* [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
@ 2024-01-04 5:39 Ma Jun
2024-01-04 15:00 ` Alex Deucher
0 siblings, 1 reply; 6+ messages in thread
From: Ma Jun @ 2024-01-04 5:39 UTC (permalink / raw)
To: amd-gfx, christian.koenig, Alexander.Deucher; +Cc: Ma Jun
Fix the warning info below during mode1 reset.
[ +0.000004] Call Trace:
[ +0.000004] <TASK>
[ +0.000006] ? show_regs+0x6e/0x80
[ +0.000011] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000005] ? __warn+0x91/0x150
[ +0.000009] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000006] ? report_bug+0x19d/0x1b0
[ +0.000013] ? handle_bug+0x46/0x80
[ +0.000012] ? exc_invalid_op+0x1d/0x80
[ +0.000011] ? asm_exc_invalid_op+0x1f/0x30
[ +0.000014] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000007] ? __flush_work.isra.0+0x208/0x390
[ +0.000007] ? _prb_read_valid+0x216/0x290
[ +0.000008] __cancel_work_timer+0x11d/0x1a0
[ +0.000007] ? try_to_grab_pending+0xe8/0x190
[ +0.000012] cancel_work_sync+0x14/0x20
[ +0.000008] amddrm_sched_stop+0x3c/0x1d0 [amd_sched]
[ +0.000032] amdgpu_device_gpu_recover+0x29a/0xe90 [amdgpu]
This warning info was printed after applying the patch
"drm/sched: Convert drm scheduler to use a work queue rather than kthread".
The root cause is that amdgpu driver tries to use the uninitialized
work_struct in the struct drm_gpu_scheduler
Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 16 +++++++++++++---
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 3 ++-
2 files changed, 15 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 4b1d5f42249f..7bac3019aa0a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5575,6 +5575,16 @@ static inline void amdgpu_device_stop_pending_resets(struct amdgpu_device *adev)
}
+static bool amdgpu_is_ring_sched_ready(struct amdgpu_ring *ring)
+{
+ if (!ring)
+ return false;
+
+ if (ring->no_scheduler || !drm_sched_wqueue_ready(&ring->sched))
+ return false;
+
+ return true;
+}
/**
* amdgpu_device_gpu_recover - reset the asic and recover scheduler
*
@@ -5700,7 +5710,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = tmp_adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_is_ring_sched_ready(ring))
continue;
drm_sched_stop(&ring->sched, job ? &job->base : NULL);
@@ -5776,7 +5786,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = tmp_adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_is_ring_sched_ready(ring))
continue;
drm_sched_start(&ring->sched, true);
@@ -6265,7 +6275,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_is_ring_sched_ready(ring))
continue;
drm_sched_start(&ring->sched, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 41266bc99345..3650e4d06e53 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -636,7 +636,8 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
ring->name);
- ring->sched.ready = !r;
+ if (!ring->no_scheduler)
+ ring->sched.ready = !r;
return r;
}
--
2.34.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
2024-01-04 5:39 [PATCH] drm/amdgpu: Fix the warning info in mode1 reset Ma Jun
@ 2024-01-04 15:00 ` Alex Deucher
0 siblings, 0 replies; 6+ messages in thread
From: Alex Deucher @ 2024-01-04 15:00 UTC (permalink / raw)
To: Ma Jun; +Cc: Alexander.Deucher, christian.koenig, amd-gfx
On Thu, Jan 4, 2024 at 12:55 AM Ma Jun <Jun.Ma2@amd.com> wrote:
>
> Fix the warning info below during mode1 reset.
> [ +0.000004] Call Trace:
> [ +0.000004] <TASK>
> [ +0.000006] ? show_regs+0x6e/0x80
> [ +0.000011] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000005] ? __warn+0x91/0x150
> [ +0.000009] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000006] ? report_bug+0x19d/0x1b0
> [ +0.000013] ? handle_bug+0x46/0x80
> [ +0.000012] ? exc_invalid_op+0x1d/0x80
> [ +0.000011] ? asm_exc_invalid_op+0x1f/0x30
> [ +0.000014] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000007] ? __flush_work.isra.0+0x208/0x390
> [ +0.000007] ? _prb_read_valid+0x216/0x290
> [ +0.000008] __cancel_work_timer+0x11d/0x1a0
> [ +0.000007] ? try_to_grab_pending+0xe8/0x190
> [ +0.000012] cancel_work_sync+0x14/0x20
> [ +0.000008] amddrm_sched_stop+0x3c/0x1d0 [amd_sched]
> [ +0.000032] amdgpu_device_gpu_recover+0x29a/0xe90 [amdgpu]
>
> This warning info was printed after applying the patch
> "drm/sched: Convert drm scheduler to use a work queue rather than kthread".
