回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list

Commit Message

Pan, Xinhui Nov. 9, 2021, 1:16 p.m. UTC

[AMD Official Use Only]

Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.

During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.

Comments

Christian König Nov. 9, 2021, 1:18 p.m. UTC | #1

Exactly that's the reason why we should have the double check in TTM 
I've mentioned in the other mail.

Christian.

Am 09.11.21 um 14:16 schrieb Pan, Xinhui:
> [AMD Official Use Only]
>
> Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.
>
> During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.
> ________________________________________
> 发件人: Pan, Xinhui <[e-mail]>
> 发送时间: 2021年11月9日 21:05
> 收件人: Koenig, Christian; [e-mail]
> 抄送: [e-mail]
> 主题: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Yes, a stable tag is needed. vulkan guys say 5.14 hit this issue too.
>
> I think that amdgpu_bo_move() does support copy from sysMem to sysMem correctly.
> maybe something below is needed.
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> index c83ef42ca702..aa63ae7ddf1e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> @@ -485,7 +485,8 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
>          }
>          if (old_mem->mem_type == TTM_PL_SYSTEM &&
>              (new_mem->mem_type == TTM_PL_TT ||
> -            new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
> +            new_mem->mem_type == AMDGPU_PL_PREEMPT ||
> +            new_mem->mem_type == TTM_PL_SYSTEM)) {
>                  ttm_bo_move_null(bo, new_mem);
>                  goto out;
>          }
>
> otherwise, amdgpu_move_blit() is called to do the system memory copy which use a wrong address.
>   206         /* Map only what can't be accessed directly */
>   207         if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) {
>   208                 *addr = amdgpu_ttm_domain_start(adev, mem->mem_type) +
>   209                         mm_cur->start;
>   210                 return 0;
>   211         }
>
> line 208, *addr is zero. So when amdgpu_copy_buffer submit job with such addr, page fault happens.
>
>
> ________________________________________
> 发件人: Koenig, Christian <[e-mail]>
> 发送时间: 2021年11月9日 20:35
> 收件人: Pan, Xinhui; [e-mail]
> 抄送: [e-mail]
> 主题: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Mhm, I'm not sure what the rational behind that is.
>
> Not moving the BO would make things less efficient, but should never
> cause a crash.
>
> Maybe we should add a CC: stable tag and push it to -fixes instead?
>
> Christian.
>
> Am 09.11.21 um 13:28 schrieb Pan, Xinhui:
>> [AMD Official Use Only]
>>
>> I hit vulkan cts test hang with navi23.
>>
>> dmesg says gmc page fault with address 0x0, 0x1000, 0x2000....
>> And some debug log also says amdgu copy one BO from system Domain to system Domain which is really weird.
>> ________________________________________
>> 发件人: Koenig, Christian <[e-mail]>
>> 发送时间: 2021年11月9日 20:20
>> 收件人: Pan, Xinhui; [e-mail]
>> 抄送: [e-mail]
>> 主题: Re: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Am 09.11.21 um 12:19 schrieb xinhui pan:
>>> After we move BO to a new memory region, we should put it to
>>> the new memory manager's lru list regardless we unlock the resv or not.
>>>
>>> Signed-off-by: xinhui pan <[e-mail]>
>> Interesting find, did you trigger that somehow or did you just stumbled
>> over it by reading the code?
>>
>> Patch is Reviewed-by: Christian König <[e-mail]>, I will
>> pick that up for drm-misc-next.
>>
>> Thanks,
>> Christian.
>>
>>> ---
>>>     drivers/gpu/drm/ttm/ttm_bo.c | 2 ++
>>>     1 file changed, 2 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
>>> index f1367107925b..e307004f0b28 100644
>>> --- a/drivers/gpu/drm/ttm/ttm_bo.c
>>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
>>> @@ -701,6 +701,8 @@ int ttm_mem_evict_first(struct ttm_device *bdev,
>>>         ret = ttm_bo_evict(bo, ctx);
>>>         if (locked)
>>>                 ttm_bo_unreserve(bo);
>>> +     else
>>> +             ttm_bo_move_to_lru_tail_unlocked(bo);
>>>
>>>         ttm_bo_put(bo);
>>>         return ret;

Pan, Xinhui Nov. 9, 2021, 1:41 p.m. UTC | #2

[AMD Official Use Only]

yes, a double check is needed.
how about change below.

