Linaro-mm-sig

linaro-mm-sig@lists.linaro.org

48 participants
6403 discussions

[syzbot] WARNING in drm_prime_destroy_file_private

by syzbot

Hello, syzbot found the following issue on: HEAD commit: ea4424be1688 Merge tag 'mtd/fixes-for-5.17-rc8' of git://g.. git tree: upstream console output: https://syzkaller.appspot.com/x/log.txt?x=14095f9e700000 kernel config: https://syzkaller.appspot.com/x/.config?x=aba0ab2928a512c2 dashboard link: https://syzkaller.appspot.com/bug?extid=2448673875b4e20db46a compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2 Unfortunately, I don't have any reproducer for this issue yet. IMPORTANT: if you fix the issue, please add the following tag to the commit: Reported-by: syzbot+2448673875b4e20db46a(a)syzkaller.appspotmail.com ------------[ cut here ]------------ WARNING: CPU: 1 PID: 16791 at drivers/gpu/drm/drm_prime.c:228 drm_prime_destroy_file_private+0x3e/0x50 drivers/gpu/drm/drm_prime.c:228 Modules linked in: CPU: 1 PID: 16791 Comm: syz-executor.5 Not tainted 5.17.0-rc7-syzkaller-00020-gea4424be1688 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:drm_prime_destroy_file_private+0x3e/0x50 drivers/gpu/drm/drm_prime.c:228 Code: 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 80 3c 02 00 75 1f 48 8b 83 90 00 00 00 48 85 c0 75 06 5b e9 e7 6c 1d fd e8 e2 6c 1d fd <0f> 0b 5b e9 da 6c 1d fd e8 05 5a 64 fd eb da 0f 1f 00 41 55 49 89 RSP: 0018:ffffc90002af79e0 EFLAGS: 00010293 RAX: 0000000000000000 RBX: ffff888025e72370 RCX: 0000000000000000 RDX: ffff8880727aa1c0 RSI: ffffffff845a788e RDI: ffff888025e72400 RBP: ffff8881471d4068 R08: 0000000000000001 R09: 0000000000000001 R10: ffffffff817e23e8 R11: 0000000000088078 R12: ffff888025e72000 R13: ffff888025e722b8 R14: ffff8881471d4098 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007ffea39ef0d8 CR3: 00000000791d8000 CR4: 00000000003506e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> drm_file_free.part.0+0x6e5/0xb80 drivers/gpu/drm/drm_file.c:291 drm_file_free drivers/gpu/drm/drm_file.c:248 [inline] drm_close_helper.isra.0+0x17d/0x1f0 drivers/gpu/drm/drm_file.c:308 drm_release+0x1e6/0x530 drivers/gpu/drm/drm_file.c:495 __fput+0x286/0x9f0 fs/file_table.c:317 task_work_run+0xdd/0x1a0 kernel/task_work.c:164 exit_task_work include/linux/task_work.h:32 [inline] do_exit+0xb29/0x2a30 kernel/exit.c:806 do_group_exit+0xd2/0x2f0 kernel/exit.c:935 get_signal+0x45a/0x2490 kernel/signal.c:2863 arch_do_signal_or_restart+0x2a9/0x1c40 arch/x86/kernel/signal.c:868 handle_signal_work kernel/entry/common.c:148 [inline] exit_to_user_mode_loop kernel/entry/common.c:172 [inline] exit_to_user_mode_prepare+0x17d/0x290 kernel/entry/common.c:207 __syscall_exit_to_user_mode_work kernel/entry/common.c:289 [inline] syscall_exit_to_user_mode+0x19/0x60 kernel/entry/common.c:300 ret_from_fork+0x15/0x30 arch/x86/entry/entry_64.S:288 RIP: 0033:0x7f836c6a0471 Code: Unable to access opcode bytes at RIP 0x7f836c6a0447. RSP: 002b:00007f836afd22f0 EFLAGS: 00000206 ORIG_RAX: 0000000000000038 RAX: 0000000000000000 RBX: 00007f836afd2700 RCX: 00007f836c6a0471 RDX: 00007f836afd29d0 RSI: 00007f836afd22f0 RDI: 00000000003d0f00 RBP: 00007ffed5e285a0 R08: 00007f836afd2700 R09: 00007f836afd2700 R10: 00007f836afd29d0 R11: 0000000000000206 R12: 00007ffed5e2840e R13: 00007ffed5e2840f R14: 00007f836afd2300 R15: 0000000000022000 </TASK> --- This report is generated by a bot. It may contain errors. See https://goo.gl/tpsmEJ for more information about syzbot. syzbot engineers can be reached at syzkaller(a)googlegroups.com. syzbot will keep track of this issue. See: https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