> The root cause is that amdgpu driver tries to use the uninitialized
> work_struct in the struct drm_gpu_scheduler
>
> Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 16 +++++++++++++---
> drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 3 ++-
> 2 files changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 4b1d5f42249f..7bac3019aa0a 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -5575,6 +5575,16 @@ static inline void amdgpu_device_stop_pending_resets(struct amdgpu_device *adev)
>
> }
>
> +static bool amdgpu_is_ring_sched_ready(struct amdgpu_ring *ring)
I would either rename this amdgpu_device_is_ring_sched_ready() for
consistency, or move it to amdgpu_ring.c and rename it to
amdgpu_ring_sched_ready().
Alex
> +{
> + if (!ring)
> + return false;
> +
> + if (ring->no_scheduler || !drm_sched_wqueue_ready(&ring->sched))
> + return false;
> +
> + return true;
> +}
> /**
> * amdgpu_device_gpu_recover - reset the asic and recover scheduler
> *
> @@ -5700,7 +5710,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = tmp_adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_is_ring_sched_ready(ring))
> continue;
>
> drm_sched_stop(&ring->sched, job ? &job->base : NULL);
> @@ -5776,7 +5786,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = tmp_adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_is_ring_sched_ready(ring))
> continue;
>
> drm_sched_start(&ring->sched, true);
> @@ -6265,7 +6275,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_is_ring_sched_ready(ring))
> continue;
>
> drm_sched_start(&ring->sched, true);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> index 41266bc99345..3650e4d06e53 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> @@ -636,7 +636,8 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
> DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
> ring->name);
>
> - ring->sched.ready = !r;
> + if (!ring->no_scheduler)
> + ring->sched.ready = !r;
> return r;
> }
>
> --
> 2.34.1
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
@ 2024-01-25 16:15 Alex Deucher
2024-01-29 15:33 ` Deucher, Alexander
0 siblings, 1 reply; 6+ messages in thread
From: Alex Deucher @ 2024-01-25 16:15 UTC (permalink / raw)
To: amd-gfx; +Cc: Alex Deucher, Ma Jun, Vitaly Prosyak
From: Ma Jun <Jun.Ma2@amd.com>
Fix the warning info below during mode1 reset.
[ +0.000004] Call Trace:
[ +0.000004] <TASK>
[ +0.000006] ? show_regs+0x6e/0x80
[ +0.000011] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000005] ? __warn+0x91/0x150
[ +0.000009] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000006] ? report_bug+0x19d/0x1b0
[ +0.000013] ? handle_bug+0x46/0x80
[ +0.000012] ? exc_invalid_op+0x1d/0x80
[ +0.000011] ? asm_exc_invalid_op+0x1f/0x30
[ +0.000014] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000007] ? __flush_work.isra.0+0x208/0x390
[ +0.000007] ? _prb_read_valid+0x216/0x290
[ +0.000008] __cancel_work_timer+0x11d/0x1a0
[ +0.000007] ? try_to_grab_pending+0xe8/0x190
[ +0.000012] cancel_work_sync+0x14/0x20
[ +0.000008] amddrm_sched_stop+0x3c/0x1d0 [amd_sched]
[ +0.000032] amdgpu_device_gpu_recover+0x29a/0xe90 [amdgpu]
This warning info was printed after applying the patch
"drm/sched: Convert drm scheduler to use a work queue rather than kthread".