As long as we detect such mismatch, it indicates another eviction is on going. return false here is reasonable.

@@ -1335,6 +1336,8 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo,
        struct dma_fence *f;
        int i;

+       if (bo->resource->mem_type != place->mem_type)
+               return false;
        /* Swapout? */
        if (bo->resource->mem_type == TTM_PL_SYSTEM)
                return true;

Christian König Nov. 9, 2021, 1:59 p.m. UTC | #3

In general the correct idea, but the wrong place to check that.

Calling amdgpu_ttm_bo_eviction_valuable() is only optional, but that 
check must be mandatory for correct operation.

This needs to be inside ttm_bo_evict_swapout_allowable().

Christian.

Am 09.11.21 um 14:41 schrieb Pan, Xinhui:
> [AMD Official Use Only]
>
> yes, a double check is needed.
> how about change below.
>
> As long as we detect such mismatch, it indicates another eviction is on going. return false here is reasonable.
>
> @@ -1335,6 +1336,8 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo,
>          struct dma_fence *f;
>          int i;
>
> +       if (bo->resource->mem_type != place->mem_type)
> +               return false;
>          /* Swapout? */
>          if (bo->resource->mem_type == TTM_PL_SYSTEM)
>                  return true;
>
> ________________________________________
> 发件人: Koenig, Christian <[e-mail]>
> 发送时间: 2021年11月9日 21:18
> 收件人: Pan, Xinhui; [e-mail]
> 抄送: [e-mail]
> 主题: Re: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Exactly that's the reason why we should have the double check in TTM
> I've mentioned in the other mail.
>
> Christian.
>
> Am 09.11.21 um 14:16 schrieb Pan, Xinhui:
>> [AMD Official Use Only]
>>
>> Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.
>>
>> During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.
>> ________________________________________
>> 发件人: Pan, Xinhui <[e-mail]>
>> 发送时间: 2021年11月9日 21:05
>> 收件人: Koenig, Christian; [e-mail]
>> 抄送: [e-mail]
>> 主题: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Yes, a stable tag is needed. vulkan guys say 5.14 hit this issue too.
>>
>> I think that amdgpu_bo_move() does support copy from sysMem to sysMem correctly.
>> maybe something below is needed.
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> index c83ef42ca702..aa63ae7ddf1e 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> @@ -485,7 +485,8 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
>>           }
>>           if (old_mem->mem_type == TTM_PL_SYSTEM &&
>>               (new_mem->mem_type == TTM_PL_TT ||
>> -            new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
>> +            new_mem->mem_type == AMDGPU_PL_PREEMPT ||
>> +            new_mem->mem_type == TTM_PL_SYSTEM)) {
>>                   ttm_bo_move_null(bo, new_mem);
>>                   goto out;
>>           }
>>
>> otherwise, amdgpu_move_blit() is called to do the system memory copy which use a wrong address.
>>    206         /* Map only what can't be accessed directly */
>>    207         if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) {
>>    208                 *addr = amdgpu_ttm_domain_start(adev, mem->mem_type) +
>>    209                         mm_cur->start;
>>    210                 return 0;
>>    211         }
>>
>> line 208, *addr is zero. So when amdgpu_copy_buffer submit job with such addr, page fault happens.
>>
>>
>> ________________________________________
>> 发件人: Koenig, Christian <[e-mail]>
>> 发送时间: 2021年11月9日 20:35
>> 收件人: Pan, Xinhui; [e-mail]
>> 抄送: [e-mail]
>> 主题: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Mhm, I'm not sure what the rational behind that is.
>>
>> Not moving the BO would make things less efficient, but should never
>> cause a crash.
>>
>> Maybe we should add a CC: stable tag and push it to -fixes instead?
>>
>> Christian.
>>
>> Am 09.11.21 um 13:28 schrieb Pan, Xinhui:
>>> [AMD Official Use Only]
>>>
>>> I hit vulkan cts test hang with navi23.
>>>
>>> dmesg says gmc page fault with address 0x0, 0x1000, 0x2000....
>>> And some debug log also says amdgu copy one BO from system Domain to system Domain which is really weird.
>>> ________________________________________
>>> 发件人: Koenig, Christian <[e-mail]>
>>> 发送时间: 2021年11月9日 20:20
>>> 收件人: Pan, Xinhui; [e-mail]
>>> 抄送: [e-mail]
>>> 主题: Re: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>>
>>> Am 09.11.21 um 12:19 schrieb xinhui pan:
>>>> After we move BO to a new memory region, we should put it to
>>>> the new memory manager's lru list regardless we unlock the resv or not.
>>>>
>>>> Signed-off-by: xinhui pan <[e-mail]>
>>> Interesting find, did you trigger that somehow or did you just stumbled
>>> over it by reading the code?
>>>
>>> Patch is Reviewed-by: Christian König <[e-mail]>, I will
>>> pick that up for drm-misc-next.
>>>
>>> Thanks,
>>> Christian.
>>>
>>>> ---
>>>>      drivers/gpu/drm/ttm/ttm_bo.c | 2 ++
>>>>      1 file changed, 2 insertions(+)
>>>>
>>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
>>>> index f1367107925b..e307004f0b28 100644
>>>> --- a/drivers/gpu/drm/ttm/ttm_bo.c
>>>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
>>>> @@ -701,6 +701,8 @@ int ttm_mem_evict_first(struct ttm_device *bdev,
>>>>          ret = ttm_bo_evict(bo, ctx);
>>>>          if (locked)
>>>>                  ttm_bo_unreserve(bo);
>>>> +     else
>>>> +             ttm_bo_move_to_lru_tail_unlocked(bo);
>>>>
>>>>          ttm_bo_put(bo);
>>>>          return ret;

