From: "Christian König" <christian.koenig@amd.com>
To: Lang Yu <lang.yu@amd.com>, amd-gfx@lists.freedesktop.org
Cc: Felix Kuehling <Felix.Kuehling@amd.com>,
Christian K nig <C3B6christian.koenig@amd.com>,
Huang Rui <ray.huang@amd.com>
Subject: Re: [PATCH] drm/kfd: fix ttm_bo_release warning
Date: Thu, 23 Sep 2021 13:39:34 +0200 [thread overview]
Message-ID: <063d904f-4083-f5db-b3d0-eb4e080b055f@amd.com> (raw)
In-Reply-To: <20210923094405.144613-1-lang.yu@amd.com>
Am 23.09.21 um 11:44 schrieb Lang Yu:
> If a BO is pinned, unpin it before freeing it.
>
> Call Trace:
> ttm_bo_put+0x30/0x50 [ttm]
> amdgpu_bo_unref+0x1e/0x30 [amdgpu]
> amdgpu_gem_object_free+0x34/0x50 [amdgpu]
> drm_gem_object_free+0x1d/0x30 [drm]
> amdgpu_amdkfd_gpuvm_free_memory_of_gpu+0x31f/0x3a0 [amdgpu]
> kfd_process_device_free_bos+0xa3/0xf0 [amdgpu]
> kfd_process_wq_release+0x224/0x2e0 [amdgpu]
> process_one_work+0x220/0x3c0
> worker_thread+0x4d/0x3f0
> kthread+0x114/0x150
> process_one_work+0x3c0/0x3c0
> kthread_park+0x90/0x90
> ret_from_fork+0x22/0x30
>
> Signed-off-by: Lang Yu <lang.yu@amd.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> index 2d6b2d77b738..7e693b064072 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
> @@ -1567,6 +1567,9 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu(
> pr_debug("Release VA 0x%llx - 0x%llx\n", mem->va,
> mem->va + bo_size * (1 + mem->aql_queue));
>
> + if (mem->bo->tbo.pin_count)
> + amdgpu_bo_unpin(mem->bo);
> +
NAK, using mem->bo->tbo.pin_count like this is illegal.
Where was the BO pinned in the first place?
Christian.
> ret = unreserve_bo_and_vms(&ctx, false, false);
>
> /* Remove from VM internal data structures */
next prev parent reply other threads:[~2021-09-23 11:39 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-23 9:44 [PATCH] drm/kfd: fix ttm_bo_release warning Lang Yu
2021-09-23 11:39 ` Christian König [this message]
2021-09-23 12:09 ` Yu, Lang
2021-09-23 12:23 ` Christian König
2021-09-23 14:24 ` Yu, Lang
2021-09-23 14:52 ` Christian König
2021-09-24 5:35 ` Yu, Lang
2021-09-24 5:42 ` Christian König
2021-09-24 5:50 ` Yu, Lang
2021-09-24 5:54 ` Christian König
2021-09-24 6:34 ` Yu, Lang
2021-09-24 6:37 ` Christian König
2021-09-24 10:37 ` Yu, Lang
2021-09-23 16:21 ` Felix Kuehling
2021-09-24 5:35 ` Yu, Lang
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=063d904f-4083-f5db-b3d0-eb4e080b055f@amd.com \
--to=christian.koenig@amd.com \
--cc=C3B6christian.koenig@amd.com \
--cc=Felix.Kuehling@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=lang.yu@amd.com \
--cc=ray.huang@amd.com \
/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).