@ static bool has_aux_ccs(struct xe_device *xe)
}
static void __emit_job_gen12_video(struct xe_sched_job *job, struct xe_lrc *lrc,
- u64 batch_addr, u32 seqno)
+ u64 batch_addr, u32 *head, u32 seqno)
{
u32 dw[MAX_JOB_SIZE_DW], i = 0;
u32 ppgtt_flag = get_ppgtt_flag(job);
@@ -304,6 +306,8 @@ static void __emit_job_gen12_video(struct xe_sched_job *job, struct xe_lrc *lrc,
struct xe_device *xe = gt_to_xe(gt);
bool decode = job->q->class == XE_ENGINE_CLASS_VIDEO_DECODE;
+ *head = lrc->ring.tail;
+
i = emit_copy_timestamp(lrc, dw, i);
dw[i++] = preparser_disable(true);
@@ -346,7 +350,8 @@ static void __emit_job_gen12_video(struct xe_sched_job *job, struct xe_lrc *lrc,
static void __emit_job_gen12_render_compute(struct xe_sched_job *job,
struct xe_lrc *lrc,
- u64 batch_addr, u32 seqno)
+ u64 batch_addr, u32 *head,
+ u32 seqno)
{
u32 dw[MAX_JOB_SIZE_DW], i = 0;
u32 ppgtt_flag = get_ppgtt_flag(job);
@@ -355,6 +360,8 @@ static void __emit_job_gen12_render_compute(struct xe_sched_job *job,
bool lacks_render = !(gt->info.engine_mask & XE_HW_ENGINE_RCS_MASK);
u32 mask_flags = 0;
+ *head = lrc->ring.tail;
+
i = emit_copy_timestamp(lrc, dw, i);
dw[i++] = preparser_disable(true);
@@ -396,11 +403,14 @@ static void __emit_job_gen12_render_compute(struct xe_sched_job *job,
}
static void emit_migration_job_gen12(struct xe_sched_job *job,
- struct xe_lrc *lrc, u32 seqno)
+ struct xe_lrc *lrc, u32 *head,
+ u32 seqno)
{
u32 saddr = xe_lrc_start_seqno_ggtt_addr(lrc);
u32 dw[MAX_JOB_SIZE_DW], i = 0;
+ *head = lrc->ring.tail;
+
i = emit_copy_timestamp(lrc, dw, i);
i = emit_store_imm_ggtt(saddr, seqno, dw, i);
@@ -434,6 +444,7 @@ static void emit_job_gen12_gsc(struct xe_sched_job *job)
__emit_job_gen12_simple(job, job->q->lrc[0],
job->ptrs[0].batch_addr,
+ &job->ptrs[0].head,
xe_sched_job_lrc_seqno(job));
}
@@ -443,6 +454,7 @@ static void emit_job_gen12_copy(struct xe_sched_job *job)
if (xe_sched_job_is_migration(job->q)) {
emit_migration_job_gen12(job, job->q->lrc[0],
+ &job->ptrs[0].head,
xe_sched_job_lrc_seqno(job));
return;
}
@@ -450,6 +462,7 @@ static void emit_job_gen12_copy(struct xe_sched_job *job)
for (i = 0; i < job->q->width; ++i)
__emit_job_gen12_simple(job, job->q->lrc[i],
job->ptrs[i].batch_addr,
+ &job->ptrs[i].head,
xe_sched_job_lrc_seqno(job));
}
@@ -461,6 +474,7 @@ static void emit_job_gen12_video(struct xe_sched_job *job)
for (i = 0; i < job->q->width; ++i)
__emit_job_gen12_video(job, job->q->lrc[i],
job->ptrs[i].batch_addr,
+ &job->ptrs[i].head,
xe_sched_job_lrc_seqno(job));
}
@@ -471,6 +485,7 @@ static void emit_job_gen12_render_compute(struct xe_sched_job *job)
for (i = 0; i < job->q->width; ++i)
__emit_job_gen12_render_compute(job, job->q->lrc[i],
job->ptrs[i].batch_addr,
+ &job->ptrs[i].head,
xe_sched_job_lrc_seqno(job));
}
diff --git a/drivers/gpu/drm/xe/xe_sched_job_types.h b/drivers/gpu/drm/xe/xe_sched_job_types.h
index dbf260dded8d..7ce58765a34a 100644
--- a/drivers/gpu/drm/xe/xe_sched_job_types.h
+++ b/drivers/gpu/drm/xe/xe_sched_job_types.h
@@ -24,6 +24,11 @@ struct xe_job_ptrs {
struct dma_fence_chain *chain_fence;
/** @batch_addr: Batch buffer address. */
u64 batch_addr;
+ /**
+ * @head: The tail pointer of the LRC (so head pointer of job) when the
+ * job was submitted
+ */
+ u32 head;
};
/**
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 06/34] drm/xe/guc: Track pending-enable source in submission state
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (4 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 05/34] drm/xe: Save off position in ring in which a job was programmed Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 07/34] drm/xe: Track LR jobs in DRM scheduler pending list Matthew Brost
` (30 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Add explicit tracking in the GuC submission state to record the source
of a pending enable (TDR vs. queue resume path vs. submission).
Disambiguating the origin lets the GuC submission state machine apply
the correct recovery/replay behavior.
This helps VF restore: when the device comes back, the state machine knows
whether the pending enable stems from timeout recovery, from a queue resume
sequence, or submission and can gate sequencing and fixups accordingly.
v4:
- Clarify commit message (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
---
drivers/gpu/drm/xe/xe_guc_submit.c | 36 ++++++++++++++++++++++++++++++
1 file changed, 36 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 16f78376f196..13746f32b231 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -69,6 +69,8 @@ exec_queue_to_guc(struct xe_exec_queue *q)
#define EXEC_QUEUE_STATE_BANNED (1 << 9)
#define EXEC_QUEUE_STATE_CHECK_TIMEOUT (1 << 10)
#define EXEC_QUEUE_STATE_EXTRA_REF (1 << 11)
+#define EXEC_QUEUE_STATE_PENDING_RESUME (1 << 12)
+#define EXEC_QUEUE_STATE_PENDING_TDR_EXIT (1 << 13)
static bool exec_queue_registered(struct xe_exec_queue *q)
{
@@ -220,6 +222,36 @@ static void set_exec_queue_extra_ref(struct xe_exec_queue *q)
atomic_or(EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
}
+static bool __maybe_unused exec_queue_pending_resume(struct xe_exec_queue *q)
+{
+ return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_RESUME;
+}
+
+static void set_exec_queue_pending_resume(struct xe_exec_queue *q)
+{
+ atomic_or(EXEC_QUEUE_STATE_PENDING_RESUME, &q->guc->state);
+}
+
+static void clear_exec_queue_pending_resume(struct xe_exec_queue *q)
+{
+ atomic_and(~EXEC_QUEUE_STATE_PENDING_RESUME, &q->guc->state);
+}
+
+static bool __maybe_unused exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
+{
+ return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_TDR_EXIT;
+}
+
+static void set_exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
+{
+ atomic_or(EXEC_QUEUE_STATE_PENDING_TDR_EXIT, &q->guc->state);
+}
+
+static void clear_exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
+{
+ atomic_and(~EXEC_QUEUE_STATE_PENDING_TDR_EXIT, &q->guc->state);
+}
+
static bool exec_queue_killed_or_banned_or_wedged(struct xe_exec_queue *q)
{
return (atomic_read(&q->guc->state) &
@@ -1334,6 +1366,7 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
return DRM_GPU_SCHED_STAT_RESET;
sched_enable:
+ set_exec_queue_pending_tdr_exit(q);
enable_scheduling(q);
rearm:
/*
@@ -1493,6 +1526,7 @@ static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg)
clear_exec_queue_suspended(q);
if (!exec_queue_enabled(q)) {
q->guc->resume_time = RESUME_PENDING;
+ set_exec_queue_pending_resume(q);
enable_scheduling(q);
}
} else {
@@ -2065,6 +2099,8 @@ static void handle_sched_done(struct xe_guc *guc, struct xe_exec_queue *q,
xe_gt_assert(guc_to_gt(guc), exec_queue_pending_enable(q));
q->guc->resume_time = ktime_get();
+ clear_exec_queue_pending_resume(q);
+ clear_exec_queue_pending_tdr_exit(q);
clear_exec_queue_pending_enable(q);
smp_wmb();
wake_up_all(&guc->ct.wq);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 07/34] drm/xe: Track LR jobs in DRM scheduler pending list
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (5 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 06/34] drm/xe/guc: Track pending-enable source in submission state Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 08/34] drm/xe: Don't change LRC ring head on job resubmission Matthew Brost
` (29 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
VF migration requires jobs to remain pending so they can be replayed
after the VF comes back. Previously, LR job fences were intentionally
signaled immediately after submission to avoid the risk of exporting
them, as these fences do not naturally signal in a timely manner and
could break dma-fence contracts. A side effect of this approach was that
LR jobs were never added to the DRM scheduler’s pending list, preventing
them from being tracked for later resubmission.
We now avoid signaling LR job fences and ensure they are never exported;
Xe already guards against exporting these internal fences. With that
guarantee in place, we can safely track LR jobs in the scheduler’s
pending list so they are eligible for resubmission during VF
post-migration recovery (and similar recovery paths).
An added benefit is that LR queues now gain the DRM scheduler’s built-in
flow control over ring usage rather than rejecting new jobs in the exec
IOCTL if the ring is full.
v2:
- Ensure DRM scheduler TDR doesn't run for LR jobs
- Stack variable for killed_or_banned_or_wedged
v4:
- Clarify commit message (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
---
drivers/gpu/drm/xe/xe_exec.c | 12 ++-------
drivers/gpu/drm/xe/xe_exec_queue.c | 19 -------------
drivers/gpu/drm/xe/xe_exec_queue.h | 2 --
drivers/gpu/drm/xe/xe_guc_submit.c | 43 ++++++++++++++++++++----------
4 files changed, 31 insertions(+), 45 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index 83897950f0da..0dc27476832b 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -124,7 +124,7 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
struct xe_validation_ctx ctx;
struct xe_sched_job *job;
struct xe_vm *vm;
- bool write_locked, skip_retry = false;
+ bool write_locked;
int err = 0;
struct xe_hw_engine_group *group;
enum xe_hw_engine_group_execution_mode mode, previous_mode;
@@ -266,12 +266,6 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
goto err_exec;
}
- if (xe_exec_queue_is_lr(q) && xe_exec_queue_ring_full(q)) {
- err = -EWOULDBLOCK; /* Aliased to -EAGAIN */
- skip_retry = true;
- goto err_exec;
- }
-
if (xe_exec_queue_uses_pxp(q)) {
err = xe_vm_validate_protected(q->vm);
if (err)
@@ -328,8 +322,6 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
xe_sched_job_init_user_fence(job, &syncs[i]);
}
- if (xe_exec_queue_is_lr(q))
- q->ring_ops->emit_job(job);
if (!xe_vm_in_lr_mode(vm))
xe_exec_queue_last_fence_set(q, vm, &job->drm.s_fence->finished);
xe_sched_job_push(job);
@@ -355,7 +347,7 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
xe_validation_ctx_fini(&ctx);
err_unlock_list:
up_read(&vm->lock);
- if (err == -EAGAIN && !skip_retry)
+ if (err == -EAGAIN)
goto retry;
err_hw_exec_mode:
if (mode == EXEC_MODE_DMA_FENCE)
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 6bfaca424ca3..81f707d2c388 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -824,25 +824,6 @@ bool xe_exec_queue_is_lr(struct xe_exec_queue *q)
!(q->flags & EXEC_QUEUE_FLAG_VM);
}
-static s32 xe_exec_queue_num_job_inflight(struct xe_exec_queue *q)
-{
- return q->lrc[0]->fence_ctx.next_seqno - xe_lrc_seqno(q->lrc[0]) - 1;
-}
-
-/**
- * xe_exec_queue_ring_full() - Whether an exec_queue's ring is full
- * @q: The exec_queue
- *
- * Return: True if the exec_queue's ring is full, false otherwise.
- */
-bool xe_exec_queue_ring_full(struct xe_exec_queue *q)
-{
- struct xe_lrc *lrc = q->lrc[0];
- s32 max_job = lrc->ring.size / MAX_JOB_SIZE_BYTES;
-
- return xe_exec_queue_num_job_inflight(q) >= max_job;
-}
-
/**
* xe_exec_queue_is_idle() - Whether an exec_queue is idle.
* @q: The exec_queue
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index 8821ceb838d0..a4dfbe858bda 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -64,8 +64,6 @@ static inline bool xe_exec_queue_uses_pxp(struct xe_exec_queue *q)
bool xe_exec_queue_is_lr(struct xe_exec_queue *q);
-bool xe_exec_queue_ring_full(struct xe_exec_queue *q);
-
bool xe_exec_queue_is_idle(struct xe_exec_queue *q);
void xe_exec_queue_kill(struct xe_exec_queue *q);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 13746f32b231..3a534d93505f 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -851,30 +851,31 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
struct xe_sched_job *job = to_xe_sched_job(drm_job);
struct xe_exec_queue *q = job->q;
struct xe_guc *guc = exec_queue_to_guc(q);
- struct dma_fence *fence = NULL;
- bool lr = xe_exec_queue_is_lr(q);
+ bool lr = xe_exec_queue_is_lr(q), killed_or_banned_or_wedged =
+ exec_queue_killed_or_banned_or_wedged(q);
xe_gt_assert(guc_to_gt(guc), !(exec_queue_destroyed(q) || exec_queue_pending_disable(q)) ||
exec_queue_banned(q) || exec_queue_suspended(q));
trace_xe_sched_job_run(job);
- if (!exec_queue_killed_or_banned_or_wedged(q) && !xe_sched_job_is_error(job)) {
+ if (!killed_or_banned_or_wedged && !xe_sched_job_is_error(job)) {
if (!exec_queue_registered(q))
register_exec_queue(q, GUC_CONTEXT_NORMAL);
- if (!lr) /* LR jobs are emitted in the exec IOCTL */
- q->ring_ops->emit_job(job);
+ q->ring_ops->emit_job(job);
submit_exec_queue(q);
}
- if (lr) {
- xe_sched_job_set_error(job, -EOPNOTSUPP);
- dma_fence_put(job->fence); /* Drop ref from xe_sched_job_arm */
- } else {
- fence = job->fence;
- }
+ /*
+ * We don't care about job-fence ordering in LR VMs because these fences
+ * are never exported; they are used solely to keep jobs on the pending
+ * list. Once a queue enters an error state, there's no need to track
+ * them.
+ */
+ if (killed_or_banned_or_wedged && lr)
+ xe_sched_job_set_error(job, -ECANCELED);
- return fence;
+ return job->fence;
}
static void guc_exec_queue_free_job(struct drm_sched_job *drm_job)
@@ -916,7 +917,8 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
xe_gt_warn(q->gt, "Pending enable/disable failed to respond\n");
xe_sched_submission_start(sched);
xe_gt_reset_async(q->gt);
- xe_sched_tdr_queue_imm(sched);
+ if (!xe_exec_queue_is_lr(q))
+ xe_sched_tdr_queue_imm(sched);
return;
}
@@ -1008,6 +1010,7 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
struct xe_exec_queue *q = ge->q;
struct xe_guc *guc = exec_queue_to_guc(q);
struct xe_gpu_scheduler *sched = &ge->sched;
+ struct xe_sched_job *job;
bool wedged = false;
xe_gt_assert(guc_to_gt(guc), xe_exec_queue_is_lr(q));
@@ -1058,7 +1061,16 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
if (!exec_queue_killed(q) && !xe_lrc_ring_is_idle(q->lrc[0]))
xe_devcoredump(q, NULL, "LR job cleanup, guc_id=%d", q->guc->id);
+ xe_hw_fence_irq_stop(q->fence_irq);
+
xe_sched_submission_start(sched);
+
+ spin_lock(&sched->base.job_list_lock);
+ list_for_each_entry(job, &sched->base.pending_list, drm.list)
+ xe_sched_job_set_error(job, -ECANCELED);
+ spin_unlock(&sched->base.job_list_lock);
+
+ xe_hw_fence_irq_start(q->fence_irq);
}
#define ADJUST_FIVE_PERCENT(__t) mul_u64_u32_div(__t, 105, 100)
@@ -1129,7 +1141,8 @@ static void enable_scheduling(struct xe_exec_queue *q)
xe_gt_warn(guc_to_gt(guc), "Schedule enable failed to respond");
set_exec_queue_banned(q);
xe_gt_reset_async(q->gt);
- xe_sched_tdr_queue_imm(&q->guc->sched);
+ if (!xe_exec_queue_is_lr(q))
+ xe_sched_tdr_queue_imm(&q->guc->sched);
}
}
@@ -1187,6 +1200,8 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
int i = 0;
bool wedged = false, skip_timeout_check;
+ xe_gt_assert(guc_to_gt(guc), !xe_exec_queue_is_lr(q));
+
/*
* TDR has fired before free job worker. Common if exec queue
* immediately closed after last fence signaled. Add back to pending
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 08/34] drm/xe: Don't change LRC ring head on job resubmission
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (6 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 07/34] drm/xe: Track LR jobs in DRM scheduler pending list Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 15:14 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent Matthew Brost
` (28 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Now that we save the job's head during submission, it's no longer
necessary to adjust the LRC ring head during resubmission. Instead, a
software-based adjustment of the tail will overwrite the old jobs in
place. For some odd reason, adjusting the LRC ring head didn't work on
parallel queues, which was causing issues in our CI.
v6:
- Also set LRC tail to head so queue is idle coming out of reset
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_guc_submit.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 3a534d93505f..70306f902ba5 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2008,11 +2008,17 @@ static void guc_exec_queue_start(struct xe_exec_queue *q)
struct xe_gpu_scheduler *sched = &q->guc->sched;
if (!exec_queue_killed_or_banned_or_wedged(q)) {
+ struct xe_sched_job *job = xe_sched_first_pending_job(sched);
int i;
trace_xe_exec_queue_resubmit(q);
- for (i = 0; i < q->width; ++i)
- xe_lrc_set_ring_head(q->lrc[i], q->lrc[i]->ring.tail);
+ if (job) {
+ for (i = 0; i < q->width; ++i) {
+ q->lrc[i]->ring.tail = job->ptrs[i].head;
+ xe_lrc_set_ring_tail(q->lrc[i],
+ xe_lrc_ring_head(q->lrc[i]));
+ }
+ }
xe_sched_resubmit_jobs(sched);
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 08/34] drm/xe: Don't change LRC ring head on job resubmission
2025-09-24 1:15 ` [PATCH v2 08/34] drm/xe: Don't change LRC ring head on job resubmission Matthew Brost
@ 2025-09-24 15:14 ` Lis, Tomasz
2025-09-25 16:12 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-24 15:14 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Now that we save the job's head during submission, it's no longer
> necessary to adjust the LRC ring head during resubmission. Instead, a
> software-based adjustment of the tail will overwrite the old jobs in
> place. For some odd reason, adjusting the LRC ring head didn't work on
> parallel queues, which was causing issues in our CI.
>
> v6:
> - Also set LRC tail to head so queue is idle coming out of reset
This is complex.. if the job got preempted, then not touching the head makes
sense. But what if the queue was removed by reset? Are the problematic
commands NOP'ed so that we really can leave the head?
Setting tails looks ok.
-Tomasz
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_guc_submit.c | 10 ++++++++--
> 1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index 3a534d93505f..70306f902ba5 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -2008,11 +2008,17 @@ static void guc_exec_queue_start(struct xe_exec_queue *q)
> struct xe_gpu_scheduler *sched = &q->guc->sched;
>
> if (!exec_queue_killed_or_banned_or_wedged(q)) {
> + struct xe_sched_job *job = xe_sched_first_pending_job(sched);
> int i;
>
> trace_xe_exec_queue_resubmit(q);
> - for (i = 0; i < q->width; ++i)
> - xe_lrc_set_ring_head(q->lrc[i], q->lrc[i]->ring.tail);
> + if (job) {
> + for (i = 0; i < q->width; ++i) {
> + q->lrc[i]->ring.tail = job->ptrs[i].head;
> + xe_lrc_set_ring_tail(q->lrc[i],
> + xe_lrc_ring_head(q->lrc[i]));
> + }
> + }
> xe_sched_resubmit_jobs(sched);
> }
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 08/34] drm/xe: Don't change LRC ring head on job resubmission
2025-09-24 15:14 ` Lis, Tomasz
@ 2025-09-25 16:12 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-25 16:12 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Wed, Sep 24, 2025 at 05:14:57PM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Now that we save the job's head during submission, it's no longer
> > necessary to adjust the LRC ring head during resubmission. Instead, a
> > software-based adjustment of the tail will overwrite the old jobs in
> > place. For some odd reason, adjusting the LRC ring head didn't work on
> > parallel queues, which was causing issues in our CI.
> >
> > v6:
> > - Also set LRC tail to head so queue is idle coming out of reset
>
> This is complex.. if the job got preempted, then not touching the head makes
>
> sense. But what if the queue was removed by reset? Are the problematic
>
> commands NOP'ed so that we really can leave the head?
>
> Setting tails looks ok.
>
So this case is a bit different than VF restore. If a job has started
but not completed across GT reset, we just ban the queue unless it is
KMD migration queue so preempted jobs are a non issue. KMD jobs could
potentially have been preempted but copy jobs are simple copies which
would be fine to run twice.
Also coming out of GT reset the queues are not registered with GuC as
this point so I believe it is fine to move KMD own LRC tail back the
original position of the job which we resubmiting.
TL;DR after some thought I believe this patch is correct.
Matt
> -Tomasz
>
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_guc_submit.c | 10 ++++++++--
> > 1 file changed, 8 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> > index 3a534d93505f..70306f902ba5 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> > @@ -2008,11 +2008,17 @@ static void guc_exec_queue_start(struct xe_exec_queue *q)
> > struct xe_gpu_scheduler *sched = &q->guc->sched;
> > if (!exec_queue_killed_or_banned_or_wedged(q)) {
> > + struct xe_sched_job *job = xe_sched_first_pending_job(sched);
> > int i;
> > trace_xe_exec_queue_resubmit(q);
> > - for (i = 0; i < q->width; ++i)
> > - xe_lrc_set_ring_head(q->lrc[i], q->lrc[i]->ring.tail);
> > + if (job) {
> > + for (i = 0; i < q->width; ++i) {
> > + q->lrc[i]->ring.tail = job->ptrs[i].head;
> > + xe_lrc_set_ring_tail(q->lrc[i],
> > + xe_lrc_ring_head(q->lrc[i]));
> > + }
> > + }
> > xe_sched_resubmit_jobs(sched);
> > }
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (7 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 08/34] drm/xe: Don't change LRC ring head on job resubmission Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 14:23 ` Lis, Tomasz
2025-09-24 18:01 ` Lucas De Marchi
2025-09-24 1:15 ` [PATCH v2 10/34] drm/xe/guc: Document GuC submission backend Matthew Brost
` (27 subsequent siblings)
36 siblings, 2 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
The LRC W/A currently checks for LRC being iomem in some places, while
in others it checks if the scratch buffer is non-NULL. This
inconsistency causes issues with the VF post-migration recovery code,
which blindly passes in a scratch buffer.
This patch standardizes the check by consistently verifying whether the
LRC is iomem to determine if the scratch buffer should be used.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_lrc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 47e9df775072..aed0a3155d1f 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -1248,7 +1248,7 @@ static int setup_bo(struct bo_setup_state *state)
static void finish_bo(struct bo_setup_state *state)
{
- if (!state->buffer)
+ if (!state->lrc->bo->vmap.is_iomem)
return;
xe_map_memcpy_to(gt_to_xe(state->lrc->gt), &state->lrc->bo->vmap,
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent
2025-09-24 1:15 ` [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent Matthew Brost
@ 2025-09-24 14:23 ` Lis, Tomasz
2025-09-24 18:01 ` Lucas De Marchi
1 sibling, 0 replies; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-24 14:23 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> The LRC W/A currently checks for LRC being iomem in some places, while
> in others it checks if the scratch buffer is non-NULL. This
> inconsistency causes issues with the VF post-migration recovery code,
> which blindly passes in a scratch buffer.
>
> This patch standardizes the check by consistently verifying whether the
> LRC is iomem to determine if the scratch buffer should be used.
No need for explicit assert as the kernel will abort by itself if
state->buffer in NULL, so this is ok.
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
-Tomasz
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_lrc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
> index 47e9df775072..aed0a3155d1f 100644
> --- a/drivers/gpu/drm/xe/xe_lrc.c
> +++ b/drivers/gpu/drm/xe/xe_lrc.c
> @@ -1248,7 +1248,7 @@ static int setup_bo(struct bo_setup_state *state)
>
> static void finish_bo(struct bo_setup_state *state)
> {
> - if (!state->buffer)
> + if (!state->lrc->bo->vmap.is_iomem)
> return;
>
> xe_map_memcpy_to(gt_to_xe(state->lrc->gt), &state->lrc->bo->vmap,
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent
2025-09-24 1:15 ` [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent Matthew Brost
2025-09-24 14:23 ` Lis, Tomasz
@ 2025-09-24 18:01 ` Lucas De Marchi
2025-09-25 20:25 ` Matthew Brost
1 sibling, 1 reply; 90+ messages in thread
From: Lucas De Marchi @ 2025-09-24 18:01 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
On Tue, Sep 23, 2025 at 06:15:36PM -0700, Matthew Brost wrote:
>The LRC W/A currently checks for LRC being iomem in some places, while
>in others it checks if the scratch buffer is non-NULL. This
>inconsistency causes issues with the VF post-migration recovery code,
>which blindly passes in a scratch buffer.
>
>This patch standardizes the check by consistently verifying whether the
>LRC is iomem to determine if the scratch buffer should be used.
>
>Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>---
> drivers/gpu/drm/xe/xe_lrc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
>diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
>index 47e9df775072..aed0a3155d1f 100644
>--- a/drivers/gpu/drm/xe/xe_lrc.c
>+++ b/drivers/gpu/drm/xe/xe_lrc.c
>@@ -1248,7 +1248,7 @@ static int setup_bo(struct bo_setup_state *state)
>
> static void finish_bo(struct bo_setup_state *state)
> {
>- if (!state->buffer)
>+ if (!state->lrc->bo->vmap.is_iomem)
Please add the fixes as the first patches and find the commits they
are fixing so they can propagate to drm-xe-fixes. With that, for a
palliative fix:
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Long-term thoughts below:
The fact that we exposed xe_lrc_setup_wa_bb_with_scratch() to users
outside xe_lrc seems completely wrong to me... it explains why
indirect_ctx was completely forgotten.
And the ENOMEM handling is in a completely different place than the
allocation which makes this a code smell.
I think this change is a palliative that we could accept, but going
forward I think it needs to be better handled: we shouldn't be
recalculating it on every lrc init, it should rather be handled like the
default_lrc and just copied in place. Then on migration what really
needs to be fixed up is the ggtt address, i.e. only these calls:
xe_lrc_write_ctx_reg(lrc,
CTX_CS_INDIRECT_CTX,
(xe_bo_ggtt_addr(lrc->bo) + state.offset) |
/* Size in CLs. */
(state.written * sizeof(u32) / 64));
and
xe_lrc_write_ctx_reg(lrc, CTX_BB_PER_CTX_PTR,
xe_bo_ggtt_addr(lrc->bo) + state.offset + 1);
without re-allocating and recalculating it over and over. I will take a
look at doing this
Lucas De Marchi
> return;
>
> xe_map_memcpy_to(gt_to_xe(state->lrc->gt), &state->lrc->bo->vmap,
>--
>2.34.1
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent
2025-09-24 18:01 ` Lucas De Marchi
@ 2025-09-25 20:25 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-25 20:25 UTC (permalink / raw)
To: Lucas De Marchi; +Cc: intel-xe
On Wed, Sep 24, 2025 at 01:01:15PM -0500, Lucas De Marchi wrote:
> On Tue, Sep 23, 2025 at 06:15:36PM -0700, Matthew Brost wrote:
> > The LRC W/A currently checks for LRC being iomem in some places, while
> > in others it checks if the scratch buffer is non-NULL. This
> > inconsistency causes issues with the VF post-migration recovery code,
> > which blindly passes in a scratch buffer.
> >
> > This patch standardizes the check by consistently verifying whether the
> > LRC is iomem to determine if the scratch buffer should be used.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_lrc.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
> > index 47e9df775072..aed0a3155d1f 100644
> > --- a/drivers/gpu/drm/xe/xe_lrc.c
> > +++ b/drivers/gpu/drm/xe/xe_lrc.c
> > @@ -1248,7 +1248,7 @@ static int setup_bo(struct bo_setup_state *state)
> >
> > static void finish_bo(struct bo_setup_state *state)
> > {
> > - if (!state->buffer)
> > + if (!state->lrc->bo->vmap.is_iomem)
>
> Please add the fixes as the first patches and find the commits they
> are fixing so they can propagate to drm-xe-fixes. With that, for a
VF migration doesn't work at all upstream, so I don't believe this
warrents a fixes tag.
> palliative fix:
>
> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
>
> Long-term thoughts below:
>
> The fact that we exposed xe_lrc_setup_wa_bb_with_scratch() to users
> outside xe_lrc seems completely wrong to me... it explains why
> indirect_ctx was completely forgotten.
>
Let me add in indirect_ctx fixup in my next rev for completeness.
> And the ENOMEM handling is in a completely different place than the
> allocation which makes this a code smell.
I have a fix for that, we will needs fixes tag for this one.
>
> I think this change is a palliative that we could accept, but going
> forward I think it needs to be better handled: we shouldn't be
> recalculating it on every lrc init, it should rather be handled like the
> default_lrc and just copied in place. Then on migration what really
> needs to be fixed up is the ggtt address, i.e. only these calls:
>
> xe_lrc_write_ctx_reg(lrc,
> CTX_CS_INDIRECT_CTX,
> (xe_bo_ggtt_addr(lrc->bo) + state.offset) |
> /* Size in CLs. */
> (state.written * sizeof(u32) / 64));
>
> and
>
> xe_lrc_write_ctx_reg(lrc, CTX_BB_PER_CTX_PTR,
> xe_bo_ggtt_addr(lrc->bo) + state.offset + 1);
>
>
> without re-allocating and recalculating it over and over. I will take a
> look at doing this
>
That is not a bad idea but also not huge deal with the way the code is
written as this isn't really a critical path anywhere it is called and
unlike default LRC which we need to save off for the GuC ADS / W/A
issued during driver load it is perfectly fine to regenerate on each LRC
init.
Matt
> Lucas De Marchi
>
> > return;
> >
> > xe_map_memcpy_to(gt_to_xe(state->lrc->gt), &state->lrc->bo->vmap,
> > --
> > 2.34.1
> >
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 10/34] drm/xe/guc: Document GuC submission backend
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (8 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 09/34] drm/xe: Make LRC W/A scratch buffer usage consistent Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 9:35 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper Matthew Brost
` (26 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Add kernel-doc to xe_guc_submit.c describing the submission path,
the per-queue single-threaded model with pause/resume, the driver shadow
state machine and lost-H2G replay, job timeout handling, recovery flows
(GT reset, PM resume, VF resume), and reclaim constraints.
v2:
- Mirror tweaks for clarity
- Add new doc to Xe rst files
v3:
- Clarify global vs per-queue stop / start
- Clarify VF resume flow
- Add section for 'Waiters during VF resume'
- Add section for 'Page-faulting queues during VF migration'
- Add section for 'GuC-ID assignment'
- Add section for 'Reference counting and final queue destruction'
v4:
- s/VF resume/VF post migration recovery (Tomasz)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
Documentation/gpu/xe/index.rst | 1 +
drivers/gpu/drm/xe/xe_guc_submit.c | 282 +++++++++++++++++++++++++++++
2 files changed, 283 insertions(+)
diff --git a/Documentation/gpu/xe/index.rst b/Documentation/gpu/xe/index.rst
index 88b22fad880e..692c544b164c 100644
--- a/Documentation/gpu/xe/index.rst
+++ b/Documentation/gpu/xe/index.rst
@@ -28,3 +28,4 @@ DG2, etc is provided to prototype the driver.
xe_device
xe-drm-usage-stats.rst
xe_configfs
+ xe_guc_submit
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 70306f902ba5..cd5e506527fe 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -46,6 +46,288 @@
#include "xe_trace.h"
#include "xe_vm.h"
+/*
+ * DOC: Overview
+ *
+ * The GuC submission backend is responsible for submitting GPU jobs to the GuC
+ * firmware, assigning per-queue GuC IDs, tracking submission state via a
+ * driver-side state machine, handling GuC-to-host (G2H) messages, tracking
+ * outstanding jobs, managing job timeouts and queue teardown, and providing
+ * recovery when GuC state is lost. It is built on top of the DRM scheduler
+ * (drm_sched).
+ *
+ * GuC ID assignment:
+ * ------------------
+ * Each queue is assigned a unique GuC ID at queue init. The ID is used in all
+ * H2G/G2H to identify the queue and remains reserved until final destruction,
+ * when the GuC is known to hold no references to it.
+ *
+ * The backend maintains a reverse map GuC-ID -> queue to resolve targets for
+ * G2H handlers and to iterate all queues when required (e.g., recovery). This
+ * map is protected by submission_state.lock, a global (per-GT) lock. Lockless
+ * lookups are acceptable in paths where the queue’s lifetime is otherwise
+ * pinned and it cannot disappear underneath the operation (e.g., G2H handlers).
+ *
+ * Basic submission flow
+ * ---------------------
+ * Submission is driven by the DRM scheduler vfunc ->run_job(). The flow is:
+ *
+ * 1) Emit the job's ring instructions.
+ * 2) Advance the LRC ring tail:
+ * - width == 1: simple memory write,
+ * - width > 1: append a GuC workqueue (WQ) item.
+ * 3) If the queue is unregistered, issue a register H2G for the context.
+ * 4) Trigger execution via a scheduler enable or context submit command.
+ * 5) Return the job's hardware fence to the DRM scheduler.
+ *
+ * Registration, scheduler enable, and submit commands are issued as host-to-GuC
+ * (H2G) messages over the Command Transport (CT) layer, like all GuC
+ * interactions.
+ *
+ * Completion path
+ * ---------------
+ * When the job's hardware fence signals, the DRM scheduler vfunc ->free_job()
+ * is called; it drops the job's reference, typically freeing it.
+ *
+ * Control-plane messages:
+ * -----------------------
+ * GuC submission scheduler messages form the control plane for queue cleanup,
+ * toggling runnability, and modifying queue properties (e.g., scheduler
+ * priority, timeslice, preemption timeout). Messages are initiated via queue
+ * vfuncs that append a control message to the queue. They are processed on the
+ * same single-threaded DRM scheduler workqueue that runs ->run_job() and
+ * ->free_job().
+ *
+ * Lockless model:
+ * ---------------
+ * ->run_job(), ->free_job(), and the message handlers execute as work items on
+ * a single-threaded DRM scheduler workqueue. Per queue, this provides built-in
+ * mutual exclusion: only one of these items can run at a time. As a result,
+ * these paths are lockless with respect to per-queue state tracking. (Global
+ * or cross-queue data structures still use their own synchronization.)
+ *
+ * Stopping / starting:
+ * --------------------
+ * The submission backend supports two scopes of quiesce control:
+ *
+ * - Per-queue stop/start:
+ * The single-threaded DRM scheduler workqueue for a specific queue can be
+ * stopped and started dynamically. Stopping synchronously quiesces that
+ * queue's worker (lets any in-flight item finish and prevents new items from
+ * starting), yielding a stable snapshot while an external operation (e.g.,
+ * job timeout handling) inspects/updates state and performs any required
+ * fixups. While stopped, no submission, message, or ->free_job() work runs
+ * for that queue. When the operation completes, the queue is started; any
+ * pending items are then processed in order on the same worker. Other queues
+ * continue to run unaffected.
+ *
+ * - Global (per-GT) stop/start:
+ * Implemented on top of the per-queue stop/start primitive: the driver
+ * stops (or starts) each queue on the GT to obtain a device-wide stable
+ * snapshot. This is used by coordinated recovery flows (GT reset, PM resume,
+ * VF post migration recovery). Queues created while the global stop is in
+ * effect (i.e., future queues) initialize in the stopped state and remain
+ * stopped until the global start. After recovery fixups are complete, a
+ * global start iterates queues to start all eligible ones and resumes normal
+ * submission.
+ *
+ * State machine:
+ * --------------
+ * The submission state machine is the driver's shadow of the GuC-visible queue
+ * state (e.g., registered, runnable, scheduler properties). It tracks the
+ * transitions we intend to make (issued as H2G commands), marking them pending
+ * until acknowledged via G2H or otherwise observed as applied. It also records
+ * the origin of each transition (->run_job(), timeout handler, explicit control
+ * message, etc.).
+ *
+ * Because H2G commands and/or GuC submission state can be lost across GT reset,
+ * PM resume, or VF post migration recovery, this bookkeeping lets recovery
+ * decide which operations to replay, which to elide, and which need fixups,
+ * restoring a consistent queue state without additional per-queue locks.
+ *
+ * Job timeouts:
+ * -------------
+ * To prevent jobs from running indefinitely and violating dma-fence signaling
+ * rules, the DRM scheduler tracks how long each job has been running. If a
+ * threshold is exceeded, it calls ->timeout_job().
+ *
+ * ->timeout_job() stops the queue, samples the LRC context timestamps to
+ * confirm the job actually started and has exceeded the allowed runtime, and
+ * then, if confirmed, signals all pending jobs' fences and initiates queue
+ * teardown. Finally, the queue is started.
+ *
+ * Job timeout handling runs on a per-GT, single-threaded recovery workqueue
+ * that is shared with other recovery paths (e.g., GT reset handling, VF
+ * resume). This guarantees only one recovery action executes at a time.
+ *
+ * Queue teardown:
+ * ---------------
+ * Teardown can be triggered by: (1) userspace closing the queue; (2) a G2H
+ * queue-reset notification; (3) a G2H memory_cat_error for the queue; or (4)
+ * in-flight jobs detected on the queue during GT reset.
+ *
+ * In all cases teardown is driven via the timeout path by setting the queue's
+ * DRM scheduler timeout to zero, forcing an immediate ->timeout_job() pass.
+ *
+ * Reference counting and final queue destruction:
+ * -----------------------------------------------
+ * Jobs reference-count the queue; queues hold a reference to the VM. When a
+ * queue's reference count reaches zero (e.g., all jobs are freed and the
+ * userspace handle is closed), the queue is not destroyed immediately because
+ * the GuC may still reference its state.
+ *
+ * Instead, a control-plane cleanup message is appended to remove GuC-side
+ * references (e.g., disable runnability, deregister). Once the final G2H
+ * confirming that GuC no longer references the queue is eligible for
+ * destruction.
+ *
+ * To avoid freeing the queue from within its own DRM scheduler workqueue (which
+ * would risk use-after-free), the actual destruction is deferred to a separate
+ * work item queued on a dedicated destruction workqueue.
+ *
+ * GT resets:
+ * ----------
+ * GT resets are triggered by catastrophic errors (e.g., CT channel failure).
+ * The GuC is reset and all GuC-side submission state is lost. Recovery proceeds
+ * as follows:
+ *
+ * 1) Quiesce:
+ * - Stop all queues (global submission stop). Per-queue workers finish any
+ * in-flight item and then stop; newly created queues during the window
+ * initialize in the stopped state.
+ * - Abort any waits on CT/G2H to avoid deadlock.
+ *
+ * 2) Sanitize driver shadow state:
+ * - For each queue, clear GuC-derived bits in the submission state machine
+ * (e.g., registered/enabled) and mark in-flight H2G transitions as lost.
+ * - Convert/flush any side effects of lost H2G.
+ *
+ * 3) Decide teardown vs. replay:
+ * - If a queue's LRC seqno indicates that a job started but did not
+ * complete, initiate teardown for that queue via the timeout path.
+ * - If no job started, keep the queue for replay.
+ *
+ * 4) Resume:
+ * - Start remaining queues; resubmit pending jobs.
+ * - Queues marked for teardown remain stopped/destroyed.
+ *
+ * The entire sequence runs on the per-GT single-threaded recovery worker,
+ * ensuring only one recovery action executes at a time; a runtime PM reference
+ * is held for the duration.
+ *
+ * PM resume:
+ * ----------
+ * PM resume assumes all GuC state is lost (the device may have been powered
+ * down). It reuses the GT reset recovery path, but executes in the context of
+ * the caller that wakes the device (runtime PM or system resume).
+ *
+ * Suspend entry:
+ * - Control-plane message work is quiesced; state toggles that require an
+ * active device are not enqueued while suspended.
+ * - Per-queue scheduler workers are stopped before the device is allowed to
+ * suspend.
+ * - Barring driver bugs, no queues should have in-flight jobs at
+ * suspend/resume..
+ *
+ * On resume, run the GT reset recovery flow and then start eligible queues.
+ *
+ * Runtime PM and state-change ordering:
+ * -------------------------------------
+ * Runtime/system PM transitions must not race with per-queue submission and
+ * state updates.
+ *
+ * Execution contexts and RPM sources:
+ * - Scheduler callbacks (->run_job(), ->free_job(), ->timeout_job()):
+ * executed with an active RPM ref held by the in-flight job.
+ * - Control-plane message work:
+ * enqueued from IOCTL paths that already hold an RPM ref; the message path
+ * itself does not get/put RPM. State toggles are only issued while active.
+ * During suspend entry, message work is quiesced and no new toggles are
+ * enqueued until after resume.
+ * - G2H handlers:
+ * dispatched with an RPM ref guaranteed by the CT layer.
+ * - Recovery phases (GT reset/VF post migration recovery):
+ * explicitly get/put an RPM ref for their duration on the per-GT recovery
+ * worker.
+ *
+ * Consequence:
+ * - All submission/state mutations run with an RPM reference. The PM core
+ * cannot enter suspend while these updates are in progress, and resume is
+ * complete before updates execute. This prevents PM state changes from
+ * racing with queue state changes.
+ *
+ * VF post migration recovery:
+ * ---------------------------
+ * VF post migration recovery resembles a GT reset, but GuC submission state is
+ * expected to persist across migration; in-flight H2G commands may be lost, and
+ * GGTT base/offsets may change. Recovery proceeds as follows:
+ *
+ * 1) Quiesce:
+ * - Stop all queues and abort waits (as with GT reset) to obtain a stable
+ * snapshot.
+ * - Queues created while VF post migration recovery is in-flight initialize
+ * in the stopped state.
+ *
+ * 2) Treat H2G as lost and prepare in-place resubmission (GuC/CT down):
+ * - Treat in-flight H2G (enable/disable, etc.) as dropped; update shadow
+ * bits to a safe baseline and tag the ops as "needs replay".
+ * - Quarantine device-visible submission state: set the GuC-visible LRC ring
+ * tail equal to the head (and, for WQ-based submission, set the WQ
+ * descriptor head == tail) so that when the GuC comes up it will not process
+ * any entries that were built with stale GGTT addresses.
+ * - Reset the software ring tail to the original value captured at the
+ * submission of the oldest pending job, so the write pointer sits exactly
+ * where that job was originally emitted.
+ *
+ * 3) Replay and resubmit once GuC/CT is live:
+ * - VF post migration recovery invokes ->run_job() for pending jobs;
+ * ->emit_job() overwrites ring instructions in place, fixes GGTT fields,
+ * then advances the LRC tail (and WQ descriptor for width > 1). Required
+ * submission H2G(s) are reissued and fresh WQ entries are written.
+ * - Queue lost control-plane operations (scheduling-state toggles, cleanup)
+ * in order via the message path.
+ * - Start the queues to process the queued control-plane operations and run
+ * the resubmitted jobs.
+ *
+ * The goal is to preserve both job and queue state; no teardown is performed
+ * in this flow. The sequence runs on the per-GT single-threaded recovery
+ * worker with a held runtime PM reference.
+ *
+ * Waiters during VF post migration recovery
+ * -----------------------------------------
+ * The submission backend frequently uses wait_event_timeout() to wait on
+ * GuC-driven conditions. Across VF migration/recovery two issues arise:
+ * 1) The timeout does not account for migration downtime and may expire
+ * prematurely, triggering undesired actions (e.g., GT reset, prematurely
+ * signaling a fence).
+ * 2) Some waits target GuC work that cannot complete until VF recovery
+ * finishes; these typically sit on the queue-stopping path.
+ *
+ * To handle this, all waiters must atomically test the "GuC down / VF-recovery
+ * in progress" condition (e.g., VF_RESFIX_BLOCKED) both before sleeping and
+ * after wakeup. The flag is coherent with VF migration: vCPUs observe it
+ * immediately on unhalt, and it is cleared only after the GuC/CT is live again.
+ * If set, the waiter must either (a) abort the wait without side effects, or
+ * (b) re-arm the wait with a fresh timeout once the GuC/CT is live. Timeouts
+ * that occur while GuC/CT are down are non-fatal—the VF-recovery path will
+ * rebuild state—and must not trigger recovery or teardown.
+ *
+ * Relation to reclaim:
+ * --------------------
+ * Jobs signal dma-fences, and the MM may wait on those fences during reclaim.
+ * As a consequence, the entire GuC submission backend (DRM scheduler callbacks,
+ * message handling, and all recovery paths) lies on the reclaim path and must
+ * be reclaim-safe.
+ *
+ * Practical implications:
+ * - No memory allocations in these paths (avoid any allocation that could
+ * recurse into reclaim or sleep).
+ * - The global submission-state lock may be taken from reclaim-tainted contexts
+ * (timeout/recovery). Any path that acquires it (including queue init/destroy)
+ * must not allocate or take locks that can recurse into reclaim while holding
+ * it; keep the critical section to state/xarray updates.
+ */
+
static struct xe_guc *
exec_queue_to_guc(struct xe_exec_queue *q)
{
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 10/34] drm/xe/guc: Document GuC submission backend
2025-09-24 1:15 ` [PATCH v2 10/34] drm/xe/guc: Document GuC submission backend Matthew Brost
@ 2025-09-24 9:35 ` Michal Wajdeczko
2025-09-24 20:20 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 9:35 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Add kernel-doc to xe_guc_submit.c describing the submission path,
> the per-queue single-threaded model with pause/resume, the driver shadow
> state machine and lost-H2G replay, job timeout handling, recovery flows
> (GT reset, PM resume, VF resume), and reclaim constraints.
can we split this patch into "general doc of submission path" that could be
reviewed and merged separately from the "VF recovery" doc that will stay
in this series ?
>
> v2:
> - Mirror tweaks for clarity
> - Add new doc to Xe rst files
> v3:
> - Clarify global vs per-queue stop / start
> - Clarify VF resume flow
> - Add section for 'Waiters during VF resume'
> - Add section for 'Page-faulting queues during VF migration'
> - Add section for 'GuC-ID assignment'
> - Add section for 'Reference counting and final queue destruction'
> v4:
> - s/VF resume/VF post migration recovery (Tomasz)
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> Documentation/gpu/xe/index.rst | 1 +
> drivers/gpu/drm/xe/xe_guc_submit.c | 282 +++++++++++++++++++++++++++++
^ permalink raw reply [flat|nested] 90+ messages in thread
* Re: [PATCH v2 10/34] drm/xe/guc: Document GuC submission backend
2025-09-24 9:35 ` Michal Wajdeczko
@ 2025-09-24 20:20 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:20 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 11:35:23AM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Add kernel-doc to xe_guc_submit.c describing the submission path,
> > the per-queue single-threaded model with pause/resume, the driver shadow
> > state machine and lost-H2G replay, job timeout handling, recovery flows
> > (GT reset, PM resume, VF resume), and reclaim constraints.
>
> can we split this patch into "general doc of submission path" that could be
> reviewed and merged separately from the "VF recovery" doc that will stay
> in this series ?
>
I don't really see the why here? Ofc merging this patch as part of this
series is a blocker but included it as largely explains everything this
series does + overall design of GuC submission which was entirely in my
head prior to this patch.
Matt
> >
> > v2:
> > - Mirror tweaks for clarity
> > - Add new doc to Xe rst files
> > v3:
> > - Clarify global vs per-queue stop / start
> > - Clarify VF resume flow
> > - Add section for 'Waiters during VF resume'
> > - Add section for 'Page-faulting queues during VF migration'
> > - Add section for 'GuC-ID assignment'
> > - Add section for 'Reference counting and final queue destruction'
> > v4:
> > - s/VF resume/VF post migration recovery (Tomasz)
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > Documentation/gpu/xe/index.rst | 1 +
> > drivers/gpu/drm/xe/xe_guc_submit.c | 282 +++++++++++++++++++++++++++++
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (9 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 10/34] drm/xe/guc: Document GuC submission backend Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 10:14 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker Matthew Brost
` (25 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Add xe_gt_sriov_vf_recovery_inprogress helper.
This helper serves as the singular point to determine whether a VF
post-migration recovery is currently in progress. Expected callers
include the GuC CT layer and the GuC submission layer. Atomically
visable as soon as vCPU are unhalted until VF recovery completes.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 ++++++++
drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 10 +++++
drivers/gpu/drm/xe/xe_memirq.c | 48 ++++++++++++++++++++++-
drivers/gpu/drm/xe/xe_memirq.h | 3 ++
5 files changed, 79 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 016c867e5e2b..c9d0e32e7a15 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -26,6 +26,7 @@
#include "xe_guc_hxg_helpers.h"
#include "xe_guc_relay.h"
#include "xe_lrc.h"
+#include "xe_memirq.h"
#include "xe_mmio.h"
#include "xe_sriov.h"
#include "xe_sriov_vf.h"
@@ -828,6 +829,7 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
struct xe_device *xe = gt_to_xe(gt);
xe_gt_assert(gt, IS_SRIOV_VF(xe));
+ xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
/*
@@ -1172,3 +1174,18 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
drm_printf(p, "\thandshake:\t%u.%u\n",
pf_version->major, pf_version->minor);
}
+
+/**
+ * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
+ * @gt: the &xe_gt
+ *
+ * Return: True if VF post migration recovery in progress, False otherwise
+ */
+bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt)
+{
+ struct xe_memirq *memirq = >_to_tile(gt)->memirq;
+
+ return IS_SRIOV_VF(gt_to_xe(gt)) &&
+ (xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
+ READ_ONCE(gt->sriov.vf.migration.recovery_inprogress));
+}
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index 0af1dc769fe0..bb5f8eace19b 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -25,6 +25,8 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
+bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
+
u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index d95857bd789b..7b10b8e1e10e 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -49,6 +49,14 @@ struct xe_gt_sriov_vf_runtime {
} *regs;
};
+/**
+ * xe_gt_sriov_vf_migration - VF migration data.
+ */
+struct xe_gt_sriov_vf_migration {
+ /** @recovery_inprogress: VF post migration recovery in progress */
+ bool recovery_inprogress;
+};
+
/**
* struct xe_gt_sriov_vf - GT level VF virtualization data.
*/
@@ -61,6 +69,8 @@ struct xe_gt_sriov_vf {
struct xe_gt_sriov_vf_selfconfig self_config;
/** @runtime: runtime data retrieved from the PF. */
struct xe_gt_sriov_vf_runtime runtime;
+ /** @migration: migration data for the VF. */
+ struct xe_gt_sriov_vf_migration migration;
};
#endif
diff --git a/drivers/gpu/drm/xe/xe_memirq.c b/drivers/gpu/drm/xe/xe_memirq.c
index 49c45ec3e83c..94d5d6859aab 100644
--- a/drivers/gpu/drm/xe/xe_memirq.c
+++ b/drivers/gpu/drm/xe/xe_memirq.c
@@ -398,6 +398,23 @@ void xe_memirq_postinstall(struct xe_memirq *memirq)
memirq_set_enable(memirq, true);
}
+static bool memirq_received_noclear(struct xe_memirq *memirq,
+ struct iosys_map *vector,
+ u16 offset, const char *name)
+{
+ u8 value;
+
+ value = iosys_map_rd(vector, offset, u8);
+ if (value) {
+ if (value != 0xff)
+ memirq_err_ratelimited(memirq,
+ "Unexpected memirq value %#x from %s at %u\n",
+ value, name, offset);
+ }
+
+ return value;
+}
+
static bool memirq_received(struct xe_memirq *memirq, struct iosys_map *vector,
u16 offset, const char *name)
{
@@ -434,8 +451,16 @@ static void memirq_dispatch_guc(struct xe_memirq *memirq, struct iosys_map *stat
if (memirq_received(memirq, status, ilog2(GUC_INTR_GUC2HOST), name))
xe_guc_irq_handler(guc, GUC_INTR_GUC2HOST);
- if (memirq_received(memirq, status, ilog2(GUC_INTR_SW_INT_0), name))
+ /*
+ * We must wait to perform the clear operation until after
+ * xe_gt_sriov_vf_start_migration_recovery() runs, to avoid race
+ * conditions where xe_gt_sriov_vf_recovery_inprogress() returns false.
+ */
+ if (memirq_received_noclear(memirq, status, ilog2(GUC_INTR_SW_INT_0),
+ name)) {
xe_guc_irq_handler(guc, GUC_INTR_SW_INT_0);
+ iosys_map_wr(status, ilog2(GUC_INTR_SW_INT_0), u8, 0x00);
+ }
}
/**
@@ -460,6 +485,27 @@ void xe_memirq_hwe_handler(struct xe_memirq *memirq, struct xe_hw_engine *hwe)
}
}
+/**
+ * xe_memirq_vf_recovery_irq_pending() - VF recovery IRQ is pending
+ * @memirq: the &xe_memirq
+ * @guc: the &xe_guc to check for IRQ
+ *
+ * Return: True if VF recovery IRQ is pending on @guc, False otherwise
+ */
+bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
+ struct xe_guc *guc)
+{
+ struct xe_gt *gt = guc_to_gt(guc);
+ struct iosys_map map;
+
+ if (xe_gt_is_media_type(gt))
+ map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_MGUC) * SZ_16);
+ else
+ map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_GUC) * SZ_16);
+
+ return iosys_map_rd(&map, ilog2(GUC_INTR_SW_INT_0), u8);
+}
+
/**
* xe_memirq_handler - The `Memory Based Interrupts`_ Handler.
* @memirq: the &xe_memirq
diff --git a/drivers/gpu/drm/xe/xe_memirq.h b/drivers/gpu/drm/xe/xe_memirq.h
index 06130650e9d6..476b8cba179d 100644
--- a/drivers/gpu/drm/xe/xe_memirq.h
+++ b/drivers/gpu/drm/xe/xe_memirq.h
@@ -25,4 +25,7 @@ void xe_memirq_handler(struct xe_memirq *memirq);
int xe_memirq_init_guc(struct xe_memirq *memirq, struct xe_guc *guc);
+bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
+ struct xe_guc *guc);
+
#endif
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper
2025-09-24 1:15 ` [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper Matthew Brost
@ 2025-09-24 10:14 ` Michal Wajdeczko
2025-09-24 19:39 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 10:14 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Add xe_gt_sriov_vf_recovery_inprogress helper.
>
> This helper serves as the singular point to determine whether a VF
hmm, this "singular" looks like a GT-level only, not global
> post-migration recovery is currently in progress. Expected callers
> include the GuC CT layer and the GuC submission layer. Atomically
> visable as soon as vCPU are unhalted until VF recovery completes.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 ++++++++
> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +
> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 10 +++++
> drivers/gpu/drm/xe/xe_memirq.c | 48 ++++++++++++++++++++++-
> drivers/gpu/drm/xe/xe_memirq.h | 3 ++
> 5 files changed, 79 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 016c867e5e2b..c9d0e32e7a15 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -26,6 +26,7 @@
> #include "xe_guc_hxg_helpers.h"
> #include "xe_guc_relay.h"
> #include "xe_lrc.h"
> +#include "xe_memirq.h"
> #include "xe_mmio.h"
> #include "xe_sriov.h"
> #include "xe_sriov_vf.h"
> @@ -828,6 +829,7 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> struct xe_device *xe = gt_to_xe(gt);
>
> xe_gt_assert(gt, IS_SRIOV_VF(xe));
> + xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
>
> set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> /*
> @@ -1172,3 +1174,18 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> drm_printf(p, "\thandshake:\t%u.%u\n",
> pf_version->major, pf_version->minor);
> }
> +
> +/**
> + * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> + * @gt: the &xe_gt
> + *
> + * Return: True if VF post migration recovery in progress, False otherwise
> + */
> +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt)
> +{
> + struct xe_memirq *memirq = >_to_tile(gt)->memirq;
> +
> + return IS_SRIOV_VF(gt_to_xe(gt)) &&
this is xe_gt_sriov_vf function, so it is expected to be called only by
the VF code, thus we should rather use xe_gt_assert here and the caller
is responsible for the IS_SRIOV_VF check
> + (xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
> + READ_ONCE(gt->sriov.vf.migration.recovery_inprogress));
> +}
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> index 0af1dc769fe0..bb5f8eace19b 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> @@ -25,6 +25,8 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
>
> +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> +
> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
> u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> index d95857bd789b..7b10b8e1e10e 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> @@ -49,6 +49,14 @@ struct xe_gt_sriov_vf_runtime {
> } *regs;
> };
>
> +/**
> + * xe_gt_sriov_vf_migration - VF migration data.
> + */
> +struct xe_gt_sriov_vf_migration {
> + /** @recovery_inprogress: VF post migration recovery in progress */
> + bool recovery_inprogress;
> +};
> +
> /**
> * struct xe_gt_sriov_vf - GT level VF virtualization data.
> */
> @@ -61,6 +69,8 @@ struct xe_gt_sriov_vf {
> struct xe_gt_sriov_vf_selfconfig self_config;
> /** @runtime: runtime data retrieved from the PF. */
> struct xe_gt_sriov_vf_runtime runtime;
> + /** @migration: migration data for the VF. */
> + struct xe_gt_sriov_vf_migration migration;
> };
>
> #endif
> diff --git a/drivers/gpu/drm/xe/xe_memirq.c b/drivers/gpu/drm/xe/xe_memirq.c
> index 49c45ec3e83c..94d5d6859aab 100644
> --- a/drivers/gpu/drm/xe/xe_memirq.c
> +++ b/drivers/gpu/drm/xe/xe_memirq.c
> @@ -398,6 +398,23 @@ void xe_memirq_postinstall(struct xe_memirq *memirq)
> memirq_set_enable(memirq, true);
> }
>
> +static bool memirq_received_noclear(struct xe_memirq *memirq,
> + struct iosys_map *vector,
> + u16 offset, const char *name)
maybe instead of duplicating code of memirq_received() in 90% just add there
the "bool clear" flag?
> +{
> + u8 value;
> +
> + value = iosys_map_rd(vector, offset, u8);
> + if (value) {
> + if (value != 0xff)
> + memirq_err_ratelimited(memirq,
> + "Unexpected memirq value %#x from %s at %u\n",
> + value, name, offset);
> + }
> +
> + return value;
> +}
> +
> static bool memirq_received(struct xe_memirq *memirq, struct iosys_map *vector,
> u16 offset, const char *name)
> {
> @@ -434,8 +451,16 @@ static void memirq_dispatch_guc(struct xe_memirq *memirq, struct iosys_map *stat
> if (memirq_received(memirq, status, ilog2(GUC_INTR_GUC2HOST), name))
> xe_guc_irq_handler(guc, GUC_INTR_GUC2HOST);
>
> - if (memirq_received(memirq, status, ilog2(GUC_INTR_SW_INT_0), name))
> + /*
> + * We must wait to perform the clear operation until after
> + * xe_gt_sriov_vf_start_migration_recovery() runs, to avoid race
> + * conditions where xe_gt_sriov_vf_recovery_inprogress() returns false.
but the VF recovery "inprogress" shall be already set in the top level
xe_sriov_vf_start_migration_recovery()
even before the GT-level recovery starts, where is this race ?
> + */
> + if (memirq_received_noclear(memirq, status, ilog2(GUC_INTR_SW_INT_0),
> + name)) {
> xe_guc_irq_handler(guc, GUC_INTR_SW_INT_0);
what if new MEMIRQ will arrive just here
is it ok that we will clear it immediately?
the whole memirq flow is that we clear irq byte first and then process
it, so if anything comes right after we finish processing will be noticed
on next iteration
I assume any races due to double migration shall be handled on the VF2GUC
communication level while sending RESFIX_START/DONE, not here
> + iosys_map_wr(status, ilog2(GUC_INTR_SW_INT_0), u8, 0x00);
> + }
> }
>
> /**
> @@ -460,6 +485,27 @@ void xe_memirq_hwe_handler(struct xe_memirq *memirq, struct xe_hw_engine *hwe)
> }
> }
>
> +/**
> + * xe_memirq_vf_recovery_irq_pending() - VF recovery IRQ is pending
this function isn't really using anything VF specific except that on the
VF the SW_INT_0 means "migrated"
maybe we can drop the _vf from function name?
xe_memirq_pending_guc(memirq, guc, bit)
> + * @memirq: the &xe_memirq
> + * @guc: the &xe_guc to check for IRQ
> + *
> + * Return: True if VF recovery IRQ is pending on @guc, False otherwise
> + */
> +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
> + struct xe_guc *guc)
> +{
> + struct xe_gt *gt = guc_to_gt(guc);
> + struct iosys_map map;
> +
> + if (xe_gt_is_media_type(gt))
> + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_MGUC) * SZ_16);
> + else
> + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_GUC) * SZ_16);
nit: maybe just calc offset conditionally?
u32 offset = is_media ? ilog2(INTR_MGUC) : ilog2(INTR_GUC);
> +
> + return iosys_map_rd(&map, ilog2(GUC_INTR_SW_INT_0), u8);
as we have helpers we should use them
return memirq_received_noclear(...)
> +}
> +
> /**
> * xe_memirq_handler - The `Memory Based Interrupts`_ Handler.
> * @memirq: the &xe_memirq
> diff --git a/drivers/gpu/drm/xe/xe_memirq.h b/drivers/gpu/drm/xe/xe_memirq.h
> index 06130650e9d6..476b8cba179d 100644
> --- a/drivers/gpu/drm/xe/xe_memirq.h
> +++ b/drivers/gpu/drm/xe/xe_memirq.h
> @@ -25,4 +25,7 @@ void xe_memirq_handler(struct xe_memirq *memirq);
>
> int xe_memirq_init_guc(struct xe_memirq *memirq, struct xe_guc *guc);
>
> +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
> + struct xe_guc *guc);
> +
> #endif
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper
2025-09-24 10:14 ` Michal Wajdeczko
@ 2025-09-24 19:39 ` Matthew Brost
2025-09-24 20:12 ` Michal Wajdeczko
0 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 19:39 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 12:14:28PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Add xe_gt_sriov_vf_recovery_inprogress helper.
> >
> > This helper serves as the singular point to determine whether a VF
>
> hmm, this "singular" looks like a GT-level only, not global
>
Yes, it is GT scoped. I will adjust the commit message.
> > post-migration recovery is currently in progress. Expected callers
> > include the GuC CT layer and the GuC submission layer. Atomically
> > visable as soon as vCPU are unhalted until VF recovery completes.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 ++++++++
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +
> > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 10 +++++
> > drivers/gpu/drm/xe/xe_memirq.c | 48 ++++++++++++++++++++++-
> > drivers/gpu/drm/xe/xe_memirq.h | 3 ++
> > 5 files changed, 79 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 016c867e5e2b..c9d0e32e7a15 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -26,6 +26,7 @@
> > #include "xe_guc_hxg_helpers.h"
> > #include "xe_guc_relay.h"
> > #include "xe_lrc.h"
> > +#include "xe_memirq.h"
> > #include "xe_mmio.h"
> > #include "xe_sriov.h"
> > #include "xe_sriov_vf.h"
> > @@ -828,6 +829,7 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> > struct xe_device *xe = gt_to_xe(gt);
> >
> > xe_gt_assert(gt, IS_SRIOV_VF(xe));
> > + xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
> >
> > set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> > /*
> > @@ -1172,3 +1174,18 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> > drm_printf(p, "\thandshake:\t%u.%u\n",
> > pf_version->major, pf_version->minor);
> > }
> > +
> > +/**
> > + * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> > + * @gt: the &xe_gt
> > + *
> > + * Return: True if VF post migration recovery in progress, False otherwise
> > + */
> > +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt)
> > +{
> > + struct xe_memirq *memirq = >_to_tile(gt)->memirq;
> > +
> > + return IS_SRIOV_VF(gt_to_xe(gt)) &&
>
> this is xe_gt_sriov_vf function, so it is expected to be called only by
> the VF code, thus we should rather use xe_gt_assert here and the caller
> is responsible for the IS_SRIOV_VF check
>
That is not how I have coded this. I blindly call this in various places
and I don't think it could at call site to determine if it is a VF as we
have if (VF) statements all over the driver. If perf is the concern, we
could move the IS_SRIOV_VF(gt_to_xe(gt)) part of function to static
inline and reset of the function in an exported function.
> > + (xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
> > + READ_ONCE(gt->sriov.vf.migration.recovery_inprogress));
> > +}
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > index 0af1dc769fe0..bb5f8eace19b 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > @@ -25,6 +25,8 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> > int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> > void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> >
> > +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> > +
> > u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> > u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
> > u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > index d95857bd789b..7b10b8e1e10e 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > @@ -49,6 +49,14 @@ struct xe_gt_sriov_vf_runtime {
> > } *regs;
> > };
> >
> > +/**
> > + * xe_gt_sriov_vf_migration - VF migration data.
> > + */
> > +struct xe_gt_sriov_vf_migration {
> > + /** @recovery_inprogress: VF post migration recovery in progress */
> > + bool recovery_inprogress;
> > +};
> > +
> > /**
> > * struct xe_gt_sriov_vf - GT level VF virtualization data.
> > */
> > @@ -61,6 +69,8 @@ struct xe_gt_sriov_vf {
> > struct xe_gt_sriov_vf_selfconfig self_config;
> > /** @runtime: runtime data retrieved from the PF. */
> > struct xe_gt_sriov_vf_runtime runtime;
> > + /** @migration: migration data for the VF. */
> > + struct xe_gt_sriov_vf_migration migration;
> > };
> >
> > #endif
> > diff --git a/drivers/gpu/drm/xe/xe_memirq.c b/drivers/gpu/drm/xe/xe_memirq.c
> > index 49c45ec3e83c..94d5d6859aab 100644
> > --- a/drivers/gpu/drm/xe/xe_memirq.c
> > +++ b/drivers/gpu/drm/xe/xe_memirq.c
> > @@ -398,6 +398,23 @@ void xe_memirq_postinstall(struct xe_memirq *memirq)
> > memirq_set_enable(memirq, true);
> > }
> >
> > +static bool memirq_received_noclear(struct xe_memirq *memirq,
> > + struct iosys_map *vector,
> > + u16 offset, const char *name)
>
> maybe instead of duplicating code of memirq_received() in 90% just add there
> the "bool clear" flag?
>
Sure.
> > +{
> > + u8 value;
> > +
> > + value = iosys_map_rd(vector, offset, u8);
> > + if (value) {
> > + if (value != 0xff)
> > + memirq_err_ratelimited(memirq,
> > + "Unexpected memirq value %#x from %s at %u\n",
> > + value, name, offset);
> > + }
> > +
> > + return value;
> > +}
> > +
> > static bool memirq_received(struct xe_memirq *memirq, struct iosys_map *vector,
> > u16 offset, const char *name)
> > {
> > @@ -434,8 +451,16 @@ static void memirq_dispatch_guc(struct xe_memirq *memirq, struct iosys_map *stat
> > if (memirq_received(memirq, status, ilog2(GUC_INTR_GUC2HOST), name))
> > xe_guc_irq_handler(guc, GUC_INTR_GUC2HOST);
> >
> > - if (memirq_received(memirq, status, ilog2(GUC_INTR_SW_INT_0), name))
> > + /*
> > + * We must wait to perform the clear operation until after
> > + * xe_gt_sriov_vf_start_migration_recovery() runs, to avoid race
> > + * conditions where xe_gt_sriov_vf_recovery_inprogress() returns false.
>
> but the VF recovery "inprogress" shall be already set in the top level
>
> xe_sriov_vf_start_migration_recovery()
>
> even before the GT-level recovery starts, where is this race ?
>
If we clear the interrupt it here, this is before the IRQ handler is
called which flips the software bit for "inprogress". There would be
window where xe_gt_sriov_vf_recovery_inprogress could return false which
is problematic.
> > + */
> > + if (memirq_received_noclear(memirq, status, ilog2(GUC_INTR_SW_INT_0),
> > + name)) {
> > xe_guc_irq_handler(guc, GUC_INTR_SW_INT_0);
>
> what if new MEMIRQ will arrive just here
>
> is it ok that we will clear it immediately?
>
> the whole memirq flow is that we clear irq byte first and then process
> it, so if anything comes right after we finish processing will be noticed
> on next iteration
>
I don't see why that matters in this case. In either case multiple IRQs
could happen and single IRQ handler runs.
> I assume any races due to double migration shall be handled on the VF2GUC
> communication level while sending RESFIX_START/DONE, not here
I don't think it should be possible to get multiple RESFIX_START IRQs
before DONE is complete. This existing code upstream seems to handle
these cases, so my series attempts to handle this too, but it seems like
something that shouldn't be possible.
> > + iosys_map_wr(status, ilog2(GUC_INTR_SW_INT_0), u8, 0x00);
> > + }
> > }
> >
> > /**
> > @@ -460,6 +485,27 @@ void xe_memirq_hwe_handler(struct xe_memirq *memirq, struct xe_hw_engine *hwe)
> > }
> > }
> >
> > +/**
> > + * xe_memirq_vf_recovery_irq_pending() - VF recovery IRQ is pending
>
> this function isn't really using anything VF specific except that on the
> VF the SW_INT_0 means "migrated"
>
> maybe we can drop the _vf from function name?
>
Sure, so 'xe_memirq_recovery_irq_pending'?
Or 'xe_memirq_sw_int0_irq_pending'?
> xe_memirq_pending_guc(memirq, guc, bit)
> > + * @memirq: the &xe_memirq
> > + * @guc: the &xe_guc to check for IRQ
> > + *
> > + * Return: True if VF recovery IRQ is pending on @guc, False otherwise
> > + */
> > +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
> > + struct xe_guc *guc)
> > +{
> > + struct xe_gt *gt = guc_to_gt(guc);
> > + struct iosys_map map;
> > +
> > + if (xe_gt_is_media_type(gt))
> > + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_MGUC) * SZ_16);
> > + else
> > + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_GUC) * SZ_16);
>
> nit: maybe just calc offset conditionally?
>
> u32 offset = is_media ? ilog2(INTR_MGUC) : ilog2(INTR_GUC);
>
Sure.
> > +
> > + return iosys_map_rd(&map, ilog2(GUC_INTR_SW_INT_0), u8);
>
> as we have helpers we should use them
>
> return memirq_received_noclear(...)
Sure.
Matt
>
> > +}
> > +
> > /**
> > * xe_memirq_handler - The `Memory Based Interrupts`_ Handler.
> > * @memirq: the &xe_memirq
> > diff --git a/drivers/gpu/drm/xe/xe_memirq.h b/drivers/gpu/drm/xe/xe_memirq.h
> > index 06130650e9d6..476b8cba179d 100644
> > --- a/drivers/gpu/drm/xe/xe_memirq.h
> > +++ b/drivers/gpu/drm/xe/xe_memirq.h
> > @@ -25,4 +25,7 @@ void xe_memirq_handler(struct xe_memirq *memirq);
> >
> > int xe_memirq_init_guc(struct xe_memirq *memirq, struct xe_guc *guc);
> >
> > +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
> > + struct xe_guc *guc);
> > +
> > #endif
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper
2025-09-24 19:39 ` Matthew Brost
@ 2025-09-24 20:12 ` Michal Wajdeczko
2025-09-24 20:30 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 20:12 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
On 9/24/2025 9:39 PM, Matthew Brost wrote:
> On Wed, Sep 24, 2025 at 12:14:28PM +0200, Michal Wajdeczko wrote:
>>
>>
>> On 9/24/2025 3:15 AM, Matthew Brost wrote:
>>> Add xe_gt_sriov_vf_recovery_inprogress helper.
>>>
>>> This helper serves as the singular point to determine whether a VF
>>
>> hmm, this "singular" looks like a GT-level only, not global
>>
>
> Yes, it is GT scoped. I will adjust the commit message.
>
>>> post-migration recovery is currently in progress. Expected callers
>>> include the GuC CT layer and the GuC submission layer. Atomically
>>> visable as soon as vCPU are unhalted until VF recovery completes.
>>>
>>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>>> ---
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 ++++++++
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 10 +++++
>>> drivers/gpu/drm/xe/xe_memirq.c | 48 ++++++++++++++++++++++-
>>> drivers/gpu/drm/xe/xe_memirq.h | 3 ++
>>> 5 files changed, 79 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> index 016c867e5e2b..c9d0e32e7a15 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> @@ -26,6 +26,7 @@
>>> #include "xe_guc_hxg_helpers.h"
>>> #include "xe_guc_relay.h"
>>> #include "xe_lrc.h"
>>> +#include "xe_memirq.h"
>>> #include "xe_mmio.h"
>>> #include "xe_sriov.h"
>>> #include "xe_sriov_vf.h"
>>> @@ -828,6 +829,7 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
>>> struct xe_device *xe = gt_to_xe(gt);
>>>
>>> xe_gt_assert(gt, IS_SRIOV_VF(xe));
>>> + xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
>>>
>>> set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
>>> /*
>>> @@ -1172,3 +1174,18 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
>>> drm_printf(p, "\thandshake:\t%u.%u\n",
>>> pf_version->major, pf_version->minor);
>>> }
>>> +
>>> +/**
>>> + * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
>>> + * @gt: the &xe_gt
>>> + *
>>> + * Return: True if VF post migration recovery in progress, False otherwise
>>> + */
>>> +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt)
>>> +{
>>> + struct xe_memirq *memirq = >_to_tile(gt)->memirq;
>>> +
>>> + return IS_SRIOV_VF(gt_to_xe(gt)) &&
>>
>> this is xe_gt_sriov_vf function, so it is expected to be called only by
>> the VF code, thus we should rather use xe_gt_assert here and the caller
>> is responsible for the IS_SRIOV_VF check
>>
>
> That is not how I have coded this. I blindly call this in various places
> and I don't think it could at call site to determine if it is a VF as we
> have if (VF) statements all over the driver. If perf is the concern, we
> could move the IS_SRIOV_VF(gt_to_xe(gt)) part of function to static
> inline and reset of the function in an exported function.
inline will work, preferable not in xe_gt_sriov_vf.h ;)
static inline xe_gt_recovery_inprogress(gt)
{
return IS_SRIOV_VF(xe) && xe_gt_sriov_vf_recovery_inprogress(gt);
}
>
>>> + (xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
>>> + READ_ONCE(gt->sriov.vf.migration.recovery_inprogress));
>>> +}
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>> index 0af1dc769fe0..bb5f8eace19b 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>> @@ -25,6 +25,8 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
>>> int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
>>> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
>>>
>>> +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
>>> +
>>> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
>>> u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
>>> u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> index d95857bd789b..7b10b8e1e10e 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> @@ -49,6 +49,14 @@ struct xe_gt_sriov_vf_runtime {
>>> } *regs;
>>> };
>>>
>>> +/**
>>> + * xe_gt_sriov_vf_migration - VF migration data.
>>> + */
>>> +struct xe_gt_sriov_vf_migration {
>>> + /** @recovery_inprogress: VF post migration recovery in progress */
>>> + bool recovery_inprogress;
>>> +};
>>> +
>>> /**
>>> * struct xe_gt_sriov_vf - GT level VF virtualization data.
>>> */
>>> @@ -61,6 +69,8 @@ struct xe_gt_sriov_vf {
>>> struct xe_gt_sriov_vf_selfconfig self_config;
>>> /** @runtime: runtime data retrieved from the PF. */
>>> struct xe_gt_sriov_vf_runtime runtime;
>>> + /** @migration: migration data for the VF. */
>>> + struct xe_gt_sriov_vf_migration migration;
>>> };
>>>
>>> #endif
>>> diff --git a/drivers/gpu/drm/xe/xe_memirq.c b/drivers/gpu/drm/xe/xe_memirq.c
>>> index 49c45ec3e83c..94d5d6859aab 100644
>>> --- a/drivers/gpu/drm/xe/xe_memirq.c
>>> +++ b/drivers/gpu/drm/xe/xe_memirq.c
>>> @@ -398,6 +398,23 @@ void xe_memirq_postinstall(struct xe_memirq *memirq)
>>> memirq_set_enable(memirq, true);
>>> }
>>>
>>> +static bool memirq_received_noclear(struct xe_memirq *memirq,
>>> + struct iosys_map *vector,
>>> + u16 offset, const char *name)
>>
>> maybe instead of duplicating code of memirq_received() in 90% just add there
>> the "bool clear" flag?
>>
>
> Sure.
>
>>> +{
>>> + u8 value;
>>> +
>>> + value = iosys_map_rd(vector, offset, u8);
>>> + if (value) {
>>> + if (value != 0xff)
>>> + memirq_err_ratelimited(memirq,
>>> + "Unexpected memirq value %#x from %s at %u\n",
>>> + value, name, offset);
>>> + }
>>> +
>>> + return value;
>>> +}
>>> +
>>> static bool memirq_received(struct xe_memirq *memirq, struct iosys_map *vector,
>>> u16 offset, const char *name)
>>> {
>>> @@ -434,8 +451,16 @@ static void memirq_dispatch_guc(struct xe_memirq *memirq, struct iosys_map *stat
>>> if (memirq_received(memirq, status, ilog2(GUC_INTR_GUC2HOST), name))
>>> xe_guc_irq_handler(guc, GUC_INTR_GUC2HOST);
>>>
>>> - if (memirq_received(memirq, status, ilog2(GUC_INTR_SW_INT_0), name))
>>> + /*
>>> + * We must wait to perform the clear operation until after
>>> + * xe_gt_sriov_vf_start_migration_recovery() runs, to avoid race
>>> + * conditions where xe_gt_sriov_vf_recovery_inprogress() returns false.
>>
>> but the VF recovery "inprogress" shall be already set in the top level
>>
>> xe_sriov_vf_start_migration_recovery()
>>
>> even before the GT-level recovery starts, where is this race ?