Chen, Guchun Jan. 11, 2022, 8:47 a.m. UTC | #4

[Public]

Hi Christian,

Looks this patch still missed in 5.16 kernel. Is it intentional?
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/ttm/ttm_bo.c?h=v5.16

Regards,
Guchun

-----Original Message-----
From: amd-gfx <[e-mail]> On Behalf Of Pan, Xinhui
Sent: Tuesday, November 9, 2021 9:16 PM
To: Koenig, Christian <[e-mail]>; [e-mail]
Cc: [e-mail]
Subject: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list

[AMD Official Use Only]

[AMD Official Use Only]

Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.

During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.

Christian König Jan. 11, 2022, 11:27 a.m. UTC | #5

IIRC we have completely dropped this patch in favor of a check at a 
different place.

Regards,
Christian.

Am 11.01.22 um 09:47 schrieb Chen, Guchun:
> [Public]
>
> Hi Christian,
>
> Looks this patch still missed in 5.16 kernel. Is it intentional?
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/ttm/ttm_bo.c?h=v5.16
>
> Regards,
> Guchun
>
> -----Original Message-----
> From: amd-gfx <[e-mail]> On Behalf Of Pan, Xinhui
> Sent: Tuesday, November 9, 2021 9:16 PM
> To: Koenig, Christian <[e-mail]>; [e-mail]
> Cc: [e-mail]
> Subject: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> [AMD Official Use Only]
>
> [AMD Official Use Only]
>
> Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.
>
> During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.
> ________________________________________
> 发件人: Pan, Xinhui <[e-mail]>
> 发送时间: 2021年11月9日 21:05
> 收件人: Koenig, Christian; [e-mail]
> 抄送: [e-mail]
> 主题: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Yes, a stable tag is needed. vulkan guys say 5.14 hit this issue too.
>
> I think that amdgpu_bo_move() does support copy from sysMem to sysMem correctly.
> maybe something below is needed.
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> index c83ef42ca702..aa63ae7ddf1e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> @@ -485,7 +485,8 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
>          }
>          if (old_mem->mem_type == TTM_PL_SYSTEM &&
>              (new_mem->mem_type == TTM_PL_TT ||
> -            new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
> +            new_mem->mem_type == AMDGPU_PL_PREEMPT ||
> +            new_mem->mem_type == TTM_PL_SYSTEM)) {
>                  ttm_bo_move_null(bo, new_mem);
>                  goto out;
>          }
>
> otherwise, amdgpu_move_blit() is called to do the system memory copy which use a wrong address.
>   206         /* Map only what can't be accessed directly */
>   207         if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) {
>   208                 *addr = amdgpu_ttm_domain_start(adev, mem->mem_type) +
>   209                         mm_cur->start;
>   210                 return 0;
