From: Greg KH <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: torvalds@linux-foundation.org, akpm@linux-foundation.org,
alan@lxorguk.ukuu.org.uk,
Thomas Hellstrom <thellstrom@vmware.com>,
Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
Jerome Glisse <jglisse@redhat.com>,
Dave Airlie <airlied@redhat.com>
Subject: [ 14/42] drm/ttm: Fix buffer object metadata accounting regression v2
Date: Thu, 14 Jun 2012 16:56:37 -0700 [thread overview]
Message-ID: <20120614235625.157134606@linuxfoundation.org> (raw)
In-Reply-To: <20120614235636.GA5275@kroah.com>
3.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Hellstrom <thellstrom@vmware.com>
commit a393c730ab69617c3291a3b0b2a228c9be2fc28c upstream.
A regression was introduced in the 3.3 rc series, commit
"drm/ttm: simplify memory accounting for ttm user v2",
causing the metadata of buffer objects created using the ttm_bo_create()
function to be accounted twice.
That causes massive leaks with the vmwgfx driver running for example
SpecViewperf Catia-03 test 2, eventually killing the app.
Furthermore, the same commit introduces a regression where
metadata accounting is leaked if a buffer object is
initialized with an illegal size. This is also fixed with this commit.
v2: Fixed an error path and removed an unused variable.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/gpu/drm/ttm/ttm_bo.c | 13 +++----------
1 file changed, 3 insertions(+), 10 deletions(-)
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -1193,6 +1193,7 @@ int ttm_bo_init(struct ttm_bo_device *bd
(*destroy)(bo);
else
kfree(bo);
+ ttm_mem_global_free(mem_glob, acc_size);
return -EINVAL;
}
bo->destroy = destroy;
@@ -1294,22 +1295,14 @@ int ttm_bo_create(struct ttm_bo_device *
struct ttm_buffer_object **p_bo)
{
struct ttm_buffer_object *bo;
- struct ttm_mem_global *mem_glob = bdev->glob->mem_glob;
size_t acc_size;
int ret;
- acc_size = ttm_bo_acc_size(bdev, size, sizeof(struct ttm_buffer_object));
- ret = ttm_mem_global_alloc(mem_glob, acc_size, false, false);
- if (unlikely(ret != 0))
- return ret;
-
bo = kzalloc(sizeof(*bo), GFP_KERNEL);
-
- if (unlikely(bo == NULL)) {
- ttm_mem_global_free(mem_glob, acc_size);
+ if (unlikely(bo == NULL))
return -ENOMEM;
- }
+ acc_size = ttm_bo_acc_size(bdev, size, sizeof(struct ttm_buffer_object));
ret = ttm_bo_init(bdev, bo, size, type, placement, page_alignment,
buffer_start, interruptible,
persistent_swap_storage, acc_size, NULL);
next prev parent reply other threads:[~2012-06-14 23:56 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-14 23:56 [ 00/42] 3.4.3-stable review Greg KH
2012-06-14 23:56 ` [ 01/42] drm/nouveau/disp: fix dithering not being enabled on some eDP macbooks Greg KH
2012-06-14 23:56 ` [ 02/42] drm/i915: Reset last_retired_head when resetting ring Greg KH
2012-06-14 23:56 ` [ 03/42] char/agp: add another Ironlake host bridge Greg KH
2012-06-14 23:56 ` [ 04/42] btree: fix tree corruption in btree_get_prev() Greg KH
2012-06-14 23:56 ` [ 05/42] powerpc/time: Sanity check of decrementer expiration is necessary Greg KH
2012-06-14 23:56 ` [ 06/42] powerpc: Fix kernel panic during kernel module load Greg KH
2012-06-14 23:56 ` [ 07/42] module_param: stop double-calling parameters Greg KH
2012-06-14 23:56 ` [ 08/42] timekeeping: Fix CLOCK_MONOTONIC inconsistency during leapsecond Greg KH
2012-06-14 23:56 ` [ 09/42] ext4: fix the free blocks calculation for ext3 file systems w/ uninit_bg Greg KH