>>
>
> If we clear the interrupt it here, this is before the IRQ handler is
> called which flips the software bit for "inprogress". There would be
> window where xe_gt_sriov_vf_recovery_inprogress could return false which
> is problematic.
but who calls that recovery_inprogress() and why should it be problematic?
note there might be still other threads that will finish some actions
(including sending CTB) that will just be stuck there until we start
the recovery, so even if recovery_inprogress() returns false for the
small window that shouldn't change the picture
>
>>> + */
>>> + if (memirq_received_noclear(memirq, status, ilog2(GUC_INTR_SW_INT_0),
>>> + name)) {
>>> xe_guc_irq_handler(guc, GUC_INTR_SW_INT_0);
>>
>> what if new MEMIRQ will arrive just here
>>
>> is it ok that we will clear it immediately?
>>
>> the whole memirq flow is that we clear irq byte first and then process
>> it, so if anything comes right after we finish processing will be noticed
>> on next iteration
>>
>
> I don't see why that matters in this cabuse. In either case multiple IRQs
> could happen and single IRQ handler runs.
>
>> I assume any races due to double migration shall be handled on the VF2GUC
>> communication level while sending RESFIX_START/DONE, not here
>
> I don't think it should be possible to get multiple RESFIX_START IRQs
I assume the SW_INT_0 is set on every migration, even in case we didn't
have a chance to read it and clear it, or we didn't start (by sending
RESFIX_START) nor finish (RESFIX_DONE)
> before DONE is complete. This existing code upstream seems to handle
> these cases, so my series attempts to handle this too, but it seems like
> something that shouldn't be possible.
>
>>> + iosys_map_wr(status, ilog2(GUC_INTR_SW_INT_0), u8, 0x00);
>>> + }
>>> }
>>>
>>> /**
>>> @@ -460,6 +485,27 @@ void xe_memirq_hwe_handler(struct xe_memirq *memirq, struct xe_hw_engine *hwe)
>>> }
>>> }
>>>
>>> +/**
>>> + * xe_memirq_vf_recovery_irq_pending() - VF recovery IRQ is pending
>>
>> this function isn't really using anything VF specific except that on the
>> VF the SW_INT_0 means "migrated"
>>
>> maybe we can drop the _vf from function name?
>>
>
> Sure, so 'xe_memirq_recovery_irq_pending'?
>
> Or 'xe_memirq_sw_int0_irq_pending'?
>
>> xe_memirq_pending_guc(memirq, guc, bit)
I would make it more generic and provide bit param
note that xe_guc_irq_handler() also takes a bit
>>> + * @memirq: the &xe_memirq
>>> + * @guc: the &xe_guc to check for IRQ
>>> + *
>>> + * Return: True if VF recovery IRQ is pending on @guc, False otherwise
>>> + */
>>> +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
>>> + struct xe_guc *guc)
>>> +{
>>> + struct xe_gt *gt = guc_to_gt(guc);
>>> + struct iosys_map map;
>>> +
>>> + if (xe_gt_is_media_type(gt))
>>> + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_MGUC) * SZ_16);
>>> + else
>>> + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_GUC) * SZ_16);
>>
>> nit: maybe just calc offset conditionally?
>>
>> u32 offset = is_media ? ilog2(INTR_MGUC) : ilog2(INTR_GUC);
>>
>
> Sure.
>
>>> +
>>> + return iosys_map_rd(&map, ilog2(GUC_INTR_SW_INT_0), u8);
>>
>> as we have helpers we should use them
>>
>> return memirq_received_noclear(...)
>
> Sure.
>
> Matt
>
>>
>>> +}
>>> +
>>> /**
>>> * xe_memirq_handler - The `Memory Based Interrupts`_ Handler.
>>> * @memirq: the &xe_memirq
>>> diff --git a/drivers/gpu/drm/xe/xe_memirq.h b/drivers/gpu/drm/xe/xe_memirq.h
>>> index 06130650e9d6..476b8cba179d 100644
>>> --- a/drivers/gpu/drm/xe/xe_memirq.h
>>> +++ b/drivers/gpu/drm/xe/xe_memirq.h
>>> @@ -25,4 +25,7 @@ void xe_memirq_handler(struct xe_memirq *memirq);
>>>
>>> int xe_memirq_init_guc(struct xe_memirq *memirq, struct xe_guc *guc);
>>>
>>> +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
>>> + struct xe_guc *guc);
>>> +
>>> #endif
>>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper
2025-09-24 20:12 ` Michal Wajdeczko
@ 2025-09-24 20:30 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:30 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 10:12:10PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 9:39 PM, Matthew Brost wrote:
> > On Wed, Sep 24, 2025 at 12:14:28PM +0200, Michal Wajdeczko wrote:
> >>
> >>
> >> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> >>> Add xe_gt_sriov_vf_recovery_inprogress helper.
> >>>
> >>> This helper serves as the singular point to determine whether a VF
> >>
> >> hmm, this "singular" looks like a GT-level only, not global
> >>
> >
> > Yes, it is GT scoped. I will adjust the commit message.
> >
> >>> post-migration recovery is currently in progress. Expected callers
> >>> include the GuC CT layer and the GuC submission layer. Atomically
> >>> visable as soon as vCPU are unhalted until VF recovery completes.
> >>>
> >>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> >>> ---
> >>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 ++++++++
> >>> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +
> >>> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 10 +++++
> >>> drivers/gpu/drm/xe/xe_memirq.c | 48 ++++++++++++++++++++++-
> >>> drivers/gpu/drm/xe/xe_memirq.h | 3 ++
> >>> 5 files changed, 79 insertions(+), 1 deletion(-)
> >>>
> >>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> >>> index 016c867e5e2b..c9d0e32e7a15 100644
> >>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> >>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> >>> @@ -26,6 +26,7 @@
> >>> #include "xe_guc_hxg_helpers.h"
> >>> #include "xe_guc_relay.h"
> >>> #include "xe_lrc.h"
> >>> +#include "xe_memirq.h"
> >>> #include "xe_mmio.h"
> >>> #include "xe_sriov.h"
> >>> #include "xe_sriov_vf.h"
> >>> @@ -828,6 +829,7 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> >>> struct xe_device *xe = gt_to_xe(gt);
> >>>
> >>> xe_gt_assert(gt, IS_SRIOV_VF(xe));
> >>> + xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
> >>>
> >>> set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> >>> /*
> >>> @@ -1172,3 +1174,18 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> >>> drm_printf(p, "\thandshake:\t%u.%u\n",
> >>> pf_version->major, pf_version->minor);
> >>> }
> >>> +
> >>> +/**
> >>> + * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> >>> + * @gt: the &xe_gt
> >>> + *
> >>> + * Return: True if VF post migration recovery in progress, False otherwise
> >>> + */
> >>> +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt)
> >>> +{
> >>> + struct xe_memirq *memirq = >_to_tile(gt)->memirq;
> >>> +
> >>> + return IS_SRIOV_VF(gt_to_xe(gt)) &&
> >>
> >> this is xe_gt_sriov_vf function, so it is expected to be called only by
> >> the VF code, thus we should rather use xe_gt_assert here and the caller
> >> is responsible for the IS_SRIOV_VF check
> >>
> >
> > That is not how I have coded this. I blindly call this in various places
> > and I don't think it could at call site to determine if it is a VF as we
> > have if (VF) statements all over the driver. If perf is the concern, we
> > could move the IS_SRIOV_VF(gt_to_xe(gt)) part of function to static
> > inline and reset of the function in an exported function.
>
> inline will work, preferable not in xe_gt_sriov_vf.h ;)
>
> static inline xe_gt_recovery_inprogress(gt)
> {
> return IS_SRIOV_VF(xe) && xe_gt_sriov_vf_recovery_inprogress(gt);
> }
>
+1
> >
> >>> + (xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
> >>> + READ_ONCE(gt->sriov.vf.migration.recovery_inprogress));
> >>> +}
> >>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> >>> index 0af1dc769fe0..bb5f8eace19b 100644
> >>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> >>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> >>> @@ -25,6 +25,8 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> >>> int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> >>> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> >>>
> >>> +bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> >>> +
> >>> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> >>> u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
> >>> u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
> >>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> >>> index d95857bd789b..7b10b8e1e10e 100644
> >>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> >>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> >>> @@ -49,6 +49,14 @@ struct xe_gt_sriov_vf_runtime {
> >>> } *regs;
> >>> };
> >>>
> >>> +/**
> >>> + * xe_gt_sriov_vf_migration - VF migration data.
> >>> + */
> >>> +struct xe_gt_sriov_vf_migration {
> >>> + /** @recovery_inprogress: VF post migration recovery in progress */
> >>> + bool recovery_inprogress;
> >>> +};
> >>> +
> >>> /**
> >>> * struct xe_gt_sriov_vf - GT level VF virtualization data.
> >>> */
> >>> @@ -61,6 +69,8 @@ struct xe_gt_sriov_vf {
> >>> struct xe_gt_sriov_vf_selfconfig self_config;
> >>> /** @runtime: runtime data retrieved from the PF. */
> >>> struct xe_gt_sriov_vf_runtime runtime;
> >>> + /** @migration: migration data for the VF. */
> >>> + struct xe_gt_sriov_vf_migration migration;
> >>> };
> >>>
> >>> #endif
> >>> diff --git a/drivers/gpu/drm/xe/xe_memirq.c b/drivers/gpu/drm/xe/xe_memirq.c
> >>> index 49c45ec3e83c..94d5d6859aab 100644
> >>> --- a/drivers/gpu/drm/xe/xe_memirq.c
> >>> +++ b/drivers/gpu/drm/xe/xe_memirq.c
> >>> @@ -398,6 +398,23 @@ void xe_memirq_postinstall(struct xe_memirq *memirq)
> >>> memirq_set_enable(memirq, true);
> >>> }
> >>>
> >>> +static bool memirq_received_noclear(struct xe_memirq *memirq,
> >>> + struct iosys_map *vector,
> >>> + u16 offset, const char *name)
> >>
> >> maybe instead of duplicating code of memirq_received() in 90% just add there
> >> the "bool clear" flag?
> >>
> >
> > Sure.
> >
> >>> +{
> >>> + u8 value;
> >>> +
> >>> + value = iosys_map_rd(vector, offset, u8);
> >>> + if (value) {
> >>> + if (value != 0xff)
> >>> + memirq_err_ratelimited(memirq,
> >>> + "Unexpected memirq value %#x from %s at %u\n",
> >>> + value, name, offset);
> >>> + }
> >>> +
> >>> + return value;
> >>> +}
> >>> +
> >>> static bool memirq_received(struct xe_memirq *memirq, struct iosys_map *vector,
> >>> u16 offset, const char *name)
> >>> {
> >>> @@ -434,8 +451,16 @@ static void memirq_dispatch_guc(struct xe_memirq *memirq, struct iosys_map *stat
> >>> if (memirq_received(memirq, status, ilog2(GUC_INTR_GUC2HOST), name))
> >>> xe_guc_irq_handler(guc, GUC_INTR_GUC2HOST);
> >>>
> >>> - if (memirq_received(memirq, status, ilog2(GUC_INTR_SW_INT_0), name))
> >>> + /*
> >>> + * We must wait to perform the clear operation until after
> >>> + * xe_gt_sriov_vf_start_migration_recovery() runs, to avoid race
> >>> + * conditions where xe_gt_sriov_vf_recovery_inprogress() returns false.
> >>
> >> but the VF recovery "inprogress" shall be already set in the top level
> >>
> >> xe_sriov_vf_start_migration_recovery()
> >>
> >> even before the GT-level recovery starts, where is this race ?
> >>
> >
> > If we clear the interrupt it here, this is before the IRQ handler is
> > called which flips the software bit for "inprogress". There would be
> > window where xe_gt_sriov_vf_recovery_inprogress could return false which
> > is problematic.
>
> but who calls that recovery_inprogress() and why should it be problematic?
>
> note there might be still other threads that will finish some actions
> (including sending CTB) that will just be stuck there until we start
> the recovery, so even if recovery_inprogress() returns false for the
> small window that shouldn't change the picture
>
No. Read 'Waiters during VF post migration recovery' in [1]. That should
explain the reason why this is required. Let me know if what is
documented there is unclear.
[1] https://patchwork.freedesktop.org/patch/676374/?series=154627&rev=2
> >
> >>> + */
> >>> + if (memirq_received_noclear(memirq, status, ilog2(GUC_INTR_SW_INT_0),
> >>> + name)) {
> >>> xe_guc_irq_handler(guc, GUC_INTR_SW_INT_0);
> >>
> >> what if new MEMIRQ will arrive just here
> >>
> >> is it ok that we will clear it immediately?
> >>
> >> the whole memirq flow is that we clear irq byte first and then process
> >> it, so if anything comes right after we finish processing will be noticed
> >> on next iteration
> >>
> >
> > I don't see why that matters in this cabuse. In either case multiple IRQs
> > could happen and single IRQ handler runs.
> >
> >> I assume any races due to double migration shall be handled on the VF2GUC
> >> communication level while sending RESFIX_START/DONE, not here
> >
> > I don't think it should be possible to get multiple RESFIX_START IRQs
>
> I assume the SW_INT_0 is set on every migration, even in case we didn't
> have a chance to read it and clear it, or we didn't start (by sending
> RESFIX_START) nor finish (RESFIX_DONE)
>
But can migration be triggered and then another one before the initial
migration completes? I just don't see how that is possible or how it
wouldn't break the world (i.e., The VF of GuC explodes somewhere).
> > before DONE is complete. This existing code upstream seems to handle
> > these cases, so my series attempts to handle this too, but it seems like
> > something that shouldn't be possible.
> >
> >>> + iosys_map_wr(status, ilog2(GUC_INTR_SW_INT_0), u8, 0x00);
> >>> + }
> >>> }
> >>>
> >>> /**
> >>> @@ -460,6 +485,27 @@ void xe_memirq_hwe_handler(struct xe_memirq *memirq, struct xe_hw_engine *hwe)
> >>> }
> >>> }
> >>>
> >>> +/**
> >>> + * xe_memirq_vf_recovery_irq_pending() - VF recovery IRQ is pending
> >>
> >> this function isn't really using anything VF specific except that on the
> >> VF the SW_INT_0 means "migrated"
> >>
> >> maybe we can drop the _vf from function name?
> >>
> >
> > Sure, so 'xe_memirq_recovery_irq_pending'?
> >
> > Or 'xe_memirq_sw_int0_irq_pending'?
>
>
> >
> >> xe_memirq_pending_guc(memirq, guc, bit)
>
> I would make it more generic and provide bit param
>
> note that xe_guc_irq_handler() also takes a bit
>
Sure.
Matt
>
> >>> + * @memirq: the &xe_memirq
> >>> + * @guc: the &xe_guc to check for IRQ
> >>> + *
> >>> + * Return: True if VF recovery IRQ is pending on @guc, False otherwise
> >>> + */
> >>> +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
> >>> + struct xe_guc *guc)
> >>> +{
> >>> + struct xe_gt *gt = guc_to_gt(guc);
> >>> + struct iosys_map map;
> >>> +
> >>> + if (xe_gt_is_media_type(gt))
> >>> + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_MGUC) * SZ_16);
> >>> + else
> >>> + map = IOSYS_MAP_INIT_OFFSET(&memirq->status, ilog2(INTR_GUC) * SZ_16);
> >>
> >> nit: maybe just calc offset conditionally?
> >>
> >> u32 offset = is_media ? ilog2(INTR_MGUC) : ilog2(INTR_GUC);
> >>
> >
> > Sure.
> >
> >>> +
> >>> + return iosys_map_rd(&map, ilog2(GUC_INTR_SW_INT_0), u8);
> >>
> >> as we have helpers we should use them
> >>
> >> return memirq_received_noclear(...)
> >
> > Sure.
> >
> > Matt
> >
> >>
> >>> +}
> >>> +
> >>> /**
> >>> * xe_memirq_handler - The `Memory Based Interrupts`_ Handler.
> >>> * @memirq: the &xe_memirq
> >>> diff --git a/drivers/gpu/drm/xe/xe_memirq.h b/drivers/gpu/drm/xe/xe_memirq.h
> >>> index 06130650e9d6..476b8cba179d 100644
> >>> --- a/drivers/gpu/drm/xe/xe_memirq.h
> >>> +++ b/drivers/gpu/drm/xe/xe_memirq.h
> >>> @@ -25,4 +25,7 @@ void xe_memirq_handler(struct xe_memirq *memirq);
> >>>
> >>> int xe_memirq_init_guc(struct xe_memirq *memirq, struct xe_guc *guc);
> >>>
> >>> +bool xe_memirq_vf_recovery_irq_pending(struct xe_memirq *memirq,
> >>> + struct xe_guc *guc);
> >>> +
> >>> #endif
> >>
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (10 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 11/34] drm/xe/vf: Add xe_gt_sriov_vf_recovery_inprogress helper Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 10:49 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 13/34] drm/xe/vf: Abort H2G sends during VF post-migration recovery Matthew Brost
` (24 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
VF recovery is a per-GT operation, so it makes sense to isolate it to a
per-GT queue. Scheduling this operation on the same worker as the GT
reset and TDR not only aligns with this design but also helps avoid race
conditions, as those operations can also modify the queue state.
v2:
- Fix lockdep splat (Adam)
- Use xe_sriov_vf_migration_supported helper
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
6 files changed, 169 insertions(+), 258 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index c9d0e32e7a15..cfb71b749e52 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -25,11 +25,15 @@
#include "xe_guc.h"
#include "xe_guc_hxg_helpers.h"
#include "xe_guc_relay.h"
+#include "xe_guc_submit.h"
+#include "xe_irq.h"
#include "xe_lrc.h"
#include "xe_memirq.h"
#include "xe_mmio.h"
+#include "xe_pm.h"
#include "xe_sriov.h"
#include "xe_sriov_vf.h"
+#include "xe_tile_sriov_vf.h"
#include "xe_uc_fw.h"
#include "xe_wopcm.h"
@@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
* Returns: 0 if the operation completed successfully, or a negative error
* code otherwise.
*/
-int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
+static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
{
struct xe_guc *guc = >->uc.guc;
int err;
@@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
* xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
* @gt: the &xe_gt struct instance
*/
-void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
+static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
{
struct xe_hw_engine *hwe;
enum xe_hw_engine_id id;
@@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
xe_default_lrc_update_memirq_regs_with_address(hwe);
}
+static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
+{
+ bool started;
+
+ xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
+
+ spin_lock(>->sriov.vf.migration.lock);
+
+ if (!gt->sriov.vf.migration.recovery_queued) {
+ gt->sriov.vf.migration.recovery_queued = true;
+ WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
+
+ started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
+ xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
+ "scheduled" : "already in progress");
+ }
+
+ spin_unlock(>->sriov.vf.migration.lock);
+}
+
/**
* xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
* or just mark that a GuC is ready for it.
@@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(xe));
xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
- set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
- /*
- * We need to be certain that if all flags were set, at least one
- * thread will notice that and schedule the recovery.
- */
- smp_mb__after_atomic();
-
xe_gt_sriov_info(gt, "ready for recovery after migration\n");
- xe_sriov_vf_start_migration_recovery(xe);
+ xe_gt_sriov_vf_start_migration_recovery(gt);
}
static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
@@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
pf_version->major, pf_version->minor);
}
+static void vf_post_migration_shutdown(struct xe_gt *gt)
+{
+ int ret = 0;
+
+ spin_lock_irq(>->sriov.vf.migration.lock);
+ gt->sriov.vf.migration.recovery_queued = false;
+ spin_unlock_irq(>->sriov.vf.migration.lock);
+
+ xe_guc_submit_pause(>->uc.guc);
+ ret |= xe_guc_submit_reset_block(>->uc.guc);
+
+ if (ret)
+ xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
+}
+
+static size_t post_migration_scratch_size(struct xe_device *xe)
+{
+ return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
+}
+
+static int vf_post_migration_fixups(struct xe_gt *gt)
+{
+ s64 shift;
+ void *buf;
+ int err;
+
+ buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
+ if (!buf)
+ return -ENOMEM;
+
+ err = xe_gt_sriov_vf_query_config(gt);
+ if (err)
+ goto out;
+
+ shift = xe_gt_sriov_vf_ggtt_shift(gt);
+ if (shift) {
+ xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
+ xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
+ err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
+ if (err)
+ goto out;
+ }
+
+out:
+ kfree(buf);
+ return err;
+}
+
+static void vf_post_migration_kickstart(struct xe_gt *gt)
+{
+ /*
+ * Make sure interrupts on the new HW are properly set. The GuC IRQ
+ * must be working at this point, since the recovery did started,
+ * but the rest was not enabled using the procedure from spec.
+ */
+ xe_irq_resume(gt_to_xe(gt));
+
+ xe_guc_submit_reset_unblock(>->uc.guc);
+ xe_guc_submit_unpause(>->uc.guc);
+}
+
+static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
+{
+ bool skip_resfix = false;
+
+ spin_lock_irq(>->sriov.vf.migration.lock);
+ if (gt->sriov.vf.migration.recovery_queued) {
+ skip_resfix = true;
+ xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
+ } else {
+ WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
+ }
+ spin_unlock_irq(>->sriov.vf.migration.lock);
+
+ return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
+}
+
+static void vf_post_migration_recovery(struct xe_gt *gt)
+{
+ struct xe_device *xe = gt_to_xe(gt);
+ int err;
+
+ xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
+
+ xe_pm_runtime_get(xe);
+ vf_post_migration_shutdown(gt);
+
+ if (!xe_sriov_vf_migration_supported(xe)) {
+ xe_gt_sriov_err(gt, "migration is not supported\n");
+ err = -ENOTRECOVERABLE;
+ goto fail;
+ }
+
+ err = vf_post_migration_fixups(gt);
+ if (err)
+ goto fail;
+
+ vf_post_migration_kickstart(gt);
+ err = vf_post_migration_notify_resfix_done(gt);
+ if (err && err != -EAGAIN)
+ goto fail;
+
+ xe_pm_runtime_put(xe);
+ xe_gt_sriov_notice(gt, "migration recovery ended\n");
+ return;
+fail:
+ xe_pm_runtime_put(xe);
+ xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
+ xe_device_declare_wedged(xe);
+}
+
+static void migration_worker_func(struct work_struct *w)
+{
+ struct xe_gt *gt = container_of(w, struct xe_gt,
+ sriov.vf.migration.worker);
+
+ vf_post_migration_recovery(gt);
+}
+
+/**
+ * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
+ * @gt: the &xe_gt
+ */
+void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
+{
+ init_rwsem(>->sriov.vf.self_config.lock);
+ spin_lock_init(>->sriov.vf.migration.lock);
+ INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
+
+ if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
+ xe_gt_sriov_info(gt, "migration not supported by this module version\n");
+}
+
/**
* xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
* @gt: the &xe_gt
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index bb5f8eace19b..2ac6775b52f0 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
int xe_gt_sriov_vf_connect(struct xe_gt *gt);
int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
-void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
-int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
+void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index 7b10b8e1e10e..53680a2f188a 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -8,6 +8,7 @@
#include <linux/rwsem.h>
#include <linux/types.h>
+#include <linux/workqueue.h>
#include "xe_uc_fw_types.h"
/**
@@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
* xe_gt_sriov_vf_migration - VF migration data.
*/
struct xe_gt_sriov_vf_migration {
+ /** @migration: VF migration recovery worker */
+ struct work_struct worker;
+ /** @lock: Protects recovery_queued */
+ spinlock_t lock;
+ /** @recovery_queued: VF post migration recovery in queued */
+ bool recovery_queued;
/** @recovery_inprogress: VF post migration recovery in progress */
bool recovery_inprogress;
};
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index da064a1e7419..7d91553c4acc 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -6,21 +6,12 @@
#include <drm/drm_debugfs.h>
#include <drm/drm_managed.h>
-#include "xe_assert.h"
-#include "xe_device.h"
#include "xe_gt.h"
-#include "xe_gt_sriov_printk.h"
#include "xe_gt_sriov_vf.h"
#include "xe_guc.h"
-#include "xe_guc_submit.h"
-#include "xe_irq.h"
-#include "xe_lrc.h"
-#include "xe_pm.h"
-#include "xe_sriov.h"
#include "xe_sriov_printk.h"
#include "xe_sriov_vf.h"
#include "xe_sriov_vf_ccs.h"
-#include "xe_tile_sriov_vf.h"
/**
* DOC: VF restore procedure in PF KMD and VF KMD
@@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
xe->sriov.vf.migration.enabled = false;
}
-static void migration_worker_func(struct work_struct *w);
-
static void vf_migration_init_early(struct xe_device *xe)
{
/*
@@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
guc_version.major, guc_version.minor);
}
- INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
-
xe->sriov.vf.migration.enabled = true;
xe_sriov_dbg(xe, "migration support enabled\n");
}
@@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
unsigned int id;
for_each_gt(gt, xe, id)
- init_rwsem(>->sriov.vf.self_config.lock);
+ xe_gt_sriov_vf_migration_init_early(gt);
vf_migration_init_early(xe);
}
-/**
- * vf_post_migration_shutdown - Stop the driver activities after VF migration.
- * @xe: the &xe_device struct instance
- *
- * After this VM is migrated and assigned to a new VF, it is running on a new
- * hardware, and therefore many hardware-dependent states and related structures
- * require fixups. Without fixups, the hardware cannot do any work, and therefore
- * all GPU pipelines are stalled.
- * Stop some of kernel activities to make the fixup process faster.
- */
-static void vf_post_migration_shutdown(struct xe_device *xe)
-{
- struct xe_gt *gt;
- unsigned int id;
- int ret = 0;
-
- for_each_gt(gt, xe, id) {
- xe_guc_submit_pause(>->uc.guc);
- ret |= xe_guc_submit_reset_block(>->uc.guc);
- }
-
- if (ret)
- drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
-}
-
-/**
- * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
- * @xe: the &xe_device struct instance
- *
- * After we have finished with all post-migration fixups, restart the driver
- * activities to continue feeding the GPU with workloads.
- */
-static void vf_post_migration_kickstart(struct xe_device *xe)
-{
- struct xe_gt *gt;
- unsigned int id;
-
- /*
- * Make sure interrupts on the new HW are properly set. The GuC IRQ
- * must be working at this point, since the recovery did started,
- * but the rest was not enabled using the procedure from spec.
- */
- xe_irq_resume(xe);
-
- for_each_gt(gt, xe, id) {
- xe_guc_submit_reset_unblock(>->uc.guc);
- xe_guc_submit_unpause(>->uc.guc);
- }
-}
-
-static bool gt_vf_post_migration_needed(struct xe_gt *gt)
-{
- return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
-}
-
-/*
- * Notify GuCs marked in flags about resource fixups apply finished.
- * @xe: the &xe_device struct instance
- * @gt_flags: flags marking to which GTs the notification shall be sent
- */
-static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
-{
- struct xe_gt *gt;
- unsigned int id;
- int err = 0;
-
- for_each_gt(gt, xe, id) {
- if (!test_bit(id, >_flags))
- continue;
- /* skip asking GuC for RESFIX exit if new recovery request arrived */
- if (gt_vf_post_migration_needed(gt))
- continue;
- err = xe_gt_sriov_vf_notify_resfix_done(gt);
- if (err)
- break;
- clear_bit(id, >_flags);
- }
-
- if (gt_flags && !err)
- drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
- return err;
-}
-
-static int vf_get_next_migrated_gt_id(struct xe_device *xe)
-{
- struct xe_gt *gt;
- unsigned int id;
-
- for_each_gt(gt, xe, id) {
- if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
- return id;
- }
- return -1;
-}
-
-static size_t post_migration_scratch_size(struct xe_device *xe)
-{
- return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
-}
-
-/**
- * Perform post-migration fixups on a single GT.
- *
- * After migration, GuC needs to be re-queried for VF configuration to check
- * if it matches previous provisioning. Most of VF provisioning shall be the
- * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
- * range has changed, we have to perform fixups - shift all GGTT references
- * used anywhere within the driver. After the fixups in this function succeed,
- * it is allowed to ask the GuC bound to this GT to continue normal operation.
- *
- * Returns: 0 if the operation completed successfully, or a negative error
- * code otherwise.
- */
-static int gt_vf_post_migration_fixups(struct xe_gt *gt)
-{
- s64 shift;
- void *buf;
- int err;
-
- buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
- if (!buf)
- return -ENOMEM;
-
- err = xe_gt_sriov_vf_query_config(gt);
- if (err)
- goto out;
-
- shift = xe_gt_sriov_vf_ggtt_shift(gt);
- if (shift) {
- xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
- xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
- err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
- if (err)
- goto out;
- }
-
-out:
- kfree(buf);
- return err;
-}
-
-static void vf_post_migration_recovery(struct xe_device *xe)
-{
- unsigned long fixed_gts = 0;
- int id, err;
-
- drm_dbg(&xe->drm, "migration recovery in progress\n");
- xe_pm_runtime_get(xe);
- vf_post_migration_shutdown(xe);
-
- if (!xe_sriov_vf_migration_supported(xe)) {
- xe_sriov_err(xe, "migration is not supported\n");
- err = -ENOTRECOVERABLE;
- goto fail;
- }
-
- while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
- struct xe_gt *gt = xe_device_get_gt(xe, id);
-
- err = gt_vf_post_migration_fixups(gt);
- if (err)
- goto fail;
-
- set_bit(id, &fixed_gts);
- }
-
- vf_post_migration_kickstart(xe);
- err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
- if (err)
- goto fail;
-
- xe_pm_runtime_put(xe);
- drm_notice(&xe->drm, "migration recovery ended\n");
- return;
-fail:
- xe_pm_runtime_put(xe);
- drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
- xe_device_declare_wedged(xe);
-}
-
-static void migration_worker_func(struct work_struct *w)
-{
- struct xe_device *xe = container_of(w, struct xe_device,
- sriov.vf.migration.worker);
-
- vf_post_migration_recovery(xe);
-}
-
-/*
- * Check if post-restore recovery is coming on any of GTs.
- * @xe: the &xe_device struct instance
- *
- * Return: True if migration recovery worker will soon be running. Any worker currently
- * executing does not affect the result.
- */
-static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
-{
- struct xe_gt *gt;
- unsigned int id;
-
- for_each_gt(gt, xe, id) {
- if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
- return true;
- }
- return false;
-}
-
-/**
- * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
- * @xe: the &xe_device to start recovery on
- *
- * This function shall be called only by VF.
- */
-void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
-{
- bool started;
-
- xe_assert(xe, IS_SRIOV_VF(xe));
-
- if (!vf_ready_to_recovery_on_any_gts(xe))
- return;
-
- started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
- drm_info(&xe->drm, "VF migration recovery %s\n", started ?
- "scheduled" : "already in progress");
-}
-
/**
* xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
* @xe: the &xe_device to initialize
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
index 9e752105ec2a..4df95266b261 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
@@ -13,7 +13,6 @@ struct xe_device;
void xe_sriov_vf_init_early(struct xe_device *xe);
int xe_sriov_vf_init_late(struct xe_device *xe);
-void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
bool xe_sriov_vf_migration_supported(struct xe_device *xe);
void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
index 426cc5841958..6a0fd0f5463e 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
@@ -33,10 +33,6 @@ struct xe_device_vf {
/** @migration: VF Migration state data */
struct {
- /** @migration.worker: VF migration recovery worker */
- struct work_struct worker;
- /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
- unsigned long gt_flags;
/**
* @migration.enabled: flag indicating if migration support
* was enabled or not due to missing prerequisites
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-24 1:15 ` [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker Matthew Brost
@ 2025-09-24 10:49 ` Michal Wajdeczko
2025-09-24 19:50 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 10:49 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> VF recovery is a per-GT operation, so it makes sense to isolate it to a
that was also my suggestion to make it per-GT, good to see it happen now
> per-GT queue. Scheduling this operation on the same worker as the GT
> reset and TDR not only aligns with this design but also helps avoid race
> conditions, as those operations can also modify the queue state.
but while the recovery is per-GT, we should still protect against that
one GT starts the recovery sooner then other GTs notice about it
>
> v2:
> - Fix lockdep splat (Adam)
> - Use xe_sriov_vf_migration_supported helper
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
> drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
> drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
> drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
> 6 files changed, 169 insertions(+), 258 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index c9d0e32e7a15..cfb71b749e52 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -25,11 +25,15 @@
> #include "xe_guc.h"
> #include "xe_guc_hxg_helpers.h"
> #include "xe_guc_relay.h"
> +#include "xe_guc_submit.h"
> +#include "xe_irq.h"
> #include "xe_lrc.h"
> #include "xe_memirq.h"
> #include "xe_mmio.h"
> +#include "xe_pm.h"
> #include "xe_sriov.h"
> #include "xe_sriov_vf.h"
> +#include "xe_tile_sriov_vf.h"
> #include "xe_uc_fw.h"
> #include "xe_wopcm.h"
>
> @@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
> * Returns: 0 if the operation completed successfully, or a negative error
> * code otherwise.
> */
> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> +static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> {
> struct xe_guc *guc = >->uc.guc;
> int err;
> @@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
> * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
> * @gt: the &xe_gt struct instance
> */
> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> +static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> {
> struct xe_hw_engine *hwe;
> enum xe_hw_engine_id id;
> @@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> xe_default_lrc_update_memirq_regs_with_address(hwe);
> }
>
> +static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
nit: if this is static then could be just:
vf_start_migration_recovery(gt)
> +{
> + bool started;
> +
> + xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> +
> + spin_lock(>->sriov.vf.migration.lock);
> +
> + if (!gt->sriov.vf.migration.recovery_queued) {
> + gt->sriov.vf.migration.recovery_queued = true;
> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
> +
> + started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
> + xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
> + "scheduled" : "already in progress");
with this .recovery_queued flag, can we hit "already in progress" case?
> + }
> +
> + spin_unlock(>->sriov.vf.migration.lock);
> +}
> +
> /**
> * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
> * or just mark that a GuC is ready for it.
> @@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> xe_gt_assert(gt, IS_SRIOV_VF(xe));
> xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
>
> - set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> - /*
> - * We need to be certain that if all flags were set, at least one
> - * thread will notice that and schedule the recovery.
> - */
> - smp_mb__after_atomic();
> -
> xe_gt_sriov_info(gt, "ready for recovery after migration\n");
> - xe_sriov_vf_start_migration_recovery(xe);
> + xe_gt_sriov_vf_start_migration_recovery(gt);
> }
>
> static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
> @@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> pf_version->major, pf_version->minor);
> }
>
> +static void vf_post_migration_shutdown(struct xe_gt *gt)
> +{
> + int ret = 0;
> +
> + spin_lock_irq(>->sriov.vf.migration.lock);
> + gt->sriov.vf.migration.recovery_queued = false;
> + spin_unlock_irq(>->sriov.vf.migration.lock);
> +
> + xe_guc_submit_pause(>->uc.guc);
> + ret |= xe_guc_submit_reset_block(>->uc.guc);
this |= seem unneeded
> +
> + if (ret)
> + xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
is this the only possible reason? maybe worth to add %pe ?
> +}
> +
> +static size_t post_migration_scratch_size(struct xe_device *xe)
> +{
> + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> +}
> +
> +static int vf_post_migration_fixups(struct xe_gt *gt)
> +{
> + s64 shift;
> + void *buf;
> + int err;
> +
> + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
> + if (!buf)
> + return -ENOMEM;
> +
> + err = xe_gt_sriov_vf_query_config(gt);
> + if (err)
> + goto out;
> +
> + shift = xe_gt_sriov_vf_ggtt_shift(gt);
> + if (shift) {
> + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> + if (err)
> + goto out;
> + }
> +
> +out:
> + kfree(buf);
> + return err;
> +}
> +
> +static void vf_post_migration_kickstart(struct xe_gt *gt)
> +{
> + /*
> + * Make sure interrupts on the new HW are properly set. The GuC IRQ
> + * must be working at this point, since the recovery did started,
> + * but the rest was not enabled using the procedure from spec.
> + */
> + xe_irq_resume(gt_to_xe(gt));
> +
> + xe_guc_submit_reset_unblock(>->uc.guc);
> + xe_guc_submit_unpause(>->uc.guc);
> +}
> +
> +static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> +{
> + bool skip_resfix = false;
> +
> + spin_lock_irq(>->sriov.vf.migration.lock);
> + if (gt->sriov.vf.migration.recovery_queued) {
> + skip_resfix = true;
> + xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
> + } else {
> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
> + }
> + spin_unlock_irq(>->sriov.vf.migration.lock);
> +
> + return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
nit: this looks cleaner:
if (skip_resfix)
return -EAGAIN;
return xe_gt_sriov_vf_notify_resfix_done(gt);
> +}
> +
> +static void vf_post_migration_recovery(struct xe_gt *gt)
> +{
> + struct xe_device *xe = gt_to_xe(gt);
> + int err;
> +
> + xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
> +
> + xe_pm_runtime_get(xe);
> + vf_post_migration_shutdown(gt);
> +
> + if (!xe_sriov_vf_migration_supported(xe)) {
> + xe_gt_sriov_err(gt, "migration is not supported\n");
> + err = -ENOTRECOVERABLE;
> + goto fail;
> + }
> +
> + err = vf_post_migration_fixups(gt);
> + if (err)
> + goto fail;
> +
> + vf_post_migration_kickstart(gt);
> + err = vf_post_migration_notify_resfix_done(gt);
> + if (err && err != -EAGAIN)
> + goto fail;
> +
> + xe_pm_runtime_put(xe);
> + xe_gt_sriov_notice(gt, "migration recovery ended\n");
> + return;
> +fail:
> + xe_pm_runtime_put(xe);
> + xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
> + xe_device_declare_wedged(xe);
> +}
> +
> +static void migration_worker_func(struct work_struct *w)
> +{
> + struct xe_gt *gt = container_of(w, struct xe_gt,
> + sriov.vf.migration.worker);
> +
> + vf_post_migration_recovery(gt);
> +}
> +
> +/**
> + * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> + * @gt: the &xe_gt
> + */
> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> +{
> + init_rwsem(>->sriov.vf.self_config.lock);
> + spin_lock_init(>->sriov.vf.migration.lock);
> + INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
> +
> + if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> + xe_gt_sriov_info(gt, "migration not supported by this module version\n");
we likely don't want to repeat that message on every GT
> +}
> +
> /**
> * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> * @gt: the &xe_gt
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> index bb5f8eace19b..2ac6775b52f0 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> @@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
> int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
> int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
>
> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
>
> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> index 7b10b8e1e10e..53680a2f188a 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> @@ -8,6 +8,7 @@
>
> #include <linux/rwsem.h>
> #include <linux/types.h>
> +#include <linux/workqueue.h>
> #include "xe_uc_fw_types.h"
>
> /**
> @@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
> * xe_gt_sriov_vf_migration - VF migration data.
> */
> struct xe_gt_sriov_vf_migration {
> + /** @migration: VF migration recovery worker */
> + struct work_struct worker;
> + /** @lock: Protects recovery_queued */
> + spinlock_t lock;
> + /** @recovery_queued: VF post migration recovery in queued */
> + bool recovery_queued;
> /** @recovery_inprogress: VF post migration recovery in progress */
> bool recovery_inprogress;
> };
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> index da064a1e7419..7d91553c4acc 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> @@ -6,21 +6,12 @@
> #include <drm/drm_debugfs.h>
> #include <drm/drm_managed.h>
>
> -#include "xe_assert.h"
> -#include "xe_device.h"
> #include "xe_gt.h"
> -#include "xe_gt_sriov_printk.h"
> #include "xe_gt_sriov_vf.h"
> #include "xe_guc.h"
> -#include "xe_guc_submit.h"
> -#include "xe_irq.h"
> -#include "xe_lrc.h"
> -#include "xe_pm.h"
> -#include "xe_sriov.h"
> #include "xe_sriov_printk.h"
> #include "xe_sriov_vf.h"
> #include "xe_sriov_vf_ccs.h"
> -#include "xe_tile_sriov_vf.h"
>
> /**
> * DOC: VF restore procedure in PF KMD and VF KMD
> @@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
> xe->sriov.vf.migration.enabled = false;
> }
>
> -static void migration_worker_func(struct work_struct *w);
> -
> static void vf_migration_init_early(struct xe_device *xe)
> {
> /*
> @@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
> guc_version.major, guc_version.minor);
> }
>
> - INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
> -
> xe->sriov.vf.migration.enabled = true;
> xe_sriov_dbg(xe, "migration support enabled\n");
> }
> @@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
> unsigned int id;
>
> for_each_gt(gt, xe, id)
> - init_rwsem(>->sriov.vf.self_config.lock);
> + xe_gt_sriov_vf_migration_init_early(gt);
still, this should be called from gt_init_early kind of functions
>
> vf_migration_init_early(xe);
> }
>
> -/**
> - * vf_post_migration_shutdown - Stop the driver activities after VF migration.
> - * @xe: the &xe_device struct instance
> - *
> - * After this VM is migrated and assigned to a new VF, it is running on a new
> - * hardware, and therefore many hardware-dependent states and related structures
> - * require fixups. Without fixups, the hardware cannot do any work, and therefore
> - * all GPU pipelines are stalled.
> - * Stop some of kernel activities to make the fixup process faster.
> - */
> -static void vf_post_migration_shutdown(struct xe_device *xe)
> -{
> - struct xe_gt *gt;
> - unsigned int id;
> - int ret = 0;
> -
> - for_each_gt(gt, xe, id) {
> - xe_guc_submit_pause(>->uc.guc);
> - ret |= xe_guc_submit_reset_block(>->uc.guc);
> - }
> -
> - if (ret)
> - drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
> -}
> -
> -/**
> - * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
> - * @xe: the &xe_device struct instance
> - *
> - * After we have finished with all post-migration fixups, restart the driver
> - * activities to continue feeding the GPU with workloads.
> - */
> -static void vf_post_migration_kickstart(struct xe_device *xe)
> -{
> - struct xe_gt *gt;
> - unsigned int id;
> -
> - /*
> - * Make sure interrupts on the new HW are properly set. The GuC IRQ
> - * must be working at this point, since the recovery did started,
> - * but the rest was not enabled using the procedure from spec.
> - */
> - xe_irq_resume(xe);
> -
> - for_each_gt(gt, xe, id) {
> - xe_guc_submit_reset_unblock(>->uc.guc);
> - xe_guc_submit_unpause(>->uc.guc);
> - }
> -}
> -
> -static bool gt_vf_post_migration_needed(struct xe_gt *gt)
> -{
> - return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
> -}
> -
> -/*
> - * Notify GuCs marked in flags about resource fixups apply finished.
> - * @xe: the &xe_device struct instance
> - * @gt_flags: flags marking to which GTs the notification shall be sent
> - */
> -static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
> -{
> - struct xe_gt *gt;
> - unsigned int id;
> - int err = 0;
> -
> - for_each_gt(gt, xe, id) {
> - if (!test_bit(id, >_flags))
> - continue;
> - /* skip asking GuC for RESFIX exit if new recovery request arrived */
> - if (gt_vf_post_migration_needed(gt))
> - continue;
> - err = xe_gt_sriov_vf_notify_resfix_done(gt);
> - if (err)
> - break;
> - clear_bit(id, >_flags);
> - }
> -
> - if (gt_flags && !err)
> - drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
> - return err;
> -}
> -
> -static int vf_get_next_migrated_gt_id(struct xe_device *xe)
> -{
> - struct xe_gt *gt;
> - unsigned int id;
> -
> - for_each_gt(gt, xe, id) {
> - if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
> - return id;
> - }
> - return -1;
> -}
> -
> -static size_t post_migration_scratch_size(struct xe_device *xe)
> -{
> - return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> -}
> -
> -/**
> - * Perform post-migration fixups on a single GT.
> - *
> - * After migration, GuC needs to be re-queried for VF configuration to check
> - * if it matches previous provisioning. Most of VF provisioning shall be the
> - * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
> - * range has changed, we have to perform fixups - shift all GGTT references
> - * used anywhere within the driver. After the fixups in this function succeed,
> - * it is allowed to ask the GuC bound to this GT to continue normal operation.
> - *
> - * Returns: 0 if the operation completed successfully, or a negative error
> - * code otherwise.
> - */
> -static int gt_vf_post_migration_fixups(struct xe_gt *gt)
> -{
> - s64 shift;
> - void *buf;
> - int err;
> -
> - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
> - if (!buf)
> - return -ENOMEM;
> -
> - err = xe_gt_sriov_vf_query_config(gt);
> - if (err)
> - goto out;
> -
> - shift = xe_gt_sriov_vf_ggtt_shift(gt);
> - if (shift) {
> - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> - if (err)
> - goto out;
> - }
> -
> -out:
> - kfree(buf);
> - return err;
> -}
> -
> -static void vf_post_migration_recovery(struct xe_device *xe)
> -{
> - unsigned long fixed_gts = 0;
> - int id, err;
> -
> - drm_dbg(&xe->drm, "migration recovery in progress\n");
> - xe_pm_runtime_get(xe);
> - vf_post_migration_shutdown(xe);
> -
> - if (!xe_sriov_vf_migration_supported(xe)) {
> - xe_sriov_err(xe, "migration is not supported\n");
> - err = -ENOTRECOVERABLE;
> - goto fail;
> - }
> -
> - while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
> - struct xe_gt *gt = xe_device_get_gt(xe, id);
> -
> - err = gt_vf_post_migration_fixups(gt);
> - if (err)
> - goto fail;
> -
> - set_bit(id, &fixed_gts);
> - }
> -
> - vf_post_migration_kickstart(xe);
> - err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
> - if (err)
> - goto fail;
> -
> - xe_pm_runtime_put(xe);
> - drm_notice(&xe->drm, "migration recovery ended\n");
> - return;
> -fail:
> - xe_pm_runtime_put(xe);
> - drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
> - xe_device_declare_wedged(xe);
> -}
> -
> -static void migration_worker_func(struct work_struct *w)
> -{
> - struct xe_device *xe = container_of(w, struct xe_device,
> - sriov.vf.migration.worker);
> -
> - vf_post_migration_recovery(xe);
> -}
> -
> -/*
> - * Check if post-restore recovery is coming on any of GTs.
> - * @xe: the &xe_device struct instance
> - *
> - * Return: True if migration recovery worker will soon be running. Any worker currently
> - * executing does not affect the result.
> - */
> -static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
> -{
> - struct xe_gt *gt;
> - unsigned int id;
> -
> - for_each_gt(gt, xe, id) {
> - if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
> - return true;
> - }
> - return false;
> -}
> -
> -/**
> - * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
> - * @xe: the &xe_device to start recovery on
> - *
> - * This function shall be called only by VF.
> - */
> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
> -{
> - bool started;
> -
> - xe_assert(xe, IS_SRIOV_VF(xe));
> -
> - if (!vf_ready_to_recovery_on_any_gts(xe))
> - return;
> -
> - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
> - drm_info(&xe->drm, "VF migration recovery %s\n", started ?
> - "scheduled" : "already in progress");
> -}
> -
> /**
> * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
> * @xe: the &xe_device to initialize
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
> index 9e752105ec2a..4df95266b261 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
> @@ -13,7 +13,6 @@ struct xe_device;
>
> void xe_sriov_vf_init_early(struct xe_device *xe);
> int xe_sriov_vf_init_late(struct xe_device *xe);
> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
> bool xe_sriov_vf_migration_supported(struct xe_device *xe);
> void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
>
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> index 426cc5841958..6a0fd0f5463e 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> @@ -33,10 +33,6 @@ struct xe_device_vf {
>
> /** @migration: VF Migration state data */
> struct {
> - /** @migration.worker: VF migration recovery worker */
> - struct work_struct worker;
> - /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
> - unsigned long gt_flags;
> /**
> * @migration.enabled: flag indicating if migration support
> * was enabled or not due to missing prerequisites
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-24 10:49 ` Michal Wajdeczko
@ 2025-09-24 19:50 ` Matthew Brost
2025-09-24 20:21 ` Michal Wajdeczko
0 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 19:50 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 12:49:25PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > VF recovery is a per-GT operation, so it makes sense to isolate it to a
>
> that was also my suggestion to make it per-GT, good to see it happen now
>
+1
> > per-GT queue. Scheduling this operation on the same worker as the GT
> > reset and TDR not only aligns with this design but also helps avoid race
> > conditions, as those operations can also modify the queue state.
>
> but while the recovery is per-GT, we should still protect against that
> one GT starts the recovery sooner then other GTs notice about it
>
Yes. There is shared state in 2 places:
- The GGTT shifting, this handled by [1] in my series.
- CCS restore on iGPU (PTL) handled by [2] [3] in my series.
[1] https://patchwork.freedesktop.org/patch/676394/?series=154627&rev=2
[2] https://patchwork.freedesktop.org/patch/676393/?series=154627&rev=2
[3] https://patchwork.freedesktop.org/patch/676397/?series=154627&rev=2
> >
> > v2:
> > - Fix lockdep splat (Adam)
> > - Use xe_sriov_vf_migration_supported helper
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
> > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
> > drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
> > drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
> > drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
> > 6 files changed, 169 insertions(+), 258 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index c9d0e32e7a15..cfb71b749e52 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -25,11 +25,15 @@
> > #include "xe_guc.h"
> > #include "xe_guc_hxg_helpers.h"
> > #include "xe_guc_relay.h"
> > +#include "xe_guc_submit.h"
> > +#include "xe_irq.h"
> > #include "xe_lrc.h"
> > #include "xe_memirq.h"
> > #include "xe_mmio.h"
> > +#include "xe_pm.h"
> > #include "xe_sriov.h"
> > #include "xe_sriov_vf.h"
> > +#include "xe_tile_sriov_vf.h"
> > #include "xe_uc_fw.h"
> > #include "xe_wopcm.h"
> >
> > @@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
> > * Returns: 0 if the operation completed successfully, or a negative error
> > * code otherwise.
> > */
> > -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> > +static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> > {
> > struct xe_guc *guc = >->uc.guc;
> > int err;
> > @@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
> > * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
> > * @gt: the &xe_gt struct instance
> > */
> > -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> > +static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> > {
> > struct xe_hw_engine *hwe;
> > enum xe_hw_engine_id id;
> > @@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> > xe_default_lrc_update_memirq_regs_with_address(hwe);
> > }
> >
> > +static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
>
> nit: if this is static then could be just:
>
> vf_start_migration_recovery(gt)
>
Sure.
> > +{
> > + bool started;
> > +
> > + xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > +
> > + spin_lock(>->sriov.vf.migration.lock);
> > +
> > + if (!gt->sriov.vf.migration.recovery_queued) {
> > + gt->sriov.vf.migration.recovery_queued = true;
> > + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
> > +
> > + started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
> > + xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
> > + "scheduled" : "already in progress");
>
> with this .recovery_queued flag, can we hit "already in progress" case?
>
This was existing code so kept it but I'm really unclear how we can get
multiple resfix IRQs without the prior resfix flow being completed.
Maybe Tomasz can clear this one up? Ideally I'd like to remove multiple
resfix IRQ handled code if this is not possible.
> > + }
> > +
> > + spin_unlock(>->sriov.vf.migration.lock);
> > +}
> > +
> > /**
> > * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
> > * or just mark that a GuC is ready for it.
> > @@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> > xe_gt_assert(gt, IS_SRIOV_VF(xe));
> > xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
> >
> > - set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> > - /*
> > - * We need to be certain that if all flags were set, at least one
> > - * thread will notice that and schedule the recovery.
> > - */
> > - smp_mb__after_atomic();
> > -
> > xe_gt_sriov_info(gt, "ready for recovery after migration\n");
> > - xe_sriov_vf_start_migration_recovery(xe);
> > + xe_gt_sriov_vf_start_migration_recovery(gt);
> > }
> >
> > static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
> > @@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> > pf_version->major, pf_version->minor);
> > }
> >
> > +static void vf_post_migration_shutdown(struct xe_gt *gt)
> > +{
> > + int ret = 0;
> > +
> > + spin_lock_irq(>->sriov.vf.migration.lock);
> > + gt->sriov.vf.migration.recovery_queued = false;
> > + spin_unlock_irq(>->sriov.vf.migration.lock);
> > +
> > + xe_guc_submit_pause(>->uc.guc);
> > + ret |= xe_guc_submit_reset_block(>->uc.guc);
>
> this |= seem unneeded
>
This is existing code copy / pasted and is removed in [4].
[4] https://patchwork.freedesktop.org/patch/676382/?series=154627&rev=2
> > +
> > + if (ret)
> > + xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
>
> is this the only possible reason? maybe worth to add %pe ?
>
Again this existing code and will be removed in [4].
So with above statements, I'd prefer just to leave as is.
> > +}
> > +
> > +static size_t post_migration_scratch_size(struct xe_device *xe)
> > +{
> > + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> > +}
> > +
> > +static int vf_post_migration_fixups(struct xe_gt *gt)
> > +{
> > + s64 shift;
> > + void *buf;
> > + int err;
> > +
> > + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
> > + if (!buf)
> > + return -ENOMEM;
> > +
> > + err = xe_gt_sriov_vf_query_config(gt);
> > + if (err)
> > + goto out;
> > +
> > + shift = xe_gt_sriov_vf_ggtt_shift(gt);
> > + if (shift) {
> > + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> > + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > + if (err)
> > + goto out;
> > + }
> > +
> > +out:
> > + kfree(buf);
> > + return err;
> > +}
> > +
> > +static void vf_post_migration_kickstart(struct xe_gt *gt)
> > +{
> > + /*
> > + * Make sure interrupts on the new HW are properly set. The GuC IRQ
> > + * must be working at this point, since the recovery did started,
> > + * but the rest was not enabled using the procedure from spec.
> > + */
> > + xe_irq_resume(gt_to_xe(gt));
> > +
> > + xe_guc_submit_reset_unblock(>->uc.guc);
> > + xe_guc_submit_unpause(>->uc.guc);
> > +}
> > +
> > +static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> > +{
> > + bool skip_resfix = false;
> > +
> > + spin_lock_irq(>->sriov.vf.migration.lock);
> > + if (gt->sriov.vf.migration.recovery_queued) {
> > + skip_resfix = true;
> > + xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
> > + } else {
> > + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
> > + }
> > + spin_unlock_irq(>->sriov.vf.migration.lock);
> > +
> > + return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
>
> nit: this looks cleaner:
>
> if (skip_resfix)
> return -EAGAIN;
>
> return xe_gt_sriov_vf_notify_resfix_done(gt);
>
Sure.
> > +}
> > +
> > +static void vf_post_migration_recovery(struct xe_gt *gt)
> > +{
> > + struct xe_device *xe = gt_to_xe(gt);
> > + int err;
> > +
> > + xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
> > +
> > + xe_pm_runtime_get(xe);
> > + vf_post_migration_shutdown(gt);
> > +
> > + if (!xe_sriov_vf_migration_supported(xe)) {
> > + xe_gt_sriov_err(gt, "migration is not supported\n");
> > + err = -ENOTRECOVERABLE;
> > + goto fail;
> > + }
> > +
> > + err = vf_post_migration_fixups(gt);
> > + if (err)
> > + goto fail;
> > +
> > + vf_post_migration_kickstart(gt);
> > + err = vf_post_migration_notify_resfix_done(gt);
> > + if (err && err != -EAGAIN)
> > + goto fail;
> > +
> > + xe_pm_runtime_put(xe);
> > + xe_gt_sriov_notice(gt, "migration recovery ended\n");
> > + return;
> > +fail:
> > + xe_pm_runtime_put(xe);
> > + xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
> > + xe_device_declare_wedged(xe);
> > +}
> > +
> > +static void migration_worker_func(struct work_struct *w)
> > +{
> > + struct xe_gt *gt = container_of(w, struct xe_gt,
> > + sriov.vf.migration.worker);
> > +
> > + vf_post_migration_recovery(gt);
> > +}
> > +
> > +/**
> > + * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> > + * @gt: the &xe_gt
> > + */
> > +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > +{
> > + init_rwsem(>->sriov.vf.self_config.lock);
> > + spin_lock_init(>->sriov.vf.migration.lock);
> > + INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
> > +
> > + if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> > + xe_gt_sriov_info(gt, "migration not supported by this module version\n");
>
> we likely don't want to repeat that message on every GT
>
So move this to xe_sriov_vf_init_early?
> > +}
> > +
> > /**
> > * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> > * @gt: the &xe_gt
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > index bb5f8eace19b..2ac6775b52f0 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > @@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
> > int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
> > int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> > int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> > -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> > -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> > void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> >
> > +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> > bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> >
> > u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > index 7b10b8e1e10e..53680a2f188a 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > @@ -8,6 +8,7 @@
> >
> > #include <linux/rwsem.h>
> > #include <linux/types.h>
> > +#include <linux/workqueue.h>
> > #include "xe_uc_fw_types.h"
> >
> > /**
> > @@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
> > * xe_gt_sriov_vf_migration - VF migration data.
> > */
> > struct xe_gt_sriov_vf_migration {
> > + /** @migration: VF migration recovery worker */
> > + struct work_struct worker;
> > + /** @lock: Protects recovery_queued */
> > + spinlock_t lock;
> > + /** @recovery_queued: VF post migration recovery in queued */
> > + bool recovery_queued;
> > /** @recovery_inprogress: VF post migration recovery in progress */
> > bool recovery_inprogress;
> > };
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > index da064a1e7419..7d91553c4acc 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > @@ -6,21 +6,12 @@
> > #include <drm/drm_debugfs.h>
> > #include <drm/drm_managed.h>
> >
> > -#include "xe_assert.h"
> > -#include "xe_device.h"
> > #include "xe_gt.h"
> > -#include "xe_gt_sriov_printk.h"
> > #include "xe_gt_sriov_vf.h"
> > #include "xe_guc.h"
> > -#include "xe_guc_submit.h"
> > -#include "xe_irq.h"
> > -#include "xe_lrc.h"
> > -#include "xe_pm.h"
> > -#include "xe_sriov.h"
> > #include "xe_sriov_printk.h"
> > #include "xe_sriov_vf.h"
> > #include "xe_sriov_vf_ccs.h"
> > -#include "xe_tile_sriov_vf.h"
> >
> > /**
> > * DOC: VF restore procedure in PF KMD and VF KMD
> > @@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
> > xe->sriov.vf.migration.enabled = false;
> > }
> >
> > -static void migration_worker_func(struct work_struct *w);
> > -
> > static void vf_migration_init_early(struct xe_device *xe)
> > {
> > /*
> > @@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
> > guc_version.major, guc_version.minor);
> > }
> >
> > - INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
> > -
> > xe->sriov.vf.migration.enabled = true;
> > xe_sriov_dbg(xe, "migration support enabled\n");
> > }
> > @@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
> > unsigned int id;
> >
> > for_each_gt(gt, xe, id)
> > - init_rwsem(>->sriov.vf.self_config.lock);
> > + xe_gt_sriov_vf_migration_init_early(gt);
>
> still, this should be called from gt_init_early kind of functions
>
Kinda a nit that I'm not convinced is worth while to have
xe_sriov_vf_init_early and then xe_gt_sriov_vf_migration_init_early
called in gt_init_early...
Matt
> >
> > vf_migration_init_early(xe);
> > }
> >
> > -/**
> > - * vf_post_migration_shutdown - Stop the driver activities after VF migration.
> > - * @xe: the &xe_device struct instance
> > - *
> > - * After this VM is migrated and assigned to a new VF, it is running on a new
> > - * hardware, and therefore many hardware-dependent states and related structures
> > - * require fixups. Without fixups, the hardware cannot do any work, and therefore
> > - * all GPU pipelines are stalled.
> > - * Stop some of kernel activities to make the fixup process faster.
> > - */
> > -static void vf_post_migration_shutdown(struct xe_device *xe)
> > -{
> > - struct xe_gt *gt;
> > - unsigned int id;
> > - int ret = 0;
> > -
> > - for_each_gt(gt, xe, id) {
> > - xe_guc_submit_pause(>->uc.guc);
> > - ret |= xe_guc_submit_reset_block(>->uc.guc);
> > - }
> > -
> > - if (ret)
> > - drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
> > -}
> > -
> > -/**
> > - * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
> > - * @xe: the &xe_device struct instance
> > - *
> > - * After we have finished with all post-migration fixups, restart the driver
> > - * activities to continue feeding the GPU with workloads.
> > - */
> > -static void vf_post_migration_kickstart(struct xe_device *xe)
> > -{
> > - struct xe_gt *gt;
> > - unsigned int id;
> > -
> > - /*
> > - * Make sure interrupts on the new HW are properly set. The GuC IRQ
> > - * must be working at this point, since the recovery did started,
> > - * but the rest was not enabled using the procedure from spec.
> > - */
> > - xe_irq_resume(xe);
> > -
> > - for_each_gt(gt, xe, id) {
> > - xe_guc_submit_reset_unblock(>->uc.guc);
> > - xe_guc_submit_unpause(>->uc.guc);
> > - }
> > -}
> > -
> > -static bool gt_vf_post_migration_needed(struct xe_gt *gt)
> > -{
> > - return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
> > -}
> > -
> > -/*
> > - * Notify GuCs marked in flags about resource fixups apply finished.
> > - * @xe: the &xe_device struct instance
> > - * @gt_flags: flags marking to which GTs the notification shall be sent
> > - */
> > -static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
> > -{
> > - struct xe_gt *gt;
> > - unsigned int id;
> > - int err = 0;
> > -
> > - for_each_gt(gt, xe, id) {
> > - if (!test_bit(id, >_flags))
> > - continue;
> > - /* skip asking GuC for RESFIX exit if new recovery request arrived */
> > - if (gt_vf_post_migration_needed(gt))
> > - continue;
> > - err = xe_gt_sriov_vf_notify_resfix_done(gt);
> > - if (err)
> > - break;
> > - clear_bit(id, >_flags);
> > - }
> > -
> > - if (gt_flags && !err)
> > - drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
> > - return err;
> > -}
> > -
> > -static int vf_get_next_migrated_gt_id(struct xe_device *xe)
> > -{
> > - struct xe_gt *gt;
> > - unsigned int id;
> > -
> > - for_each_gt(gt, xe, id) {
> > - if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
> > - return id;
> > - }
> > - return -1;
> > -}
> > -
> > -static size_t post_migration_scratch_size(struct xe_device *xe)
> > -{
> > - return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> > -}
> > -
> > -/**
> > - * Perform post-migration fixups on a single GT.
> > - *
> > - * After migration, GuC needs to be re-queried for VF configuration to check
> > - * if it matches previous provisioning. Most of VF provisioning shall be the
> > - * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
> > - * range has changed, we have to perform fixups - shift all GGTT references
> > - * used anywhere within the driver. After the fixups in this function succeed,
> > - * it is allowed to ask the GuC bound to this GT to continue normal operation.
> > - *
> > - * Returns: 0 if the operation completed successfully, or a negative error
> > - * code otherwise.
> > - */
> > -static int gt_vf_post_migration_fixups(struct xe_gt *gt)
> > -{
> > - s64 shift;
> > - void *buf;
> > - int err;
> > -
> > - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
> > - if (!buf)
> > - return -ENOMEM;
> > -
> > - err = xe_gt_sriov_vf_query_config(gt);
> > - if (err)
> > - goto out;
> > -
> > - shift = xe_gt_sriov_vf_ggtt_shift(gt);
> > - if (shift) {
> > - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> > - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > - if (err)
> > - goto out;
> > - }
> > -
> > -out:
> > - kfree(buf);
> > - return err;
> > -}
> > -
> > -static void vf_post_migration_recovery(struct xe_device *xe)
> > -{
> > - unsigned long fixed_gts = 0;
> > - int id, err;
> > -
> > - drm_dbg(&xe->drm, "migration recovery in progress\n");
> > - xe_pm_runtime_get(xe);
> > - vf_post_migration_shutdown(xe);
> > -
> > - if (!xe_sriov_vf_migration_supported(xe)) {
> > - xe_sriov_err(xe, "migration is not supported\n");
> > - err = -ENOTRECOVERABLE;
> > - goto fail;
> > - }
> > -
> > - while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
> > - struct xe_gt *gt = xe_device_get_gt(xe, id);
> > -
> > - err = gt_vf_post_migration_fixups(gt);
> > - if (err)
> > - goto fail;
> > -
> > - set_bit(id, &fixed_gts);
> > - }
> > -
> > - vf_post_migration_kickstart(xe);
> > - err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
> > - if (err)
> > - goto fail;
> > -
> > - xe_pm_runtime_put(xe);
> > - drm_notice(&xe->drm, "migration recovery ended\n");
> > - return;
> > -fail:
> > - xe_pm_runtime_put(xe);
> > - drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
> > - xe_device_declare_wedged(xe);
> > -}
> > -
> > -static void migration_worker_func(struct work_struct *w)
> > -{
> > - struct xe_device *xe = container_of(w, struct xe_device,
> > - sriov.vf.migration.worker);
> > -
> > - vf_post_migration_recovery(xe);
> > -}
> > -
> > -/*
> > - * Check if post-restore recovery is coming on any of GTs.
> > - * @xe: the &xe_device struct instance
> > - *
> > - * Return: True if migration recovery worker will soon be running. Any worker currently
> > - * executing does not affect the result.
> > - */
> > -static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
> > -{
> > - struct xe_gt *gt;
> > - unsigned int id;
> > -
> > - for_each_gt(gt, xe, id) {
> > - if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
> > - return true;
> > - }
> > - return false;
> > -}
> > -
> > -/**
> > - * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
> > - * @xe: the &xe_device to start recovery on
> > - *
> > - * This function shall be called only by VF.
> > - */
> > -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
> > -{
> > - bool started;
> > -
> > - xe_assert(xe, IS_SRIOV_VF(xe));
> > -
> > - if (!vf_ready_to_recovery_on_any_gts(xe))
> > - return;
> > -
> > - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
> > - drm_info(&xe->drm, "VF migration recovery %s\n", started ?
> > - "scheduled" : "already in progress");
> > -}
> > -
> > /**
> > * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
> > * @xe: the &xe_device to initialize
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
> > index 9e752105ec2a..4df95266b261 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
> > @@ -13,7 +13,6 @@ struct xe_device;
> >
> > void xe_sriov_vf_init_early(struct xe_device *xe);
> > int xe_sriov_vf_init_late(struct xe_device *xe);
> > -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
> > bool xe_sriov_vf_migration_supported(struct xe_device *xe);
> > void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
> >
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> > index 426cc5841958..6a0fd0f5463e 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> > @@ -33,10 +33,6 @@ struct xe_device_vf {
> >
> > /** @migration: VF Migration state data */
> > struct {
> > - /** @migration.worker: VF migration recovery worker */
> > - struct work_struct worker;
> > - /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
> > - unsigned long gt_flags;
> > /**
> > * @migration.enabled: flag indicating if migration support
> > * was enabled or not due to missing prerequisites
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-24 19:50 ` Matthew Brost
@ 2025-09-24 20:21 ` Michal Wajdeczko
2025-09-24 20:35 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 20:21 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
On 9/24/2025 9:50 PM, Matthew Brost wrote:
> On Wed, Sep 24, 2025 at 12:49:25PM +0200, Michal Wajdeczko wrote:
>>
>>
>> On 9/24/2025 3:15 AM, Matthew Brost wrote:
>>> VF recovery is a per-GT operation, so it makes sense to isolate it to a
>>
>> that was also my suggestion to make it per-GT, good to see it happen now
>>
>
> +1
>
>>> per-GT queue. Scheduling this operation on the same worker as the GT
>>> reset and TDR not only aligns with this design but also helps avoid race
>>> conditions, as those operations can also modify the queue state.
>>
>> but while the recovery is per-GT, we should still protect against that
>> one GT starts the recovery sooner then other GTs notice about it
>>
>
> Yes. There is shared state in 2 places:
>
> - The GGTT shifting, this handled by [1] in my series.
> - CCS restore on iGPU (PTL) handled by [2] [3] in my series.
>
> [1] https://patchwork.freedesktop.org/patch/676394/?series=154627&rev=2
> [2] https://patchwork.freedesktop.org/patch/676393/?series=154627&rev=2
> [3] https://patchwork.freedesktop.org/patch/676397/?series=154627&rev=2
>
>>>
>>> v2:
>>> - Fix lockdep splat (Adam)
>>> - Use xe_sriov_vf_migration_supported helper
>>>
>>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>>> ---
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
>>> drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
>>> drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
>>> drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
>>> 6 files changed, 169 insertions(+), 258 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> index c9d0e32e7a15..cfb71b749e52 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> @@ -25,11 +25,15 @@
>>> #include "xe_guc.h"
>>> #include "xe_guc_hxg_helpers.h"
>>> #include "xe_guc_relay.h"
>>> +#include "xe_guc_submit.h"
>>> +#include "xe_irq.h"
>>> #include "xe_lrc.h"
>>> #include "xe_memirq.h"
>>> #include "xe_mmio.h"
>>> +#include "xe_pm.h"
>>> #include "xe_sriov.h"
>>> #include "xe_sriov_vf.h"
>>> +#include "xe_tile_sriov_vf.h"
>>> #include "xe_uc_fw.h"
>>> #include "xe_wopcm.h"
>>>
>>> @@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
>>> * Returns: 0 if the operation completed successfully, or a negative error
>>> * code otherwise.
>>> */
>>> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
>>> +static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
>>> {
>>> struct xe_guc *guc = >->uc.guc;
>>> int err;
>>> @@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
>>> * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
>>> * @gt: the &xe_gt struct instance
>>> */
>>> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
>>> +static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
>>> {
>>> struct xe_hw_engine *hwe;
>>> enum xe_hw_engine_id id;
>>> @@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
>>> xe_default_lrc_update_memirq_regs_with_address(hwe);
>>> }
>>>
>>> +static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
>>
>> nit: if this is static then could be just:
>>
>> vf_start_migration_recovery(gt)
>>
>
> Sure.
>
>>> +{
>>> + bool started;
>>> +
>>> + xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>> +
>>> + spin_lock(>->sriov.vf.migration.lock);
>>> +
>>> + if (!gt->sriov.vf.migration.recovery_queued) {
>>> + gt->sriov.vf.migration.recovery_queued = true;
>>> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
>>> +
>>> + started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
>>> + xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
>>> + "scheduled" : "already in progress");
>>
>> with this .recovery_queued flag, can we hit "already in progress" case?
>>
>
> This was existing code so kept it but I'm really unclear how we can get
I can't find "recovery_queued" flag in upstream code
> multiple resfix IRQs without the prior resfix flow being completed.
> Maybe Tomasz can clear this one up? Ideally I'd like to remove multiple
> resfix IRQ handled code if this is not possible.
>
>>> + }
>>> +
>>> + spin_unlock(>->sriov.vf.migration.lock);
>>> +}
>>> +
>>> /**
>>> * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
>>> * or just mark that a GuC is ready for it.
>>> @@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
>>> xe_gt_assert(gt, IS_SRIOV_VF(xe));
>>> xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
>>>
>>> - set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
>>> - /*
>>> - * We need to be certain that if all flags were set, at least one
>>> - * thread will notice that and schedule the recovery.
>>> - */
>>> - smp_mb__after_atomic();
>>> -
>>> xe_gt_sriov_info(gt, "ready for recovery after migration\n");
>>> - xe_sriov_vf_start_migration_recovery(xe);
>>> + xe_gt_sriov_vf_start_migration_recovery(gt);
>>> }
>>>
>>> static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
>>> @@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
>>> pf_version->major, pf_version->minor);
>>> }
>>>
>>> +static void vf_post_migration_shutdown(struct xe_gt *gt)
>>> +{
>>> + int ret = 0;
>>> +
>>> + spin_lock_irq(>->sriov.vf.migration.lock);
>>> + gt->sriov.vf.migration.recovery_queued = false;
>>> + spin_unlock_irq(>->sriov.vf.migration.lock);
>>> +
>>> + xe_guc_submit_pause(>->uc.guc);
>>> + ret |= xe_guc_submit_reset_block(>->uc.guc);
>>
>> this |= seem unneeded
>>
>
> This is existing code copy / pasted and is removed in [4].
>
> [4] https://patchwork.freedesktop.org/patch/676382/?series=154627&rev=2
>
>>> +
>>> + if (ret)
>>> + xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
>>
>> is this the only possible reason? maybe worth to add %pe ?
>>
>
> Again this existing code and will be removed in [4].
>
> So with above statements, I'd prefer just to leave as is.
>
>>> +}
>>> +
>>> +static size_t post_migration_scratch_size(struct xe_device *xe)
>>> +{
>>> + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
>>> +}
>>> +
>>> +static int vf_post_migration_fixups(struct xe_gt *gt)
>>> +{
>>> + s64 shift;
>>> + void *buf;
>>> + int err;
>>> +
>>> + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
>>> + if (!buf)
>>> + return -ENOMEM;
>>> +
>>> + err = xe_gt_sriov_vf_query_config(gt);
>>> + if (err)
>>> + goto out;
>>> +
>>> + shift = xe_gt_sriov_vf_ggtt_shift(gt);
>>> + if (shift) {
>>> + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
>>> + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
>>> + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
>>> + if (err)
>>> + goto out;
>>> + }
>>> +
>>> +out:
>>> + kfree(buf);
>>> + return err;
>>> +}
>>> +
>>> +static void vf_post_migration_kickstart(struct xe_gt *gt)
>>> +{
>>> + /*
>>> + * Make sure interrupts on the new HW are properly set. The GuC IRQ
>>> + * must be working at this point, since the recovery did started,
>>> + * but the rest was not enabled using the procedure from spec.
>>> + */
>>> + xe_irq_resume(gt_to_xe(gt));
>>> +
>>> + xe_guc_submit_reset_unblock(>->uc.guc);
>>> + xe_guc_submit_unpause(>->uc.guc);
>>> +}
>>> +
>>> +static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
>>> +{
>>> + bool skip_resfix = false;
>>> +
>>> + spin_lock_irq(>->sriov.vf.migration.lock);
>>> + if (gt->sriov.vf.migration.recovery_queued) {
>>> + skip_resfix = true;
>>> + xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
>>> + } else {
>>> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
>>> + }
>>> + spin_unlock_irq(>->sriov.vf.migration.lock);
>>> +
>>> + return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
>>
>> nit: this looks cleaner:
>>
>> if (skip_resfix)
>> return -EAGAIN;
>>
>> return xe_gt_sriov_vf_notify_resfix_done(gt);
>>
>
> Sure.
>
>>> +}
>>> +
>>> +static void vf_post_migration_recovery(struct xe_gt *gt)
>>> +{
>>> + struct xe_device *xe = gt_to_xe(gt);
>>> + int err;
>>> +
>>> + xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
>>> +
>>> + xe_pm_runtime_get(xe);
>>> + vf_post_migration_shutdown(gt);
>>> +
>>> + if (!xe_sriov_vf_migration_supported(xe)) {
>>> + xe_gt_sriov_err(gt, "migration is not supported\n");
>>> + err = -ENOTRECOVERABLE;
>>> + goto fail;
>>> + }
>>> +
>>> + err = vf_post_migration_fixups(gt);
>>> + if (err)
>>> + goto fail;
>>> +
>>> + vf_post_migration_kickstart(gt);
>>> + err = vf_post_migration_notify_resfix_done(gt);
>>> + if (err && err != -EAGAIN)
>>> + goto fail;
>>> +
>>> + xe_pm_runtime_put(xe);
>>> + xe_gt_sriov_notice(gt, "migration recovery ended\n");
>>> + return;
>>> +fail:
>>> + xe_pm_runtime_put(xe);
>>> + xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
>>> + xe_device_declare_wedged(xe);
>>> +}
>>> +
>>> +static void migration_worker_func(struct work_struct *w)
>>> +{
>>> + struct xe_gt *gt = container_of(w, struct xe_gt,
>>> + sriov.vf.migration.worker);
>>> +
>>> + vf_post_migration_recovery(gt);
>>> +}
>>> +
>>> +/**
>>> + * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
>>> + * @gt: the &xe_gt
>>> + */
>>> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
>>> +{
>>> + init_rwsem(>->sriov.vf.self_config.lock);
>>> + spin_lock_init(>->sriov.vf.migration.lock);
>>> + INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
>>> +
>>> + if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
>>> + xe_gt_sriov_info(gt, "migration not supported by this module version\n");
>>
>> we likely don't want to repeat that message on every GT
>>
>
> So move this to xe_sriov_vf_init_early?
hmm, maybe it is even already there
>
>>> +}
>>> +
>>> /**
>>> * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
>>> * @gt: the &xe_gt
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>> index bb5f8eace19b..2ac6775b52f0 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>> @@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
>>> int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
>>> int xe_gt_sriov_vf_connect(struct xe_gt *gt);
>>> int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
>>> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
>>> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
>>> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
>>>
>>> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
>>> bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
>>>
>>> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> index 7b10b8e1e10e..53680a2f188a 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>> @@ -8,6 +8,7 @@
>>>
>>> #include <linux/rwsem.h>
>>> #include <linux/types.h>
>>> +#include <linux/workqueue.h>
>>> #include "xe_uc_fw_types.h"
>>>
>>> /**
>>> @@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
>>> * xe_gt_sriov_vf_migration - VF migration data.