>   211         }
>
> line 208, *addr is zero. So when amdgpu_copy_buffer submit job with such addr, page fault happens.
>
>
> ________________________________________
> 发件人: Koenig, Christian <[e-mail]>
> 发送时间: 2021年11月9日 20:35
> 收件人: Pan, Xinhui; [e-mail]
> 抄送: [e-mail]
> 主题: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Mhm, I'm not sure what the rational behind that is.
>
> Not moving the BO would make things less efficient, but should never cause a crash.
>
> Maybe we should add a CC: stable tag and push it to -fixes instead?
>
> Christian.
>
> Am 09.11.21 um 13:28 schrieb Pan, Xinhui:
>> [AMD Official Use Only]
>>
>> I hit vulkan cts test hang with navi23.
>>
>> dmesg says gmc page fault with address 0x0, 0x1000, 0x2000....
>> And some debug log also says amdgu copy one BO from system Domain to system Domain which is really weird.
>> ________________________________________
>> 发件人: Koenig, Christian <[e-mail]>
>> 发送时间: 2021年11月9日 20:20
>> 收件人: Pan, Xinhui; [e-mail]
>> 抄送: [e-mail]
>> 主题: Re: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Am 09.11.21 um 12:19 schrieb xinhui pan:
>>> After we move BO to a new memory region, we should put it to the new
>>> memory manager's lru list regardless we unlock the resv or not.
>>>
>>> Signed-off-by: xinhui pan <[e-mail]>
>> Interesting find, did you trigger that somehow or did you just
>> stumbled over it by reading the code?
>>
>> Patch is Reviewed-by: Christian König <[e-mail]>, I
>> will pick that up for drm-misc-next.
>>
>> Thanks,
>> Christian.
>>
>>> ---
>>>     drivers/gpu/drm/ttm/ttm_bo.c | 2 ++
>>>     1 file changed, 2 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c
>>> b/drivers/gpu/drm/ttm/ttm_bo.c index f1367107925b..e307004f0b28
>>> 100644
>>> --- a/drivers/gpu/drm/ttm/ttm_bo.c
>>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
>>> @@ -701,6 +701,8 @@ int ttm_mem_evict_first(struct ttm_device *bdev,
>>>         ret = ttm_bo_evict(bo, ctx);
>>>         if (locked)
>>>                 ttm_bo_unreserve(bo);
>>> +     else
>>> +             ttm_bo_move_to_lru_tail_unlocked(bo);
>>>
>>>         ttm_bo_put(bo);
>>>         return ret;

Chen, Guchun Jan. 12, 2022, 2:19 a.m. UTC | #6

[Public]

Hi Christian,

My BAD, I checked that discussion history of this just now. So If I read it correctly, the double check at a different place to skip evict is: " drm/ttm: Double check mem_type of BO while eviction"? It is in 5.16 kernel.

Regards,
Guchun

-----Original Message-----
From: Christian König <[e-mail]> 
Sent: Tuesday, January 11, 2022 7:27 PM
To: Chen, Guchun <[e-mail]>; Pan, Xinhui <[e-mail]>; Koenig, Christian <[e-mail]>; [e-mail]
Cc: [e-mail]
Subject: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list

IIRC we have completely dropped this patch in favor of a check at a different place.

Regards,
Christian.