4 years, 2 months

[PATCH] dma-fence: fix free sync object incorrectly

by wangshumin

The function dma_fence_free() works fine because struct dma_fence is the first member of sync object. Use `kfree` make it more reasonable. Signed-off-by: wangshumin <pdsrazor(a)gmail.com> --- drivers/dma-buf/dma-fence-array.c | 2 +- drivers/dma-buf/dma-fence-chain.c | 2 +- drivers/dma-buf/sw_sync.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/dma-buf/dma-fence-array.c b/drivers/dma-buf/dma-fence-array.c index cb1bacb5a42b..fc52d837e579 100644 --- a/drivers/dma-buf/dma-fence-array.c +++ b/drivers/dma-buf/dma-fence-array.c @@ -120,7 +120,7 @@ static void dma_fence_array_release(struct dma_fence *fence) dma_fence_put(array->fences[i]); kfree(array->fences); - dma_fence_free(fence); + kfree(array); } const struct dma_fence_ops dma_fence_array_ops = { diff --git a/drivers/dma-buf/dma-fence-chain.c b/drivers/dma-buf/dma-fence-chain.c index 06f8ef97c6e8..b29e1f22f08e 100644 --- a/drivers/dma-buf/dma-fence-chain.c +++ b/drivers/dma-buf/dma-fence-chain.c @@ -203,7 +203,7 @@ static void dma_fence_chain_release(struct dma_fence *fence) dma_fence_put(prev); dma_fence_put(chain->fence); - dma_fence_free(fence); + kfree(chain); } const struct dma_fence_ops dma_fence_chain_ops = { diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c index 348b3a9170fa..80432eeb58c3 100644 --- a/drivers/dma-buf/sw_sync.c +++ b/drivers/dma-buf/sw_sync.c @@ -142,7 +142,7 @@ static void timeline_fence_release(struct dma_fence *fence) spin_unlock_irqrestore(fence->lock, flags); sync_timeline_put(parent); - dma_fence_free(fence); + kfree(pt); } static bool timeline_fence_signaled(struct dma_fence *fence) -- 2.17.1

4 years, 2 months

completely rework the dma_resv semantic

by Christian König

Hi Daniel, just a gentle ping that you wanted to take a look at this. Not much changed compared to the last version, only a minor bugfix in the dma_resv_get_singleton error handling. Regards, Christian.