>>> */
>>> struct xe_gt_sriov_vf_migration {
>>> + /** @migration: VF migration recovery worker */
>>> + struct work_struct worker;
>>> + /** @lock: Protects recovery_queued */
>>> + spinlock_t lock;
>>> + /** @recovery_queued: VF post migration recovery in queued */
>>> + bool recovery_queued;
>>> /** @recovery_inprogress: VF post migration recovery in progress */
>>> bool recovery_inprogress;
>>> };
>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
>>> index da064a1e7419..7d91553c4acc 100644
>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
>>> @@ -6,21 +6,12 @@
>>> #include <drm/drm_debugfs.h>
>>> #include <drm/drm_managed.h>
>>>
>>> -#include "xe_assert.h"
>>> -#include "xe_device.h"
>>> #include "xe_gt.h"
>>> -#include "xe_gt_sriov_printk.h"
>>> #include "xe_gt_sriov_vf.h"
>>> #include "xe_guc.h"
>>> -#include "xe_guc_submit.h"
>>> -#include "xe_irq.h"
>>> -#include "xe_lrc.h"
>>> -#include "xe_pm.h"
>>> -#include "xe_sriov.h"
>>> #include "xe_sriov_printk.h"
>>> #include "xe_sriov_vf.h"
>>> #include "xe_sriov_vf_ccs.h"
>>> -#include "xe_tile_sriov_vf.h"
>>>
>>> /**
>>> * DOC: VF restore procedure in PF KMD and VF KMD
>>> @@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
>>> xe->sriov.vf.migration.enabled = false;
>>> }
>>>
>>> -static void migration_worker_func(struct work_struct *w);
>>> -
>>> static void vf_migration_init_early(struct xe_device *xe)
>>> {
>>> /*
>>> @@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
>>> guc_version.major, guc_version.minor);
>>> }
>>>
>>> - INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
>>> -
>>> xe->sriov.vf.migration.enabled = true;
>>> xe_sriov_dbg(xe, "migration support enabled\n");
>>> }
>>> @@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
>>> unsigned int id;
>>>
>>> for_each_gt(gt, xe, id)
>>> - init_rwsem(>->sriov.vf.self_config.lock);
>>> + xe_gt_sriov_vf_migration_init_early(gt);
>>
>> still, this should be called from gt_init_early kind of functions
>>
>
> Kinda a nit that I'm not convinced is worth while to have
> xe_sriov_vf_init_early and then xe_gt_sriov_vf_migration_init_early
> called in gt_init_early...
we have global xe level initialization, something like:
xe_device_init_early
xe_sriov_vf_init_early
xe_device_init
xe_sriov_vf_init
then per-gt initialization, that follows the native flow, something like
xe_gt_init_early
xe_gt_sriov_vf_init_early
xe_gt_sriov_vf_migration_init_early
xe_gt_init
xe_gt_sriov_vf_init
xe_gt_sriov_vf_migration_init
IMO we shouldn't jump in sriov code from xe level to gt level init on our own
>
> Matt
>
>>>
>>> vf_migration_init_early(xe);
>>> }
>>>
>>> -/**
>>> - * vf_post_migration_shutdown - Stop the driver activities after VF migration.
>>> - * @xe: the &xe_device struct instance
>>> - *
>>> - * After this VM is migrated and assigned to a new VF, it is running on a new
>>> - * hardware, and therefore many hardware-dependent states and related structures
>>> - * require fixups. Without fixups, the hardware cannot do any work, and therefore
>>> - * all GPU pipelines are stalled.
>>> - * Stop some of kernel activities to make the fixup process faster.
>>> - */
>>> -static void vf_post_migration_shutdown(struct xe_device *xe)
>>> -{
>>> - struct xe_gt *gt;
>>> - unsigned int id;
>>> - int ret = 0;
>>> -
>>> - for_each_gt(gt, xe, id) {
>>> - xe_guc_submit_pause(>->uc.guc);
>>> - ret |= xe_guc_submit_reset_block(>->uc.guc);
>>> - }
>>> -
>>> - if (ret)
>>> - drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
>>> -}
>>> -
>>> -/**
>>> - * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
>>> - * @xe: the &xe_device struct instance
>>> - *
>>> - * After we have finished with all post-migration fixups, restart the driver
>>> - * activities to continue feeding the GPU with workloads.
>>> - */
>>> -static void vf_post_migration_kickstart(struct xe_device *xe)
>>> -{
>>> - struct xe_gt *gt;
>>> - unsigned int id;
>>> -
>>> - /*
>>> - * Make sure interrupts on the new HW are properly set. The GuC IRQ
>>> - * must be working at this point, since the recovery did started,
>>> - * but the rest was not enabled using the procedure from spec.
>>> - */
>>> - xe_irq_resume(xe);
>>> -
>>> - for_each_gt(gt, xe, id) {
>>> - xe_guc_submit_reset_unblock(>->uc.guc);
>>> - xe_guc_submit_unpause(>->uc.guc);
>>> - }
>>> -}
>>> -
>>> -static bool gt_vf_post_migration_needed(struct xe_gt *gt)
>>> -{
>>> - return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
>>> -}
>>> -
>>> -/*
>>> - * Notify GuCs marked in flags about resource fixups apply finished.
>>> - * @xe: the &xe_device struct instance
>>> - * @gt_flags: flags marking to which GTs the notification shall be sent
>>> - */
>>> -static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
>>> -{
>>> - struct xe_gt *gt;
>>> - unsigned int id;
>>> - int err = 0;
>>> -
>>> - for_each_gt(gt, xe, id) {
>>> - if (!test_bit(id, >_flags))
>>> - continue;
>>> - /* skip asking GuC for RESFIX exit if new recovery request arrived */
>>> - if (gt_vf_post_migration_needed(gt))
>>> - continue;
>>> - err = xe_gt_sriov_vf_notify_resfix_done(gt);
>>> - if (err)
>>> - break;
>>> - clear_bit(id, >_flags);
>>> - }
>>> -
>>> - if (gt_flags && !err)
>>> - drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
>>> - return err;
>>> -}
>>> -
>>> -static int vf_get_next_migrated_gt_id(struct xe_device *xe)
>>> -{
>>> - struct xe_gt *gt;
>>> - unsigned int id;
>>> -
>>> - for_each_gt(gt, xe, id) {
>>> - if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
>>> - return id;
>>> - }
>>> - return -1;
>>> -}
>>> -
>>> -static size_t post_migration_scratch_size(struct xe_device *xe)
>>> -{
>>> - return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
>>> -}
>>> -
>>> -/**
>>> - * Perform post-migration fixups on a single GT.
>>> - *
>>> - * After migration, GuC needs to be re-queried for VF configuration to check
>>> - * if it matches previous provisioning. Most of VF provisioning shall be the
>>> - * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
>>> - * range has changed, we have to perform fixups - shift all GGTT references
>>> - * used anywhere within the driver. After the fixups in this function succeed,
>>> - * it is allowed to ask the GuC bound to this GT to continue normal operation.
>>> - *
>>> - * Returns: 0 if the operation completed successfully, or a negative error
>>> - * code otherwise.
>>> - */
>>> -static int gt_vf_post_migration_fixups(struct xe_gt *gt)
>>> -{
>>> - s64 shift;
>>> - void *buf;
>>> - int err;
>>> -
>>> - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
>>> - if (!buf)
>>> - return -ENOMEM;
>>> -
>>> - err = xe_gt_sriov_vf_query_config(gt);
>>> - if (err)
>>> - goto out;
>>> -
>>> - shift = xe_gt_sriov_vf_ggtt_shift(gt);
>>> - if (shift) {
>>> - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
>>> - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
>>> - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
>>> - if (err)
>>> - goto out;
>>> - }
>>> -
>>> -out:
>>> - kfree(buf);
>>> - return err;
>>> -}
>>> -
>>> -static void vf_post_migration_recovery(struct xe_device *xe)
>>> -{
>>> - unsigned long fixed_gts = 0;
>>> - int id, err;
>>> -
>>> - drm_dbg(&xe->drm, "migration recovery in progress\n");
>>> - xe_pm_runtime_get(xe);
>>> - vf_post_migration_shutdown(xe);
>>> -
>>> - if (!xe_sriov_vf_migration_supported(xe)) {
>>> - xe_sriov_err(xe, "migration is not supported\n");
>>> - err = -ENOTRECOVERABLE;
>>> - goto fail;
>>> - }
>>> -
>>> - while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
>>> - struct xe_gt *gt = xe_device_get_gt(xe, id);
>>> -
>>> - err = gt_vf_post_migration_fixups(gt);
>>> - if (err)
>>> - goto fail;
>>> -
>>> - set_bit(id, &fixed_gts);
>>> - }
>>> -
>>> - vf_post_migration_kickstart(xe);
>>> - err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
>>> - if (err)
>>> - goto fail;
>>> -
>>> - xe_pm_runtime_put(xe);
>>> - drm_notice(&xe->drm, "migration recovery ended\n");
>>> - return;
>>> -fail:
>>> - xe_pm_runtime_put(xe);
>>> - drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
>>> - xe_device_declare_wedged(xe);
>>> -}
>>> -
>>> -static void migration_worker_func(struct work_struct *w)
>>> -{
>>> - struct xe_device *xe = container_of(w, struct xe_device,
>>> - sriov.vf.migration.worker);
>>> -
>>> - vf_post_migration_recovery(xe);
>>> -}
>>> -
>>> -/*
>>> - * Check if post-restore recovery is coming on any of GTs.
>>> - * @xe: the &xe_device struct instance
>>> - *
>>> - * Return: True if migration recovery worker will soon be running. Any worker currently
>>> - * executing does not affect the result.
>>> - */
>>> -static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
>>> -{
>>> - struct xe_gt *gt;
>>> - unsigned int id;
>>> -
>>> - for_each_gt(gt, xe, id) {
>>> - if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
>>> - return true;
>>> - }
>>> - return false;
>>> -}
>>> -
>>> -/**
>>> - * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
>>> - * @xe: the &xe_device to start recovery on
>>> - *
>>> - * This function shall be called only by VF.
>>> - */
>>> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
>>> -{
>>> - bool started;
>>> -
>>> - xe_assert(xe, IS_SRIOV_VF(xe));
>>> -
>>> - if (!vf_ready_to_recovery_on_any_gts(xe))
>>> - return;
>>> -
>>> - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
>>> - drm_info(&xe->drm, "VF migration recovery %s\n", started ?
>>> - "scheduled" : "already in progress");
>>> -}
>>> -
>>> /**
>>> * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
>>> * @xe: the &xe_device to initialize
>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
>>> index 9e752105ec2a..4df95266b261 100644
>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
>>> @@ -13,7 +13,6 @@ struct xe_device;
>>>
>>> void xe_sriov_vf_init_early(struct xe_device *xe);
>>> int xe_sriov_vf_init_late(struct xe_device *xe);
>>> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
>>> bool xe_sriov_vf_migration_supported(struct xe_device *xe);
>>> void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
>>> index 426cc5841958..6a0fd0f5463e 100644
>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
>>> @@ -33,10 +33,6 @@ struct xe_device_vf {
>>>
>>> /** @migration: VF Migration state data */
>>> struct {
>>> - /** @migration.worker: VF migration recovery worker */
>>> - struct work_struct worker;
>>> - /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
>>> - unsigned long gt_flags;
>>> /**
>>> * @migration.enabled: flag indicating if migration support
>>> * was enabled or not due to missing prerequisites
>>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-24 20:21 ` Michal Wajdeczko
@ 2025-09-24 20:35 ` Matthew Brost
2025-09-25 16:27 ` Lis, Tomasz
0 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:35 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 10:21:32PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 9:50 PM, Matthew Brost wrote:
> > On Wed, Sep 24, 2025 at 12:49:25PM +0200, Michal Wajdeczko wrote:
> >>
> >>
> >> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> >>> VF recovery is a per-GT operation, so it makes sense to isolate it to a
> >>
> >> that was also my suggestion to make it per-GT, good to see it happen now
> >>
> >
> > +1
> >
> >>> per-GT queue. Scheduling this operation on the same worker as the GT
> >>> reset and TDR not only aligns with this design but also helps avoid race
> >>> conditions, as those operations can also modify the queue state.
> >>
> >> but while the recovery is per-GT, we should still protect against that
> >> one GT starts the recovery sooner then other GTs notice about it
> >>
> >
> > Yes. There is shared state in 2 places:
> >
> > - The GGTT shifting, this handled by [1] in my series.
> > - CCS restore on iGPU (PTL) handled by [2] [3] in my series.
> >
> > [1] https://patchwork.freedesktop.org/patch/676394/?series=154627&rev=2
> > [2] https://patchwork.freedesktop.org/patch/676393/?series=154627&rev=2
> > [3] https://patchwork.freedesktop.org/patch/676397/?series=154627&rev=2
> >
> >>>
> >>> v2:
> >>> - Fix lockdep splat (Adam)
> >>> - Use xe_sriov_vf_migration_supported helper
> >>>
> >>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> >>> ---
> >>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
> >>> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
> >>> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
> >>> drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
> >>> drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
> >>> drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
> >>> 6 files changed, 169 insertions(+), 258 deletions(-)
> >>>
> >>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> >>> index c9d0e32e7a15..cfb71b749e52 100644
> >>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> >>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> >>> @@ -25,11 +25,15 @@
> >>> #include "xe_guc.h"
> >>> #include "xe_guc_hxg_helpers.h"
> >>> #include "xe_guc_relay.h"
> >>> +#include "xe_guc_submit.h"
> >>> +#include "xe_irq.h"
> >>> #include "xe_lrc.h"
> >>> #include "xe_memirq.h"
> >>> #include "xe_mmio.h"
> >>> +#include "xe_pm.h"
> >>> #include "xe_sriov.h"
> >>> #include "xe_sriov_vf.h"
> >>> +#include "xe_tile_sriov_vf.h"
> >>> #include "xe_uc_fw.h"
> >>> #include "xe_wopcm.h"
> >>>
> >>> @@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
> >>> * Returns: 0 if the operation completed successfully, or a negative error
> >>> * code otherwise.
> >>> */
> >>> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> >>> +static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> >>> {
> >>> struct xe_guc *guc = >->uc.guc;
> >>> int err;
> >>> @@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
> >>> * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
> >>> * @gt: the &xe_gt struct instance
> >>> */
> >>> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> >>> +static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> >>> {
> >>> struct xe_hw_engine *hwe;
> >>> enum xe_hw_engine_id id;
> >>> @@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> >>> xe_default_lrc_update_memirq_regs_with_address(hwe);
> >>> }
> >>>
> >>> +static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
> >>
> >> nit: if this is static then could be just:
> >>
> >> vf_start_migration_recovery(gt)
> >>
> >
> > Sure.
> >
> >>> +{
> >>> + bool started;
> >>> +
> >>> + xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> >>> +
> >>> + spin_lock(>->sriov.vf.migration.lock);
> >>> +
> >>> + if (!gt->sriov.vf.migration.recovery_queued) {
> >>> + gt->sriov.vf.migration.recovery_queued = true;
> >>> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
> >>> +
> >>> + started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
> >>> + xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
> >>> + "scheduled" : "already in progress");
> >>
> >> with this .recovery_queued flag, can we hit "already in progress" case?
> >>
> >
> > This was existing code so kept it but I'm really unclear how we can get
>
> I can't find "recovery_queued" flag in upstream code
>
This is akin to comment in upstream code. Again, I'm not really
convienced this is something we need to handle as I suspect this is not
possible.
/* skip asking GuC for RESFIX exit if new recovery request arrived */
> > multiple resfix IRQs without the prior resfix flow being completed.
> > Maybe Tomasz can clear this one up? Ideally I'd like to remove multiple
> > resfix IRQ handled code if this is not possible.
> >
> >>> + }
> >>> +
> >>> + spin_unlock(>->sriov.vf.migration.lock);
> >>> +}
> >>> +
> >>> /**
> >>> * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
> >>> * or just mark that a GuC is ready for it.
> >>> @@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> >>> xe_gt_assert(gt, IS_SRIOV_VF(xe));
> >>> xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
> >>>
> >>> - set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> >>> - /*
> >>> - * We need to be certain that if all flags were set, at least one
> >>> - * thread will notice that and schedule the recovery.
> >>> - */
> >>> - smp_mb__after_atomic();
> >>> -
> >>> xe_gt_sriov_info(gt, "ready for recovery after migration\n");
> >>> - xe_sriov_vf_start_migration_recovery(xe);
> >>> + xe_gt_sriov_vf_start_migration_recovery(gt);
> >>> }
> >>>
> >>> static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
> >>> @@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> >>> pf_version->major, pf_version->minor);
> >>> }
> >>>
> >>> +static void vf_post_migration_shutdown(struct xe_gt *gt)
> >>> +{
> >>> + int ret = 0;
> >>> +
> >>> + spin_lock_irq(>->sriov.vf.migration.lock);
> >>> + gt->sriov.vf.migration.recovery_queued = false;
> >>> + spin_unlock_irq(>->sriov.vf.migration.lock);
> >>> +
> >>> + xe_guc_submit_pause(>->uc.guc);
> >>> + ret |= xe_guc_submit_reset_block(>->uc.guc);
> >>
> >> this |= seem unneeded
> >>
> >
> > This is existing code copy / pasted and is removed in [4].
> >
> > [4] https://patchwork.freedesktop.org/patch/676382/?series=154627&rev=2
> >
> >>> +
> >>> + if (ret)
> >>> + xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
> >>
> >> is this the only possible reason? maybe worth to add %pe ?
> >>
> >
> > Again this existing code and will be removed in [4].
> >
> > So with above statements, I'd prefer just to leave as is.
> >
> >>> +}
> >>> +
> >>> +static size_t post_migration_scratch_size(struct xe_device *xe)
> >>> +{
> >>> + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> >>> +}
> >>> +
> >>> +static int vf_post_migration_fixups(struct xe_gt *gt)
> >>> +{
> >>> + s64 shift;
> >>> + void *buf;
> >>> + int err;
> >>> +
> >>> + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
> >>> + if (!buf)
> >>> + return -ENOMEM;
> >>> +
> >>> + err = xe_gt_sriov_vf_query_config(gt);
> >>> + if (err)
> >>> + goto out;
> >>> +
> >>> + shift = xe_gt_sriov_vf_ggtt_shift(gt);
> >>> + if (shift) {
> >>> + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> >>> + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> >>> + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> >>> + if (err)
> >>> + goto out;
> >>> + }
> >>> +
> >>> +out:
> >>> + kfree(buf);
> >>> + return err;
> >>> +}
> >>> +
> >>> +static void vf_post_migration_kickstart(struct xe_gt *gt)
> >>> +{
> >>> + /*
> >>> + * Make sure interrupts on the new HW are properly set. The GuC IRQ
> >>> + * must be working at this point, since the recovery did started,
> >>> + * but the rest was not enabled using the procedure from spec.
> >>> + */
> >>> + xe_irq_resume(gt_to_xe(gt));
> >>> +
> >>> + xe_guc_submit_reset_unblock(>->uc.guc);
> >>> + xe_guc_submit_unpause(>->uc.guc);
> >>> +}
> >>> +
> >>> +static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> >>> +{
> >>> + bool skip_resfix = false;
> >>> +
> >>> + spin_lock_irq(>->sriov.vf.migration.lock);
> >>> + if (gt->sriov.vf.migration.recovery_queued) {
> >>> + skip_resfix = true;
> >>> + xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
> >>> + } else {
> >>> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
> >>> + }
> >>> + spin_unlock_irq(>->sriov.vf.migration.lock);
> >>> +
> >>> + return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
> >>
> >> nit: this looks cleaner:
> >>
> >> if (skip_resfix)
> >> return -EAGAIN;
> >>
> >> return xe_gt_sriov_vf_notify_resfix_done(gt);
> >>
> >
> > Sure.
> >
> >>> +}
> >>> +
> >>> +static void vf_post_migration_recovery(struct xe_gt *gt)
> >>> +{
> >>> + struct xe_device *xe = gt_to_xe(gt);
> >>> + int err;
> >>> +
> >>> + xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
> >>> +
> >>> + xe_pm_runtime_get(xe);
> >>> + vf_post_migration_shutdown(gt);
> >>> +
> >>> + if (!xe_sriov_vf_migration_supported(xe)) {
> >>> + xe_gt_sriov_err(gt, "migration is not supported\n");
> >>> + err = -ENOTRECOVERABLE;
> >>> + goto fail;
> >>> + }
> >>> +
> >>> + err = vf_post_migration_fixups(gt);
> >>> + if (err)
> >>> + goto fail;
> >>> +
> >>> + vf_post_migration_kickstart(gt);
> >>> + err = vf_post_migration_notify_resfix_done(gt);
> >>> + if (err && err != -EAGAIN)
> >>> + goto fail;
> >>> +
> >>> + xe_pm_runtime_put(xe);
> >>> + xe_gt_sriov_notice(gt, "migration recovery ended\n");
> >>> + return;
> >>> +fail:
> >>> + xe_pm_runtime_put(xe);
> >>> + xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
> >>> + xe_device_declare_wedged(xe);
> >>> +}
> >>> +
> >>> +static void migration_worker_func(struct work_struct *w)
> >>> +{
> >>> + struct xe_gt *gt = container_of(w, struct xe_gt,
> >>> + sriov.vf.migration.worker);
> >>> +
> >>> + vf_post_migration_recovery(gt);
> >>> +}
> >>> +
> >>> +/**
> >>> + * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> >>> + * @gt: the &xe_gt
> >>> + */
> >>> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> >>> +{
> >>> + init_rwsem(>->sriov.vf.self_config.lock);
> >>> + spin_lock_init(>->sriov.vf.migration.lock);
> >>> + INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
> >>> +
> >>> + if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> >>> + xe_gt_sriov_info(gt, "migration not supported by this module version\n");
> >>
> >> we likely don't want to repeat that message on every GT
> >>
> >
> > So move this to xe_sriov_vf_init_early?
>
> hmm, maybe it is even already there
>
Let me check on that. Either way will remove this from the GT layer.
> >
> >>> +}
> >>> +
> >>> /**
> >>> * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> >>> * @gt: the &xe_gt
> >>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> >>> index bb5f8eace19b..2ac6775b52f0 100644
> >>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> >>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> >>> @@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
> >>> int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
> >>> int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> >>> int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> >>> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> >>> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> >>> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> >>>
> >>> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> >>> bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> >>>
> >>> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> >>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> >>> index 7b10b8e1e10e..53680a2f188a 100644
> >>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> >>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> >>> @@ -8,6 +8,7 @@
> >>>
> >>> #include <linux/rwsem.h>
> >>> #include <linux/types.h>
> >>> +#include <linux/workqueue.h>
> >>> #include "xe_uc_fw_types.h"
> >>>
> >>> /**
> >>> @@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
> >>> * xe_gt_sriov_vf_migration - VF migration data.
> >>> */
> >>> struct xe_gt_sriov_vf_migration {
> >>> + /** @migration: VF migration recovery worker */
> >>> + struct work_struct worker;
> >>> + /** @lock: Protects recovery_queued */
> >>> + spinlock_t lock;
> >>> + /** @recovery_queued: VF post migration recovery in queued */
> >>> + bool recovery_queued;
> >>> /** @recovery_inprogress: VF post migration recovery in progress */
> >>> bool recovery_inprogress;
> >>> };
> >>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> >>> index da064a1e7419..7d91553c4acc 100644
> >>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> >>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> >>> @@ -6,21 +6,12 @@
> >>> #include <drm/drm_debugfs.h>
> >>> #include <drm/drm_managed.h>
> >>>
> >>> -#include "xe_assert.h"
> >>> -#include "xe_device.h"
> >>> #include "xe_gt.h"
> >>> -#include "xe_gt_sriov_printk.h"
> >>> #include "xe_gt_sriov_vf.h"
> >>> #include "xe_guc.h"
> >>> -#include "xe_guc_submit.h"
> >>> -#include "xe_irq.h"
> >>> -#include "xe_lrc.h"
> >>> -#include "xe_pm.h"
> >>> -#include "xe_sriov.h"
> >>> #include "xe_sriov_printk.h"
> >>> #include "xe_sriov_vf.h"
> >>> #include "xe_sriov_vf_ccs.h"
> >>> -#include "xe_tile_sriov_vf.h"
> >>>
> >>> /**
> >>> * DOC: VF restore procedure in PF KMD and VF KMD
> >>> @@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
> >>> xe->sriov.vf.migration.enabled = false;
> >>> }
> >>>
> >>> -static void migration_worker_func(struct work_struct *w);
> >>> -
> >>> static void vf_migration_init_early(struct xe_device *xe)
> >>> {
> >>> /*
> >>> @@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
> >>> guc_version.major, guc_version.minor);
> >>> }
> >>>
> >>> - INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
> >>> -
> >>> xe->sriov.vf.migration.enabled = true;
> >>> xe_sriov_dbg(xe, "migration support enabled\n");
> >>> }
> >>> @@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
> >>> unsigned int id;
> >>>
> >>> for_each_gt(gt, xe, id)
> >>> - init_rwsem(>->sriov.vf.self_config.lock);
> >>> + xe_gt_sriov_vf_migration_init_early(gt);
> >>
> >> still, this should be called from gt_init_early kind of functions
> >>
> >
> > Kinda a nit that I'm not convinced is worth while to have
> > xe_sriov_vf_init_early and then xe_gt_sriov_vf_migration_init_early
> > called in gt_init_early...
>
> we have global xe level initialization, something like:
>
> xe_device_init_early
> xe_sriov_vf_init_early
> xe_device_init
> xe_sriov_vf_init
>
> then per-gt initialization, that follows the native flow, something like
>
> xe_gt_init_early
> xe_gt_sriov_vf_init_early
> xe_gt_sriov_vf_migration_init_early
> xe_gt_init
> xe_gt_sriov_vf_init
> xe_gt_sriov_vf_migration_init
>
> IMO we shouldn't jump in sriov code from xe level to gt level init on our own
>
Ok, let me refactor this. Easy enough to do.
Matt
> >
> > Matt
> >
> >>>
> >>> vf_migration_init_early(xe);
> >>> }
> >>>
> >>> -/**
> >>> - * vf_post_migration_shutdown - Stop the driver activities after VF migration.
> >>> - * @xe: the &xe_device struct instance
> >>> - *
> >>> - * After this VM is migrated and assigned to a new VF, it is running on a new
> >>> - * hardware, and therefore many hardware-dependent states and related structures
> >>> - * require fixups. Without fixups, the hardware cannot do any work, and therefore
> >>> - * all GPU pipelines are stalled.
> >>> - * Stop some of kernel activities to make the fixup process faster.
> >>> - */
> >>> -static void vf_post_migration_shutdown(struct xe_device *xe)
> >>> -{
> >>> - struct xe_gt *gt;
> >>> - unsigned int id;
> >>> - int ret = 0;
> >>> -
> >>> - for_each_gt(gt, xe, id) {
> >>> - xe_guc_submit_pause(>->uc.guc);
> >>> - ret |= xe_guc_submit_reset_block(>->uc.guc);
> >>> - }
> >>> -
> >>> - if (ret)
> >>> - drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
> >>> -}
> >>> -
> >>> -/**
> >>> - * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
> >>> - * @xe: the &xe_device struct instance
> >>> - *
> >>> - * After we have finished with all post-migration fixups, restart the driver
> >>> - * activities to continue feeding the GPU with workloads.
> >>> - */
> >>> -static void vf_post_migration_kickstart(struct xe_device *xe)
> >>> -{
> >>> - struct xe_gt *gt;
> >>> - unsigned int id;
> >>> -
> >>> - /*
> >>> - * Make sure interrupts on the new HW are properly set. The GuC IRQ
> >>> - * must be working at this point, since the recovery did started,
> >>> - * but the rest was not enabled using the procedure from spec.
> >>> - */
> >>> - xe_irq_resume(xe);
> >>> -
> >>> - for_each_gt(gt, xe, id) {
> >>> - xe_guc_submit_reset_unblock(>->uc.guc);
> >>> - xe_guc_submit_unpause(>->uc.guc);
> >>> - }
> >>> -}
> >>> -
> >>> -static bool gt_vf_post_migration_needed(struct xe_gt *gt)
> >>> -{
> >>> - return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
> >>> -}
> >>> -
> >>> -/*
> >>> - * Notify GuCs marked in flags about resource fixups apply finished.
> >>> - * @xe: the &xe_device struct instance
> >>> - * @gt_flags: flags marking to which GTs the notification shall be sent
> >>> - */
> >>> -static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
> >>> -{
> >>> - struct xe_gt *gt;
> >>> - unsigned int id;
> >>> - int err = 0;
> >>> -
> >>> - for_each_gt(gt, xe, id) {
> >>> - if (!test_bit(id, >_flags))
> >>> - continue;
> >>> - /* skip asking GuC for RESFIX exit if new recovery request arrived */
> >>> - if (gt_vf_post_migration_needed(gt))
> >>> - continue;
> >>> - err = xe_gt_sriov_vf_notify_resfix_done(gt);
> >>> - if (err)
> >>> - break;
> >>> - clear_bit(id, >_flags);
> >>> - }
> >>> -
> >>> - if (gt_flags && !err)
> >>> - drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
> >>> - return err;
> >>> -}
> >>> -
> >>> -static int vf_get_next_migrated_gt_id(struct xe_device *xe)
> >>> -{
> >>> - struct xe_gt *gt;
> >>> - unsigned int id;
> >>> -
> >>> - for_each_gt(gt, xe, id) {
> >>> - if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
> >>> - return id;
> >>> - }
> >>> - return -1;
> >>> -}
> >>> -
> >>> -static size_t post_migration_scratch_size(struct xe_device *xe)
> >>> -{
> >>> - return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> >>> -}
> >>> -
> >>> -/**
> >>> - * Perform post-migration fixups on a single GT.
> >>> - *
> >>> - * After migration, GuC needs to be re-queried for VF configuration to check
> >>> - * if it matches previous provisioning. Most of VF provisioning shall be the
> >>> - * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
> >>> - * range has changed, we have to perform fixups - shift all GGTT references
> >>> - * used anywhere within the driver. After the fixups in this function succeed,
> >>> - * it is allowed to ask the GuC bound to this GT to continue normal operation.
> >>> - *
> >>> - * Returns: 0 if the operation completed successfully, or a negative error
> >>> - * code otherwise.
> >>> - */
> >>> -static int gt_vf_post_migration_fixups(struct xe_gt *gt)
> >>> -{
> >>> - s64 shift;
> >>> - void *buf;
> >>> - int err;
> >>> -
> >>> - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
> >>> - if (!buf)
> >>> - return -ENOMEM;
> >>> -
> >>> - err = xe_gt_sriov_vf_query_config(gt);
> >>> - if (err)
> >>> - goto out;
> >>> -
> >>> - shift = xe_gt_sriov_vf_ggtt_shift(gt);
> >>> - if (shift) {
> >>> - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> >>> - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> >>> - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> >>> - if (err)
> >>> - goto out;
> >>> - }
> >>> -
> >>> -out:
> >>> - kfree(buf);
> >>> - return err;
> >>> -}
> >>> -
> >>> -static void vf_post_migration_recovery(struct xe_device *xe)
> >>> -{
> >>> - unsigned long fixed_gts = 0;
> >>> - int id, err;
> >>> -
> >>> - drm_dbg(&xe->drm, "migration recovery in progress\n");
> >>> - xe_pm_runtime_get(xe);
> >>> - vf_post_migration_shutdown(xe);
> >>> -
> >>> - if (!xe_sriov_vf_migration_supported(xe)) {
> >>> - xe_sriov_err(xe, "migration is not supported\n");
> >>> - err = -ENOTRECOVERABLE;
> >>> - goto fail;
> >>> - }
> >>> -
> >>> - while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
> >>> - struct xe_gt *gt = xe_device_get_gt(xe, id);
> >>> -
> >>> - err = gt_vf_post_migration_fixups(gt);
> >>> - if (err)
> >>> - goto fail;
> >>> -
> >>> - set_bit(id, &fixed_gts);
> >>> - }
> >>> -
> >>> - vf_post_migration_kickstart(xe);
> >>> - err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
> >>> - if (err)
> >>> - goto fail;
> >>> -
> >>> - xe_pm_runtime_put(xe);
> >>> - drm_notice(&xe->drm, "migration recovery ended\n");
> >>> - return;
> >>> -fail:
> >>> - xe_pm_runtime_put(xe);
> >>> - drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
> >>> - xe_device_declare_wedged(xe);
> >>> -}
> >>> -
> >>> -static void migration_worker_func(struct work_struct *w)
> >>> -{
> >>> - struct xe_device *xe = container_of(w, struct xe_device,
> >>> - sriov.vf.migration.worker);
> >>> -
> >>> - vf_post_migration_recovery(xe);
> >>> -}
> >>> -
> >>> -/*
> >>> - * Check if post-restore recovery is coming on any of GTs.
> >>> - * @xe: the &xe_device struct instance
> >>> - *
> >>> - * Return: True if migration recovery worker will soon be running. Any worker currently
> >>> - * executing does not affect the result.
> >>> - */
> >>> -static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
> >>> -{
> >>> - struct xe_gt *gt;
> >>> - unsigned int id;
> >>> -
> >>> - for_each_gt(gt, xe, id) {
> >>> - if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
> >>> - return true;
> >>> - }
> >>> - return false;
> >>> -}
> >>> -
> >>> -/**
> >>> - * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
> >>> - * @xe: the &xe_device to start recovery on
> >>> - *
> >>> - * This function shall be called only by VF.
> >>> - */
> >>> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
> >>> -{
> >>> - bool started;
> >>> -
> >>> - xe_assert(xe, IS_SRIOV_VF(xe));
> >>> -
> >>> - if (!vf_ready_to_recovery_on_any_gts(xe))
> >>> - return;
> >>> -
> >>> - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
> >>> - drm_info(&xe->drm, "VF migration recovery %s\n", started ?
> >>> - "scheduled" : "already in progress");
> >>> -}
> >>> -
> >>> /**
> >>> * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
> >>> * @xe: the &xe_device to initialize
> >>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
> >>> index 9e752105ec2a..4df95266b261 100644
> >>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
> >>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
> >>> @@ -13,7 +13,6 @@ struct xe_device;
> >>>
> >>> void xe_sriov_vf_init_early(struct xe_device *xe);
> >>> int xe_sriov_vf_init_late(struct xe_device *xe);
> >>> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
> >>> bool xe_sriov_vf_migration_supported(struct xe_device *xe);
> >>> void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
> >>>
> >>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> >>> index 426cc5841958..6a0fd0f5463e 100644
> >>> --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> >>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> >>> @@ -33,10 +33,6 @@ struct xe_device_vf {
> >>>
> >>> /** @migration: VF Migration state data */
> >>> struct {
> >>> - /** @migration.worker: VF migration recovery worker */
> >>> - struct work_struct worker;
> >>> - /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
> >>> - unsigned long gt_flags;
> >>> /**
> >>> * @migration.enabled: flag indicating if migration support
> >>> * was enabled or not due to missing prerequisites
> >>
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-24 20:35 ` Matthew Brost
@ 2025-09-25 16:27 ` Lis, Tomasz
2025-09-25 16:56 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-25 16:27 UTC (permalink / raw)
To: Matthew Brost, Michal Wajdeczko; +Cc: intel-xe
On 9/24/2025 10:35 PM, Matthew Brost wrote:
> On Wed, Sep 24, 2025 at 10:21:32PM +0200, Michal Wajdeczko wrote:
>>
>> On 9/24/2025 9:50 PM, Matthew Brost wrote:
>>> On Wed, Sep 24, 2025 at 12:49:25PM +0200, Michal Wajdeczko wrote:
>>>>
>>>> On 9/24/2025 3:15 AM, Matthew Brost wrote:
>>>>> VF recovery is a per-GT operation, so it makes sense to isolate it to a
>>>> that was also my suggestion to make it per-GT, good to see it happen now
>>>>
>>> +1
>>>
>>>>> per-GT queue. Scheduling this operation on the same worker as the GT
>>>>> reset and TDR not only aligns with this design but also helps avoid race
>>>>> conditions, as those operations can also modify the queue state.
>>>> but while the recovery is per-GT, we should still protect against that
>>>> one GT starts the recovery sooner then other GTs notice about it
>>>>
>>> Yes. There is shared state in 2 places:
>>>
>>> - The GGTT shifting, this handled by [1] in my series.
>>> - CCS restore on iGPU (PTL) handled by [2] [3] in my series.
>>>
>>> [1] https://patchwork.freedesktop.org/patch/676394/?series=154627&rev=2
>>> [2] https://patchwork.freedesktop.org/patch/676393/?series=154627&rev=2
>>> [3] https://patchwork.freedesktop.org/patch/676397/?series=154627&rev=2
>>>
>>>>> v2:
>>>>> - Fix lockdep splat (Adam)
>>>>> - Use xe_sriov_vf_migration_supported helper
>>>>>
>>>>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>>>>> ---
>>>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
>>>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
>>>>> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
>>>>> drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
>>>>> drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
>>>>> drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
>>>>> 6 files changed, 169 insertions(+), 258 deletions(-)
>>>>>
>>>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>>>> index c9d0e32e7a15..cfb71b749e52 100644
>>>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>>>> @@ -25,11 +25,15 @@
>>>>> #include "xe_guc.h"
>>>>> #include "xe_guc_hxg_helpers.h"
>>>>> #include "xe_guc_relay.h"
>>>>> +#include "xe_guc_submit.h"
>>>>> +#include "xe_irq.h"
>>>>> #include "xe_lrc.h"
>>>>> #include "xe_memirq.h"
>>>>> #include "xe_mmio.h"
>>>>> +#include "xe_pm.h"
>>>>> #include "xe_sriov.h"
>>>>> #include "xe_sriov_vf.h"
>>>>> +#include "xe_tile_sriov_vf.h"
>>>>> #include "xe_uc_fw.h"
>>>>> #include "xe_wopcm.h"
>>>>>
>>>>> @@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
>>>>> * Returns: 0 if the operation completed successfully, or a negative error
>>>>> * code otherwise.
>>>>> */
>>>>> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
>>>>> +static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
>>>>> {
>>>>> struct xe_guc *guc = >->uc.guc;
>>>>> int err;
>>>>> @@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
>>>>> * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
>>>>> * @gt: the &xe_gt struct instance
>>>>> */
>>>>> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
>>>>> +static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
>>>>> {
>>>>> struct xe_hw_engine *hwe;
>>>>> enum xe_hw_engine_id id;
>>>>> @@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
>>>>> xe_default_lrc_update_memirq_regs_with_address(hwe);
>>>>> }
>>>>>
>>>>> +static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
>>>> nit: if this is static then could be just:
>>>>
>>>> vf_start_migration_recovery(gt)
>>>>
>>> Sure.
>>>
>>>>> +{
>>>>> + bool started;
>>>>> +
>>>>> + xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>>>>> +
>>>>> + spin_lock(>->sriov.vf.migration.lock);
>>>>> +
>>>>> + if (!gt->sriov.vf.migration.recovery_queued) {
>>>>> + gt->sriov.vf.migration.recovery_queued = true;
>>>>> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
>>>>> +
>>>>> + started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
>>>>> + xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
>>>>> + "scheduled" : "already in progress");
>>>> with this .recovery_queued flag, can we hit "already in progress" case?
>>>>
>>> This was existing code so kept it but I'm really unclear how we can get
>> I can't find "recovery_queued" flag in upstream code
>>
> This is akin to comment in upstream code. Again, I'm not really
> convienced this is something we need to handle as I suspect this is not
> possible.
>
> /* skip asking GuC for RESFIX exit if new recovery request arrived */
Second migration could happen while `recovery_queued` is set, I don't
see why not. Migration could happen at any point.
Interrupts tend to be treated with priority compared to workers, so if
migration will happen in that short period, this can and will be hit.
For the code near "/* skip asking GuC for RESFIX exit if new recovery
request arrived */" - while its days are numbered, let's leave this
method of double migration support for this series. We will later switch
to an improved double migration support, but that depends on GuC version
and should be developed and tested separately.
-Tomasz
>
>>> multiple resfix IRQs without the prior resfix flow being completed.
>>> Maybe Tomasz can clear this one up? Ideally I'd like to remove multiple
>>> resfix IRQ handled code if this is not possible.
>>>
>>>>> + }
>>>>> +
>>>>> + spin_unlock(>->sriov.vf.migration.lock);
>>>>> +}
>>>>> +
>>>>> /**
>>>>> * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
>>>>> * or just mark that a GuC is ready for it.
>>>>> @@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
>>>>> xe_gt_assert(gt, IS_SRIOV_VF(xe));
>>>>> xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
>>>>>
>>>>> - set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
>>>>> - /*
>>>>> - * We need to be certain that if all flags were set, at least one
>>>>> - * thread will notice that and schedule the recovery.
>>>>> - */
>>>>> - smp_mb__after_atomic();
>>>>> -
>>>>> xe_gt_sriov_info(gt, "ready for recovery after migration\n");
>>>>> - xe_sriov_vf_start_migration_recovery(xe);
>>>>> + xe_gt_sriov_vf_start_migration_recovery(gt);
>>>>> }
>>>>>
>>>>> static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
>>>>> @@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
>>>>> pf_version->major, pf_version->minor);
>>>>> }
>>>>>
>>>>> +static void vf_post_migration_shutdown(struct xe_gt *gt)
>>>>> +{
>>>>> + int ret = 0;
>>>>> +
>>>>> + spin_lock_irq(>->sriov.vf.migration.lock);
>>>>> + gt->sriov.vf.migration.recovery_queued = false;
>>>>> + spin_unlock_irq(>->sriov.vf.migration.lock);
>>>>> +
>>>>> + xe_guc_submit_pause(>->uc.guc);
>>>>> + ret |= xe_guc_submit_reset_block(>->uc.guc);
>>>> this |= seem unneeded
>>>>
>>> This is existing code copy / pasted and is removed in [4].
>>>
>>> [4] https://patchwork.freedesktop.org/patch/676382/?series=154627&rev=2
>>>
>>>>> +
>>>>> + if (ret)
>>>>> + xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
>>>> is this the only possible reason? maybe worth to add %pe ?
>>>>
>>> Again this existing code and will be removed in [4].
>>>
>>> So with above statements, I'd prefer just to leave as is.
>>>
>>>>> +}
>>>>> +
>>>>> +static size_t post_migration_scratch_size(struct xe_device *xe)
>>>>> +{
>>>>> + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
>>>>> +}
>>>>> +
>>>>> +static int vf_post_migration_fixups(struct xe_gt *gt)
>>>>> +{
>>>>> + s64 shift;
>>>>> + void *buf;
>>>>> + int err;
>>>>> +
>>>>> + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
>>>>> + if (!buf)
>>>>> + return -ENOMEM;
>>>>> +
>>>>> + err = xe_gt_sriov_vf_query_config(gt);
>>>>> + if (err)
>>>>> + goto out;
>>>>> +
>>>>> + shift = xe_gt_sriov_vf_ggtt_shift(gt);
>>>>> + if (shift) {
>>>>> + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
>>>>> + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
>>>>> + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
>>>>> + if (err)
>>>>> + goto out;
>>>>> + }
>>>>> +
>>>>> +out:
>>>>> + kfree(buf);
>>>>> + return err;
>>>>> +}
>>>>> +
>>>>> +static void vf_post_migration_kickstart(struct xe_gt *gt)
>>>>> +{
>>>>> + /*
>>>>> + * Make sure interrupts on the new HW are properly set. The GuC IRQ
>>>>> + * must be working at this point, since the recovery did started,
>>>>> + * but the rest was not enabled using the procedure from spec.
>>>>> + */
>>>>> + xe_irq_resume(gt_to_xe(gt));
>>>>> +
>>>>> + xe_guc_submit_reset_unblock(>->uc.guc);
>>>>> + xe_guc_submit_unpause(>->uc.guc);
>>>>> +}
>>>>> +
>>>>> +static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
>>>>> +{
>>>>> + bool skip_resfix = false;
>>>>> +
>>>>> + spin_lock_irq(>->sriov.vf.migration.lock);
>>>>> + if (gt->sriov.vf.migration.recovery_queued) {
>>>>> + skip_resfix = true;
>>>>> + xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
>>>>> + } else {
>>>>> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
>>>>> + }
>>>>> + spin_unlock_irq(>->sriov.vf.migration.lock);
>>>>> +
>>>>> + return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
>>>> nit: this looks cleaner:
>>>>
>>>> if (skip_resfix)
>>>> return -EAGAIN;
>>>>
>>>> return xe_gt_sriov_vf_notify_resfix_done(gt);
>>>>
>>> Sure.
>>>
>>>>> +}
>>>>> +
>>>>> +static void vf_post_migration_recovery(struct xe_gt *gt)
>>>>> +{
>>>>> + struct xe_device *xe = gt_to_xe(gt);
>>>>> + int err;
>>>>> +
>>>>> + xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
>>>>> +
>>>>> + xe_pm_runtime_get(xe);
>>>>> + vf_post_migration_shutdown(gt);
>>>>> +
>>>>> + if (!xe_sriov_vf_migration_supported(xe)) {
>>>>> + xe_gt_sriov_err(gt, "migration is not supported\n");
>>>>> + err = -ENOTRECOVERABLE;
>>>>> + goto fail;
>>>>> + }
>>>>> +
>>>>> + err = vf_post_migration_fixups(gt);
>>>>> + if (err)
>>>>> + goto fail;
>>>>> +
>>>>> + vf_post_migration_kickstart(gt);
>>>>> + err = vf_post_migration_notify_resfix_done(gt);
>>>>> + if (err && err != -EAGAIN)
>>>>> + goto fail;
>>>>> +
>>>>> + xe_pm_runtime_put(xe);
>>>>> + xe_gt_sriov_notice(gt, "migration recovery ended\n");
>>>>> + return;
>>>>> +fail:
>>>>> + xe_pm_runtime_put(xe);
>>>>> + xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
>>>>> + xe_device_declare_wedged(xe);
>>>>> +}
>>>>> +
>>>>> +static void migration_worker_func(struct work_struct *w)
>>>>> +{
>>>>> + struct xe_gt *gt = container_of(w, struct xe_gt,
>>>>> + sriov.vf.migration.worker);
>>>>> +
>>>>> + vf_post_migration_recovery(gt);
>>>>> +}
>>>>> +
>>>>> +/**
>>>>> + * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
>>>>> + * @gt: the &xe_gt
>>>>> + */
>>>>> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
>>>>> +{
>>>>> + init_rwsem(>->sriov.vf.self_config.lock);
>>>>> + spin_lock_init(>->sriov.vf.migration.lock);
>>>>> + INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
>>>>> +
>>>>> + if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
>>>>> + xe_gt_sriov_info(gt, "migration not supported by this module version\n");
>>>> we likely don't want to repeat that message on every GT
>>>>
>>> So move this to xe_sriov_vf_init_early?
>> hmm, maybe it is even already there
>>
> Let me check on that. Either way will remove this from the GT layer.
>
>>>>> +}
>>>>> +
>>>>> /**
>>>>> * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
>>>>> * @gt: the &xe_gt
>>>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>>>> index bb5f8eace19b..2ac6775b52f0 100644
>>>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
>>>>> @@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
>>>>> int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
>>>>> int xe_gt_sriov_vf_connect(struct xe_gt *gt);
>>>>> int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
>>>>> -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
>>>>> -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
>>>>> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
>>>>>
>>>>> +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
>>>>> bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
>>>>>
>>>>> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
>>>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>>>> index 7b10b8e1e10e..53680a2f188a 100644
>>>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
>>>>> @@ -8,6 +8,7 @@
>>>>>
>>>>> #include <linux/rwsem.h>
>>>>> #include <linux/types.h>
>>>>> +#include <linux/workqueue.h>
>>>>> #include "xe_uc_fw_types.h"
>>>>>
>>>>> /**
>>>>> @@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
>>>>> * xe_gt_sriov_vf_migration - VF migration data.
>>>>> */
>>>>> struct xe_gt_sriov_vf_migration {
>>>>> + /** @migration: VF migration recovery worker */
>>>>> + struct work_struct worker;
>>>>> + /** @lock: Protects recovery_queued */
>>>>> + spinlock_t lock;
>>>>> + /** @recovery_queued: VF post migration recovery in queued */
>>>>> + bool recovery_queued;
>>>>> /** @recovery_inprogress: VF post migration recovery in progress */
>>>>> bool recovery_inprogress;
>>>>> };
>>>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
>>>>> index da064a1e7419..7d91553c4acc 100644
>>>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
>>>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
>>>>> @@ -6,21 +6,12 @@
>>>>> #include <drm/drm_debugfs.h>
>>>>> #include <drm/drm_managed.h>
>>>>>
>>>>> -#include "xe_assert.h"
>>>>> -#include "xe_device.h"
>>>>> #include "xe_gt.h"
>>>>> -#include "xe_gt_sriov_printk.h"
>>>>> #include "xe_gt_sriov_vf.h"
>>>>> #include "xe_guc.h"
>>>>> -#include "xe_guc_submit.h"
>>>>> -#include "xe_irq.h"
>>>>> -#include "xe_lrc.h"
>>>>> -#include "xe_pm.h"
>>>>> -#include "xe_sriov.h"
>>>>> #include "xe_sriov_printk.h"
>>>>> #include "xe_sriov_vf.h"
>>>>> #include "xe_sriov_vf_ccs.h"
>>>>> -#include "xe_tile_sriov_vf.h"
>>>>>
>>>>> /**
>>>>> * DOC: VF restore procedure in PF KMD and VF KMD
>>>>> @@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
>>>>> xe->sriov.vf.migration.enabled = false;
>>>>> }
>>>>>
>>>>> -static void migration_worker_func(struct work_struct *w);
>>>>> -
>>>>> static void vf_migration_init_early(struct xe_device *xe)
>>>>> {
>>>>> /*
>>>>> @@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
>>>>> guc_version.major, guc_version.minor);
>>>>> }
>>>>>
>>>>> - INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
>>>>> -
>>>>> xe->sriov.vf.migration.enabled = true;
>>>>> xe_sriov_dbg(xe, "migration support enabled\n");
>>>>> }
>>>>> @@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
>>>>> unsigned int id;
>>>>>
>>>>> for_each_gt(gt, xe, id)
>>>>> - init_rwsem(>->sriov.vf.self_config.lock);
>>>>> + xe_gt_sriov_vf_migration_init_early(gt);
>>>> still, this should be called from gt_init_early kind of functions
>>>>
>>> Kinda a nit that I'm not convinced is worth while to have
>>> xe_sriov_vf_init_early and then xe_gt_sriov_vf_migration_init_early
>>> called in gt_init_early...
>> we have global xe level initialization, something like:
>>
>> xe_device_init_early
>> xe_sriov_vf_init_early
>> xe_device_init
>> xe_sriov_vf_init
>>
>> then per-gt initialization, that follows the native flow, something like
>>
>> xe_gt_init_early
>> xe_gt_sriov_vf_init_early
>> xe_gt_sriov_vf_migration_init_early
>> xe_gt_init
>> xe_gt_sriov_vf_init
>> xe_gt_sriov_vf_migration_init
>>
>> IMO we shouldn't jump in sriov code from xe level to gt level init on our own
>>
> Ok, let me refactor this. Easy enough to do.
>
> Matt
>
>>> Matt
>>>
>>>>>
>>>>> vf_migration_init_early(xe);
>>>>> }
>>>>>
>>>>> -/**
>>>>> - * vf_post_migration_shutdown - Stop the driver activities after VF migration.
>>>>> - * @xe: the &xe_device struct instance
>>>>> - *
>>>>> - * After this VM is migrated and assigned to a new VF, it is running on a new
>>>>> - * hardware, and therefore many hardware-dependent states and related structures
>>>>> - * require fixups. Without fixups, the hardware cannot do any work, and therefore
>>>>> - * all GPU pipelines are stalled.
>>>>> - * Stop some of kernel activities to make the fixup process faster.
>>>>> - */
>>>>> -static void vf_post_migration_shutdown(struct xe_device *xe)
>>>>> -{
>>>>> - struct xe_gt *gt;
>>>>> - unsigned int id;
>>>>> - int ret = 0;
>>>>> -
>>>>> - for_each_gt(gt, xe, id) {
>>>>> - xe_guc_submit_pause(>->uc.guc);
>>>>> - ret |= xe_guc_submit_reset_block(>->uc.guc);
>>>>> - }
>>>>> -
>>>>> - if (ret)
>>>>> - drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
>>>>> -}
>>>>> -
>>>>> -/**
>>>>> - * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
>>>>> - * @xe: the &xe_device struct instance
>>>>> - *
>>>>> - * After we have finished with all post-migration fixups, restart the driver
>>>>> - * activities to continue feeding the GPU with workloads.
>>>>> - */
>>>>> -static void vf_post_migration_kickstart(struct xe_device *xe)
>>>>> -{
>>>>> - struct xe_gt *gt;
>>>>> - unsigned int id;
>>>>> -
>>>>> - /*
>>>>> - * Make sure interrupts on the new HW are properly set. The GuC IRQ
>>>>> - * must be working at this point, since the recovery did started,
>>>>> - * but the rest was not enabled using the procedure from spec.
>>>>> - */
>>>>> - xe_irq_resume(xe);
>>>>> -
>>>>> - for_each_gt(gt, xe, id) {
>>>>> - xe_guc_submit_reset_unblock(>->uc.guc);
>>>>> - xe_guc_submit_unpause(>->uc.guc);
>>>>> - }
>>>>> -}
>>>>> -
>>>>> -static bool gt_vf_post_migration_needed(struct xe_gt *gt)
>>>>> -{
>>>>> - return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
>>>>> -}
>>>>> -
>>>>> -/*
>>>>> - * Notify GuCs marked in flags about resource fixups apply finished.
>>>>> - * @xe: the &xe_device struct instance
>>>>> - * @gt_flags: flags marking to which GTs the notification shall be sent
>>>>> - */
>>>>> -static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
>>>>> -{
>>>>> - struct xe_gt *gt;
>>>>> - unsigned int id;
>>>>> - int err = 0;
>>>>> -
>>>>> - for_each_gt(gt, xe, id) {
>>>>> - if (!test_bit(id, >_flags))
>>>>> - continue;
>>>>> - /* skip asking GuC for RESFIX exit if new recovery request arrived */
>>>>> - if (gt_vf_post_migration_needed(gt))
>>>>> - continue;
>>>>> - err = xe_gt_sriov_vf_notify_resfix_done(gt);
>>>>> - if (err)
>>>>> - break;
>>>>> - clear_bit(id, >_flags);
>>>>> - }
>>>>> -
>>>>> - if (gt_flags && !err)
>>>>> - drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
>>>>> - return err;
>>>>> -}
>>>>> -
>>>>> -static int vf_get_next_migrated_gt_id(struct xe_device *xe)
>>>>> -{
>>>>> - struct xe_gt *gt;
>>>>> - unsigned int id;
>>>>> -
>>>>> - for_each_gt(gt, xe, id) {
>>>>> - if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
>>>>> - return id;
>>>>> - }
>>>>> - return -1;
>>>>> -}
>>>>> -
>>>>> -static size_t post_migration_scratch_size(struct xe_device *xe)
>>>>> -{
>>>>> - return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
>>>>> -}
>>>>> -
>>>>> -/**
>>>>> - * Perform post-migration fixups on a single GT.
>>>>> - *
>>>>> - * After migration, GuC needs to be re-queried for VF configuration to check
>>>>> - * if it matches previous provisioning. Most of VF provisioning shall be the
>>>>> - * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
>>>>> - * range has changed, we have to perform fixups - shift all GGTT references
>>>>> - * used anywhere within the driver. After the fixups in this function succeed,
>>>>> - * it is allowed to ask the GuC bound to this GT to continue normal operation.
>>>>> - *
>>>>> - * Returns: 0 if the operation completed successfully, or a negative error
>>>>> - * code otherwise.
>>>>> - */
>>>>> -static int gt_vf_post_migration_fixups(struct xe_gt *gt)
>>>>> -{
>>>>> - s64 shift;
>>>>> - void *buf;
>>>>> - int err;
>>>>> -
>>>>> - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
>>>>> - if (!buf)
>>>>> - return -ENOMEM;
>>>>> -
>>>>> - err = xe_gt_sriov_vf_query_config(gt);
>>>>> - if (err)
>>>>> - goto out;
>>>>> -
>>>>> - shift = xe_gt_sriov_vf_ggtt_shift(gt);
>>>>> - if (shift) {
>>>>> - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
>>>>> - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
>>>>> - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
>>>>> - if (err)
>>>>> - goto out;
>>>>> - }
>>>>> -
>>>>> -out:
>>>>> - kfree(buf);
>>>>> - return err;
>>>>> -}
>>>>> -
>>>>> -static void vf_post_migration_recovery(struct xe_device *xe)
>>>>> -{
>>>>> - unsigned long fixed_gts = 0;
>>>>> - int id, err;
>>>>> -
>>>>> - drm_dbg(&xe->drm, "migration recovery in progress\n");
>>>>> - xe_pm_runtime_get(xe);
>>>>> - vf_post_migration_shutdown(xe);
>>>>> -
>>>>> - if (!xe_sriov_vf_migration_supported(xe)) {
>>>>> - xe_sriov_err(xe, "migration is not supported\n");
>>>>> - err = -ENOTRECOVERABLE;
>>>>> - goto fail;
>>>>> - }
>>>>> -
>>>>> - while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
>>>>> - struct xe_gt *gt = xe_device_get_gt(xe, id);
>>>>> -
>>>>> - err = gt_vf_post_migration_fixups(gt);
>>>>> - if (err)
>>>>> - goto fail;
>>>>> -
>>>>> - set_bit(id, &fixed_gts);
>>>>> - }
>>>>> -
>>>>> - vf_post_migration_kickstart(xe);
>>>>> - err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
>>>>> - if (err)
>>>>> - goto fail;
>>>>> -
>>>>> - xe_pm_runtime_put(xe);
>>>>> - drm_notice(&xe->drm, "migration recovery ended\n");
>>>>> - return;
>>>>> -fail:
>>>>> - xe_pm_runtime_put(xe);
>>>>> - drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
>>>>> - xe_device_declare_wedged(xe);
>>>>> -}
>>>>> -
>>>>> -static void migration_worker_func(struct work_struct *w)
>>>>> -{
>>>>> - struct xe_device *xe = container_of(w, struct xe_device,
>>>>> - sriov.vf.migration.worker);
>>>>> -
>>>>> - vf_post_migration_recovery(xe);
>>>>> -}
>>>>> -
>>>>> -/*
>>>>> - * Check if post-restore recovery is coming on any of GTs.
>>>>> - * @xe: the &xe_device struct instance
>>>>> - *
>>>>> - * Return: True if migration recovery worker will soon be running. Any worker currently
>>>>> - * executing does not affect the result.
>>>>> - */
>>>>> -static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
>>>>> -{
>>>>> - struct xe_gt *gt;
>>>>> - unsigned int id;
>>>>> -
>>>>> - for_each_gt(gt, xe, id) {
>>>>> - if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
>>>>> - return true;
>>>>> - }
>>>>> - return false;
>>>>> -}
>>>>> -
>>>>> -/**
>>>>> - * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
>>>>> - * @xe: the &xe_device to start recovery on
>>>>> - *
>>>>> - * This function shall be called only by VF.
>>>>> - */
>>>>> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
>>>>> -{
>>>>> - bool started;
>>>>> -
>>>>> - xe_assert(xe, IS_SRIOV_VF(xe));
>>>>> -
>>>>> - if (!vf_ready_to_recovery_on_any_gts(xe))
>>>>> - return;
>>>>> -
>>>>> - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
>>>>> - drm_info(&xe->drm, "VF migration recovery %s\n", started ?
>>>>> - "scheduled" : "already in progress");
>>>>> -}
>>>>> -
>>>>> /**
>>>>> * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
>>>>> * @xe: the &xe_device to initialize
>>>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
>>>>> index 9e752105ec2a..4df95266b261 100644
>>>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
>>>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
>>>>> @@ -13,7 +13,6 @@ struct xe_device;
>>>>>
>>>>> void xe_sriov_vf_init_early(struct xe_device *xe);
>>>>> int xe_sriov_vf_init_late(struct xe_device *xe);
>>>>> -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
>>>>> bool xe_sriov_vf_migration_supported(struct xe_device *xe);
>>>>> void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
>>>>>
>>>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
>>>>> index 426cc5841958..6a0fd0f5463e 100644
>>>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
>>>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
>>>>> @@ -33,10 +33,6 @@ struct xe_device_vf {
>>>>>
>>>>> /** @migration: VF Migration state data */
>>>>> struct {
>>>>> - /** @migration.worker: VF migration recovery worker */
>>>>> - struct work_struct worker;
>>>>> - /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
>>>>> - unsigned long gt_flags;
>>>>> /**
>>>>> * @migration.enabled: flag indicating if migration support
>>>>> * was enabled or not due to missing prerequisites
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker
2025-09-25 16:27 ` Lis, Tomasz
@ 2025-09-25 16:56 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-25 16:56 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: Michal Wajdeczko, intel-xe
On Thu, Sep 25, 2025 at 06:27:55PM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 10:35 PM, Matthew Brost wrote:
> > On Wed, Sep 24, 2025 at 10:21:32PM +0200, Michal Wajdeczko wrote:
> > >
> > > On 9/24/2025 9:50 PM, Matthew Brost wrote:
> > > > On Wed, Sep 24, 2025 at 12:49:25PM +0200, Michal Wajdeczko wrote:
> > > > >
> > > > > On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > > > > > VF recovery is a per-GT operation, so it makes sense to isolate it to a
> > > > > that was also my suggestion to make it per-GT, good to see it happen now
> > > > >
> > > > +1
> > > >
> > > > > > per-GT queue. Scheduling this operation on the same worker as the GT
> > > > > > reset and TDR not only aligns with this design but also helps avoid race
> > > > > > conditions, as those operations can also modify the queue state.
> > > > > but while the recovery is per-GT, we should still protect against that
> > > > > one GT starts the recovery sooner then other GTs notice about it
> > > > >
> > > > Yes. There is shared state in 2 places:
> > > >
> > > > - The GGTT shifting, this handled by [1] in my series.
> > > > - CCS restore on iGPU (PTL) handled by [2] [3] in my series.
> > > >
> > > > [1] https://patchwork.freedesktop.org/patch/676394/?series=154627&rev=2
> > > > [2] https://patchwork.freedesktop.org/patch/676393/?series=154627&rev=2
> > > > [3] https://patchwork.freedesktop.org/patch/676397/?series=154627&rev=2
> > > >
> > > > > > v2:
> > > > > > - Fix lockdep splat (Adam)
> > > > > > - Use xe_sriov_vf_migration_supported helper
> > > > > >
> > > > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > > > ---
> > > > > > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 170 ++++++++++++++-
> > > > > > drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
> > > > > > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 7 +
> > > > > > drivers/gpu/drm/xe/xe_sriov_vf.c | 242 +---------------------
> > > > > > drivers/gpu/drm/xe/xe_sriov_vf.h | 1 -
> > > > > > drivers/gpu/drm/xe/xe_sriov_vf_types.h | 4 -
> > > > > > 6 files changed, 169 insertions(+), 258 deletions(-)
> > > > > >
> > > > > > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > > > > > index c9d0e32e7a15..cfb71b749e52 100644
> > > > > > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > > > > > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > > > > > @@ -25,11 +25,15 @@
> > > > > > #include "xe_guc.h"
> > > > > > #include "xe_guc_hxg_helpers.h"
> > > > > > #include "xe_guc_relay.h"
> > > > > > +#include "xe_guc_submit.h"
> > > > > > +#include "xe_irq.h"
> > > > > > #include "xe_lrc.h"
> > > > > > #include "xe_memirq.h"
> > > > > > #include "xe_mmio.h"
> > > > > > +#include "xe_pm.h"
> > > > > > #include "xe_sriov.h"
> > > > > > #include "xe_sriov_vf.h"
> > > > > > +#include "xe_tile_sriov_vf.h"
> > > > > > #include "xe_uc_fw.h"
> > > > > > #include "xe_wopcm.h"
> > > > > > @@ -314,7 +318,7 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
> > > > > > * Returns: 0 if the operation completed successfully, or a negative error
> > > > > > * code otherwise.
> > > > > > */
> > > > > > -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> > > > > > +static int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt)
> > > > > > {
> > > > > > struct xe_guc *guc = >->uc.guc;
> > > > > > int err;
> > > > > > @@ -808,7 +812,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt)
> > > > > > * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
> > > > > > * @gt: the &xe_gt struct instance
> > > > > > */
> > > > > > -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> > > > > > +static void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> > > > > > {
> > > > > > struct xe_hw_engine *hwe;
> > > > > > enum xe_hw_engine_id id;
> > > > > > @@ -817,6 +821,26 @@ void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
> > > > > > xe_default_lrc_update_memirq_regs_with_address(hwe);
> > > > > > }
> > > > > > +static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
> > > > > nit: if this is static then could be just:
> > > > >
> > > > > vf_start_migration_recovery(gt)
> > > > >
> > > > Sure.
> > > >
> > > > > > +{
> > > > > > + bool started;
> > > > > > +
> > > > > > + xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > > > > > +
> > > > > > + spin_lock(>->sriov.vf.migration.lock);
> > > > > > +
> > > > > > + if (!gt->sriov.vf.migration.recovery_queued) {
> > > > > > + gt->sriov.vf.migration.recovery_queued = true;
> > > > > > + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
> > > > > > +
> > > > > > + started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
> > > > > > + xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
> > > > > > + "scheduled" : "already in progress");
> > > > > with this .recovery_queued flag, can we hit "already in progress" case?
> > > > >
> > > > This was existing code so kept it but I'm really unclear how we can get
> > > I can't find "recovery_queued" flag in upstream code
> > >
> > This is akin to comment in upstream code. Again, I'm not really
> > convienced this is something we need to handle as I suspect this is not
> > possible.
> >
> > /* skip asking GuC for RESFIX exit if new recovery request arrived */
>
> Second migration could happen while `recovery_queued` is set, I don't see
> why not. Migration could happen at any point.
>
> Interrupts tend to be treated with priority compared to workers, so if
> migration will happen in that short period, this can and will be hit.
>
I was made aware how this can occur late yesterday.
> For the code near "/* skip asking GuC for RESFIX exit if new recovery
> request arrived */" - while its days are numbered, let's leave this method
> of double migration support for this series. We will later switch to an
Ok, 'migration.recovery_queued' implements the same functionality as in
the upstream code.
> improved double migration support, but that depends on GuC version and
> should be developed and tested separately.
>
Yes, I am aware of this upcoming change too. I agree this is better
solution as 'migration.recovery_queued' or upstream version which does
the same thing tries hard to not race but it is basically impossible to
seal that race without ridiculous locking which I chose not to implement
in this series. I suggest we leave 'migration.recovery_queued' as in
this series and remove once proper GuC support lands.
Matt
> -Tomasz
>
> >
> > > > multiple resfix IRQs without the prior resfix flow being completed.
> > > > Maybe Tomasz can clear this one up? Ideally I'd like to remove multiple
> > > > resfix IRQ handled code if this is not possible.
> > > >
> > > > > > + }
> > > > > > +
> > > > > > + spin_unlock(>->sriov.vf.migration.lock);
> > > > > > +}
> > > > > > +
> > > > > > /**
> > > > > > * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
> > > > > > * or just mark that a GuC is ready for it.
> > > > > > @@ -831,15 +855,8 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt)
> > > > > > xe_gt_assert(gt, IS_SRIOV_VF(xe));
> > > > > > xe_gt_assert(gt, xe_gt_sriov_vf_recovery_inprogress(gt));
> > > > > > - set_bit(gt->info.id, &xe->sriov.vf.migration.gt_flags);
> > > > > > - /*
> > > > > > - * We need to be certain that if all flags were set, at least one
> > > > > > - * thread will notice that and schedule the recovery.
> > > > > > - */
> > > > > > - smp_mb__after_atomic();
> > > > > > -
> > > > > > xe_gt_sriov_info(gt, "ready for recovery after migration\n");
> > > > > > - xe_sriov_vf_start_migration_recovery(xe);
> > > > > > + xe_gt_sriov_vf_start_migration_recovery(gt);
> > > > > > }
> > > > > > static bool vf_is_negotiated(struct xe_gt *gt, u16 major, u16 minor)
> > > > > > @@ -1175,6 +1192,139 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> > > > > > pf_version->major, pf_version->minor);
> > > > > > }
> > > > > > +static void vf_post_migration_shutdown(struct xe_gt *gt)
> > > > > > +{
> > > > > > + int ret = 0;
> > > > > > +
> > > > > > + spin_lock_irq(>->sriov.vf.migration.lock);
> > > > > > + gt->sriov.vf.migration.recovery_queued = false;
> > > > > > + spin_unlock_irq(>->sriov.vf.migration.lock);
> > > > > > +
> > > > > > + xe_guc_submit_pause(>->uc.guc);
> > > > > > + ret |= xe_guc_submit_reset_block(>->uc.guc);
> > > > > this |= seem unneeded
> > > > >
> > > > This is existing code copy / pasted and is removed in [4].
> > > >
> > > > [4] https://patchwork.freedesktop.org/patch/676382/?series=154627&rev=2
> > > >
> > > > > > +
> > > > > > + if (ret)
> > > > > > + xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
> > > > > is this the only possible reason? maybe worth to add %pe ?
> > > > >
> > > > Again this existing code and will be removed in [4].
> > > >
> > > > So with above statements, I'd prefer just to leave as is.
> > > >
> > > > > > +}
> > > > > > +
> > > > > > +static size_t post_migration_scratch_size(struct xe_device *xe)
> > > > > > +{
> > > > > > + return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> > > > > > +}
> > > > > > +
> > > > > > +static int vf_post_migration_fixups(struct xe_gt *gt)
> > > > > > +{
> > > > > > + s64 shift;
> > > > > > + void *buf;
> > > > > > + int err;
> > > > > > +
> > > > > > + buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
> > > > > > + if (!buf)
> > > > > > + return -ENOMEM;
> > > > > > +
> > > > > > + err = xe_gt_sriov_vf_query_config(gt);
> > > > > > + if (err)
> > > > > > + goto out;
> > > > > > +
> > > > > > + shift = xe_gt_sriov_vf_ggtt_shift(gt);
> > > > > > + if (shift) {
> > > > > > + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> > > > > > + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > > > > > + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > > > > > + if (err)
> > > > > > + goto out;
> > > > > > + }
> > > > > > +
> > > > > > +out:
> > > > > > + kfree(buf);
> > > > > > + return err;
> > > > > > +}
> > > > > > +
> > > > > > +static void vf_post_migration_kickstart(struct xe_gt *gt)
> > > > > > +{
> > > > > > + /*
> > > > > > + * Make sure interrupts on the new HW are properly set. The GuC IRQ
> > > > > > + * must be working at this point, since the recovery did started,
> > > > > > + * but the rest was not enabled using the procedure from spec.
> > > > > > + */
> > > > > > + xe_irq_resume(gt_to_xe(gt));
> > > > > > +
> > > > > > + xe_guc_submit_reset_unblock(>->uc.guc);
> > > > > > + xe_guc_submit_unpause(>->uc.guc);
> > > > > > +}
> > > > > > +
> > > > > > +static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> > > > > > +{
> > > > > > + bool skip_resfix = false;
> > > > > > +
> > > > > > + spin_lock_irq(>->sriov.vf.migration.lock);
> > > > > > + if (gt->sriov.vf.migration.recovery_queued) {
> > > > > > + skip_resfix = true;
> > > > > > + xe_gt_sriov_dbg(gt, "another recovery imminent, skipped some notifications\n");
> > > > > > + } else {
> > > > > > + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
> > > > > > + }
> > > > > > + spin_unlock_irq(>->sriov.vf.migration.lock);
> > > > > > +
> > > > > > + return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
> > > > > nit: this looks cleaner:
> > > > >
> > > > > if (skip_resfix)
> > > > > return -EAGAIN;
> > > > >
> > > > > return xe_gt_sriov_vf_notify_resfix_done(gt);
> > > > >
> > > > Sure.
> > > >
> > > > > > +}
> > > > > > +
> > > > > > +static void vf_post_migration_recovery(struct xe_gt *gt)
> > > > > > +{
> > > > > > + struct xe_device *xe = gt_to_xe(gt);
> > > > > > + int err;
> > > > > > +
> > > > > > + xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
> > > > > > +
> > > > > > + xe_pm_runtime_get(xe);
> > > > > > + vf_post_migration_shutdown(gt);
> > > > > > +
> > > > > > + if (!xe_sriov_vf_migration_supported(xe)) {
> > > > > > + xe_gt_sriov_err(gt, "migration is not supported\n");
> > > > > > + err = -ENOTRECOVERABLE;
> > > > > > + goto fail;
> > > > > > + }
> > > > > > +
> > > > > > + err = vf_post_migration_fixups(gt);
> > > > > > + if (err)
> > > > > > + goto fail;
> > > > > > +
> > > > > > + vf_post_migration_kickstart(gt);
> > > > > > + err = vf_post_migration_notify_resfix_done(gt);
> > > > > > + if (err && err != -EAGAIN)
> > > > > > + goto fail;
> > > > > > +
> > > > > > + xe_pm_runtime_put(xe);
> > > > > > + xe_gt_sriov_notice(gt, "migration recovery ended\n");
> > > > > > + return;
> > > > > > +fail:
> > > > > > + xe_pm_runtime_put(xe);
> > > > > > + xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
> > > > > > + xe_device_declare_wedged(xe);
> > > > > > +}
> > > > > > +
> > > > > > +static void migration_worker_func(struct work_struct *w)
> > > > > > +{
> > > > > > + struct xe_gt *gt = container_of(w, struct xe_gt,
> > > > > > + sriov.vf.migration.worker);
> > > > > > +
> > > > > > + vf_post_migration_recovery(gt);
> > > > > > +}
> > > > > > +
> > > > > > +/**
> > > > > > + * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> > > > > > + * @gt: the &xe_gt
> > > > > > + */
> > > > > > +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > > > > > +{
> > > > > > + init_rwsem(>->sriov.vf.self_config.lock);
> > > > > > + spin_lock_init(>->sriov.vf.migration.lock);
> > > > > > + INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
> > > > > > +
> > > > > > + if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> > > > > > + xe_gt_sriov_info(gt, "migration not supported by this module version\n");
> > > > > we likely don't want to repeat that message on every GT
> > > > >
> > > > So move this to xe_sriov_vf_init_early?
> > > hmm, maybe it is even already there
> > >
> > Let me check on that. Either way will remove this from the GT layer.
> >
> > > > > > +}
> > > > > > +
> > > > > > /**
> > > > > > * xe_gt_sriov_vf_recovery_inprogress() - VF post migration recovery in progress
> > > > > > * @gt: the &xe_gt
> > > > > > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > > > > > index bb5f8eace19b..2ac6775b52f0 100644
> > > > > > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > > > > > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > > > > > @@ -21,10 +21,9 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
> > > > > > int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
> > > > > > int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> > > > > > int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> > > > > > -void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
> > > > > > -int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
> > > > > > void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> > > > > > +void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> > > > > > bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> > > > > > u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> > > > > > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > > > > > index 7b10b8e1e10e..53680a2f188a 100644
> > > > > > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > > > > > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > > > > > @@ -8,6 +8,7 @@
> > > > > > #include <linux/rwsem.h>
> > > > > > #include <linux/types.h>
> > > > > > +#include <linux/workqueue.h>
> > > > > > #include "xe_uc_fw_types.h"
> > > > > > /**
> > > > > > @@ -53,6 +54,12 @@ struct xe_gt_sriov_vf_runtime {
> > > > > > * xe_gt_sriov_vf_migration - VF migration data.
> > > > > > */
> > > > > > struct xe_gt_sriov_vf_migration {
> > > > > > + /** @migration: VF migration recovery worker */
> > > > > > + struct work_struct worker;
> > > > > > + /** @lock: Protects recovery_queued */
> > > > > > + spinlock_t lock;
> > > > > > + /** @recovery_queued: VF post migration recovery in queued */
> > > > > > + bool recovery_queued;
> > > > > > /** @recovery_inprogress: VF post migration recovery in progress */
> > > > > > bool recovery_inprogress;
> > > > > > };
> > > > > > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > > > > > index da064a1e7419..7d91553c4acc 100644
> > > > > > --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> > > > > > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > > > > > @@ -6,21 +6,12 @@
> > > > > > #include <drm/drm_debugfs.h>
> > > > > > #include <drm/drm_managed.h>
> > > > > > -#include "xe_assert.h"
> > > > > > -#include "xe_device.h"
> > > > > > #include "xe_gt.h"
> > > > > > -#include "xe_gt_sriov_printk.h"
> > > > > > #include "xe_gt_sriov_vf.h"
> > > > > > #include "xe_guc.h"
> > > > > > -#include "xe_guc_submit.h"
> > > > > > -#include "xe_irq.h"
> > > > > > -#include "xe_lrc.h"
> > > > > > -#include "xe_pm.h"
> > > > > > -#include "xe_sriov.h"
> > > > > > #include "xe_sriov_printk.h"
> > > > > > #include "xe_sriov_vf.h"
> > > > > > #include "xe_sriov_vf_ccs.h"
> > > > > > -#include "xe_tile_sriov_vf.h"
> > > > > > /**
> > > > > > * DOC: VF restore procedure in PF KMD and VF KMD
> > > > > > @@ -158,8 +149,6 @@ static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
> > > > > > xe->sriov.vf.migration.enabled = false;
> > > > > > }
> > > > > > -static void migration_worker_func(struct work_struct *w);
> > > > > > -
> > > > > > static void vf_migration_init_early(struct xe_device *xe)
> > > > > > {
> > > > > > /*
> > > > > > @@ -184,8 +173,6 @@ static void vf_migration_init_early(struct xe_device *xe)
> > > > > > guc_version.major, guc_version.minor);
> > > > > > }
> > > > > > - INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
> > > > > > -
> > > > > > xe->sriov.vf.migration.enabled = true;
> > > > > > xe_sriov_dbg(xe, "migration support enabled\n");
> > > > > > }
> > > > > > @@ -200,238 +187,11 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
> > > > > > unsigned int id;
> > > > > > for_each_gt(gt, xe, id)
> > > > > > - init_rwsem(>->sriov.vf.self_config.lock);
> > > > > > + xe_gt_sriov_vf_migration_init_early(gt);
> > > > > still, this should be called from gt_init_early kind of functions
> > > > >
> > > > Kinda a nit that I'm not convinced is worth while to have
> > > > xe_sriov_vf_init_early and then xe_gt_sriov_vf_migration_init_early
> > > > called in gt_init_early...
> > > we have global xe level initialization, something like:
> > >
> > > xe_device_init_early
> > > xe_sriov_vf_init_early
> > > xe_device_init
> > > xe_sriov_vf_init
> > >
> > > then per-gt initialization, that follows the native flow, something like
> > >
> > > xe_gt_init_early
> > > xe_gt_sriov_vf_init_early
> > > xe_gt_sriov_vf_migration_init_early
> > > xe_gt_init
> > > xe_gt_sriov_vf_init
> > > xe_gt_sriov_vf_migration_init
> > >
> > > IMO we shouldn't jump in sriov code from xe level to gt level init on our own
> > >
> > Ok, let me refactor this. Easy enough to do.
> >
> > Matt
> >
> > > > Matt
> > > >
> > > > > > vf_migration_init_early(xe);
> > > > > > }
> > > > > > -/**
> > > > > > - * vf_post_migration_shutdown - Stop the driver activities after VF migration.