The root cause is that amdgpu driver tries to use the uninitialized
work_struct in the struct drm_gpu_scheduler
v2:
- Rename the function to amdgpu_ring_sched_ready and move it to
amdgpu_ring.c (Alex)
v3:
- Fix a few more checks based on Vitaly's patch (Alex)
Fixes: 11b3b9f461c5 ("drm/sched: Check scheduler ready before calling timeout handling")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 8 ++++----
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 12 ++++++------
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 14 +++++++++++++-
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +-
5 files changed, 25 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index 899e31e3a5e8..3a3f3ce09f00 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -290,7 +290,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
struct amdgpu_ring *ring = &adev->gfx.compute_ring[i];
- if (!(ring && drm_sched_wqueue_ready(&ring->sched)))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
/* stop secheduler and drain ring. */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index e485dd3357c6..1afbb2e932c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -1678,7 +1678,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_wqueue_stop(&ring->sched);
}
@@ -1694,7 +1694,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_wqueue_start(&ring->sched);
}
@@ -1916,8 +1916,8 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
ring = adev->rings[val];
- if (!ring || !ring->funcs->preempt_ib ||
- !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring) ||
+ !ring->funcs->preempt_ib)
return -EINVAL;
/* the last preemption failed */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 1a04ccba9542..7ff17df7a5ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5042,7 +5042,7 @@ bool amdgpu_device_has_job_running(struct amdgpu_device *adev)
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
spin_lock(&ring->sched.job_list_lock);
@@ -5181,7 +5181,7 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
/* Clear job fence from fence drv to avoid force_completion
@@ -5648,7 +5648,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = tmp_adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_stop(&ring->sched, job ? &job->base : NULL);
@@ -5717,7 +5717,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = tmp_adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_start(&ring->sched, true);
@@ -6072,7 +6072,7 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_stop(&ring->sched, NULL);
@@ -6214,7 +6214,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_start(&ring->sched, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 45424ebf9681..9ae386e9d41d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -634,7 +634,8 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
ring->name);
- ring->sched.ready = !r;
+ if (!ring->no_scheduler)
+ ring->sched.ready = !r;
return r;
}
@@ -717,3 +718,14 @@ void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring)
if (ring->is_sw_ring)
amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_DE);
}
+
+bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring)
+{
+ if (!ring)
+ return false;
+
+ if (ring->no_scheduler || !drm_sched_wqueue_ready(&ring->sched))
+ return false;
+
+ return true;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index bbb53720a018..fe1a61eb6e4c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -450,5 +450,5 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
int amdgpu_ib_pool_init(struct amdgpu_device *adev);
void amdgpu_ib_pool_fini(struct amdgpu_device *adev);
int amdgpu_ib_ring_tests(struct amdgpu_device *adev);
-
+bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring);
#endif
--
2.42.0
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
2024-01-25 16:15 Alex Deucher
@ 2024-01-29 15:33 ` Deucher, Alexander
2024-01-29 19:28 ` vitaly prosyak
2024-01-30 2:40 ` Ma, Jun
0 siblings, 2 replies; 6+ messages in thread
From: Deucher, Alexander @ 2024-01-29 15:33 UTC (permalink / raw)
To: amd-gfx@lists.freedesktop.org; +Cc: Ma, Jun, Prosyak, Vitaly
[-- Attachment #1: Type: text/plain, Size: 9157 bytes --]
[AMD Official Use Only - General]
Ping?
________________________________
From: Deucher, Alexander <Alexander.Deucher@amd.com>
Sent: Thursday, January 25, 2024 11:15 AM
To: amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org>
Cc: Ma, Jun <Jun.Ma2@amd.com>; Deucher, Alexander <Alexander.Deucher@amd.com>; Prosyak, Vitaly <Vitaly.Prosyak@amd.com>
Subject: [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
From: Ma Jun <Jun.Ma2@amd.com>
Fix the warning info below during mode1 reset.