4 years, 2 months

Re: [Linaro-mm-sig] [PATCH 18/24] dma-buf: add enum dma_resv_usage v3

by Daniel Vetter

On Tue, Dec 07, 2021 at 01:34:05PM +0100, Christian König wrote: > This change adds the dma_resv_usage enum and allows us to specify why a > dma_resv object is queried for its containing fences. > > Additional to that a dma_resv_usage_rw() helper function is added to aid > retrieving the fences for a read or write userspace submission. > > This is then deployed to the different query functions of the dma_resv > object and all of their users. When the write paratermer was previously > true we now use DMA_RESV_USAGE_WRITE and DMA_RESV_USAGE_READ otherwise. > > v2: add KERNEL/OTHER in separate patch > v3: some kerneldoc suggestions by Daniel > > Signed-off-by: Christian König <christian.koenig(a)amd.com> Just commenting on the kerneldoc here. > diff --git a/include/linux/dma-resv.h b/include/linux/dma-resv.h > index 40ac9d486f8f..d96d8ca9af56 100644 > --- a/include/linux/dma-resv.h > +++ b/include/linux/dma-resv.h > @@ -49,6 +49,49 @@ extern struct ww_class reservation_ww_class; > > struct dma_resv_list; > > +/** > + * enum dma_resv_usage - how the fences from a dma_resv obj are used > + * > + * This enum describes the different use cases for a dma_resv object and > + * controls which fences are returned when queried. We need to link here to both dma_buf.resv and from there to here. Also we had a fair amount of text in the old dma_resv fields which should probably be included here. > + */ > +enum dma_resv_usage { > + /** > + * @DMA_RESV_USAGE_WRITE: Implicit write synchronization. > + * > + * This should only be used for userspace command submissions which add > + * an implicit write dependency. > + */ > + DMA_RESV_USAGE_WRITE, > + > + /** > + * @DMA_RESV_USAGE_READ: Implicit read synchronization. > + * > + * This should only be used for userspace command submissions which add > + * an implicit read dependency. I think the above would benefit from at least a link each to &dma_buf.resv for further discusion. Plus the READ flag needs a huge warning that in general it does _not_ guarantee that neither there's no writes possible, nor that the writes can be assumed mistakes and dropped (on buffer moves e.g.). Drivers can only make further assumptions for driver-internal dma_resv objects (e.g. on vm/pagetables) or when the fences are all fences of the same driver (e.g. the special sync rules amd has that takes the fence owner into account). We have this documented in the dma_buf.resv rules, but since it came up again in a discussion with Thomas H. somewhere, it's better to hammer this in a few more time. Specically in generally ignoring READ fences for buffer moves (well the copy job, memory freeing still has to wait for all of them) is a correctness bug. Maybe include a big warning that really the difference between READ and WRITE should only matter for implicit sync, and _not_ for anything else the kernel does. I'm assuming the actual replacement is all mechanical, so I skipped that one for now, that's for next year :-) -Daniel > + */ > + DMA_RESV_USAGE_READ, > +}; > + > +/** > + * dma_resv_usage_rw - helper for implicit sync > + * @write: true if we create a new implicit sync write > + * > + * This returns the implicit synchronization usage for write or read accesses, > + * see enum dma_resv_usage. > + */ > +static inline enum dma_resv_usage dma_resv_usage_rw(bool write) > +{ > + /* This looks confusing at first sight, but is indeed correct. > + * > + * The rational is that new write operations needs to wait for the > + * existing read and write operations to finish. > + * But a new read operation only needs to wait for the existing write > + * operations to finish. > + */ > + return write ? DMA_RESV_USAGE_READ : DMA_RESV_USAGE_WRITE; > +} > + > /** > * struct dma_resv - a reservation object manages fences for a buffer > * > @@ -147,8 +190,8 @@ struct dma_resv_iter { > /** @obj: The dma_resv object we iterate over */ > struct dma_resv *obj; > > - /** @all_fences: If all fences should be returned */ > - bool all_fences; > + /** @usage: Controls which fences are returned */ > + enum dma_resv_usage usage; > > /** @fence: the currently handled fence */ > struct dma_fence *fence; > @@ -178,14 +221,14 @@ struct dma_fence *dma_resv_iter_next(struct dma_resv_iter *cursor); > * dma_resv_iter_begin - initialize a dma_resv_iter object > * @cursor: The dma_resv_iter object to initialize > * @obj: The dma_resv object which we want to iterate over > - * @all_fences: If all fences should be returned or just the exclusive one > + * @usage: controls which fences to include, see enum dma_resv_usage. > */ > static inline void dma_resv_iter_begin(struct dma_resv_iter *cursor, > struct dma_resv *obj, > - bool all_fences) > + enum dma_resv_usage usage) > { > cursor->obj = obj; > - cursor->all_fences = all_fences; > + cursor->usage = usage; > cursor->fence = NULL; > } > > @@ -242,7 +285,7 @@ static inline bool dma_resv_iter_is_restarted(struct dma_resv_iter *cursor) > * dma_resv_for_each_fence - fence iterator > * @cursor: a struct dma_resv_iter pointer > * @obj: a dma_resv object pointer > - * @all_fences: true if all fences should be returned > + * @usage: controls which fences to return > * @fence: the current fence > * > * Iterate over the fences in a struct dma_resv object while holding the > @@ -251,8 +294,8 @@ static inline bool dma_resv_iter_is_restarted(struct dma_resv_iter *cursor) > * valid as long as the lock is held and so no extra reference to the fence is > * taken. > */ > -#define dma_resv_for_each_fence(cursor, obj, all_fences, fence) \ > - for (dma_resv_iter_begin(cursor, obj, all_fences), \ > +#define dma_resv_for_each_fence(cursor, obj, usage, fence) \ > + for (dma_resv_iter_begin(cursor, obj, usage), \ > fence = dma_resv_iter_first(cursor); fence; \ > fence = dma_resv_iter_next(cursor)) > > @@ -419,14 +462,14 @@ void dma_resv_add_shared_fence(struct dma_resv *obj, struct dma_fence *fence); > void dma_resv_replace_fences(struct dma_resv *obj, uint64_t context, > struct dma_fence *fence); > void dma_resv_add_excl_fence(struct dma_resv *obj, struct dma_fence *fence); > -int dma_resv_get_fences(struct dma_resv *obj, bool write, > +int dma_resv_get_fences(struct dma_resv *obj, enum dma_resv_usage usage, > unsigned int *num_fences, struct dma_fence ***fences); > -int dma_resv_get_singleton(struct dma_resv *obj, bool write, > +int dma_resv_get_singleton(struct dma_resv *obj, enum dma_resv_usage usage, > struct dma_fence **fence); > int dma_resv_copy_fences(struct dma_resv *dst, struct dma_resv *src); > -long dma_resv_wait_timeout(struct dma_resv *obj, bool wait_all, bool intr, > - unsigned long timeout); > -bool dma_resv_test_signaled(struct dma_resv *obj, bool test_all); > +long dma_resv_wait_timeout(struct dma_resv *obj, enum dma_resv_usage usage, > + bool intr, unsigned long timeout); > +bool dma_resv_test_signaled(struct dma_resv *obj, enum dma_resv_usage usage); > void dma_resv_describe(struct dma_resv *obj, struct seq_file *seq); > > #endif /* _LINUX_RESERVATION_H */ > -- > 2.25.1 > -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch

4 years, 2 months

[PATCH] ASoC: qcom: Fix error code in lpass_platform_copy()

by Dan Carpenter

The copy_to/from_user() functions return the number of bytes remaining to be copied. This function needs to return negative error codes because snd_soc_pcm_component_copy_user() treats positive returns as success in soc_component_ret(). Fixes: 7d7209557b67 ("ASoC: qcom: Add support for codec dma driver") Signed-off-by: Dan Carpenter <dan.carpenter(a)oracle.com> --- sound/soc/qcom/lpass-platform.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/sound/soc/qcom/lpass-platform.c b/sound/soc/qcom/lpass-platform.c index bf180a594c19..620312529c2f 100644 --- a/sound/soc/qcom/lpass-platform.c +++ b/sound/soc/qcom/lpass-platform.c @@ -1228,15 +1228,19 @@ static int lpass_platform_copy(struct snd_soc_component *component, channel * (rt->dma_bytes / rt->channels)); if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) { - if (is_cdc_dma_port(dai_id)) + if (is_cdc_dma_port(dai_id)) { ret = copy_from_user_toio(dma_buf, buf, bytes); - else - ret = copy_from_user((void __force *)dma_buf, buf, bytes); + } else { + if (copy_from_user((void __force *)dma_buf, buf, bytes)) + ret = -EFAULT; + } } else if (substream->stream == SNDRV_PCM_STREAM_CAPTURE) { - if (is_cdc_dma_port(dai_id)) + if (is_cdc_dma_port(dai_id)) { ret = copy_to_user_fromio(buf, dma_buf, bytes); - else - ret = copy_to_user(buf, (void __force *)dma_buf, bytes); + } else { + if (copy_to_user(buf, (void __force *)dma_buf, bytes)) + ret = -EFAULT; + } } return ret; -- 2.20.1