> > > > > > - * @xe: the &xe_device struct instance
> > > > > > - *
> > > > > > - * After this VM is migrated and assigned to a new VF, it is running on a new
> > > > > > - * hardware, and therefore many hardware-dependent states and related structures
> > > > > > - * require fixups. Without fixups, the hardware cannot do any work, and therefore
> > > > > > - * all GPU pipelines are stalled.
> > > > > > - * Stop some of kernel activities to make the fixup process faster.
> > > > > > - */
> > > > > > -static void vf_post_migration_shutdown(struct xe_device *xe)
> > > > > > -{
> > > > > > - struct xe_gt *gt;
> > > > > > - unsigned int id;
> > > > > > - int ret = 0;
> > > > > > -
> > > > > > - for_each_gt(gt, xe, id) {
> > > > > > - xe_guc_submit_pause(>->uc.guc);
> > > > > > - ret |= xe_guc_submit_reset_block(>->uc.guc);
> > > > > > - }
> > > > > > -
> > > > > > - if (ret)
> > > > > > - drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
> > > > > > -}
> > > > > > -
> > > > > > -/**
> > > > > > - * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
> > > > > > - * @xe: the &xe_device struct instance
> > > > > > - *
> > > > > > - * After we have finished with all post-migration fixups, restart the driver
> > > > > > - * activities to continue feeding the GPU with workloads.
> > > > > > - */
> > > > > > -static void vf_post_migration_kickstart(struct xe_device *xe)
> > > > > > -{
> > > > > > - struct xe_gt *gt;
> > > > > > - unsigned int id;
> > > > > > -
> > > > > > - /*
> > > > > > - * Make sure interrupts on the new HW are properly set. The GuC IRQ
> > > > > > - * must be working at this point, since the recovery did started,
> > > > > > - * but the rest was not enabled using the procedure from spec.
> > > > > > - */
> > > > > > - xe_irq_resume(xe);
> > > > > > -
> > > > > > - for_each_gt(gt, xe, id) {
> > > > > > - xe_guc_submit_reset_unblock(>->uc.guc);
> > > > > > - xe_guc_submit_unpause(>->uc.guc);
> > > > > > - }
> > > > > > -}
> > > > > > -
> > > > > > -static bool gt_vf_post_migration_needed(struct xe_gt *gt)
> > > > > > -{
> > > > > > - return test_bit(gt->info.id, >_to_xe(gt)->sriov.vf.migration.gt_flags);
> > > > > > -}
> > > > > > -
> > > > > > -/*
> > > > > > - * Notify GuCs marked in flags about resource fixups apply finished.
> > > > > > - * @xe: the &xe_device struct instance
> > > > > > - * @gt_flags: flags marking to which GTs the notification shall be sent
> > > > > > - */
> > > > > > -static int vf_post_migration_notify_resfix_done(struct xe_device *xe, unsigned long gt_flags)
> > > > > > -{
> > > > > > - struct xe_gt *gt;
> > > > > > - unsigned int id;
> > > > > > - int err = 0;
> > > > > > -
> > > > > > - for_each_gt(gt, xe, id) {
> > > > > > - if (!test_bit(id, >_flags))
> > > > > > - continue;
> > > > > > - /* skip asking GuC for RESFIX exit if new recovery request arrived */
> > > > > > - if (gt_vf_post_migration_needed(gt))
> > > > > > - continue;
> > > > > > - err = xe_gt_sriov_vf_notify_resfix_done(gt);
> > > > > > - if (err)
> > > > > > - break;
> > > > > > - clear_bit(id, >_flags);
> > > > > > - }
> > > > > > -
> > > > > > - if (gt_flags && !err)
> > > > > > - drm_dbg(&xe->drm, "another recovery imminent, skipped some notifications\n");
> > > > > > - return err;
> > > > > > -}
> > > > > > -
> > > > > > -static int vf_get_next_migrated_gt_id(struct xe_device *xe)
> > > > > > -{
> > > > > > - struct xe_gt *gt;
> > > > > > - unsigned int id;
> > > > > > -
> > > > > > - for_each_gt(gt, xe, id) {
> > > > > > - if (test_and_clear_bit(id, &xe->sriov.vf.migration.gt_flags))
> > > > > > - return id;
> > > > > > - }
> > > > > > - return -1;
> > > > > > -}
> > > > > > -
> > > > > > -static size_t post_migration_scratch_size(struct xe_device *xe)
> > > > > > -{
> > > > > > - return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
> > > > > > -}
> > > > > > -
> > > > > > -/**
> > > > > > - * Perform post-migration fixups on a single GT.
> > > > > > - *
> > > > > > - * After migration, GuC needs to be re-queried for VF configuration to check
> > > > > > - * if it matches previous provisioning. Most of VF provisioning shall be the
> > > > > > - * same, except GGTT range, since GGTT is not virtualized per-VF. If GGTT
> > > > > > - * range has changed, we have to perform fixups - shift all GGTT references
> > > > > > - * used anywhere within the driver. After the fixups in this function succeed,
> > > > > > - * it is allowed to ask the GuC bound to this GT to continue normal operation.
> > > > > > - *
> > > > > > - * Returns: 0 if the operation completed successfully, or a negative error
> > > > > > - * code otherwise.
> > > > > > - */
> > > > > > -static int gt_vf_post_migration_fixups(struct xe_gt *gt)
> > > > > > -{
> > > > > > - s64 shift;
> > > > > > - void *buf;
> > > > > > - int err;
> > > > > > -
> > > > > > - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
> > > > > > - if (!buf)
> > > > > > - return -ENOMEM;
> > > > > > -
> > > > > > - err = xe_gt_sriov_vf_query_config(gt);
> > > > > > - if (err)
> > > > > > - goto out;
> > > > > > -
> > > > > > - shift = xe_gt_sriov_vf_ggtt_shift(gt);
> > > > > > - if (shift) {
> > > > > > - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> > > > > > - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > > > > > - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > > > > > - if (err)
> > > > > > - goto out;
> > > > > > - }
> > > > > > -
> > > > > > -out:
> > > > > > - kfree(buf);
> > > > > > - return err;
> > > > > > -}
> > > > > > -
> > > > > > -static void vf_post_migration_recovery(struct xe_device *xe)
> > > > > > -{
> > > > > > - unsigned long fixed_gts = 0;
> > > > > > - int id, err;
> > > > > > -
> > > > > > - drm_dbg(&xe->drm, "migration recovery in progress\n");
> > > > > > - xe_pm_runtime_get(xe);
> > > > > > - vf_post_migration_shutdown(xe);
> > > > > > -
> > > > > > - if (!xe_sriov_vf_migration_supported(xe)) {
> > > > > > - xe_sriov_err(xe, "migration is not supported\n");
> > > > > > - err = -ENOTRECOVERABLE;
> > > > > > - goto fail;
> > > > > > - }
> > > > > > -
> > > > > > - while (id = vf_get_next_migrated_gt_id(xe), id >= 0) {
> > > > > > - struct xe_gt *gt = xe_device_get_gt(xe, id);
> > > > > > -
> > > > > > - err = gt_vf_post_migration_fixups(gt);
> > > > > > - if (err)
> > > > > > - goto fail;
> > > > > > -
> > > > > > - set_bit(id, &fixed_gts);
> > > > > > - }
> > > > > > -
> > > > > > - vf_post_migration_kickstart(xe);
> > > > > > - err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
> > > > > > - if (err)
> > > > > > - goto fail;
> > > > > > -
> > > > > > - xe_pm_runtime_put(xe);
> > > > > > - drm_notice(&xe->drm, "migration recovery ended\n");
> > > > > > - return;
> > > > > > -fail:
> > > > > > - xe_pm_runtime_put(xe);
> > > > > > - drm_err(&xe->drm, "migration recovery failed (%pe)\n", ERR_PTR(err));
> > > > > > - xe_device_declare_wedged(xe);
> > > > > > -}
> > > > > > -
> > > > > > -static void migration_worker_func(struct work_struct *w)
> > > > > > -{
> > > > > > - struct xe_device *xe = container_of(w, struct xe_device,
> > > > > > - sriov.vf.migration.worker);
> > > > > > -
> > > > > > - vf_post_migration_recovery(xe);
> > > > > > -}
> > > > > > -
> > > > > > -/*
> > > > > > - * Check if post-restore recovery is coming on any of GTs.
> > > > > > - * @xe: the &xe_device struct instance
> > > > > > - *
> > > > > > - * Return: True if migration recovery worker will soon be running. Any worker currently
> > > > > > - * executing does not affect the result.
> > > > > > - */
> > > > > > -static bool vf_ready_to_recovery_on_any_gts(struct xe_device *xe)
> > > > > > -{
> > > > > > - struct xe_gt *gt;
> > > > > > - unsigned int id;
> > > > > > -
> > > > > > - for_each_gt(gt, xe, id) {
> > > > > > - if (test_bit(id, &xe->sriov.vf.migration.gt_flags))
> > > > > > - return true;
> > > > > > - }
> > > > > > - return false;
> > > > > > -}
> > > > > > -
> > > > > > -/**
> > > > > > - * xe_sriov_vf_start_migration_recovery - Start VF migration recovery.
> > > > > > - * @xe: the &xe_device to start recovery on
> > > > > > - *
> > > > > > - * This function shall be called only by VF.
> > > > > > - */
> > > > > > -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
> > > > > > -{
> > > > > > - bool started;
> > > > > > -
> > > > > > - xe_assert(xe, IS_SRIOV_VF(xe));
> > > > > > -
> > > > > > - if (!vf_ready_to_recovery_on_any_gts(xe))
> > > > > > - return;
> > > > > > -
> > > > > > - started = queue_work(xe->sriov.wq, &xe->sriov.vf.migration.worker);
> > > > > > - drm_info(&xe->drm, "VF migration recovery %s\n", started ?
> > > > > > - "scheduled" : "already in progress");
> > > > > > -}
> > > > > > -
> > > > > > /**
> > > > > > * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
> > > > > > * @xe: the &xe_device to initialize
> > > > > > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
> > > > > > index 9e752105ec2a..4df95266b261 100644
> > > > > > --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
> > > > > > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
> > > > > > @@ -13,7 +13,6 @@ struct xe_device;
> > > > > > void xe_sriov_vf_init_early(struct xe_device *xe);
> > > > > > int xe_sriov_vf_init_late(struct xe_device *xe);
> > > > > > -void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
> > > > > > bool xe_sriov_vf_migration_supported(struct xe_device *xe);
> > > > > > void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
> > > > > > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> > > > > > index 426cc5841958..6a0fd0f5463e 100644
> > > > > > --- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> > > > > > +++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
> > > > > > @@ -33,10 +33,6 @@ struct xe_device_vf {
> > > > > > /** @migration: VF Migration state data */
> > > > > > struct {
> > > > > > - /** @migration.worker: VF migration recovery worker */
> > > > > > - struct work_struct worker;
> > > > > > - /** @migration.gt_flags: Per-GT request flags for VF migration recovery */
> > > > > > - unsigned long gt_flags;
> > > > > > /**
> > > > > > * @migration.enabled: flag indicating if migration support
> > > > > > * was enabled or not due to missing prerequisites
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 13/34] drm/xe/vf: Abort H2G sends during VF post-migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (11 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 12/34] drm/xe/vf: Make VF recovery run on per-GT worker Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 11:00 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 14/34] drm/xe/vf: Remove memory allocations from VF post migration recovery Matthew Brost
` (23 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
While VF post-migration recovery is in progress, abort H2G sends with
-ECANCEL. These messages are treated as lost, and TLB invalidation
errors are suppressed. During this phase, the H2G channel is down, and
VF recovery requires the CT lock to proceed.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_guc_ct.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 47079ab9922c..661ab1bf4502 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -25,6 +25,7 @@
#include "xe_gt_printk.h"
#include "xe_gt_sriov_pf_control.h"
#include "xe_gt_sriov_pf_monitor.h"
+#include "xe_gt_sriov_vf.h"
#include "xe_guc.h"
#include "xe_guc_log.h"
#include "xe_guc_relay.h"
@@ -851,7 +852,7 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
u32 len, u32 g2h_len, u32 num_g2h,
struct g2h_fence *g2h_fence)
{
- struct xe_gt *gt __maybe_unused = ct_to_gt(ct);
+ struct xe_gt *gt = ct_to_gt(ct);
u16 seqno;
int ret;
@@ -872,7 +873,8 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
goto out;
}
- if (ct->state == XE_GUC_CT_STATE_STOPPED) {
+ if (ct->state == XE_GUC_CT_STATE_STOPPED ||
+ xe_gt_sriov_vf_recovery_inprogress(gt)) {
ret = -ECANCELED;
goto out;
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 13/34] drm/xe/vf: Abort H2G sends during VF post-migration recovery
2025-09-24 1:15 ` [PATCH v2 13/34] drm/xe/vf: Abort H2G sends during VF post-migration recovery Matthew Brost
@ 2025-09-24 11:00 ` Michal Wajdeczko
2025-09-24 20:01 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 11:00 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> While VF post-migration recovery is in progress, abort H2G sends with
should be abort or give the caller a second chance ?
btw, do we care about extra errors being logged in such case ?
> -ECANCEL. These messages are treated as lost, and TLB invalidation
> errors are suppressed. During this phase, the H2G channel is down, and
> VF recovery requires the CT lock to proceed.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_guc_ct.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> index 47079ab9922c..661ab1bf4502 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> @@ -25,6 +25,7 @@
> #include "xe_gt_printk.h"
> #include "xe_gt_sriov_pf_control.h"
> #include "xe_gt_sriov_pf_monitor.h"
> +#include "xe_gt_sriov_vf.h"
> #include "xe_guc.h"
> #include "xe_guc_log.h"
> #include "xe_guc_relay.h"
> @@ -851,7 +852,7 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
> u32 len, u32 g2h_len, u32 num_g2h,
> struct g2h_fence *g2h_fence)
> {
> - struct xe_gt *gt __maybe_unused = ct_to_gt(ct);
> + struct xe_gt *gt = ct_to_gt(ct);
> u16 seqno;
> int ret;
>
> @@ -872,7 +873,8 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
> goto out;
> }
>
> - if (ct->state == XE_GUC_CT_STATE_STOPPED) {
> + if (ct->state == XE_GUC_CT_STATE_STOPPED ||
> + xe_gt_sriov_vf_recovery_inprogress(gt)) {
this still looks like a hack
maybe we should either explicitly STOP CTB once we start the recovery,
or if that's too much, maybe introduce HALT CTB state?
btw, what should happen to already received G2H?
shouldn't we process them out when starting recovery?
and maybe we should return different error code to clearly
distinguish between our explicit STOP state and implicit recovery?
> ret = -ECANCELED;
> goto out;
> }
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 13/34] drm/xe/vf: Abort H2G sends during VF post-migration recovery
2025-09-24 11:00 ` Michal Wajdeczko
@ 2025-09-24 20:01 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:01 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 01:00:57PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > While VF post-migration recovery is in progress, abort H2G sends with
>
> should be abort or give the caller a second chance ?
>
> btw, do we care about extra errors being logged in such case ?
>
All submission H2G are designed to be lost and state machine recovers.
TLB invalidation H2G are also designed to be lost.
If we have a case where H2G need to be relible,
xe_guc_ct_send_recv_no_fail should be used. I code that function early
in Xe, is unused so it probably doesn't work, but conceptually we should
be able to make H2G relible across GT reset or VF migration.
IMO we audit all H2G is a follow up and if there is need fixup
xe_guc_ct_send_recv_no_fail and use in cases where a H2G needs to be
relible.
> > -ECANCEL. These messages are treated as lost, and TLB invalidation
> > errors are suppressed. During this phase, the H2G channel is down, and
> > VF recovery requires the CT lock to proceed.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_guc_ct.c | 6 ++++--
> > 1 file changed, 4 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> > index 47079ab9922c..661ab1bf4502 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> > @@ -25,6 +25,7 @@
> > #include "xe_gt_printk.h"
> > #include "xe_gt_sriov_pf_control.h"
> > #include "xe_gt_sriov_pf_monitor.h"
> > +#include "xe_gt_sriov_vf.h"
> > #include "xe_guc.h"
> > #include "xe_guc_log.h"
> > #include "xe_guc_relay.h"
> > @@ -851,7 +852,7 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
> > u32 len, u32 g2h_len, u32 num_g2h,
> > struct g2h_fence *g2h_fence)
> > {
> > - struct xe_gt *gt __maybe_unused = ct_to_gt(ct);
> > + struct xe_gt *gt = ct_to_gt(ct);
> > u16 seqno;
> > int ret;
> >
> > @@ -872,7 +873,8 @@ static int __guc_ct_send_locked(struct xe_guc_ct *ct, const u32 *action,
> > goto out;
> > }
> >
> > - if (ct->state == XE_GUC_CT_STATE_STOPPED) {
> > + if (ct->state == XE_GUC_CT_STATE_STOPPED ||
> > + xe_gt_sriov_vf_recovery_inprogress(gt)) {
>
> this still looks like a hack
>
It is not and is required. We need to pop out of __guc_ct_send_locked
during VF migration without taking the CT lock (calling STOP CTB).
Consider the case where H2G is full (no credits) and VF migration
happens. If we try to call STOP CTB this requires CT mutex which is
being held by __guc_ct_send_locked, __guc_ct_send_locked eventually
times out and tries to trigger a GT reset which is not desired. IMO the
flow of locklessly, immediately being able to detect a VF migration
happend in the CT code is correct.
> maybe we should either explicitly STOP CTB once we start the recovery,
> or if that's too much, maybe introduce HALT CTB state?
>
We do STOP the CT as soon as we can in VF migration recovery but this
cover the case mentioned above.
> btw, what should happen to already received G2H?
> shouldn't we process them out when starting recovery?
>
Yes. xe_guc_ct_flush process all G2H before stopping the CT.
> and maybe we should return different error code to clearly
> distinguish between our explicit STOP state and implicit recovery?
>
-ECANCELED is the correct return code which the upper layer (e.g., TLB
invalidation) understand that is fine the H2G was dropped.
Matt
> > ret = -ECANCELED;
> > goto out;
> > }
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 14/34] drm/xe/vf: Remove memory allocations from VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (12 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 13/34] drm/xe/vf: Abort H2G sends during VF post-migration recovery Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-26 1:35 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 15/34] drm/xe/vf: Close multi-GT GGTT shift race Matthew Brost
` (22 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
VF post migration recovery is the path of dma-fence signaling / reclaim,
avoid memory allocations in this path.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 29 ++++++++++++++---------
drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +-
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 2 ++
drivers/gpu/drm/xe/xe_sriov.c | 8 +++++--
drivers/gpu/drm/xe/xe_sriov_vf.c | 14 ++++++++---
drivers/gpu/drm/xe/xe_sriov_vf.h | 2 +-
6 files changed, 39 insertions(+), 18 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index cfb71b749e52..8304c26c076e 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1214,17 +1214,13 @@ static size_t post_migration_scratch_size(struct xe_device *xe)
static int vf_post_migration_fixups(struct xe_gt *gt)
{
+ void *buf = gt->sriov.vf.migration.lrc_wa_bb;
s64 shift;
- void *buf;
int err;
- buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
- if (!buf)
- return -ENOMEM;
-
err = xe_gt_sriov_vf_query_config(gt);
if (err)
- goto out;
+ return err;
shift = xe_gt_sriov_vf_ggtt_shift(gt);
if (shift) {
@@ -1232,12 +1228,10 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
if (err)
- goto out;
+ return err;
}
-out:
- kfree(buf);
- return err;
+ return 0;
}
static void vf_post_migration_kickstart(struct xe_gt *gt)
@@ -1314,15 +1308,28 @@ static void migration_worker_func(struct work_struct *w)
/**
* xe_gt_sriov_vf_migration_init_early() - VF post migration init early
* @gt: the &xe_gt
+ *
+ * Return 0 on success, errno on failure
*/
-void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
+int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
{
+ void *buf;
+
+ buf = drmm_kmalloc(>_to_xe(gt)->drm,
+ post_migration_scratch_size(gt_to_xe(gt)),
+ GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
+
+ gt->sriov.vf.migration.lrc_wa_bb = buf;
init_rwsem(>->sriov.vf.self_config.lock);
spin_lock_init(>->sriov.vf.migration.lock);
INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
xe_gt_sriov_info(gt, "migration not supported by this module version\n");
+
+ return 0;
}
/**
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index 2ac6775b52f0..195dbebe941e 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -23,7 +23,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt);
int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
-void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
+int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index 53680a2f188a..496b657119de 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -58,6 +58,8 @@ struct xe_gt_sriov_vf_migration {
struct work_struct worker;
/** @lock: Protects recovery_queued */
spinlock_t lock;
+ /** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
+ void *lrc_wa_bb;
/** @recovery_queued: VF post migration recovery in queued */
bool recovery_queued;
/** @recovery_inprogress: VF post migration recovery in progress */
diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c
index 7d2d6de2aabf..358a35c80d5a 100644
--- a/drivers/gpu/drm/xe/xe_sriov.c
+++ b/drivers/gpu/drm/xe/xe_sriov.c
@@ -116,8 +116,12 @@ int xe_sriov_init(struct xe_device *xe)
return err;
}
- if (IS_SRIOV_VF(xe))
- xe_sriov_vf_init_early(xe);
+ if (IS_SRIOV_VF(xe)) {
+ int err = xe_sriov_vf_init_early(xe);
+
+ if (err)
+ return err;
+ }
xe_assert(xe, !xe->sriov.wq);
xe->sriov.wq = alloc_workqueue("xe-sriov-wq", 0, 0);
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 7d91553c4acc..e622a7c562c4 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -180,16 +180,24 @@ static void vf_migration_init_early(struct xe_device *xe)
/**
* xe_sriov_vf_init_early - Initialize SR-IOV VF specific data.
* @xe: the &xe_device to initialize
+ *
+ * Return: 0 on success or a negative error code on failure.
*/
-void xe_sriov_vf_init_early(struct xe_device *xe)
+int xe_sriov_vf_init_early(struct xe_device *xe)
{
struct xe_gt *gt;
unsigned int id;
+ int err;
- for_each_gt(gt, xe, id)
- xe_gt_sriov_vf_migration_init_early(gt);
+ for_each_gt(gt, xe, id) {
+ err = xe_gt_sriov_vf_migration_init_early(gt);
+ if (err)
+ return err;
+ }
vf_migration_init_early(xe);
+
+ return 0;
}
/**
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
index 4df95266b261..13969c6910ce 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
@@ -11,7 +11,7 @@
struct dentry;
struct xe_device;
-void xe_sriov_vf_init_early(struct xe_device *xe);
+int xe_sriov_vf_init_early(struct xe_device *xe);
int xe_sriov_vf_init_late(struct xe_device *xe);
bool xe_sriov_vf_migration_supported(struct xe_device *xe);
void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 14/34] drm/xe/vf: Remove memory allocations from VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 14/34] drm/xe/vf: Remove memory allocations from VF post migration recovery Matthew Brost
@ 2025-09-26 1:35 ` Lis, Tomasz
2025-09-26 1:43 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-26 1:35 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> VF post migration recovery is the path of dma-fence signaling / reclaim,
> avoid memory allocations in this path.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 29 ++++++++++++++---------
> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +-
> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 2 ++
> drivers/gpu/drm/xe/xe_sriov.c | 8 +++++--
> drivers/gpu/drm/xe/xe_sriov_vf.c | 14 ++++++++---
> drivers/gpu/drm/xe/xe_sriov_vf.h | 2 +-
> 6 files changed, 39 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index cfb71b749e52..8304c26c076e 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -1214,17 +1214,13 @@ static size_t post_migration_scratch_size(struct xe_device *xe)
>
> static int vf_post_migration_fixups(struct xe_gt *gt)
> {
> + void *buf = gt->sriov.vf.migration.lrc_wa_bb;
can we just call it 'scratch', or 'recovery_scratch'? The current use is
limited to lrc, but besides temp buffer for WAs it also stores MEMIRQ regs.
This name makes it look like something more important than it is, and
does not describe all uses.
Besides that name, LGTM.
-Tomasz
> s64 shift;
> - void *buf;
> int err;
>
> - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
> - if (!buf)
> - return -ENOMEM;
> -
> err = xe_gt_sriov_vf_query_config(gt);
> if (err)
> - goto out;
> + return err;
>
> shift = xe_gt_sriov_vf_ggtt_shift(gt);
> if (shift) {
> @@ -1232,12 +1228,10 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> if (err)
> - goto out;
> + return err;
> }
>
> -out:
> - kfree(buf);
> - return err;
> + return 0;
> }
>
> static void vf_post_migration_kickstart(struct xe_gt *gt)
> @@ -1314,15 +1308,28 @@ static void migration_worker_func(struct work_struct *w)
> /**
> * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> * @gt: the &xe_gt
> + *
> + * Return 0 on success, errno on failure
> */
> -void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> +int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> {
> + void *buf;
> +
> + buf = drmm_kmalloc(>_to_xe(gt)->drm,
> + post_migration_scratch_size(gt_to_xe(gt)),
> + GFP_KERNEL);
> + if (!buf)
> + return -ENOMEM;
> +
> + gt->sriov.vf.migration.lrc_wa_bb = buf;
> init_rwsem(>->sriov.vf.self_config.lock);
> spin_lock_init(>->sriov.vf.migration.lock);
> INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
>
> if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> xe_gt_sriov_info(gt, "migration not supported by this module version\n");
> +
> + return 0;
> }
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> index 2ac6775b52f0..195dbebe941e 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> @@ -23,7 +23,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
>
> -void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> +int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
>
> u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> index 53680a2f188a..496b657119de 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> @@ -58,6 +58,8 @@ struct xe_gt_sriov_vf_migration {
> struct work_struct worker;
> /** @lock: Protects recovery_queued */
> spinlock_t lock;
> + /** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
> + void *lrc_wa_bb;
> /** @recovery_queued: VF post migration recovery in queued */
> bool recovery_queued;
> /** @recovery_inprogress: VF post migration recovery in progress */
> diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c
> index 7d2d6de2aabf..358a35c80d5a 100644
> --- a/drivers/gpu/drm/xe/xe_sriov.c
> +++ b/drivers/gpu/drm/xe/xe_sriov.c
> @@ -116,8 +116,12 @@ int xe_sriov_init(struct xe_device *xe)
> return err;
> }
>
> - if (IS_SRIOV_VF(xe))
> - xe_sriov_vf_init_early(xe);
> + if (IS_SRIOV_VF(xe)) {
> + int err = xe_sriov_vf_init_early(xe);
> +
> + if (err)
> + return err;
> + }
>
> xe_assert(xe, !xe->sriov.wq);
> xe->sriov.wq = alloc_workqueue("xe-sriov-wq", 0, 0);
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> index 7d91553c4acc..e622a7c562c4 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> @@ -180,16 +180,24 @@ static void vf_migration_init_early(struct xe_device *xe)
> /**
> * xe_sriov_vf_init_early - Initialize SR-IOV VF specific data.
> * @xe: the &xe_device to initialize
> + *
> + * Return: 0 on success or a negative error code on failure.
> */
> -void xe_sriov_vf_init_early(struct xe_device *xe)
> +int xe_sriov_vf_init_early(struct xe_device *xe)
> {
> struct xe_gt *gt;
> unsigned int id;
> + int err;
>
> - for_each_gt(gt, xe, id)
> - xe_gt_sriov_vf_migration_init_early(gt);
> + for_each_gt(gt, xe, id) {
> + err = xe_gt_sriov_vf_migration_init_early(gt);
> + if (err)
> + return err;
> + }
>
> vf_migration_init_early(xe);
> +
> + return 0;
> }
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
> index 4df95266b261..13969c6910ce 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
> @@ -11,7 +11,7 @@
> struct dentry;
> struct xe_device;
>
> -void xe_sriov_vf_init_early(struct xe_device *xe);
> +int xe_sriov_vf_init_early(struct xe_device *xe);
> int xe_sriov_vf_init_late(struct xe_device *xe);
> bool xe_sriov_vf_migration_supported(struct xe_device *xe);
> void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 14/34] drm/xe/vf: Remove memory allocations from VF post migration recovery
2025-09-26 1:35 ` Lis, Tomasz
@ 2025-09-26 1:43 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-26 1:43 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Fri, Sep 26, 2025 at 03:35:33AM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > VF post migration recovery is the path of dma-fence signaling / reclaim,
> > avoid memory allocations in this path.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 29 ++++++++++++++---------
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 +-
> > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 2 ++
> > drivers/gpu/drm/xe/xe_sriov.c | 8 +++++--
> > drivers/gpu/drm/xe/xe_sriov_vf.c | 14 ++++++++---
> > drivers/gpu/drm/xe/xe_sriov_vf.h | 2 +-
> > 6 files changed, 39 insertions(+), 18 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index cfb71b749e52..8304c26c076e 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -1214,17 +1214,13 @@ static size_t post_migration_scratch_size(struct xe_device *xe)
> > static int vf_post_migration_fixups(struct xe_gt *gt)
> > {
> > + void *buf = gt->sriov.vf.migration.lrc_wa_bb;
>
> can we just call it 'scratch', or 'recovery_scratch'? The current use is
> limited to lrc, but besides temp buffer for WAs it also stores MEMIRQ regs.
>
> This name makes it look like something more important than it is, and does
> not describe all uses.
>
> Besides that name, LGTM.
Yes, that is a better name. Will change.
Matt
>
> -Tomasz
>
> > s64 shift;
> > - void *buf;
> > int err;
> > - buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_ATOMIC);
> > - if (!buf)
> > - return -ENOMEM;
> > -
> > err = xe_gt_sriov_vf_query_config(gt);
> > if (err)
> > - goto out;
> > + return err;
> > shift = xe_gt_sriov_vf_ggtt_shift(gt);
> > if (shift) {
> > @@ -1232,12 +1228,10 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> > xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > if (err)
> > - goto out;
> > + return err;
> > }
> > -out:
> > - kfree(buf);
> > - return err;
> > + return 0;
> > }
> > static void vf_post_migration_kickstart(struct xe_gt *gt)
> > @@ -1314,15 +1308,28 @@ static void migration_worker_func(struct work_struct *w)
> > /**
> > * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> > * @gt: the &xe_gt
> > + *
> > + * Return 0 on success, errno on failure
> > */
> > -void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > +int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > {
> > + void *buf;
> > +
> > + buf = drmm_kmalloc(>_to_xe(gt)->drm,
> > + post_migration_scratch_size(gt_to_xe(gt)),
> > + GFP_KERNEL);
> > + if (!buf)
> > + return -ENOMEM;
> > +
> > + gt->sriov.vf.migration.lrc_wa_bb = buf;
> > init_rwsem(>->sriov.vf.self_config.lock);
> > spin_lock_init(>->sriov.vf.migration.lock);
> > INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
> > if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> > xe_gt_sriov_info(gt, "migration not supported by this module version\n");
> > +
> > + return 0;
> > }
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > index 2ac6775b52f0..195dbebe941e 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > @@ -23,7 +23,7 @@ int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> > int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> > void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> > -void xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> > +int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt);
> > bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt);
> > u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt);
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > index 53680a2f188a..496b657119de 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > @@ -58,6 +58,8 @@ struct xe_gt_sriov_vf_migration {
> > struct work_struct worker;
> > /** @lock: Protects recovery_queued */
> > spinlock_t lock;
> > + /** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
> > + void *lrc_wa_bb;
> > /** @recovery_queued: VF post migration recovery in queued */
> > bool recovery_queued;
> > /** @recovery_inprogress: VF post migration recovery in progress */
> > diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c
> > index 7d2d6de2aabf..358a35c80d5a 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov.c
> > +++ b/drivers/gpu/drm/xe/xe_sriov.c
> > @@ -116,8 +116,12 @@ int xe_sriov_init(struct xe_device *xe)
> > return err;
> > }
> > - if (IS_SRIOV_VF(xe))
> > - xe_sriov_vf_init_early(xe);
> > + if (IS_SRIOV_VF(xe)) {
> > + int err = xe_sriov_vf_init_early(xe);
> > +
> > + if (err)
> > + return err;
> > + }
> > xe_assert(xe, !xe->sriov.wq);
> > xe->sriov.wq = alloc_workqueue("xe-sriov-wq", 0, 0);
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > index 7d91553c4acc..e622a7c562c4 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> > @@ -180,16 +180,24 @@ static void vf_migration_init_early(struct xe_device *xe)
> > /**
> > * xe_sriov_vf_init_early - Initialize SR-IOV VF specific data.
> > * @xe: the &xe_device to initialize
> > + *
> > + * Return: 0 on success or a negative error code on failure.
> > */
> > -void xe_sriov_vf_init_early(struct xe_device *xe)
> > +int xe_sriov_vf_init_early(struct xe_device *xe)
> > {
> > struct xe_gt *gt;
> > unsigned int id;
> > + int err;
> > - for_each_gt(gt, xe, id)
> > - xe_gt_sriov_vf_migration_init_early(gt);
> > + for_each_gt(gt, xe, id) {
> > + err = xe_gt_sriov_vf_migration_init_early(gt);
> > + if (err)
> > + return err;
> > + }
> > vf_migration_init_early(xe);
> > +
> > + return 0;
> > }
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
> > index 4df95266b261..13969c6910ce 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
> > @@ -11,7 +11,7 @@
> > struct dentry;
> > struct xe_device;
> > -void xe_sriov_vf_init_early(struct xe_device *xe);
> > +int xe_sriov_vf_init_early(struct xe_device *xe);
> > int xe_sriov_vf_init_late(struct xe_device *xe);
> > bool xe_sriov_vf_migration_supported(struct xe_device *xe);
> > void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 15/34] drm/xe/vf: Close multi-GT GGTT shift race
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (13 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 14/34] drm/xe/vf: Remove memory allocations from VF post migration recovery Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-26 2:33 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 16/34] drm/xe/vf: Teardown VF post migration worker on driver unload Matthew Brost
` (21 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
As multi-GT VF post-migration recovery can run in parallel on different
workqueues, but both GTs point to the same GGTT, only one GT needs to
shift the GGTT. However, both GTs need to know when this step has
completed. To coordinate this, share the VF config lock among all GTs
that share a GGTT, and perform the GGTT shift under this lock.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 95 +++++++++--------------
drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 11 ++-
drivers/gpu/drm/xe/xe_guc.c | 2 +-
drivers/gpu/drm/xe/xe_tile_sriov_vf.c | 6 +-
drivers/gpu/drm/xe/xe_tile_sriov_vf.h | 1 -
6 files changed, 51 insertions(+), 67 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 8304c26c076e..807fdced0228 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -436,16 +436,19 @@ u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt)
return value;
}
-static int vf_get_ggtt_info(struct xe_gt *gt)
+static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
{
struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
+ struct xe_gt_sriov_vf_selfconfig *primary_config =
+ >_to_tile(gt)->primary_gt->sriov.vf.self_config;
struct xe_guc *guc = >->uc.guc;
u64 start, size;
+ s64 shift;
int err;
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
- down_write(&config->lock);
+ down_write(config->lock);
err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
if (unlikely(err))
@@ -465,13 +468,17 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
start, start + size - 1, size / SZ_1K);
- config->ggtt_shift = start - (s64)config->ggtt_base;
+ shift = start - (s64)primary_config->ggtt_base;
config->ggtt_base = start;
config->ggtt_size = size;
+ if (recovery)
+ primary_config->ggtt_base = start;
err = config->ggtt_size ? 0 : -ENODATA;
+ if (!err && shift && recovery)
+ xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
out:
- up_write(&config->lock);
+ up_write(config->lock);
return err;
}
@@ -485,7 +492,7 @@ static int vf_get_lmem_info(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
- down_write(&config->lock);
+ down_write(config->lock);
err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
if (unlikely(err))
@@ -505,7 +512,7 @@ static int vf_get_lmem_info(struct xe_gt *gt)
err = config->lmem_size ? 0 : -ENODATA;
out:
- up_write(&config->lock);
+ up_write(config->lock);
return err;
}
@@ -518,7 +525,7 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
- down_write(&config->lock);
+ down_write(config->lock);
err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
if (unlikely(err))
@@ -549,7 +556,7 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
err = config->num_ctxs ? 0 : -ENODATA;
out:
- up_write(&config->lock);
+ up_write(config->lock);
return err;
}
@@ -564,17 +571,18 @@ static void vf_cache_gmdid(struct xe_gt *gt)
/**
* xe_gt_sriov_vf_query_config - Query SR-IOV config data over MMIO.
* @gt: the &xe_gt
+ * @recovery: VF post migration recovery path
*
* This function is for VF use only.
*
* Return: 0 on success or a negative error code on failure.
*/
-int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
+int xe_gt_sriov_vf_query_config(struct xe_gt *gt, bool recovery)
{
struct xe_device *xe = gt_to_xe(gt);
int err;
- err = vf_get_ggtt_info(gt);
+ err = vf_get_ggtt_info(gt, recovery);
if (unlikely(err))
return err;
@@ -610,10 +618,10 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
- down_read(&config->lock);
+ down_read(config->lock);
xe_gt_assert(gt, config->num_ctxs);
val = config->num_ctxs;
- up_read(&config->lock);
+ up_read(config->lock);
return val;
}
@@ -634,10 +642,10 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
- down_read(&config->lock);
+ down_read(config->lock);
xe_gt_assert(gt, config->lmem_size);
val = config->lmem_size;
- up_read(&config->lock);
+ up_read(config->lock);
return val;
}
@@ -656,11 +664,9 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
u64 val;
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
- xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
+ lockdep_assert_held(config->lock);
- down_read(&config->lock);
val = config->ggtt_size;
- up_read(&config->lock);
return val;
}
@@ -680,34 +686,10 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
-
- down_read(&config->lock);
xe_gt_assert(gt, config->ggtt_size);
- val = config->ggtt_base;
- up_read(&config->lock);
-
- return val;
-}
+ lockdep_assert_held(config->lock);
-/**
- * xe_gt_sriov_vf_ggtt_shift - Return shift in GGTT range due to VF migration
- * @gt: the &xe_gt struct instance
- *
- * This function is for VF use only.
- *
- * Return: The shift value; could be negative
- */
-s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
-{
- struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
- s64 val;
-
- xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
- xe_gt_assert(gt, xe_gt_is_main_type(gt));
-
- down_read(&config->lock);
- val = config->ggtt_shift;
- up_read(&config->lock);
+ val = config->ggtt_base;
return val;
}
@@ -1115,7 +1097,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
- down_read(&config->lock);
+ down_read(config->lock);
drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
config->ggtt_base,
config->ggtt_base + config->ggtt_size - 1);
@@ -1123,8 +1105,6 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
string_get_size(config->ggtt_size, 1, STRING_UNITS_2, buf, sizeof(buf));
drm_printf(p, "GGTT size:\t%llu (%s)\n", config->ggtt_size, buf);
- drm_printf(p, "GGTT shift on last restore:\t%lld\n", config->ggtt_shift);
-
if (IS_DGFX(xe) && xe_gt_is_main_type(gt)) {
string_get_size(config->lmem_size, 1, STRING_UNITS_2, buf, sizeof(buf));
drm_printf(p, "LMEM size:\t%llu (%s)\n", config->lmem_size, buf);
@@ -1132,7 +1112,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
- up_read(&config->lock);
+ up_read(config->lock);
}
/**
@@ -1215,21 +1195,16 @@ static size_t post_migration_scratch_size(struct xe_device *xe)
static int vf_post_migration_fixups(struct xe_gt *gt)
{
void *buf = gt->sriov.vf.migration.lrc_wa_bb;
- s64 shift;
int err;
- err = xe_gt_sriov_vf_query_config(gt);
+ err = xe_gt_sriov_vf_query_config(gt, true);
if (err)
return err;
- shift = xe_gt_sriov_vf_ggtt_shift(gt);
- if (shift) {
- xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
- xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
- err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
- if (err)
- return err;
- }
+ xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
+ err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
+ if (err)
+ return err;
return 0;
}
@@ -1313,6 +1288,7 @@ static void migration_worker_func(struct work_struct *w)
*/
int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
{
+ struct xe_tile *tile = gt_to_tile(gt);
void *buf;
buf = drmm_kmalloc(>_to_xe(gt)->drm,
@@ -1322,7 +1298,10 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
return -ENOMEM;
gt->sriov.vf.migration.lrc_wa_bb = buf;
- init_rwsem(>->sriov.vf.self_config.lock);
+ if (xe_gt_is_main_type(gt))
+ init_rwsem(>->sriov.vf.self_config.__lock);
+ gt->sriov.vf.self_config.lock =
+ &tile->primary_gt->sriov.vf.self_config.__lock;
spin_lock_init(>->sriov.vf.migration.lock);
INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index 195dbebe941e..535237003915 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -18,7 +18,7 @@ int xe_gt_sriov_vf_bootstrap(struct xe_gt *gt);
void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
struct xe_uc_fw_version *wanted,
struct xe_uc_fw_version *found);
-int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
+int xe_gt_sriov_vf_query_config(struct xe_gt *gt, bool recovery);
int xe_gt_sriov_vf_connect(struct xe_gt *gt);
int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
@@ -31,7 +31,6 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt);
u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt);
-s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt);
u32 xe_gt_sriov_vf_read32(struct xe_gt *gt, struct xe_reg reg);
void xe_gt_sriov_vf_write32(struct xe_gt *gt, struct xe_reg reg, u32 val);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index 496b657119de..61484c7c9a36 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -19,16 +19,19 @@ struct xe_gt_sriov_vf_selfconfig {
u64 ggtt_base;
/** @ggtt_size: assigned size of the GGTT region. */
u64 ggtt_size;
- /** @ggtt_shift: difference in ggtt_base on last migration */
- s64 ggtt_shift;
/** @lmem_size: assigned size of the LMEM. */
u64 lmem_size;
/** @num_ctxs: assigned number of GuC submission context IDs. */
u16 num_ctxs;
/** @num_dbs: assigned number of GuC doorbells IDs. */
u16 num_dbs;
- /** @lock: lock for protecting access to all selfconfig fields. */
- struct rw_semaphore lock;
+ /** @__lock: lock for protecting access to all selfconfig fields. */
+ struct rw_semaphore __lock;
+ /**
+ * @lock: pointer to lock for protecting access to all selfconfig
+ * fields, all GTs point to primary GT.
+ */
+ struct rw_semaphore *lock;
};
/**
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 00789844ea4d..ac60da51da2c 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -712,7 +712,7 @@ static int vf_guc_init_noalloc(struct xe_guc *guc)
if (err)
return err;
- err = xe_gt_sriov_vf_query_config(gt);
+ err = xe_gt_sriov_vf_query_config(gt, false);
if (err)
return err;
diff --git a/drivers/gpu/drm/xe/xe_tile_sriov_vf.c b/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
index f221dbed16f0..dc6221fc0520 100644
--- a/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
@@ -40,7 +40,7 @@ static int vf_init_ggtt_balloons(struct xe_tile *tile)
*
* Return: 0 on success or a negative error code on failure.
*/
-int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
+static int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
{
u64 ggtt_base = xe_gt_sriov_vf_ggtt_base(tile->primary_gt);
u64 ggtt_size = xe_gt_sriov_vf_ggtt(tile->primary_gt);
@@ -100,12 +100,16 @@ int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
static int vf_balloon_ggtt(struct xe_tile *tile)
{
+ struct xe_gt_sriov_vf_selfconfig *config =
+ &tile->primary_gt->sriov.vf.self_config;
struct xe_ggtt *ggtt = tile->mem.ggtt;
int err;
+ down_read(config->lock);
mutex_lock(&ggtt->lock);
err = xe_tile_sriov_vf_balloon_ggtt_locked(tile);
mutex_unlock(&ggtt->lock);
+ up_read(config->lock);
return err;
}
diff --git a/drivers/gpu/drm/xe/xe_tile_sriov_vf.h b/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
index 93eb043171e8..4ee68d1fb28e 100644
--- a/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
@@ -11,7 +11,6 @@
struct xe_tile;
int xe_tile_sriov_vf_prepare_ggtt(struct xe_tile *tile);
-int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile);
void xe_tile_sriov_vf_deballoon_ggtt_locked(struct xe_tile *tile);
void xe_tile_sriov_vf_fixup_ggtt_nodes(struct xe_tile *tile, s64 shift);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 15/34] drm/xe/vf: Close multi-GT GGTT shift race
2025-09-24 1:15 ` [PATCH v2 15/34] drm/xe/vf: Close multi-GT GGTT shift race Matthew Brost
@ 2025-09-26 2:33 ` Lis, Tomasz
2025-09-26 19:09 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-26 2:33 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> As multi-GT VF post-migration recovery can run in parallel on different
> workqueues, but both GTs point to the same GGTT, only one GT needs to
> shift the GGTT. However, both GTs need to know when this step has
> completed. To coordinate this, share the VF config lock among all GTs
> that share a GGTT, and perform the GGTT shift under this lock.
The description does not mention removal of ggtt_shift variable; this
removal is not related to the locking change, so should be separately
mentioned.
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 95 +++++++++--------------
> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 11 ++-
> drivers/gpu/drm/xe/xe_guc.c | 2 +-
> drivers/gpu/drm/xe/xe_tile_sriov_vf.c | 6 +-
> drivers/gpu/drm/xe/xe_tile_sriov_vf.h | 1 -
> 6 files changed, 51 insertions(+), 67 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 8304c26c076e..807fdced0228 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -436,16 +436,19 @@ u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt)
> return value;
> }
>
> -static int vf_get_ggtt_info(struct xe_gt *gt)
> +static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
> {
> struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> + struct xe_gt_sriov_vf_selfconfig *primary_config =
> + >_to_tile(gt)->primary_gt->sriov.vf.self_config;
> struct xe_guc *guc = >->uc.guc;
> u64 start, size;
> + s64 shift;
> int err;
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> - down_write(&config->lock);
> + down_write(config->lock);
>
> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
> if (unlikely(err))
> @@ -465,13 +468,17 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
> xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
> start, start + size - 1, size / SZ_1K);
>
> - config->ggtt_shift = start - (s64)config->ggtt_base;
> + shift = start - (s64)primary_config->ggtt_base;
> config->ggtt_base = start;
> config->ggtt_size = size;
> + if (recovery)
> + primary_config->ggtt_base = start;
> err = config->ggtt_size ? 0 : -ENODATA;
>
> + if (!err && shift && recovery)
> + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> out:
> - up_write(&config->lock);
> + up_write(config->lock);
> return err;
> }
>
> @@ -485,7 +492,7 @@ static int vf_get_lmem_info(struct xe_gt *gt)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> - down_write(&config->lock);
> + down_write(config->lock);
>
> err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
> if (unlikely(err))
> @@ -505,7 +512,7 @@ static int vf_get_lmem_info(struct xe_gt *gt)
> err = config->lmem_size ? 0 : -ENODATA;
>
> out:
> - up_write(&config->lock);
> + up_write(config->lock);
> return err;
> }
>
> @@ -518,7 +525,7 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> - down_write(&config->lock);
> + down_write(config->lock);
>
> err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
> if (unlikely(err))
> @@ -549,7 +556,7 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
> err = config->num_ctxs ? 0 : -ENODATA;
>
> out:
> - up_write(&config->lock);
> + up_write(config->lock);
> return err;
> }
>
> @@ -564,17 +571,18 @@ static void vf_cache_gmdid(struct xe_gt *gt)
> /**
> * xe_gt_sriov_vf_query_config - Query SR-IOV config data over MMIO.
> * @gt: the &xe_gt
> + * @recovery: VF post migration recovery path
> *
> * This function is for VF use only.
> *
> * Return: 0 on success or a negative error code on failure.
> */
> -int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
> +int xe_gt_sriov_vf_query_config(struct xe_gt *gt, bool recovery)
> {
> struct xe_device *xe = gt_to_xe(gt);
> int err;
>
> - err = vf_get_ggtt_info(gt);
> + err = vf_get_ggtt_info(gt, recovery);
> if (unlikely(err))
> return err;
>
> @@ -610,10 +618,10 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
>
> - down_read(&config->lock);
> + down_read(config->lock);
> xe_gt_assert(gt, config->num_ctxs);
> val = config->num_ctxs;
> - up_read(&config->lock);
> + up_read(config->lock);
>
> return val;
> }
> @@ -634,10 +642,10 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
>
> - down_read(&config->lock);
> + down_read(config->lock);
> xe_gt_assert(gt, config->lmem_size);
> val = config->lmem_size;
> - up_read(&config->lock);
> + up_read(config->lock);
>
> return val;
> }
> @@ -656,11 +664,9 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
> u64 val;
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> - xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> + lockdep_assert_held(config->lock);
>
> - down_read(&config->lock);
> val = config->ggtt_size;
> - up_read(&config->lock);
>
> return val;
> }
> @@ -680,34 +686,10 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> -
> - down_read(&config->lock);
> xe_gt_assert(gt, config->ggtt_size);
> - val = config->ggtt_base;
> - up_read(&config->lock);
> -
> - return val;
> -}
> + lockdep_assert_held(config->lock);
>
> -/**
> - * xe_gt_sriov_vf_ggtt_shift - Return shift in GGTT range due to VF migration
> - * @gt: the &xe_gt struct instance
> - *
> - * This function is for VF use only.
> - *
> - * Return: The shift value; could be negative
> - */
> -s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
> -{
> - struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> - s64 val;
> -
> - xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> - xe_gt_assert(gt, xe_gt_is_main_type(gt));
> -
> - down_read(&config->lock);
> - val = config->ggtt_shift;
> - up_read(&config->lock);
> + val = config->ggtt_base;
>
> return val;
> }
> @@ -1115,7 +1097,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
>
> xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
>
> - down_read(&config->lock);
> + down_read(config->lock);
> drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
> config->ggtt_base,
> config->ggtt_base + config->ggtt_size - 1);
> @@ -1123,8 +1105,6 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
> string_get_size(config->ggtt_size, 1, STRING_UNITS_2, buf, sizeof(buf));
> drm_printf(p, "GGTT size:\t%llu (%s)\n", config->ggtt_size, buf);
>
> - drm_printf(p, "GGTT shift on last restore:\t%lld\n", config->ggtt_shift);
> -
Right.. so by that we're losing a useful debug value.
I'm not very attached to it as I did not originated the idea to have it
there.
IMO the GGTT config is printed to dmesg and that's enough, shift can be
computed from that.
For the main change - the lock and the multi-GT support - looks good, no
issues.
-Tomasz
> if (IS_DGFX(xe) && xe_gt_is_main_type(gt)) {
> string_get_size(config->lmem_size, 1, STRING_UNITS_2, buf, sizeof(buf));
> drm_printf(p, "LMEM size:\t%llu (%s)\n", config->lmem_size, buf);
> @@ -1132,7 +1112,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
>
> drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
> drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
> - up_read(&config->lock);
> + up_read(config->lock);
> }
>
> /**
> @@ -1215,21 +1195,16 @@ static size_t post_migration_scratch_size(struct xe_device *xe)
> static int vf_post_migration_fixups(struct xe_gt *gt)
> {
> void *buf = gt->sriov.vf.migration.lrc_wa_bb;
> - s64 shift;
> int err;
>
> - err = xe_gt_sriov_vf_query_config(gt);
> + err = xe_gt_sriov_vf_query_config(gt, true);
> if (err)
> return err;
>
> - shift = xe_gt_sriov_vf_ggtt_shift(gt);
> - if (shift) {
> - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> - if (err)
> - return err;
> - }
> + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> + if (err)
> + return err;
>
> return 0;
> }
> @@ -1313,6 +1288,7 @@ static void migration_worker_func(struct work_struct *w)
> */
> int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> {
> + struct xe_tile *tile = gt_to_tile(gt);
> void *buf;
>
> buf = drmm_kmalloc(>_to_xe(gt)->drm,
> @@ -1322,7 +1298,10 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> return -ENOMEM;
>
> gt->sriov.vf.migration.lrc_wa_bb = buf;
> - init_rwsem(>->sriov.vf.self_config.lock);
> + if (xe_gt_is_main_type(gt))
> + init_rwsem(>->sriov.vf.self_config.__lock);
> + gt->sriov.vf.self_config.lock =
> + &tile->primary_gt->sriov.vf.self_config.__lock;
> spin_lock_init(>->sriov.vf.migration.lock);
> INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> index 195dbebe941e..535237003915 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> @@ -18,7 +18,7 @@ int xe_gt_sriov_vf_bootstrap(struct xe_gt *gt);
> void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
> struct xe_uc_fw_version *wanted,
> struct xe_uc_fw_version *found);
> -int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
> +int xe_gt_sriov_vf_query_config(struct xe_gt *gt, bool recovery);
> int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> @@ -31,7 +31,6 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
> u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
> u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt);
> u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt);
> -s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt);
>
> u32 xe_gt_sriov_vf_read32(struct xe_gt *gt, struct xe_reg reg);
> void xe_gt_sriov_vf_write32(struct xe_gt *gt, struct xe_reg reg, u32 val);
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> index 496b657119de..61484c7c9a36 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> @@ -19,16 +19,19 @@ struct xe_gt_sriov_vf_selfconfig {
> u64 ggtt_base;
> /** @ggtt_size: assigned size of the GGTT region. */
> u64 ggtt_size;
> - /** @ggtt_shift: difference in ggtt_base on last migration */
> - s64 ggtt_shift;
> /** @lmem_size: assigned size of the LMEM. */
> u64 lmem_size;
> /** @num_ctxs: assigned number of GuC submission context IDs. */
> u16 num_ctxs;
> /** @num_dbs: assigned number of GuC doorbells IDs. */
> u16 num_dbs;
> - /** @lock: lock for protecting access to all selfconfig fields. */
> - struct rw_semaphore lock;
> + /** @__lock: lock for protecting access to all selfconfig fields. */
> + struct rw_semaphore __lock;
> + /**
> + * @lock: pointer to lock for protecting access to all selfconfig
> + * fields, all GTs point to primary GT.
> + */
> + struct rw_semaphore *lock;
> };
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> index 00789844ea4d..ac60da51da2c 100644
> --- a/drivers/gpu/drm/xe/xe_guc.c
> +++ b/drivers/gpu/drm/xe/xe_guc.c
> @@ -712,7 +712,7 @@ static int vf_guc_init_noalloc(struct xe_guc *guc)
> if (err)
> return err;
>
> - err = xe_gt_sriov_vf_query_config(gt);
> + err = xe_gt_sriov_vf_query_config(gt, false);
> if (err)
> return err;
>
> diff --git a/drivers/gpu/drm/xe/xe_tile_sriov_vf.c b/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
> index f221dbed16f0..dc6221fc0520 100644
> --- a/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
> @@ -40,7 +40,7 @@ static int vf_init_ggtt_balloons(struct xe_tile *tile)
> *
> * Return: 0 on success or a negative error code on failure.
> */
> -int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
> +static int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
> {
> u64 ggtt_base = xe_gt_sriov_vf_ggtt_base(tile->primary_gt);
> u64 ggtt_size = xe_gt_sriov_vf_ggtt(tile->primary_gt);
> @@ -100,12 +100,16 @@ int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
>
> static int vf_balloon_ggtt(struct xe_tile *tile)
> {
> + struct xe_gt_sriov_vf_selfconfig *config =
> + &tile->primary_gt->sriov.vf.self_config;
> struct xe_ggtt *ggtt = tile->mem.ggtt;
> int err;
>
> + down_read(config->lock);
> mutex_lock(&ggtt->lock);
> err = xe_tile_sriov_vf_balloon_ggtt_locked(tile);
> mutex_unlock(&ggtt->lock);
> + up_read(config->lock);
>
> return err;
> }
> diff --git a/drivers/gpu/drm/xe/xe_tile_sriov_vf.h b/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
> index 93eb043171e8..4ee68d1fb28e 100644
> --- a/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
> @@ -11,7 +11,6 @@
> struct xe_tile;
>
> int xe_tile_sriov_vf_prepare_ggtt(struct xe_tile *tile);
> -int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile);
> void xe_tile_sriov_vf_deballoon_ggtt_locked(struct xe_tile *tile);
> void xe_tile_sriov_vf_fixup_ggtt_nodes(struct xe_tile *tile, s64 shift);
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 15/34] drm/xe/vf: Close multi-GT GGTT shift race
2025-09-26 2:33 ` Lis, Tomasz
@ 2025-09-26 19:09 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-26 19:09 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Fri, Sep 26, 2025 at 04:33:36AM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > As multi-GT VF post-migration recovery can run in parallel on different
> > workqueues, but both GTs point to the same GGTT, only one GT needs to
> > shift the GGTT. However, both GTs need to know when this step has
> > completed. To coordinate this, share the VF config lock among all GTs
> > that share a GGTT, and perform the GGTT shift under this lock.
>
> The description does not mention removal of ggtt_shift variable; this
> removal is not related to the locking change, so should be separately
> mentioned.
>
The shift doesn't really need to be stored when shift is done under
config lock, rather just calculated there. I mention something like this
in the commit message.
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 95 +++++++++--------------
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 3 +-
> > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 11 ++-
> > drivers/gpu/drm/xe/xe_guc.c | 2 +-
> > drivers/gpu/drm/xe/xe_tile_sriov_vf.c | 6 +-
> > drivers/gpu/drm/xe/xe_tile_sriov_vf.h | 1 -
> > 6 files changed, 51 insertions(+), 67 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 8304c26c076e..807fdced0228 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -436,16 +436,19 @@ u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt)
> > return value;
> > }
> > -static int vf_get_ggtt_info(struct xe_gt *gt)
> > +static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
> > {
> > struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > + struct xe_gt_sriov_vf_selfconfig *primary_config =
> > + >_to_tile(gt)->primary_gt->sriov.vf.self_config;
> > struct xe_guc *guc = >->uc.guc;
> > u64 start, size;
> > + s64 shift;
> > int err;
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > - down_write(&config->lock);
> > + down_write(config->lock);
> > err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_GGTT_START_KEY, &start);
> > if (unlikely(err))
> > @@ -465,13 +468,17 @@ static int vf_get_ggtt_info(struct xe_gt *gt)
> > xe_gt_sriov_dbg_verbose(gt, "GGTT %#llx-%#llx = %lluK\n",
> > start, start + size - 1, size / SZ_1K);
> > - config->ggtt_shift = start - (s64)config->ggtt_base;
> > + shift = start - (s64)primary_config->ggtt_base;
> > config->ggtt_base = start;
> > config->ggtt_size = size;
> > + if (recovery)
> > + primary_config->ggtt_base = start;
> > err = config->ggtt_size ? 0 : -ENODATA;
> > + if (!err && shift && recovery)
> > + xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> > out:
> > - up_write(&config->lock);
> > + up_write(config->lock);
> > return err;
> > }
> > @@ -485,7 +492,7 @@ static int vf_get_lmem_info(struct xe_gt *gt)
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > - down_write(&config->lock);
> > + down_write(config->lock);
> > err = guc_action_query_single_klv64(guc, GUC_KLV_VF_CFG_LMEM_SIZE_KEY, &size);
> > if (unlikely(err))
> > @@ -505,7 +512,7 @@ static int vf_get_lmem_info(struct xe_gt *gt)
> > err = config->lmem_size ? 0 : -ENODATA;
> > out:
> > - up_write(&config->lock);
> > + up_write(config->lock);
> > return err;
> > }
> > @@ -518,7 +525,7 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > - down_write(&config->lock);
> > + down_write(config->lock);
> > err = guc_action_query_single_klv32(guc, GUC_KLV_VF_CFG_NUM_CONTEXTS_KEY, &num_ctxs);
> > if (unlikely(err))
> > @@ -549,7 +556,7 @@ static int vf_get_submission_cfg(struct xe_gt *gt)
> > err = config->num_ctxs ? 0 : -ENODATA;
> > out:
> > - up_write(&config->lock);
> > + up_write(config->lock);
> > return err;
> > }
> > @@ -564,17 +571,18 @@ static void vf_cache_gmdid(struct xe_gt *gt)
> > /**
> > * xe_gt_sriov_vf_query_config - Query SR-IOV config data over MMIO.
> > * @gt: the &xe_gt
> > + * @recovery: VF post migration recovery path
> > *
> > * This function is for VF use only.
> > *
> > * Return: 0 on success or a negative error code on failure.
> > */
> > -int xe_gt_sriov_vf_query_config(struct xe_gt *gt)
> > +int xe_gt_sriov_vf_query_config(struct xe_gt *gt, bool recovery)
> > {
> > struct xe_device *xe = gt_to_xe(gt);
> > int err;
> > - err = vf_get_ggtt_info(gt);
> > + err = vf_get_ggtt_info(gt, recovery);
> > if (unlikely(err))
> > return err;
> > @@ -610,10 +618,10 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt)
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > - down_read(&config->lock);
> > + down_read(config->lock);
> > xe_gt_assert(gt, config->num_ctxs);
> > val = config->num_ctxs;
> > - up_read(&config->lock);
> > + up_read(config->lock);
> > return val;
> > }
> > @@ -634,10 +642,10 @@ u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt)
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > - down_read(&config->lock);
> > + down_read(config->lock);
> > xe_gt_assert(gt, config->lmem_size);
> > val = config->lmem_size;
> > - up_read(&config->lock);
> > + up_read(config->lock);
> > return val;
> > }
> > @@ -656,11 +664,9 @@ u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt)
> > u64 val;
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > - xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > + lockdep_assert_held(config->lock);
> > - down_read(&config->lock);
> > val = config->ggtt_size;
> > - up_read(&config->lock);
> > return val;
> > }
> > @@ -680,34 +686,10 @@ u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt)
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > xe_gt_assert(gt, gt->sriov.vf.guc_version.major);
> > -
> > - down_read(&config->lock);
> > xe_gt_assert(gt, config->ggtt_size);
> > - val = config->ggtt_base;
> > - up_read(&config->lock);
> > -
> > - return val;
> > -}
> > + lockdep_assert_held(config->lock);
> > -/**
> > - * xe_gt_sriov_vf_ggtt_shift - Return shift in GGTT range due to VF migration
> > - * @gt: the &xe_gt struct instance
> > - *
> > - * This function is for VF use only.
> > - *
> > - * Return: The shift value; could be negative
> > - */
> > -s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt)
> > -{
> > - struct xe_gt_sriov_vf_selfconfig *config = >->sriov.vf.self_config;
> > - s64 val;
> > -
> > - xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > - xe_gt_assert(gt, xe_gt_is_main_type(gt));
> > -
> > - down_read(&config->lock);
> > - val = config->ggtt_shift;
> > - up_read(&config->lock);
> > + val = config->ggtt_base;
> > return val;
> > }
> > @@ -1115,7 +1097,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
> > xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
> > - down_read(&config->lock);
> > + down_read(config->lock);
> > drm_printf(p, "GGTT range:\t%#llx-%#llx\n",
> > config->ggtt_base,
> > config->ggtt_base + config->ggtt_size - 1);
> > @@ -1123,8 +1105,6 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
> > string_get_size(config->ggtt_size, 1, STRING_UNITS_2, buf, sizeof(buf));
> > drm_printf(p, "GGTT size:\t%llu (%s)\n", config->ggtt_size, buf);
> > - drm_printf(p, "GGTT shift on last restore:\t%lld\n", config->ggtt_shift);
> > -
>
> Right.. so by that we're losing a useful debug value.
>
> I'm not very attached to it as I did not originated the idea to have it
> there.
>
> IMO the GGTT config is printed to dmesg and that's enough, shift can be
> computed from that.
>
I added this message [1] so when shift the occurs we can see it
happening and also see which GT won the shift race. I've manually
verified on BMG either GT can safely win this race.
Matt
[1] https://patchwork.freedesktop.org/patch/676386/?series=154627&rev=2
>
> For the main change - the lock and the multi-GT support - looks good, no
> issues.
>
> -Tomasz
>
> > if (IS_DGFX(xe) && xe_gt_is_main_type(gt)) {
> > string_get_size(config->lmem_size, 1, STRING_UNITS_2, buf, sizeof(buf));
> > drm_printf(p, "LMEM size:\t%llu (%s)\n", config->lmem_size, buf);
> > @@ -1132,7 +1112,7 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p)
> > drm_printf(p, "GuC contexts:\t%u\n", config->num_ctxs);
> > drm_printf(p, "GuC doorbells:\t%u\n", config->num_dbs);
> > - up_read(&config->lock);
> > + up_read(config->lock);
> > }
> > /**
> > @@ -1215,21 +1195,16 @@ static size_t post_migration_scratch_size(struct xe_device *xe)
> > static int vf_post_migration_fixups(struct xe_gt *gt)
> > {
> > void *buf = gt->sriov.vf.migration.lrc_wa_bb;
> > - s64 shift;
> > int err;
> > - err = xe_gt_sriov_vf_query_config(gt);
> > + err = xe_gt_sriov_vf_query_config(gt, true);
> > if (err)
> > return err;
> > - shift = xe_gt_sriov_vf_ggtt_shift(gt);
> > - if (shift) {
> > - xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> > - xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > - err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > - if (err)
> > - return err;
> > - }
> > + xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > + err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > + if (err)
> > + return err;
> > return 0;
> > }
> > @@ -1313,6 +1288,7 @@ static void migration_worker_func(struct work_struct *w)
> > */
> > int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > {
> > + struct xe_tile *tile = gt_to_tile(gt);
> > void *buf;
> > buf = drmm_kmalloc(>_to_xe(gt)->drm,
> > @@ -1322,7 +1298,10 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > return -ENOMEM;
> > gt->sriov.vf.migration.lrc_wa_bb = buf;
> > - init_rwsem(>->sriov.vf.self_config.lock);
> > + if (xe_gt_is_main_type(gt))
> > + init_rwsem(>->sriov.vf.self_config.__lock);
> > + gt->sriov.vf.self_config.lock =
> > + &tile->primary_gt->sriov.vf.self_config.__lock;
> > spin_lock_init(>->sriov.vf.migration.lock);
> > INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > index 195dbebe941e..535237003915 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> > @@ -18,7 +18,7 @@ int xe_gt_sriov_vf_bootstrap(struct xe_gt *gt);
> > void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
> > struct xe_uc_fw_version *wanted,
> > struct xe_uc_fw_version *found);
> > -int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
> > +int xe_gt_sriov_vf_query_config(struct xe_gt *gt, bool recovery);
> > int xe_gt_sriov_vf_connect(struct xe_gt *gt);
> > int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
> > void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> > @@ -31,7 +31,6 @@ u16 xe_gt_sriov_vf_guc_ids(struct xe_gt *gt);
> > u64 xe_gt_sriov_vf_lmem(struct xe_gt *gt);
> > u64 xe_gt_sriov_vf_ggtt(struct xe_gt *gt);
> > u64 xe_gt_sriov_vf_ggtt_base(struct xe_gt *gt);
> > -s64 xe_gt_sriov_vf_ggtt_shift(struct xe_gt *gt);
> > u32 xe_gt_sriov_vf_read32(struct xe_gt *gt, struct xe_reg reg);
> > void xe_gt_sriov_vf_write32(struct xe_gt *gt, struct xe_reg reg, u32 val);
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > index 496b657119de..61484c7c9a36 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > @@ -19,16 +19,19 @@ struct xe_gt_sriov_vf_selfconfig {
> > u64 ggtt_base;
> > /** @ggtt_size: assigned size of the GGTT region. */
> > u64 ggtt_size;
> > - /** @ggtt_shift: difference in ggtt_base on last migration */
> > - s64 ggtt_shift;
> > /** @lmem_size: assigned size of the LMEM. */
> > u64 lmem_size;
> > /** @num_ctxs: assigned number of GuC submission context IDs. */
> > u16 num_ctxs;
> > /** @num_dbs: assigned number of GuC doorbells IDs. */
> > u16 num_dbs;
> > - /** @lock: lock for protecting access to all selfconfig fields. */
> > - struct rw_semaphore lock;
> > + /** @__lock: lock for protecting access to all selfconfig fields. */
> > + struct rw_semaphore __lock;
> > + /**
> > + * @lock: pointer to lock for protecting access to all selfconfig
> > + * fields, all GTs point to primary GT.
> > + */
> > + struct rw_semaphore *lock;
> > };
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> > index 00789844ea4d..ac60da51da2c 100644
> > --- a/drivers/gpu/drm/xe/xe_guc.c
> > +++ b/drivers/gpu/drm/xe/xe_guc.c
> > @@ -712,7 +712,7 @@ static int vf_guc_init_noalloc(struct xe_guc *guc)
> > if (err)
> > return err;
> > - err = xe_gt_sriov_vf_query_config(gt);
> > + err = xe_gt_sriov_vf_query_config(gt, false);
> > if (err)
> > return err;
> > diff --git a/drivers/gpu/drm/xe/xe_tile_sriov_vf.c b/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
> > index f221dbed16f0..dc6221fc0520 100644
> > --- a/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_tile_sriov_vf.c
> > @@ -40,7 +40,7 @@ static int vf_init_ggtt_balloons(struct xe_tile *tile)
> > *
> > * Return: 0 on success or a negative error code on failure.
> > */
> > -int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
> > +static int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
> > {
> > u64 ggtt_base = xe_gt_sriov_vf_ggtt_base(tile->primary_gt);
> > u64 ggtt_size = xe_gt_sriov_vf_ggtt(tile->primary_gt);
> > @@ -100,12 +100,16 @@ int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile)
> > static int vf_balloon_ggtt(struct xe_tile *tile)
> > {
> > + struct xe_gt_sriov_vf_selfconfig *config =
> > + &tile->primary_gt->sriov.vf.self_config;
> > struct xe_ggtt *ggtt = tile->mem.ggtt;
> > int err;
> > + down_read(config->lock);
> > mutex_lock(&ggtt->lock);
> > err = xe_tile_sriov_vf_balloon_ggtt_locked(tile);
> > mutex_unlock(&ggtt->lock);
> > + up_read(config->lock);
> > return err;
> > }
> > diff --git a/drivers/gpu/drm/xe/xe_tile_sriov_vf.h b/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
> > index 93eb043171e8..4ee68d1fb28e 100644
> > --- a/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
> > +++ b/drivers/gpu/drm/xe/xe_tile_sriov_vf.h
> > @@ -11,7 +11,6 @@
> > struct xe_tile;
> > int xe_tile_sriov_vf_prepare_ggtt(struct xe_tile *tile);
> > -int xe_tile_sriov_vf_balloon_ggtt_locked(struct xe_tile *tile);
> > void xe_tile_sriov_vf_deballoon_ggtt_locked(struct xe_tile *tile);
> > void xe_tile_sriov_vf_fixup_ggtt_nodes(struct xe_tile *tile, s64 shift);
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 16/34] drm/xe/vf: Teardown VF post migration worker on driver unload
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (14 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 15/34] drm/xe/vf: Close multi-GT GGTT shift race Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-26 15:40 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 17/34] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery Matthew Brost
` (20 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Be cautious and ensure the VF post-migration worker is not running
during driver unload.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 +++++++++++++++--
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 4 +++-
2 files changed, 18 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 807fdced0228..4eaffad6ebcf 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -811,7 +811,8 @@ static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
spin_lock(>->sriov.vf.migration.lock);
- if (!gt->sriov.vf.migration.recovery_queued) {
+ if (!gt->sriov.vf.migration.recovery_queued ||
+ !gt->sriov.vf.migration.recovery_teardown) {
gt->sriov.vf.migration.recovery_queued = true;
WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
@@ -1280,6 +1281,17 @@ static void migration_worker_func(struct work_struct *w)
vf_post_migration_recovery(gt);
}
+static void vf_migration_fini(struct drm_device *drm, void *arg)
+{
+ struct xe_gt *gt = arg;
+
+ spin_lock_irq(>->sriov.vf.migration.lock);
+ gt->sriov.vf.migration.recovery_teardown = true;
+ spin_unlock_irq(>->sriov.vf.migration.lock);
+
+ cancel_work_sync(>->sriov.vf.migration.worker);
+}
+
/**
* xe_gt_sriov_vf_migration_init_early() - VF post migration init early
* @gt: the &xe_gt
@@ -1308,7 +1320,8 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
xe_gt_sriov_info(gt, "migration not supported by this module version\n");
- return 0;
+ return drmm_add_action_or_reset(>_to_xe(gt)->drm,
+ vf_migration_fini, gt);
}
/**
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index 61484c7c9a36..beb9978336bb 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -59,10 +59,12 @@ struct xe_gt_sriov_vf_runtime {
struct xe_gt_sriov_vf_migration {
/** @migration: VF migration recovery worker */
struct work_struct worker;
- /** @lock: Protects recovery_queued */
+ /** @lock: Protects recovery_queued, teardown */
spinlock_t lock;
/** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
void *lrc_wa_bb;
+ /** @recovery_teardown: VF post migration recovery is being torn down */
+ bool recovery_teardown;
/** @recovery_queued: VF post migration recovery in queued */
bool recovery_queued;
/** @recovery_inprogress: VF post migration recovery in progress */
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 16/34] drm/xe/vf: Teardown VF post migration worker on driver unload
2025-09-24 1:15 ` [PATCH v2 16/34] drm/xe/vf: Teardown VF post migration worker on driver unload Matthew Brost
@ 2025-09-26 15:40 ` Lis, Tomasz
2025-09-26 19:13 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-26 15:40 UTC (permalink / raw)
To: Matthew Brost, intel-xe
[-- Attachment #1: Type: text/plain, Size: 3458 bytes --]
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Be cautious and ensure the VF post-migration worker is not running
> during driver unload.
>
> Signed-off-by: Matthew Brost<matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 +++++++++++++++--
> drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 4 +++-
> 2 files changed, 18 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 807fdced0228..4eaffad6ebcf 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -811,7 +811,8 @@ static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
>
> spin_lock(>->sriov.vf.migration.lock);
>
> - if (!gt->sriov.vf.migration.recovery_queued) {
> + if (!gt->sriov.vf.migration.recovery_queued ||
> + !gt->sriov.vf.migration.recovery_teardown) {
We're registering `vf_migration_fini` very early in the init. That means
it will be called very late. With that in mind, is it even possible to hit?
During `xe_gt_sriov_vf_migration_init_early`, interrupts are not enabled
yet. Doesn't that mean they are disabled already when vf_migration_fini
is called?
Both ggtt_fini_early and guc_submit_fini should be finished by then; so
if the recovery was running, it already crashed or errored out at this
point.
So, maybe register the fini later? We need to `init` before IRQs are
enabled, but for `fini` - it should be before `exec_queue_lookup` is
torn down.
-Tomasz
> gt->sriov.vf.migration.recovery_queued = true;
> WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
>
> @@ -1280,6 +1281,17 @@ static void migration_worker_func(struct work_struct *w)
> vf_post_migration_recovery(gt);
> }
>
> +static void vf_migration_fini(struct drm_device *drm, void *arg)
> +{
> + struct xe_gt *gt = arg;
> +
> + spin_lock_irq(>->sriov.vf.migration.lock);
> + gt->sriov.vf.migration.recovery_teardown = true;
> + spin_unlock_irq(>->sriov.vf.migration.lock);
> +
> + cancel_work_sync(>->sriov.vf.migration.worker);
> +}
> +
> /**
> * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> * @gt: the &xe_gt
> @@ -1308,7 +1320,8 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> xe_gt_sriov_info(gt, "migration not supported by this module version\n");
>
> - return 0;
> + return drmm_add_action_or_reset(>_to_xe(gt)->drm,
> + vf_migration_fini, gt);
> }
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> index 61484c7c9a36..beb9978336bb 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> @@ -59,10 +59,12 @@ struct xe_gt_sriov_vf_runtime {
> struct xe_gt_sriov_vf_migration {
> /** @migration: VF migration recovery worker */
> struct work_struct worker;
> - /** @lock: Protects recovery_queued */
> + /** @lock: Protects recovery_queued, teardown */
> spinlock_t lock;
> /** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
> void *lrc_wa_bb;
> + /** @recovery_teardown: VF post migration recovery is being torn down */
> + bool recovery_teardown;
> /** @recovery_queued: VF post migration recovery in queued */
> bool recovery_queued;
> /** @recovery_inprogress: VF post migration recovery in progress */
[-- Attachment #2: Type: text/html, Size: 4117 bytes --]
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 16/34] drm/xe/vf: Teardown VF post migration worker on driver unload
2025-09-26 15:40 ` Lis, Tomasz
@ 2025-09-26 19:13 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-26 19:13 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Fri, Sep 26, 2025 at 05:40:17PM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Be cautious and ensure the VF post-migration worker is not running
> > during driver unload.
> >
> > Signed-off-by: Matthew Brost<matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 17 +++++++++++++++--
> > drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 4 +++-
> > 2 files changed, 18 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 807fdced0228..4eaffad6ebcf 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -811,7 +811,8 @@ static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
> > spin_lock(>->sriov.vf.migration.lock);
> > - if (!gt->sriov.vf.migration.recovery_queued) {
> > + if (!gt->sriov.vf.migration.recovery_queued ||
> > + !gt->sriov.vf.migration.recovery_teardown) {
>
> We're registering `vf_migration_fini` very early in the init. That means it
> will be called very late. With that in mind, is it even possible to hit?
>
> During `xe_gt_sriov_vf_migration_init_early`, interrupts are not enabled
> yet. Doesn't that mean they are disabled already when vf_migration_fini is
> called?
>
> Both ggtt_fini_early and guc_submit_fini should be finished by then; so if
> the recovery was running, it already crashed or errored out at this point.
>
> So, maybe register the fini later? We need to `init` before IRQs are
> enabled, but for `fini` - it should be before `exec_queue_lookup` is torn
> down.
Yea, most this a bit later is probably a better idea. Let me look at
that.
