AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Roger He <Hongbo.He-5C7GfCeVMHo@public.gmane.org>
To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org,
	dri-devel-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org
Cc: Roger He <Hongbo.He-5C7GfCeVMHo@public.gmane.org>
Subject: [PATCH 4/4] drm/ttm: free one in huge pool even shrink request less than one element
Date: Tue, 21 Nov 2017 17:32:22 +0800	[thread overview]
Message-ID: <1511256742-5601-5-git-send-email-Hongbo.He@amd.com> (raw)
In-Reply-To: <1511256742-5601-1-git-send-email-Hongbo.He-5C7GfCeVMHo@public.gmane.org>

Change-Id: Id8bd4d1ecff9f3ab14355e2dbd1c59b9fe824e01
Signed-off-by: Roger He <Hongbo.He@amd.com>
---
 drivers/gpu/drm/ttm/ttm_page_alloc.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc.c b/drivers/gpu/drm/ttm/ttm_page_alloc.c
index 90546fd..c194a51 100644
--- a/drivers/gpu/drm/ttm/ttm_page_alloc.c
+++ b/drivers/gpu/drm/ttm/ttm_page_alloc.c
@@ -453,11 +453,13 @@ ttm_pool_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
 		pool = &_manager->pools[(i + pool_offset)%NUM_POOLS];
 		/* OK to use static buffer since global mutex is held. */
 		nr_free_pool = (nr_free >> pool->order);
-		if (nr_free_pool == 0)
-			continue;
+		if (!nr_free_pool && pool->order)
+			nr_free_pool = 1;
 
 		shrink_pages = ttm_page_pool_free(pool, nr_free_pool, true);
 		freed += ((nr_free_pool - shrink_pages) << pool->order);
+		if (freed > sc->nr_to_scan)
+			break;
 	}
 	mutex_unlock(&lock);
 	return freed;
-- 
2.7.4

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

  parent reply	other threads:[~2017-11-21  9:32 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-11-21  9:32 [PATCH 0/4] *** fix memory leak for HUGE PAGE *** Roger He
2017-11-21  9:32 ` [PATCH 1/4] drm/ttm: add page order in page pool Roger He
2017-11-21  9:52   ` Christian König
     [not found]   ` <1511256742-5601-2-git-send-email-Hongbo.He-5C7GfCeVMHo@public.gmane.org>
2017-11-23 20:07     ` kbuild test robot
2017-11-23 21:49   ` kbuild test robot
2017-11-21  9:32 ` [PATCH 2/4] drm/ttm: use NUM_PAGES_TO_ALLOC always Roger He
     [not found]   ` <1511256742-5601-3-git-send-email-Hongbo.He-5C7GfCeVMHo@public.gmane.org>
2017-11-21  9:41     ` Christian König
2017-11-21  9:32 ` [PATCH 3/4] drm/ttm: add page order support in ttm_pages_put Roger He
     [not found]   ` <1511256742-5601-4-git-send-email-Hongbo.He-5C7GfCeVMHo@public.gmane.org>
2017-11-21  9:52     ` Christian König
     [not found] ` <1511256742-5601-1-git-send-email-Hongbo.He-5C7GfCeVMHo@public.gmane.org>
2017-11-21  9:32   ` Roger He [this message]
  -- strict thread matches above, loose matches on Subject: below --
2017-11-22  5:36 [PATCH 1/4] drm/ttm: add page order in page pool Roger He
     [not found] ` <1511329016-552-1-git-send-email-Hongbo.He-5C7GfCeVMHo@public.gmane.org>
2017-11-22  5:36   ` [PATCH 4/4] drm/ttm: free one in huge pool even shrink request less than one element Roger He

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=1511256742-5601-5-git-send-email-Hongbo.He@amd.com \
    --to=hongbo.he-5c7gfcevmho@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    --cc=dri-devel-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