From: Matthew Auld <matthew.william.auld@gmail.com> To: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Cc: "Intel Graphics Development" <intel-gfx@lists.freedesktop.org>, "Christian König" <christian.koenig@amd.com>, "ML dri-devel" <dri-devel@lists.freedesktop.org> Subject: Re: [Intel-gfx] [PATCH v3 09/12] drm/ttm: Document and optimize ttm_bo_pipeline_gutting() Date: Tue, 25 May 2021 12:00:08 +0100 [thread overview] Message-ID: <CAM0jSHO59Kr534HC-i816yRMJ22-XNN-AHwdoSFtK5KBiD99=g@mail.gmail.com> (raw) In-Reply-To: <20210521153253.518037-10-thomas.hellstrom@linux.intel.com> On Fri, 21 May 2021 at 16:33, Thomas Hellström <thomas.hellstrom@linux.intel.com> wrote: > > If the bo is idle when calling ttm_bo_pipeline_gutting(), we unnecessarily > create a ghost object and push it out to delayed destroy. > Fix this by adding a path for idle, and document the function. > > Also avoid having the bo end up in a bad state vulnerable to user-space > triggered kernel BUGs if the call to ttm_tt_create() fails. > > Finally reuse ttm_bo_pipeline_gutting() in ttm_bo_evict(). > > Cc: Christian König <christian.koenig@amd.com> > Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> > --- > drivers/gpu/drm/ttm/ttm_bo.c | 20 ++++++------ > drivers/gpu/drm/ttm/ttm_bo_util.c | 52 ++++++++++++++++++++++++++++--- > drivers/gpu/drm/ttm/ttm_tt.c | 5 +++ > include/drm/ttm/ttm_tt.h | 10 ++++++ > 4 files changed, 73 insertions(+), 14 deletions(-) > > diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c > index ca1b098b6a56..a8fa3375b8aa 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo.c > +++ b/drivers/gpu/drm/ttm/ttm_bo.c > @@ -501,10 +501,15 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo, > bdev->funcs->evict_flags(bo, &placement); > > if (!placement.num_placement && !placement.num_busy_placement) { > - ttm_bo_wait(bo, false, false); > + ret = ttm_bo_wait(bo, true, false); > + if (ret) > + return ret; > > - ttm_bo_cleanup_memtype_use(bo); > - return ttm_tt_create(bo, false); > + /* > + * Since we've already synced, this frees backing store > + * immediately. > + */ > + return ttm_bo_pipeline_gutting(bo); > } > > ret = ttm_bo_mem_space(bo, &placement, &evict_mem, ctx); > @@ -974,13 +979,8 @@ int ttm_bo_validate(struct ttm_buffer_object *bo, > /* > * Remove the backing store if no placement is given. > */ > - if (!placement->num_placement && !placement->num_busy_placement) { > - ret = ttm_bo_pipeline_gutting(bo); > - if (ret) > - return ret; > - > - return ttm_tt_create(bo, false); > - } > + if (!placement->num_placement && !placement->num_busy_placement) > + return ttm_bo_pipeline_gutting(bo); > > /* > * Check whether we need to move buffer. > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > index 4a7d3d672f9a..7fa9b3a852eb 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > @@ -585,26 +585,70 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo, > } > EXPORT_SYMBOL(ttm_bo_move_accel_cleanup); > > +/** > + * ttm_bo_pipeline_gutting - purge the contents of a bo > + * @bo: The buffer object > + * > + * Purge the contents of a bo, async if the bo is not idle. > + * After a successful call, the bo is left unpopulated in > + * system placement. The function may wait uninterruptible > + * for idle on OOM. > + * > + * Return: 0 if successful, negative error code on failure. > + */ > int ttm_bo_pipeline_gutting(struct ttm_buffer_object *bo) > { > static const struct ttm_place sys_mem = { .mem_type = TTM_PL_SYSTEM }; > struct ttm_buffer_object *ghost; > + struct ttm_tt *ttm; > int ret; > > - ret = ttm_buffer_object_transfer(bo, &ghost); > + /* If already idle, no need for ghost object dance. */ > + ret = ttm_bo_wait(bo, false, true); > + if (ret != -EBUSY) { > + if (!bo->ttm) { > + ret = ttm_tt_create(bo, true); Why do we now unconditionally add clearing? Below also. > + if (ret) > + return ret; > + } else { > + ttm_tt_unpopulate(bo->bdev, bo->ttm); > + if (bo->type == ttm_bo_type_device) > + ttm_tt_mark_for_clear(bo->ttm); > + } > + ttm_resource_free(bo, &bo->mem); > + ttm_resource_alloc(bo, &sys_mem, &bo->mem); > + > + return 