Matt
>
> -Tomasz
>
> > gt->sriov.vf.migration.recovery_queued = true;
> > WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
> > @@ -1280,6 +1281,17 @@ static void migration_worker_func(struct work_struct *w)
> > vf_post_migration_recovery(gt);
> > }
> > +static void vf_migration_fini(struct drm_device *drm, void *arg)
> > +{
> > + struct xe_gt *gt = arg;
> > +
> > + spin_lock_irq(>->sriov.vf.migration.lock);
> > + gt->sriov.vf.migration.recovery_teardown = true;
> > + spin_unlock_irq(>->sriov.vf.migration.lock);
> > +
> > + cancel_work_sync(>->sriov.vf.migration.worker);
> > +}
> > +
> > /**
> > * xe_gt_sriov_vf_migration_init_early() - VF post migration init early
> > * @gt: the &xe_gt
> > @@ -1308,7 +1320,8 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
> > if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
> > xe_gt_sriov_info(gt, "migration not supported by this module version\n");
> > - return 0;
> > + return drmm_add_action_or_reset(>_to_xe(gt)->drm,
> > + vf_migration_fini, gt);
> > }
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > index 61484c7c9a36..beb9978336bb 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
> > @@ -59,10 +59,12 @@ struct xe_gt_sriov_vf_runtime {
> > struct xe_gt_sriov_vf_migration {
> > /** @migration: VF migration recovery worker */
> > struct work_struct worker;
> > - /** @lock: Protects recovery_queued */
> > + /** @lock: Protects recovery_queued, teardown */
> > spinlock_t lock;
> > /** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
> > void *lrc_wa_bb;
> > + /** @recovery_teardown: VF post migration recovery is being torn down */
> > + bool recovery_teardown;
> > /** @recovery_queued: VF post migration recovery in queued */
> > bool recovery_queued;
> > /** @recovery_inprogress: VF post migration recovery in progress */
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 17/34] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (15 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 16/34] drm/xe/vf: Teardown VF post migration worker on driver unload Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-27 2:59 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 18/34] drm/xe/vf: Wakeup in GuC backend on " Matthew Brost
` (19 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
With well behaved software, this should never happen.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt.c | 9 -------
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 7 -----
drivers/gpu/drm/xe/xe_guc_submit.c | 41 +++--------------------------
drivers/gpu/drm/xe/xe_guc_submit.h | 3 ---
4 files changed, 4 insertions(+), 56 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 3e0ad7e5b5df..1a4796cd885b 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -803,11 +803,6 @@ static int do_gt_restart(struct xe_gt *gt)
return 0;
}
-static int gt_wait_reset_unblock(struct xe_gt *gt)
-{
- return xe_guc_wait_reset_unblock(>->uc.guc);
-}
-
static int gt_reset(struct xe_gt *gt)
{
unsigned int fw_ref;
@@ -822,10 +817,6 @@ static int gt_reset(struct xe_gt *gt)
xe_gt_info(gt, "reset started\n");
- err = gt_wait_reset_unblock(gt);
- if (!err)
- xe_gt_warn(gt, "reset block failed to get lifted");
-
xe_pm_runtime_get(gt_to_xe(gt));
if (xe_fault_inject_gt_reset()) {
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 4eaffad6ebcf..b8e02bba7360 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1175,17 +1175,11 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
static void vf_post_migration_shutdown(struct xe_gt *gt)
{
- int ret = 0;
-
spin_lock_irq(>->sriov.vf.migration.lock);
gt->sriov.vf.migration.recovery_queued = false;
spin_unlock_irq(>->sriov.vf.migration.lock);
xe_guc_submit_pause(>->uc.guc);
- ret |= xe_guc_submit_reset_block(>->uc.guc);
-
- if (ret)
- xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
}
static size_t post_migration_scratch_size(struct xe_device *xe)
@@ -1219,7 +1213,6 @@ static void vf_post_migration_kickstart(struct xe_gt *gt)
*/
xe_irq_resume(gt_to_xe(gt));
- xe_guc_submit_reset_unblock(>->uc.guc);
xe_guc_submit_unpause(>->uc.guc);
}
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index cd5e506527fe..b82976f031e5 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -27,6 +27,7 @@
#include "xe_gt.h"
#include "xe_gt_clock.h"
#include "xe_gt_printk.h"
+#include "xe_gt_sriov_vf.h"
#include "xe_guc.h"
#include "xe_guc_capture.h"
#include "xe_guc_ct.h"
@@ -2182,47 +2183,13 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q)
}
}
-/**
- * xe_guc_submit_reset_block - Disallow reset calls on given GuC.
- * @guc: the &xe_guc struct instance
- */
-int xe_guc_submit_reset_block(struct xe_guc *guc)
-{
- return atomic_fetch_or(1, &guc->submission_state.reset_blocked);
-}
-
-/**
- * xe_guc_submit_reset_unblock - Allow back reset calls on given GuC.
- * @guc: the &xe_guc struct instance
- */
-void xe_guc_submit_reset_unblock(struct xe_guc *guc)
-{
- atomic_set_release(&guc->submission_state.reset_blocked, 0);
- wake_up_all(&guc->ct.wq);
-}
-
-static int guc_submit_reset_is_blocked(struct xe_guc *guc)
-{
- return atomic_read_acquire(&guc->submission_state.reset_blocked);
-}
-
-/* Maximum time of blocking reset */
-#define RESET_BLOCK_PERIOD_MAX (HZ * 5)
-
-/**
- * xe_guc_wait_reset_unblock - Wait until reset blocking flag is lifted, or timeout.
- * @guc: the &xe_guc struct instance
- */
-int xe_guc_wait_reset_unblock(struct xe_guc *guc)
-{
- return wait_event_timeout(guc->ct.wq,
- !guc_submit_reset_is_blocked(guc), RESET_BLOCK_PERIOD_MAX);
-}
-
int xe_guc_submit_reset_prepare(struct xe_guc *guc)
{
int ret;
+ if (WARN_ON_ONCE(xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))))
+ return 0;
+
if (!guc->submission_state.initialized)
return 0;
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 5b4a0a6fd818..f535fe3895e5 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -22,9 +22,6 @@ void xe_guc_submit_stop(struct xe_guc *guc);
int xe_guc_submit_start(struct xe_guc *guc);
void xe_guc_submit_pause(struct xe_guc *guc);
void xe_guc_submit_unpause(struct xe_guc *guc);
-int xe_guc_submit_reset_block(struct xe_guc *guc);
-void xe_guc_submit_reset_unblock(struct xe_guc *guc);
-int xe_guc_wait_reset_unblock(struct xe_guc *guc);
void xe_guc_submit_wedge(struct xe_guc *guc);
int xe_guc_read_stopped(struct xe_guc *guc);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 17/34] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 17/34] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery Matthew Brost
@ 2025-09-27 2:59 ` Lis, Tomasz
2025-09-27 22:33 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-27 2:59 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> With well behaved software, this should never happen.
Not a very reassuring description.
But the patch doesn't really lead to ignoring that situation as that
sentence suggests; there is a bailout from reset if VF recovery is in
progress.
This will work fine (for migration, it's actually better as we don't
need any reset of the new HW), and adds way less code. I'd just improve
the description.
-Tomasz
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt.c | 9 -------
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 7 -----
> drivers/gpu/drm/xe/xe_guc_submit.c | 41 +++--------------------------
> drivers/gpu/drm/xe/xe_guc_submit.h | 3 ---
> 4 files changed, 4 insertions(+), 56 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
> index 3e0ad7e5b5df..1a4796cd885b 100644
> --- a/drivers/gpu/drm/xe/xe_gt.c
> +++ b/drivers/gpu/drm/xe/xe_gt.c
> @@ -803,11 +803,6 @@ static int do_gt_restart(struct xe_gt *gt)
> return 0;
> }
>
> -static int gt_wait_reset_unblock(struct xe_gt *gt)
> -{
> - return xe_guc_wait_reset_unblock(>->uc.guc);
> -}
> -
> static int gt_reset(struct xe_gt *gt)
> {
> unsigned int fw_ref;
> @@ -822,10 +817,6 @@ static int gt_reset(struct xe_gt *gt)
>
> xe_gt_info(gt, "reset started\n");
>
> - err = gt_wait_reset_unblock(gt);
> - if (!err)
> - xe_gt_warn(gt, "reset block failed to get lifted");
> -
> xe_pm_runtime_get(gt_to_xe(gt));
>
> if (xe_fault_inject_gt_reset()) {
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 4eaffad6ebcf..b8e02bba7360 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -1175,17 +1175,11 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
>
> static void vf_post_migration_shutdown(struct xe_gt *gt)
> {
> - int ret = 0;
> -
> spin_lock_irq(>->sriov.vf.migration.lock);
> gt->sriov.vf.migration.recovery_queued = false;
> spin_unlock_irq(>->sriov.vf.migration.lock);
>
> xe_guc_submit_pause(>->uc.guc);
> - ret |= xe_guc_submit_reset_block(>->uc.guc);
> -
> - if (ret)
> - xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
> }
>
> static size_t post_migration_scratch_size(struct xe_device *xe)
> @@ -1219,7 +1213,6 @@ static void vf_post_migration_kickstart(struct xe_gt *gt)
> */
> xe_irq_resume(gt_to_xe(gt));
>
> - xe_guc_submit_reset_unblock(>->uc.guc);
> xe_guc_submit_unpause(>->uc.guc);
> }
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index cd5e506527fe..b82976f031e5 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -27,6 +27,7 @@
> #include "xe_gt.h"
> #include "xe_gt_clock.h"
> #include "xe_gt_printk.h"
> +#include "xe_gt_sriov_vf.h"
> #include "xe_guc.h"
> #include "xe_guc_capture.h"
> #include "xe_guc_ct.h"
> @@ -2182,47 +2183,13 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q)
> }
> }
>
> -/**
> - * xe_guc_submit_reset_block - Disallow reset calls on given GuC.
> - * @guc: the &xe_guc struct instance
> - */
> -int xe_guc_submit_reset_block(struct xe_guc *guc)
> -{
> - return atomic_fetch_or(1, &guc->submission_state.reset_blocked);
> -}
> -
> -/**
> - * xe_guc_submit_reset_unblock - Allow back reset calls on given GuC.
> - * @guc: the &xe_guc struct instance
> - */
> -void xe_guc_submit_reset_unblock(struct xe_guc *guc)
> -{
> - atomic_set_release(&guc->submission_state.reset_blocked, 0);
> - wake_up_all(&guc->ct.wq);
> -}
> -
> -static int guc_submit_reset_is_blocked(struct xe_guc *guc)
> -{
> - return atomic_read_acquire(&guc->submission_state.reset_blocked);
> -}
> -
> -/* Maximum time of blocking reset */
> -#define RESET_BLOCK_PERIOD_MAX (HZ * 5)
> -
> -/**
> - * xe_guc_wait_reset_unblock - Wait until reset blocking flag is lifted, or timeout.
> - * @guc: the &xe_guc struct instance
> - */
> -int xe_guc_wait_reset_unblock(struct xe_guc *guc)
> -{
> - return wait_event_timeout(guc->ct.wq,
> - !guc_submit_reset_is_blocked(guc), RESET_BLOCK_PERIOD_MAX);
> -}
> -
> int xe_guc_submit_reset_prepare(struct xe_guc *guc)
> {
> int ret;
>
> + if (WARN_ON_ONCE(xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))))
> + return 0;
> +
> if (!guc->submission_state.initialized)
> return 0;
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
> index 5b4a0a6fd818..f535fe3895e5 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.h
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.h
> @@ -22,9 +22,6 @@ void xe_guc_submit_stop(struct xe_guc *guc);
> int xe_guc_submit_start(struct xe_guc *guc);
> void xe_guc_submit_pause(struct xe_guc *guc);
> void xe_guc_submit_unpause(struct xe_guc *guc);
> -int xe_guc_submit_reset_block(struct xe_guc *guc);
> -void xe_guc_submit_reset_unblock(struct xe_guc *guc);
> -int xe_guc_wait_reset_unblock(struct xe_guc *guc);
> void xe_guc_submit_wedge(struct xe_guc *guc);
>
> int xe_guc_read_stopped(struct xe_guc *guc);
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 17/34] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery
2025-09-27 2:59 ` Lis, Tomasz
@ 2025-09-27 22:33 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-27 22:33 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Sat, Sep 27, 2025 at 04:59:26AM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > With well behaved software, this should never happen.
>
> Not a very reassuring description.
>
> But the patch doesn't really lead to ignoring that situation as that
> sentence suggests; there is a bailout from reset if VF recovery is in
> progress.
>
> This will work fine (for migration, it's actually better as we don't need
> any reset of the new HW), and adds way less code. I'd just improve the
> description.
>
Let me make the description a bit more verbose.
Matt
> -Tomasz
>
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt.c | 9 -------
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 7 -----
> > drivers/gpu/drm/xe/xe_guc_submit.c | 41 +++--------------------------
> > drivers/gpu/drm/xe/xe_guc_submit.h | 3 ---
> > 4 files changed, 4 insertions(+), 56 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
> > index 3e0ad7e5b5df..1a4796cd885b 100644
> > --- a/drivers/gpu/drm/xe/xe_gt.c
> > +++ b/drivers/gpu/drm/xe/xe_gt.c
> > @@ -803,11 +803,6 @@ static int do_gt_restart(struct xe_gt *gt)
> > return 0;
> > }
> > -static int gt_wait_reset_unblock(struct xe_gt *gt)
> > -{
> > - return xe_guc_wait_reset_unblock(>->uc.guc);
> > -}
> > -
> > static int gt_reset(struct xe_gt *gt)
> > {
> > unsigned int fw_ref;
> > @@ -822,10 +817,6 @@ static int gt_reset(struct xe_gt *gt)
> > xe_gt_info(gt, "reset started\n");
> > - err = gt_wait_reset_unblock(gt);
> > - if (!err)
> > - xe_gt_warn(gt, "reset block failed to get lifted");
> > -
> > xe_pm_runtime_get(gt_to_xe(gt));
> > if (xe_fault_inject_gt_reset()) {
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 4eaffad6ebcf..b8e02bba7360 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -1175,17 +1175,11 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
> > static void vf_post_migration_shutdown(struct xe_gt *gt)
> > {
> > - int ret = 0;
> > -
> > spin_lock_irq(>->sriov.vf.migration.lock);
> > gt->sriov.vf.migration.recovery_queued = false;
> > spin_unlock_irq(>->sriov.vf.migration.lock);
> > xe_guc_submit_pause(>->uc.guc);
> > - ret |= xe_guc_submit_reset_block(>->uc.guc);
> > -
> > - if (ret)
> > - xe_gt_sriov_info(gt, "migration recovery encountered ongoing reset\n");
> > }
> > static size_t post_migration_scratch_size(struct xe_device *xe)
> > @@ -1219,7 +1213,6 @@ static void vf_post_migration_kickstart(struct xe_gt *gt)
> > */
> > xe_irq_resume(gt_to_xe(gt));
> > - xe_guc_submit_reset_unblock(>->uc.guc);
> > xe_guc_submit_unpause(>->uc.guc);
> > }
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> > index cd5e506527fe..b82976f031e5 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> > @@ -27,6 +27,7 @@
> > #include "xe_gt.h"
> > #include "xe_gt_clock.h"
> > #include "xe_gt_printk.h"
> > +#include "xe_gt_sriov_vf.h"
> > #include "xe_guc.h"
> > #include "xe_guc_capture.h"
> > #include "xe_guc_ct.h"
> > @@ -2182,47 +2183,13 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q)
> > }
> > }
> > -/**
> > - * xe_guc_submit_reset_block - Disallow reset calls on given GuC.
> > - * @guc: the &xe_guc struct instance
> > - */
> > -int xe_guc_submit_reset_block(struct xe_guc *guc)
> > -{
> > - return atomic_fetch_or(1, &guc->submission_state.reset_blocked);
> > -}
> > -
> > -/**
> > - * xe_guc_submit_reset_unblock - Allow back reset calls on given GuC.
> > - * @guc: the &xe_guc struct instance
> > - */
> > -void xe_guc_submit_reset_unblock(struct xe_guc *guc)
> > -{
> > - atomic_set_release(&guc->submission_state.reset_blocked, 0);
> > - wake_up_all(&guc->ct.wq);
> > -}
> > -
> > -static int guc_submit_reset_is_blocked(struct xe_guc *guc)
> > -{
> > - return atomic_read_acquire(&guc->submission_state.reset_blocked);
> > -}
> > -
> > -/* Maximum time of blocking reset */
> > -#define RESET_BLOCK_PERIOD_MAX (HZ * 5)
> > -
> > -/**
> > - * xe_guc_wait_reset_unblock - Wait until reset blocking flag is lifted, or timeout.
> > - * @guc: the &xe_guc struct instance
> > - */
> > -int xe_guc_wait_reset_unblock(struct xe_guc *guc)
> > -{
> > - return wait_event_timeout(guc->ct.wq,
> > - !guc_submit_reset_is_blocked(guc), RESET_BLOCK_PERIOD_MAX);
> > -}
> > -
> > int xe_guc_submit_reset_prepare(struct xe_guc *guc)
> > {
> > int ret;
> > + if (WARN_ON_ONCE(xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))))
> > + return 0;
> > +
> > if (!guc->submission_state.initialized)
> > return 0;
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
> > index 5b4a0a6fd818..f535fe3895e5 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.h
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.h
> > @@ -22,9 +22,6 @@ void xe_guc_submit_stop(struct xe_guc *guc);
> > int xe_guc_submit_start(struct xe_guc *guc);
> > void xe_guc_submit_pause(struct xe_guc *guc);
> > void xe_guc_submit_unpause(struct xe_guc *guc);
> > -int xe_guc_submit_reset_block(struct xe_guc *guc);
> > -void xe_guc_submit_reset_unblock(struct xe_guc *guc);
> > -int xe_guc_wait_reset_unblock(struct xe_guc *guc);
> > void xe_guc_submit_wedge(struct xe_guc *guc);
> > int xe_guc_read_stopped(struct xe_guc *guc);
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 18/34] drm/xe/vf: Wakeup in GuC backend on VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (16 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 17/34] drm/xe/vf: Don't allow GT reset to be queued during VF post migration recovery Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-25 19:06 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift Matthew Brost
` (18 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
If VF post-migration recovery is in progress, the recovery flow will
rebuild all GuC submission state. In this case, exit all waiters to
ensure that submission queue scheduling can also be paused. Avoid taking
any adverse actions after aborting the wait.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 3 ++
drivers/gpu/drm/xe/xe_guc_submit.c | 51 +++++++++++++++++++++--------
2 files changed, 41 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index b8e02bba7360..a96e0dd65bc1 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -815,6 +815,9 @@ static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
!gt->sriov.vf.migration.recovery_teardown) {
gt->sriov.vf.migration.recovery_queued = true;
WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
+ smp_wmb(); /* Ensure above write visable before wake */
+
+ wake_up_all(>->uc.guc.ct.wq);
started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index b82976f031e5..52b86cab4ec5 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -984,6 +984,9 @@ static u32 wq_space_until_wrap(struct xe_exec_queue *q)
return (WQ_SIZE - q->guc->wqi_tail);
}
+#define vf_recovery(guc) \
+ xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))
+
static int wq_wait_for_space(struct xe_exec_queue *q, u32 wqi_size)
{
struct xe_guc *guc = exec_queue_to_guc(q);
@@ -993,7 +996,7 @@ static int wq_wait_for_space(struct xe_exec_queue *q, u32 wqi_size)
#define AVAILABLE_SPACE \
CIRC_SPACE(q->guc->wqi_tail, q->guc->wqi_head, WQ_SIZE)
- if (wqi_size > AVAILABLE_SPACE) {
+ if (wqi_size > AVAILABLE_SPACE && !vf_recovery(guc)) {
try_again:
q->guc->wqi_head = parallel_read(xe, map, wq_desc.head);
if (wqi_size > AVAILABLE_SPACE) {
@@ -1192,9 +1195,10 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
ret = wait_event_timeout(guc->ct.wq,
(!exec_queue_pending_enable(q) &&
!exec_queue_pending_disable(q)) ||
- xe_guc_read_stopped(guc),
+ xe_guc_read_stopped(guc) ||
+ vf_recovery(guc),
HZ * 5);
- if (!ret) {
+ if (!ret && !vf_recovery(guc)) {
struct xe_gpu_scheduler *sched = &q->guc->sched;
xe_gt_warn(q->gt, "Pending enable/disable failed to respond\n");
@@ -1297,6 +1301,10 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
bool wedged = false;
xe_gt_assert(guc_to_gt(guc), xe_exec_queue_is_lr(q));
+
+ if (vf_recovery(guc))
+ return;
+
trace_xe_exec_queue_lr_cleanup(q);
if (!exec_queue_killed(q))
@@ -1329,7 +1337,11 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
*/
ret = wait_event_timeout(guc->ct.wq,
!exec_queue_pending_disable(q) ||
- xe_guc_read_stopped(guc), HZ * 5);
+ xe_guc_read_stopped(guc) ||
+ vf_recovery(guc), HZ * 5);
+ if (vf_recovery(guc))
+ return;
+
if (!ret) {
xe_gt_warn(q->gt, "Schedule disable failed to respond, guc_id=%d\n",
q->guc->id);
@@ -1419,8 +1431,9 @@ static void enable_scheduling(struct xe_exec_queue *q)
ret = wait_event_timeout(guc->ct.wq,
!exec_queue_pending_enable(q) ||
- xe_guc_read_stopped(guc), HZ * 5);
- if (!ret || xe_guc_read_stopped(guc)) {
+ xe_guc_read_stopped(guc) ||
+ vf_recovery(guc), HZ * 5);
+ if ((!ret && !vf_recovery(guc)) || xe_guc_read_stopped(guc)) {
xe_gt_warn(guc_to_gt(guc), "Schedule enable failed to respond");
set_exec_queue_banned(q);
xe_gt_reset_async(q->gt);
@@ -1491,7 +1504,8 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
* list so job can be freed and kick scheduler ensuring free job is not
* lost.
*/
- if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &job->fence->flags))
+ if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &job->fence->flags) ||
+ vf_recovery(guc))
return DRM_GPU_SCHED_STAT_NO_HANG;
/* Kill the run_job entry point */
@@ -1543,7 +1557,10 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
ret = wait_event_timeout(guc->ct.wq,
(!exec_queue_pending_enable(q) &&
!exec_queue_pending_disable(q)) ||
- xe_guc_read_stopped(guc), HZ * 5);
+ xe_guc_read_stopped(guc) ||
+ vf_recovery(guc), HZ * 5);
+ if (vf_recovery(guc))
+ goto handle_vf_resume;
if (!ret || xe_guc_read_stopped(guc))
goto trigger_reset;
@@ -1568,7 +1585,10 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
smp_rmb();
ret = wait_event_timeout(guc->ct.wq,
!exec_queue_pending_disable(q) ||
- xe_guc_read_stopped(guc), HZ * 5);
+ xe_guc_read_stopped(guc) ||
+ vf_recovery(guc), HZ * 5);
+ if (vf_recovery(guc))
+ goto handle_vf_resume;
if (!ret || xe_guc_read_stopped(guc)) {
trigger_reset:
if (!ret)
@@ -1673,6 +1693,7 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
* some thought, do this in a follow up.
*/
xe_sched_submission_start(sched);
+handle_vf_resume:
return DRM_GPU_SCHED_STAT_NO_HANG;
}
@@ -1794,8 +1815,9 @@ static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg)
if (guc_exec_queue_allowed_to_change_state(q) && !exec_queue_suspended(q) &&
exec_queue_enabled(q)) {
- wait_event(guc->ct.wq, (q->guc->resume_time != RESUME_PENDING ||
- xe_guc_read_stopped(guc)) && !exec_queue_pending_disable(q));
+ wait_event(guc->ct.wq, vf_recovery(guc) ||
+ ((q->guc->resume_time != RESUME_PENDING ||
+ xe_guc_read_stopped(guc)) && !exec_queue_pending_disable(q)));
if (!xe_guc_read_stopped(guc)) {
s64 since_resume_ms =
@@ -1922,7 +1944,7 @@ static int guc_exec_queue_init(struct xe_exec_queue *q)
q->entity = &ge->entity;
- if (xe_guc_read_stopped(guc))
+ if (xe_guc_read_stopped(guc) || vf_recovery(guc))
xe_sched_stop(sched);
mutex_unlock(&guc->submission_state.lock);
@@ -2075,12 +2097,15 @@ static int guc_exec_queue_suspend_wait(struct xe_exec_queue *q)
* suspend_pending upon kill but to be paranoid but races in which
* suspend_pending is set after kill also check kill here.
*/
+retry:
ret = wait_event_interruptible_timeout(q->guc->suspend_wait,
!READ_ONCE(q->guc->suspend_pending) ||
exec_queue_killed(q) ||
xe_guc_read_stopped(guc),
HZ * 5);
+ if (vf_recovery(guc) && !xe_device_wedged((guc_to_xe(guc))))
+ goto retry;
if (!ret) {
xe_gt_warn(guc_to_gt(guc),
"Suspend fence, guc_id=%d, failed to respond",
@@ -2187,7 +2212,7 @@ int xe_guc_submit_reset_prepare(struct xe_guc *guc)
{
int ret;
- if (WARN_ON_ONCE(xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))))
+ if (WARN_ON_ONCE(vf_recovery(guc)))
return 0;
if (!guc->submission_state.initialized)
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 18/34] drm/xe/vf: Wakeup in GuC backend on VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 18/34] drm/xe/vf: Wakeup in GuC backend on " Matthew Brost
@ 2025-09-25 19:06 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-25 19:06 UTC (permalink / raw)
To: intel-xe
On Tue, Sep 23, 2025 at 06:15:45PM -0700, Matthew Brost wrote:
> If VF post-migration recovery is in progress, the recovery flow will
> rebuild all GuC submission state. In this case, exit all waiters to
> ensure that submission queue scheduling can also be paused. Avoid taking
> any adverse actions after aborting the wait.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 3 ++
> drivers/gpu/drm/xe/xe_guc_submit.c | 51 +++++++++++++++++++++--------
> 2 files changed, 41 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index b8e02bba7360..a96e0dd65bc1 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -815,6 +815,9 @@ static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
> !gt->sriov.vf.migration.recovery_teardown) {
> gt->sriov.vf.migration.recovery_queued = true;
> WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
> + smp_wmb(); /* Ensure above write visable before wake */
> +
> + wake_up_all(>->uc.guc.ct.wq);
>
> started = queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
> xe_gt_sriov_info(gt, "VF migration recovery %s\n", started ?
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index b82976f031e5..52b86cab4ec5 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -984,6 +984,9 @@ static u32 wq_space_until_wrap(struct xe_exec_queue *q)
> return (WQ_SIZE - q->guc->wqi_tail);
> }
>
> +#define vf_recovery(guc) \
> + xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))
> +
> static int wq_wait_for_space(struct xe_exec_queue *q, u32 wqi_size)
> {
> struct xe_guc *guc = exec_queue_to_guc(q);
> @@ -993,7 +996,7 @@ static int wq_wait_for_space(struct xe_exec_queue *q, u32 wqi_size)
>
> #define AVAILABLE_SPACE \
> CIRC_SPACE(q->guc->wqi_tail, q->guc->wqi_head, WQ_SIZE)
> - if (wqi_size > AVAILABLE_SPACE) {
> + if (wqi_size > AVAILABLE_SPACE && !vf_recovery(guc)) {
> try_again:
> q->guc->wqi_head = parallel_read(xe, map, wq_desc.head);
> if (wqi_size > AVAILABLE_SPACE) {
> @@ -1192,9 +1195,10 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
> ret = wait_event_timeout(guc->ct.wq,
> (!exec_queue_pending_enable(q) &&
> !exec_queue_pending_disable(q)) ||
> - xe_guc_read_stopped(guc),
> + xe_guc_read_stopped(guc) ||
> + vf_recovery(guc),
> HZ * 5);
> - if (!ret) {
> + if (!ret && !vf_recovery(guc)) {
> struct xe_gpu_scheduler *sched = &q->guc->sched;
>
> xe_gt_warn(q->gt, "Pending enable/disable failed to respond\n");
> @@ -1297,6 +1301,10 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
> bool wedged = false;
>
> xe_gt_assert(guc_to_gt(guc), xe_exec_queue_is_lr(q));
> +
> + if (vf_recovery(guc))
> + return;
> +
> trace_xe_exec_queue_lr_cleanup(q);
>
> if (!exec_queue_killed(q))
> @@ -1329,7 +1337,11 @@ static void xe_guc_exec_queue_lr_cleanup(struct work_struct *w)
> */
> ret = wait_event_timeout(guc->ct.wq,
> !exec_queue_pending_disable(q) ||
> - xe_guc_read_stopped(guc), HZ * 5);
> + xe_guc_read_stopped(guc) ||
> + vf_recovery(guc), HZ * 5);
> + if (vf_recovery(guc))
> + return;
> +
> if (!ret) {
> xe_gt_warn(q->gt, "Schedule disable failed to respond, guc_id=%d\n",
> q->guc->id);
> @@ -1419,8 +1431,9 @@ static void enable_scheduling(struct xe_exec_queue *q)
>
> ret = wait_event_timeout(guc->ct.wq,
> !exec_queue_pending_enable(q) ||
> - xe_guc_read_stopped(guc), HZ * 5);
> - if (!ret || xe_guc_read_stopped(guc)) {
> + xe_guc_read_stopped(guc) ||
> + vf_recovery(guc), HZ * 5);
> + if ((!ret && !vf_recovery(guc)) || xe_guc_read_stopped(guc)) {
> xe_gt_warn(guc_to_gt(guc), "Schedule enable failed to respond");
> set_exec_queue_banned(q);
> xe_gt_reset_async(q->gt);
> @@ -1491,7 +1504,8 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
> * list so job can be freed and kick scheduler ensuring free job is not
> * lost.
> */
> - if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &job->fence->flags))
> + if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &job->fence->flags) ||
> + vf_recovery(guc))
> return DRM_GPU_SCHED_STAT_NO_HANG;
>
> /* Kill the run_job entry point */
> @@ -1543,7 +1557,10 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
> ret = wait_event_timeout(guc->ct.wq,
> (!exec_queue_pending_enable(q) &&
> !exec_queue_pending_disable(q)) ||
> - xe_guc_read_stopped(guc), HZ * 5);
> + xe_guc_read_stopped(guc) ||
> + vf_recovery(guc), HZ * 5);
> + if (vf_recovery(guc))
> + goto handle_vf_resume;
> if (!ret || xe_guc_read_stopped(guc))
> goto trigger_reset;
>
> @@ -1568,7 +1585,10 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
> smp_rmb();
> ret = wait_event_timeout(guc->ct.wq,
> !exec_queue_pending_disable(q) ||
> - xe_guc_read_stopped(guc), HZ * 5);
> + xe_guc_read_stopped(guc) ||
> + vf_recovery(guc), HZ * 5);
> + if (vf_recovery(guc))
> + goto handle_vf_resume;
> if (!ret || xe_guc_read_stopped(guc)) {
> trigger_reset:
> if (!ret)
> @@ -1673,6 +1693,7 @@ guc_exec_queue_timedout_job(struct drm_sched_job *drm_job)
> * some thought, do this in a follow up.
> */
> xe_sched_submission_start(sched);
> +handle_vf_resume:
> return DRM_GPU_SCHED_STAT_NO_HANG;
> }
>
> @@ -1794,8 +1815,9 @@ static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg)
>
> if (guc_exec_queue_allowed_to_change_state(q) && !exec_queue_suspended(q) &&
> exec_queue_enabled(q)) {
> - wait_event(guc->ct.wq, (q->guc->resume_time != RESUME_PENDING ||
> - xe_guc_read_stopped(guc)) && !exec_queue_pending_disable(q));
> + wait_event(guc->ct.wq, vf_recovery(guc) ||
> + ((q->guc->resume_time != RESUME_PENDING ||
> + xe_guc_read_stopped(guc)) && !exec_queue_pending_disable(q)));
>
> if (!xe_guc_read_stopped(guc)) {
> s64 since_resume_ms =
> @@ -1922,7 +1944,7 @@ static int guc_exec_queue_init(struct xe_exec_queue *q)
>
> q->entity = &ge->entity;
>
> - if (xe_guc_read_stopped(guc))
> + if (xe_guc_read_stopped(guc) || vf_recovery(guc))
> xe_sched_stop(sched);
>
> mutex_unlock(&guc->submission_state.lock);
> @@ -2075,12 +2097,15 @@ static int guc_exec_queue_suspend_wait(struct xe_exec_queue *q)
> * suspend_pending upon kill but to be paranoid but races in which
> * suspend_pending is set after kill also check kill here.
> */
> +retry:
> ret = wait_event_interruptible_timeout(q->guc->suspend_wait,
> !READ_ONCE(q->guc->suspend_pending) ||
> exec_queue_killed(q) ||
> xe_guc_read_stopped(guc),
> HZ * 5);
>
> + if (vf_recovery(guc) && !xe_device_wedged((guc_to_xe(guc))))
> + goto retry;
This retry loop can lead to a deadlock on VF post-migration recovery. It
appears this work queue / item not finishing during VF post-migration
recovery can interfere with VF post-migration recovery worker getting
scheduled. I have solution to return -EAGAIN if vf_recovery is true
which the caller catches and reschedules the preempt fence work item.
This appears to break the potential deadlock esnuring the VF
post-migration recovery work item gets scheduled.
Matt
> if (!ret) {
> xe_gt_warn(guc_to_gt(guc),
> "Suspend fence, guc_id=%d, failed to respond",
> @@ -2187,7 +2212,7 @@ int xe_guc_submit_reset_prepare(struct xe_guc *guc)
> {
> int ret;
>
> - if (WARN_ON_ONCE(xe_gt_sriov_vf_recovery_inprogress(guc_to_gt(guc))))
> + if (WARN_ON_ONCE(vf_recovery(guc)))
> return 0;
>
> if (!guc->submission_state.initialized)
> --
> 2.34.1
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (17 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 18/34] drm/xe/vf: Wakeup in GuC backend on " Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-27 3:16 ` Lis, Tomasz
2025-09-27 11:06 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 20/34] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register Matthew Brost
` (17 subsequent siblings)
36 siblings, 2 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
A little extra debug on GGTT shift printing the GGT shift value is
helpful for VF post-migration recovery.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index a96e0dd65bc1..e79ab4a2a273 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -475,8 +475,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
primary_config->ggtt_base = start;
err = config->ggtt_size ? 0 : -ENODATA;
- if (!err && shift && recovery)
+ if (!err && shift && recovery) {
+ xe_gt_sriov_info(gt, "Shifting GGTT %lld -> 0x%016llx\n",
+ shift, config->ggtt_base);
xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
+ }
out:
up_write(config->lock);
return err;
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift
2025-09-24 1:15 ` [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift Matthew Brost
@ 2025-09-27 3:16 ` Lis, Tomasz
2025-09-27 11:06 ` Michal Wajdeczko
1 sibling, 0 replies; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-27 3:16 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> A little extra debug on GGTT shift printing the GGT shift value is
> helpful for VF post-migration recovery.
Right, good. It is possible to compute the shift from "GGTT" message
printed earlier and the same during first probe, but this has higher log
level and also allows to confirm that node fixups were entered.
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
-Tomasz
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index a96e0dd65bc1..e79ab4a2a273 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -475,8 +475,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
> primary_config->ggtt_base = start;
> err = config->ggtt_size ? 0 : -ENODATA;
>
> - if (!err && shift && recovery)
> + if (!err && shift && recovery) {
> + xe_gt_sriov_info(gt, "Shifting GGTT %lld -> 0x%016llx\n",
> + shift, config->ggtt_base);
> xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
> + }
> out:
> up_write(config->lock);
> return err;
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift
2025-09-24 1:15 ` [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift Matthew Brost
2025-09-27 3:16 ` Lis, Tomasz
@ 2025-09-27 11:06 ` Michal Wajdeczko
2025-09-27 22:56 ` Matthew Brost
1 sibling, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-27 11:06 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> A little extra debug on GGTT shift printing the GGT shift value is
s/GGT/GGTT
and maybe consider some rewording as I'm not sure it's right
> helpful for VF post-migration recovery.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index a96e0dd65bc1..e79ab4a2a273 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -475,8 +475,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
> primary_config->ggtt_base = start;
> err = config->ggtt_size ? 0 : -ENODATA;
>
> - if (!err && shift && recovery)
> + if (!err && shift && recovery) {
> + xe_gt_sriov_info(gt, "Shifting GGTT %lld -> 0x%016llx\n",
> + shift, config->ggtt_base);
since this is "info" level message, maybe we can make it as:
"Shifting GGTT base by %lld to 0x%016llx\n"
> xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
and print this from within this tile based function, where we could
use tile-based printk function (instead of currently used GT-based)
> + }
> out:
> up_write(config->lock);
> return err;
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift
2025-09-27 11:06 ` Michal Wajdeczko
@ 2025-09-27 22:56 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-27 22:56 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Sat, Sep 27, 2025 at 01:06:32PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > A little extra debug on GGTT shift printing the GGT shift value is
>
> s/GGT/GGTT
>
Yes.
> and maybe consider some rewording as I'm not sure it's right
>
Let me reword this.
> > helpful for VF post-migration recovery.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 5 ++++-
> > 1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index a96e0dd65bc1..e79ab4a2a273 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -475,8 +475,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
> > primary_config->ggtt_base = start;
> > err = config->ggtt_size ? 0 : -ENODATA;
> >
> > - if (!err && shift && recovery)
> > + if (!err && shift && recovery) {
> > + xe_gt_sriov_info(gt, "Shifting GGTT %lld -> 0x%016llx\n",
> > + shift, config->ggtt_base);
>
> since this is "info" level message, maybe we can make it as:
>
> "Shifting GGTT base by %lld to 0x%016llx\n"
>
Sure.
> > xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
>
> and print this from within this tile based function, where we could
> use tile-based printk function (instead of currently used GT-based)
>
Not sure about this one. It is helpful to see which GT is performing the
shift which manually can be used to verify either GT doing the shift.
Matt
> > + }
> > out:
> > up_write(config->lock);
> > return err;
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 20/34] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (18 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 19/34] drm/xe/vf: Extra debug on GGTT shift Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 11:15 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 21/34] drm/xe/vf: Stop and flush CTs in VF post migration recovery Matthew Brost
` (16 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
The only case where the GuC submission backend cannot reason 100%
correctly is when a GuC context is registered during VF post-migration
recovery. In this scenario, it's possible that the GuC context register
H2G is processed, but the immediately following schedule-enable H2G gets
lost.
A double register is harmless when using `GUC_HXG_TYPE_EVENT`, as GuC
simply drops the duplicate H2G. To keep things simple, use
`GUC_HXG_TYPE_EVENT` for all context registrations on VFs.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_guc_ct.c | 32 ++++++++++++++++++++++++--------
1 file changed, 24 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 661ab1bf4502..aa9fb08f7856 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -737,6 +737,26 @@ static u16 next_ct_seqno(struct xe_guc_ct *ct, bool is_g2h_fence)
return seqno;
}
+#define MAKE_ACTION(type, __action) \
+({ \
+ FIELD_PREP(GUC_HXG_MSG_0_TYPE, type) | \
+ FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION | \
+ GUC_HXG_EVENT_MSG_0_DATA0, __action); \
+})
+
+static bool vf_action_can_safely_fail(struct xe_device *xe, u32 action)
+{
+ /*
+ * If we are VF resuming, we can't exactly track if a context
+ * registration has been completed in the GuC state machine, it is
+ * harmless to resend as it will just fail silently if
+ * GUC_HXG_TYPE_EVENT is used.
+ */
+ return IS_SRIOV_VF(xe) &&
+ (action == XE_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC ||
+ action == XE_GUC_ACTION_REGISTER_CONTEXT);
+}
+
#define H2G_CT_HEADERS (GUC_CTB_HDR_LEN + 1) /* one DW CTB header and one DW HxG header */
static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
@@ -808,18 +828,14 @@ static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
FIELD_PREP(GUC_CTB_MSG_0_NUM_DWORDS, len) |
FIELD_PREP(GUC_CTB_MSG_0_FENCE, ct_fence_value);
if (want_response) {
- cmd[1] =
- FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_REQUEST) |
- FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
- GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
+ cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_REQUEST, action[0]);
+ } else if (vf_action_can_safely_fail(xe, action[0])) {
+ cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_EVENT, action[0]);
} else {
fast_req_track(ct, ct_fence_value,
FIELD_GET(GUC_HXG_EVENT_MSG_0_ACTION, action[0]));
- cmd[1] =
- FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_FAST_REQUEST) |
- FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
- GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
+ cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_FAST_REQUEST, action[0]);
}
/* H2G header in cmd[1] replaces action[0] so: */
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 20/34] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register
2025-09-24 1:15 ` [PATCH v2 20/34] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register Matthew Brost
@ 2025-09-24 11:15 ` Michal Wajdeczko
2025-09-24 20:16 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 11:15 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> The only case where the GuC submission backend cannot reason 100%
> correctly is when a GuC context is registered during VF post-migration
> recovery. In this scenario, it's possible that the GuC context register
> H2G is processed, but the immediately following schedule-enable H2G gets
> lost.
what do you mean by "lost"?
if it was not processed by the SRC GuC, then it is still in CTB and shall
be processed by DST GuC, unless we strip it from the CTB by ourselves, but
in this case we should notice that, no?
>
> A double register is harmless when using `GUC_HXG_TYPE_EVENT`, as GuC
> simply drops the duplicate H2G. To keep things simple, use
> `GUC_HXG_TYPE_EVENT` for all context registrations on VFs.
with this we will miss any early errors from the GuC on non-migration cases
while unlikely, IMO we shouldn't just hide potential errors
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_guc_ct.c | 32 ++++++++++++++++++++++++--------
> 1 file changed, 24 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> index 661ab1bf4502..aa9fb08f7856 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> @@ -737,6 +737,26 @@ static u16 next_ct_seqno(struct xe_guc_ct *ct, bool is_g2h_fence)
> return seqno;
> }
>
> +#define MAKE_ACTION(type, __action) \
> +({ \
> + FIELD_PREP(GUC_HXG_MSG_0_TYPE, type) | \
> + FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION | \
> + GUC_HXG_EVENT_MSG_0_DATA0, __action); \
> +})
> +
> +static bool vf_action_can_safely_fail(struct xe_device *xe, u32 action)
> +{
> + /*
> + * If we are VF resuming, we can't exactly track if a context
> + * registration has been completed in the GuC state machine, it is
> + * harmless to resend as it will just fail silently if
> + * GUC_HXG_TYPE_EVENT is used.
> + */
> + return IS_SRIOV_VF(xe) &&
> + (action == XE_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC ||
> + action == XE_GUC_ACTION_REGISTER_CONTEXT);
> +}
> +
> #define H2G_CT_HEADERS (GUC_CTB_HDR_LEN + 1) /* one DW CTB header and one DW HxG header */
>
> static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
> @@ -808,18 +828,14 @@ static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
> FIELD_PREP(GUC_CTB_MSG_0_NUM_DWORDS, len) |
> FIELD_PREP(GUC_CTB_MSG_0_FENCE, ct_fence_value);
> if (want_response) {
> - cmd[1] =
> - FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_REQUEST) |
> - FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
> - GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
> + cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_REQUEST, action[0]);
> + } else if (vf_action_can_safely_fail(xe, action[0])) {
> + cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_EVENT, action[0]);
> } else {
> fast_req_track(ct, ct_fence_value,
> FIELD_GET(GUC_HXG_EVENT_MSG_0_ACTION, action[0]));
>
> - cmd[1] =
> - FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_FAST_REQUEST) |
> - FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
> - GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
> + cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_FAST_REQUEST, action[0]);
> }
>
> /* H2G header in cmd[1] replaces action[0] so: */
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 20/34] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register
2025-09-24 11:15 ` Michal Wajdeczko
@ 2025-09-24 20:16 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:16 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 01:15:12PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > The only case where the GuC submission backend cannot reason 100%
> > correctly is when a GuC context is registered during VF post-migration
> > recovery. In this scenario, it's possible that the GuC context register
> > H2G is processed, but the immediately following schedule-enable H2G gets
> > lost.
>
> what do you mean by "lost"?
>
> if it was not processed by the SRC GuC, then it is still in CTB and shall
> be processed by DST GuC, unless we strip it from the CTB by ourselves, but
> in this case we should notice that, no?
>
We do strip out the CTBs and GuC submission state machine will reply register H2G.
> >
> > A double register is harmless when using `GUC_HXG_TYPE_EVENT`, as GuC
> > simply drops the duplicate H2G. To keep things simple, use
> > `GUC_HXG_TYPE_EVENT` for all context registrations on VFs.
>
> with this we will miss any early errors from the GuC on non-migration cases
> while unlikely, IMO we shouldn't just hide potential errors
>
I agree this isn't ideal but this check is present for non-VFs. The
registration code is generic so if we had a protocol problem our CI
would quickly catch this on non-VFs. A fast request failure is fatal to
the KMD as we take down th CT and issue GT reset when this happens.
IMO we live with this limitation.
Matt
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_guc_ct.c | 32 ++++++++++++++++++++++++--------
> > 1 file changed, 24 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> > index 661ab1bf4502..aa9fb08f7856 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> > @@ -737,6 +737,26 @@ static u16 next_ct_seqno(struct xe_guc_ct *ct, bool is_g2h_fence)
> > return seqno;
> > }
> >
> > +#define MAKE_ACTION(type, __action) \
> > +({ \
> > + FIELD_PREP(GUC_HXG_MSG_0_TYPE, type) | \
> > + FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION | \
> > + GUC_HXG_EVENT_MSG_0_DATA0, __action); \
> > +})
> > +
> > +static bool vf_action_can_safely_fail(struct xe_device *xe, u32 action)
> > +{
> > + /*
> > + * If we are VF resuming, we can't exactly track if a context
> > + * registration has been completed in the GuC state machine, it is
> > + * harmless to resend as it will just fail silently if
> > + * GUC_HXG_TYPE_EVENT is used.
> > + */
> > + return IS_SRIOV_VF(xe) &&
> > + (action == XE_GUC_ACTION_REGISTER_CONTEXT_MULTI_LRC ||
> > + action == XE_GUC_ACTION_REGISTER_CONTEXT);
> > +}
> > +
> > #define H2G_CT_HEADERS (GUC_CTB_HDR_LEN + 1) /* one DW CTB header and one DW HxG header */
> >
> > static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
> > @@ -808,18 +828,14 @@ static int h2g_write(struct xe_guc_ct *ct, const u32 *action, u32 len,
> > FIELD_PREP(GUC_CTB_MSG_0_NUM_DWORDS, len) |
> > FIELD_PREP(GUC_CTB_MSG_0_FENCE, ct_fence_value);
> > if (want_response) {
> > - cmd[1] =
> > - FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_REQUEST) |
> > - FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
> > - GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
> > + cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_REQUEST, action[0]);
> > + } else if (vf_action_can_safely_fail(xe, action[0])) {
> > + cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_EVENT, action[0]);
> > } else {
> > fast_req_track(ct, ct_fence_value,
> > FIELD_GET(GUC_HXG_EVENT_MSG_0_ACTION, action[0]));
> >
> > - cmd[1] =
> > - FIELD_PREP(GUC_HXG_MSG_0_TYPE, GUC_HXG_TYPE_FAST_REQUEST) |
> > - FIELD_PREP(GUC_HXG_EVENT_MSG_0_ACTION |
> > - GUC_HXG_EVENT_MSG_0_DATA0, action[0]);
> > + cmd[1] = MAKE_ACTION(GUC_HXG_TYPE_FAST_REQUEST, action[0]);
> > }
> >
> > /* H2G header in cmd[1] replaces action[0] so: */
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 21/34] drm/xe/vf: Stop and flush CTs in VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (19 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 20/34] drm/xe/vf: Use GUC_HXG_TYPE_EVENT for GuC context register Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 11:21 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 22/34] drm/xe/vf: Reset TLB invalidations during " Matthew Brost
` (15 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Flushing CTs (i.e., progressing all pending G2H messages) gives VF
post-migration recovery an accurate view of which H2G messages the GuC
has processed, enabling the GuC submission state machine to correctly
rebuild all state.
Also, stop all CT traffic, as the CT is not live during VF
post-migration recovery.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 2 ++
drivers/gpu/drm/xe/xe_guc_ct.c | 11 +++++++++++
drivers/gpu/drm/xe/xe_guc_ct.h | 1 +
3 files changed, 14 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index e79ab4a2a273..071c2c5e0d0a 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -23,6 +23,7 @@
#include "xe_gt_sriov_vf.h"
#include "xe_gt_sriov_vf_types.h"
#include "xe_guc.h"
+#include "xe_guc_ct.h"
#include "xe_guc_hxg_helpers.h"
#include "xe_guc_relay.h"
#include "xe_guc_submit.h"
@@ -1185,6 +1186,7 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
gt->sriov.vf.migration.recovery_queued = false;
spin_unlock_irq(>->sriov.vf.migration.lock);
+ xe_guc_ct_flush(>->uc.guc.ct);
xe_guc_submit_pause(>->uc.guc);
}
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index aa9fb08f7856..d4a132b81b86 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -574,6 +574,17 @@ void xe_guc_ct_disable(struct xe_guc_ct *ct)
stop_g2h_handler(ct);
}
+/**
+ * xe_guc_ct_flush - Flush all processing of G2H / H2G
+ * @ct: the &xe_guc_ct
+ */
+void xe_guc_ct_flush(struct xe_guc_ct *ct)
+{
+ receive_g2h(ct);
+ wake_up_all(&ct->g2h_fence_wq);
+ xe_guc_ct_stop(ct);
+}
+
/**
* xe_guc_ct_stop - Set GuC to stopped state
* @ct: the &xe_guc_ct
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
index d6c81325a76c..1e32a59817cc 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.h
+++ b/drivers/gpu/drm/xe/xe_guc_ct.h
@@ -17,6 +17,7 @@ int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
int xe_guc_ct_enable(struct xe_guc_ct *ct);
void xe_guc_ct_disable(struct xe_guc_ct *ct);
void xe_guc_ct_stop(struct xe_guc_ct *ct);
+void xe_guc_ct_flush(struct xe_guc_ct *ct);
void xe_guc_ct_fast_path(struct xe_guc_ct *ct);
struct xe_guc_ct_snapshot *xe_guc_ct_snapshot_capture(struct xe_guc_ct *ct);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 21/34] drm/xe/vf: Stop and flush CTs in VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 21/34] drm/xe/vf: Stop and flush CTs in VF post migration recovery Matthew Brost
@ 2025-09-24 11:21 ` Michal Wajdeczko
2025-09-24 20:12 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 11:21 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Flushing CTs (i.e., progressing all pending G2H messages) gives VF
> post-migration recovery an accurate view of which H2G messages the GuC
> has processed, enabling the GuC submission state machine to correctly
> rebuild all state.
>
> Also, stop all CT traffic, as the CT is not live during VF
> post-migration recovery.
this is what I was looking for, some comments below, but LGTM
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 2 ++
> drivers/gpu/drm/xe/xe_guc_ct.c | 11 +++++++++++
> drivers/gpu/drm/xe/xe_guc_ct.h | 1 +
> 3 files changed, 14 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index e79ab4a2a273..071c2c5e0d0a 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -23,6 +23,7 @@
> #include "xe_gt_sriov_vf.h"
> #include "xe_gt_sriov_vf_types.h"
> #include "xe_guc.h"
> +#include "xe_guc_ct.h"
> #include "xe_guc_hxg_helpers.h"
> #include "xe_guc_relay.h"
> #include "xe_guc_submit.h"
> @@ -1185,6 +1186,7 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
> gt->sriov.vf.migration.recovery_queued = false;
> spin_unlock_irq(>->sriov.vf.migration.lock);
>
> + xe_guc_ct_flush(>->uc.guc.ct);
> xe_guc_submit_pause(>->uc.guc);
> }
>
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> index aa9fb08f7856..d4a132b81b86 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> @@ -574,6 +574,17 @@ void xe_guc_ct_disable(struct xe_guc_ct *ct)
> stop_g2h_handler(ct);
> }
>
> +/**
> + * xe_guc_ct_flush - Flush all processing of G2H / H2G
Flush and stop ?
> + * @ct: the &xe_guc_ct
> + */
> +void xe_guc_ct_flush(struct xe_guc_ct *ct)
> +{
do we care about ct being not initialized yet?
> + receive_g2h(ct);
> + wake_up_all(&ct->g2h_fence_wq);
this is part of the guc_ct_change_state() called by below stop
isn't that sufficient ?
> + xe_guc_ct_stop(ct);
> +}
> +
> /**
> * xe_guc_ct_stop - Set GuC to stopped state
> * @ct: the &xe_guc_ct
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
> index d6c81325a76c..1e32a59817cc 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.h
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.h
> @@ -17,6 +17,7 @@ int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
> int xe_guc_ct_enable(struct xe_guc_ct *ct);
> void xe_guc_ct_disable(struct xe_guc_ct *ct);
> void xe_guc_ct_stop(struct xe_guc_ct *ct);
> +void xe_guc_ct_flush(struct xe_guc_ct *ct);
> void xe_guc_ct_fast_path(struct xe_guc_ct *ct);
>
> struct xe_guc_ct_snapshot *xe_guc_ct_snapshot_capture(struct xe_guc_ct *ct);
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 21/34] drm/xe/vf: Stop and flush CTs in VF post migration recovery
2025-09-24 11:21 ` Michal Wajdeczko
@ 2025-09-24 20:12 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:12 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 01:21:19PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Flushing CTs (i.e., progressing all pending G2H messages) gives VF
> > post-migration recovery an accurate view of which H2G messages the GuC
> > has processed, enabling the GuC submission state machine to correctly
> > rebuild all state.
> >
> > Also, stop all CT traffic, as the CT is not live during VF
> > post-migration recovery.
>
> this is what I was looking for, some comments below, but LGTM
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 2 ++
> > drivers/gpu/drm/xe/xe_guc_ct.c | 11 +++++++++++
> > drivers/gpu/drm/xe/xe_guc_ct.h | 1 +
> > 3 files changed, 14 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index e79ab4a2a273..071c2c5e0d0a 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -23,6 +23,7 @@
> > #include "xe_gt_sriov_vf.h"
> > #include "xe_gt_sriov_vf_types.h"
> > #include "xe_guc.h"
> > +#include "xe_guc_ct.h"
> > #include "xe_guc_hxg_helpers.h"
> > #include "xe_guc_relay.h"
> > #include "xe_guc_submit.h"
> > @@ -1185,6 +1186,7 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
> > gt->sriov.vf.migration.recovery_queued = false;
> > spin_unlock_irq(>->sriov.vf.migration.lock);
> >
> > + xe_guc_ct_flush(>->uc.guc.ct);
> > xe_guc_submit_pause(>->uc.guc);
> > }
> >
> > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> > index aa9fb08f7856..d4a132b81b86 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> > @@ -574,6 +574,17 @@ void xe_guc_ct_disable(struct xe_guc_ct *ct)
> > stop_g2h_handler(ct);
> > }
> >
> > +/**
> > + * xe_guc_ct_flush - Flush all processing of G2H / H2G
>
> Flush and stop ?
>
Sure.
> > + * @ct: the &xe_guc_ct
> > + */
> > +void xe_guc_ct_flush(struct xe_guc_ct *ct)
> > +{
>
> do we care about ct being not initialized yet?
>
I don't see how a VF can be moved if the driver isn't loaded. Maybe can
an assert here to ensure this assumption is correct?
> > + receive_g2h(ct);
> > + wake_up_all(&ct->g2h_fence_wq);
>
> this is part of the guc_ct_change_state() called by below stop
> isn't that sufficient ?
>
Agree not needed, let me drop this.
Matt
> > + xe_guc_ct_stop(ct);
> > +}
> > +
> > /**
> > * xe_guc_ct_stop - Set GuC to stopped state
> > * @ct: the &xe_guc_ct
> > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
> > index d6c81325a76c..1e32a59817cc 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_ct.h
> > +++ b/drivers/gpu/drm/xe/xe_guc_ct.h
> > @@ -17,6 +17,7 @@ int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
> > int xe_guc_ct_enable(struct xe_guc_ct *ct);
> > void xe_guc_ct_disable(struct xe_guc_ct *ct);
> > void xe_guc_ct_stop(struct xe_guc_ct *ct);
> > +void xe_guc_ct_flush(struct xe_guc_ct *ct);
> > void xe_guc_ct_fast_path(struct xe_guc_ct *ct);
> >
> > struct xe_guc_ct_snapshot *xe_guc_ct_snapshot_capture(struct xe_guc_ct *ct);
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 22/34] drm/xe/vf: Reset TLB invalidations during VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (20 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 21/34] drm/xe/vf: Stop and flush CTs in VF post migration recovery Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-27 3:43 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 23/34] drm/xe/vf: Kickstart after resfix in " Matthew Brost
` (14 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
TLB invalidations can be lost during VF post-migration recovery. Since
the VF has migrated, these invalidations are no longer needed.
Reset the TLB invalidation frontend, which will signal all pending
fences.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 071c2c5e0d0a..01399a871f12 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -35,6 +35,7 @@
#include "xe_sriov.h"
#include "xe_sriov_vf.h"
#include "xe_tile_sriov_vf.h"
+#include "xe_tlb_inval.h"
#include "xe_uc_fw.h"
#include "xe_wopcm.h"
@@ -1187,6 +1188,7 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
spin_unlock_irq(>->sriov.vf.migration.lock);
xe_guc_ct_flush(>->uc.guc.ct);
+ xe_tlb_inval_reset(>->tlb_inval);
xe_guc_submit_pause(>->uc.guc);
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 22/34] drm/xe/vf: Reset TLB invalidations during VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 22/34] drm/xe/vf: Reset TLB invalidations during " Matthew Brost
@ 2025-09-27 3:43 ` Lis, Tomasz
2025-09-27 22:29 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-27 3:43 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> TLB invalidations can be lost during VF post-migration recovery. Since
> the VF has migrated, these invalidations are no longer needed.
>
> Reset the TLB invalidation frontend, which will signal all pending
> fences.
Can something other than timeout wait on these fences? Because if they
trigger
additional processes, then it would make sense to wait with this.
If this can trigger submissions, then doing this after submit_pause
would make sense.
If can trigger even more, then moving this to `kickstart` (to after
recovery) may be a good idea.
-Tomasz
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 071c2c5e0d0a..01399a871f12 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -35,6 +35,7 @@
> #include "xe_sriov.h"
> #include "xe_sriov_vf.h"
> #include "xe_tile_sriov_vf.h"
> +#include "xe_tlb_inval.h"
> #include "xe_uc_fw.h"
> #include "xe_wopcm.h"
>
> @@ -1187,6 +1188,7 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
> spin_unlock_irq(>->sriov.vf.migration.lock);
>
> xe_guc_ct_flush(>->uc.guc.ct);
> + xe_tlb_inval_reset(>->tlb_inval);
> xe_guc_submit_pause(>->uc.guc);
> }
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* Re: [PATCH v2 22/34] drm/xe/vf: Reset TLB invalidations during VF post migration recovery
2025-09-27 3:43 ` Lis, Tomasz
@ 2025-09-27 22:29 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-27 22:29 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Sat, Sep 27, 2025 at 05:43:57AM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > TLB invalidations can be lost during VF post-migration recovery. Since
> > the VF has migrated, these invalidations are no longer needed.
> >
> > Reset the TLB invalidation frontend, which will signal all pending
> > fences.
>
> Can something other than timeout wait on these fences? Because if they
> trigger
The hardware via a G2H - xe_guc_ct_flush_and_stop find any remaining
G2H, the timeout mechanism, the CT channel being stopped or via
xe_tlb_inval_reset are th ways tlb invalidation fences signal.
>
> additional processes, then it would make sense to wait with this.
>
> If this can trigger submissions, then doing this after submit_pause would
> make sense.
>
Yes, fence signaling can trigger submissions. It is harmless where it is
but logically perhaps after xe_guc_submit_pause makes a bit more sense.
Will move this.
> If can trigger even more, then moving this to `kickstart` (to after
> recovery) may be a good idea.
>
> -Tomasz
>
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 071c2c5e0d0a..01399a871f12 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -35,6 +35,7 @@
> > #include "xe_sriov.h"
> > #include "xe_sriov_vf.h"
> > #include "xe_tile_sriov_vf.h"
> > +#include "xe_tlb_inval.h"
> > #include "xe_uc_fw.h"
> > #include "xe_wopcm.h"
> > @@ -1187,6 +1188,7 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
> > spin_unlock_irq(>->sriov.vf.migration.lock);
> > xe_guc_ct_flush(>->uc.guc.ct);
> > + xe_tlb_inval_reset(>->tlb_inval);
> > xe_guc_submit_pause(>->uc.guc);
> > }
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 23/34] drm/xe/vf: Kickstart after resfix in VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (21 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 22/34] drm/xe/vf: Reset TLB invalidations during " Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-27 11:21 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 24/34] drm/xe/vf: Start CTs before resfix " Matthew Brost
` (13 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
GuC needs to be live for the GuC submission state machine to resubmit
anything lost during VF post-migration recovery. Therefore, move the
kickstart step after `resfix` to ensure proper resubmission.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 18 ++++++++++--------
1 file changed, 10 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 01399a871f12..2a077a98e92e 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1216,13 +1216,6 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
static void vf_post_migration_kickstart(struct xe_gt *gt)
{
- /*
- * Make sure interrupts on the new HW are properly set. The GuC IRQ
- * must be working at this point, since the recovery did started,
- * but the rest was not enabled using the procedure from spec.
- */
- xe_irq_resume(gt_to_xe(gt));
-
xe_guc_submit_unpause(>->uc.guc);
}
@@ -1239,6 +1232,14 @@ static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
}
spin_unlock_irq(>->sriov.vf.migration.lock);
+ /*
+ * Make sure interrupts on the new HW are properly set. The GuC IRQ
+ * must be working at this point, since the recovery did started,
+ * but the rest was not enabled using the procedure from spec.
+ */
+ if (!skip_resfix)
+ xe_irq_resume(gt_to_xe(gt));
+
return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
}
@@ -1262,11 +1263,12 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
if (err)
goto fail;
- vf_post_migration_kickstart(gt);
err = vf_post_migration_notify_resfix_done(gt);
if (err && err != -EAGAIN)
goto fail;
+ vf_post_migration_kickstart(gt);
+
xe_pm_runtime_put(xe);
xe_gt_sriov_notice(gt, "migration recovery ended\n");
return;
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 23/34] drm/xe/vf: Kickstart after resfix in VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 23/34] drm/xe/vf: Kickstart after resfix in " Matthew Brost
@ 2025-09-27 11:21 ` Lis, Tomasz
0 siblings, 0 replies; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-27 11:21 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> GuC needs to be live for the GuC submission state machine to resubmit
> anything lost during VF post-migration recovery. Therefore, move the
> kickstart step after `resfix` to ensure proper resubmission.
All good.
Reviewed-by: Tomasz Lis <tomasz.lis@intel.com>
-Tomasz
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 18 ++++++++++--------
> 1 file changed, 10 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 01399a871f12..2a077a98e92e 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -1216,13 +1216,6 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
>
> static void vf_post_migration_kickstart(struct xe_gt *gt)
> {
> - /*
> - * Make sure interrupts on the new HW are properly set. The GuC IRQ
> - * must be working at this point, since the recovery did started,
> - * but the rest was not enabled using the procedure from spec.
> - */
> - xe_irq_resume(gt_to_xe(gt));
> -
> xe_guc_submit_unpause(>->uc.guc);
> }
>
> @@ -1239,6 +1232,14 @@ static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> }
> spin_unlock_irq(>->sriov.vf.migration.lock);
>
> + /*
> + * Make sure interrupts on the new HW are properly set. The GuC IRQ
> + * must be working at this point, since the recovery did started,
> + * but the rest was not enabled using the procedure from spec.
> + */
> + if (!skip_resfix)
> + xe_irq_resume(gt_to_xe(gt));
> +
> return skip_resfix ? -EAGAIN : xe_gt_sriov_vf_notify_resfix_done(gt);
> }
>
> @@ -1262,11 +1263,12 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
> if (err)
> goto fail;
>
> - vf_post_migration_kickstart(gt);
> err = vf_post_migration_notify_resfix_done(gt);
> if (err && err != -EAGAIN)
> goto fail;
>
> + vf_post_migration_kickstart(gt);
> +
> xe_pm_runtime_put(xe);
> xe_gt_sriov_notice(gt, "migration recovery ended\n");
> return;
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 24/34] drm/xe/vf: Start CTs before resfix VF post migration recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (22 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 23/34] drm/xe/vf: Kickstart after resfix in " Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 11:50 ` Michal Wajdeczko
2025-09-24 1:15 ` [PATCH v2 25/34] drm/xe/vf: Abort VF post migration recovery on failure Matthew Brost
` (12 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Before `resfix`, all CTs stuck in the H2G queue need to be squashed, as
they may contain stale or invalid data.
Starting the CTs clears all H2Gs in the queue. Any lost H2Gs are
resubmitted by the GuC submission state machine.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 7 +++++
drivers/gpu/drm/xe/xe_guc_ct.c | 48 +++++++++++++++++++++++------
drivers/gpu/drm/xe/xe_guc_ct.h | 1 +
3 files changed, 46 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 2a077a98e92e..34e57f44fbf4 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1214,6 +1214,11 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
return 0;
}
+static void vf_post_migration_rearm(struct xe_gt *gt)
+{
+ xe_guc_ct_restart(>->uc.guc.ct);
+}
+
static void vf_post_migration_kickstart(struct xe_gt *gt)
{
xe_guc_submit_unpause(>->uc.guc);
@@ -1263,6 +1268,8 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
if (err)
goto fail;
+ vf_post_migration_rearm(gt);
+
err = vf_post_migration_notify_resfix_done(gt);
if (err && err != -EAGAIN)
goto fail;
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index d4a132b81b86..fa55e9c9de3a 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -501,7 +501,7 @@ static void ct_exit_safe_mode(struct xe_guc_ct *ct)
xe_gt_dbg(ct_to_gt(ct), "GuC CT safe-mode disabled\n");
}
-int xe_guc_ct_enable(struct xe_guc_ct *ct)
+static int __xe_guc_ct_start(struct xe_guc_ct *ct, bool needs_register)
{
struct xe_device *xe = ct_to_xe(ct);
struct xe_gt *gt = ct_to_gt(ct);
@@ -513,17 +513,19 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct)
guc_ct_ctb_h2g_init(xe, &ct->ctbs.h2g, &ct->bo->vmap);
guc_ct_ctb_g2h_init(xe, &ct->ctbs.g2h, &ct->bo->vmap);
- err = guc_ct_ctb_h2g_register(ct);
- if (err)
- goto err_out;
+ if (needs_register) {
+ err = guc_ct_ctb_h2g_register(ct);
+ if (err)
+ goto err_out;
- err = guc_ct_ctb_g2h_register(ct);
- if (err)
- goto err_out;
+ err = guc_ct_ctb_g2h_register(ct);
+ if (err)
+ goto err_out;
- err = guc_ct_control_toggle(ct, true);
- if (err)
- goto err_out;
+ err = guc_ct_control_toggle(ct, true);
+ if (err)
+ goto err_out;
+ }
guc_ct_change_state(ct, XE_GUC_CT_STATE_ENABLED);
@@ -555,6 +557,32 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct)
return err;
}
+/**
+ * xe_guc_ct_restart() - Restart GuC CT
+ * @ct: the &xe_guc_ct
+ *
+ * Restart GuC CT to an empty state without issuing a CT register MMIO command.
+ *
+ * Return: 0 on success, or a negative errno on failure.
+ */
+int xe_guc_ct_restart(struct xe_guc_ct *ct)
+{
+ return __xe_guc_ct_start(ct, false);
+}
+
+/**
+ * xe_guc_ct_enable() - Enable GuC CT
+ * @ct: the &xe_guc_ct
+ *
+ * Enable GuC CT to an empty state and issue a CT register MMIO command.
+ *
+ * Return: 0 on success, or a negative errno on failure.
+ */
+int xe_guc_ct_enable(struct xe_guc_ct *ct)
+{
+ return __xe_guc_ct_start(ct, true);
+}
+
static void stop_g2h_handler(struct xe_guc_ct *ct)
{
cancel_work_sync(&ct->g2h_worker);
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
index 1e32a59817cc..506616ec1db0 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.h
+++ b/drivers/gpu/drm/xe/xe_guc_ct.h
@@ -15,6 +15,7 @@ int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct);
int xe_guc_ct_init(struct xe_guc_ct *ct);
int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
int xe_guc_ct_enable(struct xe_guc_ct *ct);
+int xe_guc_ct_restart(struct xe_guc_ct *ct);
void xe_guc_ct_disable(struct xe_guc_ct *ct);
void xe_guc_ct_stop(struct xe_guc_ct *ct);
void xe_guc_ct_flush(struct xe_guc_ct *ct);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 24/34] drm/xe/vf: Start CTs before resfix VF post migration recovery
2025-09-24 1:15 ` [PATCH v2 24/34] drm/xe/vf: Start CTs before resfix " Matthew Brost
@ 2025-09-24 11:50 ` Michal Wajdeczko
2025-09-24 20:10 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Michal Wajdeczko @ 2025-09-24 11:50 UTC (permalink / raw)
To: Matthew Brost, intel-xe
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Before `resfix`, all CTs stuck in the H2G queue need to be squashed, as
> they may contain stale or invalid data.
>
> Starting the CTs clears all H2Gs in the queue. Any lost H2Gs are
> resubmitted by the GuC submission state machine.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 7 +++++
> drivers/gpu/drm/xe/xe_guc_ct.c | 48 +++++++++++++++++++++++------
> drivers/gpu/drm/xe/xe_guc_ct.h | 1 +
> 3 files changed, 46 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 2a077a98e92e..34e57f44fbf4 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -1214,6 +1214,11 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> return 0;
> }
>
> +static void vf_post_migration_rearm(struct xe_gt *gt)
> +{
> + xe_guc_ct_restart(>->uc.guc.ct);
> +}
> +
> static void vf_post_migration_kickstart(struct xe_gt *gt)
> {
> xe_guc_submit_unpause(>->uc.guc);
> @@ -1263,6 +1268,8 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
> if (err)
> goto fail;
>
> + vf_post_migration_rearm(gt);
> +
> err = vf_post_migration_notify_resfix_done(gt);
> if (err && err != -EAGAIN)
> goto fail;
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> index d4a132b81b86..fa55e9c9de3a 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> @@ -501,7 +501,7 @@ static void ct_exit_safe_mode(struct xe_guc_ct *ct)
> xe_gt_dbg(ct_to_gt(ct), "GuC CT safe-mode disabled\n");
> }
>
> -int xe_guc_ct_enable(struct xe_guc_ct *ct)
> +static int __xe_guc_ct_start(struct xe_guc_ct *ct, bool needs_register)
> {
> struct xe_device *xe = ct_to_xe(ct);
> struct xe_gt *gt = ct_to_gt(ct);
> @@ -513,17 +513,19 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct)
> guc_ct_ctb_h2g_init(xe, &ct->ctbs.h2g, &ct->bo->vmap);
> guc_ct_ctb_g2h_init(xe, &ct->ctbs.g2h, &ct->bo->vmap);
I'm not sure that this is aligned with the spec that after CTB
being registered (here needs_register==false) we will modify the
shared CTB descriptor, where GuC owns its head/tail data.
maybe on the restart() case we should just erase what was left
unprocessed by the GuC from our tail to GuC's head in H2G CTB?
as extra bonus we can then look what will be exactly "lost"
>
> - err = guc_ct_ctb_h2g_register(ct);
> - if (err)
> - goto err_out;
> + if (needs_register) {
> + err = guc_ct_ctb_h2g_register(ct);
> + if (err)
> + goto err_out;
>
> - err = guc_ct_ctb_g2h_register(ct);
> - if (err)
> - goto err_out;
> + err = guc_ct_ctb_g2h_register(ct);
> + if (err)
> + goto err_out;
>
> - err = guc_ct_control_toggle(ct, true);
> - if (err)
> - goto err_out;
> + err = guc_ct_control_toggle(ct, true);
> + if (err)
> + goto err_out;
> + }
>
> guc_ct_change_state(ct, XE_GUC_CT_STATE_ENABLED);
>
> @@ -555,6 +557,32 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct)
> return err;
> }
>
> +/**
> + * xe_guc_ct_restart() - Restart GuC CT
> + * @ct: the &xe_guc_ct
> + *
> + * Restart GuC CT to an empty state without issuing a CT register MMIO command.
> + *
> + * Return: 0 on success, or a negative errno on failure.
> + */
> +int xe_guc_ct_restart(struct xe_guc_ct *ct)
> +{
> + return __xe_guc_ct_start(ct, false);
> +}
> +
> +/**
> + * xe_guc_ct_enable() - Enable GuC CT
> + * @ct: the &xe_guc_ct
> + *
> + * Enable GuC CT to an empty state and issue a CT register MMIO command.
> + *
> + * Return: 0 on success, or a negative errno on failure.
> + */
> +int xe_guc_ct_enable(struct xe_guc_ct *ct)
> +{
> + return __xe_guc_ct_start(ct, true);
> +}
> +
> static void stop_g2h_handler(struct xe_guc_ct *ct)
> {
> cancel_work_sync(&ct->g2h_worker);
> diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
> index 1e32a59817cc..506616ec1db0 100644
> --- a/drivers/gpu/drm/xe/xe_guc_ct.h
> +++ b/drivers/gpu/drm/xe/xe_guc_ct.h
> @@ -15,6 +15,7 @@ int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct);
> int xe_guc_ct_init(struct xe_guc_ct *ct);
> int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
> int xe_guc_ct_enable(struct xe_guc_ct *ct);
> +int xe_guc_ct_restart(struct xe_guc_ct *ct);
> void xe_guc_ct_disable(struct xe_guc_ct *ct);
> void xe_guc_ct_stop(struct xe_guc_ct *ct);
> void xe_guc_ct_flush(struct xe_guc_ct *ct);
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 24/34] drm/xe/vf: Start CTs before resfix VF post migration recovery
2025-09-24 11:50 ` Michal Wajdeczko
@ 2025-09-24 20:10 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 20:10 UTC (permalink / raw)
To: Michal Wajdeczko; +Cc: intel-xe
On Wed, Sep 24, 2025 at 01:50:06PM +0200, Michal Wajdeczko wrote:
>
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Before `resfix`, all CTs stuck in the H2G queue need to be squashed, as
> > they may contain stale or invalid data.
> >
> > Starting the CTs clears all H2Gs in the queue. Any lost H2Gs are
> > resubmitted by the GuC submission state machine.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 7 +++++
> > drivers/gpu/drm/xe/xe_guc_ct.c | 48 +++++++++++++++++++++++------
> > drivers/gpu/drm/xe/xe_guc_ct.h | 1 +
> > 3 files changed, 46 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 2a077a98e92e..34e57f44fbf4 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -1214,6 +1214,11 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> > return 0;
> > }
> >
> > +static void vf_post_migration_rearm(struct xe_gt *gt)
> > +{
> > + xe_guc_ct_restart(>->uc.guc.ct);
> > +}
> > +
> > static void vf_post_migration_kickstart(struct xe_gt *gt)
> > {
> > xe_guc_submit_unpause(>->uc.guc);
> > @@ -1263,6 +1268,8 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
> > if (err)
> > goto fail;
> >
> > + vf_post_migration_rearm(gt);
> > +
> > err = vf_post_migration_notify_resfix_done(gt);
> > if (err && err != -EAGAIN)
> > goto fail;
> > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
> > index d4a132b81b86..fa55e9c9de3a 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_ct.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_ct.c
> > @@ -501,7 +501,7 @@ static void ct_exit_safe_mode(struct xe_guc_ct *ct)
> > xe_gt_dbg(ct_to_gt(ct), "GuC CT safe-mode disabled\n");
> > }
> >
> > -int xe_guc_ct_enable(struct xe_guc_ct *ct)
> > +static int __xe_guc_ct_start(struct xe_guc_ct *ct, bool needs_register)
> > {
> > struct xe_device *xe = ct_to_xe(ct);
> > struct xe_gt *gt = ct_to_gt(ct);
> > @@ -513,17 +513,19 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct)
> > guc_ct_ctb_h2g_init(xe, &ct->ctbs.h2g, &ct->bo->vmap);
> > guc_ct_ctb_g2h_init(xe, &ct->ctbs.g2h, &ct->bo->vmap);
>
> I'm not sure that this is aligned with the spec that after CTB
Not to rant to much about specs, but those are frequently wrong.
> being registered (here needs_register==false) we will modify the
> shared CTB descriptor, where GuC owns its head/tail data.
>
> maybe on the restart() case we should just erase what was left
> unprocessed by the GuC from our tail to GuC's head in H2G CTB?
This definietly works but it in perhaps is safer to just squash
everything into NOPs rather than mess with the head / tail. The GuC
didn't like messing with LRC head / tail so maybe it follows we
shouldn't mess with the CT head / tail too. Let me change this as it
pretty straight forward to do this.
>
> as extra bonus we can then look what will be exactly "lost"
I did consider this but that is some extra complexity which really is
not required as GuC submission state machine already tracks everything.
Also once the CT is stopped or we detect a resfix we just throw H2G away
without programming them into the buffer so this really doesn't buy us
anything.
Matt
>
> >
> > - err = guc_ct_ctb_h2g_register(ct);
> > - if (err)
> > - goto err_out;
> > + if (needs_register) {
> > + err = guc_ct_ctb_h2g_register(ct);
> > + if (err)
> > + goto err_out;
> >
> > - err = guc_ct_ctb_g2h_register(ct);
> > - if (err)
> > - goto err_out;
> > + err = guc_ct_ctb_g2h_register(ct);
> > + if (err)
> > + goto err_out;
> >
> > - err = guc_ct_control_toggle(ct, true);
> > - if (err)
> > - goto err_out;
> > + err = guc_ct_control_toggle(ct, true);
> > + if (err)
> > + goto err_out;
> > + }
> >
> > guc_ct_change_state(ct, XE_GUC_CT_STATE_ENABLED);
> >
> > @@ -555,6 +557,32 @@ int xe_guc_ct_enable(struct xe_guc_ct *ct)
> > return err;
> > }
> >
> > +/**
> > + * xe_guc_ct_restart() - Restart GuC CT
> > + * @ct: the &xe_guc_ct
> > + *
> > + * Restart GuC CT to an empty state without issuing a CT register MMIO command.
> > + *
> > + * Return: 0 on success, or a negative errno on failure.
> > + */
> > +int xe_guc_ct_restart(struct xe_guc_ct *ct)
> > +{
> > + return __xe_guc_ct_start(ct, false);
> > +}
> > +
> > +/**
> > + * xe_guc_ct_enable() - Enable GuC CT
> > + * @ct: the &xe_guc_ct
> > + *
> > + * Enable GuC CT to an empty state and issue a CT register MMIO command.