Am 11.01.22 um 09:47 schrieb Chen, Guchun:
> [Public]
>
> Hi Christian,
>
> Looks this patch still missed in 5.16 kernel. Is it intentional?
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.
> kernel.org%2Fpub%2Fscm%2Flinux%2Fkernel%2Fgit%2Ftorvalds%2Flinux.git%2
> Ftree%2Fdrivers%2Fgpu%2Fdrm%2Fttm%2Fttm_bo.c%3Fh%3Dv5.16&amp;data=04%7
> C01%7CGuchun.Chen%40amd.com%7Cf3b7f4971dc8405b0c2908d9d4f55547%7C3dd89
> 61fe4884e608e11a82d994e183d%7C0%7C0%7C637774972434004088%7CUnknown%7CT
> WFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI
> 6Mn0%3D%7C3000&amp;sdata=vbuBPHO40J2HGt7abzfzC0nC1DQa62qal5S6TXBRj4w%3
> D&amp;reserved=0
>
> Regards,
> Guchun
>
> -----Original Message-----
> From: amd-gfx <[e-mail]> On Behalf Of 
> Pan, Xinhui
> Sent: Tuesday, November 9, 2021 9:16 PM
> To: Koenig, Christian <[e-mail]>; 
> [e-mail]
> Cc: [e-mail]
> Subject: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru 
> list
>
> [AMD Official Use Only]
>
> [AMD Official Use Only]
>
> Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.
>
> During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.
> ________________________________________
> 发件人: Pan, Xinhui <[e-mail]>
> 发送时间: 2021年11月9日 21:05
> 收件人: Koenig, Christian; [e-mail]
> 抄送: [e-mail]
> 主题: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Yes, a stable tag is needed. vulkan guys say 5.14 hit this issue too.
>
> I think that amdgpu_bo_move() does support copy from sysMem to sysMem correctly.
> maybe something below is needed.
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c 
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> index c83ef42ca702..aa63ae7ddf1e 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
> @@ -485,7 +485,8 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
>          }
>          if (old_mem->mem_type == TTM_PL_SYSTEM &&
>              (new_mem->mem_type == TTM_PL_TT ||
> -            new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
> +            new_mem->mem_type == AMDGPU_PL_PREEMPT ||
> +            new_mem->mem_type == TTM_PL_SYSTEM)) {
>                  ttm_bo_move_null(bo, new_mem);
>                  goto out;
>          }
>
> otherwise, amdgpu_move_blit() is called to do the system memory copy which use a wrong address.
>   206         /* Map only what can't be accessed directly */
>   207         if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) {
>   208                 *addr = amdgpu_ttm_domain_start(adev, mem->mem_type) +
>   209                         mm_cur->start;
>   210                 return 0;
>   211         }
>
> line 208, *addr is zero. So when amdgpu_copy_buffer submit job with such addr, page fault happens.
>
>
> ________________________________________
> 发件人: Koenig, Christian <[e-mail]>
> 发送时间: 2021年11月9日 20:35
> 收件人: Pan, Xinhui; [e-mail]
> 抄送: [e-mail]
> 主题: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> Mhm, I'm not sure what the rational behind that is.
>
> Not moving the BO would make things less efficient, but should never cause a crash.
>
> Maybe we should add a CC: stable tag and push it to -fixes instead?
>
> Christian.
>
> Am 09.11.21 um 13:28 schrieb Pan, Xinhui:
>> [AMD Official Use Only]
>>
>> I hit vulkan cts test hang with navi23.
>>
>> dmesg says gmc page fault with address 0x0, 0x1000, 0x2000....
>> And some debug log also says amdgu copy one BO from system Domain to system Domain which is really weird.
>> ________________________________________
>> 发件人: Koenig, Christian <[e-mail]>
>> 发送时间: 2021年11月9日 20:20
>> 收件人: Pan, Xinhui; [e-mail]
>> 抄送: [e-mail]
>> 主题: Re: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Am 09.11.21 um 12:19 schrieb xinhui pan:
>>> After we move BO to a new memory region, we should put it to the new 
>>> memory manager's lru list regardless we unlock the resv or not.
>>>
>>> Signed-off-by: xinhui pan <[e-mail]>
>> Interesting find, did you trigger that somehow or did you just 
>> stumbled over it by reading the code?
>>
>> Patch is Reviewed-by: Christian König <[e-mail]>, I 
>> will pick that up for drm-misc-next.
>>
>> Thanks,
>> Christian.
>>
>>> ---
>>>     drivers/gpu/drm/ttm/ttm_bo.c | 2 ++
>>>     1 file changed, 2 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c 
>>> b/drivers/gpu/drm/ttm/ttm_bo.c index f1367107925b..e307004f0b28
>>> 100644
>>> --- a/drivers/gpu/drm/ttm/ttm_bo.c
>>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
>>> @@ -701,6 +701,8 @@ int ttm_mem_evict_first(struct ttm_device *bdev,
>>>         ret = ttm_bo_evict(bo, ctx);
>>>         if (locked)
>>>                 ttm_bo_unreserve(bo);
>>> +     else
>>> +             ttm_bo_move_to_lru_tail_unlocked(bo);
>>>
>>>         ttm_bo_put(bo);
>>>         return ret;