0; > + } > + > + /* > + * We need an unpopulated ttm_tt after giving our current one, > + * if any, to the ghost object. And we can't afford to fail > + * creating one *after* the operation. > + */ > + > + ttm = bo->ttm; > + bo->ttm = NULL; > + ret = ttm_tt_create(bo, true); > + swap(bo->ttm, ttm); > if (ret) > return ret; > > + ret = ttm_buffer_object_transfer(bo, &ghost); > + if (ret) { > + ttm_tt_destroy(bo->bdev, ttm); > + return ret; > + } > + > ret = dma_resv_copy_fences(&ghost->base._resv, bo->base.resv); > /* Last resort, wait for the BO to be idle when we are OOM */ > if (ret) > ttm_bo_wait(bo, false, false); > > - ttm_resource_alloc(bo, &sys_mem, &bo->mem); > - bo->ttm = NULL; > - > dma_resv_unlock(&ghost->base._resv); > ttm_bo_put(ghost); > + bo->ttm = ttm; > + ttm_resource_alloc(bo, &sys_mem, &bo->mem); > > return 0; > } > diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c > index 0e41227116b1..913b330a234b 100644 > --- a/drivers/gpu/drm/ttm/ttm_tt.c > +++ b/drivers/gpu/drm/ttm/ttm_tt.c > @@ -134,6 +134,11 @@ void ttm_tt_destroy_common(struct ttm_device *bdev, struct ttm_tt *ttm) > } > EXPORT_SYMBOL(ttm_tt_destroy_common); > > +void ttm_tt_mark_for_clear(struct ttm_tt *ttm) > +{ > + ttm->page_flags |= TTM_PAGE_FLAG_ZERO_ALLOC; > +} > + > void ttm_tt_destroy(struct ttm_device *bdev, struct ttm_tt *ttm) > { > bdev->funcs->ttm_tt_destroy(bdev, ttm); > diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h > index 3102059db726..daa9c4cf48bb 100644 > --- a/include/drm/ttm/ttm_tt.h > +++ b/include/drm/ttm/ttm_tt.h > @@ -170,6 +170,16 @@ int ttm_tt_populate(struct ttm_device *bdev, struct ttm_tt *ttm, struct ttm_oper > */ > void ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm); > > +/** > + * ttm_tt_mark_for_clear - Mark pages for clearing on populate. > + * > + * @ttm: Pointer to the ttm_tt structure > + * > + * Marks pages for clearing so that the next time the page vector is > + * populated, the pages will be cleared. > + */ > +void ttm_tt_mark_for_clear(struct ttm_tt *ttm); > + > void ttm_tt_mgr_init(unsigned long num_pages, unsigned long num_dma32_pages); > > struct ttm_kmap_iter *ttm_kmap_iter_tt_init(struct ttm_kmap_iter_tt *iter_tt, > -- > 2.31.1 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
WARNING: multiple messages have this Message-ID (diff)
From: Matthew Auld <matthew.william.auld@gmail.com> To: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Cc: "Intel Graphics Development" <intel-gfx@lists.freedesktop.org>, "Christian König" <christian.koenig@amd.com>, "ML dri-devel" <dri-devel@lists.freedesktop.org> Subject: Re: [Intel-gfx] [PATCH v3 09/12] drm/ttm: Document and optimize ttm_bo_pipeline_gutting() Date: Tue, 25 May 2021 12:00:08 +0100 [thread overview] Message-ID: <CAM0jSHO59Kr534HC-i816yRMJ22-XNN-AHwdoSFtK5KBiD99=g@mail.gmail.com> (raw) In-Reply-To: <20210521153253.518037-10-thomas.hellstrom@linux.intel.com> On Fri, 21 May 2021 at 16:33, Thomas Hellström <thomas.hellstrom@linux.intel.com> wrote: > > If the bo is idle when calling ttm_bo_pipeline_gutting(), we unnecessarily > create a ghost object and push it out to delayed destroy. > Fix this by adding a path for idle, and document the function. > > Also avoid having the bo end up in a bad state vulnerable to user-space > triggered kernel BUGs if the call to ttm_tt_create() fails. > > Finally reuse ttm_bo_pipeline_gutting() in ttm_bo_evict(). > > Cc: Christian König <christian.koenig@amd.com> > Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> > --- > drivers/gpu/drm/ttm/ttm_bo.c | 20 ++++++------ > drivers/gpu/drm/ttm/ttm_bo_util.c | 52 ++++++++++++++++++++++++++++--- > drivers/gpu/drm/ttm/ttm_tt.c | 5 +++ > include/drm/ttm/ttm_tt.h | 10 ++++++ > 4 files changed, 73 insertions(+), 14 deletions(-) > > diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c > index ca1b098b6a56..a8fa3375b8aa 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo.c > +++ b/drivers/gpu/drm/ttm/ttm_bo.c > @@ -501,10 +501,15 @@ static int ttm_bo_evict(struct ttm_buffer_object *bo, > bdev->funcs->evict_flags(bo, &placement); > > if (!placement.num_placement && !placement.num_busy_placement) { > - ttm_bo_wait(bo, false, false); > + ret = ttm_bo_wait(bo, true, false); > + if (ret) > + return ret; > > - ttm_bo_cleanup_memtype_use(bo); > - return ttm_tt_create(bo, false); > + /* > + * Since we've already synced, this frees backing store > + * immediately. > + */ > + return ttm_bo_pipeline_gutting(bo); > } > > ret = ttm_bo_mem_space(bo, &placement, &evict_mem, ctx); > @@ -974,13 +979,8 @@ int ttm_bo_validate(struct ttm_buffer_object *bo, > /* > * Remove the backing store if no placement is given. > */ > - if (!placement->num_placement && !placement->num_busy_placement) { > - ret = ttm_bo_pipeline_gutting(bo); > - if (ret) > - return ret; > - > - return ttm_tt_create(bo, false); > - } > + if (!placement->num_placement && !placement->num_busy_placement) > + return ttm_bo_pipeline_gutting(bo); > > /* > * Check whether we need to move buffer. > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > index 4a7d3d672f9a..7fa9b3a852eb 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > @@ -585,26 +585,70 @@ int ttm_bo_move_accel_cleanup(struct ttm_buffer_object *bo, > } > EXPORT_SYMBOL(ttm_bo_move_accel_cleanup); > > +/** > + * ttm_bo_pipeline_gutting - purge the contents of a bo > + * @bo: The buffer object > + * > + * Purge the contents of a bo, async if the bo is not idle. > + * After a successful call, the bo is left unpopulated in > + * system placement. The function may wait uninterruptible > + * for idle on OOM. > + * > + * Return: 0 if successful, negative error code on failure. > + */ > int ttm_bo_pipeline_gutting(struct ttm_buffer_object *bo) > { > static const struct ttm_place sys_mem = { .mem_type = TTM_PL_SYSTEM }; > struct ttm_buffer_object *ghost; > + struct ttm_tt *ttm; > int ret; > > - ret = ttm_buffer_object_transfer(bo, &ghost); > + /* If already idle, no need for ghost object dance. */ > + ret = ttm_bo_wait(bo, false, true); > + if (ret != -EBUSY) { > + if (!bo->ttm) { > + ret = ttm_tt_create(bo, true); Why do we now unconditionally add clearing? Below also. > + if (ret) > + return ret; > + } else { > + ttm_tt_unpopulate(bo->bdev, bo->ttm); > + if (bo->type == ttm_bo_type_device) > + ttm_tt_mark_for_clear(bo->ttm); > + } > + ttm_resource_free(bo, &bo->mem); > + ttm_resource_alloc(bo, &sys_mem, &bo->mem); > + > + return 0; > + } > + > + /* > + * We need an unpopulated ttm_tt after giving our current one, > + * if any, to the ghost object. And we can't afford to fail > + * creating one *after* the operation. > + */ > + > + ttm = bo->ttm; > + bo->ttm = NULL; > + ret = ttm_tt_create(bo, true); > + swap(bo->ttm, ttm); > if (ret) > return ret; > > + ret = ttm_buffer_object_transfer(bo, &ghost); > + if (ret) { > + ttm_tt_destroy(bo->bdev, ttm); > + return ret; > + } > + > ret = dma_resv_copy_fences(&ghost->base._resv, bo->base.resv); > /* Last resort, wait for the BO to be idle when we are OOM */ > if (ret) > ttm_bo_wait(bo, false, false); > > - ttm_resource_alloc(bo, &sys_mem, &bo->mem); > - bo->ttm = NULL; > - > dma_resv_unlock(&ghost->base._resv); > ttm_bo_put(ghost); > + bo->ttm = ttm; > + ttm_resource_alloc(bo, &sys_mem, &bo->mem); > > return 0; > } > diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c > index 0e41227116b1..913b330a234b 100644 > --- a/drivers/gpu/drm/ttm/ttm_tt.c > +++ b/drivers/gpu/drm/ttm/ttm_tt.c > @@ -134,6 +134,11 @@ void ttm_tt_destroy_common(struct ttm_device *bdev, struct ttm_tt *ttm) > } > EXPORT_SYMBOL(ttm_tt_destroy_common); > > +void ttm_tt_mark_for_clear(struct ttm_tt *ttm) > +{ > + ttm->page_flags |= TTM_PAGE_FLAG_ZERO_ALLOC; > +} > + > void ttm_tt_destroy(struct ttm_device *bdev, struct ttm_tt *ttm) > { > bdev->funcs->ttm_tt_destroy(bdev, ttm); > diff --git a/include/drm/ttm/ttm_tt.h b/include/drm/ttm/ttm_tt.h > index 3102059db726..daa9c4cf48bb 100644 > --- a/include/drm/ttm/ttm_tt.h > +++ b/include/drm/ttm/ttm_tt.h > @@ -170,6 +170,16 @@ int ttm_tt_populate(struct ttm_device *bdev, struct ttm_tt *ttm, struct ttm_oper > */ > void ttm_tt_unpopulate(struct ttm_device *bdev, struct ttm_tt *ttm); > > +/** > + * ttm_tt_mark_for_clear - Mark pages