AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Yang, Philip" <Philip.Yang-5C7GfCeVMHo@public.gmane.org>
To: "amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
	<amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>
Cc: "Yang, Philip" <Philip.Yang-5C7GfCeVMHo@public.gmane.org>
Subject: [PATCH 1/3] drm/amdkfd: support concurrent userptr update for HMM
Date: Tue, 5 Mar 2019 18:09:35 +0000	[thread overview]
Message-ID: <20190305180915.15216-2-Philip.Yang@amd.com> (raw)
In-Reply-To: <20190305180915.15216-1-Philip.Yang-5C7GfCeVMHo@public.gmane.org>

Userptr restore may have concurrent userptr invalidation after
hmm_vma_fault adds the range to the hmm->ranges list, needs call
hmm_vma_range_done to remove the range from hmm->ranges list first,
then reschedule the restore worker. Otherwise hmm_vma_fault will add
same range to the list, this will cause loop in the list because
range->next point to range itself.

Add function untrack_invalid_user_pages to reduce code duplication.

Change-Id: I31407739dc10554f8e418c7a0e0415d3d95552f1
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  | 28 ++++++++++++++-----
 1 file changed, 21 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 314c048fcac6..783d760ccfe3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1731,6 +1731,23 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info,
 	return 0;
 }
 
+/* Remove invalid userptr BOs from hmm track list
+ *
+ * Stop HMM track the userptr update
+ */
+static void untrack_invalid_user_pages(struct amdkfd_process_info *process_info)
+{
+	struct kgd_mem *mem, *tmp_mem;
+	struct amdgpu_bo *bo;
+
+	list_for_each_entry_safe(mem, tmp_mem,
+				 &process_info->userptr_inval_list,
+				 validate_list.head) {
+		bo = mem->bo;
+		amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm);
+	}
+}
+
 /* Validate invalid userptr BOs
  *
  * Validates BOs on the userptr_inval_list, and moves them back to the
@@ -1848,12 +1865,7 @@ static int validate_invalid_user_pages(struct amdkfd_process_info *process_info)
 out_free:
 	kfree(pd_bo_list_entries);
 out_no_mem:
-	list_for_each_entry_safe(mem, tmp_mem,
-				 &process_info->userptr_inval_list,
-				 validate_list.head) {
-		bo = mem->bo;
-		amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm);
-	}
+	untrack_invalid_user_pages(process_info);
 
 	return ret;
 }
@@ -1897,8 +1909,10 @@ static void amdgpu_amdkfd_restore_userptr_worker(struct work_struct *work)
 	 * and we can just restart the queues.
 	 */
 	if (!list_empty(&process_info->userptr_inval_list)) {
-		if (atomic_read(&process_info->evicted_bos) != evicted_bos)
+		if (atomic_read(&process_info->evicted_bos) != evicted_bos) {
+			untrack_invalid_user_pages(process_info);
 			goto unlock_out; /* Concurrent eviction, try again */
+		}
 
 		if (validate_invalid_user_pages(process_info))
 			goto unlock_out;
-- 
2.17.1

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

  parent reply	other threads:[~2019-03-05 18:09 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-05 18:09 [PATCH 0/3] handle userptr corner cases with HMM path Yang, Philip
     [not found] ` <20190305180915.15216-1-Philip.Yang-5C7GfCeVMHo@public.gmane.org>
2019-03-05 18:09   ` Yang, Philip [this message]
     [not found]     ` <20190305180915.15216-2-Philip.Yang-5C7GfCeVMHo@public.gmane.org>
2019-03-06 20:01       ` [PATCH 1/3] drm/amdkfd: support concurrent userptr update for HMM Kuehling, Felix
     [not found]         ` <772b1109-b6db-05cc-4f36-6428724ae4a3-5C7GfCeVMHo@public.gmane.org>
2019-03-07  0:54           ` Yang, Philip
2019-03-05 18:09   ` [PATCH 2/3] drm/amdgpu: support userptr cross VMAs case with HMM Yang, Philip
     [not found]     ` <20190305180915.15216-3-Philip.Yang-5C7GfCeVMHo@public.gmane.org>
2019-03-06 20:11       ` Kuehling, Felix
     [not found]         ` <f51eaecd-8d50-0eea-aac6-2905f945a2c2-5C7GfCeVMHo@public.gmane.org>
2019-03-07  1:12           ` Yang, Philip
2019-03-05 18:09   ` [PATCH 3/3] drm/amdgpu: more descriptive message if HMM not enabled Yang, Philip
     [not found]     ` <20190305180915.15216-4-Philip.Yang-5C7GfCeVMHo@public.gmane.org>
2019-03-06  9:05       ` Michel Dänzer
     [not found]         ` <9ac0b1ab-891e-9fe6-667e-f77a244dd862-otUistvHUpPR7s880joybQ@public.gmane.org>
2019-03-06 14:42           ` Yang, Philip
     [not found]             ` <baa08fbf-8bb0-6ad7-9e62-b9ebdf88cf72-5C7GfCeVMHo@public.gmane.org>
2019-03-06 14:53               ` Michel Dänzer

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=20190305180915.15216-2-Philip.Yang@amd.com \
    --to=philip.yang-5c7gfcevmho@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.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