From: "Christian König" <ckoenig.leichtzumerken@gmail.com>
To: "Daniel Vetter" <daniel@ffwll.ch>,
"Christian König" <christian.koenig@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>,
dri-devel@lists.freedesktop.org, amd-gfx@lists.freedesktop.org
Subject: Re: [PATCH 8/9] drm/ttm: Don't count pages in SG BOs against pages_limit
Date: Wed, 14 Apr 2021 12:49:17 +0200 [thread overview]
Message-ID: <7ac3a5e9-abf6-6ea5-af21-25c04e984d57@gmail.com> (raw)
In-Reply-To: <YHbDQcr6Yl1Lljep@phenom.ffwll.local>
Am 14.04.21 um 12:26 schrieb Daniel Vetter:
> On Wed, Apr 14, 2021 at 11:19:41AM +0200, Christian König wrote:
>> Am 14.04.21 um 11:15 schrieb Daniel Vetter:
>>> On Wed, Apr 14, 2021 at 08:51:51AM +0200, Christian König wrote:
>>>> Am 14.04.21 um 08:48 schrieb Felix Kuehling:
>>>>> Pages in SG BOs were not allocated by TTM. So don't count them against
>>>>> TTM's pages limit.
>>>>>
>>>>> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
>>>> Reviewed-by: Christian König <christian.koenig@amd.com>
>>>>
>>>> Going to pick that one up for inclusion in drm-misc-next.
>>> See my other email, but why do we need this? A bit more explanation is imo
>>> needed here at least, since we still need to guarantee that allocations
>>> don't over the limit in total for all gpu buffers together. At least until
>>> the shrinker has landed.
>>>
>>> And this here just opens up the barn door without any explanation why it's
>>> ok.
>> The SG based BOs might not even be backed by pages. E.g. exported VRAM.
>>
>> So either they are exported by a driver which should have accounted for the
>> allocation, exported by TTM which already did the accounting or doesn't even
>> point to pages at all.
>>
>> This is really a bug fix to recreate the behavior we had before moving the
>> accounting to this place.
> Throw that into the commit message and a-b: me. Ideally with a Fixes: line
> or so pointing at the offending commit that broke stuff. Commit messages
> should really go into more detail when there's an entire story behind a
> small change like this one.
Sorry I though that this would be obvious :)
I've already pushed the patch in the morning, but going to keep that in
mind for the next time.
Christian.
> -Daniel
>
>> Christian.
>>
>>> -Daniel
>>>
>>>> Regards,
>>>> Christian.
>>>>
>>>>> ---
>>>>> drivers/gpu/drm/ttm/ttm_tt.c | 27 ++++++++++++++++++---------
>>>>> 1 file changed, 18 insertions(+), 9 deletions(-)
>>>>>
>>>>> diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
>>>>> index 5d8820725b75..e8b8c3257392 100644
>>>>> --- a/drivers/gpu/drm/ttm/ttm_tt.c
>>>>> +++ b/drivers/gpu/drm/ttm/ttm_tt.c
>>>>> @@ -317,9 +317,12 @@ int ttm_tt_populate(struct ttm_device *bdev,
>>>>> if (ttm_tt_is_populated(ttm))
>>>>> return 0;
>>>>> - atomic_long_add(ttm->num_pages, &ttm_pages_allocated);
>>>>> - if (bdev->pool.use_dma32)
>>>>> - atomic_long_add(ttm->num_pages, &ttm_dma32_pages_allocated);
>>>>> + if (!(ttm->page_flags & TTM_PAGE_FLAG_SG)) {
>>>>> + atomic_long_add(ttm->num_pages, &ttm_pages_allocated);
>>>>> + if (bdev->pool.use_dma32)
>>>>> + atomic_long_add(ttm->num_pages,
>>>>> + &ttm_dma32_pages_allocated);
>>>>> + }
>>>>> while (atomic_long_read(&ttm_pages_allocated) > ttm_pages_limit ||
>>>>> atomic_long_read(&ttm_dma32_pages_allocated) >
>>>>> @@ -350,9 +353,12 @@ int ttm_tt_populate(struct ttm_device *bdev,
