GFP_TRANSHUGE tries very hard to allocate huge pages, which can result
in long delays with high memory pressure. I have observed firefox
freezing for up to around a minute due to this while restic was taking
a full system backup.
Since we don't really need huge pages, use GFP_TRANSHUGE_LIGHT |
__GFP_NORETRY instead, in order to fail quickly when there are no huge
pages available.
Set __GFP_KSWAPD_RECLAIM as well, in order for huge pages to be freed
up in the background if necessary.
With these changes, I'm no longer seeing freezes during a restic backup.
Cc: stable@vger.kernel.org
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
the free mem space and the lower limit both include two parts:
system memory and swap space.
For the OOM triggered by TTM, that is the case as below:
first swap space is full of swapped out pages and soon
system memory also is filled up with ttm pages. and then
any memory allocation request will run into OOM.
to cover two cases:
a. if no swap disk at all or free swap space is under swap mem
limit but available system mem is bigger than sys mem limit,
allow TTM allocation;
b. if the available system mem is less than sys mem limit but
free swap space is bigger than swap mem limit, allow TTM
allocation.
v2: merge two memory limit(swap and system) into one
v3: keep original behavior except ttm_opt_ctx->flags with
TTM_OPT_FLAG_FORCE_ALLOC
v4: always set force_alloc as tx->flags & TTM_OPT_FLAG_FORCE_ALLOC
v5: add an attribute for lower_mem_limit
v6: set lower_mem_limit as 0 to keep original behavior
Signed-off-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The pointer is available as ttm->bdev->glob as well.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
when ttm_mem_global_alloc_page fails, we should call
ttm_mem_global_free_page to update memory count for
the ttm pages which already run ttm_mem_global_alloc_page
successfully
Signed-off-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This to allow drivers to choose to avoid OOM invocation and handle
page allocation failures instead.
v2:
Remove extra new lines.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
A few more fixes for 4.16, nothing major.
A few more fixes for 4.16. This is on top of the pull request from
last week. Most notable change here is a fix to the link order for
the now separate from amdgpu GPU scheduler to fix crashes when the
modules are build into the kernel rather than as modules.
* 'drm-next-4.16' of git://people.freedesktop.org/~agd5f/linux:
drm: fix gpu scheduler link order
drm/amd/display: Demote error print to debug print when ATOM impl missing
drm/amdgpu: Avoid leaking PM domain on driver unbind (v2)
drm/amd/amdgpu: Add Polaris version check
drm/amdgpu: Reenable manual GPU reset from sysfs
drm/amdgpu: disable MMHUB power gating on raven
drm/ttm: Don't unreserve swapped BOs that were previously reserved
drm/ttm: Don't add swapped BOs to swap-LRU list
drm/amdgpu: only check for ECC on Vega10
drm/amd/powerplay: Fix smu_table_entry.handle type
drm/ttm: add VADDR_FLAG_UPDATED_COUNT to correctly update dma_page global count
drm/radeon: fill in rb backend map on evergreen/ni.
drm/amdgpu/gfx9: fix ngg enablement to clear gds reserved memory (v2)
drm/ttm: only free pages rather than update global memory count together
drm/amdgpu: fix CPU based VM updates
drm/amdgpu: fix typo in amdgpu_vce_validate_bo
drm/amdgpu: fix amdgpu_vm_pasid_fault_credit
drm/ttm: check the return value of register_shrinker
drm/radeon: fix sparse warning: Should it be static?
-----BEGIN PGP SIGNATURE-----
iQEcBAABAgAGBQJaW+iVAAoJEHm+PkMAQRiGCDsIAJALNpX7odTx/8y+yCSWbpBH
E57iwr4rmnI6tXJY6gqBUWTYnjAcf4b8IsHGCO6q3WIE3l/kt+m3eA21a32mF2Db
/bfPGTOWu5LoOnFqzgH2kiFuC3Y474toxpld2YtkQWYxi5W7SUtIHi/jGgkUprth
g15yPfwYgotJd/gpmPfBDMPlYDYvLlnPYbTG6ZWdMbg39m2RF2m0BdQ6aBFLHvbJ
IN0tjCM6hrLFBP0+6Zn60pevUW9/AFYotZn2ankNTk5QVCQm14rgQIP+Pfoa5WpE
I25r0DbkG2jKJCq+tlgIJjxHKD37GEDMc4T8/5Y8CNNeT9Q8si9EWvznjaAPazw=
=o5gx
-----END PGP SIGNATURE-----
BackMerge tag 'v4.15-rc8' into drm-next
Linux 4.15-rc8
Daniel requested this for so the intel CI won't fall over on drm-next
so often.
if ttm_get_pages or ttm_mem_global_alloc_page fail, should not update
global memory count.