[ +0.000004] Call Trace:
[ +0.000004] <TASK>
[ +0.000006] ? show_regs+0x6e/0x80
[ +0.000011] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000005] ? __warn+0x91/0x150
[ +0.000009] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000006] ? report_bug+0x19d/0x1b0
[ +0.000013] ? handle_bug+0x46/0x80
[ +0.000012] ? exc_invalid_op+0x1d/0x80
[ +0.000011] ? asm_exc_invalid_op+0x1f/0x30
[ +0.000014] ? __flush_work.isra.0+0x2e8/0x390
[ +0.000007] ? __flush_work.isra.0+0x208/0x390
[ +0.000007] ? _prb_read_valid+0x216/0x290
[ +0.000008] __cancel_work_timer+0x11d/0x1a0
[ +0.000007] ? try_to_grab_pending+0xe8/0x190
[ +0.000012] cancel_work_sync+0x14/0x20
[ +0.000008] amddrm_sched_stop+0x3c/0x1d0 [amd_sched]
[ +0.000032] amdgpu_device_gpu_recover+0x29a/0xe90 [amdgpu]
This warning info was printed after applying the patch
"drm/sched: Convert drm scheduler to use a work queue rather than kthread".
The root cause is that amdgpu driver tries to use the uninitialized
work_struct in the struct drm_gpu_scheduler
v2:
- Rename the function to amdgpu_ring_sched_ready and move it to
amdgpu_ring.c (Alex)
v3:
- Fix a few more checks based on Vitaly's patch (Alex)
Fixes: 11b3b9f461c5 ("drm/sched: Check scheduler ready before calling timeout handling")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
.../gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 2 +-
drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 8 ++++----
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 12 ++++++------
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 14 +++++++++++++-
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +-
5 files changed, 25 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
index 899e31e3a5e8..3a3f3ce09f00 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
@@ -290,7 +290,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
struct amdgpu_ring *ring = &adev->gfx.compute_ring[i];
- if (!(ring && drm_sched_wqueue_ready(&ring->sched)))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
/* stop secheduler and drain ring. */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index e485dd3357c6..1afbb2e932c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -1678,7 +1678,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_wqueue_stop(&ring->sched);
}
@@ -1694,7 +1694,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_wqueue_start(&ring->sched);
}
@@ -1916,8 +1916,8 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
ring = adev->rings[val];
- if (!ring || !ring->funcs->preempt_ib ||
- !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring) ||
+ !ring->funcs->preempt_ib)
return -EINVAL;
/* the last preemption failed */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 1a04ccba9542..7ff17df7a5ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5042,7 +5042,7 @@ bool amdgpu_device_has_job_running(struct amdgpu_device *adev)
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
spin_lock(&ring->sched.job_list_lock);
@@ -5181,7 +5181,7 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
/* Clear job fence from fence drv to avoid force_completion
@@ -5648,7 +5648,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = tmp_adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_stop(&ring->sched, job ? &job->base : NULL);
@@ -5717,7 +5717,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = tmp_adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_start(&ring->sched, true);
@@ -6072,7 +6072,7 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_stop(&ring->sched, NULL);
@@ -6214,7 +6214,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
- if (!ring || !drm_sched_wqueue_ready(&ring->sched))
+ if (!amdgpu_ring_sched_ready(ring))
continue;
drm_sched_start(&ring->sched, true);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 45424ebf9681..9ae386e9d41d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -634,7 +634,8 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
ring->name);
- ring->sched.ready = !r;
+ if (!ring->no_scheduler)
+ ring->sched.ready = !r;
return r;
}
@@ -717,3 +718,14 @@ void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring)
if (ring->is_sw_ring)
amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_DE);
}
+
+bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring)
+{
+ if (!ring)
+ return false;
+
+ if (ring->no_scheduler || !drm_sched_wqueue_ready(&ring->sched))
+ return false;
+
+ return true;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index bbb53720a018..fe1a61eb6e4c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -450,5 +450,5 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
int amdgpu_ib_pool_init(struct amdgpu_device *adev);
void amdgpu_ib_pool_fini(struct amdgpu_device *adev);
int amdgpu_ib_ring_tests(struct amdgpu_device *adev);
-
+bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring);
#endif
--
2.42.0
[-- Attachment #2: Type: text/html, Size: 17967 bytes --]
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
2024-01-29 15:33 ` Deucher, Alexander
@ 2024-01-29 19:28 ` vitaly prosyak
2024-01-30 2:40 ` Ma, Jun
1 sibling, 0 replies; 6+ messages in thread
From: vitaly prosyak @ 2024-01-29 19:28 UTC (permalink / raw)
To: Deucher, Alexander, amd-gfx@lists.freedesktop.org
Cc: Ma, Jun, Prosyak, Vitaly
[-- Attachment #1: Type: text/plain, Size: 12047 bytes --]
The change looks good to me.