for clearing on populate. > + * > + * @ttm: Pointer to the ttm_tt structure > + * > + * Marks pages for clearing so that the next time the page vector is > + * populated, the pages will be cleared. > + */ > +void ttm_tt_mark_for_clear(struct ttm_tt *ttm); > + > void ttm_tt_mgr_init(unsigned long num_pages, unsigned long num_dma32_pages); > > struct ttm_kmap_iter *ttm_kmap_iter_tt_init(struct ttm_kmap_iter_tt *iter_tt, > -- > 2.31.1 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2021-05-25 11:00 UTC|newest] Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-05-21 15:32 [PATCH v3 00/12] drm/i915: Move LMEM (VRAM) management over to TTM Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 01/12] drm/i915: Untangle the vma pages_mutex Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 02/12] drm/i915: Don't free shared locks while shared Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 03/12] drm/i915: Fix i915_sg_page_sizes to record dma segments rather than physical pages Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 04/12] drm/i915/ttm Initialize the ttm device and memory managers Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 05/12] drm/i915/ttm: Embed a ttm buffer object in the i915 gem object Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 06/12] drm/ttm: Add a generic TTM memcpy move for page-based iomem Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-25 9:18 ` Matthew Auld 2021-05-25 9:18 ` Matthew Auld 2021-05-25 9:32 ` Thomas Hellström 2021-05-25 9:32 ` Thomas Hellström 2021-05-25 9:58 ` Matthew Auld 2021-05-25 9:58 ` Matthew Auld 2021-05-25 10:07 ` Thomas Hellström 2021-05-25 10:07 ` Thomas Hellström 2021-05-25 15:48 ` Christian König 2021-05-25 15:48 ` Christian König 2021-05-26 7:39 ` Thomas Hellström 2021-05-26 7:39 ` Thomas Hellström 2021-05-26 10:45 ` Christian König 2021-05-26 10:45 ` Christian König 2021-05-26 10:57 ` Thomas Hellström 2021-05-26 10:57 ` Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 07/12] drm, drm/i915: Move the memcpy_from_wc functionality to core drm Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-24 16:45 ` Matthew Auld 2021-05-24 16:45 ` Matthew Auld 2021-05-24 18:12 ` Thomas Hellström 2021-05-24 18:12 ` Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 08/12] drm/ttm: Use drm_memcpy_from_wc_dbm for TTM bo moves Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-24 18:16 ` Matthew Auld 2021-05-24 18:16 ` [Intel-gfx] " Matthew Auld 2021-05-24 18:47 ` Thomas Hellström 2021-05-24 18:47 ` [Intel-gfx] " Thomas Hellström 2021-05-26 12:48 ` Christian König 2021-05-26 12:48 ` [Intel-gfx] " Christian König 2021-05-21 15:32 ` [PATCH v3 09/12] drm/ttm: Document and optimize ttm_bo_pipeline_gutting() Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-25 11:00 ` Matthew Auld [this message] 2021-05-25 11:00 ` Matthew Auld 2021-05-25 13:37 ` Thomas Hellström 2021-05-25 13:37 ` Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 10/12] drm/ttm, drm/amdgpu: Allow the driver some control over swapping Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 11/12] drm/i915/ttm: Introduce a TTM i915 gem object backend Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 15:32 ` [PATCH v3 12/12] drm/i915/lmem: Verify checks for lmem residency Thomas Hellström 2021-05-21 15:32 ` [Intel-gfx] " Thomas Hellström 2021-05-21 16:06 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Move LMEM (VRAM) management over to TTM (rev3) Patchwork 2021-05-21 16:09 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork 2021-05-21 16:36 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork 2021-05-24 0:10 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork
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='CAM0jSHO59Kr534HC-i816yRMJ22-XNN-AHwdoSFtK5KBiD99=g@mail.gmail.com' \ --to=matthew.william.auld@gmail.com \ --cc=christian.koenig@amd.com \ --cc=dri-devel@lists.freedesktop.org \ --cc=intel-gfx@lists.freedesktop.org \ --cc=thomas.hellstrom@linux.intel.com \ /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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.