> > + *
> > + * Return: 0 on success, or a negative errno on failure.
> > + */
> > +int xe_guc_ct_enable(struct xe_guc_ct *ct)
> > +{
> > + return __xe_guc_ct_start(ct, true);
> > +}
> > +
> > static void stop_g2h_handler(struct xe_guc_ct *ct)
> > {
> > cancel_work_sync(&ct->g2h_worker);
> > diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
> > index 1e32a59817cc..506616ec1db0 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_ct.h
> > +++ b/drivers/gpu/drm/xe/xe_guc_ct.h
> > @@ -15,6 +15,7 @@ int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct);
> > int xe_guc_ct_init(struct xe_guc_ct *ct);
> > int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
> > int xe_guc_ct_enable(struct xe_guc_ct *ct);
> > +int xe_guc_ct_restart(struct xe_guc_ct *ct);
> > void xe_guc_ct_disable(struct xe_guc_ct *ct);
> > void xe_guc_ct_stop(struct xe_guc_ct *ct);
> > void xe_guc_ct_flush(struct xe_guc_ct *ct);
>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 25/34] drm/xe/vf: Abort VF post migration recovery on failure
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (23 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 24/34] drm/xe/vf: Start CTs before resfix " Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-27 11:54 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 26/34] drm/xe/vf: Replay GuC submission state on pause / unpause Matthew Brost
` (11 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
If VF post-migration recovery fails, the device is wedged. However,
submission queues still need to be enabled for proper cleanup. In such
cases, call into the GuC submission backend to restart all queues that
were previously paused.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 10 ++++++++++
drivers/gpu/drm/xe/xe_guc_submit.c | 20 ++++++++++++++++++++
drivers/gpu/drm/xe/xe_guc_submit.h | 1 +
3 files changed, 31 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 34e57f44fbf4..a987560de2c7 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1224,6 +1224,15 @@ static void vf_post_migration_kickstart(struct xe_gt *gt)
xe_guc_submit_unpause(>->uc.guc);
}
+static void vf_post_migration_abort(struct xe_gt *gt)
+{
+ spin_lock_irq(>->sriov.vf.migration.lock);
+ WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
+ spin_unlock_irq(>->sriov.vf.migration.lock);
+
+ xe_guc_submit_pause_abort(>->uc.guc);
+}
+
static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
{
bool skip_resfix = false;
@@ -1280,6 +1289,7 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
xe_gt_sriov_notice(gt, "migration recovery ended\n");
return;
fail:
+ vf_post_migration_abort(gt);
xe_pm_runtime_put(xe);
xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
xe_device_declare_wedged(xe);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 52b86cab4ec5..8bee65dd9ca6 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2345,6 +2345,26 @@ void xe_guc_submit_unpause(struct xe_guc *guc)
wake_up_all(&guc->ct.wq);
}
+/**
+ * xe_guc_submit_abort - Avort all paused submission task on given GuC.
+ * @guc: the &xe_guc struct instance whose scheduler is to be aborted
+ */
+void xe_guc_submit_pause_abort(struct xe_guc *guc)
+{
+ struct xe_exec_queue *q;
+ unsigned long index;
+
+ mutex_lock(&guc->submission_state.lock);
+ xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) {
+ struct xe_gpu_scheduler *sched = &q->guc->sched;
+
+ xe_sched_submission_start(sched);
+ if (exec_queue_killed_or_banned_or_wedged(q))
+ xe_guc_exec_queue_trigger_cleanup(q);
+ }
+ mutex_unlock(&guc->submission_state.lock);
+}
+
static struct xe_exec_queue *
g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id)
{
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index f535fe3895e5..fe82c317048e 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -22,6 +22,7 @@ void xe_guc_submit_stop(struct xe_guc *guc);
int xe_guc_submit_start(struct xe_guc *guc);
void xe_guc_submit_pause(struct xe_guc *guc);
void xe_guc_submit_unpause(struct xe_guc *guc);
+void xe_guc_submit_pause_abort(struct xe_guc *guc);
void xe_guc_submit_wedge(struct xe_guc *guc);
int xe_guc_read_stopped(struct xe_guc *guc);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 25/34] drm/xe/vf: Abort VF post migration recovery on failure
2025-09-24 1:15 ` [PATCH v2 25/34] drm/xe/vf: Abort VF post migration recovery on failure Matthew Brost
@ 2025-09-27 11:54 ` Lis, Tomasz
2025-09-27 22:38 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-27 11:54 UTC (permalink / raw)
To: Matthew Brost, intel-xe
[-- Attachment #1: Type: text/plain, Size: 3533 bytes --]
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> If VF post-migration recovery fails, the device is wedged. However,
> submission queues still need to be enabled for proper cleanup. In such
> cases, call into the GuC submission backend to restart all queues that
> were previously paused.
>
> Signed-off-by: Matthew Brost<matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 10 ++++++++++
> drivers/gpu/drm/xe/xe_guc_submit.c | 20 ++++++++++++++++++++
> drivers/gpu/drm/xe/xe_guc_submit.h | 1 +
> 3 files changed, 31 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 34e57f44fbf4..a987560de2c7 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -1224,6 +1224,15 @@ static void vf_post_migration_kickstart(struct xe_gt *gt)
> xe_guc_submit_unpause(>->uc.guc);
> }
>
> +static void vf_post_migration_abort(struct xe_gt *gt)
> +{
> + spin_lock_irq(>->sriov.vf.migration.lock);
> + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
> + spin_unlock_irq(>->sriov.vf.migration.lock);
> +
> + xe_guc_submit_pause_abort(>->uc.guc);
> +}
> +
> static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> {
> bool skip_resfix = false;
> @@ -1280,6 +1289,7 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
> xe_gt_sriov_notice(gt, "migration recovery ended\n");
> return;
> fail:
> + vf_post_migration_abort(gt);
> xe_pm_runtime_put(xe);
> xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
> xe_device_declare_wedged(xe);
We could alter the order here to make sure all queues have cleanup
triggered. Now only the ones which were previously killed/banned/wedged
have that.
Not sure if it would bring any benefits though.
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index 52b86cab4ec5..8bee65dd9ca6 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -2345,6 +2345,26 @@ void xe_guc_submit_unpause(struct xe_guc *guc)
> wake_up_all(&guc->ct.wq);
> }
>
> +/**
> + * xe_guc_submit_abort - Avort all paused submission task on given GuC.
typo - Avort
-Tomasz
> + * @guc: the &xe_guc struct instance whose scheduler is to be aborted
> + */
> +void xe_guc_submit_pause_abort(struct xe_guc *guc)
> +{
> + struct xe_exec_queue *q;
> + unsigned long index;
> +
> + mutex_lock(&guc->submission_state.lock);
> + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) {
> + struct xe_gpu_scheduler *sched = &q->guc->sched;
> +
> + xe_sched_submission_start(sched);
> + if (exec_queue_killed_or_banned_or_wedged(q))
> + xe_guc_exec_queue_trigger_cleanup(q);
> + }
> + mutex_unlock(&guc->submission_state.lock);
> +}
> +
> static struct xe_exec_queue *
> g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id)
> {
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
> index f535fe3895e5..fe82c317048e 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.h
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.h
> @@ -22,6 +22,7 @@ void xe_guc_submit_stop(struct xe_guc *guc);
> int xe_guc_submit_start(struct xe_guc *guc);
> void xe_guc_submit_pause(struct xe_guc *guc);
> void xe_guc_submit_unpause(struct xe_guc *guc);
> +void xe_guc_submit_pause_abort(struct xe_guc *guc);
> void xe_guc_submit_wedge(struct xe_guc *guc);
>
> int xe_guc_read_stopped(struct xe_guc *guc);
[-- Attachment #2: Type: text/html, Size: 4286 bytes --]
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 25/34] drm/xe/vf: Abort VF post migration recovery on failure
2025-09-27 11:54 ` Lis, Tomasz
@ 2025-09-27 22:38 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-27 22:38 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Sat, Sep 27, 2025 at 01:54:35PM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > If VF post-migration recovery fails, the device is wedged. However,
> > submission queues still need to be enabled for proper cleanup. In such
> > cases, call into the GuC submission backend to restart all queues that
> > were previously paused.
> >
> > Signed-off-by: Matthew Brost<matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 10 ++++++++++
> > drivers/gpu/drm/xe/xe_guc_submit.c | 20 ++++++++++++++++++++
> > drivers/gpu/drm/xe/xe_guc_submit.h | 1 +
> > 3 files changed, 31 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 34e57f44fbf4..a987560de2c7 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -1224,6 +1224,15 @@ static void vf_post_migration_kickstart(struct xe_gt *gt)
> > xe_guc_submit_unpause(>->uc.guc);
> > }
> > +static void vf_post_migration_abort(struct xe_gt *gt)
> > +{
> > + spin_lock_irq(>->sriov.vf.migration.lock);
> > + WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, false);
> > + spin_unlock_irq(>->sriov.vf.migration.lock);
> > +
> > + xe_guc_submit_pause_abort(>->uc.guc);
> > +}
> > +
> > static int vf_post_migration_notify_resfix_done(struct xe_gt *gt)
> > {
> > bool skip_resfix = false;
> > @@ -1280,6 +1289,7 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
> > xe_gt_sriov_notice(gt, "migration recovery ended\n");
> > return;
> > fail:
> > + vf_post_migration_abort(gt);
> > xe_pm_runtime_put(xe);
> > xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
> > xe_device_declare_wedged(xe);
>
> We could alter the order here to make sure all queues have cleanup
> triggered. Now only the ones which were previously killed/banned/wedged have
> that.
>
> Not sure if it would bring any benefits though.
>
In general doing things like this (doing things behind users back) has
proven to be a bad idea. Also in wedge mode == 2, we intentionally stop
all queues from being torn down until driver load - this enables SV to
come in and run tools to debug hardware issue or we can examine general
hardware state without it changing.
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> > index 52b86cab4ec5..8bee65dd9ca6 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> > @@ -2345,6 +2345,26 @@ void xe_guc_submit_unpause(struct xe_guc *guc)
> > wake_up_all(&guc->ct.wq);
> > }
> > +/**
> > + * xe_guc_submit_abort - Avort all paused submission task on given GuC.
>
> typo - Avort
>
Yea, will fix.
Matt
> -Tomasz
>
> > + * @guc: the &xe_guc struct instance whose scheduler is to be aborted
> > + */
> > +void xe_guc_submit_pause_abort(struct xe_guc *guc)
> > +{
> > + struct xe_exec_queue *q;
> > + unsigned long index;
> > +
> > + mutex_lock(&guc->submission_state.lock);
> > + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) {
> > + struct xe_gpu_scheduler *sched = &q->guc->sched;
> > +
> > + xe_sched_submission_start(sched);
> > + if (exec_queue_killed_or_banned_or_wedged(q))
> > + xe_guc_exec_queue_trigger_cleanup(q);
> > + }
> > + mutex_unlock(&guc->submission_state.lock);
> > +}
> > +
> > static struct xe_exec_queue *
> > g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id)
> > {
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
> > index f535fe3895e5..fe82c317048e 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.h
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.h
> > @@ -22,6 +22,7 @@ void xe_guc_submit_stop(struct xe_guc *guc);
> > int xe_guc_submit_start(struct xe_guc *guc);
> > void xe_guc_submit_pause(struct xe_guc *guc);
> > void xe_guc_submit_unpause(struct xe_guc *guc);
> > +void xe_guc_submit_pause_abort(struct xe_guc *guc);
> > void xe_guc_submit_wedge(struct xe_guc *guc);
> > int xe_guc_read_stopped(struct xe_guc *guc);
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 26/34] drm/xe/vf: Replay GuC submission state on pause / unpause
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (24 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 25/34] drm/xe/vf: Abort VF post migration recovery on failure Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-27 13:33 ` Lis, Tomasz
2025-09-24 1:15 ` [PATCH v2 27/34] drm/xe: Move queue init before LRC creation Matthew Brost
` (10 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Fixup GuC submission pause / unpause functions to properly replay any
possible state lost during VF post migration recovery.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gpu_scheduler.c | 14 ++
drivers/gpu/drm/xe/xe_gpu_scheduler.h | 2 +
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 1 +
drivers/gpu/drm/xe/xe_guc_exec_queue_types.h | 15 ++
drivers/gpu/drm/xe/xe_guc_submit.c | 225 +++++++++++++++++--
drivers/gpu/drm/xe/xe_guc_submit.h | 1 +
drivers/gpu/drm/xe/xe_sched_job_types.h | 4 +
7 files changed, 247 insertions(+), 15 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.c b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
index 455ccaf17314..af300adc7e1a 100644
--- a/drivers/gpu/drm/xe/xe_gpu_scheduler.c
+++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
@@ -135,3 +135,17 @@ void xe_sched_add_msg_locked(struct xe_gpu_scheduler *sched,
list_add_tail(&msg->link, &sched->msgs);
xe_sched_process_msg_queue(sched);
}
+
+/**
+ * xe_sched_add_msg_head() - Xe GPU scheduler add message to head of list
+ * @sched: Xe GPU scheduler
+ * @msg: Message to add
+ */
+void xe_sched_add_msg_head(struct xe_gpu_scheduler *sched,
+ struct xe_sched_msg *msg)
+{
+ lockdep_assert_held(&sched->base.job_list_lock);
+
+ list_add(&msg->link, &sched->msgs);
+ xe_sched_process_msg_queue(sched);
+}
diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.h b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
index e548b2aed95a..010003a6103a 100644
--- a/drivers/gpu/drm/xe/xe_gpu_scheduler.h
+++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
@@ -29,6 +29,8 @@ void xe_sched_add_msg(struct xe_gpu_scheduler *sched,
struct xe_sched_msg *msg);
void xe_sched_add_msg_locked(struct xe_gpu_scheduler *sched,
struct xe_sched_msg *msg);
+void xe_sched_add_msg_head(struct xe_gpu_scheduler *sched,
+ struct xe_sched_msg *msg);
static inline void xe_sched_msg_lock(struct xe_gpu_scheduler *sched)
{
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index a987560de2c7..91e7dbe80ab2 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1217,6 +1217,7 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
static void vf_post_migration_rearm(struct xe_gt *gt)
{
xe_guc_ct_restart(>->uc.guc.ct);
+ xe_guc_submit_unpause_prepare(>->uc.guc);
}
static void vf_post_migration_kickstart(struct xe_gt *gt)
diff --git a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
index c30c0e3ccbbb..a3b034e4b205 100644
--- a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
@@ -51,6 +51,21 @@ struct xe_guc_exec_queue {
wait_queue_head_t suspend_wait;
/** @suspend_pending: a suspend of the exec_queue is pending */
bool suspend_pending;
+ /**
+ * @needs_cleanup: Needs a cleanup message during VF post migration
+ * recovery.
+ */
+ bool needs_cleanup;
+ /**
+ * @needs_suspend: Needs a suspend message during VF post migration
+ * recovery.
+ */
+ bool needs_suspend;
+ /**
+ * @needs_resume: Needs a resume message during VF post migration
+ * recovery.
+ */
+ bool needs_resume;
};
#endif
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 8bee65dd9ca6..b112a4a91a5b 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -425,6 +425,11 @@ static void set_exec_queue_destroyed(struct xe_exec_queue *q)
atomic_or(EXEC_QUEUE_STATE_DESTROYED, &q->guc->state);
}
+static void clear_exec_queue_destroyed(struct xe_exec_queue *q)
+{
+ atomic_and(~EXEC_QUEUE_STATE_DESTROYED, &q->guc->state);
+}
+
static bool exec_queue_banned(struct xe_exec_queue *q)
{
return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_BANNED;
@@ -505,7 +510,12 @@ static void set_exec_queue_extra_ref(struct xe_exec_queue *q)
atomic_or(EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
}
-static bool __maybe_unused exec_queue_pending_resume(struct xe_exec_queue *q)
+static void clear_exec_queue_extra_ref(struct xe_exec_queue *q)
+{
+ atomic_and(~EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
+}
+
+static bool exec_queue_pending_resume(struct xe_exec_queue *q)
{
return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_RESUME;
}
@@ -520,7 +530,7 @@ static void clear_exec_queue_pending_resume(struct xe_exec_queue *q)
atomic_and(~EXEC_QUEUE_STATE_PENDING_RESUME, &q->guc->state);
}
-static bool __maybe_unused exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
+static bool exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
{
return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_TDR_EXIT;
}
@@ -1080,7 +1090,7 @@ static void wq_item_append(struct xe_exec_queue *q)
}
#define RESUME_PENDING ~0x0ull
-static void submit_exec_queue(struct xe_exec_queue *q)
+static void submit_exec_queue(struct xe_exec_queue *q, struct xe_sched_job *job)
{
struct xe_guc *guc = exec_queue_to_guc(q);
struct xe_lrc *lrc = q->lrc[0];
@@ -1092,10 +1102,13 @@ static void submit_exec_queue(struct xe_exec_queue *q)
xe_gt_assert(guc_to_gt(guc), exec_queue_registered(q));
- if (xe_exec_queue_is_parallel(q))
- wq_item_append(q);
- else
- xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
+ if (!job->skip_emit || job->last_replay) {
+ if (xe_exec_queue_is_parallel(q))
+ wq_item_append(q);
+ else
+ xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
+ job->last_replay = false;
+ }
if (exec_queue_suspended(q) && !xe_exec_queue_is_parallel(q))
return;
@@ -1148,8 +1161,10 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
if (!killed_or_banned_or_wedged && !xe_sched_job_is_error(job)) {
if (!exec_queue_registered(q))
register_exec_queue(q, GUC_CONTEXT_NORMAL);
- q->ring_ops->emit_job(job);
- submit_exec_queue(q);
+ if (!job->skip_emit)
+ q->ring_ops->emit_job(job);
+ submit_exec_queue(q, job);
+ job->skip_emit = false;
}
/*
@@ -1860,6 +1875,7 @@ static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg)
#define RESUME 4
#define OPCODE_MASK 0xf
#define MSG_LOCKED BIT(8)
+#define MSG_HEAD BIT(9)
static void guc_exec_queue_process_msg(struct xe_sched_msg *msg)
{
@@ -1984,12 +2000,24 @@ static void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg
msg->private_data = q;
trace_xe_sched_msg_add(msg);
- if (opcode & MSG_LOCKED)
+ if (opcode & MSG_HEAD)
+ xe_sched_add_msg_head(&q->guc->sched, msg);
+ else if (opcode & MSG_LOCKED)
xe_sched_add_msg_locked(&q->guc->sched, msg);
else
xe_sched_add_msg(&q->guc->sched, msg);
}
+static void guc_exec_queue_try_add_msg_head(struct xe_exec_queue *q,
+ struct xe_sched_msg *msg,
+ u32 opcode)
+{
+ if (!list_empty(&msg->link))
+ return;
+
+ guc_exec_queue_add_msg(q, msg, opcode | MSG_LOCKED | MSG_HEAD);
+}
+
static bool guc_exec_queue_try_add_msg(struct xe_exec_queue *q,
struct xe_sched_msg *msg,
u32 opcode)
@@ -2264,6 +2292,93 @@ void xe_guc_submit_stop(struct xe_guc *guc)
}
+/*
+ * This function is quite complex but only real way to ensure no state is lost
+ * during VF resume flows. The function scans the queue state, make adjustments
+ * as needed, and queues jobs / messages which replayed upon unpause.
+ */
+static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
+{
+ struct xe_gpu_scheduler *sched = &q->guc->sched;
+ struct xe_sched_job *job;
+ bool pending_enable, pending_disable, pending_resume;
+ int i;
+
+ lockdep_assert_held(&guc->submission_state.lock);
+
+ /* Stop scheduling + flush any DRM scheduler operations */
+ xe_sched_submission_stop(sched);
+ if (xe_exec_queue_is_lr(q))
+ cancel_work_sync(&q->guc->lr_tdr);
+ else
+ cancel_delayed_work_sync(&sched->base.work_tdr);
+
+ pending_enable = exec_queue_pending_enable(q);
+ pending_resume = exec_queue_pending_resume(q);
+
+ if (pending_enable && pending_resume)
+ q->guc->needs_resume = true;
+
+ if (pending_enable && !pending_resume &&
+ !exec_queue_pending_tdr_exit(q)) {
+ clear_exec_queue_registered(q);
+ if (xe_exec_queue_is_lr(q))
+ xe_exec_queue_put(q);
+ }
+
+ if (pending_enable) {
+ clear_exec_queue_enabled(q);
+ clear_exec_queue_pending_resume(q);
+ clear_exec_queue_pending_tdr_exit(q);
+ clear_exec_queue_pending_enable(q);
+ }
+
+ if (exec_queue_destroyed(q) && exec_queue_registered(q)) {
+ clear_exec_queue_destroyed(q);
+ if (exec_queue_extra_ref(q))
+ xe_exec_queue_put(q);
+ else
+ q->guc->needs_cleanup = true;
+ clear_exec_queue_extra_ref(q);
+ }
+
+ pending_disable = exec_queue_pending_disable(q);
+
+ if (pending_disable && exec_queue_suspended(q)) {
+ clear_exec_queue_suspended(q);
+ q->guc->needs_suspend = true;
+ }
+
+ if (pending_disable) {
+ if (!pending_enable)
+ set_exec_queue_enabled(q);
+ clear_exec_queue_pending_disable(q);
+ clear_exec_queue_check_timeout(q);
+ }
+
+ q->guc->resume_time = 0;
+
+ if (xe_exec_queue_is_parallel(q)) {
+ struct xe_device *xe = guc_to_xe(guc);
+ struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]);
+
+ for (i = 0; i < WQ_SIZE / sizeof(u32); ++i)
+ parallel_write(xe, map, wq[i],
+ FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_NOOP) |
+ FIELD_PREP(WQ_LEN_MASK, 0));
+ }
+
+ job = xe_sched_first_pending_job(sched);
+ if (job) {
+ /*
+ * Adjust software tail so jobs submitted overwrite previous
+ * position in ring buffer with new GGTT addresses.
+ */
+ for (i = 0; i < q->width; ++i)
+ q->lrc[i]->ring.tail = job->ptrs[i].head;
+ }
+}
+
/**
* xe_guc_submit_pause - Stop further runs of submission tasks on given GuC.
* @guc: the &xe_guc struct instance whose scheduler is to be disabled
@@ -2273,8 +2388,12 @@ void xe_guc_submit_pause(struct xe_guc *guc)
struct xe_exec_queue *q;
unsigned long index;
+ xe_gt_assert(guc_to_gt(guc), vf_recovery(guc));
+
+ mutex_lock(&guc->submission_state.lock);
xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
- xe_sched_submission_stop_async(&q->guc->sched);
+ guc_exec_queue_pause(guc, q);
+ mutex_unlock(&guc->submission_state.lock);
}
static void guc_exec_queue_start(struct xe_exec_queue *q)
@@ -2323,11 +2442,87 @@ int xe_guc_submit_start(struct xe_guc *guc)
return 0;
}
-static void guc_exec_queue_unpause(struct xe_exec_queue *q)
+static void guc_exec_queue_unpause_prepare(struct xe_guc *guc,
+ struct xe_exec_queue *q)
{
struct xe_gpu_scheduler *sched = &q->guc->sched;
+ struct drm_sched_job *s_job;
+ struct xe_sched_job *job = NULL;
+
+ list_for_each_entry(s_job, &sched->base.pending_list, list) {
+ job = to_xe_sched_job(s_job);
+
+ q->ring_ops->emit_job(job);
+ job->skip_emit = true;
+ }
+
+ if (job)
+ job->last_replay = true;
+}
+
+/**
+ * xe_guc_submit_unpause_prepare - Prepare unpause submission tasks on given GuC.
+ * @guc: the &xe_guc struct instance whose scheduler is to be prepared for unpause
+ */
+void xe_guc_submit_unpause_prepare(struct xe_guc *guc)
+{
+ struct xe_exec_queue *q;
+ unsigned long index;
+
+ xe_gt_assert(guc_to_gt(guc), vf_recovery(guc));
+
+ mutex_lock(&guc->submission_state.lock);
+ xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
+ guc_exec_queue_unpause_prepare(guc, q);
+ mutex_unlock(&guc->submission_state.lock);
+}
+
+static void guc_exec_queue_unpause(struct xe_guc *guc, struct xe_exec_queue *q)
+{
+ struct xe_gpu_scheduler *sched = &q->guc->sched;
+ struct xe_sched_msg *msg;
+ bool needs_tdr = exec_queue_killed_or_banned_or_wedged(q);
+
+ lockdep_assert_held(&guc->submission_state.lock);
+
+ xe_sched_resubmit_jobs(sched);
+
+ if (q->guc->needs_cleanup) {
+ msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
+
+ guc_exec_queue_add_msg(q, msg, CLEANUP);
+ q->guc->needs_cleanup = false;
+ }
+
+ if (q->guc->needs_suspend) {
+ msg = q->guc->static_msgs + STATIC_MSG_SUSPEND;
+
+ xe_sched_msg_lock(sched);
+ guc_exec_queue_try_add_msg_head(q, msg, SUSPEND);
+ xe_sched_msg_unlock(sched);
+
+ q->guc->needs_suspend = false;
+ }
+
+ /*
+ * The resume must be in the message queue before the suspend as it is
+ * not possible for a resume to be issued if a suspend pending is, but
+ * the inverse is possible.
+ */
+ if (q->guc->needs_resume) {
+ msg = q->guc->static_msgs + STATIC_MSG_RESUME;
+
+ xe_sched_msg_lock(sched);
+ guc_exec_queue_try_add_msg_head(q, msg, RESUME);
+ xe_sched_msg_unlock(sched);
+
+ q->guc->needs_resume = false;
+ }
xe_sched_submission_start(sched);
+ if (needs_tdr)
+ xe_guc_exec_queue_trigger_cleanup(q);
+ xe_sched_submission_resume_tdr(sched);
}
/**
@@ -2339,10 +2534,10 @@ void xe_guc_submit_unpause(struct xe_guc *guc)
struct xe_exec_queue *q;
unsigned long index;
+ mutex_lock(&guc->submission_state.lock);
xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
- guc_exec_queue_unpause(q);
-
- wake_up_all(&guc->ct.wq);
+ guc_exec_queue_unpause(guc, q);
+ mutex_unlock(&guc->submission_state.lock);
}
/**
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index fe82c317048e..b49a2748ec46 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -22,6 +22,7 @@ void xe_guc_submit_stop(struct xe_guc *guc);
int xe_guc_submit_start(struct xe_guc *guc);
void xe_guc_submit_pause(struct xe_guc *guc);
void xe_guc_submit_unpause(struct xe_guc *guc);
+void xe_guc_submit_unpause_prepare(struct xe_guc *guc);
void xe_guc_submit_pause_abort(struct xe_guc *guc);
void xe_guc_submit_wedge(struct xe_guc *guc);
diff --git a/drivers/gpu/drm/xe/xe_sched_job_types.h b/drivers/gpu/drm/xe/xe_sched_job_types.h
index 7ce58765a34a..13e7a12b03ad 100644
--- a/drivers/gpu/drm/xe/xe_sched_job_types.h
+++ b/drivers/gpu/drm/xe/xe_sched_job_types.h
@@ -63,6 +63,10 @@ struct xe_sched_job {
bool ring_ops_flush_tlb;
/** @ggtt: mapped in ggtt. */
bool ggtt;
+ /** @skip_emit: skip emitting the job */
+ bool skip_emit;
+ /** @last_replay: last job being replayed */
+ bool last_replay;
/** @ptrs: per instance pointers. */
struct xe_job_ptrs ptrs[];
};
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 26/34] drm/xe/vf: Replay GuC submission state on pause / unpause
2025-09-24 1:15 ` [PATCH v2 26/34] drm/xe/vf: Replay GuC submission state on pause / unpause Matthew Brost
@ 2025-09-27 13:33 ` Lis, Tomasz
2025-09-27 23:11 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: Lis, Tomasz @ 2025-09-27 13:33 UTC (permalink / raw)
To: Matthew Brost, intel-xe
[-- Attachment #1: Type: text/plain, Size: 15757 bytes --]
On 9/24/2025 3:15 AM, Matthew Brost wrote:
> Fixup GuC submission pause / unpause functions to properly replay any
> possible state lost during VF post migration recovery.
>
> Signed-off-by: Matthew Brost<matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gpu_scheduler.c | 14 ++
> drivers/gpu/drm/xe/xe_gpu_scheduler.h | 2 +
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 1 +
> drivers/gpu/drm/xe/xe_guc_exec_queue_types.h | 15 ++
> drivers/gpu/drm/xe/xe_guc_submit.c | 225 +++++++++++++++++--
> drivers/gpu/drm/xe/xe_guc_submit.h | 1 +
> drivers/gpu/drm/xe/xe_sched_job_types.h | 4 +
> 7 files changed, 247 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.c b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> index 455ccaf17314..af300adc7e1a 100644
> --- a/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> @@ -135,3 +135,17 @@ void xe_sched_add_msg_locked(struct xe_gpu_scheduler *sched,
> list_add_tail(&msg->link, &sched->msgs);
> xe_sched_process_msg_queue(sched);
> }
> +
> +/**
> + * xe_sched_add_msg_head() - Xe GPU scheduler add message to head of list
> + * @sched: Xe GPU scheduler
> + * @msg: Message to add
> + */
> +void xe_sched_add_msg_head(struct xe_gpu_scheduler *sched,
> + struct xe_sched_msg *msg)
> +{
> + lockdep_assert_held(&sched->base.job_list_lock);
> +
> + list_add(&msg->link, &sched->msgs);
> + xe_sched_process_msg_queue(sched);
> +}
> diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.h b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
> index e548b2aed95a..010003a6103a 100644
> --- a/drivers/gpu/drm/xe/xe_gpu_scheduler.h
> +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
> @@ -29,6 +29,8 @@ void xe_sched_add_msg(struct xe_gpu_scheduler *sched,
> struct xe_sched_msg *msg);
> void xe_sched_add_msg_locked(struct xe_gpu_scheduler *sched,
> struct xe_sched_msg *msg);
> +void xe_sched_add_msg_head(struct xe_gpu_scheduler *sched,
> + struct xe_sched_msg *msg);
>
> static inline void xe_sched_msg_lock(struct xe_gpu_scheduler *sched)
> {
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index a987560de2c7..91e7dbe80ab2 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -1217,6 +1217,7 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> static void vf_post_migration_rearm(struct xe_gt *gt)
> {
> xe_guc_ct_restart(>->uc.guc.ct);
> + xe_guc_submit_unpause_prepare(>->uc.guc);
> }
>
> static void vf_post_migration_kickstart(struct xe_gt *gt)
> diff --git a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
> index c30c0e3ccbbb..a3b034e4b205 100644
> --- a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
> +++ b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
> @@ -51,6 +51,21 @@ struct xe_guc_exec_queue {
> wait_queue_head_t suspend_wait;
> /** @suspend_pending: a suspend of the exec_queue is pending */
> bool suspend_pending;
> + /**
> + * @needs_cleanup: Needs a cleanup message during VF post migration
> + * recovery.
> + */
> + bool needs_cleanup;
> + /**
> + * @needs_suspend: Needs a suspend message during VF post migration
> + * recovery.
> + */
> + bool needs_suspend;
> + /**
> + * @needs_resume: Needs a resume message during VF post migration
> + * recovery.
> + */
> + bool needs_resume;
> };
>
> #endif
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> index 8bee65dd9ca6..b112a4a91a5b 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> @@ -425,6 +425,11 @@ static void set_exec_queue_destroyed(struct xe_exec_queue *q)
> atomic_or(EXEC_QUEUE_STATE_DESTROYED, &q->guc->state);
> }
>
> +static void clear_exec_queue_destroyed(struct xe_exec_queue *q)
> +{
> + atomic_and(~EXEC_QUEUE_STATE_DESTROYED, &q->guc->state);
> +}
> +
> static bool exec_queue_banned(struct xe_exec_queue *q)
> {
> return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_BANNED;
> @@ -505,7 +510,12 @@ static void set_exec_queue_extra_ref(struct xe_exec_queue *q)
> atomic_or(EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
> }
>
> -static bool __maybe_unused exec_queue_pending_resume(struct xe_exec_queue *q)
> +static void clear_exec_queue_extra_ref(struct xe_exec_queue *q)
> +{
> + atomic_and(~EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
> +}
> +
> +static bool exec_queue_pending_resume(struct xe_exec_queue *q)
> {
> return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_RESUME;
> }
> @@ -520,7 +530,7 @@ static void clear_exec_queue_pending_resume(struct xe_exec_queue *q)
> atomic_and(~EXEC_QUEUE_STATE_PENDING_RESUME, &q->guc->state);
> }
>
> -static bool __maybe_unused exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
> +static bool exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
> {
> return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_TDR_EXIT;
> }
> @@ -1080,7 +1090,7 @@ static void wq_item_append(struct xe_exec_queue *q)
> }
>
> #define RESUME_PENDING ~0x0ull
> -static void submit_exec_queue(struct xe_exec_queue *q)
> +static void submit_exec_queue(struct xe_exec_queue *q, struct xe_sched_job *job)
> {
> struct xe_guc *guc = exec_queue_to_guc(q);
> struct xe_lrc *lrc = q->lrc[0];
> @@ -1092,10 +1102,13 @@ static void submit_exec_queue(struct xe_exec_queue *q)
>
> xe_gt_assert(guc_to_gt(guc), exec_queue_registered(q));
>
> - if (xe_exec_queue_is_parallel(q))
> - wq_item_append(q);
> - else
> - xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
> + if (!job->skip_emit || job->last_replay) {
> + if (xe_exec_queue_is_parallel(q))
> + wq_item_append(q);
> + else
> + xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
> + job->last_replay = false;
> + }
>
> if (exec_queue_suspended(q) && !xe_exec_queue_is_parallel(q))
> return;
> @@ -1148,8 +1161,10 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
> if (!killed_or_banned_or_wedged && !xe_sched_job_is_error(job)) {
> if (!exec_queue_registered(q))
> register_exec_queue(q, GUC_CONTEXT_NORMAL);
> - q->ring_ops->emit_job(job);
> - submit_exec_queue(q);
> + if (!job->skip_emit)
> + q->ring_ops->emit_job(job);
> + submit_exec_queue(q, job);
> + job->skip_emit = false;
> }
>
> /*
> @@ -1860,6 +1875,7 @@ static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg)
> #define RESUME 4
> #define OPCODE_MASK 0xf
> #define MSG_LOCKED BIT(8)
> +#define MSG_HEAD BIT(9)
>
> static void guc_exec_queue_process_msg(struct xe_sched_msg *msg)
> {
> @@ -1984,12 +2000,24 @@ static void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg
> msg->private_data = q;
>
> trace_xe_sched_msg_add(msg);
> - if (opcode & MSG_LOCKED)
> + if (opcode & MSG_HEAD)
> + xe_sched_add_msg_head(&q->guc->sched, msg);
> + else if (opcode & MSG_LOCKED)
> xe_sched_add_msg_locked(&q->guc->sched, msg);
> else
> xe_sched_add_msg(&q->guc->sched, msg);
> }
>
> +static void guc_exec_queue_try_add_msg_head(struct xe_exec_queue *q,
> + struct xe_sched_msg *msg,
> + u32 opcode)
> +{
> + if (!list_empty(&msg->link))
> + return;
> +
> + guc_exec_queue_add_msg(q, msg, opcode | MSG_LOCKED | MSG_HEAD);
> +}
> +
> static bool guc_exec_queue_try_add_msg(struct xe_exec_queue *q,
> struct xe_sched_msg *msg,
> u32 opcode)
> @@ -2264,6 +2292,93 @@ void xe_guc_submit_stop(struct xe_guc *guc)
>
> }
>
> +/*
> + * This function is quite complex but only real way to ensure no state is lost
> + * during VF resume flows. The function scans the queue state, make adjustments
> + * as needed, and queues jobs / messages which replayed upon unpause.
> + */
> +static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
> +{
> + struct xe_gpu_scheduler *sched = &q->guc->sched;
> + struct xe_sched_job *job;
> + bool pending_enable, pending_disable, pending_resume;
> + int i;
> +
> + lockdep_assert_held(&guc->submission_state.lock);
> +
> + /* Stop scheduling + flush any DRM scheduler operations */
> + xe_sched_submission_stop(sched);
> + if (xe_exec_queue_is_lr(q))
> + cancel_work_sync(&q->guc->lr_tdr);
> + else
> + cancel_delayed_work_sync(&sched->base.work_tdr);
We're doing the same cancelling in `__guc_exec_queue_destroy_async()`,
maybe close it into a function?
> +
> + pending_enable = exec_queue_pending_enable(q);
> + pending_resume = exec_queue_pending_resume(q);
> +
> + if (pending_enable && pending_resume)
> + q->guc->needs_resume = true;
> +
> + if (pending_enable && !pending_resume &&
> + !exec_queue_pending_tdr_exit(q)) {
> + clear_exec_queue_registered(q);
> + if (xe_exec_queue_is_lr(q))
> + xe_exec_queue_put(q);
> + }
> +
> + if (pending_enable) {
> + clear_exec_queue_enabled(q);
> + clear_exec_queue_pending_resume(q);
> + clear_exec_queue_pending_tdr_exit(q);
> + clear_exec_queue_pending_enable(q);
> + }
> +
> + if (exec_queue_destroyed(q) && exec_queue_registered(q)) {
> + clear_exec_queue_destroyed(q);
> + if (exec_queue_extra_ref(q))
> + xe_exec_queue_put(q);
> + else
> + q->guc->needs_cleanup = true;
> + clear_exec_queue_extra_ref(q);
> + }
> +
> + pending_disable = exec_queue_pending_disable(q);
> +
> + if (pending_disable && exec_queue_suspended(q)) {
> + clear_exec_queue_suspended(q);
> + q->guc->needs_suspend = true;
> + }
> +
> + if (pending_disable) {
> + if (!pending_enable)
> + set_exec_queue_enabled(q);
> + clear_exec_queue_pending_disable(q);
> + clear_exec_queue_check_timeout(q);
> + }
maybe we can close the above into a separate function as well?
ie. guc_exec_queue_undo_unfinished_state_change()?
guc_exec_queue_revert_pending_state_change()?
That would make this function easier to read, but also describe what
we're doing.
Then, a counterfunction could be ripped out of guc_exec_queue_unpause().
> +
> + q->guc->resume_time = 0;
> +
> + if (xe_exec_queue_is_parallel(q)) {
> + struct xe_device *xe = guc_to_xe(guc);
> + struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]);
> +
> + for (i = 0; i < WQ_SIZE / sizeof(u32); ++i)
> + parallel_write(xe, map, wq[i],
> + FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_NOOP) |
> + FIELD_PREP(WQ_LEN_MASK, 0));
ok so for parallel wq we're NOP'ing everything and adding the items back
at new positions? Maybe a comment here would help in understanding that.
-Tomasz
> + }
> +
> + job = xe_sched_first_pending_job(sched);
> + if (job) {
> + /*
> + * Adjust software tail so jobs submitted overwrite previous
> + * position in ring buffer with new GGTT addresses.
> + */
> + for (i = 0; i < q->width; ++i)
> + q->lrc[i]->ring.tail = job->ptrs[i].head;
> + }
> +}
> +
> /**
> * xe_guc_submit_pause - Stop further runs of submission tasks on given GuC.
> * @guc: the &xe_guc struct instance whose scheduler is to be disabled
> @@ -2273,8 +2388,12 @@ void xe_guc_submit_pause(struct xe_guc *guc)
> struct xe_exec_queue *q;
> unsigned long index;
>
> + xe_gt_assert(guc_to_gt(guc), vf_recovery(guc));
> +
> + mutex_lock(&guc->submission_state.lock);
> xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> - xe_sched_submission_stop_async(&q->guc->sched);
> + guc_exec_queue_pause(guc, q);
> + mutex_unlock(&guc->submission_state.lock);
> }
>
> static void guc_exec_queue_start(struct xe_exec_queue *q)
> @@ -2323,11 +2442,87 @@ int xe_guc_submit_start(struct xe_guc *guc)
> return 0;
> }
>
> -static void guc_exec_queue_unpause(struct xe_exec_queue *q)
> +static void guc_exec_queue_unpause_prepare(struct xe_guc *guc,
> + struct xe_exec_queue *q)
> {
> struct xe_gpu_scheduler *sched = &q->guc->sched;
> + struct drm_sched_job *s_job;
> + struct xe_sched_job *job = NULL;
> +
> + list_for_each_entry(s_job, &sched->base.pending_list, list) {
> + job = to_xe_sched_job(s_job);
> +
> + q->ring_ops->emit_job(job);
> + job->skip_emit = true;
> + }
> +
> + if (job)
> + job->last_replay = true;
> +}
> +
> +/**
> + * xe_guc_submit_unpause_prepare - Prepare unpause submission tasks on given GuC.
> + * @guc: the &xe_guc struct instance whose scheduler is to be prepared for unpause
> + */
> +void xe_guc_submit_unpause_prepare(struct xe_guc *guc)
> +{
> + struct xe_exec_queue *q;
> + unsigned long index;
> +
> + xe_gt_assert(guc_to_gt(guc), vf_recovery(guc));
> +
> + mutex_lock(&guc->submission_state.lock);
> + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> + guc_exec_queue_unpause_prepare(guc, q);
> + mutex_unlock(&guc->submission_state.lock);
> +}
> +
> +static void guc_exec_queue_unpause(struct xe_guc *guc, struct xe_exec_queue *q)
> +{
> + struct xe_gpu_scheduler *sched = &q->guc->sched;
> + struct xe_sched_msg *msg;
> + bool needs_tdr = exec_queue_killed_or_banned_or_wedged(q);
> +
> + lockdep_assert_held(&guc->submission_state.lock);
> +
> + xe_sched_resubmit_jobs(sched);
> +
> + if (q->guc->needs_cleanup) {
> + msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
> +
> + guc_exec_queue_add_msg(q, msg, CLEANUP);
> + q->guc->needs_cleanup = false;
> + }
> +
> + if (q->guc->needs_suspend) {
> + msg = q->guc->static_msgs + STATIC_MSG_SUSPEND;
> +
> + xe_sched_msg_lock(sched);
> + guc_exec_queue_try_add_msg_head(q, msg, SUSPEND);
> + xe_sched_msg_unlock(sched);
> +
> + q->guc->needs_suspend = false;
> + }
> +
> + /*
> + * The resume must be in the message queue before the suspend as it is
> + * not possible for a resume to be issued if a suspend pending is, but
> + * the inverse is possible.
> + */
> + if (q->guc->needs_resume) {
> + msg = q->guc->static_msgs + STATIC_MSG_RESUME;
> +
> + xe_sched_msg_lock(sched);
> + guc_exec_queue_try_add_msg_head(q, msg, RESUME);
> + xe_sched_msg_unlock(sched);
> +
> + q->guc->needs_resume = false;
> + }
>
> xe_sched_submission_start(sched);
> + if (needs_tdr)
> + xe_guc_exec_queue_trigger_cleanup(q);
> + xe_sched_submission_resume_tdr(sched);
> }
>
> /**
> @@ -2339,10 +2534,10 @@ void xe_guc_submit_unpause(struct xe_guc *guc)
> struct xe_exec_queue *q;
> unsigned long index;
>
> + mutex_lock(&guc->submission_state.lock);
> xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> - guc_exec_queue_unpause(q);
> -
> - wake_up_all(&guc->ct.wq);
> + guc_exec_queue_unpause(guc, q);
> + mutex_unlock(&guc->submission_state.lock);
> }
>
> /**
> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
> index fe82c317048e..b49a2748ec46 100644
> --- a/drivers/gpu/drm/xe/xe_guc_submit.h
> +++ b/drivers/gpu/drm/xe/xe_guc_submit.h
> @@ -22,6 +22,7 @@ void xe_guc_submit_stop(struct xe_guc *guc);
> int xe_guc_submit_start(struct xe_guc *guc);
> void xe_guc_submit_pause(struct xe_guc *guc);
> void xe_guc_submit_unpause(struct xe_guc *guc);
> +void xe_guc_submit_unpause_prepare(struct xe_guc *guc);
> void xe_guc_submit_pause_abort(struct xe_guc *guc);
> void xe_guc_submit_wedge(struct xe_guc *guc);
>
> diff --git a/drivers/gpu/drm/xe/xe_sched_job_types.h b/drivers/gpu/drm/xe/xe_sched_job_types.h
> index 7ce58765a34a..13e7a12b03ad 100644
> --- a/drivers/gpu/drm/xe/xe_sched_job_types.h
> +++ b/drivers/gpu/drm/xe/xe_sched_job_types.h
> @@ -63,6 +63,10 @@ struct xe_sched_job {
> bool ring_ops_flush_tlb;
> /** @ggtt: mapped in ggtt. */
> bool ggtt;
> + /** @skip_emit: skip emitting the job */
> + bool skip_emit;
> + /** @last_replay: last job being replayed */
> + bool last_replay;
> /** @ptrs: per instance pointers. */
> struct xe_job_ptrs ptrs[];
> };
[-- Attachment #2: Type: text/html, Size: 16505 bytes --]
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 26/34] drm/xe/vf: Replay GuC submission state on pause / unpause
2025-09-27 13:33 ` Lis, Tomasz
@ 2025-09-27 23:11 ` Matthew Brost
0 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-27 23:11 UTC (permalink / raw)
To: Lis, Tomasz; +Cc: intel-xe
On Sat, Sep 27, 2025 at 03:33:43PM +0200, Lis, Tomasz wrote:
>
> On 9/24/2025 3:15 AM, Matthew Brost wrote:
> > Fixup GuC submission pause / unpause functions to properly replay any
> > possible state lost during VF post migration recovery.
> >
> > Signed-off-by: Matthew Brost<matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gpu_scheduler.c | 14 ++
> > drivers/gpu/drm/xe/xe_gpu_scheduler.h | 2 +
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 1 +
> > drivers/gpu/drm/xe/xe_guc_exec_queue_types.h | 15 ++
> > drivers/gpu/drm/xe/xe_guc_submit.c | 225 +++++++++++++++++--
> > drivers/gpu/drm/xe/xe_guc_submit.h | 1 +
> > drivers/gpu/drm/xe/xe_sched_job_types.h | 4 +
> > 7 files changed, 247 insertions(+), 15 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.c b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> > index 455ccaf17314..af300adc7e1a 100644
> > --- a/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> > +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> > @@ -135,3 +135,17 @@ void xe_sched_add_msg_locked(struct xe_gpu_scheduler *sched,
> > list_add_tail(&msg->link, &sched->msgs);
> > xe_sched_process_msg_queue(sched);
> > }
> > +
> > +/**
> > + * xe_sched_add_msg_head() - Xe GPU scheduler add message to head of list
> > + * @sched: Xe GPU scheduler
> > + * @msg: Message to add
> > + */
> > +void xe_sched_add_msg_head(struct xe_gpu_scheduler *sched,
> > + struct xe_sched_msg *msg)
> > +{
> > + lockdep_assert_held(&sched->base.job_list_lock);
> > +
> > + list_add(&msg->link, &sched->msgs);
> > + xe_sched_process_msg_queue(sched);
> > +}
> > diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.h b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
> > index e548b2aed95a..010003a6103a 100644
> > --- a/drivers/gpu/drm/xe/xe_gpu_scheduler.h
> > +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
> > @@ -29,6 +29,8 @@ void xe_sched_add_msg(struct xe_gpu_scheduler *sched,
> > struct xe_sched_msg *msg);
> > void xe_sched_add_msg_locked(struct xe_gpu_scheduler *sched,
> > struct xe_sched_msg *msg);
> > +void xe_sched_add_msg_head(struct xe_gpu_scheduler *sched,
> > + struct xe_sched_msg *msg);
> > static inline void xe_sched_msg_lock(struct xe_gpu_scheduler *sched)
> > {
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index a987560de2c7..91e7dbe80ab2 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -1217,6 +1217,7 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> > static void vf_post_migration_rearm(struct xe_gt *gt)
> > {
> > xe_guc_ct_restart(>->uc.guc.ct);
> > + xe_guc_submit_unpause_prepare(>->uc.guc);
> > }
> > static void vf_post_migration_kickstart(struct xe_gt *gt)
> > diff --git a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
> > index c30c0e3ccbbb..a3b034e4b205 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
> > +++ b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
> > @@ -51,6 +51,21 @@ struct xe_guc_exec_queue {
> > wait_queue_head_t suspend_wait;
> > /** @suspend_pending: a suspend of the exec_queue is pending */
> > bool suspend_pending;
> > + /**
> > + * @needs_cleanup: Needs a cleanup message during VF post migration
> > + * recovery.
> > + */
> > + bool needs_cleanup;
> > + /**
> > + * @needs_suspend: Needs a suspend message during VF post migration
> > + * recovery.
> > + */
> > + bool needs_suspend;
> > + /**
> > + * @needs_resume: Needs a resume message during VF post migration
> > + * recovery.
> > + */
> > + bool needs_resume;
> > };
> > #endif
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
> > index 8bee65dd9ca6..b112a4a91a5b 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.c
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
> > @@ -425,6 +425,11 @@ static void set_exec_queue_destroyed(struct xe_exec_queue *q)
> > atomic_or(EXEC_QUEUE_STATE_DESTROYED, &q->guc->state);
> > }
> > +static void clear_exec_queue_destroyed(struct xe_exec_queue *q)
> > +{
> > + atomic_and(~EXEC_QUEUE_STATE_DESTROYED, &q->guc->state);
> > +}
> > +
> > static bool exec_queue_banned(struct xe_exec_queue *q)
> > {
> > return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_BANNED;
> > @@ -505,7 +510,12 @@ static void set_exec_queue_extra_ref(struct xe_exec_queue *q)
> > atomic_or(EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
> > }
> > -static bool __maybe_unused exec_queue_pending_resume(struct xe_exec_queue *q)
> > +static void clear_exec_queue_extra_ref(struct xe_exec_queue *q)
> > +{
> > + atomic_and(~EXEC_QUEUE_STATE_EXTRA_REF, &q->guc->state);
> > +}
> > +
> > +static bool exec_queue_pending_resume(struct xe_exec_queue *q)
> > {
> > return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_RESUME;
> > }
> > @@ -520,7 +530,7 @@ static void clear_exec_queue_pending_resume(struct xe_exec_queue *q)
> > atomic_and(~EXEC_QUEUE_STATE_PENDING_RESUME, &q->guc->state);
> > }
> > -static bool __maybe_unused exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
> > +static bool exec_queue_pending_tdr_exit(struct xe_exec_queue *q)
> > {
> > return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_PENDING_TDR_EXIT;
> > }
> > @@ -1080,7 +1090,7 @@ static void wq_item_append(struct xe_exec_queue *q)
> > }
> > #define RESUME_PENDING ~0x0ull
> > -static void submit_exec_queue(struct xe_exec_queue *q)
> > +static void submit_exec_queue(struct xe_exec_queue *q, struct xe_sched_job *job)
> > {
> > struct xe_guc *guc = exec_queue_to_guc(q);
> > struct xe_lrc *lrc = q->lrc[0];
> > @@ -1092,10 +1102,13 @@ static void submit_exec_queue(struct xe_exec_queue *q)
> > xe_gt_assert(guc_to_gt(guc), exec_queue_registered(q));
> > - if (xe_exec_queue_is_parallel(q))
> > - wq_item_append(q);
> > - else
> > - xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
> > + if (!job->skip_emit || job->last_replay) {
> > + if (xe_exec_queue_is_parallel(q))
> > + wq_item_append(q);
> > + else
> > + xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
> > + job->last_replay = false;
> > + }
> > if (exec_queue_suspended(q) && !xe_exec_queue_is_parallel(q))
> > return;
> > @@ -1148,8 +1161,10 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
> > if (!killed_or_banned_or_wedged && !xe_sched_job_is_error(job)) {
> > if (!exec_queue_registered(q))
> > register_exec_queue(q, GUC_CONTEXT_NORMAL);
> > - q->ring_ops->emit_job(job);
> > - submit_exec_queue(q);
> > + if (!job->skip_emit)
> > + q->ring_ops->emit_job(job);
> > + submit_exec_queue(q, job);
> > + job->skip_emit = false;
> > }
> > /*
> > @@ -1860,6 +1875,7 @@ static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg)
> > #define RESUME 4
> > #define OPCODE_MASK 0xf
> > #define MSG_LOCKED BIT(8)
> > +#define MSG_HEAD BIT(9)
> > static void guc_exec_queue_process_msg(struct xe_sched_msg *msg)
> > {
> > @@ -1984,12 +2000,24 @@ static void guc_exec_queue_add_msg(struct xe_exec_queue *q, struct xe_sched_msg
> > msg->private_data = q;
> > trace_xe_sched_msg_add(msg);
> > - if (opcode & MSG_LOCKED)
> > + if (opcode & MSG_HEAD)
> > + xe_sched_add_msg_head(&q->guc->sched, msg);
> > + else if (opcode & MSG_LOCKED)
> > xe_sched_add_msg_locked(&q->guc->sched, msg);
> > else
> > xe_sched_add_msg(&q->guc->sched, msg);
> > }
> > +static void guc_exec_queue_try_add_msg_head(struct xe_exec_queue *q,
> > + struct xe_sched_msg *msg,
> > + u32 opcode)
> > +{
> > + if (!list_empty(&msg->link))
> > + return;
> > +
> > + guc_exec_queue_add_msg(q, msg, opcode | MSG_LOCKED | MSG_HEAD);
> > +}
> > +
> > static bool guc_exec_queue_try_add_msg(struct xe_exec_queue *q,
> > struct xe_sched_msg *msg,
> > u32 opcode)
> > @@ -2264,6 +2292,93 @@ void xe_guc_submit_stop(struct xe_guc *guc)
> > }
> > +/*
> > + * This function is quite complex but only real way to ensure no state is lost
> > + * during VF resume flows. The function scans the queue state, make adjustments
> > + * as needed, and queues jobs / messages which replayed upon unpause.
> > + */
> > +static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
> > +{
> > + struct xe_gpu_scheduler *sched = &q->guc->sched;
> > + struct xe_sched_job *job;
> > + bool pending_enable, pending_disable, pending_resume;
> > + int i;
> > +
> > + lockdep_assert_held(&guc->submission_state.lock);
> > +
> > + /* Stop scheduling + flush any DRM scheduler operations */
> > + xe_sched_submission_stop(sched);
> > + if (xe_exec_queue_is_lr(q))
> > + cancel_work_sync(&q->guc->lr_tdr);
> > + else
> > + cancel_delayed_work_sync(&sched->base.work_tdr);
>
> We're doing the same cancelling in `__guc_exec_queue_destroy_async()`, maybe
> close it into a function?
>
I could but in follow up I'm going to drop &q->guc->lr_tdr and just use
DRM scheduler TDR so it will just be cancel_delayed_work_sync. I have
the code for this lying around and there a couple of upcoming features
this will get simplified if &q->guc->lr_tdr is droppped.
> > +
> > + pending_enable = exec_queue_pending_enable(q);
> > + pending_resume = exec_queue_pending_resume(q);
> > +
> > + if (pending_enable && pending_resume)
> > + q->guc->needs_resume = true;
> > +
> > + if (pending_enable && !pending_resume &&
> > + !exec_queue_pending_tdr_exit(q)) {
> > + clear_exec_queue_registered(q);
> > + if (xe_exec_queue_is_lr(q))
> > + xe_exec_queue_put(q);
> > + }
> > +
> > + if (pending_enable) {
> > + clear_exec_queue_enabled(q);
> > + clear_exec_queue_pending_resume(q);
> > + clear_exec_queue_pending_tdr_exit(q);
> > + clear_exec_queue_pending_enable(q);
> > + }
> > +
> > + if (exec_queue_destroyed(q) && exec_queue_registered(q)) {
> > + clear_exec_queue_destroyed(q);
> > + if (exec_queue_extra_ref(q))
> > + xe_exec_queue_put(q);
> > + else
> > + q->guc->needs_cleanup = true;
> > + clear_exec_queue_extra_ref(q);
> > + }
> > +
> > + pending_disable = exec_queue_pending_disable(q);
> > +
> > + if (pending_disable && exec_queue_suspended(q)) {
> > + clear_exec_queue_suspended(q);
> > + q->guc->needs_suspend = true;
> > + }
> > +
> > + if (pending_disable) {
> > + if (!pending_enable)
> > + set_exec_queue_enabled(q);
> > + clear_exec_queue_pending_disable(q);
> > + clear_exec_queue_check_timeout(q);
> > + }
>
> maybe we can close the above into a separate function as well?
>
> ie. guc_exec_queue_undo_unfinished_state_change()?
>
Do you mean all the parsing of exec queue state into a single function?
That seems reasonable. I don't really want to abstract each individual
if statement here as that in IMO that is too much abstraction and hard
to figure out what is exactly going on.
> guc_exec_queue_revert_pending_state_change()?
>
> That would make this function easier to read, but also describe what we're
> doing.
>
> Then, a counterfunction could be ripped out of guc_exec_queue_unpause().
>
> > +
> > + q->guc->resume_time = 0;
> > +
> > + if (xe_exec_queue_is_parallel(q)) {
> > + struct xe_device *xe = guc_to_xe(guc);
> > + struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]);
> > +
> > + for (i = 0; i < WQ_SIZE / sizeof(u32); ++i)
> > + parallel_write(xe, map, wq[i],
> > + FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_NOOP) |
> > + FIELD_PREP(WQ_LEN_MASK, 0));
>
> ok so for parallel wq we're NOP'ing everything and adding the items back at
> new positions? Maybe a comment here would help in understanding that.
>
Yes, the GuC didn't like messing with the head / tail. I can add a
comment.
Matt
> -Tomasz
>
> > + }
> > +
> > + job = xe_sched_first_pending_job(sched);
> > + if (job) {
> > + /*
> > + * Adjust software tail so jobs submitted overwrite previous
> > + * position in ring buffer with new GGTT addresses.
> > + */
> > + for (i = 0; i < q->width; ++i)
> > + q->lrc[i]->ring.tail = job->ptrs[i].head;
> > + }
> > +}
> > +
> > /**
> > * xe_guc_submit_pause - Stop further runs of submission tasks on given GuC.
> > * @guc: the &xe_guc struct instance whose scheduler is to be disabled
> > @@ -2273,8 +2388,12 @@ void xe_guc_submit_pause(struct xe_guc *guc)
> > struct xe_exec_queue *q;
> > unsigned long index;
> > + xe_gt_assert(guc_to_gt(guc), vf_recovery(guc));
> > +
> > + mutex_lock(&guc->submission_state.lock);
> > xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> > - xe_sched_submission_stop_async(&q->guc->sched);
> > + guc_exec_queue_pause(guc, q);
> > + mutex_unlock(&guc->submission_state.lock);
> > }
> > static void guc_exec_queue_start(struct xe_exec_queue *q)
> > @@ -2323,11 +2442,87 @@ int xe_guc_submit_start(struct xe_guc *guc)
> > return 0;
> > }
> > -static void guc_exec_queue_unpause(struct xe_exec_queue *q)
> > +static void guc_exec_queue_unpause_prepare(struct xe_guc *guc,
> > + struct xe_exec_queue *q)
> > {
> > struct xe_gpu_scheduler *sched = &q->guc->sched;
> > + struct drm_sched_job *s_job;
> > + struct xe_sched_job *job = NULL;
> > +
> > + list_for_each_entry(s_job, &sched->base.pending_list, list) {
> > + job = to_xe_sched_job(s_job);
> > +
> > + q->ring_ops->emit_job(job);
> > + job->skip_emit = true;
> > + }
> > +
> > + if (job)
> > + job->last_replay = true;
> > +}
> > +
> > +/**
> > + * xe_guc_submit_unpause_prepare - Prepare unpause submission tasks on given GuC.
> > + * @guc: the &xe_guc struct instance whose scheduler is to be prepared for unpause
> > + */
> > +void xe_guc_submit_unpause_prepare(struct xe_guc *guc)
> > +{
> > + struct xe_exec_queue *q;
> > + unsigned long index;
> > +
> > + xe_gt_assert(guc_to_gt(guc), vf_recovery(guc));
> > +
> > + mutex_lock(&guc->submission_state.lock);
> > + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> > + guc_exec_queue_unpause_prepare(guc, q);
> > + mutex_unlock(&guc->submission_state.lock);
> > +}
> > +
> > +static void guc_exec_queue_unpause(struct xe_guc *guc, struct xe_exec_queue *q)
> > +{
> > + struct xe_gpu_scheduler *sched = &q->guc->sched;
> > + struct xe_sched_msg *msg;
> > + bool needs_tdr = exec_queue_killed_or_banned_or_wedged(q);
> > +
> > + lockdep_assert_held(&guc->submission_state.lock);
> > +
> > + xe_sched_resubmit_jobs(sched);
> > +
> > + if (q->guc->needs_cleanup) {
> > + msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
> > +
> > + guc_exec_queue_add_msg(q, msg, CLEANUP);
> > + q->guc->needs_cleanup = false;
> > + }
> > +
> > + if (q->guc->needs_suspend) {
> > + msg = q->guc->static_msgs + STATIC_MSG_SUSPEND;
> > +
> > + xe_sched_msg_lock(sched);
> > + guc_exec_queue_try_add_msg_head(q, msg, SUSPEND);
> > + xe_sched_msg_unlock(sched);
> > +
> > + q->guc->needs_suspend = false;
> > + }
> > +
> > + /*
> > + * The resume must be in the message queue before the suspend as it is
> > + * not possible for a resume to be issued if a suspend pending is, but
> > + * the inverse is possible.
> > + */
> > + if (q->guc->needs_resume) {
> > + msg = q->guc->static_msgs + STATIC_MSG_RESUME;
> > +
> > + xe_sched_msg_lock(sched);
> > + guc_exec_queue_try_add_msg_head(q, msg, RESUME);
> > + xe_sched_msg_unlock(sched);
> > +
> > + q->guc->needs_resume = false;
> > + }
> > xe_sched_submission_start(sched);
> > + if (needs_tdr)
> > + xe_guc_exec_queue_trigger_cleanup(q);
> > + xe_sched_submission_resume_tdr(sched);
> > }
> > /**
> > @@ -2339,10 +2534,10 @@ void xe_guc_submit_unpause(struct xe_guc *guc)
> > struct xe_exec_queue *q;
> > unsigned long index;
> > + mutex_lock(&guc->submission_state.lock);
> > xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
> > - guc_exec_queue_unpause(q);
> > -
> > - wake_up_all(&guc->ct.wq);
> > + guc_exec_queue_unpause(guc, q);
> > + mutex_unlock(&guc->submission_state.lock);
> > }
> > /**
> > diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
> > index fe82c317048e..b49a2748ec46 100644
> > --- a/drivers/gpu/drm/xe/xe_guc_submit.h
> > +++ b/drivers/gpu/drm/xe/xe_guc_submit.h
> > @@ -22,6 +22,7 @@ void xe_guc_submit_stop(struct xe_guc *guc);
> > int xe_guc_submit_start(struct xe_guc *guc);
> > void xe_guc_submit_pause(struct xe_guc *guc);
> > void xe_guc_submit_unpause(struct xe_guc *guc);
> > +void xe_guc_submit_unpause_prepare(struct xe_guc *guc);
> > void xe_guc_submit_pause_abort(struct xe_guc *guc);
> > void xe_guc_submit_wedge(struct xe_guc *guc);
> > diff --git a/drivers/gpu/drm/xe/xe_sched_job_types.h b/drivers/gpu/drm/xe/xe_sched_job_types.h
> > index 7ce58765a34a..13e7a12b03ad 100644
> > --- a/drivers/gpu/drm/xe/xe_sched_job_types.h
> > +++ b/drivers/gpu/drm/xe/xe_sched_job_types.h
> > @@ -63,6 +63,10 @@ struct xe_sched_job {
> > bool ring_ops_flush_tlb;
> > /** @ggtt: mapped in ggtt. */
> > bool ggtt;
> > + /** @skip_emit: skip emitting the job */
> > + bool skip_emit;
> > + /** @last_replay: last job being replayed */
> > + bool last_replay;
> > /** @ptrs: per instance pointers. */
> > struct xe_job_ptrs ptrs[];
> > };
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 27/34] drm/xe: Move queue init before LRC creation
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (25 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 26/34] drm/xe/vf: Replay GuC submission state on pause / unpause Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 28/34] drm/xe/vf: Add debug prints for GuC replaying state during VF recovery Matthew Brost
` (9 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
A queue must be in the submission backend's tracking state before the
LRC is created to avoid a race condition where the LRC's GGTT addresses
are not properly fixed up during VF post-migration recovery.
Move the queue initialization—which adds the queue to the submission
backend's tracking state—before LRC creation.
v2:
- Wait on VF GGTT fixes before creating LRC (testing)
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_exec_queue.c | 43 +++++++++++++++++------
drivers/gpu/drm/xe/xe_execlist.c | 2 +-
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 39 +++++++++++++++++++-
drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 2 ++
drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h | 5 +++
drivers/gpu/drm/xe/xe_guc_submit.c | 2 +-
drivers/gpu/drm/xe/xe_lrc.h | 10 ++++++
7 files changed, 90 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 81f707d2c388..3db8e64d9d13 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -15,6 +15,7 @@
#include "xe_dep_scheduler.h"
#include "xe_device.h"
#include "xe_gt.h"
+#include "xe_gt_sriov_vf.h"
#include "xe_hw_engine_class_sysfs.h"
#include "xe_hw_engine_group.h"
#include "xe_hw_fence.h"
@@ -179,17 +180,32 @@ static int __xe_exec_queue_init(struct xe_exec_queue *q)
flags |= XE_LRC_CREATE_RUNALONE;
}
+ err = q->ops->init(q);
+ if (err)
+ return err;
+
+ /*
+ * This must occur after q->ops->init to avoid race conditions during VF
+ * post-migration recovery, as the fixups for the LRC GGTT addresses
+ * depend on the queue being present in the backend tracking structure.
+ *
+ * In addition to above, we must wait on inflight GGTT changes to
+ * avoid writing out stale values here.
+ */
+ xe_gt_sriov_vf_wait_valid_ggtt(q->gt);
for (i = 0; i < q->width; ++i) {
- q->lrc[i] = xe_lrc_create(q->hwe, q->vm, SZ_16K, q->msix_vec, flags);
- if (IS_ERR(q->lrc[i])) {
- err = PTR_ERR(q->lrc[i]);
+ struct xe_lrc *lrc;
+
+ lrc = xe_lrc_create(q->hwe, q->vm, xe_lrc_ring_size(),
+ q->msix_vec, flags);
+ if (IS_ERR(lrc)) {
+ err = PTR_ERR(lrc);
goto err_lrc;
}
- }
- err = q->ops->init(q);
- if (err)
- goto err_lrc;
+ /* Pairs with READ_ONCE to xe_exec_queue_contexts_hwsp_rebase */
+ WRITE_ONCE(q->lrc[i], lrc);
+ }
return 0;
@@ -1095,9 +1111,16 @@ int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch)
int err = 0;
for (i = 0; i < q->width; ++i) {
- xe_lrc_update_memirq_regs_with_address(q->lrc[i], q->hwe, scratch);
- xe_lrc_update_hwctx_regs_with_address(q->lrc[i]);
- err = xe_lrc_setup_wa_bb_with_scratch(q->lrc[i], q->hwe, scratch);
+ struct xe_lrc *lrc;
+
+ /* Pairs with WRITE_ONCE in __xe_exec_queue_init */
+ lrc = READ_ONCE(q->lrc[i]);
+ if (!lrc)
+ continue;
+
+ xe_lrc_update_memirq_regs_with_address(lrc, q->hwe, scratch);
+ xe_lrc_update_hwctx_regs_with_address(lrc);
+ err = xe_lrc_setup_wa_bb_with_scratch(lrc, q->hwe, scratch);
if (err)
break;
}
diff --git a/drivers/gpu/drm/xe/xe_execlist.c b/drivers/gpu/drm/xe/xe_execlist.c
index f83d421ac9d3..769d05517f93 100644
--- a/drivers/gpu/drm/xe/xe_execlist.c
+++ b/drivers/gpu/drm/xe/xe_execlist.c
@@ -339,7 +339,7 @@ static int execlist_exec_queue_init(struct xe_exec_queue *q)
const struct drm_sched_init_args args = {
.ops = &drm_sched_ops,
.num_rqs = 1,
- .credit_limit = q->lrc[0]->ring.size / MAX_JOB_SIZE_BYTES,
+ .credit_limit = xe_lrc_ring_size() / MAX_JOB_SIZE_BYTES,
.hang_limit = XE_SCHED_HANG_LIMIT,
.timeout = XE_SCHED_JOB_TIMEOUT,
.name = q->hwe->name,
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 91e7dbe80ab2..5e70f3f93d13 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -482,6 +482,11 @@ static int vf_get_ggtt_info(struct xe_gt *gt, bool recovery)
shift, config->ggtt_base);
xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
}
+
+ WRITE_ONCE(gt->sriov.vf.migration.ggtt_need_fixes, false);
+ smp_wmb(); /* Ensure above write visible before wake */
+ wake_up_all(>->sriov.vf.migration.wq);
+
out:
up_write(config->lock);
return err;
@@ -820,7 +825,8 @@ static void xe_gt_sriov_vf_start_migration_recovery(struct xe_gt *gt)
!gt->sriov.vf.migration.recovery_teardown) {
gt->sriov.vf.migration.recovery_queued = true;
WRITE_ONCE(gt->sriov.vf.migration.recovery_inprogress, true);
- smp_wmb(); /* Ensure above write visable before wake */
+ WRITE_ONCE(gt->sriov.vf.migration.ggtt_need_fixes, true);
+ smp_wmb(); /* Ensure above writes visable before wake */
wake_up_all(>->uc.guc.ct.wq);
@@ -1339,6 +1345,7 @@ int xe_gt_sriov_vf_migration_init_early(struct xe_gt *gt)
&tile->primary_gt->sriov.vf.self_config.__lock;
spin_lock_init(>->sriov.vf.migration.lock);
INIT_WORK(>->sriov.vf.migration.worker, migration_worker_func);
+ init_waitqueue_head(>->sriov.vf.migration.wq);
if (!xe_sriov_vf_migration_supported(gt_to_xe(gt)))
xe_gt_sriov_info(gt, "migration not supported by this module version\n");
@@ -1361,3 +1368,33 @@ bool xe_gt_sriov_vf_recovery_inprogress(struct xe_gt *gt)
(xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
READ_ONCE(gt->sriov.vf.migration.recovery_inprogress));
}
+
+static bool vf_valid_ggtt(struct xe_gt *gt)
+{
+ struct xe_memirq *memirq = >_to_tile(gt)->memirq;
+
+ xe_gt_assert(gt, IS_SRIOV_VF(gt_to_xe(gt)));
+
+ if (xe_memirq_vf_recovery_irq_pending(memirq, >->uc.guc) ||
+ READ_ONCE(gt->sriov.vf.migration.ggtt_need_fixes))
+ return false;
+
+ return true;
+}
+
+/**
+ * xe_gt_sriov_vf_wait_valid_ggtt() - VF wait for valid GGTT addresses
+ * @gt: the &xe_gt
+ */
+void xe_gt_sriov_vf_wait_valid_ggtt(struct xe_gt *gt)
+{
+ int ret;
+
+ if (!IS_SRIOV_VF(gt_to_xe(gt)))
+ return;
+
+ ret = wait_event_interruptible_timeout(gt->sriov.vf.migration.wq,
+ vf_valid_ggtt(gt),
+ HZ * 5);
+ XE_WARN_ON(!ret);
+}
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index 535237003915..d5b53ba1d4ca 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -39,4 +39,6 @@ void xe_gt_sriov_vf_print_config(struct xe_gt *gt, struct drm_printer *p);
void xe_gt_sriov_vf_print_runtime(struct xe_gt *gt, struct drm_printer *p);
void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p);
+void xe_gt_sriov_vf_wait_valid_ggtt(struct xe_gt *gt);
+
#endif
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
index beb9978336bb..79bb9d5d62bd 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf_types.h
@@ -8,6 +8,7 @@
#include <linux/rwsem.h>
#include <linux/types.h>
+#include <linux/wait.h>
#include <linux/workqueue.h>
#include "xe_uc_fw_types.h"
@@ -61,6 +62,8 @@ struct xe_gt_sriov_vf_migration {
struct work_struct worker;
/** @lock: Protects recovery_queued, teardown */
spinlock_t lock;
+ /** @wq: wait queue for migration fixes */
+ wait_queue_head_t wq;
/** @lrc_wa_bb: Scratch memory for LRC WA BB in recovery */
void *lrc_wa_bb;
/** @recovery_teardown: VF post migration recovery is being torn down */
@@ -69,6 +72,8 @@ struct xe_gt_sriov_vf_migration {
bool recovery_queued;
/** @recovery_inprogress: VF post migration recovery in progress */
bool recovery_inprogress;
+ /** @ggtt_need_fixes: VF GGTT needs fixes */
+ bool ggtt_need_fixes;
};
/**
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index b112a4a91a5b..f01f2d6f97ed 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1938,7 +1938,7 @@ static int guc_exec_queue_init(struct xe_exec_queue *q)
timeout = (q->vm && xe_vm_in_lr_mode(q->vm)) ? MAX_SCHEDULE_TIMEOUT :
msecs_to_jiffies(q->sched_props.job_timeout_ms);
err = xe_sched_init(&ge->sched, &drm_sched_ops, &xe_sched_ops,
- NULL, q->lrc[0]->ring.size / MAX_JOB_SIZE_BYTES, 64,
+ NULL, xe_lrc_ring_size() / MAX_JOB_SIZE_BYTES, 64,
timeout, guc_to_gt(guc)->ordered_wq, NULL,
q->name, gt_to_xe(q->gt)->drm.dev);
if (err)
diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h
index 188565465779..5fb6c74bdab5 100644
--- a/drivers/gpu/drm/xe/xe_lrc.h
+++ b/drivers/gpu/drm/xe/xe_lrc.h
@@ -74,6 +74,16 @@ static inline void xe_lrc_put(struct xe_lrc *lrc)
kref_put(&lrc->refcount, xe_lrc_destroy);
}
+/**
+ * xe_lrc_ring_size() - Xe LRC ring size
+ *
+ * Return: Size of LRC size
+ */
+static inline size_t xe_lrc_ring_size(void)
+{
+ return SZ_16K;
+}
+
size_t xe_gt_lrc_size(struct xe_gt *gt, enum xe_engine_class class);
u32 xe_lrc_pphwsp_offset(struct xe_lrc *lrc);
u32 xe_lrc_regs_offset(struct xe_lrc *lrc);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 28/34] drm/xe/vf: Add debug prints for GuC replaying state during VF recovery
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (26 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 27/34] drm/xe: Move queue init before LRC creation Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 29/34] drm/xe/vf: Workaround for race condition in GuC firmware during VF pause Matthew Brost
` (8 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
Helpful to manually verify the GuC state machine can correctly replay
the state during a VF post-migration recovery. All replay paths have
been manually verified as triggered and working during testing.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_guc_submit.c | 18 +++++++++++++++++-
1 file changed, 17 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index f01f2d6f97ed..496cd1f2d83f 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2316,14 +2316,19 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
pending_enable = exec_queue_pending_enable(q);
pending_resume = exec_queue_pending_resume(q);
- if (pending_enable && pending_resume)
+ if (pending_enable && pending_resume) {
q->guc->needs_resume = true;
+ xe_gt_dbg(guc_to_gt(guc), "Replay RESUME - guc_id=%d",
+ q->guc->id);
+ }
if (pending_enable && !pending_resume &&
!exec_queue_pending_tdr_exit(q)) {
clear_exec_queue_registered(q);
if (xe_exec_queue_is_lr(q))
xe_exec_queue_put(q);
+ xe_gt_dbg(guc_to_gt(guc), "Replay REGISTER - guc_id=%d",
+ q->guc->id);
}
if (pending_enable) {
@@ -2331,6 +2336,8 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
clear_exec_queue_pending_resume(q);
clear_exec_queue_pending_tdr_exit(q);
clear_exec_queue_pending_enable(q);
+ xe_gt_dbg(guc_to_gt(guc), "Replay ENABLE - guc_id=%d",
+ q->guc->id);
}
if (exec_queue_destroyed(q) && exec_queue_registered(q)) {
@@ -2340,6 +2347,8 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
else
q->guc->needs_cleanup = true;
clear_exec_queue_extra_ref(q);
+ xe_gt_dbg(guc_to_gt(guc), "Replay CLEANUP - guc_id=%d",
+ q->guc->id);
}
pending_disable = exec_queue_pending_disable(q);
@@ -2347,6 +2356,8 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
if (pending_disable && exec_queue_suspended(q)) {
clear_exec_queue_suspended(q);
q->guc->needs_suspend = true;
+ xe_gt_dbg(guc_to_gt(guc), "Replay SUSPEND - guc_id=%d",
+ q->guc->id);
}
if (pending_disable) {
@@ -2354,6 +2365,8 @@ static void guc_exec_queue_pause(struct xe_guc *guc, struct xe_exec_queue *q)
set_exec_queue_enabled(q);
clear_exec_queue_pending_disable(q);
clear_exec_queue_check_timeout(q);
+ xe_gt_dbg(guc_to_gt(guc), "Replay DISABLE - guc_id=%d",
+ q->guc->id);
}
q->guc->resume_time = 0;
@@ -2452,6 +2465,9 @@ static void guc_exec_queue_unpause_prepare(struct xe_guc *guc,
list_for_each_entry(s_job, &sched->base.pending_list, list) {
job = to_xe_sched_job(s_job);
+ xe_gt_dbg(guc_to_gt(guc), "Replay JOB - guc_id=%d, seqno=%d",
+ q->guc->id, xe_sched_job_seqno(job));
+
q->ring_ops->emit_job(job);
job->skip_emit = true;
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 29/34] drm/xe/vf: Workaround for race condition in GuC firmware during VF pause
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (27 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 28/34] drm/xe/vf: Add debug prints for GuC replaying state during VF recovery Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 30/34] drm/xe: Use PPGTT addresses for TLB invalidation to avoid GGTT fixups Matthew Brost
` (7 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
A race condition exists where a paused VF's H2G request can be processed
and subsequently rejected. This rejection results in a FAST_REQ failure
being delivered to the KMD, which then terminates the CT via a dead
worker and triggers a GT reset—an undesirable outcome.