Signed-off-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Tested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This fixes the build warning:
"ignoring return value of 'register_shrinker', declared with
attribute warn_unused_result [-Wunused-result]"
Signed-off-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In the function ttm_page_alloc_init, kzalloc call is made for variable
_manager, we need to check its return value, it may return NULL.
Signed-off-by: Xiongwei Song <sxwjean@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
forward the operation context to ttm_tt_populate as well,
and the ultimate goal is swapout enablement for reserved BOs.
v2: squash in fix for vboxvideo
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
forward the operation context to ttm_mem_global_alloc_page as well,
and the ultimate goal is swapout enablement for reserved BOs.
Here reserved BOs refer to all the BOs which share same reservation object
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
fix calculation.
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
shrink_pages is in unit of Order after ttm_page_pool_free,
but it is used by nr_free in next round so need change
it into native page unit
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
fix calculation.
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
shrink_pages is in unit of Order after ttm_page_pool_free,
but it is used by nr_free in next round so need change
it into native page unit
Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When we detect consecutive allocation of pages swap them to avoid
accidentally freeing them as huge page.
v2: use swap
v3: check if it's really the first allocated page
v4: don't touch the loop variable
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When we detect consecutive allocation of pages swap them to avoid
accidentally freeing them as huge page.
v2: use swap
v3: check if it's really the first allocated page
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
e.g. shrink reqeust is less than 512, the logic will skip huge pool
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
to indicate page order for each element in the pool
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When we detect consecutive allocation of pages swap them to avoid
accidentally freeing them as huge page.
v2: use swap
v3: check if it's really the first allocated page
v4: don't touch the loop variable
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When we detect consecutive allocation of pages swap them to avoid
accidentally freeing them as huge page.
v2: use swap
v3: check if it's really the first allocated page
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Roger He <Hongbo.He@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
e.g. shrink reqeust is less than 512, the logic will skip huge pool
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
to indicate page order for each element in the pool
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Roger He <Hongbo.He@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixes for 4.15. Highlights:
- DC fixes for S3, gamma, audio, pageflipping, etc.
- fix a regression in radeon from kfd removal
- fix a ttm regression with swiotlb disabled
- misc other fixes
* 'drm-fixes-4.15' of git://people.freedesktop.org/~agd5f/linux: (36 commits)
drm/radeon: remove init of CIK VMIDs 8-16 for amdkfd
drm/ttm: fix populate_and_map() functions once more
drm/amd/display: USB-C / thunderbolt dock specific workaround
drm/amd/display: Switch to drm_atomic_helper_wait_for_flip_done
drm/amd/display: fix gamma setting
drm/amd/display: Do not put drm_atomic_state on resume
drm/amd/display: Fix couple more inconsistent NULL checks in dc_resource
drm/amd/display: Fix potential NULL and mem leak in create_links
drm/amd/display: Fix hubp check in set_cursor_position
drm/amd/display: Fix use before NULL check in validate_timing
drm/amd/display: Bunch of smatch error and warning fixes in DC
drm/amd/display: Fix amdgpu_dm bugs found by smatch
drm/amd/display: try to find matching audio inst for enc inst first
drm/amd/display: fix seq issue: turn on clock before programming afmt.
drm/amd/display: fix memory leaks on error exit return
drm/amd/display: check plane state before validating fbc
drm/amd/display: Do DC mode-change check when adding CRTCs
drm/amd/display: Revert noisy assert messages
drm/amd/display: fix split viewport rounding error
drm/amd/display: Check aux channel before MST resume
...
This reverts "drm/ttm: Fix configuration error around populate_and_map()
functions".
This fix has gone into the wrong direction. Those helpers should be
available even when neither CONFIG_INTEL_IOMMU nor CONFIG_SWIOTLB are
set.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Memory allocation failure should generally be handled gracefully by
callers. In particular, with transparent hugepage support, attempts
to allocate huge pages can fail under memory pressure, but the callers
fall back to allocating individual pages instead. In that case, there
would be spurious
[TTM] Unable to get page %u
error messages in dmesg.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We don't use compound pages at the moment. Take this into account when
freeing them.