4 years, 2 months

[PATCH] dma-fance: Add dma_fence_assert_in_signalling_section

by Daniel Vetter

Useful for checking for dma-fence signalling annotations since they don't quite nest as freely as we'd like to. Cc: Matthew Brost <matthew.brost(a)intel.com> Signed-off-by: Daniel Vetter <daniel.vetter(a)intel.com> Cc: Sumit Semwal <sumit.semwal(a)linaro.org> Cc: Gustavo Padovan <gustavo(a)padovan.org> Cc: "Christian König" <christian.koenig(a)amd.com> Cc: linux-media(a)vger.kernel.org Cc: linaro-mm-sig(a)lists.linaro.org --- drivers/dma-buf/dma-fence.c | 19 +++++++++++++++++++ include/linux/dma-fence.h | 2 ++ 2 files changed, 21 insertions(+) diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c index 066400ed8841..2b7c3fc965e6 100644 --- a/drivers/dma-buf/dma-fence.c +++ b/drivers/dma-buf/dma-fence.c @@ -307,6 +307,25 @@ bool dma_fence_begin_signalling(void) } EXPORT_SYMBOL(dma_fence_begin_signalling); +/** + * dma_fence_assert_in_signalling_section - check fence signalling annotations + * + * Since dma_fence_begin_signalling() and dma_fence_end_signalling() are built + * using lockdep annotations they have limitations on how freely they can be + * nested. Specifically, they cannot be on both inside and outside of locked + * sections, which in practice means the annotations often have to be pushed out + * to the top level callers. + * + * To ensure low-level functions are only called with the correction + * annotations, this function can be used to check for that. + */ +void dma_fence_assert_in_signalling_section(void) +{ + if (!in_atomic()) + lockdep_assert(lock_is_held(&dma_fence_lockdep_map)); +} +EXPORT_SYMBOL(dma_fence_assert_in_signalling_section); + /** * dma_fence_end_signalling - end a critical DMA fence signalling section * @cookie: opaque cookie from dma_fence_begin_signalling() diff --git a/include/linux/dma-fence.h b/include/linux/dma-fence.h index 775cdc0b4f24..7179a5692f72 100644 --- a/include/linux/dma-fence.h +++ b/include/linux/dma-fence.h @@ -356,6 +356,7 @@ dma_fence_get_rcu_safe(struct dma_fence __rcu **fencep) #ifdef CONFIG_LOCKDEP bool dma_fence_begin_signalling(void); +void dma_fence_assert_in_signalling_section(void); void dma_fence_end_signalling(bool cookie); void __dma_fence_might_wait(void); #else @@ -363,6 +364,7 @@ static inline bool dma_fence_begin_signalling(void) { return true; } +static inline void dma_fence_assert_in_signalling_section(void) {} static inline void dma_fence_end_signalling(bool cookie) {} static inline void __dma_fence_might_wait(void) {} #endif -- 2.34.1

4 years, 2 months

[PATCH] dma-buf: check the return value of kstrdup()

by xkernel.wang＠foxmail.com

From: Xiaoke Wang <xkernel.wang(a)foxmail.com> kstrdup() is a memory allocation function which can return NULL when some internaly memory errors happen. It is better to check the return value of it to prevent further wrong memory access. Signed-off-by: Xiaoke Wang <xkernel.wang(a)foxmail.com> --- drivers/dma-buf/selftest.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/dma-buf/selftest.c b/drivers/dma-buf/selftest.c index c60b694..2c29e2a 100644 --- a/drivers/dma-buf/selftest.c +++ b/drivers/dma-buf/selftest.c @@ -50,6 +50,9 @@ static bool apply_subtest_filter(const char *caller, const char *name) bool result = true; filter = kstrdup(__st_filter, GFP_KERNEL); + if (!filter) + return false; + for (sep = filter; (tok = strsep(&sep, ","));) { bool allow = true; char *sl; --