This workaround mitigates the issue by checking if a VF post-migration
recovery is in progress and aborting these adverse actions accordingly.
The GuC firmware will address this bug in an upcoming release. Once that
version is available and VF migration depends on it, this workaround can
be safely removed.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_guc_ct.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index fa55e9c9de3a..794028c9d4b3 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -1391,6 +1391,10 @@ static int parse_g2h_response(struct xe_guc_ct *ct, u32 *msg, u32 len)
fast_req_report(ct, fence);
+ /* FIXME: W/A race in the GuC, will get in firmware soon */
+ if (xe_gt_sriov_vf_recovery_inprogress(gt))
+ return 0;
+
CT_DEAD(ct, NULL, PARSE_G2H_RESPONSE);
return -EPROTO;
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 30/34] drm/xe: Use PPGTT addresses for TLB invalidation to avoid GGTT fixups
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (28 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 29/34] drm/xe/vf: Workaround for race condition in GuC firmware during VF pause Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 31/34] drm/xe/vf: Use primary GT ordered work queue on media GT on PTL VF Matthew Brost
` (6 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
The migrate VM builds the CCS metadata save/restore batch buffer (BB) in
advance and retains it so the GuC can submit it directly when saving a
VM’s state.
When a VM migrates between VFs, the GGTT base can change. Any GGTT-based
addresses embedded in the BB would then have to be parsed and patched.
Use PPGTT addresses in the BB (including for TLB invalidation) so the BB
remains GGTT-agnostic and requires no address fixups during migration.
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_migrate.c | 28 ++++++++++++++++++++--------
1 file changed, 20 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 1d667fa36cf3..ad03afb5145f 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -980,15 +980,27 @@ struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate)
return migrate->q->lrc[0];
}
-static int emit_flush_invalidate(struct xe_exec_queue *q, u32 *dw, int i,
- u32 flags)
+static u64 migrate_vm_ppgtt_addr_tlb_inval(void)
{
- struct xe_lrc *lrc = xe_exec_queue_lrc(q);
+ /*
+ * The migrate VM is self-referential so it can modify its own PTEs (see
+ * pte_update_size() or emit_pte() functions). We reserve NUM_KERNEL_PDE
+ * entries for kernel operations (copies, clears, CCS migrate), and
+ * suballocate the rest to user operations (binds/unbinds). With
+ * NUM_KERNEL_PDE = 15, NUM_KERNEL_PDE - 1 is already used for PTE updates,
+ * so assign NUM_KERNEL_PDE - 2 for TLB invalidation.
+ */
+ return (NUM_KERNEL_PDE - 2) * XE_PAGE_SIZE;
+}
+
+static int emit_flush_invalidate(u32 *dw, int i, u32 flags)
+{
+ u64 addr = migrate_vm_ppgtt_addr_tlb_inval();
+
dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW |
MI_FLUSH_IMM_DW | flags;
- dw[i++] = lower_32_bits(xe_lrc_start_seqno_ggtt_addr(lrc)) |
- MI_FLUSH_DW_USE_GTT;
- dw[i++] = upper_32_bits(xe_lrc_start_seqno_ggtt_addr(lrc));
+ dw[i++] = lower_32_bits(addr);
+ dw[i++] = upper_32_bits(addr);
dw[i++] = MI_NOOP;
dw[i++] = MI_NOOP;
@@ -1101,11 +1113,11 @@ int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
emit_pte(m, bb, ccs_pt, false, false, &ccs_it, ccs_size, src);
- bb->len = emit_flush_invalidate(q, bb->cs, bb->len, flush_flags);
+ bb->len = emit_flush_invalidate(bb->cs, bb->len, flush_flags);
flush_flags = xe_migrate_ccs_copy(m, bb, src_L0_ofs, src_is_pltt,
src_L0_ofs, dst_is_pltt,
src_L0, ccs_ofs, true);
- bb->len = emit_flush_invalidate(q, bb->cs, bb->len, flush_flags);
+ bb->len = emit_flush_invalidate(bb->cs, bb->len, flush_flags);
size -= src_L0;
}
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 31/34] drm/xe/vf: Use primary GT ordered work queue on media GT on PTL VF
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (29 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 30/34] drm/xe: Use PPGTT addresses for TLB invalidation to avoid GGTT fixups Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:15 ` [PATCH v2 32/34] drm/xe/vf: Ensure media GT VF recovery runs after primary GT on PTL Matthew Brost
` (5 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
VF CCS restore is a primary GT operation on which the media GT depends.
Therefore, it doesn't make much sense to run these operations in
parallel. To address this, point the media GT's ordered work queue to
the primary GT's ordered work queue on platforms that require (PTL VFs)
CCS restore as part of VF post-migration recovery.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_device_types.h | 2 ++
drivers/gpu/drm/xe/xe_gt.c | 16 ++++++++++------
drivers/gpu/drm/xe/xe_gt.h | 2 +-
drivers/gpu/drm/xe/xe_pci.c | 6 +++++-
drivers/gpu/drm/xe/xe_pci_types.h | 1 +
drivers/gpu/drm/xe/xe_tile.c | 2 +-
6 files changed, 20 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index a6c361db11d9..af92c8cd43b6 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -318,6 +318,8 @@ struct xe_device {
u8 skip_mtcfg:1;
/** @info.skip_pcode: skip access to PCODE uC */
u8 skip_pcode:1;
+ /** @info.needs_shared_vf_gt_wq: needs shared GT WQ on VF */
+ u8 needs_shared_vf_gt_wq:1;
} info;
/** @wa_active: keep track of active workarounds */
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 1a4796cd885b..5ab283fb24ea 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -72,7 +72,7 @@ static void gt_fini(struct drm_device *drm, void *arg)
destroy_workqueue(gt->ordered_wq);
}
-struct xe_gt *xe_gt_alloc(struct xe_tile *tile)
+struct xe_gt *xe_gt_alloc(struct xe_tile *tile, bool use_primary_wq)
{
struct xe_gt *gt;
int err;
@@ -82,12 +82,16 @@ struct xe_gt *xe_gt_alloc(struct xe_tile *tile)
return ERR_PTR(-ENOMEM);
gt->tile = tile;
- gt->ordered_wq = alloc_ordered_workqueue("gt-ordered-wq",
- WQ_MEM_RECLAIM);
+ if (use_primary_wq) {
+ gt->ordered_wq = tile->primary_gt->ordered_wq;
+ } else {
+ gt->ordered_wq = alloc_ordered_workqueue("gt-ordered-wq",
+ WQ_MEM_RECLAIM);
- err = drmm_add_action_or_reset(>_to_xe(gt)->drm, gt_fini, gt);
- if (err)
- return ERR_PTR(err);
+ err = drmm_add_action_or_reset(>_to_xe(gt)->drm, gt_fini, gt);
+ if (err)
+ return ERR_PTR(err);
+ }
return gt;
}
diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h
index 41880979f4de..31090ee47591 100644
--- a/drivers/gpu/drm/xe/xe_gt.h
+++ b/drivers/gpu/drm/xe/xe_gt.h
@@ -27,7 +27,7 @@ static inline bool xe_fault_inject_gt_reset(void)
return IS_ENABLED(CONFIG_DEBUG_FS) && should_fail(>_reset_failure, 1);
}
-struct xe_gt *xe_gt_alloc(struct xe_tile *tile);
+struct xe_gt *xe_gt_alloc(struct xe_tile *tile, bool use_primary_wq);
int xe_gt_init_early(struct xe_gt *gt);
int xe_gt_init(struct xe_gt *gt);
void xe_gt_mmio_init(struct xe_gt *gt);
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 3f42b91efa28..25a1d96a68e7 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -347,6 +347,7 @@ static const struct xe_device_desc ptl_desc = {
.has_sriov = true,
.max_gt_per_tile = 2,
.needs_scratch = true,
+ .needs_shared_vf_gt_wq = true,
};
#undef PLATFORM
@@ -598,6 +599,7 @@ static int xe_info_init_early(struct xe_device *xe,
xe->info.skip_mtcfg = desc->skip_mtcfg;
xe->info.skip_pcode = desc->skip_pcode;
xe->info.needs_scratch = desc->needs_scratch;
+ xe->info.needs_shared_vf_gt_wq = desc->needs_shared_vf_gt_wq;
xe->info.probe_display = IS_ENABLED(CONFIG_DRM_XE_DISPLAY) &&
xe_modparam.probe_display &&
@@ -766,7 +768,9 @@ static int xe_info_init(struct xe_device *xe,
* Allocate and setup media GT for platforms with standalone
* media.
*/
- tile->media_gt = xe_gt_alloc(tile);
+ tile->media_gt = xe_gt_alloc(tile,
+ xe->info.needs_shared_vf_gt_wq &&
+ IS_SRIOV_VF(xe));
if (IS_ERR(tile->media_gt))
return PTR_ERR(tile->media_gt);
diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h
index 9b9766a3baa3..b11bf6abda5b 100644
--- a/drivers/gpu/drm/xe/xe_pci_types.h
+++ b/drivers/gpu/drm/xe/xe_pci_types.h
@@ -48,6 +48,7 @@ struct xe_device_desc {
u8 skip_guc_pc:1;
u8 skip_mtcfg:1;
u8 skip_pcode:1;
+ u8 needs_shared_vf_gt_wq:1;
};
struct xe_graphics_desc {
diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c
index d49ba3401963..a982732a8056 100644
--- a/drivers/gpu/drm/xe/xe_tile.c
+++ b/drivers/gpu/drm/xe/xe_tile.c
@@ -149,7 +149,7 @@ int xe_tile_init_early(struct xe_tile *tile, struct xe_device *xe, u8 id)
if (err)
return err;
- tile->primary_gt = xe_gt_alloc(tile);
+ tile->primary_gt = xe_gt_alloc(tile, false);
if (IS_ERR(tile->primary_gt))
return PTR_ERR(tile->primary_gt);
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 32/34] drm/xe/vf: Ensure media GT VF recovery runs after primary GT on PTL
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (30 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 31/34] drm/xe/vf: Use primary GT ordered work queue on media GT on PTL VF Matthew Brost
@ 2025-09-24 1:15 ` Matthew Brost
2025-09-24 1:16 ` [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses Matthew Brost
` (4 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:15 UTC (permalink / raw)
To: intel-xe
It is possible that the media GT's VF post-migration recovery work item
gets scheduled before the primary GT's work item. Since the media GT
depends on the primary GT's work item to complete CCS restore, if the
media GT's work item is scheduled first, detect this condition and
re-queue the media GT's work item for a later time.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 29 +++++++++++++++++++++++++++--
1 file changed, 27 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 5e70f3f93d13..46d1d1fa91b2 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -1187,8 +1187,22 @@ void xe_gt_sriov_vf_print_version(struct xe_gt *gt, struct drm_printer *p)
pf_version->major, pf_version->minor);
}
-static void vf_post_migration_shutdown(struct xe_gt *gt)
+static bool vf_post_migration_shutdown(struct xe_gt *gt)
{
+ struct xe_device *xe = gt_to_xe(gt);
+
+ /*
+ * On platforms where CCS must be restored by the primary GT, the media
+ * GT's VF post-migration recovery must run afterward. Detect this case
+ * and re-queue the media GT's restore work item if necessary.
+ */
+ if (xe->info.needs_shared_vf_gt_wq && xe_gt_is_media_type(gt)) {
+ struct xe_gt *primary_gt = gt_to_tile(gt)->primary_gt;
+
+ if (xe_gt_sriov_vf_recovery_inprogress(primary_gt))
+ return true;
+ }
+
spin_lock_irq(>->sriov.vf.migration.lock);
gt->sriov.vf.migration.recovery_queued = false;
spin_unlock_irq(>->sriov.vf.migration.lock);
@@ -1196,6 +1210,8 @@ static void vf_post_migration_shutdown(struct xe_gt *gt)
xe_guc_ct_flush(>->uc.guc.ct);
xe_tlb_inval_reset(>->tlb_inval);
xe_guc_submit_pause(>->uc.guc);
+
+ return false;
}
static size_t post_migration_scratch_size(struct xe_device *xe)
@@ -1268,11 +1284,14 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
{
struct xe_device *xe = gt_to_xe(gt);
int err;
+ bool retry;
xe_gt_sriov_dbg(gt, "migration recovery in progress\n");
xe_pm_runtime_get(xe);
- vf_post_migration_shutdown(gt);
+ retry = vf_post_migration_shutdown(gt);
+ if (retry)
+ goto queue;
if (!xe_sriov_vf_migration_supported(xe)) {
xe_gt_sriov_err(gt, "migration is not supported\n");
@@ -1300,6 +1319,12 @@ static void vf_post_migration_recovery(struct xe_gt *gt)
xe_pm_runtime_put(xe);
xe_gt_sriov_err(gt, "migration recovery failed (%pe)\n", ERR_PTR(err));
xe_device_declare_wedged(xe);
+ return;
+
+queue:
+ xe_gt_sriov_info(gt, "Re-queuing GT recovery\n");
+ queue_work(gt->ordered_wq, >->sriov.vf.migration.worker);
+ xe_pm_runtime_put(xe);
}
static void migration_worker_func(struct work_struct *w)
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (31 preceding siblings ...)
2025-09-24 1:15 ` [PATCH v2 32/34] drm/xe/vf: Ensure media GT VF recovery runs after primary GT on PTL Matthew Brost
@ 2025-09-24 1:16 ` Matthew Brost
2025-09-24 4:04 ` K V P, Satyanarayana
2025-09-24 1:16 ` [PATCH v2 34/34] drm/xe/guc: Increase wait timeout to 2sec after BUSY reply from GuC Matthew Brost
` (3 subsequent siblings)
36 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:16 UTC (permalink / raw)
To: intel-xe
Rebase the CCS save/restore BB's GGTT addresses during VF post-migration
recovery by setting the software ring tail to zero, the LRC ring head to
zero, and rewriting the jump-to-BB instructions.
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
---
drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 4 ++++
drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 24 ++++++++++++++++++++++++
drivers/gpu/drm/xe/xe_sriov_vf_ccs.h | 1 +
3 files changed, 29 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index 46d1d1fa91b2..d8d96dd76c73 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -34,6 +34,7 @@
#include "xe_pm.h"
#include "xe_sriov.h"
#include "xe_sriov_vf.h"
+#include "xe_sriov_vf_ccs.h"
#include "xe_tile_sriov_vf.h"
#include "xe_tlb_inval.h"
#include "xe_uc_fw.h"
@@ -1228,6 +1229,9 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
if (err)
return err;
+ if (xe_gt_is_main_type(gt))
+ xe_sriov_vf_ccs_rebase(gt_to_xe(gt));
+
xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
if (err)
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 8dec616c37c9..bffad45e655f 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -175,6 +175,15 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
u32 dw[10], i = 0;
+ /*
+ * XXX: Save/restore fixes — for some reason, the GuC only accepts the
+ * save/restore context if the LRC head pointer is zero. This is evident
+ * from repeated VF migrations failing when the LRC head pointer is
+ * non-zero.
+ */
+ lrc->ring.tail = 0;
+ xe_lrc_set_ring_head(lrc, 0);
+
dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;
dw[i++] = MI_BATCH_BUFFER_START | XE_INSTR_NUM_DW(3);
dw[i++] = lower_32_bits(addr);
@@ -186,6 +195,21 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
}
+void xe_sriov_vf_ccs_rebase(struct xe_device *xe)
+{
+ enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+
+ if (!IS_VF_CCS_READY(xe))
+ return;
+
+ for_each_ccs_rw_ctx(ctx_id) {
+ struct xe_sriov_vf_ccs_ctx *ctx =
+ &xe->sriov.vf.ccs.contexts[ctx_id];
+
+ ccs_rw_update_ring(ctx);
+ }
+}
+
static int register_save_restore_context(struct xe_sriov_vf_ccs_ctx *ctx)
{
int ctx_type;
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
index 0745c0ff0228..f8ca6efce9ee 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
@@ -18,6 +18,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe);
int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
+void xe_sriov_vf_ccs_rebase(struct xe_device *xe);
void xe_sriov_vf_ccs_print(struct xe_device *xe, struct drm_printer *p);
static inline bool xe_sriov_vf_ccs_ready(struct xe_device *xe)
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* Re: [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses
2025-09-24 1:16 ` [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses Matthew Brost
@ 2025-09-24 4:04 ` K V P, Satyanarayana
2025-09-24 6:32 ` Matthew Brost
0 siblings, 1 reply; 90+ messages in thread
From: K V P, Satyanarayana @ 2025-09-24 4:04 UTC (permalink / raw)
To: intel-xe
On 24-09-2025 06:46, Matthew Brost wrote:
> Rebase the CCS save/restore BB's GGTT addresses during VF post-migration
> recovery by setting the software ring tail to zero, the LRC ring head to
> zero, and rewriting the jump-to-BB instructions.
>
> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 4 ++++
> drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 24 ++++++++++++++++++++++++
> drivers/gpu/drm/xe/xe_sriov_vf_ccs.h | 1 +
> 3 files changed, 29 insertions(+)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> index 46d1d1fa91b2..d8d96dd76c73 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -34,6 +34,7 @@
> #include "xe_pm.h"
> #include "xe_sriov.h"
> #include "xe_sriov_vf.h"
> +#include "xe_sriov_vf_ccs.h"
> #include "xe_tile_sriov_vf.h"
> #include "xe_tlb_inval.h"
> #include "xe_uc_fw.h"
> @@ -1228,6 +1229,9 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> if (err)
> return err;
>
> + if (xe_gt_is_main_type(gt))
> + xe_sriov_vf_ccs_rebase(gt_to_xe(gt));
> +
> xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> if (err)
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
> index 8dec616c37c9..bffad45e655f 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
> @@ -175,6 +175,15 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
> struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
> u32 dw[10], i = 0;
>
> + /*
> + * XXX: Save/restore fixes — for some reason, the GuC only accepts the
> + * save/restore context if the LRC head pointer is zero. This is evident
> + * from repeated VF migrations failing when the LRC head pointer is
> + * non-zero.
> + */
> + lrc->ring.tail = 0;
> + xe_lrc_set_ring_head(lrc, 0);
> +
On lrc_init(), the ring may be pre-populated with some GPU instructions.
Should not we keep them as it is? Making tail=0 will overwrite the
pre-populated GPU instructions (May be some WAs??).
-Satya.> dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;
> dw[i++] = MI_BATCH_BUFFER_START | XE_INSTR_NUM_DW(3);
> dw[i++] = lower_32_bits(addr);
> @@ -186,6 +195,21 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
> xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
> }
>
> +void xe_sriov_vf_ccs_rebase(struct xe_device *xe)
> +{
> + enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
> +
> + if (!IS_VF_CCS_READY(xe))
> + return;
> +
> + for_each_ccs_rw_ctx(ctx_id) {
> + struct xe_sriov_vf_ccs_ctx *ctx =
> + &xe->sriov.vf.ccs.contexts[ctx_id];
> +
> + ccs_rw_update_ring(ctx);
> + }
> +}
> +
> static int register_save_restore_context(struct xe_sriov_vf_ccs_ctx *ctx)
> {
> int ctx_type;
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
> index 0745c0ff0228..f8ca6efce9ee 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
> @@ -18,6 +18,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe);
> int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
> int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
> int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
> +void xe_sriov_vf_ccs_rebase(struct xe_device *xe);
> void xe_sriov_vf_ccs_print(struct xe_device *xe, struct drm_printer *p);
>
> static inline bool xe_sriov_vf_ccs_ready(struct xe_device *xe)
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses
2025-09-24 4:04 ` K V P, Satyanarayana
@ 2025-09-24 6:32 ` Matthew Brost
2025-09-24 6:36 ` K V P, Satyanarayana
0 siblings, 1 reply; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 6:32 UTC (permalink / raw)
To: K V P, Satyanarayana; +Cc: intel-xe
On Wed, Sep 24, 2025 at 09:34:25AM +0530, K V P, Satyanarayana wrote:
>
>
> On 24-09-2025 06:46, Matthew Brost wrote:
> > Rebase the CCS save/restore BB's GGTT addresses during VF post-migration
> > recovery by setting the software ring tail to zero, the LRC ring head to
> > zero, and rewriting the jump-to-BB instructions.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 4 ++++
> > drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 24 ++++++++++++++++++++++++
> > drivers/gpu/drm/xe/xe_sriov_vf_ccs.h | 1 +
> > 3 files changed, 29 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > index 46d1d1fa91b2..d8d96dd76c73 100644
> > --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> > @@ -34,6 +34,7 @@
> > #include "xe_pm.h"
> > #include "xe_sriov.h"
> > #include "xe_sriov_vf.h"
> > +#include "xe_sriov_vf_ccs.h"
> > #include "xe_tile_sriov_vf.h"
> > #include "xe_tlb_inval.h"
> > #include "xe_uc_fw.h"
> > @@ -1228,6 +1229,9 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
> > if (err)
> > return err;
> > + if (xe_gt_is_main_type(gt))
> > + xe_sriov_vf_ccs_rebase(gt_to_xe(gt));
> > +
> > xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
> > err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
> > if (err)
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
> > index 8dec616c37c9..bffad45e655f 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
> > @@ -175,6 +175,15 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
> > struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
> > u32 dw[10], i = 0;
> > + /*
> > + * XXX: Save/restore fixes — for some reason, the GuC only accepts the
> > + * save/restore context if the LRC head pointer is zero. This is evident
> > + * from repeated VF migrations failing when the LRC head pointer is
> > + * non-zero.
> > + */
> > + lrc->ring.tail = 0;
> > + xe_lrc_set_ring_head(lrc, 0);
> > +
> On lrc_init(), the ring may be pre-populated with some GPU instructions.
> Should not we keep them as it is? Making tail=0 will overwrite the
> pre-populated GPU instructions (May be some WAs??).
> -Satya.> dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;
I’ve thought about that. W/A should be applied through a W/A BB
mechanism, not via the ring. If I recall correctly, the original code
for programming instructions in the ring came from Faith’s original Xe
PoC code from over four years ago, which we’ve never really revisited.
In my opinion, that code should be removed.
I could include the removal in this series, but I’d prefer not to expand
its scope. We can address it in a follow-up. I also don’t want this
series to grow in complexity by retaining what is almost certainly
useless code.
Matt
> > dw[i++] = MI_BATCH_BUFFER_START | XE_INSTR_NUM_DW(3);
> > dw[i++] = lower_32_bits(addr);
> > @@ -186,6 +195,21 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
> > xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
> > }
> > +void xe_sriov_vf_ccs_rebase(struct xe_device *xe)
> > +{
> > + enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
> > +
> > + if (!IS_VF_CCS_READY(xe))
> > + return;
> > +
> > + for_each_ccs_rw_ctx(ctx_id) {
> > + struct xe_sriov_vf_ccs_ctx *ctx =
> > + &xe->sriov.vf.ccs.contexts[ctx_id];
> > +
> > + ccs_rw_update_ring(ctx);
> > + }
> > +}
> > +
> > static int register_save_restore_context(struct xe_sriov_vf_ccs_ctx *ctx)
> > {
> > int ctx_type;
> > diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
> > index 0745c0ff0228..f8ca6efce9ee 100644
> > --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
> > +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
> > @@ -18,6 +18,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe);
> > int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
> > int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
> > int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
> > +void xe_sriov_vf_ccs_rebase(struct xe_device *xe);
> > void xe_sriov_vf_ccs_print(struct xe_device *xe, struct drm_printer *p);
> > static inline bool xe_sriov_vf_ccs_ready(struct xe_device *xe)
>
^ permalink raw reply [flat|nested] 90+ messages in thread* Re: [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses
2025-09-24 6:32 ` Matthew Brost
@ 2025-09-24 6:36 ` K V P, Satyanarayana
0 siblings, 0 replies; 90+ messages in thread
From: K V P, Satyanarayana @ 2025-09-24 6:36 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
On 24-09-2025 12:02, Matthew Brost wrote:
> On Wed, Sep 24, 2025 at 09:34:25AM +0530, K V P, Satyanarayana wrote:
>>
>>
>> On 24-09-2025 06:46, Matthew Brost wrote:
>>> Rebase the CCS save/restore BB's GGTT addresses during VF post-migration
>>> recovery by setting the software ring tail to zero, the LRC ring head to
>>> zero, and rewriting the jump-to-BB instructions.
>>>
>>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>>> ---
>>> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 4 ++++
>>> drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 24 ++++++++++++++++++++++++
>>> drivers/gpu/drm/xe/xe_sriov_vf_ccs.h | 1 +
>>> 3 files changed, 29 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> index 46d1d1fa91b2..d8d96dd76c73 100644
>>> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
>>> @@ -34,6 +34,7 @@
>>> #include "xe_pm.h"
>>> #include "xe_sriov.h"
>>> #include "xe_sriov_vf.h"
>>> +#include "xe_sriov_vf_ccs.h"
>>> #include "xe_tile_sriov_vf.h"
>>> #include "xe_tlb_inval.h"
>>> #include "xe_uc_fw.h"
>>> @@ -1228,6 +1229,9 @@ static int vf_post_migration_fixups(struct xe_gt *gt)
>>> if (err)
>>> return err;
>>> + if (xe_gt_is_main_type(gt))
>>> + xe_sriov_vf_ccs_rebase(gt_to_xe(gt));
>>> +
>>> xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
>>> err = xe_guc_contexts_hwsp_rebase(>->uc.guc, buf);
>>> if (err)
>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
>>> index 8dec616c37c9..bffad45e655f 100644
>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
>>> @@ -175,6 +175,15 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
>>> struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
>>> u32 dw[10], i = 0;
>>> + /*
>>> + * XXX: Save/restore fixes — for some reason, the GuC only accepts the
>>> + * save/restore context if the LRC head pointer is zero. This is evident
>>> + * from repeated VF migrations failing when the LRC head pointer is
>>> + * non-zero.
>>> + */
>>> + lrc->ring.tail = 0;
>>> + xe_lrc_set_ring_head(lrc, 0);
>>> +
>> On lrc_init(), the ring may be pre-populated with some GPU instructions.
>> Should not we keep them as it is? Making tail=0 will overwrite the
>> pre-populated GPU instructions (May be some WAs??).
>> -Satya.> dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;
>
> I’ve thought about that. W/A should be applied through a W/A BB
> mechanism, not via the ring. If I recall correctly, the original code
> for programming instructions in the ring came from Faith’s original Xe
> PoC code from over four years ago, which we’ve never really revisited.
>
> In my opinion, that code should be removed.
>
> I could include the removal in this series, but I’d prefer not to expand
> its scope. We can address it in a follow-up. I also don’t want this
> series to grow in complexity by retaining what is almost certainly
> useless code.
>
> Matt
LGTM.
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>>
>>> dw[i++] = MI_BATCH_BUFFER_START | XE_INSTR_NUM_DW(3);
>>> dw[i++] = lower_32_bits(addr);
>>> @@ -186,6 +195,21 @@ static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
>>> xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
>>> }
>>> +void xe_sriov_vf_ccs_rebase(struct xe_device *xe)
>>> +{
>>> + enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
>>> +
>>> + if (!IS_VF_CCS_READY(xe))
>>> + return;
>>> +
>>> + for_each_ccs_rw_ctx(ctx_id) {
>>> + struct xe_sriov_vf_ccs_ctx *ctx =
>>> + &xe->sriov.vf.ccs.contexts[ctx_id];
>>> +
>>> + ccs_rw_update_ring(ctx);
>>> + }
>>> +}
>>> +
>>> static int register_save_restore_context(struct xe_sriov_vf_ccs_ctx *ctx)
>>> {
>>> int ctx_type;
>>> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
>>> index 0745c0ff0228..f8ca6efce9ee 100644
>>> --- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
>>> +++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
>>> @@ -18,6 +18,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe);
>>> int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
>>> int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
>>> int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
>>> +void xe_sriov_vf_ccs_rebase(struct xe_device *xe);
>>> void xe_sriov_vf_ccs_print(struct xe_device *xe, struct drm_printer *p);
>>> static inline bool xe_sriov_vf_ccs_ready(struct xe_device *xe)
>>
^ permalink raw reply [flat|nested] 90+ messages in thread
* [PATCH v2 34/34] drm/xe/guc: Increase wait timeout to 2sec after BUSY reply from GuC
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (32 preceding siblings ...)
2025-09-24 1:16 ` [PATCH v2 33/34] drm/xe/vf: Rebase CCS save/restore BB GGTT addresses Matthew Brost
@ 2025-09-24 1:16 ` Matthew Brost
2025-09-24 1:29 ` ✓ CI.KUnit: success for VF migration redesign (rev2) Patchwork
` (2 subsequent siblings)
36 siblings, 0 replies; 90+ messages in thread
From: Matthew Brost @ 2025-09-24 1:16 UTC (permalink / raw)
To: intel-xe
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Some VF2GUC actions may take longer to process. Increase default timeout
after received BUSY indication to 2sec to cover all worst case scenarios.
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
---
drivers/gpu/drm/xe/xe_guc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index ac60da51da2c..d029f349dafa 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -1472,7 +1472,7 @@ int xe_guc_mmio_send_recv(struct xe_guc *guc, const u32 *request,
BUILD_BUG_ON((GUC_HXG_TYPE_RESPONSE_SUCCESS ^ GUC_HXG_TYPE_RESPONSE_FAILURE) != 1);
ret = xe_mmio_wait32(mmio, reply_reg, resp_mask, resp_mask,
- 1000000, &header, false);
+ 2000000, &header, false);
if (unlikely(FIELD_GET(GUC_HXG_MSG_0_ORIGIN, header) !=
GUC_HXG_ORIGIN_GUC))
--
2.34.1
^ permalink raw reply related [flat|nested] 90+ messages in thread* ✓ CI.KUnit: success for VF migration redesign (rev2)
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (33 preceding siblings ...)
2025-09-24 1:16 ` [PATCH v2 34/34] drm/xe/guc: Increase wait timeout to 2sec after BUSY reply from GuC Matthew Brost
@ 2025-09-24 1:29 ` Patchwork
2025-09-24 2:14 ` ✗ Xe.CI.BAT: failure " Patchwork
2025-09-24 7:37 ` ✗ Xe.CI.Full: " Patchwork
36 siblings, 0 replies; 90+ messages in thread
From: Patchwork @ 2025-09-24 1:29 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
== Series Details ==
Series: VF migration redesign (rev2)
URL : https://patchwork.freedesktop.org/series/154627/
State : success
== Summary ==
+ trap cleanup EXIT
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig
[01:28:43] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[01:28:47] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[01:29:17] Starting KUnit Kernel (1/1)...
[01:29:17] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[01:29:17] ================== guc_buf (11 subtests) ===================
[01:29:17] [PASSED] test_smallest
[01:29:17] [PASSED] test_largest
[01:29:17] [PASSED] test_granular
[01:29:17] [PASSED] test_unique
[01:29:17] [PASSED] test_overlap
[01:29:17] [PASSED] test_reusable
[01:29:17] [PASSED] test_too_big
[01:29:17] [PASSED] test_flush
[01:29:17] [PASSED] test_lookup
[01:29:17] [PASSED] test_data
[01:29:17] [PASSED] test_class
[01:29:17] ===================== [PASSED] guc_buf =====================
[01:29:17] =================== guc_dbm (7 subtests) ===================
[01:29:17] [PASSED] test_empty
[01:29:17] [PASSED] test_default
[01:29:17] ======================== test_size ========================
[01:29:17] [PASSED] 4
[01:29:17] [PASSED] 8
[01:29:17] [PASSED] 32
[01:29:17] [PASSED] 256
[01:29:17] ==================== [PASSED] test_size ====================
[01:29:17] ======================= test_reuse ========================
[01:29:17] [PASSED] 4
[01:29:17] [PASSED] 8
[01:29:17] [PASSED] 32
[01:29:17] [PASSED] 256
[01:29:17] =================== [PASSED] test_reuse ====================
[01:29:17] =================== test_range_overlap ====================
[01:29:17] [PASSED] 4
[01:29:17] [PASSED] 8
[01:29:17] [PASSED] 32
[01:29:17] [PASSED] 256
[01:29:17] =============== [PASSED] test_range_overlap ================
[01:29:17] =================== test_range_compact ====================
[01:29:17] [PASSED] 4
[01:29:17] [PASSED] 8
[01:29:17] [PASSED] 32
[01:29:17] [PASSED] 256
[01:29:17] =============== [PASSED] test_range_compact ================
[01:29:17] ==================== test_range_spare =====================
[01:29:17] [PASSED] 4
[01:29:17] [PASSED] 8
[01:29:17] [PASSED] 32
[01:29:17] [PASSED] 256
[01:29:17] ================ [PASSED] test_range_spare =================
[01:29:17] ===================== [PASSED] guc_dbm =====================
[01:29:17] =================== guc_idm (6 subtests) ===================
[01:29:17] [PASSED] bad_init
[01:29:17] [PASSED] no_init
[01:29:17] [PASSED] init_fini
[01:29:17] [PASSED] check_used
[01:29:17] [PASSED] check_quota
[01:29:17] [PASSED] check_all
[01:29:17] ===================== [PASSED] guc_idm =====================
[01:29:17] ================== no_relay (3 subtests) ===================
[01:29:17] [PASSED] xe_drops_guc2pf_if_not_ready
[01:29:17] [PASSED] xe_drops_guc2vf_if_not_ready
[01:29:17] [PASSED] xe_rejects_send_if_not_ready
[01:29:17] ==================== [PASSED] no_relay =====================
[01:29:17] ================== pf_relay (14 subtests) ==================
[01:29:17] [PASSED] pf_rejects_guc2pf_too_short
[01:29:17] [PASSED] pf_rejects_guc2pf_too_long
[01:29:17] [PASSED] pf_rejects_guc2pf_no_payload
[01:29:17] [PASSED] pf_fails_no_payload
[01:29:17] [PASSED] pf_fails_bad_origin
[01:29:17] [PASSED] pf_fails_bad_type
[01:29:17] [PASSED] pf_txn_reports_error
[01:29:17] [PASSED] pf_txn_sends_pf2guc
[01:29:17] [PASSED] pf_sends_pf2guc
[01:29:17] [SKIPPED] pf_loopback_nop
[01:29:17] [SKIPPED] pf_loopback_echo
[01:29:17] [SKIPPED] pf_loopback_fail
[01:29:17] [SKIPPED] pf_loopback_busy
[01:29:17] [SKIPPED] pf_loopback_retry
[01:29:17] ==================== [PASSED] pf_relay =====================
[01:29:17] ================== vf_relay (3 subtests) ===================
[01:29:17] [PASSED] vf_rejects_guc2vf_too_short
[01:29:17] [PASSED] vf_rejects_guc2vf_too_long
[01:29:17] [PASSED] vf_rejects_guc2vf_no_payload
[01:29:17] ==================== [PASSED] vf_relay =====================
[01:29:17] ===================== lmtt (1 subtest) =====================
[01:29:17] ======================== test_ops =========================
[01:29:17] [PASSED] 2-level
[01:29:17] [PASSED] multi-level
[01:29:17] ==================== [PASSED] test_ops =====================
[01:29:17] ====================== [PASSED] lmtt =======================
[01:29:17] ================= pf_service (11 subtests) =================
[01:29:17] [PASSED] pf_negotiate_any
[01:29:17] [PASSED] pf_negotiate_base_match
[01:29:17] [PASSED] pf_negotiate_base_newer
[01:29:17] [PASSED] pf_negotiate_base_next
[01:29:17] [SKIPPED] pf_negotiate_base_older
[01:29:17] [PASSED] pf_negotiate_base_prev
[01:29:17] [PASSED] pf_negotiate_latest_match
[01:29:17] [PASSED] pf_negotiate_latest_newer
[01:29:17] [PASSED] pf_negotiate_latest_next
[01:29:17] [SKIPPED] pf_negotiate_latest_older
[01:29:17] [SKIPPED] pf_negotiate_latest_prev
[01:29:17] =================== [PASSED] pf_service ====================
[01:29:17] ================= xe_guc_g2g (2 subtests) ==================
[01:29:17] ============== xe_live_guc_g2g_kunit_default ==============
[01:29:17] ========= [SKIPPED] xe_live_guc_g2g_kunit_default ==========
[01:29:17] ============== xe_live_guc_g2g_kunit_allmem ===============
[01:29:17] ========== [SKIPPED] xe_live_guc_g2g_kunit_allmem ==========
[01:29:17] =================== [SKIPPED] xe_guc_g2g ===================
[01:29:17] =================== xe_mocs (2 subtests) ===================
[01:29:17] ================ xe_live_mocs_kernel_kunit ================
[01:29:17] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============
[01:29:17] ================ xe_live_mocs_reset_kunit =================
[01:29:17] ============ [SKIPPED] xe_live_mocs_reset_kunit ============
[01:29:17] ==================== [SKIPPED] xe_mocs =====================
[01:29:17] ================= xe_migrate (2 subtests) ==================
[01:29:17] ================= xe_migrate_sanity_kunit =================
[01:29:17] ============ [SKIPPED] xe_migrate_sanity_kunit =============
[01:29:17] ================== xe_validate_ccs_kunit ==================
[01:29:17] ============= [SKIPPED] xe_validate_ccs_kunit ==============
[01:29:17] =================== [SKIPPED] xe_migrate ===================
[01:29:17] ================== xe_dma_buf (1 subtest) ==================
[01:29:17] ==================== xe_dma_buf_kunit =====================
[01:29:17] ================ [SKIPPED] xe_dma_buf_kunit ================
[01:29:17] =================== [SKIPPED] xe_dma_buf ===================
[01:29:17] ================= xe_bo_shrink (1 subtest) =================
[01:29:17] =================== xe_bo_shrink_kunit ====================
[01:29:17] =============== [SKIPPED] xe_bo_shrink_kunit ===============
[01:29:17] ================== [SKIPPED] xe_bo_shrink ==================
[01:29:17] ==================== xe_bo (2 subtests) ====================
[01:29:17] ================== xe_ccs_migrate_kunit ===================
[01:29:17] ============== [SKIPPED] xe_ccs_migrate_kunit ==============
[01:29:17] ==================== xe_bo_evict_kunit ====================
[01:29:17] =============== [SKIPPED] xe_bo_evict_kunit ================
[01:29:17] ===================== [SKIPPED] xe_bo ======================
[01:29:17] ==================== args (11 subtests) ====================
[01:29:17] [PASSED] count_args_test
[01:29:17] [PASSED] call_args_example
[01:29:17] [PASSED] call_args_test
[01:29:17] [PASSED] drop_first_arg_example
[01:29:17] [PASSED] drop_first_arg_test
[01:29:17] [PASSED] first_arg_example
[01:29:17] [PASSED] first_arg_test
[01:29:17] [PASSED] last_arg_example
[01:29:17] [PASSED] last_arg_test
[01:29:17] [PASSED] pick_arg_example
[01:29:17] [PASSED] sep_comma_example
[01:29:17] ====================== [PASSED] args =======================
[01:29:17] =================== xe_pci (3 subtests) ====================
[01:29:17] ==================== check_graphics_ip ====================
[01:29:17] [PASSED] 12.00 Xe_LP
[01:29:17] [PASSED] 12.10 Xe_LP+
[01:29:17] [PASSED] 12.55 Xe_HPG
[01:29:17] [PASSED] 12.60 Xe_HPC
[01:29:17] [PASSED] 12.70 Xe_LPG
[01:29:17] [PASSED] 12.71 Xe_LPG
[01:29:17] [PASSED] 12.74 Xe_LPG+
[01:29:17] [PASSED] 20.01 Xe2_HPG
[01:29:17] [PASSED] 20.02 Xe2_HPG
[01:29:17] [PASSED] 20.04 Xe2_LPG
[01:29:17] [PASSED] 30.00 Xe3_LPG
[01:29:17] [PASSED] 30.01 Xe3_LPG
[01:29:17] [PASSED] 30.03 Xe3_LPG
[01:29:17] ================ [PASSED] check_graphics_ip ================
[01:29:17] ===================== check_media_ip ======================
[01:29:17] [PASSED] 12.00 Xe_M
[01:29:17] [PASSED] 12.55 Xe_HPM
[01:29:17] [PASSED] 13.00 Xe_LPM+
[01:29:17] [PASSED] 13.01 Xe2_HPM
[01:29:17] [PASSED] 20.00 Xe2_LPM
[01:29:17] [PASSED] 30.00 Xe3_LPM
[01:29:17] [PASSED] 30.02 Xe3_LPM
[01:29:17] ================= [PASSED] check_media_ip ==================
[01:29:17] ================= check_platform_gt_count =================
[01:29:17] [PASSED] 0x9A60 (TIGERLAKE)
[01:29:17] [PASSED] 0x9A68 (TIGERLAKE)
[01:29:17] [PASSED] 0x9A70 (TIGERLAKE)
[01:29:17] [PASSED] 0x9A40 (TIGERLAKE)
[01:29:17] [PASSED] 0x9A49 (TIGERLAKE)
[01:29:17] [PASSED] 0x9A59 (TIGERLAKE)
[01:29:17] [PASSED] 0x9A78 (TIGERLAKE)
[01:29:17] [PASSED] 0x9AC0 (TIGERLAKE)
[01:29:17] [PASSED] 0x9AC9 (TIGERLAKE)
[01:29:17] [PASSED] 0x9AD9 (TIGERLAKE)
[01:29:17] [PASSED] 0x9AF8 (TIGERLAKE)
[01:29:17] [PASSED] 0x4C80 (ROCKETLAKE)
[01:29:17] [PASSED] 0x4C8A (ROCKETLAKE)
[01:29:17] [PASSED] 0x4C8B (ROCKETLAKE)
[01:29:17] [PASSED] 0x4C8C (ROCKETLAKE)
[01:29:17] [PASSED] 0x4C90 (ROCKETLAKE)
[01:29:17] [PASSED] 0x4C9A (ROCKETLAKE)
[01:29:17] [PASSED] 0x4680 (ALDERLAKE_S)
[01:29:17] [PASSED] 0x4682 (ALDERLAKE_S)
[01:29:17] [PASSED] 0x4688 (ALDERLAKE_S)
[01:29:17] [PASSED] 0x468A (ALDERLAKE_S)
[01:29:17] [PASSED] 0x468B (ALDERLAKE_S)
[01:29:17] [PASSED] 0x4690 (ALDERLAKE_S)
[01:29:17] [PASSED] 0x4692 (ALDERLAKE_S)
[01:29:17] [PASSED] 0x4693 (ALDERLAKE_S)
[01:29:17] [PASSED] 0x46A0 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46A1 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46A2 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46A3 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46A6 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46A8 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46AA (ALDERLAKE_P)
[01:29:17] [PASSED] 0x462A (ALDERLAKE_P)
[01:29:17] [PASSED] 0x4626 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x4628 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46B0 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46B1 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46B2 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46B3 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46C0 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46C1 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46C2 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46C3 (ALDERLAKE_P)
[01:29:17] [PASSED] 0x46D0 (ALDERLAKE_N)
[01:29:17] [PASSED] 0x46D1 (ALDERLAKE_N)
[01:29:17] [PASSED] 0x46D2 (ALDERLAKE_N)
[01:29:17] [PASSED] 0x46D3 (ALDERLAKE_N)
[01:29:17] [PASSED] 0x46D4 (ALDERLAKE_N)
[01:29:17] [PASSED] 0xA721 (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7A1 (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7A9 (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7AC (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7AD (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA720 (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7A0 (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7A8 (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7AA (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA7AB (ALDERLAKE_P)
[01:29:17] [PASSED] 0xA780 (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA781 (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA782 (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA783 (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA788 (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA789 (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA78A (ALDERLAKE_S)
[01:29:17] [PASSED] 0xA78B (ALDERLAKE_S)
[01:29:17] [PASSED] 0x4905 (DG1)
[01:29:17] [PASSED] 0x4906 (DG1)
[01:29:17] [PASSED] 0x4907 (DG1)
[01:29:17] [PASSED] 0x4908 (DG1)
[01:29:17] [PASSED] 0x4909 (DG1)
[01:29:17] [PASSED] 0x56C0 (DG2)
[01:29:17] [PASSED] 0x56C2 (DG2)
[01:29:17] [PASSED] 0x56C1 (DG2)
[01:29:17] [PASSED] 0x7D51 (METEORLAKE)
[01:29:17] [PASSED] 0x7DD1 (METEORLAKE)
[01:29:17] [PASSED] 0x7D41 (METEORLAKE)
[01:29:17] [PASSED] 0x7D67 (METEORLAKE)
[01:29:17] [PASSED] 0xB640 (METEORLAKE)
[01:29:17] [PASSED] 0x56A0 (DG2)
[01:29:17] [PASSED] 0x56A1 (DG2)
[01:29:17] [PASSED] 0x56A2 (DG2)
[01:29:17] [PASSED] 0x56BE (DG2)
[01:29:17] [PASSED] 0x56BF (DG2)
[01:29:17] [PASSED] 0x5690 (DG2)
[01:29:17] [PASSED] 0x5691 (DG2)
[01:29:17] [PASSED] 0x5692 (DG2)
[01:29:17] [PASSED] 0x56A5 (DG2)
[01:29:17] [PASSED] 0x56A6 (DG2)
[01:29:17] [PASSED] 0x56B0 (DG2)
[01:29:17] [PASSED] 0x56B1 (DG2)
[01:29:17] [PASSED] 0x56BA (DG2)
[01:29:17] [PASSED] 0x56BB (DG2)
[01:29:17] [PASSED] 0x56BC (DG2)
[01:29:17] [PASSED] 0x56BD (DG2)
[01:29:17] [PASSED] 0x5693 (DG2)
[01:29:17] [PASSED] 0x5694 (DG2)
[01:29:17] [PASSED] 0x5695 (DG2)
[01:29:17] [PASSED] 0x56A3 (DG2)
[01:29:17] [PASSED] 0x56A4 (DG2)
[01:29:17] [PASSED] 0x56B2 (DG2)
[01:29:17] [PASSED] 0x56B3 (DG2)
[01:29:17] [PASSED] 0x5696 (DG2)
[01:29:17] [PASSED] 0x5697 (DG2)
[01:29:17] [PASSED] 0xB69 (PVC)
[01:29:17] [PASSED] 0xB6E (PVC)
[01:29:17] [PASSED] 0xBD4 (PVC)
[01:29:17] [PASSED] 0xBD5 (PVC)
[01:29:17] [PASSED] 0xBD6 (PVC)
[01:29:17] [PASSED] 0xBD7 (PVC)
[01:29:17] [PASSED] 0xBD8 (PVC)
[01:29:17] [PASSED] 0xBD9 (PVC)
[01:29:17] [PASSED] 0xBDA (PVC)
[01:29:17] [PASSED] 0xBDB (PVC)
[01:29:17] [PASSED] 0xBE0 (PVC)
[01:29:17] [PASSED] 0xBE1 (PVC)
[01:29:17] [PASSED] 0xBE5 (PVC)
[01:29:17] [PASSED] 0x7D40 (METEORLAKE)
[01:29:17] [PASSED] 0x7D45 (METEORLAKE)
[01:29:17] [PASSED] 0x7D55 (METEORLAKE)
[01:29:17] [PASSED] 0x7D60 (METEORLAKE)
[01:29:17] [PASSED] 0x7DD5 (METEORLAKE)
[01:29:17] [PASSED] 0x6420 (LUNARLAKE)
[01:29:17] [PASSED] 0x64A0 (LUNARLAKE)
[01:29:17] [PASSED] 0x64B0 (LUNARLAKE)
[01:29:17] [PASSED] 0xE202 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE209 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE20B (BATTLEMAGE)
[01:29:17] [PASSED] 0xE20C (BATTLEMAGE)
[01:29:17] [PASSED] 0xE20D (BATTLEMAGE)
[01:29:17] [PASSED] 0xE210 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE211 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE212 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE216 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE220 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE221 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE222 (BATTLEMAGE)
[01:29:17] [PASSED] 0xE223 (BATTLEMAGE)
[01:29:17] [PASSED] 0xB080 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB081 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB082 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB083 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB084 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB085 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB086 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB087 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB08F (PANTHERLAKE)
[01:29:17] [PASSED] 0xB090 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB0A0 (PANTHERLAKE)
[01:29:17] [PASSED] 0xB0B0 (PANTHERLAKE)
[01:29:17] [PASSED] 0xFD80 (PANTHERLAKE)
[01:29:17] [PASSED] 0xFD81 (PANTHERLAKE)
[01:29:17] ============= [PASSED] check_platform_gt_count =============
[01:29:17] ===================== [PASSED] xe_pci ======================
[01:29:17] =================== xe_rtp (2 subtests) ====================
[01:29:17] =============== xe_rtp_process_to_sr_tests ================
[01:29:17] [PASSED] coalesce-same-reg
[01:29:17] [PASSED] no-match-no-add
[01:29:17] [PASSED] match-or
[01:29:17] [PASSED] match-or-xfail
[01:29:17] [PASSED] no-match-no-add-multiple-rules
[01:29:17] [PASSED] two-regs-two-entries
[01:29:17] [PASSED] clr-one-set-other
[01:29:17] [PASSED] set-field
[01:29:17] [PASSED] conflict-duplicate
[01:29:17] [PASSED] conflict-not-disjoint
[01:29:17] [PASSED] conflict-reg-type
[01:29:17] =========== [PASSED] xe_rtp_process_to_sr_tests ============
[01:29:17] ================== xe_rtp_process_tests ===================
[01:29:17] [PASSED] active1
[01:29:17] [PASSED] active2
[01:29:17] [PASSED] active-inactive
[01:29:17] [PASSED] inactive-active
[01:29:17] [PASSED] inactive-1st_or_active-inactive
[01:29:17] [PASSED] inactive-2nd_or_active-inactive
[01:29:17] [PASSED] inactive-last_or_active-inactive
[01:29:17] [PASSED] inactive-no_or_active-inactive
[01:29:17] ============== [PASSED] xe_rtp_process_tests ===============
[01:29:17] ===================== [PASSED] xe_rtp ======================
[01:29:17] ==================== xe_wa (1 subtest) =====================
[01:29:17] ======================== xe_wa_gt =========================
[01:29:17] [PASSED] TIGERLAKE B0
[01:29:17] [PASSED] DG1 A0
[01:29:17] [PASSED] DG1 B0
[01:29:17] [PASSED] ALDERLAKE_S A0
[01:29:17] [PASSED] ALDERLAKE_S B0
stty: 'standard input': Inappropriate ioctl for device
[01:29:17] [PASSED] ALDERLAKE_S C0
[01:29:17] [PASSED] ALDERLAKE_S D0
[01:29:17] [PASSED] ALDERLAKE_P A0
[01:29:17] [PASSED] ALDERLAKE_P B0
[01:29:17] [PASSED] ALDERLAKE_P C0
[01:29:17] [PASSED] ALDERLAKE_S RPLS D0
[01:29:17] [PASSED] ALDERLAKE_P RPLU E0
[01:29:17] [PASSED] DG2 G10 C0
[01:29:17] [PASSED] DG2 G11 B1
[01:29:17] [PASSED] DG2 G12 A1
[01:29:17] [PASSED] METEORLAKE 12.70(Xe_LPG) A0 13.00(Xe_LPM+) A0
[01:29:17] [PASSED] METEORLAKE 12.71(Xe_LPG) A0 13.00(Xe_LPM+) A0
[01:29:17] [PASSED] METEORLAKE 12.74(Xe_LPG+) A0 13.00(Xe_LPM+) A0
[01:29:17] [PASSED] LUNARLAKE 20.04(Xe2_LPG) A0 20.00(Xe2_LPM) A0
[01:29:17] [PASSED] LUNARLAKE 20.04(Xe2_LPG) B0 20.00(Xe2_LPM) A0
[01:29:17] [PASSED] BATTLEMAGE 20.01(Xe2_HPG) A0 13.01(Xe2_HPM) A1
[01:29:17] [PASSED] PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
[01:29:17] ==================== [PASSED] xe_wa_gt =====================
[01:29:17] ====================== [PASSED] xe_wa ======================
[01:29:17] ============================================================
[01:29:17] Testing complete. Ran 306 tests: passed: 288, skipped: 18
[01:29:17] Elapsed time: 33.657s total, 4.236s configuring, 29.055s building, 0.322s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig
[01:29:17] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[01:29:19] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[01:29:42] Starting KUnit Kernel (1/1)...
[01:29:42] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[01:29:42] ============ drm_test_pick_cmdline (2 subtests) ============
[01:29:42] [PASSED] drm_test_pick_cmdline_res_1920_1080_60
[01:29:42] =============== drm_test_pick_cmdline_named ===============
[01:29:42] [PASSED] NTSC
[01:29:42] [PASSED] NTSC-J
[01:29:42] [PASSED] PAL
[01:29:42] [PASSED] PAL-M
[01:29:42] =========== [PASSED] drm_test_pick_cmdline_named ===========
[01:29:42] ============== [PASSED] drm_test_pick_cmdline ==============
[01:29:42] == drm_test_atomic_get_connector_for_encoder (1 subtest) ===
[01:29:42] [PASSED] drm_test_drm_atomic_get_connector_for_encoder
[01:29:42] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ====
[01:29:42] =========== drm_validate_clone_mode (2 subtests) ===========
[01:29:42] ============== drm_test_check_in_clone_mode ===============
[01:29:42] [PASSED] in_clone_mode
[01:29:42] [PASSED] not_in_clone_mode
[01:29:42] ========== [PASSED] drm_test_check_in_clone_mode ===========
[01:29:42] =============== drm_test_check_valid_clones ===============
[01:29:42] [PASSED] not_in_clone_mode
[01:29:42] [PASSED] valid_clone
[01:29:42] [PASSED] invalid_clone
[01:29:42] =========== [PASSED] drm_test_check_valid_clones ===========
[01:29:42] ============= [PASSED] drm_validate_clone_mode =============
[01:29:42] ============= drm_validate_modeset (1 subtest) =============
[01:29:42] [PASSED] drm_test_check_connector_changed_modeset
[01:29:42] ============== [PASSED] drm_validate_modeset ===============
[01:29:42] ====== drm_test_bridge_get_current_state (2 subtests) ======
[01:29:42] [PASSED] drm_test_drm_bridge_get_current_state_atomic
[01:29:42] [PASSED] drm_test_drm_bridge_get_current_state_legacy
[01:29:42] ======== [PASSED] drm_test_bridge_get_current_state ========
[01:29:42] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ======
[01:29:42] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic
[01:29:42] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled
[01:29:42] [PASSED] drm_test_drm_bridge_helper_reset_crtc_legacy
[01:29:42] ======== [PASSED] drm_test_bridge_helper_reset_crtc ========
[01:29:42] ============== drm_bridge_alloc (2 subtests) ===============
[01:29:42] [PASSED] drm_test_drm_bridge_alloc_basic
[01:29:42] [PASSED] drm_test_drm_bridge_alloc_get_put
[01:29:42] ================ [PASSED] drm_bridge_alloc =================
[01:29:42] ================== drm_buddy (7 subtests) ==================
[01:29:42] [PASSED] drm_test_buddy_alloc_limit
[01:29:42] [PASSED] drm_test_buddy_alloc_optimistic
[01:29:42] [PASSED] drm_test_buddy_alloc_pessimistic
[01:29:42] [PASSED] drm_test_buddy_alloc_pathological
[01:29:42] [PASSED] drm_test_buddy_alloc_contiguous
[01:29:42] [PASSED] drm_test_buddy_alloc_clear
[01:29:42] [PASSED] drm_test_buddy_alloc_range_bias
[01:29:42] ==================== [PASSED] drm_buddy ====================
[01:29:42] ============= drm_cmdline_parser (40 subtests) =============
[01:29:42] [PASSED] drm_test_cmdline_force_d_only
[01:29:42] [PASSED] drm_test_cmdline_force_D_only_dvi
[01:29:42] [PASSED] drm_test_cmdline_force_D_only_hdmi
[01:29:42] [PASSED] drm_test_cmdline_force_D_only_not_digital
[01:29:42] [PASSED] drm_test_cmdline_force_e_only
[01:29:42] [PASSED] drm_test_cmdline_res
[01:29:42] [PASSED] drm_test_cmdline_res_vesa
[01:29:42] [PASSED] drm_test_cmdline_res_vesa_rblank
[01:29:42] [PASSED] drm_test_cmdline_res_rblank
[01:29:42] [PASSED] drm_test_cmdline_res_bpp
[01:29:42] [PASSED] drm_test_cmdline_res_refresh
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_margins
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital
[01:29:42] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on
[01:29:42] [PASSED] drm_test_cmdline_res_margins_force_on
[01:29:42] [PASSED] drm_test_cmdline_res_vesa_margins
[01:29:42] [PASSED] drm_test_cmdline_name
[01:29:42] [PASSED] drm_test_cmdline_name_bpp
[01:29:42] [PASSED] drm_test_cmdline_name_option
[01:29:42] [PASSED] drm_test_cmdline_name_bpp_option
[01:29:42] [PASSED] drm_test_cmdline_rotate_0
[01:29:42] [PASSED] drm_test_cmdline_rotate_90
[01:29:42] [PASSED] drm_test_cmdline_rotate_180
[01:29:42] [PASSED] drm_test_cmdline_rotate_270
[01:29:42] [PASSED] drm_test_cmdline_hmirror
[01:29:42] [PASSED] drm_test_cmdline_vmirror
[01:29:42] [PASSED] drm_test_cmdline_margin_options
[01:29:42] [PASSED] drm_test_cmdline_multiple_options
[01:29:42] [PASSED] drm_test_cmdline_bpp_extra_and_option
[01:29:42] [PASSED] drm_test_cmdline_extra_and_option
[01:29:42] [PASSED] drm_test_cmdline_freestanding_options
[01:29:42] [PASSED] drm_test_cmdline_freestanding_force_e_and_options
[01:29:42] [PASSED] drm_test_cmdline_panel_orientation
[01:29:42] ================ drm_test_cmdline_invalid =================
[01:29:42] [PASSED] margin_only
[01:29:42] [PASSED] interlace_only
[01:29:42] [PASSED] res_missing_x
[01:29:42] [PASSED] res_missing_y
[01:29:42] [PASSED] res_bad_y
[01:29:42] [PASSED] res_missing_y_bpp
[01:29:42] [PASSED] res_bad_bpp
[01:29:42] [PASSED] res_bad_refresh
[01:29:42] [PASSED] res_bpp_refresh_force_on_off
[01:29:42] [PASSED] res_invalid_mode
[01:29:42] [PASSED] res_bpp_wrong_place_mode
[01:29:42] [PASSED] name_bpp_refresh
[01:29:42] [PASSED] name_refresh
[01:29:42] [PASSED] name_refresh_wrong_mode
[01:29:42] [PASSED] name_refresh_invalid_mode
[01:29:42] [PASSED] rotate_multiple
[01:29:42] [PASSED] rotate_invalid_val
[01:29:42] [PASSED] rotate_truncated
[01:29:42] [PASSED] invalid_option
[01:29:42] [PASSED] invalid_tv_option
[01:29:42] [PASSED] truncated_tv_option
[01:29:42] ============ [PASSED] drm_test_cmdline_invalid =============
[01:29:42] =============== drm_test_cmdline_tv_options ===============
[01:29:42] [PASSED] NTSC
[01:29:42] [PASSED] NTSC_443
[01:29:42] [PASSED] NTSC_J
[01:29:42] [PASSED] PAL
[01:29:42] [PASSED] PAL_M
[01:29:42] [PASSED] PAL_N
[01:29:42] [PASSED] SECAM
[01:29:42] [PASSED] MONO_525
[01:29:42] [PASSED] MONO_625
[01:29:42] =========== [PASSED] drm_test_cmdline_tv_options ===========
[01:29:42] =============== [PASSED] drm_cmdline_parser ================
[01:29:42] ========== drmm_connector_hdmi_init (20 subtests) ==========
[01:29:42] [PASSED] drm_test_connector_hdmi_init_valid
[01:29:42] [PASSED] drm_test_connector_hdmi_init_bpc_8
[01:29:42] [PASSED] drm_test_connector_hdmi_init_bpc_10
[01:29:42] [PASSED] drm_test_connector_hdmi_init_bpc_12
[01:29:42] [PASSED] drm_test_connector_hdmi_init_bpc_invalid
[01:29:42] [PASSED] drm_test_connector_hdmi_init_bpc_null
[01:29:42] [PASSED] drm_test_connector_hdmi_init_formats_empty
[01:29:42] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb
[01:29:42] === drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[01:29:42] [PASSED] supported_formats=0x9 yuv420_allowed=1
[01:29:42] [PASSED] supported_formats=0x9 yuv420_allowed=0
[01:29:42] [PASSED] supported_formats=0x3 yuv420_allowed=1
[01:29:42] [PASSED] supported_formats=0x3 yuv420_allowed=0
[01:29:42] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[01:29:42] [PASSED] drm_test_connector_hdmi_init_null_ddc
[01:29:42] [PASSED] drm_test_connector_hdmi_init_null_product
[01:29:42] [PASSED] drm_test_connector_hdmi_init_null_vendor
[01:29:42] [PASSED] drm_test_connector_hdmi_init_product_length_exact
[01:29:42] [PASSED] drm_test_connector_hdmi_init_product_length_too_long
[01:29:42] [PASSED] drm_test_connector_hdmi_init_product_valid
[01:29:42] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact
[01:29:42] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long
[01:29:42] [PASSED] drm_test_connector_hdmi_init_vendor_valid
[01:29:42] ========= drm_test_connector_hdmi_init_type_valid =========
[01:29:42] [PASSED] HDMI-A
[01:29:42] [PASSED] HDMI-B
[01:29:42] ===== [PASSED] drm_test_connector_hdmi_init_type_valid =====
[01:29:42] ======== drm_test_connector_hdmi_init_type_invalid ========
[01:29:42] [PASSED] Unknown
[01:29:42] [PASSED] VGA
[01:29:42] [PASSED] DVI-I
[01:29:42] [PASSED] DVI-D
[01:29:42] [PASSED] DVI-A
[01:29:42] [PASSED] Composite
[01:29:42] [PASSED] SVIDEO
[01:29:42] [PASSED] LVDS
[01:29:42] [PASSED] Component
[01:29:42] [PASSED] DIN
[01:29:42] [PASSED] DP
[01:29:42] [PASSED] TV
[01:29:42] [PASSED] eDP
[01:29:42] [PASSED] Virtual
[01:29:42] [PASSED] DSI
[01:29:42] [PASSED] DPI
[01:29:42] [PASSED] Writeback
[01:29:42] [PASSED] SPI
[01:29:42] [PASSED] USB
[01:29:42] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ====
[01:29:42] ============ [PASSED] drmm_connector_hdmi_init =============
[01:29:42] ============= drmm_connector_init (3 subtests) =============
[01:29:42] [PASSED] drm_test_drmm_connector_init
[01:29:42] [PASSED] drm_test_drmm_connector_init_null_ddc
[01:29:42] ========= drm_test_drmm_connector_init_type_valid =========
[01:29:42] [PASSED] Unknown
[01:29:42] [PASSED] VGA
[01:29:42] [PASSED] DVI-I
[01:29:42] [PASSED] DVI-D
[01:29:42] [PASSED] DVI-A
[01:29:42] [PASSED] Composite
[01:29:42] [PASSED] SVIDEO
[01:29:42] [PASSED] LVDS
[01:29:42] [PASSED] Component
[01:29:42] [PASSED] DIN
[01:29:42] [PASSED] DP
[01:29:42] [PASSED] HDMI-A
[01:29:42] [PASSED] HDMI-B
[01:29:42] [PASSED] TV
[01:29:42] [PASSED] eDP
[01:29:42] [PASSED] Virtual
[01:29:42] [PASSED] DSI
[01:29:42] [PASSED] DPI
[01:29:42] [PASSED] Writeback
[01:29:42] [PASSED] SPI
[01:29:42] [PASSED] USB
[01:29:42] ===== [PASSED] drm_test_drmm_connector_init_type_valid =====
[01:29:42] =============== [PASSED] drmm_connector_init ===============
[01:29:42] ========= drm_connector_dynamic_init (6 subtests) ==========
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_init
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_init_not_added
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_init_properties
[01:29:42] ===== drm_test_drm_connector_dynamic_init_type_valid ======
[01:29:42] [PASSED] Unknown
[01:29:42] [PASSED] VGA
[01:29:42] [PASSED] DVI-I
[01:29:42] [PASSED] DVI-D
[01:29:42] [PASSED] DVI-A
[01:29:42] [PASSED] Composite
[01:29:42] [PASSED] SVIDEO
[01:29:42] [PASSED] LVDS
[01:29:42] [PASSED] Component
[01:29:42] [PASSED] DIN
[01:29:42] [PASSED] DP
[01:29:42] [PASSED] HDMI-A
[01:29:42] [PASSED] HDMI-B
[01:29:42] [PASSED] TV
[01:29:42] [PASSED] eDP
[01:29:42] [PASSED] Virtual
[01:29:42] [PASSED] DSI
[01:29:42] [PASSED] DPI
[01:29:42] [PASSED] Writeback
[01:29:42] [PASSED] SPI
[01:29:42] [PASSED] USB
[01:29:42] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid ==
[01:29:42] ======== drm_test_drm_connector_dynamic_init_name =========
[01:29:42] [PASSED] Unknown
[01:29:42] [PASSED] VGA
[01:29:42] [PASSED] DVI-I
[01:29:42] [PASSED] DVI-D
[01:29:42] [PASSED] DVI-A
[01:29:42] [PASSED] Composite
[01:29:42] [PASSED] SVIDEO
[01:29:42] [PASSED] LVDS
[01:29:42] [PASSED] Component
[01:29:42] [PASSED] DIN
[01:29:42] [PASSED] DP
[01:29:42] [PASSED] HDMI-A
[01:29:42] [PASSED] HDMI-B
[01:29:42] [PASSED] TV
[01:29:42] [PASSED] eDP
[01:29:42] [PASSED] Virtual
[01:29:42] [PASSED] DSI
[01:29:42] [PASSED] DPI
[01:29:42] [PASSED] Writeback
[01:29:42] [PASSED] SPI
[01:29:42] [PASSED] USB
[01:29:42] ==== [PASSED] drm_test_drm_connector_dynamic_init_name =====
[01:29:42] =========== [PASSED] drm_connector_dynamic_init ============
[01:29:42] ==== drm_connector_dynamic_register_early (4 subtests) =====
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_early_defer
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object
[01:29:42] ====== [PASSED] drm_connector_dynamic_register_early =======
[01:29:42] ======= drm_connector_dynamic_register (7 subtests) ========
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_on_list
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_no_defer
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_no_init
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_mode_object
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_sysfs
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name
[01:29:42] [PASSED] drm_test_drm_connector_dynamic_register_debugfs
[01:29:42] ========= [PASSED] drm_connector_dynamic_register ==========
[01:29:42] = drm_connector_attach_broadcast_rgb_property (2 subtests) =
[01:29:42] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property
[01:29:42] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector
[01:29:42] === [PASSED] drm_connector_attach_broadcast_rgb_property ===
[01:29:42] ========== drm_get_tv_mode_from_name (2 subtests) ==========
[01:29:42] ========== drm_test_get_tv_mode_from_name_valid ===========
[01:29:42] [PASSED] NTSC
[01:29:42] [PASSED] NTSC-443
[01:29:42] [PASSED] NTSC-J
[01:29:42] [PASSED] PAL
[01:29:42] [PASSED] PAL-M
[01:29:42] [PASSED] PAL-N
[01:29:42] [PASSED] SECAM
[01:29:42] [PASSED] Mono
[01:29:42] ====== [PASSED] drm_test_get_tv_mode_from_name_valid =======
[01:29:42] [PASSED] drm_test_get_tv_mode_from_name_truncated
[01:29:42] ============ [PASSED] drm_get_tv_mode_from_name ============
[01:29:42] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) =
[01:29:42] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb
[01:29:42] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc
[01:29:42] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1
[01:29:42] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc
[01:29:42] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1
[01:29:42] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double
[01:29:42] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid =
[01:29:42] [PASSED] VIC 96
[01:29:42] [PASSED] VIC 97
[01:29:42] [PASSED] VIC 101
[01:29:42] [PASSED] VIC 102
[01:29:42] [PASSED] VIC 106
[01:29:42] [PASSED] VIC 107
[01:29:42] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid ===
[01:29:42] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc
[01:29:42] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc
[01:29:42] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc
[01:29:42] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc
[01:29:42] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc
[01:29:42] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ====
[01:29:42] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) ==
[01:29:42] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name ====
[01:29:42] [PASSED] Automatic
[01:29:42] [PASSED] Full
[01:29:42] [PASSED] Limited 16:235
[01:29:42] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name ===
[01:29:42] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid
[01:29:42] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ====
[01:29:42] == drm_hdmi_connector_get_output_format_name (2 subtests) ==
[01:29:42] === drm_test_drm_hdmi_connector_get_output_format_name ====
[01:29:42] [PASSED] RGB
[01:29:42] [PASSED] YUV 4:2:0
[01:29:42] [PASSED] YUV 4:2:2
[01:29:42] [PASSED] YUV 4:4:4
[01:29:42] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name ===
[01:29:42] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid
[01:29:42] ==== [PASSED] drm_hdmi_connector_get_output_format_name ====
[01:29:42] ============= drm_damage_helper (21 subtests) ==============
[01:29:42] [PASSED] drm_test_damage_iter_no_damage
[01:29:42] [PASSED] drm_test_damage_iter_no_damage_fractional_src
[01:29:42] [PASSED] drm_test_damage_iter_no_damage_src_moved
[01:29:42] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved
[01:29:42] [PASSED] drm_test_damage_iter_no_damage_not_visible
[01:29:42] [PASSED] drm_test_damage_iter_no_damage_no_crtc
[01:29:42] [PASSED] drm_test_damage_iter_no_damage_no_fb
[01:29:42] [PASSED] drm_test_damage_iter_simple_damage
[01:29:42] [PASSED] drm_test_damage_iter_single_damage
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_intersect_src
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_outside_src
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_fractional_src
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_src_moved
[01:29:42] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved
[01:29:42] [PASSED] drm_test_damage_iter_damage
[01:29:42] [PASSED] drm_test_damage_iter_damage_one_intersect
[01:29:42] [PASSED] drm_test_damage_iter_damage_one_outside
[01:29:42] [PASSED] drm_test_damage_iter_damage_src_moved
[01:29:42] [PASSED] drm_test_damage_iter_damage_not_visible
[01:29:42] ================ [PASSED] drm_damage_helper ================
[01:29:42] ============== drm_dp_mst_helper (3 subtests) ==============
[01:29:42] ============== drm_test_dp_mst_calc_pbn_mode ==============
[01:29:42] [PASSED] Clock 154000 BPP 30 DSC disabled
[01:29:42] [PASSED] Clock 234000 BPP 30 DSC disabled
[01:29:42] [PASSED] Clock 297000 BPP 24 DSC disabled
[01:29:42] [PASSED] Clock 332880 BPP 24 DSC enabled
[01:29:42] [PASSED] Clock 324540 BPP 24 DSC enabled
[01:29:42] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ==========
[01:29:42] ============== drm_test_dp_mst_calc_pbn_div ===============
[01:29:42] [PASSED] Link rate 2000000 lane count 4
[01:29:42] [PASSED] Link rate 2000000 lane count 2
[01:29:42] [PASSED] Link rate 2000000 lane count 1
[01:29:42] [PASSED] Link rate 1350000 lane count 4
[01:29:42] [PASSED] Link rate 1350000 lane count 2
[01:29:42] [PASSED] Link rate 1350000 lane count 1
[01:29:42] [PASSED] Link rate 1000000 lane count 4
[01:29:42] [PASSED] Link rate 1000000 lane count 2
[01:29:42] [PASSED] Link rate 1000000 lane count 1
[01:29:42] [PASSED] Link rate 810000 lane count 4
[01:29:42] [PASSED] Link rate 810000 lane count 2
[01:29:42] [PASSED] Link rate 810000 lane count 1
[01:29:42] [PASSED] Link rate 540000 lane count 4
[01:29:42] [PASSED] Link rate 540000 lane count 2
[01:29:42] [PASSED] Link rate 540000 lane count 1
[01:29:42] [PASSED] Link rate 270000 lane count 4
[01:29:42] [PASSED] Link rate 270000 lane count 2
[01:29:42] [PASSED] Link rate 270000 lane count 1
[01:29:42] [PASSED] Link rate 162000 lane count 4
[01:29:42] [PASSED] Link rate 162000 lane count 2
[01:29:42] [PASSED] Link rate 162000 lane count 1
[01:29:42] ========== [PASSED] drm_test_dp_mst_calc_pbn_div ===========
[01:29:42] ========= drm_test_dp_mst_sideband_msg_req_decode =========
[01:29:42] [PASSED] DP_ENUM_PATH_RESOURCES with port number
[01:29:42] [PASSED] DP_POWER_UP_PHY with port number
[01:29:42] [PASSED] DP_POWER_DOWN_PHY with port number
[01:29:42] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks
[01:29:42] [PASSED] DP_ALLOCATE_PAYLOAD with port number
[01:29:42] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI
[01:29:42] [PASSED] DP_ALLOCATE_PAYLOAD with PBN
[01:29:42] [PASSED] DP_QUERY_PAYLOAD with port number
[01:29:42] [PASSED] DP_QUERY_PAYLOAD with VCPI
[01:29:42] [PASSED] DP_REMOTE_DPCD_READ with port number
[01:29:42] [PASSED] DP_REMOTE_DPCD_READ with DPCD address
[01:29:42] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes
[01:29:42] [PASSED] DP_REMOTE_DPCD_WRITE with port number
[01:29:42] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address
[01:29:42] [PASSED] DP_REMOTE_DPCD_WRITE with data array
[01:29:42] [PASSED] DP_REMOTE_I2C_READ with port number
[01:29:42] [PASSED] DP_REMOTE_I2C_READ with I2C device ID
[01:29:42] [PASSED] DP_REMOTE_I2C_READ with transactions array
[01:29:42] [PASSED] DP_REMOTE_I2C_WRITE with port number
[01:29:42] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID
[01:29:42] [PASSED] DP_REMOTE_I2C_WRITE with data array
[01:29:42] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID
[01:29:42] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID
[01:29:42] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event
[01:29:42] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event
[01:29:42] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior
[01:29:42] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior
[01:29:42] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode =====
[01:29:42] ================ [PASSED] drm_dp_mst_helper ================
[01:29:42] ================== drm_exec (7 subtests) ===================
[01:29:42] [PASSED] sanitycheck
[01:29:42] [PASSED] test_lock
[01:29:42] [PASSED] test_lock_unlock
[01:29:42] [PASSED] test_duplicates
[01:29:42] [PASSED] test_prepare
[01:29:42] [PASSED] test_prepare_array
[01:29:42] [PASSED] test_multiple_loops
[01:29:42] ==================== [PASSED] drm_exec =====================
[01:29:42] =========== drm_format_helper_test (17 subtests) ===========
[01:29:42] ============== drm_test_fb_xrgb8888_to_gray8 ==============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ==========
[01:29:42] ============= drm_test_fb_xrgb8888_to_rgb332 ==============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ==========
[01:29:42] ============= drm_test_fb_xrgb8888_to_rgb565 ==============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ==========
[01:29:42] ============ drm_test_fb_xrgb8888_to_xrgb1555 =============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 =========
[01:29:42] ============ drm_test_fb_xrgb8888_to_argb1555 =============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 =========
[01:29:42] ============ drm_test_fb_xrgb8888_to_rgba5551 =============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 =========
[01:29:42] ============= drm_test_fb_xrgb8888_to_rgb888 ==============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ==========
[01:29:42] ============= drm_test_fb_xrgb8888_to_bgr888 ==============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ==========
[01:29:42] ============ drm_test_fb_xrgb8888_to_argb8888 =============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 =========
[01:29:42] =========== drm_test_fb_xrgb8888_to_xrgb2101010 ===========
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 =======
[01:29:42] =========== drm_test_fb_xrgb8888_to_argb2101010 ===========
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 =======
[01:29:42] ============== drm_test_fb_xrgb8888_to_mono ===============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ========== [PASSED] drm_test_fb_xrgb8888_to_mono ===========
[01:29:42] ==================== drm_test_fb_swab =====================
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ================ [PASSED] drm_test_fb_swab =================
[01:29:42] ============ drm_test_fb_xrgb8888_to_xbgr8888 =============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 =========
[01:29:42] ============ drm_test_fb_xrgb8888_to_abgr8888 =============
[01:29:42] [PASSED] single_pixel_source_buffer
[01:29:42] [PASSED] single_pixel_clip_rectangle
[01:29:42] [PASSED] well_known_colors
[01:29:42] [PASSED] destination_pitch
[01:29:42] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 =========
[01:29:42] ================= drm_test_fb_clip_offset =================
[01:29:42] [PASSED] pass through
[01:29:42] [PASSED] horizontal offset
[01:29:42] [PASSED] vertical offset
[01:29:42] [PASSED] horizontal and vertical offset
[01:29:42] [PASSED] horizontal offset (custom pitch)
[01:29:42] [PASSED] vertical offset (custom pitch)
[01:29:42] [PASSED] horizontal and vertical offset (custom pitch)
[01:29:42] ============= [PASSED] drm_test_fb_clip_offset =============
[01:29:42] =================== drm_test_fb_memcpy ====================
[01:29:42] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258)
[01:29:42] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258)
[01:29:42] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559)
[01:29:42] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258)
[01:29:42] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258)
[01:29:42] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559)
[01:29:42] [PASSED] well_known_colors: XB24 little-endian (0x34324258)
[01:29:42] [PASSED] well_known_colors: XRA8 little-endian (0x38415258)
[01:29:42] [PASSED] well_known_colors: YU24 little-endian (0x34325559)
[01:29:42] [PASSED] destination_pitch: XB24 little-endian (0x34324258)
[01:29:42] [PASSED] destination_pitch: XRA8 little-endian (0x38415258)
[01:29:42] [PASSED] destination_pitch: YU24 little-endian (0x34325559)
[01:29:42] =============== [PASSED] drm_test_fb_memcpy ================
[01:29:42] ============= [PASSED] drm_format_helper_test ==============
[01:29:42] ================= drm_format (18 subtests) =================
[01:29:42] [PASSED] drm_test_format_block_width_invalid
[01:29:42] [PASSED] drm_test_format_block_width_one_plane
[01:29:42] [PASSED] drm_test_format_block_width_two_plane
[01:29:42] [PASSED] drm_test_format_block_width_three_plane
[01:29:42] [PASSED] drm_test_format_block_width_tiled
[01:29:42] [PASSED] drm_test_format_block_height_invalid
[01:29:42] [PASSED] drm_test_format_block_height_one_plane
[01:29:42] [PASSED] drm_test_format_block_height_two_plane
[01:29:42] [PASSED] drm_test_format_block_height_three_plane
[01:29:42] [PASSED] drm_test_format_block_height_tiled
[01:29:42] [PASSED] drm_test_format_min_pitch_invalid
[01:29:42] [PASSED] drm_test_format_min_pitch_one_plane_8bpp
[01:29:42] [PASSED] drm_test_format_min_pitch_one_plane_16bpp
[01:29:42] [PASSED] drm_test_format_min_pitch_one_plane_24bpp
[01:29:42] [PASSED] drm_test_format_min_pitch_one_plane_32bpp
[01:29:42] [PASSED] drm_test_format_min_pitch_two_plane
[01:29:42] [PASSED] drm_test_format_min_pitch_three_plane_8bpp
[01:29:42] [PASSED] drm_test_format_min_pitch_tiled
[01:29:42] =================== [PASSED] drm_format ====================
[01:29:42] ============== drm_framebuffer (10 subtests) ===============
[01:29:42] ========== drm_test_framebuffer_check_src_coords ==========
[01:29:42] [PASSED] Success: source fits into fb
[01:29:42] [PASSED] Fail: overflowing fb with x-axis coordinate
[01:29:42] [PASSED] Fail: overflowing fb with y-axis coordinate
[01:29:42] [PASSED] Fail: overflowing fb with source width
[01:29:42] [PASSED] Fail: overflowing fb with source height
[01:29:42] ====== [PASSED] drm_test_framebuffer_check_src_coords ======
[01:29:42] [PASSED] drm_test_framebuffer_cleanup
[01:29:42] =============== drm_test_framebuffer_create ===============
[01:29:42] [PASSED] ABGR8888 normal sizes
[01:29:42] [PASSED] ABGR8888 max sizes
[01:29:42] [PASSED] ABGR8888 pitch greater than min required
[01:29:42] [PASSED] ABGR8888 pitch less than min required
[01:29:42] [PASSED] ABGR8888 Invalid width
[01:29:42] [PASSED] ABGR8888 Invalid buffer handle
[01:29:42] [PASSED] No pixel format
[01:29:42] [PASSED] ABGR8888 Width 0
[01:29:42] [PASSED] ABGR8888 Height 0
[01:29:42] [PASSED] ABGR8888 Out of bound height * pitch combination
[01:29:42] [PASSED] ABGR8888 Large buffer offset
[01:29:42] [PASSED] ABGR8888 Buffer offset for inexistent plane
[01:29:42] [PASSED] ABGR8888 Invalid flag
[01:29:42] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers
[01:29:42] [PASSED] ABGR8888 Valid buffer modifier
[01:29:42] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE)
[01:29:42] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] NV12 Normal sizes
[01:29:42] [PASSED] NV12 Max sizes
[01:29:42] [PASSED] NV12 Invalid pitch
[01:29:42] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag
[01:29:42] [PASSED] NV12 different modifier per-plane
[01:29:42] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE
[01:29:42] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] NV12 Modifier for inexistent plane
[01:29:42] [PASSED] NV12 Handle for inexistent plane
[01:29:42] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier
[01:29:42] [PASSED] YVU420 Normal sizes
[01:29:42] [PASSED] YVU420 Max sizes
[01:29:42] [PASSED] YVU420 Invalid pitch
[01:29:42] [PASSED] YVU420 Different pitches
[01:29:42] [PASSED] YVU420 Different buffer offsets/pitches
[01:29:42] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS
[01:29:42] [PASSED] YVU420 Valid modifier
[01:29:42] [PASSED] YVU420 Different modifiers per plane
[01:29:42] [PASSED] YVU420 Modifier for inexistent plane
[01:29:42] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR)
[01:29:42] [PASSED] X0L2 Normal sizes
[01:29:42] [PASSED] X0L2 Max sizes
[01:29:42] [PASSED] X0L2 Invalid pitch
[01:29:42] [PASSED] X0L2 Pitch greater than minimum required
[01:29:42] [PASSED] X0L2 Handle for inexistent plane
[01:29:42] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set
[01:29:42] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set
[01:29:42] [PASSED] X0L2 Valid modifier
[01:29:42] [PASSED] X0L2 Modifier for inexistent plane
[01:29:42] =========== [PASSED] drm_test_framebuffer_create ===========
[01:29:42] [PASSED] drm_test_framebuffer_free
[01:29:42] [PASSED] drm_test_framebuffer_init
[01:29:42] [PASSED] drm_test_framebuffer_init_bad_format
[01:29:42] [PASSED] drm_test_framebuffer_init_dev_mismatch
[01:29:42] [PASSED] drm_test_framebuffer_lookup
[01:29:42] [PASSED] drm_test_framebuffer_lookup_inexistent
[01:29:42] [PASSED] drm_test_framebuffer_modifiers_not_supported
[01:29:42] ================= [PASSED] drm_framebuffer =================
[01:29:42] ================ drm_gem_shmem (8 subtests) ================
[01:29:42] [PASSED] drm_gem_shmem_test_obj_create
[01:29:42] [PASSED] drm_gem_shmem_test_obj_create_private
[01:29:42] [PASSED] drm_gem_shmem_test_pin_pages
[01:29:42] [PASSED] drm_gem_shmem_test_vmap
[01:29:42] [PASSED] drm_gem_shmem_test_get_pages_sgt
[01:29:42] [PASSED] drm_gem_shmem_test_get_sg_table
[01:29:42] [PASSED] drm_gem_shmem_test_madvise
[01:29:42] [PASSED] drm_gem_shmem_test_purge
[01:29:42] ================== [PASSED] drm_gem_shmem ==================
[01:29:42] === drm_atomic_helper_connector_hdmi_check (27 subtests) ===
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1
[01:29:42] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420 =======
[01:29:42] [PASSED] Automatic
[01:29:42] [PASSED] Full
[01:29:42] [PASSED] Limited 16:235
[01:29:42] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 ===
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed
[01:29:42] [PASSED] drm_test_check_disable_connector
[01:29:42] [PASSED] drm_test_check_hdmi_funcs_reject_rate
[01:29:42] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb
[01:29:42] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420
[01:29:42] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422
[01:29:42] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420
[01:29:42] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420
[01:29:42] [PASSED] drm_test_check_output_bpc_crtc_mode_changed
[01:29:42] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed
[01:29:42] [PASSED] drm_test_check_output_bpc_dvi
[01:29:42] [PASSED] drm_test_check_output_bpc_format_vic_1
[01:29:42] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only
[01:29:42] [PASSED] drm_test_check_output_bpc_format_display_rgb_only
[01:29:42] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only
[01:29:42] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only
[01:29:42] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc
[01:29:42] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc
[01:29:42] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc
[01:29:42] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ======
[01:29:42] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ====
[01:29:42] [PASSED] drm_test_check_broadcast_rgb_value
[01:29:42] [PASSED] drm_test_check_bpc_8_value
[01:29:42] [PASSED] drm_test_check_bpc_10_value
[01:29:42] [PASSED] drm_test_check_bpc_12_value
[01:29:42] [PASSED] drm_test_check_format_value
[01:29:42] [PASSED] drm_test_check_tmds_char_value
[01:29:42] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ======
[01:29:42] = drm_atomic_helper_connector_hdmi_mode_valid (4 subtests) =
[01:29:42] [PASSED] drm_test_check_mode_valid
[01:29:42] [PASSED] drm_test_check_mode_valid_reject
[01:29:42] [PASSED] drm_test_check_mode_valid_reject_rate
[01:29:42] [PASSED] drm_test_check_mode_valid_reject_max_clock
[01:29:42] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid ===
[01:29:42] ================= drm_managed (2 subtests) =================
[01:29:42] [PASSED] drm_test_managed_release_action
[01:29:42] [PASSED] drm_test_managed_run_action
[01:29:42] =================== [PASSED] drm_managed ===================
[01:29:42] =================== drm_mm (6 subtests) ====================
[01:29:42] [PASSED] drm_test_mm_init
[01:29:42] [PASSED] drm_test_mm_debug
[01:29:42] [PASSED] drm_test_mm_align32
[01:29:42] [PASSED] drm_test_mm_align64
[01:29:42] [PASSED] drm_test_mm_lowest
[01:29:42] [PASSED] drm_test_mm_highest
[01:29:42] ===================== [PASSED] drm_mm ======================
[01:29:42] ============= drm_modes_analog_tv (5 subtests) =============
[01:29:42] [PASSED] drm_test_modes_analog_tv_mono_576i
[01:29:42] [PASSED] drm_test_modes_analog_tv_ntsc_480i
[01:29:42] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined
[01:29:42] [PASSED] drm_test_modes_analog_tv_pal_576i
[01:29:42] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined
[01:29:42] =============== [PASSED] drm_modes_analog_tv ===============
[01:29:42] ============== drm_plane_helper (2 subtests) ===============
[01:29:42] =============== drm_test_check_plane_state ================
[01:29:42] [PASSED] clipping_simple
[01:29:42] [PASSED] clipping_rotate_reflect
[01:29:42] [PASSED] positioning_simple
[01:29:42] [PASSED] upscaling
[01:29:42] [PASSED] downscaling
[01:29:42] [PASSED] rounding1
[01:29:42] [PASSED] rounding2
[01:29:42] [PASSED] rounding3
[01:29:42] [PASSED] rounding4
[01:29:42] =========== [PASSED] drm_test_check_plane_state ============
[01:29:42] =========== drm_test_check_invalid_plane_state ============
[01:29:42] [PASSED] positioning_invalid
[01:29:42] [PASSED] upscaling_invalid
[01:29:42] [PASSED] downscaling_invalid
[01:29:42] ======= [PASSED] drm_test_check_invalid_plane_state ========
[01:29:42] ================ [PASSED] drm_plane_helper =================
[01:29:42] ====== drm_connector_helper_tv_get_modes (1 subtest) =======
[01:29:42] ====== drm_test_connector_helper_tv_get_modes_check =======
[01:29:42] [PASSED] None
[01:29:42] [PASSED] PAL
[01:29:42] [PASSED] NTSC
[01:29:42] [PASSED] Both, NTSC Default
[01:29:42] [PASSED] Both, PAL Default
[01:29:42] [PASSED] Both, NTSC Default, with PAL on command-line
[01:29:42] [PASSED] Both, PAL Default, with NTSC on command-line
[01:29:42] == [PASSED] drm_test_connector_helper_tv_get_modes_check ===
[01:29:42] ======== [PASSED] drm_connector_helper_tv_get_modes ========
[01:29:42] ================== drm_rect (9 subtests) ===================
[01:29:42] [PASSED] drm_test_rect_clip_scaled_div_by_zero
[01:29:42] [PASSED] drm_test_rect_clip_scaled_not_clipped
[01:29:42] [PASSED] drm_test_rect_clip_scaled_clipped
[01:29:42] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned
[01:29:42] ================= drm_test_rect_intersect =================
[01:29:42] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0
[01:29:42] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1
[01:29:42] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0
[01:29:42] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1
[01:29:42] [PASSED] right x left: 2x1+0+0 x 3x1+1+0
[01:29:42] [PASSED] left x right: 3x1+1+0 x 2x1+0+0
[01:29:42] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1
[01:29:42] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0
[01:29:42] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1
[01:29:42] [PASSED] touching side: 1x1+0+0 x 1x1+1+0
[01:29:42] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0
[01:29:42] [PASSED] inside another: 2x2+0+0 x 1x1+1+1
[01:29:42] [PASSED] far away: 1x1+0+0 x 1x1+3+6
[01:29:42] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10
[01:29:42] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10
[01:29:42] ============= [PASSED] drm_test_rect_intersect =============
[01:29:42] ================ drm_test_rect_calc_hscale ================
[01:29:42] [PASSED] normal use
[01:29:42] [PASSED] out of max range
[01:29:42] [PASSED] out of min range
[01:29:42] [PASSED] zero dst
[01:29:42] [PASSED] negative src
[01:29:42] [PASSED] negative dst
[01:29:42] ============ [PASSED] drm_test_rect_calc_hscale ============
[01:29:42] ================ drm_test_rect_calc_vscale ================
[01:29:42] [PASSED] normal use
[01:29:42] [PASSED] out of max range
[01:29:42] [PASSED] out of min range
[01:29:42] [PASSED] zero dst
[01:29:42] [PASSED] negative src
stty: 'standard input': Inappropriate ioctl for device
[01:29:42] [PASSED] negative dst
[01:29:42] ============ [PASSED] drm_test_rect_calc_vscale ============
[01:29:42] ================== drm_test_rect_rotate ===================
[01:29:42] [PASSED] reflect-x
[01:29:42] [PASSED] reflect-y
[01:29:42] [PASSED] rotate-0
[01:29:42] [PASSED] rotate-90
[01:29:42] [PASSED] rotate-180
[01:29:42] [PASSED] rotate-270
[01:29:42] ============== [PASSED] drm_test_rect_rotate ===============
[01:29:42] ================ drm_test_rect_rotate_inv =================
[01:29:42] [PASSED] reflect-x
[01:29:42] [PASSED] reflect-y
[01:29:42] [PASSED] rotate-0
[01:29:42] [PASSED] rotate-90
[01:29:42] [PASSED] rotate-180
[01:29:42] [PASSED] rotate-270
[01:29:42] ============ [PASSED] drm_test_rect_rotate_inv =============
[01:29:42] ==================== [PASSED] drm_rect =====================
[01:29:42] ============ drm_sysfb_modeset_test (1 subtest) ============
[01:29:42] ============ drm_test_sysfb_build_fourcc_list =============
[01:29:42] [PASSED] no native formats
[01:29:42] [PASSED] XRGB8888 as native format
[01:29:42] [PASSED] remove duplicates
[01:29:42] [PASSED] convert alpha formats
[01:29:42] [PASSED] random formats
[01:29:42] ======== [PASSED] drm_test_sysfb_build_fourcc_list =========
[01:29:42] ============= [PASSED] drm_sysfb_modeset_test ==============
[01:29:42] ============================================================
[01:29:42] Testing complete. Ran 621 tests: passed: 621
[01:29:43] Elapsed time: 25.531s total, 1.660s configuring, 23.704s building, 0.155s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig
[01:29:43] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[01:29:44] Building KUnit Kernel ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
Building with:
$ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48
[01:29:53] Starting KUnit Kernel (1/1)...