Signed-off-by: Christian König <christian.koenig@amd.comd>
Reviewed-and-Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add a new huge page pool and try to allocate from it when it makes sense.
v2: avoid compound pages for now
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Make it easier to add huge page pool.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Try to allocate huge pages when it makes sense.
v2: avoid compound pages for now
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Instead of mapping them bit by bit map/unmap all consecutive
pages as in one call.
v2: test for consecutive pages instead of using compound page order.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Totally surprisingly this is more efficient than doing it page by page.
Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Correctly handle different page sizes in the memory accounting.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Nobody is actually using that, remove it.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fixed kbuild errors when IOMMU/SWIOTLB are disabled.
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
These functions replace a section of common code found
in radeon/amdgpu drivers (and possibly others) as part
of the ttm_tt_*populate() callbacks.
v2: squash in fix for sw iommu from Tom
Signed-off-by: Tom St Denis <tom.stdenis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When fail to get needed page for pool, need to put allocated pages
into pool. But current code has a miscalculation of allocated pages,
correct it.
Signed-off-by: Xiangliang.Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
set_memory_* functions have moved to set_memory.h. Switch to this
explicitly.
[akpm@linux-foundation.org: track drivers/gpu/drm/i915/i915_gem_gtt.c linux-next changes]
Link: http://lkml.kernel.org/r/1488920133-27229-8-git-send-email-labbott@redhat.com
Signed-off-by: Laura Abbott <labbott@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It tries to do fancy things with excluding agp support if ttm is
built-in, but agp isn't. Instead just express this depency like drm
does and use CONFIG_AGP everywhere.
Also use the neat Makefile magic to make the entire ttm_agp_backend
file optional.
v2: Use IS_ENABLED(CONFIG_AGP) as suggested by Ville
v3: Review from Emil.
v4: Actually get it right as spotted by 0-day.
Cc: Emil Velikov <emil.l.velikov@gmail.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1459337046-25882-1-git-send-email-daniel.vetter@ffwll.ch
Andrew Morton wrote:
> On Wed, 12 Nov 2014 13:08:55 +0900 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> wrote:
>
> > Andrew Morton wrote:
> > > Poor ttm guys - this is a bit of a trap we set for them.
> >
> > Commit a91576d791 ("drm/ttm: Pass GFP flags in order to avoid deadlock.")
> > changed to use sc->gfp_mask rather than GFP_KERNEL.
> >
> > - pages_to_free = kmalloc(npages_to_free * sizeof(struct page *),
> > - GFP_KERNEL);
> > + pages_to_free = kmalloc(npages_to_free * sizeof(struct page *), gfp);
> >
> > But this bug is caused by sc->gfp_mask containing some flags which are not
> > in GFP_KERNEL, right? Then, I think
> >
> > - pages_to_free = kmalloc(npages_to_free * sizeof(struct page *), gfp);
> > + pages_to_free = kmalloc(npages_to_free * sizeof(struct page *), gfp & GFP_KERNEL);
> >
> > would hide this bug.
> >
> > But I think we should use GFP_ATOMIC (or drop __GFP_WAIT flag)
>
> Well no - ttm_page_pool_free() should stop calling kmalloc altogether.
> Just do
>
> struct page *pages_to_free[16];
>
> and rework the code to free 16 pages at a time. Easy.
Well, ttm code wants to process 512 pages at a time for performance.
Memory footprint increased by 512 * sizeof(struct page *) buffer is
only 4096 bytes. What about using static buffer like below?
----------
>From d3cb5393c9c8099d6b37e769f78c31af1541fe8c Mon Sep 17 00:00:00 2001
From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Date: Thu, 13 Nov 2014 22:21:54 +0900
Subject: [PATCH] drm/ttm: Avoid memory allocation from shrinker functions.
Commit a91576d791 ("drm/ttm: Pass GFP flags in order to avoid
deadlock.") caused BUG_ON() due to sc->gfp_mask containing flags
which are not in GFP_KERNEL.
https://bugzilla.kernel.org/show_bug.cgi?id=87891
Changing from sc->gfp_mask to (sc->gfp_mask & GFP_KERNEL) would
avoid the BUG_ON(), but avoiding memory allocation from shrinker
function is better and reliable fix.
Shrinker function is already serialized by global lock, and
clean up function is called after shrinker function is unregistered.
Thus, we can use static buffer when called from shrinker function
and clean up function.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: stable <stable@kernel.org> [2.6.35+]
Signed-off-by: Dave Airlie <airlied@redhat.com>
Kees Cook