4 years, 2 months

[PATCH] drm/amdgpu: check vm bo eviction valuable at last

by Qiang Yu

Workstation application ANSA/META get this error dmesg: [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA (-16) This is caused by: 1. create a 256MB buffer in invisible VRAM 2. CPU map the buffer and access it causes vm_fault and try to move it to visible VRAM 3. force visible VRAM space and traverse all VRAM bos to check if evicting this bo is valuable 4. when checking a VM bo (in invisible VRAM), amdgpu_vm_evictable() will set amdgpu_vm->evicting, but latter due to not in visible VRAM, won't really evict it so not add it to amdgpu_vm->evicted 5. before next CS to clear the amdgpu_vm->evicting, user VM ops ioctl will pass amdgpu_vm_ready() (check amdgpu_vm->evicted) but fail in amdgpu_vm_bo_update_mapping() (check amdgpu_vm->evicting) and get this error log This error won't affect functionality as next CS will finish the waiting VM ops. But we'd better make the amdgpu_vm->evicting correctly reflact the vm status and clear the error log. Signed-off-by: Qiang Yu <qiang.yu(a)amd.com> --- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 85 ++++++++++++++----------- 1 file changed, 47 insertions(+), 38 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 5a32ee66d8c8..88a27911054f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -1306,45 +1306,11 @@ uint64_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm, return flags; } -/* - * amdgpu_ttm_bo_eviction_valuable - Check to see if we can evict a buffer - * object. - * - * Return true if eviction is sensible. Called by ttm_mem_evict_first() on - * behalf of ttm_bo_mem_force_space() which tries to evict buffer objects until - * it can find space for a new object and by ttm_bo_force_list_clean() which is - * used to clean out a memory space. - */ -static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo, - const struct ttm_place *place) +static bool amdgpu_ttm_mem_eviction_valuable(struct ttm_buffer_object *bo, + const struct ttm_place *place) { unsigned long num_pages = bo->resource->num_pages; struct amdgpu_res_cursor cursor; - struct dma_resv_list *flist; - struct dma_fence *f; - int i; - - /* Swapout? */ - if (bo->resource->mem_type == TTM_PL_SYSTEM) - return true; - - if (bo->type == ttm_bo_type_kernel && - !amdgpu_vm_evictable(ttm_to_amdgpu_bo(bo))) - return false; - - /* If bo is a KFD BO, check if the bo belongs to the current process. - * If true, then return false as any KFD process needs all its BOs to - * be resident to run successfully - */ - flist = dma_resv_shared_list(bo->base.resv); - if (flist) { - for (i = 0; i < flist->shared_count; ++i) { - f = rcu_dereference_protected(flist->shared[i], - dma_resv_held(bo->base.resv)); - if (amdkfd_fence_check_mm(f, current->mm)) - return false; - } - } switch (bo->resource->mem_type) { case AMDGPU_PL_PREEMPT: @@ -1377,10 +1343,53 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo, return false; default: - break; + return ttm_bo_eviction_valuable(bo, place); } +} - return ttm_bo_eviction_valuable(bo, place); +/* + * amdgpu_ttm_bo_eviction_valuable - Check to see if we can evict a buffer + * object. + * + * Return true if eviction is sensible. Called by ttm_mem_evict_first() on + * behalf of ttm_bo_mem_force_space() which tries to evict buffer objects until + * it can find space for a new object and by ttm_bo_force_list_clean() which is + * used to clean out a memory space. + */ +static bool amdgpu_ttm_bo_eviction_valuable(struct ttm_buffer_object *bo, + const struct ttm_place *place) +{ + struct dma_resv_list *flist; + struct dma_fence *f; + int i; + + /* Swapout? */ + if (bo->resource->mem_type == TTM_PL_SYSTEM) + return true; + + /* If bo is a KFD BO, check if the bo belongs to the current process. + * If true, then return false as any KFD process needs all its BOs to + * be resident to run successfully + */ + flist = dma_resv_shared_list(bo->base.resv); + if (flist) { + for (i = 0; i < flist->shared_count; ++i) { + f = rcu_dereference_protected(flist->shared[i], + dma_resv_held(bo->base.resv)); + if (amdkfd_fence_check_mm(f, current->mm)) + return false; + } + } + + /* Check by different mem type. */ + if (!amdgpu_ttm_mem_eviction_valuable(bo, place)) + return false; + + /* VM bo should be checked at last because it will mark VM evicting. */ + if (bo->type == ttm_bo_type_kernel) + return amdgpu_vm_evictable(ttm_to_amdgpu_bo(bo)); + + return true; } static void amdgpu_ttm_vram_mm_access(struct amdgpu_device *adev, loff_t pos, -- 2.25.1