2012-06-14 23:56 ` [ 10/42] x86/uv: Fix UV2 BAU legacy mode Greg KH
2012-06-14 23:56 ` [ 11/42] x86, MCE, AMD: Make APIC LVT thresholding interrupt optional Greg KH
2012-06-14 23:56 ` [ 12/42] hwrng: atmel-rng - fix race condition leading to repeated bits Greg KH
2012-06-14 23:56 ` [ 13/42] crypto: aesni-intel - fix unaligned cbc decrypt for x86-32 Greg KH
2012-06-14 23:56 ` Greg KH [this message]
2012-06-14 23:56 ` [ 15/42] Btrfs: fall back to non-inline if we dont have enough space Greg KH
2012-06-14 23:56 ` [ 16/42] iwlwifi: disable WoWLAN if !CONFIG_PM_SLEEP Greg KH
2012-06-14 23:56 ` [ 17/42] iwlwifi: unregister LEDs if mac80211 registration fails Greg KH
2012-06-14 23:56 ` [ 18/42] iwlwifi: dont mess up the SCD when removing a key Greg KH
2012-06-14 23:56 ` [ 19/42] iwlwifi: disable the buggy chain extension feature in HW Greg KH
2012-06-14 23:56 ` [ 20/42] mac80211: fix error in station state transitions during reconfig Greg KH
2012-06-14 23:56 ` [ 21/42] mac80211: clean up remain-on-channel on interface stop Greg KH
2012-06-14 23:56 ` [ 22/42] mac80211: fix non RCU-safe sta_list manipulation Greg KH
2012-06-14 23:56 ` [ 23/42] ASoC: wm8994: Ensure all AIFnCLK events are run from the _late variants Greg KH
2012-06-14 23:56 ` [ 24/42] ASoC: wm8994: Apply volume updates with clocks enabled Greg KH
2012-06-14 23:56 ` [ 25/42] bcma: add ext PA workaround for BCM4331 and BCM43431 Greg KH
2012-06-14 23:56 ` [ 26/42] ALSA: HDA: Pin fixup for Zotac Z68 motherboard Greg KH
2012-06-14 23:56 ` [ 27/42] cfg80211: fix interface combinations check Greg KH
2012-06-14 23:56 ` [ 28/42] rt2x00: use atomic variable for seqno Greg KH
2012-06-14 23:56 ` [ 29/42] wireless: rt2x00: rt2800usb add more devices ids Greg KH
2012-06-14 23:56 ` [ 30/42] wireless: rt2x00: rt2800usb more devices were identified Greg KH
2012-06-14 23:56 ` [ 31/42] net: sierra_net: device IDs for Aircard 320U++ Greg KH
2012-06-14 23:56 ` [ 32/42] can: c_can: fix "BUG! echo_skb is occupied!" during transmit Greg KH
2012-06-14 23:56 ` [ 33/42] can: c_can: fix an interrupt thrash issue with c_can driver Greg KH
2012-06-14 23:56 ` [ 34/42] can: c_can: fix race condition in c_can_open() Greg KH
2012-06-14 23:56 ` [ 35/42] gma500: dont register the ACPI video bus Greg KH
2012-06-14 23:56 ` [ 36/42] acpi_video: fix leaking PCI references Greg KH
2012-06-14 23:57 ` [ 37/42] sched: Fix the relax_domain_level boot parameter Greg KH
2012-06-14 23:57 ` [ 38/42] drm/radeon: fix tiling and command stream checking on evergreen v3 Greg KH
2012-06-14 23:57 ` [ 39/42] drm/i915: Mark the ringbuffers as being in the GTT domain Greg KH
2012-06-14 23:57 ` [ 40/42] fuse: fix stat call on 32 bit platforms Greg KH
2012-06-14 23:57 ` [ 41/42] libata: add a host flag to ignore detected ATA devices Greg KH
2012-06-14 23:57 ` [ 42/42] ata_piix: defer disks to the Hyper-V drivers by default Greg KH
2012-06-15 11:16 ` From all Hyper-V admins: Big thanks! Re: [ 41/42] libata + [ 42/42] ata_piix: defer disks to the Hyper-V drivers by default has been added to the 3.4-stable tree Victor Miasnikov
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=20120614235625.157134606@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=airlied@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=jglisse@redhat.com \
--cc=konrad.wilk@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=thellstrom@vmware.com \
--cc=torvalds@linux-foundation.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).