>>>>> return 0;
>>>>> error:
>>>>> - atomic_long_sub(ttm->num_pages, &ttm_pages_allocated);
>>>>> - if (bdev->pool.use_dma32)
>>>>> - atomic_long_sub(ttm->num_pages, &ttm_dma32_pages_allocated);
>>>>> + if (!(ttm->page_flags & TTM_PAGE_FLAG_SG)) {
>>>>> + atomic_long_sub(ttm->num_pages, &ttm_pages_allocated);
>>>>> + if (bdev->pool.use_dma32)
>>>>> + atomic_long_sub(ttm->num_pages,
>>>>> + &ttm_dma32_pages_allocated);
>>>>> + }
>>>>> return ret;
>>>>> }
>>>>> EXPORT_SYMBOL(ttm_tt_populate);
>>>>> @@ -382,9 +388,12 @@ void ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm)
>>>>> else
>>>>> ttm_pool_free(&bdev->pool, ttm);
>>>>> - atomic_long_sub(ttm->num_pages, &ttm_pages_allocated);
>>>>> - if (bdev->pool.use_dma32)
>>>>> - atomic_long_sub(ttm->num_pages, &ttm_dma32_pages_allocated);
>>>>> + if (!(ttm->page_flags & TTM_PAGE_FLAG_SG)) {
>>>>> + atomic_long_sub(ttm->num_pages, &ttm_pages_allocated);
>>>>> + if (bdev->pool.use_dma32)
>>>>> + atomic_long_sub(ttm->num_pages,
>>>>> + &ttm_dma32_pages_allocated);
>>>>> + }
>>>>> ttm->page_flags &= ~TTM_PAGE_FLAG_PRIV_POPULATED;
>>>>> }
>>>> _______________________________________________
>>>> dri-devel mailing list
>>>> dri-devel@lists.freedesktop.org
>>>> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.freedesktop.org%2Fmailman%2Flistinfo%2Fdri-devel&data=04%7C01%7Cchristian.koenig%40amd.com%7C3075d7fd16644322a13608d8ff25d59b%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637539885255795187%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=KOnHA1CbNFjjMZR2rgHmGkH%2B7C84YCtA6u9V1wBAay4%3D&reserved=0
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
next prev parent reply other threads:[~2021-04-14 10:49 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-14 6:47 [PATCH 0/9] Implement multi-GPU DMA mappings for KFD Felix Kuehling
2021-04-14 6:47 ` [PATCH 1/9] drm/amdgpu: Rename kfd_bo_va_list to kfd_mem_attachment Felix Kuehling
2021-04-14 6:47 ` [PATCH 2/9] drm/amdgpu: Keep a bo-reference per-attachment Felix Kuehling
2021-04-14 6:47 ` [PATCH 3/9] drm/amdgpu: Simplify AQL queue mapping Felix Kuehling
2021-04-14 6:47 ` [PATCH 4/9] drm/amdgpu: Add multi-GPU DMA mapping helpers Felix Kuehling
2021-04-14 6:48 ` [PATCH 5/9] drm/amdgpu: DMA map/unmap when updating GPU mappings Felix Kuehling
2021-04-14 6:48 ` [PATCH 6/9] drm/amdgpu: Move kfd_mem_attach outside reservation Felix Kuehling
2021-04-14 6:48 ` [PATCH 7/9] drm/amdgpu: Add DMA mapping of GTT BOs Felix Kuehling
2021-04-14 6:48 ` [PATCH 8/9] drm/ttm: Don't count pages in SG BOs against pages_limit Felix Kuehling
2021-04-14 6:51 ` Christian König
2021-04-14 9:15 ` Daniel Vetter
2021-04-14 9:19 ` Christian König
2021-04-14 10:26 ` Daniel Vetter
2021-04-14 10:49 ` Christian König [this message]
2021-04-14 12:25 ` Daniel Vetter
2021-04-14 12:43 ` Christian König
2021-04-14 12:47 ` Daniel Vetter
2021-04-14 12:49 ` Christian König
2021-04-14 14:41 ` Felix Kuehling
2021-04-14 6:48 ` [PATCH 9/9] drm/amdgpu: Lock the attached dmabuf in unpopulate Felix Kuehling
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=7ac3a5e9-abf6-6ea5-af21-25c04e984d57@gmail.com \
--to=ckoenig.leichtzumerken@gmail.com \
--cc=Felix.Kuehling@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
/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;
as well as URLs for NNTP newsgroup(s).