4 years, 2 months

ACCOUNT WARNING: Undelivered mails linaro-mm-sig@lists.linaro.org Kindly rectify

by Noreply＠lists.linaro.org

4 years, 2 months

[RFC v2 0/6] Proposal for a GPU cgroup controller

by T.J. Mercier

This patch series revisits the proposal for a GPU cgroup controller to track and limit memory allocations by various device/allocator subsystems. The patch series also contains a simple prototype to illustrate how Android intends to implement DMA-BUF allocator attribution using the GPU cgroup controller. The prototype does not include resource limit enforcements. Changelog: v2: See the previous revision of this change submitted by Hridya Valsaraju at: https://lore.kernel.org/all/20220115010622.3185921-1-hridya@google.com/ Move dma-buf cgroup charge transfer from a dma_buf_op defined by every heap to a single dma-buf function for all heaps per Daniel Vetter and Christian König. Pointers to struct gpucg and struct gpucg_device tracking the current associations were added to the dma_buf struct to achieve this. Fix incorrect Kconfig help section indentation per Randy Dunlap. History of the GPU cgroup controller ==================================== The GPU/DRM cgroup controller came into being when a consensus[1] was reached that the resources it tracked were unsuitable to be integrated into memcg. Originally, the proposed controller was specific to the DRM subsystem and was intended to track GEM buffers and GPU-specific resources[2]. In order to help establish a unified memory accounting model for all GPU and all related subsystems, Daniel Vetter put forth a suggestion to move it out of the DRM subsystem so that it can be used by other DMA-BUF exporters as well[3]. This RFC proposes an interface that does the same. [1]: https://patchwork.kernel.org/project/dri-devel/cover/20190501140438.9506-1-… [2]: https://lore.kernel.org/amd-gfx/20210126214626.16260-1-brian.welty@intel.co… [3]: https://lore.kernel.org/amd-gfx/YCVOl8%2F87bqRSQei@phenom.ffwll.local/ T.J. Mercier (6): gpu: rfc: Proposal for a GPU cgroup controller cgroup: gpu: Add a cgroup controller for allocator attribution of GPU memory dmabuf: Use the GPU cgroup charge/uncharge APIs dmabuf: heaps: export system_heap buffers with GPU cgroup charging dmabuf: Add gpu cgroup charge transfer function android: binder: Add a buffer flag to relinquish ownership of fds Documentation/gpu/rfc/gpu-cgroup.rst | 195 +++++++++++++++++ Documentation/gpu/rfc/index.rst | 4 + drivers/android/binder.c | 26 +++ drivers/dma-buf/dma-buf.c | 100 +++++++++ drivers/dma-buf/dma-heap.c | 27 +++ drivers/dma-buf/heaps/system_heap.c | 3 + include/linux/cgroup_gpu.h | 127 +++++++++++ include/linux/cgroup_subsys.h | 4 + include/linux/dma-buf.h | 22 +- include/linux/dma-heap.h | 11 + include/uapi/linux/android/binder.h | 1 + init/Kconfig | 7 + kernel/cgroup/Makefile | 1 + kernel/cgroup/gpu.c | 304 +++++++++++++++++++++++++++ 14 files changed, 830 insertions(+), 2 deletions(-) create mode 100644 Documentation/gpu/rfc/gpu-cgroup.rst create mode 100644 include/linux/cgroup_gpu.h create mode 100644 kernel/cgroup/gpu.c -- 2.35.1.265.g69c8d7142f-goog

4 years, 2 months

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Linaro-mm-sig