Validated now on kernel 5.7 ( branch origin/adeucher/amd-staging-drm-next)
Thanks, Vitaly
On 2024-01-29 10:33, Deucher, Alexander wrote:
>
> [AMD Official Use Only - General]
>
>
> Ping?
>
> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> *From:* Deucher, Alexander <Alexander.Deucher@amd.com>
> *Sent:* Thursday, January 25, 2024 11:15 AM
> *To:* amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org>
> *Cc:* Ma, Jun <Jun.Ma2@amd.com>; Deucher, Alexander <Alexander.Deucher@amd.com>; Prosyak, Vitaly <Vitaly.Prosyak@amd.com>
> *Subject:* [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
>
> From: Ma Jun <Jun.Ma2@amd.com>
>
> Fix the warning info below during mode1 reset.
> [ +0.000004] Call Trace:
> [ +0.000004] <TASK>
> [ +0.000006] ? show_regs+0x6e/0x80
> [ +0.000011] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000005] ? __warn+0x91/0x150
> [ +0.000009] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000006] ? report_bug+0x19d/0x1b0
> [ +0.000013] ? handle_bug+0x46/0x80
> [ +0.000012] ? exc_invalid_op+0x1d/0x80
> [ +0.000011] ? asm_exc_invalid_op+0x1f/0x30
> [ +0.000014] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000007] ? __flush_work.isra.0+0x208/0x390
> [ +0.000007] ? _prb_read_valid+0x216/0x290
> [ +0.000008] __cancel_work_timer+0x11d/0x1a0
> [ +0.000007] ? try_to_grab_pending+0xe8/0x190
> [ +0.000012] cancel_work_sync+0x14/0x20
> [ +0.000008] amddrm_sched_stop+0x3c/0x1d0 [amd_sched]
> [ +0.000032] amdgpu_device_gpu_recover+0x29a/0xe90 [amdgpu]
>
> This warning info was printed after applying the patch
> "drm/sched: Convert drm scheduler to use a work queue rather than kthread".
> The root cause is that amdgpu driver tries to use the uninitialized
> work_struct in the struct drm_gpu_scheduler
>
> v2:
> - Rename the function to amdgpu_ring_sched_ready and move it to
> amdgpu_ring.c (Alex)
> v3:
> - Fix a few more checks based on Vitaly's patch (Alex)
>
> Fixes: 11b3b9f461c5 ("drm/sched: Check scheduler ready before calling timeout handling")
> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
> Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
> Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
> ---
> .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 2 +-
> drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 8 ++++----
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 12 ++++++------
> drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 14 +++++++++++++-
> drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +-
> 5 files changed, 25 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
> index 899e31e3a5e8..3a3f3ce09f00 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
> @@ -290,7 +290,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus
> for (i = 0; i < adev->gfx.num_compute_rings; i++) {
> struct amdgpu_ring *ring = &adev->gfx.compute_ring[i];
>
> - if (!(ring && drm_sched_wqueue_ready(&ring->sched)))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> /* stop secheduler and drain ring. */
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> index e485dd3357c6..1afbb2e932c6 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> @@ -1678,7 +1678,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
> for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
> drm_sched_wqueue_stop(&ring->sched);
> }
> @@ -1694,7 +1694,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
> for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
> drm_sched_wqueue_start(&ring->sched);
> }
> @@ -1916,8 +1916,8 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
>
> ring = adev->rings[val];
>
> - if (!ring || !ring->funcs->preempt_ib ||
> - !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring) ||
> + !ring->funcs->preempt_ib)
> return -EINVAL;
>
> /* the last preemption failed */
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 1a04ccba9542..7ff17df7a5ce 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -5042,7 +5042,7 @@ bool amdgpu_device_has_job_running(struct amdgpu_device *adev)
