<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <meta name="Generator" content="Microsoft Word 15 (filtered medium)"> <style></style> </head> <body bgcolor="white" lang="EN-US" link="blue" vlink="purple"> <div class="WordSection1"> Hi Christian，<o:p></o:p> <o:p> </o:p> With the series patch set , amdgpu_vm_validate_pt_bos occasionally evicted amdgpu BOs failed and can’t <o:p></o:p> find the valid first busy bo . Another problem is that during the first BOs get lock period will run into deadlock .<o:p></o:p> <o:p> </o:p> /* check if other user occupy memory too long time */<o:p></o:p> if (!first_bo || !request_resv || !request_resv->lock.ctx) {<o:p></o:p> if (first_bo)<o:p></o:p> ttm_bo_put(first_bo);<o:p></o:p> return -EBUSY;<o:p></o:p> }<o:p></o:p> if (first_bo->resv == request_resv) {<o:p></o:p> ttm_bo_put(first_bo);<o:p></o:p> return -EBUSY;<o:p></o:p> }<o:p></o:p> if (ctx->interruptible)<o:p></o:p> ret = ww_mutex_lock_interruptible(&first_bo->resv->lock,<o:p></o:p> request_resv->lock.ctx);<o:p></o:p> else<o:p></o:p> ret = ww_mutex_lock(&first_bo->resv->lock, request_resv->lock.ctx);<o:p></o:p> if (ret) {<o:p></o:p> ttm_bo_put(first_bo);<o:p></o:p> if (ret == -EDEADLK) {<o:p></o:p> ret = -EAGAIN;<o:p></o:p> }<o:p></o:p> <o:p> </o:p> return ret;<o:p></o:p> } <o:p></o:p> <o:p> </o:p> Thanks<o:p></o:p> Prike<o:p></o:p> <o:p> </o:p> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in"> From: Christian König <ckoenig.leichtzumerken@gmail.com> Sent: Wednesday, May 15, 2019 3:05 PM To: Liang, Prike <Prike.Liang@amd.com>; Marek Olšák <maraeo@gmail.com> Cc: Zhou, David(ChunMing) <David1.Zhou@amd.com>; dri-devel <dri-devel@lists.freedesktop.org>; amd-gfx mailing list <amd-gfx@lists.freedesktop.org> Subject: Re: [PATCH 11/11] drm/amdgpu: stop removing BOs from the LRU during CS<o:p></o:p> </div> </div> <o:p> </o:p> [CAUTION: External Email] <o:p></o:p> <div> <div> Hi Prike, no, that can lead to massive problems in a real OOM situation and is not something we can do here. Christian. Am 15.05.19 um 04:00 schrieb Liang, Prike:<o:p></o:p> </div> <blockquote style="margin-top:5.0pt;margin-bottom:5.0pt"> Hi Christian ,<o:p></o:p> <o:p></o:p> I just wonder when encounter ENOMEM error during pin amdgpu BOs can we retry validate again as below.<o:p></o:p> With the following simply patch the Abaqus pinned issue not observed. <o:p></o:p> <o:p></o:p> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c<o:p></o:p> index 11cbf63..72a32f5 100644<o:p></o:p> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c<o:p></o:p> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c<o:p></o:p> @@ -902,11 +902,15 @@ int amdgpu_bo_pin_restricted(struct amdgpu_bo *bo, u32 domain,<o:p></o:p> bo->placements[i].lpfn = lpfn;<o:p></o:p> bo->placements[i].flags |= TTM_PL_FLAG_NO_EVICT;<o:p></o:p> }<o:p></o:p> -<o:p></o:p> +retry:<o:p></o:p> r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);<o:p></o:p> if (unlikely(r)) {<o:p></o:p> - dev_err(adev->dev, "%p pin failed\n", bo);<o:p></o:p> - goto error;<o:p></o:p> + if (r == -ENOMEM){<o:p></o:p> + goto retry;<o:p></o:p> + } else {<o:p></o:p> + dev_err(adev->dev, "%p pin failed\n", bo);<o:p></o:p> + goto error;<o:p></o:p> + }<o:p></o:p> }<o:p></o:p> <o:p></o:p> bo->pin_count = 1;<o:p></o:p> <o:p></o:p> <o:p></o:p> Thanks,<o:p></o:p> Prike<o:p></o:p> <o:p></o:p> <div> <div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in"> From: Marek Olšák <a href="mailto:maraeo@gmail.com"><maraeo@gmail.com></a> Sent: Wednesday, May 15, 2019 3:33 AM To: Christian König <a href="mailto:ckoenig.leichtzumerken@gmail.com"><ckoenig.leichtzumerken@gmail.com></a> Cc: Zhou, David(ChunMing) <a href="mailto:David1.Zhou@amd.com"><David1.Zhou@amd.com></a>; Liang, Prike <a href="mailto:Prike.Liang@amd.com"><Prike.Liang@amd.com></a>; dri-devel <a href="mailto:dri-devel@lists.freedesktop.org"> <dri-devel@lists.freedesktop.org></a>; amd-gfx mailing list <a href="mailto:amd-gfx@lists.freedesktop.org"> <amd-gfx@lists.freedesktop.org></a> Subject: Re: [PATCH 11/11] drm/amdgpu: stop removing BOs from the LRU during CS<o:p></o:p> </div> </div> <o:p></o:p> [CAUTION: External Email] <o:p></o:p> <div> <div> <div> <div> This series fixes the OOM errors. However, if I torture the kernel driver more, I can get it to deadlock and end up with unkillable processes. I can also get an OOM error. I just ran the test 5 times:<o:p></o:p> </div> <div> <o:p></o:p> </div> <div> AMD_DEBUG=testgdsmm glxgears & AMD_DEBUG=testgdsmm glxgears & AMD_DEBUG=testgdsmm glxgears & AMD_DEBUG=testgdsmm glxgears & AMD_DEBUG=testgdsmm glxgears<o:p></o:p> </div> <o:p></o:p> <div> Marek<o:p></o:p> </div> </div> </div> <o:p></o:p> <div> <div> On Tue, May 14, 2019 at 8:31 AM Christian König <<a href="mailto:ckoenig.leichtzumerken@gmail.com">ckoenig.leichtzumerken@gmail.com</a>> wrote:<o:p></o:p> </div> <blockquote style="border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-top:5.0pt;margin-right:0in;margin-bottom:5.0pt"> This avoids OOM situations when we have lots of threads submitting at the same time. Signed-off-by: Christian König <<a href="mailto:christian.koenig@amd.com" target="_blank">christian.koenig@amd.com</a>> --- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index fff558cf385b..f9240a94217b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -648,7 +648,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, } r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true, - &duplicates, true); + &duplicates, false); if (unlikely(r != 0)) { if (r != -ERESTARTSYS) DRM_ERROR("ttm_eu_reserve_buffers failed.\n"); -- 2.17.1 _______________________________________________ amd-gfx mailing list <a href="mailto:amd-gfx@lists.freedesktop.org" target="_blank">amd-gfx@lists.freedesktop.org</a> <a href="https://lists.freedesktop.org/mailman/listinfo/amd-gfx" target="_blank">https://lists.freedesktop.org/mailman/listinfo/amd-gfx</a><o:p></o:p> </blockquote> </div> </div> </blockquote> <o:p> </o:p> </div> </div> </body> </html>