[01:29:53] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[01:29:54] ================= ttm_device (5 subtests) ==================
[01:29:54] [PASSED] ttm_device_init_basic
[01:29:54] [PASSED] ttm_device_init_multiple
[01:29:54] [PASSED] ttm_device_fini_basic
[01:29:54] [PASSED] ttm_device_init_no_vma_man
[01:29:54] ================== ttm_device_init_pools ==================
[01:29:54] [PASSED] No DMA allocations, no DMA32 required
[01:29:54] [PASSED] DMA allocations, DMA32 required
[01:29:54] [PASSED] No DMA allocations, DMA32 required
[01:29:54] [PASSED] DMA allocations, no DMA32 required
[01:29:54] ============== [PASSED] ttm_device_init_pools ==============
[01:29:54] =================== [PASSED] ttm_device ====================
[01:29:54] ================== ttm_pool (8 subtests) ===================
[01:29:54] ================== ttm_pool_alloc_basic ===================
[01:29:54] [PASSED] One page
[01:29:54] [PASSED] More than one page
[01:29:54] [PASSED] Above the allocation limit
[01:29:54] [PASSED] One page, with coherent DMA mappings enabled
[01:29:54] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[01:29:54] ============== [PASSED] ttm_pool_alloc_basic ===============
[01:29:54] ============== ttm_pool_alloc_basic_dma_addr ==============
[01:29:54] [PASSED] One page
[01:29:54] [PASSED] More than one page
[01:29:54] [PASSED] Above the allocation limit
[01:29:54] [PASSED] One page, with coherent DMA mappings enabled
[01:29:54] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[01:29:54] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ==========
[01:29:54] [PASSED] ttm_pool_alloc_order_caching_match
[01:29:54] [PASSED] ttm_pool_alloc_caching_mismatch
[01:29:54] [PASSED] ttm_pool_alloc_order_mismatch
[01:29:54] [PASSED] ttm_pool_free_dma_alloc
[01:29:54] [PASSED] ttm_pool_free_no_dma_alloc
[01:29:54] [PASSED] ttm_pool_fini_basic
[01:29:54] ==================== [PASSED] ttm_pool =====================
[01:29:54] ================ ttm_resource (8 subtests) =================
[01:29:54] ================= ttm_resource_init_basic =================
[01:29:54] [PASSED] Init resource in TTM_PL_SYSTEM
[01:29:54] [PASSED] Init resource in TTM_PL_VRAM
[01:29:54] [PASSED] Init resource in a private placement
[01:29:54] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags
[01:29:54] ============= [PASSED] ttm_resource_init_basic =============
[01:29:54] [PASSED] ttm_resource_init_pinned
[01:29:54] [PASSED] ttm_resource_fini_basic
[01:29:54] [PASSED] ttm_resource_manager_init_basic
[01:29:54] [PASSED] ttm_resource_manager_usage_basic
[01:29:54] [PASSED] ttm_resource_manager_set_used_basic
[01:29:54] [PASSED] ttm_sys_man_alloc_basic
[01:29:54] [PASSED] ttm_sys_man_free_basic
[01:29:54] ================== [PASSED] ttm_resource ===================
[01:29:54] =================== ttm_tt (15 subtests) ===================
[01:29:54] ==================== ttm_tt_init_basic ====================
[01:29:54] [PASSED] Page-aligned size
[01:29:54] [PASSED] Extra pages requested
[01:29:54] ================ [PASSED] ttm_tt_init_basic ================
[01:29:54] [PASSED] ttm_tt_init_misaligned
[01:29:54] [PASSED] ttm_tt_fini_basic
[01:29:54] [PASSED] ttm_tt_fini_sg
[01:29:54] [PASSED] ttm_tt_fini_shmem
[01:29:54] [PASSED] ttm_tt_create_basic
[01:29:54] [PASSED] ttm_tt_create_invalid_bo_type
[01:29:54] [PASSED] ttm_tt_create_ttm_exists
[01:29:54] [PASSED] ttm_tt_create_failed
[01:29:54] [PASSED] ttm_tt_destroy_basic
[01:29:54] [PASSED] ttm_tt_populate_null_ttm
[01:29:54] [PASSED] ttm_tt_populate_populated_ttm
[01:29:54] [PASSED] ttm_tt_unpopulate_basic
[01:29:54] [PASSED] ttm_tt_unpopulate_empty_ttm
[01:29:54] [PASSED] ttm_tt_swapin_basic
[01:29:54] ===================== [PASSED] ttm_tt ======================
[01:29:54] =================== ttm_bo (14 subtests) ===================
[01:29:54] =========== ttm_bo_reserve_optimistic_no_ticket ===========
[01:29:54] [PASSED] Cannot be interrupted and sleeps
[01:29:54] [PASSED] Cannot be interrupted, locks straight away
[01:29:54] [PASSED] Can be interrupted, sleeps
[01:29:54] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket =======
[01:29:54] [PASSED] ttm_bo_reserve_locked_no_sleep
[01:29:54] [PASSED] ttm_bo_reserve_no_wait_ticket
[01:29:54] [PASSED] ttm_bo_reserve_double_resv
[01:29:54] [PASSED] ttm_bo_reserve_interrupted
[01:29:54] [PASSED] ttm_bo_reserve_deadlock
[01:29:54] [PASSED] ttm_bo_unreserve_basic
[01:29:54] [PASSED] ttm_bo_unreserve_pinned
[01:29:54] [PASSED] ttm_bo_unreserve_bulk
[01:29:54] [PASSED] ttm_bo_fini_basic
[01:29:54] [PASSED] ttm_bo_fini_shared_resv
[01:29:54] [PASSED] ttm_bo_pin_basic
[01:29:54] [PASSED] ttm_bo_pin_unpin_resource
[01:29:54] [PASSED] ttm_bo_multiple_pin_one_unpin
[01:29:54] ===================== [PASSED] ttm_bo ======================
[01:29:54] ============== ttm_bo_validate (21 subtests) ===============
[01:29:54] ============== ttm_bo_init_reserved_sys_man ===============
[01:29:54] [PASSED] Buffer object for userspace
[01:29:54] [PASSED] Kernel buffer object
[01:29:54] [PASSED] Shared buffer object
[01:29:54] ========== [PASSED] ttm_bo_init_reserved_sys_man ===========
[01:29:54] ============== ttm_bo_init_reserved_mock_man ==============
[01:29:54] [PASSED] Buffer object for userspace
[01:29:54] [PASSED] Kernel buffer object
[01:29:54] [PASSED] Shared buffer object
[01:29:54] ========== [PASSED] ttm_bo_init_reserved_mock_man ==========
[01:29:54] [PASSED] ttm_bo_init_reserved_resv
[01:29:54] ================== ttm_bo_validate_basic ==================
[01:29:54] [PASSED] Buffer object for userspace
[01:29:54] [PASSED] Kernel buffer object
[01:29:54] [PASSED] Shared buffer object
[01:29:54] ============== [PASSED] ttm_bo_validate_basic ==============
[01:29:54] [PASSED] ttm_bo_validate_invalid_placement
[01:29:54] ============= ttm_bo_validate_same_placement ==============
[01:29:54] [PASSED] System manager
[01:29:54] [PASSED] VRAM manager
[01:29:54] ========= [PASSED] ttm_bo_validate_same_placement ==========
[01:29:54] [PASSED] ttm_bo_validate_failed_alloc
[01:29:54] [PASSED] ttm_bo_validate_pinned
[01:29:54] [PASSED] ttm_bo_validate_busy_placement
[01:29:54] ================ ttm_bo_validate_multihop =================
[01:29:54] [PASSED] Buffer object for userspace
[01:29:54] [PASSED] Kernel buffer object
[01:29:54] [PASSED] Shared buffer object
[01:29:54] ============ [PASSED] ttm_bo_validate_multihop =============
[01:29:54] ========== ttm_bo_validate_no_placement_signaled ==========
[01:29:54] [PASSED] Buffer object in system domain, no page vector
[01:29:54] [PASSED] Buffer object in system domain with an existing page vector
[01:29:54] ====== [PASSED] ttm_bo_validate_no_placement_signaled ======
[01:29:54] ======== ttm_bo_validate_no_placement_not_signaled ========
[01:29:54] [PASSED] Buffer object for userspace
[01:29:54] [PASSED] Kernel buffer object
[01:29:54] [PASSED] Shared buffer object
[01:29:54] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ====
[01:29:54] [PASSED] ttm_bo_validate_move_fence_signaled
[01:29:54] ========= ttm_bo_validate_move_fence_not_signaled =========
[01:29:54] [PASSED] Waits for GPU
[01:29:54] [PASSED] Tries to lock straight away
[01:29:54] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled =====
[01:29:54] [PASSED] ttm_bo_validate_happy_evict
[01:29:54] [PASSED] ttm_bo_validate_all_pinned_evict
[01:29:54] [PASSED] ttm_bo_validate_allowed_only_evict
[01:29:54] [PASSED] ttm_bo_validate_deleted_evict
[01:29:54] [PASSED] ttm_bo_validate_busy_domain_evict
[01:29:54] [PASSED] ttm_bo_validate_evict_gutting
[01:29:54] [PASSED] ttm_bo_validate_recrusive_evict
stty: 'standard input': Inappropriate ioctl for device
[01:29:54] ================= [PASSED] ttm_bo_validate =================
[01:29:54] ============================================================
[01:29:54] Testing complete. Ran 101 tests: passed: 101
[01:29:54] Elapsed time: 11.071s total, 1.734s configuring, 9.122s building, 0.188s running
+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel
^ permalink raw reply [flat|nested] 90+ messages in thread* ✗ Xe.CI.BAT: failure for VF migration redesign (rev2)
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (34 preceding siblings ...)
2025-09-24 1:29 ` ✓ CI.KUnit: success for VF migration redesign (rev2) Patchwork
@ 2025-09-24 2:14 ` Patchwork
2025-09-24 7:37 ` ✗ Xe.CI.Full: " Patchwork
36 siblings, 0 replies; 90+ messages in thread
From: Patchwork @ 2025-09-24 2:14 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 2453 bytes --]
== Series Details ==
Series: VF migration redesign (rev2)
URL : https://patchwork.freedesktop.org/series/154627/
State : failure
== Summary ==
CI Bug Log - changes from xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a_BAT -> xe-pw-154627v2_BAT
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with xe-pw-154627v2_BAT absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in xe-pw-154627v2_BAT, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
to document this new failure mode, which will reduce false positives in CI.
Participating hosts (11 -> 9)
------------------------------
Missing (2): bat-adlp-vm bat-ptl-vm
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in xe-pw-154627v2_BAT:
### IGT changes ###
#### Possible regressions ####
* igt@sriov_basic@enable-vfs-autoprobe-on:
- bat-adlp-7: [PASS][1] -> [ABORT][2] +1 other test abort
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/bat-adlp-7/igt@sriov_basic@enable-vfs-autoprobe-on.html
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/bat-adlp-7/igt@sriov_basic@enable-vfs-autoprobe-on.html
Known issues
------------
Here are the changes found in xe-pw-154627v2_BAT that come from known issues:
### IGT changes ###
#### Possible fixes ####
* igt@kms_flip@basic-flip-vs-dpms:
- bat-adlp-7: [DMESG-WARN][3] ([Intel XE#4543]) -> [PASS][4] +1 other test pass
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/bat-adlp-7/igt@kms_flip@basic-flip-vs-dpms.html
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/bat-adlp-7/igt@kms_flip@basic-flip-vs-dpms.html
[Intel XE#4543]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4543
Build changes
-------------
* Linux: xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a -> xe-pw-154627v2
IGT_8550: 4f8c7886ad02e116804ec08714f17bce1755c6e4 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a: af3cdefd0a1ad2dad29fcde0854ccbce494cc28a
xe-pw-154627v2: 154627v2
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/index.html
[-- Attachment #2: Type: text/html, Size: 3059 bytes --]
^ permalink raw reply [flat|nested] 90+ messages in thread* ✗ Xe.CI.Full: failure for VF migration redesign (rev2)
2025-09-24 1:15 [PATCH v2 00/34] VF migration redesign Matthew Brost
` (35 preceding siblings ...)
2025-09-24 2:14 ` ✗ Xe.CI.BAT: failure " Patchwork
@ 2025-09-24 7:37 ` Patchwork
36 siblings, 0 replies; 90+ messages in thread
From: Patchwork @ 2025-09-24 7:37 UTC (permalink / raw)
To: Matthew Brost; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 75086 bytes --]
== Series Details ==
Series: VF migration redesign (rev2)
URL : https://patchwork.freedesktop.org/series/154627/
State : failure
== Summary ==
CI Bug Log - changes from xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a_FULL -> xe-pw-154627v2_FULL
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with xe-pw-154627v2_FULL absolutely need to be
verified manually.
If you think the reported changes have nothing to do with the changes
introduced in xe-pw-154627v2_FULL, please notify your bug team (I915-ci-infra@lists.freedesktop.org) to allow them
to document this new failure mode, which will reduce false positives in CI.
Participating hosts (4 -> 4)
------------------------------
No changes in participating hosts
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in xe-pw-154627v2_FULL:
### IGT changes ###
#### Possible regressions ####
* igt@xe_exec_compute_mode@non-blocking:
- shard-bmg: [PASS][1] -> [INCOMPLETE][2]
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_exec_compute_mode@non-blocking.html
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-2/igt@xe_exec_compute_mode@non-blocking.html
- shard-dg2-set2: [PASS][3] -> [INCOMPLETE][4]
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-434/igt@xe_exec_compute_mode@non-blocking.html
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@xe_exec_compute_mode@non-blocking.html
- shard-lnl: [PASS][5] -> [FAIL][6]
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-lnl-5/igt@xe_exec_compute_mode@non-blocking.html
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-4/igt@xe_exec_compute_mode@non-blocking.html
* igt@xe_exec_reset@cm-gt-reset:
- shard-adlp: [PASS][7] -> [FAIL][8] +1 other test fail
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-6/igt@xe_exec_reset@cm-gt-reset.html
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@xe_exec_reset@cm-gt-reset.html
* igt@xe_live_ktest@xe_bo:
- shard-bmg: NOTRUN -> [ABORT][9] +1 other test abort
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@xe_live_ktest@xe_bo.html
* igt@xe_pmu@all-fn-engine-activity-load:
- shard-adlp: NOTRUN -> [ABORT][10]
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_pmu@all-fn-engine-activity-load.html
* igt@xe_pmu@engine-activity-most-load-idle:
- shard-adlp: [PASS][11] -> [ABORT][12] +19 other tests abort
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-3/igt@xe_pmu@engine-activity-most-load-idle.html
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-8/igt@xe_pmu@engine-activity-most-load-idle.html
#### Suppressed ####
The following results come from untrusted machines, tests, or statuses.
They do not affect the overall result.
* {igt@kms_pipe_stress@stress-xrgb8888-4tiled}:
- shard-bmg: [TIMEOUT][13] -> [FAIL][14]
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-7/igt@kms_pipe_stress@stress-xrgb8888-4tiled.html
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_pipe_stress@stress-xrgb8888-4tiled.html
* {igt@xe_pmu@engine-activity-suspend}:
- shard-adlp: [PASS][15] -> [ABORT][16] +11 other tests abort
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-3/igt@xe_pmu@engine-activity-suspend.html
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_pmu@engine-activity-suspend.html
Known issues
------------
Here are the changes found in xe-pw-154627v2_FULL that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@kms_big_fb@4-tiled-64bpp-rotate-90:
- shard-bmg: NOTRUN -> [SKIP][17] ([Intel XE#2327]) +1 other test skip
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_big_fb@4-tiled-64bpp-rotate-90.html
* igt@kms_big_fb@linear-32bpp-rotate-270:
- shard-dg2-set2: NOTRUN -> [SKIP][18] ([Intel XE#316])
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_big_fb@linear-32bpp-rotate-270.html
* igt@kms_big_fb@y-tiled-16bpp-rotate-0:
- shard-bmg: NOTRUN -> [SKIP][19] ([Intel XE#1124]) +3 other tests skip
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_big_fb@y-tiled-16bpp-rotate-0.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip:
- shard-adlp: NOTRUN -> [DMESG-FAIL][20] ([Intel XE#4543])
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip:
- shard-dg2-set2: NOTRUN -> [SKIP][21] ([Intel XE#1124]) +8 other tests skip
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip.html
* igt@kms_big_fb@yf-tiled-8bpp-rotate-270:
- shard-adlp: NOTRUN -> [SKIP][22] ([Intel XE#1124])
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_big_fb@yf-tiled-8bpp-rotate-270.html
* igt@kms_big_fb@yf-tiled-addfb-size-offset-overflow:
- shard-dg2-set2: NOTRUN -> [SKIP][23] ([Intel XE#607]) +1 other test skip
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_big_fb@yf-tiled-addfb-size-offset-overflow.html
* igt@kms_big_fb@yf-tiled-addfb-size-overflow:
- shard-bmg: NOTRUN -> [SKIP][24] ([Intel XE#610])
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html
* igt@kms_bw@connected-linear-tiling-3-displays-3840x2160p:
- shard-dg2-set2: NOTRUN -> [SKIP][25] ([Intel XE#2191])
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@kms_bw@connected-linear-tiling-3-displays-3840x2160p.html
- shard-lnl: NOTRUN -> [SKIP][26] ([Intel XE#2191])
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_bw@connected-linear-tiling-3-displays-3840x2160p.html
* igt@kms_bw@connected-linear-tiling-4-displays-2160x1440p:
- shard-bmg: NOTRUN -> [SKIP][27] ([Intel XE#2314] / [Intel XE#2894])
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_bw@connected-linear-tiling-4-displays-2160x1440p.html
* igt@kms_bw@linear-tiling-4-displays-3840x2160p:
- shard-dg2-set2: NOTRUN -> [SKIP][28] ([Intel XE#367])
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_bw@linear-tiling-4-displays-3840x2160p.html
* igt@kms_ccs@bad-aux-stride-y-tiled-gen12-rc-ccs-cc@pipe-c-hdmi-a-1:
- shard-adlp: NOTRUN -> [SKIP][29] ([Intel XE#787]) +5 other tests skip
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_ccs@bad-aux-stride-y-tiled-gen12-rc-ccs-cc@pipe-c-hdmi-a-1.html
* igt@kms_ccs@bad-aux-stride-y-tiled-gen12-rc-ccs-cc@pipe-d-hdmi-a-1:
- shard-adlp: NOTRUN -> [SKIP][30] ([Intel XE#455] / [Intel XE#787]) +3 other tests skip
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_ccs@bad-aux-stride-y-tiled-gen12-rc-ccs-cc@pipe-d-hdmi-a-1.html
* igt@kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc@pipe-b-dp-2:
- shard-dg2-set2: NOTRUN -> [SKIP][31] ([Intel XE#787]) +160 other tests skip
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc@pipe-b-dp-2.html
* igt@kms_ccs@crc-primary-rotation-180-y-tiled-gen12-rc-ccs-cc:
- shard-bmg: NOTRUN -> [SKIP][32] ([Intel XE#2887]) +8 other tests skip
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_ccs@crc-primary-rotation-180-y-tiled-gen12-rc-ccs-cc.html
* igt@kms_ccs@crc-sprite-planes-basic-4-tiled-bmg-ccs:
- shard-dg2-set2: NOTRUN -> [SKIP][33] ([Intel XE#2907]) +1 other test skip
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_ccs@crc-sprite-planes-basic-4-tiled-bmg-ccs.html
* igt@kms_ccs@crc-sprite-planes-basic-4-tiled-mtl-rc-ccs@pipe-d-dp-2:
- shard-dg2-set2: NOTRUN -> [SKIP][34] ([Intel XE#455] / [Intel XE#787]) +33 other tests skip
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_ccs@crc-sprite-planes-basic-4-tiled-mtl-rc-ccs@pipe-d-dp-2.html
* igt@kms_ccs@random-ccs-data-4-tiled-lnl-ccs@pipe-d-hdmi-a-3:
- shard-bmg: NOTRUN -> [SKIP][35] ([Intel XE#2652] / [Intel XE#787]) +8 other tests skip
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_ccs@random-ccs-data-4-tiled-lnl-ccs@pipe-d-hdmi-a-3.html
* igt@kms_chamelium_color@ctm-0-50:
- shard-dg2-set2: NOTRUN -> [SKIP][36] ([Intel XE#306])
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_chamelium_color@ctm-0-50.html
* igt@kms_chamelium_edid@dp-edid-change-during-hibernate:
- shard-bmg: NOTRUN -> [SKIP][37] ([Intel XE#2252]) +7 other tests skip
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_chamelium_edid@dp-edid-change-during-hibernate.html
* igt@kms_chamelium_hpd@vga-hpd:
- shard-dg2-set2: NOTRUN -> [SKIP][38] ([Intel XE#373]) +7 other tests skip
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_chamelium_hpd@vga-hpd.html
* igt@kms_content_protection@dp-mst-type-1:
- shard-dg2-set2: NOTRUN -> [SKIP][39] ([Intel XE#307]) +1 other test skip
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_content_protection@dp-mst-type-1.html
* igt@kms_content_protection@srm@pipe-a-dp-2:
- shard-bmg: NOTRUN -> [FAIL][40] ([Intel XE#1178])
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_content_protection@srm@pipe-a-dp-2.html
* igt@kms_cursor_crc@cursor-offscreen-512x512:
- shard-dg2-set2: NOTRUN -> [SKIP][41] ([Intel XE#308]) +1 other test skip
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_cursor_crc@cursor-offscreen-512x512.html
* igt@kms_cursor_crc@cursor-onscreen-512x170:
- shard-bmg: NOTRUN -> [SKIP][42] ([Intel XE#2321])
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_cursor_crc@cursor-onscreen-512x170.html
* igt@kms_cursor_crc@cursor-rapid-movement-32x10:
- shard-bmg: NOTRUN -> [SKIP][43] ([Intel XE#2320]) +1 other test skip
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_cursor_crc@cursor-rapid-movement-32x10.html
* igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions-varying-size:
- shard-bmg: [PASS][44] -> [SKIP][45] ([Intel XE#2291]) +3 other tests skip
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions-varying-size.html
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions-varying-size.html
* igt@kms_cursor_legacy@short-busy-flip-before-cursor-toggle:
- shard-bmg: NOTRUN -> [SKIP][46] ([Intel XE#2286])
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_cursor_legacy@short-busy-flip-before-cursor-toggle.html
* igt@kms_dirtyfb@fbc-dirtyfb-ioctl:
- shard-bmg: NOTRUN -> [SKIP][47] ([Intel XE#5428])
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_dirtyfb@fbc-dirtyfb-ioctl.html
* igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-6:
- shard-dg2-set2: NOTRUN -> [SKIP][48] ([Intel XE#4494] / [i915#3804])
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-6.html
* igt@kms_dp_aux_dev:
- shard-bmg: [PASS][49] -> [SKIP][50] ([Intel XE#3009])
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_dp_aux_dev.html
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_dp_aux_dev.html
* igt@kms_dp_link_training@uhbr-mst:
- shard-dg2-set2: NOTRUN -> [SKIP][51] ([Intel XE#4356])
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_dp_link_training@uhbr-mst.html
* igt@kms_dp_linktrain_fallback@dsc-fallback:
- shard-dg2-set2: NOTRUN -> [SKIP][52] ([Intel XE#4331])
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_dp_linktrain_fallback@dsc-fallback.html
* igt@kms_dsc@dsc-fractional-bpp-with-bpc:
- shard-bmg: NOTRUN -> [SKIP][53] ([Intel XE#2244])
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_dsc@dsc-fractional-bpp-with-bpc.html
* igt@kms_fbcon_fbt@psr:
- shard-bmg: NOTRUN -> [SKIP][54] ([Intel XE#776])
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_fbcon_fbt@psr.html
* igt@kms_fbcon_fbt@psr-suspend:
- shard-dg2-set2: NOTRUN -> [SKIP][55] ([Intel XE#776]) +1 other test skip
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_fbcon_fbt@psr-suspend.html
* igt@kms_feature_discovery@dp-mst:
- shard-bmg: NOTRUN -> [SKIP][56] ([Intel XE#2375])
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_feature_discovery@dp-mst.html
* igt@kms_feature_discovery@psr2:
- shard-dg2-set2: NOTRUN -> [SKIP][57] ([Intel XE#1135])
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_feature_discovery@psr2.html
* igt@kms_flip@2x-dpms-vs-vblank-race:
- shard-adlp: NOTRUN -> [SKIP][58] ([Intel XE#310])
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_flip@2x-dpms-vs-vblank-race.html
* igt@kms_flip@2x-plain-flip-ts-check-interruptible:
- shard-bmg: [PASS][59] -> [SKIP][60] ([Intel XE#2316]) +4 other tests skip
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_flip@2x-plain-flip-ts-check-interruptible.html
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_flip@2x-plain-flip-ts-check-interruptible.html
* igt@kms_flip@flip-vs-dpms-on-nop@b-hdmi-a1:
- shard-adlp: [PASS][61] -> [DMESG-WARN][62] ([Intel XE#4543]) +6 other tests dmesg-warn
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@kms_flip@flip-vs-dpms-on-nop@b-hdmi-a1.html
[62]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@kms_flip@flip-vs-dpms-on-nop@b-hdmi-a1.html
* igt@kms_flip@flip-vs-expired-vblank@b-edp1:
- shard-lnl: [PASS][63] -> [FAIL][64] ([Intel XE#301]) +1 other test fail
[63]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-lnl-4/igt@kms_flip@flip-vs-expired-vblank@b-edp1.html
[64]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_flip@flip-vs-expired-vblank@b-edp1.html
* igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling:
- shard-bmg: NOTRUN -> [SKIP][65] ([Intel XE#2293] / [Intel XE#2380])
[65]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling.html
* igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling@pipe-a-valid-mode:
- shard-bmg: NOTRUN -> [SKIP][66] ([Intel XE#2293])
[66]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling@pipe-a-valid-mode.html
* igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling:
- shard-lnl: NOTRUN -> [SKIP][67] ([Intel XE#1397] / [Intel XE#1745])
[67]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling.html
* igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling@pipe-a-default-mode:
- shard-lnl: NOTRUN -> [SKIP][68] ([Intel XE#1397])
[68]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling@pipe-a-default-mode.html
* igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling:
- shard-dg2-set2: NOTRUN -> [SKIP][69] ([Intel XE#455]) +7 other tests skip
[69]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling.html
* igt@kms_flip_tiling@flip-change-tiling@pipe-b-hdmi-a-1-x-to-y:
- shard-adlp: [PASS][70] -> [FAIL][71] ([Intel XE#1874])
[70]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-b-hdmi-a-1-x-to-y.html
[71]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-b-hdmi-a-1-x-to-y.html
* igt@kms_flip_tiling@flip-change-tiling@pipe-b-hdmi-a-1-y-to-y:
- shard-adlp: [PASS][72] -> [DMESG-FAIL][73] ([Intel XE#4543])
[72]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-b-hdmi-a-1-y-to-y.html
[73]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-b-hdmi-a-1-y-to-y.html
* igt@kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-blt:
- shard-bmg: NOTRUN -> [SKIP][74] ([Intel XE#2311]) +15 other tests skip
[74]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-blt.html
* igt@kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-onoff:
- shard-dg2-set2: NOTRUN -> [SKIP][75] ([Intel XE#651]) +25 other tests skip
[75]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-onoff.html
* igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render:
- shard-adlp: [PASS][76] -> [DMESG-WARN][77] ([Intel XE#2953] / [Intel XE#4173])
[76]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-9/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render.html
[77]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc:
- shard-bmg: NOTRUN -> [SKIP][78] ([Intel XE#5390]) +4 other tests skip
[78]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-rte:
- shard-bmg: NOTRUN -> [SKIP][79] ([Intel XE#5427])
[79]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_frontbuffer_tracking@fbc-2p-rte.html
* igt@kms_frontbuffer_tracking@fbc-tiling-y:
- shard-dg2-set2: NOTRUN -> [SKIP][80] ([Intel XE#658]) +1 other test skip
[80]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@kms_frontbuffer_tracking@fbc-tiling-y.html
- shard-bmg: NOTRUN -> [SKIP][81] ([Intel XE#2352])
[81]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_frontbuffer_tracking@fbc-tiling-y.html
* igt@kms_frontbuffer_tracking@fbcdrrs-slowdraw:
- shard-adlp: NOTRUN -> [SKIP][82] ([Intel XE#651])
[82]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_frontbuffer_tracking@fbcdrrs-slowdraw.html
* igt@kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-render:
- shard-adlp: NOTRUN -> [SKIP][83] ([Intel XE#653]) +2 other tests skip
[83]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-render.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-render:
- shard-adlp: NOTRUN -> [SKIP][84] ([Intel XE#656]) +1 other test skip
[84]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@psr-2p-primscrn-indfb-plflip-blt:
- shard-bmg: NOTRUN -> [SKIP][85] ([Intel XE#2313]) +10 other tests skip
[85]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_frontbuffer_tracking@psr-2p-primscrn-indfb-plflip-blt.html
* igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-blt:
- shard-lnl: NOTRUN -> [SKIP][86] ([Intel XE#656])
[86]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-blt.html
* igt@kms_frontbuffer_tracking@psr-slowdraw:
- shard-dg2-set2: NOTRUN -> [SKIP][87] ([Intel XE#653]) +27 other tests skip
[87]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_frontbuffer_tracking@psr-slowdraw.html
* igt@kms_hdr@bpc-switch-suspend:
- shard-adlp: NOTRUN -> [SKIP][88] ([Intel XE#455])
[88]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_hdr@bpc-switch-suspend.html
* igt@kms_plane_cursor@overlay@pipe-a-hdmi-a-6-size-64:
- shard-dg2-set2: NOTRUN -> [FAIL][89] ([Intel XE#616]) +1 other test fail
[89]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_plane_cursor@overlay@pipe-a-hdmi-a-6-size-64.html
* igt@kms_plane_multiple@2x-tiling-none:
- shard-bmg: [PASS][90] -> [SKIP][91] ([Intel XE#4596])
[90]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-8/igt@kms_plane_multiple@2x-tiling-none.html
[91]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_plane_multiple@2x-tiling-none.html
* igt@kms_plane_multiple@tiling-y:
- shard-dg2-set2: NOTRUN -> [SKIP][92] ([Intel XE#5020])
[92]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_plane_multiple@tiling-y.html
* igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5@pipe-c:
- shard-lnl: NOTRUN -> [SKIP][93] ([Intel XE#2763]) +3 other tests skip
[93]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5@pipe-c.html
* igt@kms_pm_backlight@fade-with-suspend:
- shard-dg2-set2: NOTRUN -> [SKIP][94] ([Intel XE#870])
[94]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_pm_backlight@fade-with-suspend.html
* igt@kms_pm_dc@dc6-psr:
- shard-bmg: NOTRUN -> [SKIP][95] ([Intel XE#2392])
[95]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_pm_dc@dc6-psr.html
* igt@kms_pm_dc@dc9-dpms:
- shard-adlp: NOTRUN -> [SKIP][96] ([Intel XE#734])
[96]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_pm_dc@dc9-dpms.html
* igt@kms_psr2_sf@fbc-pr-cursor-plane-move-continuous-sf:
- shard-bmg: NOTRUN -> [SKIP][97] ([Intel XE#1406] / [Intel XE#1489]) +7 other tests skip
[97]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_psr2_sf@fbc-pr-cursor-plane-move-continuous-sf.html
* igt@kms_psr2_sf@fbc-psr2-overlay-plane-move-continuous-exceed-fully-sf:
- shard-adlp: NOTRUN -> [SKIP][98] ([Intel XE#1406] / [Intel XE#1489])
[98]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_psr2_sf@fbc-psr2-overlay-plane-move-continuous-exceed-fully-sf.html
* igt@kms_psr2_sf@pr-overlay-plane-move-continuous-exceed-fully-sf:
- shard-dg2-set2: NOTRUN -> [SKIP][99] ([Intel XE#1406] / [Intel XE#1489]) +7 other tests skip
[99]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_psr2_sf@pr-overlay-plane-move-continuous-exceed-fully-sf.html
* igt@kms_psr2_su@page_flip-p010:
- shard-dg2-set2: NOTRUN -> [SKIP][100] ([Intel XE#1122] / [Intel XE#1406])
[100]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_psr2_su@page_flip-p010.html
* igt@kms_psr@fbc-pr-sprite-plane-onoff:
- shard-adlp: NOTRUN -> [SKIP][101] ([Intel XE#1406] / [Intel XE#2850] / [Intel XE#929]) +1 other test skip
[101]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_psr@fbc-pr-sprite-plane-onoff.html
* igt@kms_psr@fbc-psr2-suspend:
- shard-bmg: NOTRUN -> [SKIP][102] ([Intel XE#1406] / [Intel XE#2234] / [Intel XE#2850]) +2 other tests skip
[102]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_psr@fbc-psr2-suspend.html
* igt@kms_psr@pr-primary-page-flip:
- shard-dg2-set2: NOTRUN -> [SKIP][103] ([Intel XE#1406] / [Intel XE#2850] / [Intel XE#929]) +8 other tests skip
[103]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_psr@pr-primary-page-flip.html
* igt@kms_rotation_crc@primary-y-tiled-reflect-x-0:
- shard-bmg: NOTRUN -> [SKIP][104] ([Intel XE#2330])
[104]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_rotation_crc@primary-y-tiled-reflect-x-0.html
* igt@kms_rotation_crc@primary-yf-tiled-reflect-x-180:
- shard-adlp: NOTRUN -> [SKIP][105] ([Intel XE#1127])
[105]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_rotation_crc@primary-yf-tiled-reflect-x-180.html
* igt@kms_rotation_crc@primary-yf-tiled-reflect-x-270:
- shard-bmg: NOTRUN -> [SKIP][106] ([Intel XE#3414] / [Intel XE#3904])
[106]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_rotation_crc@primary-yf-tiled-reflect-x-270.html
* igt@kms_rotation_crc@sprite-rotation-90-pos-100-0:
- shard-dg2-set2: NOTRUN -> [SKIP][107] ([Intel XE#3414])
[107]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@kms_rotation_crc@sprite-rotation-90-pos-100-0.html
* igt@kms_scaling_modes@scaling-mode-full-aspect:
- shard-bmg: NOTRUN -> [SKIP][108] ([Intel XE#2413])
[108]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_scaling_modes@scaling-mode-full-aspect.html
* igt@kms_setmode@invalid-clone-single-crtc:
- shard-bmg: [PASS][109] -> [SKIP][110] ([Intel XE#1435])
[109]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-7/igt@kms_setmode@invalid-clone-single-crtc.html
[110]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_setmode@invalid-clone-single-crtc.html
* igt@kms_vrr@flip-suspend:
- shard-bmg: NOTRUN -> [SKIP][111] ([Intel XE#1499])
[111]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_vrr@flip-suspend.html
* igt@kms_vrr@lobf:
- shard-dg2-set2: NOTRUN -> [SKIP][112] ([Intel XE#2168])
[112]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@kms_vrr@lobf.html
* igt@xe_compute_preempt@compute-preempt-many-all-ram@engine-drm_xe_engine_class_compute:
- shard-dg2-set2: NOTRUN -> [FAIL][113] ([Intel XE#5890]) +1 other test fail
[113]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_compute_preempt@compute-preempt-many-all-ram@engine-drm_xe_engine_class_compute.html
* igt@xe_copy_basic@mem-set-linear-0xfffe:
- shard-dg2-set2: NOTRUN -> [SKIP][114] ([Intel XE#1126])
[114]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_copy_basic@mem-set-linear-0xfffe.html
* igt@xe_eudebug@basic-vm-access-parameters:
- shard-dg2-set2: NOTRUN -> [SKIP][115] ([Intel XE#4837]) +13 other tests skip
[115]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_eudebug@basic-vm-access-parameters.html
* igt@xe_eudebug_online@single-step:
- shard-bmg: NOTRUN -> [SKIP][116] ([Intel XE#4837]) +4 other tests skip
[116]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_eudebug_online@single-step.html
* igt@xe_eudebug_online@stopped-thread:
- shard-adlp: NOTRUN -> [SKIP][117] ([Intel XE#4837] / [Intel XE#5565]) +1 other test skip
[117]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_eudebug_online@stopped-thread.html
* igt@xe_eudebug_online@writes-caching-sram-bb-sram-target-vram:
- shard-lnl: NOTRUN -> [SKIP][118] ([Intel XE#4837])
[118]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@xe_eudebug_online@writes-caching-sram-bb-sram-target-vram.html
* igt@xe_exec_basic@multigpu-no-exec-null-defer-mmap:
- shard-dg2-set2: [PASS][119] -> [SKIP][120] ([Intel XE#1392]) +5 other tests skip
[119]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-436/igt@xe_exec_basic@multigpu-no-exec-null-defer-mmap.html
[120]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_exec_basic@multigpu-no-exec-null-defer-mmap.html
* igt@xe_exec_basic@multigpu-no-exec-null-rebind:
- shard-bmg: NOTRUN -> [SKIP][121] ([Intel XE#2322]) +3 other tests skip
[121]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@xe_exec_basic@multigpu-no-exec-null-rebind.html
* igt@xe_exec_basic@multigpu-once-bindexecqueue-userptr:
- shard-dg2-set2: NOTRUN -> [SKIP][122] ([Intel XE#1392])
[122]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_exec_basic@multigpu-once-bindexecqueue-userptr.html
* igt@xe_exec_basic@multigpu-once-userptr-invalidate-race:
- shard-adlp: NOTRUN -> [SKIP][123] ([Intel XE#1392] / [Intel XE#5575]) +1 other test skip
[123]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_exec_basic@multigpu-once-userptr-invalidate-race.html
* igt@xe_exec_fault_mode@many-execqueues-userptr-invalidate-race-imm:
- shard-dg2-set2: NOTRUN -> [SKIP][124] ([Intel XE#288]) +22 other tests skip
[124]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@xe_exec_fault_mode@many-execqueues-userptr-invalidate-race-imm.html
* igt@xe_exec_fault_mode@twice-bindexecqueue-userptr-invalidate:
- shard-adlp: NOTRUN -> [SKIP][125] ([Intel XE#288] / [Intel XE#5561]) +1 other test skip
[125]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_exec_fault_mode@twice-bindexecqueue-userptr-invalidate.html
* igt@xe_exec_mix_modes@exec-simple-batch-store-lr:
- shard-dg2-set2: NOTRUN -> [SKIP][126] ([Intel XE#2360]) +1 other test skip
[126]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@xe_exec_mix_modes@exec-simple-batch-store-lr.html
* igt@xe_exec_system_allocator@many-execqueues-mmap-new-huge:
- shard-dg2-set2: NOTRUN -> [SKIP][127] ([Intel XE#4915]) +214 other tests skip
[127]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_exec_system_allocator@many-execqueues-mmap-new-huge.html
* igt@xe_exec_system_allocator@process-many-execqueues-mmap-free-huge:
- shard-bmg: NOTRUN -> [SKIP][128] ([Intel XE#4943]) +5 other tests skip
[128]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@xe_exec_system_allocator@process-many-execqueues-mmap-free-huge.html
* igt@xe_exec_system_allocator@threads-many-execqueues-mmap-free-huge:
- shard-lnl: NOTRUN -> [SKIP][129] ([Intel XE#4943]) +1 other test skip
[129]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@xe_exec_system_allocator@threads-many-execqueues-mmap-free-huge.html
* igt@xe_exec_system_allocator@twice-large-new-race:
- shard-adlp: NOTRUN -> [SKIP][130] ([Intel XE#4915]) +26 other tests skip
[130]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_exec_system_allocator@twice-large-new-race.html
* igt@xe_live_ktest@xe_bo@xe_ccs_migrate_kunit:
- shard-bmg: NOTRUN -> [SKIP][131] ([Intel XE#2229])
[131]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@xe_live_ktest@xe_bo@xe_ccs_migrate_kunit.html
* igt@xe_oa@buffer-fill:
- shard-adlp: NOTRUN -> [SKIP][132] ([Intel XE#3573]) +1 other test skip
[132]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_oa@buffer-fill.html
* igt@xe_oa@closed-fd-and-unmapped-access:
- shard-dg2-set2: NOTRUN -> [SKIP][133] ([Intel XE#3573]) +6 other tests skip
[133]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_oa@closed-fd-and-unmapped-access.html
* igt@xe_pm@d3cold-multiple-execs:
- shard-bmg: NOTRUN -> [SKIP][134] ([Intel XE#2284]) +1 other test skip
[134]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_pm@d3cold-multiple-execs.html
* igt@xe_pm@s2idle-d3hot-basic-exec:
- shard-adlp: [PASS][135] -> [DMESG-WARN][136] ([Intel XE#2953] / [Intel XE#4173] / [Intel XE#4504])
[135]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-9/igt@xe_pm@s2idle-d3hot-basic-exec.html
[136]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@xe_pm@s2idle-d3hot-basic-exec.html
* igt@xe_pm@s4-d3cold-basic-exec:
- shard-dg2-set2: NOTRUN -> [SKIP][137] ([Intel XE#2284] / [Intel XE#366])
[137]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@xe_pm@s4-d3cold-basic-exec.html
* igt@xe_pmu@fn-engine-activity-load:
- shard-dg2-set2: NOTRUN -> [SKIP][138] ([Intel XE#4650])
[138]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@xe_pmu@fn-engine-activity-load.html
* igt@xe_pxp@display-pxp-fb:
- shard-dg2-set2: NOTRUN -> [SKIP][139] ([Intel XE#4733]) +1 other test skip
[139]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_pxp@display-pxp-fb.html
* igt@xe_pxp@pxp-termination-key-update-post-rpm:
- shard-adlp: NOTRUN -> [SKIP][140] ([Intel XE#4733] / [Intel XE#5594])
[140]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_pxp@pxp-termination-key-update-post-rpm.html
* igt@xe_query@multigpu-query-invalid-size:
- shard-bmg: NOTRUN -> [SKIP][141] ([Intel XE#944])
[141]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_query@multigpu-query-invalid-size.html
* igt@xe_sriov_auto_provisioning@selfconfig-reprovision-reduce-numvfs:
- shard-bmg: NOTRUN -> [DMESG-WARN][142] ([Intel XE#6190]) +3 other tests dmesg-warn
[142]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@xe_sriov_auto_provisioning@selfconfig-reprovision-reduce-numvfs.html
- shard-dg2-set2: NOTRUN -> [SKIP][143] ([Intel XE#4130]) +1 other test skip
[143]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@xe_sriov_auto_provisioning@selfconfig-reprovision-reduce-numvfs.html
#### Possible fixes ####
* igt@kms_async_flips@alternate-sync-async-flip:
- shard-bmg: [FAIL][144] ([Intel XE#3718] / [Intel XE#6078]) -> [PASS][145]
[144]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_async_flips@alternate-sync-async-flip.html
[145]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_async_flips@alternate-sync-async-flip.html
* igt@kms_async_flips@alternate-sync-async-flip@pipe-a-dp-2:
- shard-bmg: [FAIL][146] -> [PASS][147]
[146]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_async_flips@alternate-sync-async-flip@pipe-a-dp-2.html
[147]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_async_flips@alternate-sync-async-flip@pipe-a-dp-2.html
* igt@kms_async_flips@async-flip-suspend-resume@pipe-c-hdmi-a-1:
- shard-adlp: [DMESG-WARN][148] ([Intel XE#4543]) -> [PASS][149] +2 other tests pass
[148]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-6/igt@kms_async_flips@async-flip-suspend-resume@pipe-c-hdmi-a-1.html
[149]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@kms_async_flips@async-flip-suspend-resume@pipe-c-hdmi-a-1.html
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs:
- shard-dg2-set2: [INCOMPLETE][150] ([Intel XE#1727] / [Intel XE#3113] / [Intel XE#4345]) -> [PASS][151]
[150]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-466/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs.html
[151]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs.html
* igt@kms_cursor_crc@cursor-suspend@pipe-a-hdmi-a-1:
- shard-adlp: [DMESG-WARN][152] ([Intel XE#2953] / [Intel XE#4173]) -> [PASS][153] +1 other test pass
[152]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-3/igt@kms_cursor_crc@cursor-suspend@pipe-a-hdmi-a-1.html
[153]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@kms_cursor_crc@cursor-suspend@pipe-a-hdmi-a-1.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
- shard-bmg: [DMESG-WARN][154] ([Intel XE#5354]) -> [PASS][155]
[154]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
[155]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
* igt@kms_flip@2x-wf_vblank-ts-check:
- shard-bmg: [SKIP][156] ([Intel XE#2316]) -> [PASS][157] +1 other test pass
[156]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-6/igt@kms_flip@2x-wf_vblank-ts-check.html
[157]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_flip@2x-wf_vblank-ts-check.html
* igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1:
- shard-lnl: [FAIL][158] ([Intel XE#301]) -> [PASS][159] +1 other test pass
[158]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-lnl-2/igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1.html
[159]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@kms_flip@flip-vs-expired-vblank-interruptible@b-edp1.html
* igt@kms_flip_tiling@flip-change-tiling@pipe-c-hdmi-a-1-x-to-x:
- shard-adlp: [DMESG-FAIL][160] ([Intel XE#4543]) -> [PASS][161]
[160]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-c-hdmi-a-1-x-to-x.html
[161]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-c-hdmi-a-1-x-to-x.html
* igt@kms_flip_tiling@flip-change-tiling@pipe-d-hdmi-a-1-y-to-x:
- shard-adlp: [FAIL][162] ([Intel XE#1874]) -> [PASS][163]
[162]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-d-hdmi-a-1-y-to-x.html
[163]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@kms_flip_tiling@flip-change-tiling@pipe-d-hdmi-a-1-y-to-x.html
* igt@kms_hdr@static-toggle-suspend:
- shard-bmg: [SKIP][164] ([Intel XE#1503]) -> [PASS][165] +1 other test pass
[164]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-6/igt@kms_hdr@static-toggle-suspend.html
[165]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_hdr@static-toggle-suspend.html
* igt@kms_plane_multiple@tiling-none:
- shard-bmg: [TIMEOUT][166] -> [PASS][167] +1 other test pass
[166]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@kms_plane_multiple@tiling-none.html
[167]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@kms_plane_multiple@tiling-none.html
* igt@xe_exec_basic@multigpu-many-execqueues-many-vm-userptr-invalidate-race:
- shard-dg2-set2: [SKIP][168] ([Intel XE#1392]) -> [PASS][169] +5 other tests pass
[168]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-432/igt@xe_exec_basic@multigpu-many-execqueues-many-vm-userptr-invalidate-race.html
[169]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-436/igt@xe_exec_basic@multigpu-many-execqueues-many-vm-userptr-invalidate-race.html
* igt@xe_exec_compute_mode@many-execqueues-userptr-free:
- shard-bmg: [FAIL][170] ([Intel XE#5625]) -> [PASS][171]
[170]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_exec_compute_mode@many-execqueues-userptr-free.html
[171]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_exec_compute_mode@many-execqueues-userptr-free.html
* igt@xe_exec_fault_mode@many-execqueues-userptr-imm:
- shard-bmg: [FAIL][172] ([Intel XE#6050]) -> [PASS][173]
[172]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_exec_fault_mode@many-execqueues-userptr-imm.html
[173]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_exec_fault_mode@many-execqueues-userptr-imm.html
* igt@xe_exec_reset@parallel-gt-reset:
- shard-bmg: [DMESG-WARN][174] ([Intel XE#3876]) -> [PASS][175]
[174]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_exec_reset@parallel-gt-reset.html
[175]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_exec_reset@parallel-gt-reset.html
* {igt@xe_exec_system_allocator@many-large-execqueues-new-prefetch}:
- shard-bmg: [CRASH][176] ([Intel XE#6192]) -> [PASS][177] +9 other tests pass
[176]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-7/igt@xe_exec_system_allocator@many-large-execqueues-new-prefetch.html
[177]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-8/igt@xe_exec_system_allocator@many-large-execqueues-new-prefetch.html
* {igt@xe_exec_system_allocator@many-new-prefetch}:
- shard-lnl: [CRASH][178] ([Intel XE#6192]) -> [PASS][179] +4 other tests pass
[178]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-lnl-2/igt@xe_exec_system_allocator@many-new-prefetch.html
[179]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-lnl-1/igt@xe_exec_system_allocator@many-new-prefetch.html
* igt@xe_exec_system_allocator@process-many-stride-mmap-remap-dontunmap-eocheck:
- shard-bmg: [ABORT][180] ([Intel XE#5790]) -> [PASS][181]
[180]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_exec_system_allocator@process-many-stride-mmap-remap-dontunmap-eocheck.html
[181]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_exec_system_allocator@process-many-stride-mmap-remap-dontunmap-eocheck.html
* igt@xe_exec_threads@threads-shared-vm-userptr-invalidate:
- shard-bmg: [DMESG-FAIL][182] ([Intel XE#3876]) -> [PASS][183]
[182]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_exec_threads@threads-shared-vm-userptr-invalidate.html
[183]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_exec_threads@threads-shared-vm-userptr-invalidate.html
* igt@xe_module_load@load:
- shard-adlp: ([PASS][184], [PASS][185], [PASS][186], [PASS][187], [PASS][188], [PASS][189], [PASS][190], [PASS][191], [PASS][192], [PASS][193], [PASS][194], [PASS][195], [PASS][196], [PASS][197], [PASS][198], [PASS][199], [PASS][200], [PASS][201], [PASS][202], [PASS][203], [PASS][204], [PASS][205], [SKIP][206], [PASS][207], [PASS][208]) ([Intel XE#378] / [Intel XE#5612]) -> ([PASS][209], [PASS][210], [PASS][211], [PASS][212], [PASS][213], [PASS][214], [PASS][215], [PASS][216], [PASS][217], [PASS][218], [PASS][219], [PASS][220], [PASS][221], [PASS][222], [PASS][223], [PASS][224], [PASS][225], [PASS][226], [PASS][227], [PASS][228], [PASS][229], [PASS][230], [PASS][231], [PASS][232], [PASS][233])
[184]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-6/igt@xe_module_load@load.html
[185]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@xe_module_load@load.html
[186]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@xe_module_load@load.html
[187]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@xe_module_load@load.html
[188]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-9/igt@xe_module_load@load.html
[189]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-9/igt@xe_module_load@load.html
[190]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-4/igt@xe_module_load@load.html
[191]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-1/igt@xe_module_load@load.html
[192]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-3/igt@xe_module_load@load.html
[193]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-9/igt@xe_module_load@load.html
[194]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-8/igt@xe_module_load@load.html
[195]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-8/igt@xe_module_load@load.html
[196]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-8/igt@xe_module_load@load.html
[197]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-1/igt@xe_module_load@load.html
[198]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-1/igt@xe_module_load@load.html
[199]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-3/igt@xe_module_load@load.html
[200]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-3/igt@xe_module_load@load.html
[201]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-1/igt@xe_module_load@load.html
[202]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-2/igt@xe_module_load@load.html
[203]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-2/igt@xe_module_load@load.html
[204]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-6/igt@xe_module_load@load.html
[205]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-6/igt@xe_module_load@load.html
[206]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-2/igt@xe_module_load@load.html
[207]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-2/igt@xe_module_load@load.html
[208]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-adlp-8/igt@xe_module_load@load.html
[209]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-8/igt@xe_module_load@load.html
[210]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@xe_module_load@load.html
[211]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@xe_module_load@load.html
[212]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-2/igt@xe_module_load@load.html
[213]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-8/igt@xe_module_load@load.html
[214]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-8/igt@xe_module_load@load.html
[215]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-3/igt@xe_module_load@load.html
[216]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-3/igt@xe_module_load@load.html
[217]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-3/igt@xe_module_load@load.html
[218]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-8/igt@xe_module_load@load.html
[219]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-9/igt@xe_module_load@load.html
[220]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-9/igt@xe_module_load@load.html
[221]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-9/igt@xe_module_load@load.html
[222]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-9/igt@xe_module_load@load.html
[223]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-1/igt@xe_module_load@load.html
[224]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-1/igt@xe_module_load@load.html
[225]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-1/igt@xe_module_load@load.html
[226]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-1/igt@xe_module_load@load.html
[227]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_module_load@load.html
[228]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_module_load@load.html
[229]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_module_load@load.html
[230]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-6/igt@xe_module_load@load.html
[231]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@xe_module_load@load.html
[232]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@xe_module_load@load.html
[233]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-adlp-4/igt@xe_module_load@load.html
- shard-dg2-set2: ([PASS][234], [PASS][235], [PASS][236], [PASS][237], [PASS][238], [PASS][239], [PASS][240], [PASS][241], [PASS][242], [PASS][243], [PASS][244], [PASS][245], [PASS][246], [PASS][247], [PASS][248], [PASS][249], [PASS][250], [SKIP][251], [PASS][252], [PASS][253], [PASS][254]) ([Intel XE#378]) -> ([PASS][255], [PASS][256], [PASS][257], [PASS][258], [PASS][259], [PASS][260], [PASS][261], [PASS][262], [PASS][263], [PASS][264], [PASS][265], [PASS][266], [PASS][267], [PASS][268], [PASS][269], [PASS][270], [PASS][271], [PASS][272], [PASS][273], [PASS][274])
[234]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-463/igt@xe_module_load@load.html
[235]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-463/igt@xe_module_load@load.html
[236]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-463/igt@xe_module_load@load.html
[237]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-466/igt@xe_module_load@load.html
[238]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-466/igt@xe_module_load@load.html
[239]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-436/igt@xe_module_load@load.html
[240]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-432/igt@xe_module_load@load.html
[241]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-432/igt@xe_module_load@load.html
[242]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-432/igt@xe_module_load@load.html
[243]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-436/igt@xe_module_load@load.html
[244]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-436/igt@xe_module_load@load.html
[245]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-433/igt@xe_module_load@load.html
[246]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-433/igt@xe_module_load@load.html
[247]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-464/igt@xe_module_load@load.html
[248]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-464/igt@xe_module_load@load.html
[249]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-464/igt@xe_module_load@load.html
[250]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-434/igt@xe_module_load@load.html
[251]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-434/igt@xe_module_load@load.html
[252]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-434/igt@xe_module_load@load.html
[253]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-434/igt@xe_module_load@load.html
[254]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-434/igt@xe_module_load@load.html
[255]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_module_load@load.html
[256]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_module_load@load.html
[257]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_module_load@load.html
[258]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-434/igt@xe_module_load@load.html
[259]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-434/igt@xe_module_load@load.html
[260]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-434/igt@xe_module_load@load.html
[261]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-436/igt@xe_module_load@load.html
[262]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-436/igt@xe_module_load@load.html
[263]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@xe_module_load@load.html
[264]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-463/igt@xe_module_load@load.html
[265]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-436/igt@xe_module_load@load.html
[266]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-464/igt@xe_module_load@load.html
[267]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-464/igt@xe_module_load@load.html
[268]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-464/igt@xe_module_load@load.html
[269]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_module_load@load.html
[270]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_module_load@load.html
[271]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@xe_module_load@load.html
[272]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@xe_module_load@load.html
[273]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-466/igt@xe_module_load@load.html
[274]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-433/igt@xe_module_load@load.html
* igt@xe_pm@s2idle-vm-bind-unbind-all:
- shard-bmg: [DMESG-FAIL][275] -> [PASS][276]
[275]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-5/igt@xe_pm@s2idle-vm-bind-unbind-all.html
[276]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-4/igt@xe_pm@s2idle-vm-bind-unbind-all.html
#### Warnings ####
* igt@kms_content_protection@srm:
- shard-bmg: [SKIP][277] ([Intel XE#2341]) -> [FAIL][278] ([Intel XE#1178])
[277]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-6/igt@kms_content_protection@srm.html
[278]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_content_protection@srm.html
* igt@kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render:
- shard-bmg: [SKIP][279] ([Intel XE#2311]) -> [SKIP][280] ([Intel XE#2312]) +10 other tests skip
[279]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-8/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render.html
[280]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@drrs-2p-primscrn-indfb-pgflip-blt:
- shard-bmg: [SKIP][281] ([Intel XE#2312]) -> [SKIP][282] ([Intel XE#2311]) +8 other tests skip
[281]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-6/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-indfb-pgflip-blt.html
[282]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-indfb-pgflip-blt.html
* igt@kms_frontbuffer_tracking@fbc-2p-primscrn-shrfb-pgflip-blt:
- shard-bmg: [SKIP][283] ([Intel XE#5390]) -> [SKIP][284] ([Intel XE#2312]) +3 other tests skip
[283]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-shrfb-pgflip-blt.html
[284]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-shrfb-pgflip-blt.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff:
- shard-bmg: [SKIP][285] ([Intel XE#2312]) -> [SKIP][286] ([Intel XE#5390]) +7 other tests skip
[285]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-6/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff.html
[286]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-shrfb-plflip-blt:
- shard-bmg: [SKIP][287] ([Intel XE#2313]) -> [SKIP][288] ([Intel XE#2312]) +12 other tests skip
[287]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-7/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-shrfb-plflip-blt.html
[288]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-6/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-shrfb-plflip-blt.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt:
- shard-bmg: [SKIP][289] ([Intel XE#2312]) -> [SKIP][290] ([Intel XE#2313]) +6 other tests skip
[289]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-6/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt.html
[290]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-1/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt.html
* igt@kms_hdr@brightness-with-hdr:
- shard-bmg: [SKIP][291] ([Intel XE#3544]) -> [SKIP][292] ([Intel XE#3374] / [Intel XE#3544])
[291]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-1/igt@kms_hdr@brightness-with-hdr.html
[292]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-2/igt@kms_hdr@brightness-with-hdr.html
* igt@kms_tiled_display@basic-test-pattern:
- shard-bmg: [SKIP][293] ([Intel XE#2426]) -> [FAIL][294] ([Intel XE#1729])
[293]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-4/igt@kms_tiled_display@basic-test-pattern.html
[294]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_tiled_display@basic-test-pattern.html
* igt@kms_tiled_display@basic-test-pattern-with-chamelium:
- shard-bmg: [SKIP][295] ([Intel XE#2426]) -> [SKIP][296] ([Intel XE#2509])
[295]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-2/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html
[296]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html
* igt@xe_fault_injection@probe-fail-guc-xe_guc_ct_send_recv:
- shard-bmg: [ABORT][297] ([Intel XE#5466] / [Intel XE#5530]) -> [ABORT][298] ([Intel XE#4917] / [Intel XE#5466] / [Intel XE#5530])
[297]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-bmg-3/igt@xe_fault_injection@probe-fail-guc-xe_guc_ct_send_recv.html
[298]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-bmg-5/igt@xe_fault_injection@probe-fail-guc-xe_guc_ct_send_recv.html
* igt@xe_peer2peer@write:
- shard-dg2-set2: [FAIL][299] ([Intel XE#1173]) -> [SKIP][300] ([Intel XE#1061])
[299]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a/shard-dg2-436/igt@xe_peer2peer@write.html
[300]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/shard-dg2-432/igt@xe_peer2peer@write.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[Intel XE#1061]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1061
[Intel XE#1122]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1122
[Intel XE#1124]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1124
[Intel XE#1126]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1126
[Intel XE#1127]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1127
[Intel XE#1135]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1135
[Intel XE#1173]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1173
[Intel XE#1178]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1178
[Intel XE#1392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1392
[Intel XE#1397]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1397
[Intel XE#1406]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1406
[Intel XE#1435]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1435
[Intel XE#1489]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1489
[Intel XE#1499]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1499
[Intel XE#1503]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1503
[Intel XE#1727]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1727
[Intel XE#1729]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1729
[Intel XE#1745]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1745
[Intel XE#1874]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1874
[Intel XE#2168]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2168
[Intel XE#2191]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2191
[Intel XE#2229]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2229
[Intel XE#2234]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2234
[Intel XE#2244]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2244
[Intel XE#2252]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2252
[Intel XE#2284]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2284
[Intel XE#2286]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2286
[Intel XE#2291]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2291
[Intel XE#2293]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2293
[Intel XE#2311]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2311
[Intel XE#2312]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2312
[Intel XE#2313]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2313
[Intel XE#2314]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2314
[Intel XE#2316]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2316
[Intel XE#2320]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2320
[Intel XE#2321]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2321
[Intel XE#2322]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2322
[Intel XE#2327]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2327
[Intel XE#2330]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2330
[Intel XE#2341]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2341
[Intel XE#2352]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2352
[Intel XE#2360]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2360
[Intel XE#2375]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2375
[Intel XE#2380]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2380
[Intel XE#2392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2392
[Intel XE#2413]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2413
[Intel XE#2426]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2426
[Intel XE#2509]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2509
[Intel XE#2652]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2652
[Intel XE#2763]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2763
[Intel XE#2850]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2850
[Intel XE#288]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/288
[Intel XE#2887]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2887
[Intel XE#2894]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2894
[Intel XE#2907]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2907
[Intel XE#2953]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2953
[Intel XE#3009]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3009
[Intel XE#301]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/301
[Intel XE#306]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/306
[Intel XE#307]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/307
[Intel XE#308]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/308
[Intel XE#310]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/310
[Intel XE#3113]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3113
[Intel XE#316]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/316
[Intel XE#3374]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3374
[Intel XE#3414]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3414
[Intel XE#3544]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3544
[Intel XE#3573]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3573
[Intel XE#366]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/366
[Intel XE#367]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/367
[Intel XE#3718]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3718
[Intel XE#373]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/373
[Intel XE#378]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/378
[Intel XE#3876]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3876
[Intel XE#3904]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3904
[Intel XE#4130]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4130
[Intel XE#4173]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4173
[Intel XE#4331]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4331
[Intel XE#4345]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4345
[Intel XE#4356]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4356
[Intel XE#4494]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4494
[Intel XE#4504]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4504
[Intel XE#4543]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4543
[Intel XE#455]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/455
[Intel XE#4596]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4596
[Intel XE#4650]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4650
[Intel XE#4733]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4733
[Intel XE#4837]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4837
[Intel XE#4915]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4915
[Intel XE#4917]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4917
[Intel XE#4943]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4943
[Intel XE#5020]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5020
[Intel XE#5300]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5300
[Intel XE#5354]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5354
[Intel XE#5390]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5390
[Intel XE#5427]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5427
[Intel XE#5428]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5428
[Intel XE#5466]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5466
[Intel XE#5530]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5530
[Intel XE#5561]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5561
[Intel XE#5565]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5565
[Intel XE#5575]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5575
[Intel XE#5594]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5594
[Intel XE#5612]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5612
[Intel XE#5625]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5625
[Intel XE#5786]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5786
[Intel XE#5790]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5790
[Intel XE#5890]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5890
[Intel XE#6050]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6050
[Intel XE#607]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/607
[Intel XE#6078]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6078
[Intel XE#610]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/610
[Intel XE#616]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/616
[Intel XE#6190]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6190
[Intel XE#6192]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/6192
[Intel XE#651]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/651
[Intel XE#653]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/653
[Intel XE#656]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/656
[Intel XE#658]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/658
[Intel XE#734]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/734
[Intel XE#776]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/776
[Intel XE#787]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/787
[Intel XE#870]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/870
[Intel XE#929]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/929
[Intel XE#944]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/944
[i915#3804]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/3804
Build changes
-------------
* Linux: xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a -> xe-pw-154627v2
IGT_8550: 4f8c7886ad02e116804ec08714f17bce1755c6e4 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-3822-af3cdefd0a1ad2dad29fcde0854ccbce494cc28a: af3cdefd0a1ad2dad29fcde0854ccbce494cc28a
xe-pw-154627v2: 154627v2
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154627v2/index.html
[-- Attachment #2: Type: text/html, Size: 84450 bytes --]
^ permalink raw reply [flat|nested] 90+ messages in thread