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> spin_lock(&ring->sched.job_list_lock);
> @@ -5181,7 +5181,7 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> /* Clear job fence from fence drv to avoid force_completion
> @@ -5648,7 +5648,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = tmp_adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_stop(&ring->sched, job ? &job->base : NULL);
> @@ -5717,7 +5717,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = tmp_adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_start(&ring->sched, true);
> @@ -6072,7 +6072,7 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_stop(&ring->sched, NULL);
> @@ -6214,7 +6214,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_start(&ring->sched, true);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> index 45424ebf9681..9ae386e9d41d 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> @@ -634,7 +634,8 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
> DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
> ring->name);
>
> - ring->sched.ready = !r;
> + if (!ring->no_scheduler)
> + ring->sched.ready = !r;
> return r;
> }
>
> @@ -717,3 +718,14 @@ void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring)
> if (ring->is_sw_ring)
> amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_DE);
> }
> +
> +bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring)
> +{
> + if (!ring)
> + return false;
> +
> + if (ring->no_scheduler || !drm_sched_wqueue_ready(&ring->sched))
> + return false;
> +
> + return true;
> +}
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> index bbb53720a018..fe1a61eb6e4c 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> @@ -450,5 +450,5 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
> int amdgpu_ib_pool_init(struct amdgpu_device *adev);
> void amdgpu_ib_pool_fini(struct amdgpu_device *adev);
> int amdgpu_ib_ring_tests(struct amdgpu_device *adev);
> -
> +bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring);
> #endif
> --
> 2.42.0
>
[-- Attachment #2: Type: text/html, Size: 23588 bytes --]
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
2024-01-29 15:33 ` Deucher, Alexander
2024-01-29 19:28 ` vitaly prosyak
@ 2024-01-30 2:40 ` Ma, Jun
1 sibling, 0 replies; 6+ messages in thread
From: Ma, Jun @ 2024-01-30 2:40 UTC (permalink / raw)
To: Deucher, Alexander, amd-gfx@lists.freedesktop.org
Cc: Ma, Jun, Prosyak, Vitaly
Reviewed-by: Ma Jun <Jun.Ma2@amd.com>
Regards,
Ma Jun
On 1/29/2024 11:33 PM, Deucher, Alexander wrote:
> [AMD Official Use Only - General]
>
>
> Ping?
>
> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> *From:* Deucher, Alexander <Alexander.Deucher@amd.com>
> *Sent:* Thursday, January 25, 2024 11:15 AM
> *To:* amd-gfx@lists.freedesktop.org <amd-gfx@lists.freedesktop.org>
> *Cc:* Ma, Jun <Jun.Ma2@amd.com>; Deucher, Alexander <Alexander.Deucher@amd.com>; Prosyak, Vitaly <Vitaly.Prosyak@amd.com>
> *Subject:* [PATCH] drm/amdgpu: Fix the warning info in mode1 reset
>
> From: Ma Jun <Jun.Ma2@amd.com>
>
> Fix the warning info below during mode1 reset.
> [ +0.000004] Call Trace:
> [ +0.000004] <TASK>
> [ +0.000006] ? show_regs+0x6e/0x80
> [ +0.000011] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000005] ? __warn+0x91/0x150
> [ +0.000009] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000006] ? report_bug+0x19d/0x1b0
> [ +0.000013] ? handle_bug+0x46/0x80
> [ +0.000012] ? exc_invalid_op+0x1d/0x80
> [ +0.000011] ? asm_exc_invalid_op+0x1f/0x30
> [ +0.000014] ? __flush_work.isra.0+0x2e8/0x390
> [ +0.000007] ? __flush_work.isra.0+0x208/0x390
> [ +0.000007] ? _prb_read_valid+0x216/0x290
> [ +0.000008] __cancel_work_timer+0x11d/0x1a0
> [ +0.000007] ? try_to_grab_pending+0xe8/0x190
> [ +0.000012] cancel_work_sync+0x14/0x20
> [ +0.000008] amddrm_sched_stop+0x3c/0x1d0 [amd_sched]
> [ +0.000032] amdgpu_device_gpu_recover+0x29a/0xe90 [amdgpu]
>
> This warning info was printed after applying the patch
> "drm/sched: Convert drm scheduler to use a work queue rather than kthread".
> The root cause is that amdgpu driver tries to use the uninitialized
> work_struct in the struct drm_gpu_scheduler
>
> v2:
> - Rename the function to amdgpu_ring_sched_ready and move it to
> amdgpu_ring.c (Alex)
> v3:
> - Fix a few more checks based on Vitaly's patch (Alex)
>
> Fixes: 11b3b9f461c5 ("drm/sched: Check scheduler ready before calling timeout handling")
> Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
> Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
> Signed-off-by: Ma Jun <Jun.Ma2@amd.com>
> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
> ---
> .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c | 2 +-
> drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 8 ++++----
> drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 12 ++++++------
> drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 14 +++++++++++++-
> drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +-
> 5 files changed, 25 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
> index 899e31e3a5e8..3a3f3ce09f00 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c
> @@ -290,7 +290,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus
> for (i = 0; i < adev->gfx.num_compute_rings; i++) {
> struct amdgpu_ring *ring = &adev->gfx.compute_ring[i];
>
> - if (!(ring && drm_sched_wqueue_ready(&ring->sched)))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> /* stop secheduler and drain ring. */
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> index e485dd3357c6..1afbb2e932c6 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
> @@ -1678,7 +1678,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
> for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
> drm_sched_wqueue_stop(&ring->sched);
> }
> @@ -1694,7 +1694,7 @@ static int amdgpu_debugfs_test_ib_show(struct seq_file *m, void *unused)
> for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
> drm_sched_wqueue_start(&ring->sched);
> }
> @@ -1916,8 +1916,8 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
>
> ring = adev->rings[val];
>
> - if (!ring || !ring->funcs->preempt_ib ||
> - !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring) ||
> + !ring->funcs->preempt_ib)
> return -EINVAL;
>
> /* the last preemption failed */
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index 1a04ccba9542..7ff17df7a5ce 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -5042,7 +5042,7 @@ bool amdgpu_device_has_job_running(struct amdgpu_device *adev)
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> spin_lock(&ring->sched.job_list_lock);
> @@ -5181,7 +5181,7 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> /* Clear job fence from fence drv to avoid force_completion
> @@ -5648,7 +5648,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = tmp_adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_stop(&ring->sched, job ? &job->base : NULL);
> @@ -5717,7 +5717,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = tmp_adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_start(&ring->sched, true);
> @@ -6072,7 +6072,7 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_stop(&ring->sched, NULL);
> @@ -6214,7 +6214,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
> for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
> struct amdgpu_ring *ring = adev->rings[i];
>
> - if (!ring || !drm_sched_wqueue_ready(&ring->sched))
> + if (!amdgpu_ring_sched_ready(ring))
> continue;
>
> drm_sched_start(&ring->sched, true);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> index 45424ebf9681..9ae386e9d41d 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
> @@ -634,7 +634,8 @@ int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
> DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
> ring->name);
>
> - ring->sched.ready = !r;
> + if (!ring->no_scheduler)
> + ring->sched.ready = !r;
> return r;
> }
>
> @@ -717,3 +718,14 @@ void amdgpu_ring_ib_on_emit_de(struct amdgpu_ring *ring)
> if (ring->is_sw_ring)
> amdgpu_sw_ring_ib_mark_offset(ring, AMDGPU_MUX_OFFSET_TYPE_DE);
> }
> +
> +bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring)
> +{
> + if (!ring)
> + return false;
> +
> + if (ring->no_scheduler || !drm_sched_wqueue_ready(&ring->sched))
> + return false;
> +
> + return true;
> +}
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> index bbb53720a018..fe1a61eb6e4c 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> @@ -450,5 +450,5 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
> int amdgpu_ib_pool_init(struct amdgpu_device *adev);
> void amdgpu_ib_pool_fini(struct amdgpu_device *adev);
> int amdgpu_ib_ring_tests(struct amdgpu_device *adev);
> -
> +bool amdgpu_ring_sched_ready(struct amdgpu_ring *ring);
> #endif
> --
> 2.42.0
>
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2024-01-30 2:41 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-01-04 5:39 [PATCH] drm/amdgpu: Fix the warning info in mode1 reset Ma Jun
2024-01-04 15:00 ` Alex Deucher
-- strict thread matches above, loose matches on Subject: below --
2024-01-25 16:15 Alex Deucher
2024-01-29 15:33 ` Deucher, Alexander
2024-01-29 19:28 ` vitaly prosyak
2024-01-30 2:40 ` Ma, Jun
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox