[PATCH] Revert "drm/amdgpu: Increase KIQ invalidate_tlbs timeout"

[PATCH] Revert "drm/amdgpu: Increase KIQ invalidate_tlbs timeout"

[Jay Cornwall]

This reverts commit fdb90033846e2f23dfaaa01dc47fec7b94704d0e.

Reportedly causing unknown issue in memory management code:

[  128.047288] amdgpu 0000:65:00.0: amdgpu: Failed to map peer:0000:46:00.0 mem_domain:2
[...]
[  137.815340] WARNING: CPU: 81 PID: 1006 at drivers/gpu/drm/ttm/ttm_bo.c:613 ttm_bo_unpin+0x7e/0x90 [ttm]

Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Cc: Kent Russell <kent.russell@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 16 ++++------------
 2 files changed, 5 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 3cdb5f8325aa..ffca74a476da 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -356,6 +356,7 @@ enum amdgpu_kiq_irq {
 	AMDGPU_CP_KIQ_IRQ_DRIVER0 = 0,
 	AMDGPU_CP_KIQ_IRQ_LAST
 };
+#define SRIOV_USEC_TIMEOUT  1200000 /* wait 12 * 100ms for SRIOV */
 #define MAX_KIQ_REG_WAIT       5000 /* in usecs, 5ms */
 #define MAX_KIQ_REG_BAILOUT_INTERVAL   5 /* in msecs, 5ms */
 #define MAX_KIQ_REG_TRY 1000
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index c46e36a0cd9c..464625282872 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -699,10 +699,12 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid,
 				   uint32_t flush_type, bool all_hub,
 				   uint32_t inst)
 {
+	u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT :
+		adev->usec_timeout;
 	struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst];
 	unsigned int ndw;
-	int r, cnt = 0;
+	int r;
 	uint32_t seq;
 
 	/*
@@ -759,17 +761,7 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid,
 
 		amdgpu_ring_commit(ring);
 		spin_unlock(&adev->gfx.kiq[inst].ring_lock);
-
-		r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
-
-		might_sleep();
-		while (r < 1 && cnt++ < MAX_KIQ_REG_TRY &&
-		       !amdgpu_reset_pending(adev->reset_domain)) {
-			msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
-			r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
-		}
-
-		if (cnt > MAX_KIQ_REG_TRY) {
+		if (amdgpu_fence_wait_polling(ring, seq, usec_timeout) < 1) {
 			dev_err(adev->dev, "timeout waiting for kiq fence\n");
 			r = -ETIME;
 		}
-- 
2.34.1

RE: [PATCH] Revert "drm/amdgpu: Increase KIQ invalidate_tlbs timeout"

[Russell, Kent]

[Public]

Reviewed-by: Kent Russell <kent.russell@amd.com>



> -----Original Message-----
> From: Cornwall, Jay <Jay.Cornwall@amd.com>
> Sent: Friday, March 28, 2025 12:05 AM
> To: amd-gfx@lists.freedesktop.org
> Cc: Cornwall, Jay <Jay.Cornwall@amd.com>; Russell, Kent
> <Kent.Russell@amd.com>
> Subject: [PATCH] Revert "drm/amdgpu: Increase KIQ invalidate_tlbs timeout"
>
> This reverts commit fdb90033846e2f23dfaaa01dc47fec7b94704d0e.
>
> Reportedly causing unknown issue in memory management code:
>
> [  128.047288] amdgpu 0000:65:00.0: amdgpu: Failed to map peer:0000:46:00.0
> mem_domain:2
> [...]
> [  137.815340] WARNING: CPU: 81 PID: 1006 at drivers/gpu/drm/ttm/ttm_bo.c:613
> ttm_bo_unpin+0x7e/0x90 [ttm]
>
> Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
> Cc: Kent Russell <kent.russell@amd.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 +
>  drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 16 ++++------------
>  2 files changed, 5 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> index 3cdb5f8325aa..ffca74a476da 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
> @@ -356,6 +356,7 @@ enum amdgpu_kiq_irq {
>       AMDGPU_CP_KIQ_IRQ_DRIVER0 = 0,
>       AMDGPU_CP_KIQ_IRQ_LAST
>  };
> +#define SRIOV_USEC_TIMEOUT  1200000 /* wait 12 * 100ms for SRIOV */
>  #define MAX_KIQ_REG_WAIT       5000 /* in usecs, 5ms */
>  #define MAX_KIQ_REG_BAILOUT_INTERVAL   5 /* in msecs, 5ms */
>  #define MAX_KIQ_REG_TRY 1000
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> index c46e36a0cd9c..464625282872 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
> @@ -699,10 +699,12 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct
> amdgpu_device *adev, uint16_t pasid,
>                                  uint32_t flush_type, bool all_hub,
>                                  uint32_t inst)
>  {
> +     u32 usec_timeout = amdgpu_sriov_vf(adev) ? SRIOV_USEC_TIMEOUT :
> +             adev->usec_timeout;
>       struct amdgpu_ring *ring = &adev->gfx.kiq[inst].ring;
>       struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst];
>       unsigned int ndw;
> -     int r, cnt = 0;
> +     int r;
>       uint32_t seq;
>
>       /*
> @@ -759,17 +761,7 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct
> amdgpu_device *adev, uint16_t pasid,
>
>               amdgpu_ring_commit(ring);
>               spin_unlock(&adev->gfx.kiq[inst].ring_lock);
> -
> -             r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
> -
> -             might_sleep();
> -             while (r < 1 && cnt++ < MAX_KIQ_REG_TRY &&
> -                    !amdgpu_reset_pending(adev->reset_domain)) {
> -                     msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
> -                     r = amdgpu_fence_wait_polling(ring, seq,
> MAX_KIQ_REG_WAIT);
> -             }
> -
> -             if (cnt > MAX_KIQ_REG_TRY) {
> +             if (amdgpu_fence_wait_polling(ring, seq, usec_timeout) < 1) {
>                       dev_err(adev->dev, "timeout waiting for kiq fence\n");
>                       r = -ETIME;
>               }
> --
> 2.34.1