Christian König Jan. 12, 2022, 7 a.m. UTC | #7

Yeah, that should probably be the right one.

Christian.

Am 12.01.22 um 03:19 schrieb Chen, Guchun:
> [Public]
>
> Hi Christian,
>
> My BAD, I checked that discussion history of this just now. So If I read it correctly, the double check at a different place to skip evict is: " drm/ttm: Double check mem_type of BO while eviction"? It is in 5.16 kernel.
>
> Regards,
> Guchun
>
> -----Original Message-----
> From: Christian König <[e-mail]>
> Sent: Tuesday, January 11, 2022 7:27 PM
> To: Chen, Guchun <[e-mail]>; Pan, Xinhui <[e-mail]>; Koenig, Christian <[e-mail]>; [e-mail]
> Cc: [e-mail]
> Subject: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>
> IIRC we have completely dropped this patch in favor of a check at a different place.
>
> Regards,
> Christian.
>
> Am 11.01.22 um 09:47 schrieb Chen, Guchun:
>> [Public]
>>
>> Hi Christian,
>>
>> Looks this patch still missed in 5.16 kernel. Is it intentional?
>> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.
>> kernel.org%2Fpub%2Fscm%2Flinux%2Fkernel%2Fgit%2Ftorvalds%2Flinux.git%2
>> Ftree%2Fdrivers%2Fgpu%2Fdrm%2Fttm%2Fttm_bo.c%3Fh%3Dv5.16&amp;data=04%7
>> C01%7CGuchun.Chen%40amd.com%7Cf3b7f4971dc8405b0c2908d9d4f55547%7C3dd89
>> 61fe4884e608e11a82d994e183d%7C0%7C0%7C637774972434004088%7CUnknown%7CT
>> WFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI
>> 6Mn0%3D%7C3000&amp;sdata=vbuBPHO40J2HGt7abzfzC0nC1DQa62qal5S6TXBRj4w%3
>> D&amp;reserved=0
>>
>> Regards,
>> Guchun
>>
>> -----Original Message-----
>> From: amd-gfx <[e-mail]> On Behalf Of
>> Pan, Xinhui
>> Sent: Tuesday, November 9, 2021 9:16 PM
>> To: Koenig, Christian <[e-mail]>;
>> [e-mail]
>> Cc: [e-mail]
>> Subject: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru
>> list
>>
>> [AMD Official Use Only]
>>
>> [AMD Official Use Only]
>>
>> Actually this patch does not totally fix the mismatch of lru list with mem_type as mem_type is changed in ->move() and lru list is changed after that.
>>
>> During this small period, another eviction could still happed and evict this mismatched BO from sMam(say, its lru list is on vram domain) to sMem.
>> ________________________________________
>> 发件人: Pan, Xinhui <[e-mail]>
>> 发送时间: 2021年11月9日 21:05
>> 收件人: Koenig, Christian; [e-mail]
>> 抄送: [e-mail]
>> 主题: 回复: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Yes, a stable tag is needed. vulkan guys say 5.14 hit this issue too.
>>
>> I think that amdgpu_bo_move() does support copy from sysMem to sysMem correctly.
>> maybe something below is needed.
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> index c83ef42ca702..aa63ae7ddf1e 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
>> @@ -485,7 +485,8 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
>>           }
>>           if (old_mem->mem_type == TTM_PL_SYSTEM &&
>>               (new_mem->mem_type == TTM_PL_TT ||
>> -            new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
>> +            new_mem->mem_type == AMDGPU_PL_PREEMPT ||
>> +            new_mem->mem_type == TTM_PL_SYSTEM)) {
>>                   ttm_bo_move_null(bo, new_mem);
>>                   goto out;
>>           }
>>
>> otherwise, amdgpu_move_blit() is called to do the system memory copy which use a wrong address.
>>    206         /* Map only what can't be accessed directly */
>>    207         if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) {
>>    208                 *addr = amdgpu_ttm_domain_start(adev, mem->mem_type) +
>>    209                         mm_cur->start;
>>    210                 return 0;
>>    211         }
>>
>> line 208, *addr is zero. So when amdgpu_copy_buffer submit job with such addr, page fault happens.
>>
>>
>> ________________________________________
>> 发件人: Koenig, Christian <[e-mail]>
>> 发送时间: 2021年11月9日 20:35
>> 收件人: Pan, Xinhui; [e-mail]
>> 抄送: [e-mail]
>> 主题: Re: 回复: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>
>> Mhm, I'm not sure what the rational behind that is.
>>
>> Not moving the BO would make things less efficient, but should never cause a crash.
>>
>> Maybe we should add a CC: stable tag and push it to -fixes instead?
>>
>> Christian.
>>
>> Am 09.11.21 um 13:28 schrieb Pan, Xinhui:
>>> [AMD Official Use Only]
>>>
>>> I hit vulkan cts test hang with navi23.
>>>
>>> dmesg says gmc page fault with address 0x0, 0x1000, 0x2000....
>>> And some debug log also says amdgu copy one BO from system Domain to system Domain which is really weird.
>>> ________________________________________
>>> 发件人: Koenig, Christian <[e-mail]>
>>> 发送时间: 2021年11月9日 20:20
>>> 收件人: Pan, Xinhui; [e-mail]
>>> 抄送: [e-mail]
>>> 主题: Re: [PATCH] drm/ttm: Put BO in its memory manager's lru list
>>>
>>> Am 09.11.21 um 12:19 schrieb xinhui pan:
>>>> After we move BO to a new memory region, we should put it to the new
>>>> memory manager's lru list regardless we unlock the resv or not.
>>>>
>>>> Signed-off-by: xinhui pan <[e-mail]>
>>> Interesting find, did you trigger that somehow or did you just
>>> stumbled over it by reading the code?
>>>
>>> Patch is Reviewed-by: Christian König <[e-mail]>, I
>>> will pick that up for drm-misc-next.
>>>
>>> Thanks,
>>> Christian.
>>>
>>>> ---
>>>>      drivers/gpu/drm/ttm/ttm_bo.c | 2 ++
>>>>      1 file changed, 2 insertions(+)
>>>>
>>>> diff --git a/drivers/gpu/drm/ttm/ttm_bo.c
>>>> b/drivers/gpu/drm/ttm/ttm_bo.c index f1367107925b..e307004f0b28
>>>> 100644
>>>> --- a/drivers/gpu/drm/ttm/ttm_bo.c
>>>> +++ b/drivers/gpu/drm/ttm/ttm_bo.c
>>>> @@ -701,6 +701,8 @@ int ttm_mem_evict_first(struct ttm_device *bdev,
>>>>          ret = ttm_bo_evict(bo, ctx);
>>>>          if (locked)
>>>>                  ttm_bo_unreserve(bo);
>>>> +     else
>>>> +             ttm_bo_move_to_lru_tail_unlocked(bo);
>>>>
>>>>          ttm_bo_put(bo);
>>>>          return ret;

Patch

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index c83ef42ca702..aa63ae7ddf1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -485,7 +485,8 @@  static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
        }
        if (old_mem->mem_type == TTM_PL_SYSTEM &&
            (new_mem->mem_type == TTM_PL_TT ||
-            new_mem->mem_type == AMDGPU_PL_PREEMPT)) {
+            new_mem->mem_type == AMDGPU_PL_PREEMPT ||
+            new_mem->mem_type == TTM_PL_SYSTEM)) {
                ttm_bo_move_null(bo, new_mem);
                goto out;
        }