From: Matthew Brost <matthew.brost@intel.com>
To: Tomasz Lis <tomasz.lis@intel.com>
Cc: intel-xe@lists.freedesktop.org,
"Michał Winiarski" <michal.winiarski@intel.com>,
"Michał Wajdeczko" <michal.wajdeczko@intel.com>,
"Piotr Piórkowski" <piotr.piorkowski@intel.com>
Subject: Re: [PATCH v2 1/4] drm/xe/vf: Fix GuC FW check for VF migration support
Date: Wed, 15 Oct 2025 19:58:29 -0700 [thread overview]
Message-ID: <aPBfVRjz5AraRGsM@lstrano-desk.jf.intel.com> (raw)
In-Reply-To: <20251015002755.720992-2-tomasz.lis@intel.com>
On Wed, Oct 15, 2025 at 02:27:52AM +0200, Tomasz Lis wrote:
> The check was done before GuC ABI version could be acquired.
> Comparing only to zeros provides very stable results, though
> not the ones expected.
>
> This change dislodged part of the VF migration support check
> and moved it to after GuC handshake.
>
I thought the handshake has performed as part of read_gmdid earlier in
the driver probe. Still confused by how this got broke but anyways, I
can confirm this fixed the VF probe issue:
Tested-by: Matthew Brost <matthew.brost@intel.com>
As I'm confused, I'll leave the RB to someone else.
Matt
> Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 44 +++++++++++++++++++++++++++++
> drivers/gpu/drm/xe/xe_gt_sriov_vf.h | 1 +
> drivers/gpu/drm/xe/xe_guc.c | 2 ++
> drivers/gpu/drm/xe/xe_sriov_vf.c | 10 -------
> 4 files changed, 47 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 46518e629ba3..95c10de0732f 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
> @@ -314,6 +314,50 @@ static int guc_action_vf_notify_resfix_done(struct xe_guc *guc)
> return ret > 0 ? -EPROTO : ret;
> }
>
> +static void vf_disable_migration(struct xe_gt *gt, const char *fmt, ...)
> +{
> + struct xe_device *xe = gt_to_xe(gt);
> + struct va_format vaf;
> + va_list va_args;
> +
> + xe_gt_assert(gt, IS_SRIOV_VF(xe));
> +
> + va_start(va_args, fmt);
> + vaf.fmt = fmt;
> + vaf.va = &va_args;
> + xe_gt_sriov_notice(gt, "migration disabled: %pV\n", &vaf);
> + va_end(va_args);
> +
> + xe->sriov.vf.migration.enabled = false;
> +}
> +
> +#define NEEDS_SW_CCS_MIGRATION_WA(xe) \
> + ((xe)->info.platform == XE_PANTHERLAKE)
> +
> +/**
> + * xe_gt_sriov_vf_guc_check_migration_support - Check for disable migration due to GuC.
> + * @gt: the &xe_gt struct instance linked to target GuC
> + *
> + * Performs late disable of VF migration feature in case GuC FW cannot support it.
> + */
> +void xe_gt_sriov_vf_guc_check_migration_support(struct xe_gt *gt)
> +{
> + struct xe_device *xe = gt_to_xe(gt);
> +
> + if (!xe_sriov_vf_migration_supported(xe))
> + return;
> +
> + if (NEEDS_SW_CCS_MIGRATION_WA(xe)) {
> + struct xe_uc_fw_version guc_version;
> +
> + xe_gt_sriov_vf_guc_versions(gt, NULL, &guc_version);
> + if (MAKE_GUC_VER_STRUCT(guc_version) < MAKE_GUC_VER(1, 23, 0))
> + return vf_disable_migration(gt,
> + "CCS migration requires GuC ABI >= 1.23 but only %u.%u found",
> + guc_version.major, guc_version.minor);
> + }
> +}
> +
> /**
> * vf_notify_resfix_done - Notify GuC about resource fixups apply completed.
> * @gt: the &xe_gt struct instance linked to target GuC
> diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> index af40276790fa..60a3b9b05b20 100644
> --- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> +++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
> @@ -26,6 +26,7 @@ void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
> int xe_gt_sriov_vf_init_early(struct xe_gt *gt);
> int xe_gt_sriov_vf_init(struct xe_gt *gt);
> bool xe_gt_sriov_vf_recovery_pending(struct xe_gt *gt);
> +void xe_gt_sriov_vf_guc_check_migration_support(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);
> diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> index d94490979adc..3c4e64233b3a 100644
> --- a/drivers/gpu/drm/xe/xe_guc.c
> +++ b/drivers/gpu/drm/xe/xe_guc.c
> @@ -713,6 +713,8 @@ static int vf_guc_init_noalloc(struct xe_guc *guc)
> if (err)
> return err;
>
> + xe_gt_sriov_vf_guc_check_migration_support(gt);
> +
> err = xe_gt_sriov_vf_query_config(gt);
> if (err)
> return err;
> diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
> index 911d5720917b..5fb042c05112 100644
> --- a/drivers/gpu/drm/xe/xe_sriov_vf.c
> +++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
> @@ -163,16 +163,6 @@ static void vf_migration_init_early(struct xe_device *xe)
> return vf_disable_migration(xe, "requires gfx version >= 20, but only %u found",
> GRAPHICS_VER(xe));
>
> - if (!IS_DGFX(xe)) {
> - struct xe_uc_fw_version guc_version;
> -
> - xe_gt_sriov_vf_guc_versions(xe_device_get_gt(xe, 0), NULL, &guc_version);
> - if (MAKE_GUC_VER_STRUCT(guc_version) < MAKE_GUC_VER(1, 23, 0))
> - return vf_disable_migration(xe,
> - "CCS migration requires GuC ABI >= 1.23 but only %u.%u found",
> - guc_version.major, guc_version.minor);
> - }
> -
> xe->sriov.vf.migration.enabled = true;
> xe_sriov_dbg(xe, "migration support enabled\n");
> }
> --
> 2.25.1
>
next prev parent reply other threads:[~2025-10-16 2:58 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-15 0:27 [PATCH v2 0/4] drm/xe/vf: Minor fixes to post-migration recovery Tomasz Lis
2025-10-15 0:27 ` [PATCH v2 1/4] drm/xe/vf: Fix GuC FW check for VF migration support Tomasz Lis
2025-10-16 2:58 ` Matthew Brost [this message]
2025-10-15 0:27 ` [PATCH v2 2/4] drm/xe/vf: Skip fixups on VF migration before getting GGTT info Tomasz Lis
2025-10-15 0:27 ` [PATCH v2 3/4] drm/xe: Assert that VF will never use fixed placement of BOs Tomasz Lis
2025-10-15 0:27 ` [PATCH v2 4/4] drm/xe/vf: Do not disable VF migration on ATS-M Tomasz Lis
2025-10-15 0:33 ` ✓ CI.KUnit: success for drm/xe/vf: Minor fixes to post-migration recovery (rev2) Patchwork
2025-10-15 1:13 ` ✓ Xe.CI.BAT: " Patchwork
2025-10-15 13:02 ` ✗ Xe.CI.Full: failure " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=aPBfVRjz5AraRGsM@lstrano-desk.jf.intel.com \
--to=matthew.brost@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=michal.wajdeczko@intel.com \
--cc=michal.winiarski@intel.com \
--cc=piotr.piorkowski@intel.com \
--cc=tomasz.lis@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox