Skip to contentSkip to conversation

drm/amdgpu/sriov: give 8s for recover vram under RUNTIME

PATCH In review Emily Deng 2 participants 3 messages Aug 7, 2018 – Aug 8, 2018 2 messages not in the archive

Subsystems (2)
lore
Share & tools
lore

Emily Deng

author
#0 lore raw

PATCH 1/1 drm/amdgpu/sriov: give 8s for recover vram under RUNTIME

Under runtime, the wait fence time could be quite long when other VFs are in exclusive mode.

SWDEV-161490

Change-Id Ifc32d56ca7fde01b1f4fe2b0db6959b51909008a
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +1 -1
index 1d933db..ef82ad1 100644
@@ -3124,7 +3124,7 @@ static int amdgpu_device_handle_vram_lost(struct amdgpu_device *adev)
 	long tmo;
 
 	if (amdgpu_sriov_runtime(adev))
-		tmo = msecs_to_jiffies(amdgpu_lockup_timeout);
+		tmo = msecs_to_jiffies(8000);
 	else
 		tmo = msecs_to_jiffies(100);
 
Signature · 6 lines
2.7.4

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): 1533637327-18411-1-git-send-email-Emily.Deng-5C7GfCeVMHo@public.gmane.org

#2 lore raw
Quoted 8 lines · Emily Deng

Under runtime, the wait fence time could be quite long when other VFs are in exclusive mode.

SWDEV-161490

Change-Id Ifc32d56ca7fde01b1f4fe2b0db6959b51909008a

Seems pretty long. Is this value based on any evidence (e.g., worse case length of time slices, etc.) or just a long value that happens to work? Might be nice to provide a bit more context in the commit message. E.g., extend the timeout for recovering vram bos from shadows on sr-iov to cover the worst case scenario for timeslices and VFs.

Alex

Quoted 20 lines
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +1 -2
index 1d933db..ef82ad1 100644
@@ -3124,7 +3124,7 @@ static int amdgpu_device_handle_vram_lost(struct amdgpu_device *adev)
        long tmo;
 
        if (amdgpu_sriov_runtime(adev))
-               tmo = msecs_to_jiffies(amdgpu_lockup_timeout);
+               tmo = msecs_to_jiffies(8000);
        else
                tmo = msecs_to_jiffies(100);
 
--

2.7.4

Quoted 4 lines

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

A message that is not in the archive (sent off-list or not yet received): CADnq5_NyWQWYCbmrxoKe35AT0qQxp9knZ6F7aApbROe_g_Dcgw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

#4 lore raw

Hi Alex,
Thanks for your review.
The host gim driver gives every VF fullaccess_timeout is 3000, so for 4 vf, the worst case is 3*3000(9000), if the vf number is more than 4,
then the time will be longer, so 8000 is not the worst time.
But with using 8000, it will pass all the TDR test, so it is Ok. Will send a patch to modify the commit message as your suggestion.

Best wishes
Emily Deng

-----Original Message-----
From: Alex Deucher <[email protected]>
Sent: Wednesday, August 8, 2018 12:42 AM
To: Deng, Emily <[email protected]>
Cc: amd-gfx list <[email protected]>; Liu, Monk <[email protected]>
Subject: Re: [PATCH] drm/amdgpu/sriov: give 8s for recover vram under RUNTIME

Quoted 8 lines · Emily Deng

Under runtime, the wait fence time could be quite long when other VFs are in exclusive mode.

SWDEV-161490

Change-Id Ifc32d56ca7fde01b1f4fe2b0db6959b51909008a

Seems pretty long. Is this value based on any evidence (e.g., worse case length of time slices, etc.) or just a long value that happens to work? Might be nice to provide a bit more context in the commit message. E.g., extend the timeout for recovering vram bos from shadows on sr-iov to cover the worst case scenario for timeslices and VFs.

Alex

Quoted 21 lines
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +1 -2
index 1d933db..ef82ad1 100644
@@ -3124,7 +3124,7 @@ static int amdgpu_device_handle_vram_lost(struct amdgpu_device *adev)
        long tmo;
 
        if (amdgpu_sriov_runtime(adev))
-               tmo = msecs_to_jiffies(amdgpu_lockup_timeout);
+               tmo = msecs_to_jiffies(8000);
        else
                tmo = msecs_to_jiffies(100);
 
--

2.7.4

Quoted 4 lines

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx