On Tue, 2026-07-07 at 14:52 +0200, Christian König wrote: > On 7/7/26 11:53, Thomas Hellström wrote: > > On Tue, 2026-07-07 at 10:48 +0200, Christian König wrote: > > > On 7/7/26 08:56, Thomas Hellström wrote: > > > > On Mon, 2026-07-06 at 20:23 +0200, Christian König wrote: > > > > > On 7/6/26 19:01, Thomas Hellström wrote: > ... > > > > > > The problem is that locking the dma_resv object for release is > > > mandatory not only for proper DMA-buf import handling but for > > > tons of > > > other use cases as well. In other words DMA-buf imports, page > > > tables, > > > per process buffers who use the same dma_resv object as the root > > > PD > > > etc... > > > > > > By now we I don't think we can't avoid locking the dma_resv > > > object in > > > the delete path, so a completeion event or any other method which > > > blocks for resource cleanup without deadlock detection won't work > > > as > > > far as I can see. > > > > So in the TTM code itself we have only trylocks, so that works. For > > the > > other ones, trylocking the individualized resv in release_notify() > > or > > delete_mem_notify() should cover all / most cases except the dma- > > buf > > one? xe is doing exactly that. For imported dma-bufs it might even > > possible to release the resource before cleaning up the bo > > attachment. > > > > One problem remains for shrinkers, though, The memory allocation in > > individualization becomes a problem. We'd either need to annotate > > those > > allocations with PF_MEMALLOC flags or skip shrinking zombies. > > That still sounds like a massive hack to me.
What is a massive hack? The allocation modification or the _resv trylock? The latter is already done in the code, and using for example memalloc_nofs_save() will avoid the problematic reclaims so that we might trade stalling memory reclaims for explicit syncs on memory pressure. > > Especially the release notification is something I wanted to move > under the correct lock (_resv) in the long term because drivers tend > to mess that up. A _resv trylock would always succeed. And we're already doing that. > > > > > Since it's a layer violation to pass a drm_exec in an augmented > > > > dma-buf > > > > map we will be needing something at the dma-buf level to handle > > > > that, > > > > and given that, it also makes sense to base the eviction > > > > mechanism > > > > on > > > > it. > > > > > > Well I disagree. The DMA-buf invalidation notification > > > intentionally > > > doesn't have a ww_acquire_ctx. > > > > > > Background is that when this is called exporters have often > > > already > > > passed the point of no return and the importers just *must* act > > > on > > > the request and can't say 'Na, sorry I can't take the locks I > > > need'. > > > > I'm not referring to the invalidation. I'm thinking of the > > following > > sequence: > > > > xe_bo_validate() // Importer. Finds it's an evicted dma-buf. > > dma_buf_map(); > > xe_bo_validate(); // Exporter. Short on memory. > > ttm_bo_evict(); // Exporter wants to evict but lacks a > > ticket > > // and is limited to trylock and loses. > > > > The remedy is to optionally pass a ticket in an augmented > > dma_buf_map(), but then it can't be a drm_exec. dma_buf > > invalidation > > stays as it is. > > Hui? dma_buf_map_attachment() is called with the reservation lock > held for quite a while now. > > If a ticket is used for locking or not is up to the importer. And > that TTM eviction is limited to trylock is what we try to solve with > this patch set. > > So I absolutely don't get what you try to solve here. So how does the exporter, that performs the eviction of its own VRAM to satisfy the VRAM allocation of map_attachment(), get hold of the importer's drm_exec that it needs for the eviction ww_transaction? Aside from fishing out of the ww_mutex itself and assume it's always part of a drm_exec, which is IMO a complete no-go. You need to pass it through an map_attachment call modified to carry the drm_exec, except it can't be a drm_exec because that creates a dma-buf dependency on drm. Or translated to code: This drm_exec here: https://elixir.bootlin.com/linux/v7.1.2/source/drivers/gpu/drm/xe/xe_dma_buf.c#L103 Needs to be the importer's drm_exec so the exporter can pass it to TTM when it calls xe_bo_validate(). This is actually the main argument for introducing the dma_buf transaction helper to begin with. The ability to sidestep the ttm refcount removal is just an additional opportunity. > > > > > The ability to reuse the TTM refcount for now would just be a > > > > way > > > > to > > > > separate out the zombie handling from this series until we can > > > > do > > > > it > > > > properly. After all this series is about enabling sleeping ww > > > > locks. > > > > > > Well the problem is that I don't see how it should work with the > > > TTM > > > refcount. > > > > > > This basically means that for each lock operation has an ops > > > table > > > with put/get, in that case it would just easier to make the > > > ww_mutex > > > ref counted. > > > > Each object type embedding a ww_mutex needs an ops table pointed to > > by > > the lock operation. But get() is not needed, only put(). (See the > > RFC), > > I don't think that this is a good ide at all. This clearly just > complicates things even further. > > > If we refcount the ww mutex, instead it would need a separate > > destructor for each object type. We'd save a pointer chase on put > > but > > need extra care on object creation and destruction, so I'm not sure > > it's easier. > > As far as I can see that approach doesn't needs an ops pointer at > all. You just save the contented ww_mutex object in the drm_exec > container instead of the full GEM object reference. > > Let me give that a try a see how far we get. But that doesn't solve the dma_buf problem? You still have the dma_buf dependency on drm_exec? FWIW, I asked Claude to rebase this series on the dma_buf transaction helper: https://gitlab.freedesktop.org/thomash/xe-vibe/-/tree/ttm_exec2 Except for the dma_buf transaction helper this yields a notably smaller diff in the drm tree. Thomas > > Regards, > Christian. > > > > > /Thomas > > > > > > > > > > Regards, > > > Christian. > > > > > > > > > > > Thanks, > > > > Thomas > > > > > > > > > > > > > > > > > > Regards, > > > > > Christian. > > > > > > > > > > > > > > > > > > > > > > Also for reference: (Section at the end and follow-up > > > > > > messages) > > > > > > https://lore.kernel.org/all/[email protected]/ > > > > > > > > > > > > Thanks, > > > > > > Thomas > > > > > > > > > > > > > > > > > > > Regards, > > > > > > > Christian. > > > > > > > > > > > > > > > > > > > > > > > /Thomas > > > > > > > > > > > > > > > > > > > > > > > > > > Signed-off-by: tChristian König > > > > > > > > > <[email protected]> > > > > > > > > > Signed-off-by: Natalie Vock <[email protected]> > > > > > > > > > --- > > > > > > > > > drivers/gpu/drm/i915/gem/i915_gem_ttm.c | > > > > > > > > > 35 > > > > > > > > > +++-- > > > > > > > > > - > > > > > > > > > drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c | > > > > > > > > > 8 > > > > > > > > > +- > > > > > > > > > drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c | > > > > > > > > > 2 - > > > > > > > > > drivers/gpu/drm/ttm/ttm_bo.c | > > > > > > > > > 135 > > > > > > > > > +++++++++++-- > > > > > > > > > ---------- > > > > > > > > > drivers/gpu/drm/ttm/ttm_bo_internal.h | > > > > > > > > > 17 > > > > > > > > > ++- > > > > > > > > > drivers/gpu/drm/ttm/ttm_bo_util.c | > > > > > > > > > 15 > > > > > > > > > ++- > > > > > > > > > drivers/gpu/drm/xe/xe_bo.c | > > > > > > > > > 2 > > > > > > > > > +- > > > > > > > > > include/drm/ttm/ttm_bo.h | > > > > > > > > > 9 - > > > > > > > > > - > > > > > > > > > 8 files changed, 111 insertions(+), 112 deletions(-) > > > > > > > > > > > > > > > > > > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c > > > > > > > > > b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c > > > > > > > > > index df3fcc2b1248e..642296602de69 100644 > > > > > > > > > --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c > > > > > > > > > +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c > > > > > > > > > @@ -939,7 +939,7 @@ void i915_ttm_adjust_lru(struct > > > > > > > > > drm_i915_gem_object *obj) > > > > > > > > > * Don't manipulate the TTM LRUs while in > > > > > > > > > TTM bo > > > > > > > > > destruction. > > > > > > > > > * We're called through > > > > > > > > > i915_ttm_delete_mem_notify(). > > > > > > > > > */ > > > > > > > > > - if (!kref_read(&bo->kref)) > > > > > > > > > + if (!kref_read(&bo->base.refcount)) > > > > > > > > > return; > > > > > > > > > > > > > > > > > > /* > > > > > > > > > @@ -957,30 +957,21 @@ void i915_ttm_adjust_lru(struct > > > > > > > > > drm_i915_gem_object *obj) > > > > > > > > > * > > > > > > > > > * TODO: consider maybe also bumping the > > > > > > > > > shrinker > > > > > > > > > list > > > > > > > > > here > > > > > > > > > when we have > > > > > > > > > * already unpinned it, which should give us > > > > > > > > > something > > > > > > > > > more > > > > > > > > > like an LRU. > > > > > > > > > - * > > > > > > > > > - * TODO: There is a small window of > > > > > > > > > opportunity > > > > > > > > > for > > > > > > > > > this > > > > > > > > > function to > > > > > > > > > - * get called from eviction after we've > > > > > > > > > dropped > > > > > > > > > the > > > > > > > > > last > > > > > > > > > GEM > > > > > > > > > refcount, > > > > > > > > > - * but before the TTM deleted flag is set on > > > > > > > > > the > > > > > > > > > object. > > > > > > > > > Avoid > > > > > > > > > - * adjusting the shrinker list in such > > > > > > > > > cases, > > > > > > > > > since > > > > > > > > > the > > > > > > > > > object is > > > > > > > > > - * not available to the shrinker anyway due > > > > > > > > > to > > > > > > > > > its > > > > > > > > > zero > > > > > > > > > refcount. > > > > > > > > > - * To fix this properly we should move to a > > > > > > > > > TTM > > > > > > > > > shrinker > > > > > > > > > LRU > > > > > > > > > list for > > > > > > > > > - * these objects. > > > > > > > > > */ > > > > > > > > > - if (kref_get_unless_zero(&obj- > > > > > > > > > >base.refcount)) { > > > > > > > > > - if (shrinkable != obj- > > > > > > > > > > mm.ttm_shrinkable) { > > > > > > > > > - if (shrinkable) { > > > > > > > > > - if (obj->mm.madv == > > > > > > > > > I915_MADV_WILLNEED) > > > > > > > > > - > > > > > > > > > __i915_gem_o > > > > > > > > > bjec > > > > > > > > > t_ma > > > > > > > > > ke_s > > > > > > > > > hrinkable(obj); > > > > > > > > > - else > > > > > > > > > - > > > > > > > > > __i915_gem_o > > > > > > > > > bjec > > > > > > > > > t_ma > > > > > > > > > ke_p > > > > > > > > > urgeable(obj); > > > > > > > > > - } else { > > > > > > > > > - > > > > > > > > > i915_gem_object_make > > > > > > > > > _uns > > > > > > > > > hrin > > > > > > > > > kabl > > > > > > > > > e(obj); > > > > > > > > > - } > > > > > > > > > - > > > > > > > > > - obj->mm.ttm_shrinkable = > > > > > > > > > shrinkable; > > > > > > > > > + i915_gem_object_get(obj); > > > > > > > > > + if (shrinkable != obj->mm.ttm_shrinkable) { > > > > > > > > > + if (shrinkable) { > > > > > > > > > + if (obj->mm.madv == > > > > > > > > > I915_MADV_WILLNEED) > > > > > > > > > + __i915_gem_object_ma > > > > > > > > > ke_s > > > > > > > > > hrin > > > > > > > > > kabl > > > > > > > > > e(ob > > > > > > > > > j); > > > > > > > > > + else > > > > > > > > > + __i915_gem_object_ma > > > > > > > > > ke_p > > > > > > > > > urge > > > > > > > > > able > > > > > > > > > (obj > > > > > > > > > ); > > > > > > > > > + } else { > > > > > > > > > + i915_gem_object_make_unshrin > > > > > > > > > kabl > > > > > > > > > e(ob > > > > > > > > > j); > > > > > > > > > } > > > > > > > > > - i915_gem_object_put(obj); > > > > > > > > > + > > > > > > > > > + obj->mm.ttm_shrinkable = shrinkable; > > > > > > > > > } > > > > > > > > > + i915_gem_object_put(obj); > > > > > > > > > > > > > > > > > > /* > > > > > > > > > * Put on the correct LRU list depending on > > > > > > > > > the > > > > > > > > > MADV > > > > > > > > > status > > > > > > > > > diff --git > > > > > > > > > a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > > > > > > > > > b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > > > > > > > > > index 56ad8ef325840..904cb4da6c9b3 100644 > > > > > > > > > --- > > > > > > > > > a/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > > > > > > > > > +++ > > > > > > > > > b/drivers/gpu/drm/ttm/tests/ttm_bo_validate_test.c > > > > > > > > > @@ -127,7 +127,7 @@ static void > > > > > > > > > ttm_bo_init_reserved_sys_man(struct > > > > > > > > > kunit *test) > > > > > > > > > dma_resv_unlock(bo->base.resv); > > > > > > > > > > > > > > > > > > KUNIT_EXPECT_EQ(test, err, 0); > > > > > > > > > - KUNIT_EXPECT_EQ(test, kref_read(&bo->kref), > > > > > > > > > 1); > > > > > > > > > + KUNIT_EXPECT_EQ(test, kref_read(&bo- > > > > > > > > > > base.refcount), > > > > > > > > > 1); > > > > > > > > > KUNIT_EXPECT_PTR_EQ(test, bo->bdev, priv- > > > > > > > > > > ttm_dev); > > > > > > > > > KUNIT_EXPECT_EQ(test, bo->type, bo_type); > > > > > > > > > KUNIT_EXPECT_EQ(test, bo->page_alignment, > > > > > > > > > PAGE_SIZE); > > > > > > > > > @@ -176,7 +176,7 @@ static void > > > > > > > > > ttm_bo_init_reserved_mock_man(struct > > > > > > > > > kunit *test) > > > > > > > > > dma_resv_unlock(bo->base.resv); > > > > > > > > > > > > > > > > > > KUNIT_EXPECT_EQ(test, err, 0); > > > > > > > > > - KUNIT_EXPECT_EQ(test, kref_read(&bo->kref), > > > > > > > > > 1); > > > > > > > > > + KUNIT_EXPECT_EQ(test, kref_read(&bo- > > > > > > > > > > base.refcount), > > > > > > > > > 1); > > > > > > > > > KUNIT_EXPECT_PTR_EQ(test, bo->bdev, priv- > > > > > > > > > > ttm_dev); > > > > > > > > > KUNIT_EXPECT_EQ(test, bo->type, bo_type); > > > > > > > > > KUNIT_EXPECT_EQ(test, ctx.bytes_moved, > > > > > > > > > size); > > > > > > > > > @@ -969,6 +969,8 @@ static void > > > > > > > > > ttm_bo_validate_allowed_only_evict(struct kunit > > > > > > > > > *test) > > > > > > > > > ttm_mock_manager_fini(priv->ttm_dev, > > > > > > > > > mem_multihop); > > > > > > > > > } > > > > > > > > > > > > > > > > > > +extern const struct drm_gem_object_funcs > > > > > > > > > ttm_deleted_object_funcs; > > > > > > > > > + > > > > > > > > > static void ttm_bo_validate_deleted_evict(struct > > > > > > > > > kunit > > > > > > > > > *test) > > > > > > > > > { > > > > > > > > > struct ttm_operation_ctx ctx_init = { }, > > > > > > > > > ctx_val = > > > > > > > > > { }; > > > > > > > > > @@ -999,7 +1001,7 @@ static void > > > > > > > > > ttm_bo_validate_deleted_evict(struct > > > > > > > > > kunit *test) > > > > > > > > > KUNIT_EXPECT_EQ(test, > > > > > > > > > ttm_resource_manager_usage(man), > > > > > > > > > big); > > > > > > > > > > > > > > > > > > dma_resv_unlock(bo_big->base.resv); > > > > > > > > > - bo_big->deleted = true; > > > > > > > > > + bo_big->base.funcs = > > > > > > > > > &ttm_deleted_object_funcs; > > > > > > > > > > > > > > > > > > bo_small = ttm_bo_kunit_init(test, test- > > > > > > > > > >priv, > > > > > > > > > small, > > > > > > > > > NULL); > > > > > > > > > bo_small->type = bo_type; > > > > > > > > > diff --git > > > > > > > > > a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c > > > > > > > > > b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c > > > > > > > > > index 5cfe8f3f80d75..b7ab19e0e4b2b 100644 > > > > > > > > > --- a/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c > > > > > > > > > +++ b/drivers/gpu/drm/ttm/tests/ttm_kunit_helpers.c > > > > > > > > > @@ -189,8 +189,6 @@ struct ttm_buffer_object > > > > > > > > > *ttm_bo_kunit_init(struct kunit *test, > > > > > > > > > bo->bdev = devs->ttm_dev; > > > > > > > > > bo->destroy = dummy_ttm_bo_destroy; > > > > > > > > > > > > > > > > > > - kref_init(&bo->kref); > > > > > > > > > - > > > > > > > > > return bo; > > > > > > > > > } > > > > > > > > > EXPORT_SYMBOL_GPL(ttm_bo_kunit_init); > > > > > > > > > diff --git a/drivers/gpu/drm/ttm/ttm_bo.c > > > > > > > > > b/drivers/gpu/drm/ttm/ttm_bo.c > > > > > > > > > index 3980f376e3ba4..2b470c1746f60 100644 > > > > > > > > > --- a/drivers/gpu/drm/ttm/ttm_bo.c > > > > > > > > > +++ b/drivers/gpu/drm/ttm/ttm_bo.c > > > > > > > > > @@ -246,88 +246,84 @@ static void > > > > > > > > > ttm_bo_delayed_delete(struct > > > > > > > > > work_struct *work) > > > > > > > > > ttm_bo_put(bo); > > > > > > > > > } > > > > > > > > > > > > > > > > > > -static void ttm_bo_release(struct kref *kref) > > > > > > > > > +/* > > > > > > > > > + * All other callbacks should never ever be called > > > > > > > > > on a > > > > > > > > > deleted > > > > > > > > > TTM > > > > > > > > > object. > > > > > > > > > + */ > > > > > > > > > +const struct drm_gem_object_funcs > > > > > > > > > ttm_deleted_object_funcs = > > > > > > > > > { > > > > > > > > > + .free = ttm_bo_free > > > > > > > > > +}; > > > > > > > > > +EXPORT_SYMBOL_FOR_TESTS_ONLY(ttm_deleted_object_func > > > > > > > > > s); > > > > > > > > > + > > > > > > > > > +/* Returns true if the BO is about to get deleted */ > > > > > > > > > +static bool ttm_bo_is_zombie(struct > > > > > > > > > ttm_buffer_object > > > > > > > > > *bo) > > > > > > > > > +{ > > > > > > > > > + return bo->base.funcs == > > > > > > > > > &ttm_deleted_object_funcs; > > > > > > > > > +} > > > > > > > > > + > > > > > > > > > +void ttm_bo_fini(struct ttm_buffer_object *bo) > > > > > > > > > { > > > > > > > > > - struct ttm_buffer_object *bo = > > > > > > > > > - container_of(kref, struct > > > > > > > > > ttm_buffer_object, > > > > > > > > > kref); > > > > > > > > > struct ttm_device *bdev = bo->bdev; > > > > > > > > > int ret; > > > > > > > > > > > > > > > > > > WARN_ON_ONCE(bo->pin_count); > > > > > > > > > WARN_ON_ONCE(bo->bulk_move); > > > > > > > > > > > > > > > > > > - if (!bo->deleted) { > > > > > > > > > - ret = ttm_bo_individualize_resv(bo); > > > > > > > > > - if (ret) { > > > > > > > > > - /* Last resort, if we fail > > > > > > > > > to > > > > > > > > > allocate > > > > > > > > > memory for the > > > > > > > > > - * fences block for the BO > > > > > > > > > to > > > > > > > > > become > > > > > > > > > idle > > > > > > > > > - */ > > > > > > > > > - dma_resv_wait_timeout(bo- > > > > > > > > > > base.resv, > > > > > > > > > - > > > > > > > > > DMA_RESV_USAGE_BOOKKEEP, false, > > > > > > > > > - 30 * > > > > > > > > > HZ); > > > > > > > > > - } > > > > > > > > > + ret = ttm_bo_individualize_resv(bo); > > > > > > > > > + if (ret) { > > > > > > > > > + /* Last resort, if we fail to > > > > > > > > > allocate > > > > > > > > > memory > > > > > > > > > for > > > > > > > > > the > > > > > > > > > + * fences block for the BO to become > > > > > > > > > idle > > > > > > > > > + */ > > > > > > > > > + dma_resv_wait_timeout(bo->base.resv, > > > > > > > > > DMA_RESV_USAGE_BOOKKEEP, > > > > > > > > > + false, 30 * > > > > > > > > > HZ); > > > > > > > > > + } > > > > > > > > > > > > > > > > > > - if (bdev->funcs->release_notify) > > > > > > > > > - bdev->funcs- > > > > > > > > > >release_notify(bo); > > > > > > > > > + if (bo->bdev->funcs->release_notify) > > > > > > > > > + bo->bdev->funcs->release_notify(bo); > > > > > > > > > > > > > > > > > > - drm_vma_offset_remove(bdev- > > > > > > > > > >vma_manager, > > > > > > > > > &bo- > > > > > > > > > > base.vma_node); > > > > > > > > > - ttm_mem_io_free(bdev, bo->resource); > > > > > > > > > + drm_vma_offset_remove(bdev->vma_manager, > > > > > > > > > &bo- > > > > > > > > > > base.vma_node); > > > > > > > > > + ttm_mem_io_free(bdev, bo->resource); > > > > > > > > > > > > > > > > > > - if (!dma_resv_test_signaled(&bo- > > > > > > > > > > base._resv, > > > > > > > > > - > > > > > > > > > DMA_RESV_USAGE_BOOKKEEP) > > > > > > > > > > > > > > > > > > > > - (want_init_on_free() && (bo->ttm > > > > > > > > > != > > > > > > > > > NULL)) > > > > > > > > > > > > > > > > > > > > - bo->type == ttm_bo_type_sg || > > > > > > > > > - !dma_resv_trylock(bo- > > > > > > > > > >base.resv)) { > > > > > > > > > - /* The BO is not idle, > > > > > > > > > resurrect > > > > > > > > > it > > > > > > > > > for > > > > > > > > > delayed destroy */ > > > > > > > > > - ttm_bo_flush_all_fences(bo); > > > > > > > > > - bo->deleted = true; > > > > > > > > > + if (!dma_resv_test_signaled(&bo->base._resv, > > > > > > > > > DMA_RESV_USAGE_BOOKKEEP) || > > > > > > > > > + (want_init_on_free() && (bo->ttm != > > > > > > > > > NULL)) > > > > > > > > > > > > > > > > > > > > + bo->type == ttm_bo_type_sg || > > > > > > > > > + !dma_resv_trylock(bo->base.resv)) { > > > > > > > > > + /* The BO is not idle, resurrect it > > > > > > > > > for > > > > > > > > > delayed > > > > > > > > > destroy */ > > > > > > > > > + ttm_bo_flush_all_fences(bo); > > > > > > > > > > > > > > > > > > - spin_lock(&bdev->lru_lock); > > > > > > > > > - > > > > > > > > > - /* > > > > > > > > > - * Make pinned bos > > > > > > > > > immediately > > > > > > > > > available > > > > > > > > > to > > > > > > > > > - * shrinkers, now that they > > > > > > > > > are > > > > > > > > > queued > > > > > > > > > for > > > > > > > > > - * destruction. > > > > > > > > > - * > > > > > > > > > - * FIXME: QXL is triggering > > > > > > > > > this. > > > > > > > > > Can be > > > > > > > > > removed when the > > > > > > > > > - * driver is fixed. > > > > > > > > > - */ > > > > > > > > > - if (bo->pin_count) { > > > > > > > > > - bo->pin_count = 0; > > > > > > > > > - > > > > > > > > > ttm_resource_move_to > > > > > > > > > _lru > > > > > > > > > _tai > > > > > > > > > l(bo- > > > > > > > > > > resource); > > > > > > > > > - } > > > > > > > > > + spin_lock(&bo->bdev->lru_lock); > > > > > > > > > > > > > > > > > > - kref_init(&bo->kref); > > > > > > > > > - spin_unlock(&bdev- > > > > > > > > > >lru_lock); > > > > > > > > > + /* > > > > > > > > > + * Make pinned bos immediately > > > > > > > > > available > > > > > > > > > to > > > > > > > > > + * shrinkers, now that they are > > > > > > > > > queued > > > > > > > > > for > > > > > > > > > + * destruction. > > > > > > > > > + * > > > > > > > > > + * FIXME: QXL is triggering this. > > > > > > > > > Can be > > > > > > > > > removed > > > > > > > > > when the > > > > > > > > > + * driver is fixed. > > > > > > > > > + */ > > > > > > > > > + if (bo->pin_count) { > > > > > > > > > + bo->pin_count = 0; > > > > > > > > > + ttm_resource_move_to_lru_tai > > > > > > > > > l(bo > > > > > > > > > - > > > > > > > > > > resource); > > > > > > > > > + } > > > > > > > > > > > > > > > > > > - INIT_WORK(&bo- > > > > > > > > > >delayed_delete, > > > > > > > > > ttm_bo_delayed_delete); > > > > > > > > > + kref_init(&bo->base.refcount); > > > > > > > > > + bo->base.funcs = > > > > > > > > > &ttm_deleted_object_funcs; > > > > > > > > > + spin_unlock(&bo->bdev->lru_lock); > > > > > > > > > > > > > > > > > > - /* Schedule the worker on > > > > > > > > > the > > > > > > > > > closest > > > > > > > > > NUMA > > > > > > > > > node. This > > > > > > > > > - * improves performance > > > > > > > > > since > > > > > > > > > system > > > > > > > > > memory > > > > > > > > > might be > > > > > > > > > - * cleared on free and that > > > > > > > > > is > > > > > > > > > best > > > > > > > > > done > > > > > > > > > on > > > > > > > > > a CPU core > > > > > > > > > - * close to it. > > > > > > > > > - */ > > > > > > > > > - queue_work_node(bdev- > > > > > > > > > >pool.nid, > > > > > > > > > bdev- > > > > > > > > > > wq, > > > > > > > > > &bo->delayed_delete); > > > > > > > > > - return; > > > > > > > > > - } > > > > > > > > > + INIT_WORK(&bo->delayed_delete, > > > > > > > > > ttm_bo_delayed_delete); > > > > > > > > > > > > > > > > > > + /* Schedule the worker on the > > > > > > > > > closest > > > > > > > > > NUMA > > > > > > > > > node. > > > > > > > > > This > > > > > > > > > + * improves performance since system > > > > > > > > > memory > > > > > > > > > might be > > > > > > > > > + * cleared on free and that is best > > > > > > > > > done > > > > > > > > > on > > > > > > > > > a > > > > > > > > > CPU > > > > > > > > > core > > > > > > > > > + * close to it. > > > > > > > > > + */ > > > > > > > > > + queue_work_node(bdev->pool.nid, > > > > > > > > > bdev- > > > > > > > > > > wq, > > > > > > > > > &bo- > > > > > > > > > > delayed_delete); > > > > > > > > > + } else { > > > > > > > > > ttm_bo_cleanup_memtype_use(bo); > > > > > > > > > dma_resv_unlock(bo->base.resv); > > > > > > > > > - } > > > > > > > > > > > > > > > > > > - atomic_dec(&ttm_glob.bo_count); > > > > > > > > > - bo->destroy(bo); > > > > > > > > > -} > > > > > > > > > - > > > > > > > > > -/* TODO: remove! */ > > > > > > > > > -void ttm_bo_put(struct ttm_buffer_object *bo) > > > > > > > > > -{ > > > > > > > > > - kref_put(&bo->kref, ttm_bo_release); > > > > > > > > > -} > > > > > > > > > - > > > > > > > > > -void ttm_bo_fini(struct ttm_buffer_object *bo) > > > > > > > > > -{ > > > > > > > > > - ttm_bo_put(bo); > > > > > > > > > + atomic_dec(&ttm_glob.bo_count); > > > > > > > > > + bo->destroy(bo); > > > > > > > > > + } > > > > > > > > > } > > > > > > > > > EXPORT_SYMBOL(ttm_bo_fini); > > > > > > > > > > > > > > > > > > @@ -470,7 +466,7 @@ int ttm_bo_evict_first(struct > > > > > > > > > ttm_device > > > > > > > > > *bdev, > > > > > > > > > struct ttm_resource_manager *man > > > > > > > > > if (!bo->resource || bo->resource->mem_type > > > > > > > > > != > > > > > > > > > mem_type) > > > > > > > > > goto out_bo_moved; > > > > > > > > > > > > > > > > > > - if (bo->deleted) { > > > > > > > > > + if (ttm_bo_is_zombie(bo)) { > > > > > > > > > ret = ttm_bo_wait_ctx(bo, ctx); > > > > > > > > > if (!ret) > > > > > > > > > ttm_bo_cleanup_memtype_use(b > > > > > > > > > o); > > > > > > > > > @@ -524,7 +520,7 @@ static s64 ttm_bo_evict_cb(struct > > > > > > > > > ttm_lru_walk > > > > > > > > > *walk, struct ttm_buffer_object * > > > > > > > > > if (bo->pin_count || !bo->bdev->funcs- > > > > > > > > > > eviction_valuable(bo, > > > > > > > > > evict_walk->place)) > > > > > > > > > return 0; > > > > > > > > > > > > > > > > > > - if (bo->deleted) { > > > > > > > > > + if (ttm_bo_is_zombie(bo)) { > > > > > > > > > lret = ttm_bo_wait_ctx(bo, walk- > > > > > > > > > > arg.ctx); > > > > > > > > > if (!lret) > > > > > > > > > ttm_bo_cleanup_memtype_use(b > > > > > > > > > o); > > > > > > > > > @@ -624,7 +620,6 @@ static int > > > > > > > > > ttm_bo_evict_alloc(struct > > > > > > > > > ttm_device > > > > > > > > > *bdev, > > > > > > > > > void ttm_bo_pin(struct ttm_buffer_object *bo) > > > > > > > > > { > > > > > > > > > dma_resv_assert_held(bo->base.resv); > > > > > > > > > - WARN_ON_ONCE(!kref_read(&bo->kref)); > > > > > > > > > spin_lock(&bo->bdev->lru_lock); > > > > > > > > > if (bo->resource) > > > > > > > > > ttm_resource_del_bulk_move(bo- > > > > > > > > > >resource, > > > > > > > > > bo); > > > > > > > > > @@ -643,7 +638,6 @@ EXPORT_SYMBOL(ttm_bo_pin); > > > > > > > > > void ttm_bo_unpin(struct ttm_buffer_object *bo) > > > > > > > > > { > > > > > > > > > dma_resv_assert_held(bo->base.resv); > > > > > > > > > - WARN_ON_ONCE(!kref_read(&bo->kref)); > > > > > > > > > if (WARN_ON_ONCE(!bo->pin_count)) > > > > > > > > > return; > > > > > > > > > > > > > > > > > > @@ -934,7 +928,6 @@ int ttm_bo_init_reserved(struct > > > > > > > > > ttm_device > > > > > > > > > *bdev, > > > > > > > > > struct ttm_buffer_object *bo, > > > > > > > > > { > > > > > > > > > int ret; > > > > > > > > > > > > > > > > > > - kref_init(&bo->kref); > > > > > > > > > bo->bdev = bdev; > > > > > > > > > bo->type = type; > > > > > > > > > bo->page_alignment = alignment; > > > > > > > > > @@ -1131,8 +1124,8 @@ ttm_bo_swapout_cb(struct > > > > > > > > > ttm_lru_walk > > > > > > > > > *walk, > > > > > > > > > struct ttm_buffer_object *bo) > > > > > > > > > goto out; > > > > > > > > > } > > > > > > > > > > > > > > > > > > - if (bo->deleted) { > > > > > > > > > - pgoff_t num_pages = tt->num_pages; > > > > > > > > > + if (ttm_bo_is_zombie(bo)) { > > > > > > > > > + pgoff_t num_pages = bo->ttm- > > > > > > > > > >num_pages; > > > > > > > > > > > > > > > > > > ret = ttm_bo_wait_ctx(bo, ctx); > > > > > > > > > if (ret) > > > > > > > > > diff --git a/drivers/gpu/drm/ttm/ttm_bo_internal.h > > > > > > > > > b/drivers/gpu/drm/ttm/ttm_bo_internal.h > > > > > > > > > index e0d48eac74b03..ded2a47be0bcb 100644 > > > > > > > > > --- a/drivers/gpu/drm/ttm/ttm_bo_internal.h > > > > > > > > > +++ b/drivers/gpu/drm/ttm/ttm_bo_internal.h > > > > > > > > > @@ -27,6 +27,14 @@ > > > > > > > > > > > > > > > > > > #include <drm/ttm/ttm_bo.h> > > > > > > > > > > > > > > > > > > +static inline void ttm_bo_free(struct drm_gem_object > > > > > > > > > *gobj) > > > > > > > > > +{ > > > > > > > > > + struct ttm_buffer_object *bo = > > > > > > > > > container_of(gobj, > > > > > > > > > typeof(*bo), base); > > > > > > > > > + > > > > > > > > > + atomic_dec(&ttm_glob.bo_count); > > > > > > > > > + bo->destroy(bo); > > > > > > > > > +} > > > > > > > > > + > > > > > > > > > /** > > > > > > > > > * ttm_bo_get - reference a struct ttm_buffer_object > > > > > > > > > * > > > > > > > > > @@ -34,7 +42,7 @@ > > > > > > > > > */ > > > > > > > > > static inline void ttm_bo_get(struct > > > > > > > > > ttm_buffer_object > > > > > > > > > *bo) > > > > > > > > > { > > > > > > > > > - kref_get(&bo->kref); > > > > > > > > > + drm_gem_object_get(&bo->base); > > > > > > > > > } > > > > > > > > > > > > > > > > > > /** > > > > > > > > > @@ -50,11 +58,14 @@ static inline void > > > > > > > > > ttm_bo_get(struct > > > > > > > > > ttm_buffer_object *bo) > > > > > > > > > static inline __must_check struct ttm_buffer_object > > > > > > > > > * > > > > > > > > > ttm_bo_get_unless_zero(struct ttm_buffer_object *bo) > > > > > > > > > { > > > > > > > > > - if (!kref_get_unless_zero(&bo->kref)) > > > > > > > > > + if (!kref_get_unless_zero(&bo- > > > > > > > > > >base.refcount)) > > > > > > > > > return NULL; > > > > > > > > > return bo; > > > > > > > > > } > > > > > > > > > > > > > > > > > > -void ttm_bo_put(struct ttm_buffer_object *bo); > > > > > > > > > +static inline void ttm_bo_put(struct > > > > > > > > > ttm_buffer_object > > > > > > > > > *bo) > > > > > > > > > +{ > > > > > > > > > + drm_gem_object_put(&bo->base); > > > > > > > > > +} > > > > > > > > > > > > > > > > > > #endif > > > > > > > > > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c > > > > > > > > > b/drivers/gpu/drm/ttm/ttm_bo_util.c > > > > > > > > > index 3e3c201a02226..7ed085adf1c9b 100644 > > > > > > > > > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > > > > > > > > > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > > > > > > > > > @@ -41,6 +41,18 @@ > > > > > > > > > > > > > > > > > > #include "ttm_bo_internal.h" > > > > > > > > > > > > > > > > > > +static void ttm_transfer_object_free(struct > > > > > > > > > drm_gem_object > > > > > > > > > *obj) > > > > > > > > > +{ > > > > > > > > > + struct ttm_buffer_object *bo = > > > > > > > > > + container_of(obj, struct > > > > > > > > > ttm_buffer_object, > > > > > > > > > base); > > > > > > > > > + > > > > > > > > > + ttm_bo_fini(bo); > > > > > > > > > +} > > > > > > > > > + > > > > > > > > > +const struct drm_gem_object_funcs > > > > > > > > > ttm_transfer_object_funcs > > > > > > > > > = { > > > > > > > > > + .free = ttm_transfer_object_free, > > > > > > > > > +}; > > > > > > > > > + > > > > > > > > > struct ttm_transfer_obj { > > > > > > > > > struct ttm_buffer_object base; > > > > > > > > > struct ttm_buffer_object *bo; > > > > > > > > > @@ -247,7 +259,8 @@ static int > > > > > > > > > ttm_buffer_object_transfer(struct > > > > > > > > > ttm_buffer_object *bo, > > > > > > > > > atomic_inc(&ttm_glob.bo_count); > > > > > > > > > drm_vma_node_reset(&fbo- > > > > > > > > > >base.base.vma_node); > > > > > > > > > > > > > > > > > > - kref_init(&fbo->base.kref); > > > > > > > > > + kref_init(&fbo->base.base.refcount); > > > > > > > > > + fbo->base.base.funcs = > > > > > > > > > &ttm_transfer_object_funcs; > > > > > > > > > fbo->base.destroy = &ttm_transfered_destroy; > > > > > > > > > fbo->base.pin_count = 0; > > > > > > > > > if (bo->type != ttm_bo_type_sg) > > > > > > > > > diff --git a/drivers/gpu/drm/xe/xe_bo.c > > > > > > > > > b/drivers/gpu/drm/xe/xe_bo.c > > > > > > > > > index 85e6d9a0f575b..5843f850339c7 100644 > > > > > > > > > --- a/drivers/gpu/drm/xe/xe_bo.c > > > > > > > > > +++ b/drivers/gpu/drm/xe/xe_bo.c > > > > > > > > > @@ -1651,7 +1651,7 @@ static bool > > > > > > > > > xe_ttm_bo_lock_in_destructor(struct > > > > > > > > > ttm_buffer_object *ttm_bo) > > > > > > > > > struct xe_device *xe = > > > > > > > > > ttm_to_xe_device(ttm_bo- > > > > > > > > > > bdev); > > > > > > > > > bool locked; > > > > > > > > > > > > > > > > > > - xe_assert(xe, !kref_read(&ttm_bo->kref)); > > > > > > > > > + xe_assert(xe, !kref_read(&ttm_bo- > > > > > > > > > > base.refcount)); > > > > > > > > > > > > > > > > > > /* > > > > > > > > > * We can typically only race with TTM > > > > > > > > > trylocking > > > > > > > > > under > > > > > > > > > the > > > > > > > > > diff --git a/include/drm/ttm/ttm_bo.h > > > > > > > > > b/include/drm/ttm/ttm_bo.h > > > > > > > > > index 8310bc3d55f90..1eae9eea5ff32 100644 > > > > > > > > > --- a/include/drm/ttm/ttm_bo.h > > > > > > > > > +++ b/include/drm/ttm/ttm_bo.h > > > > > > > > > @@ -78,11 +78,8 @@ enum ttm_bo_type { > > > > > > > > > * @type: The bo type. > > > > > > > > > * @page_alignment: Page alignment. > > > > > > > > > * @destroy: Destruction function. If NULL, kfree is > > > > > > > > > used. > > > > > > > > > - * @kref: Reference count of this buffer object. > > > > > > > > > When > > > > > > > > > this > > > > > > > > > refcount > > > > > > > > > reaches > > > > > > > > > - * zero, the object is destroyed or put on the > > > > > > > > > delayed > > > > > > > > > delete > > > > > > > > > list. > > > > > > > > > * @resource: structure describing current > > > > > > > > > placement. > > > > > > > > > * @ttm: TTM structure holding system pages. > > > > > > > > > - * @deleted: True if the object is only a zombie and > > > > > > > > > already > > > > > > > > > deleted. > > > > > > > > > * @bulk_move: The bulk move object. > > > > > > > > > * @priority: Priority for LRU, BOs with lower > > > > > > > > > priority > > > > > > > > > are > > > > > > > > > evicted > > > > > > > > > first. > > > > > > > > > * @pin_count: Pin count. > > > > > > > > > @@ -109,17 +106,11 @@ struct ttm_buffer_object { > > > > > > > > > uint32_t page_alignment; > > > > > > > > > void (*destroy) (struct ttm_buffer_object > > > > > > > > > *); > > > > > > > > > > > > > > > > > > - /* > > > > > > > > > - * Members not needing protection. > > > > > > > > > - */ > > > > > > > > > - struct kref kref; > > > > > > > > > - > > > > > > > > > /* > > > > > > > > > * Members protected by the > > > > > > > > > bo::resv::reserved > > > > > > > > > lock. > > > > > > > > > */ > > > > > > > > > struct ttm_resource *resource; > > > > > > > > > struct ttm_tt *ttm; > > > > > > > > > - bool deleted; > > > > > > > > > struct ttm_lru_bulk_move *bulk_move; > > > > > > > > > unsigned priority; > > > > > > > > > unsigned pin_count;
