From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
To: intel-gfx@lists.freedesktop.org
Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
dri-devel@lists.freedesktop.org
Subject: [Intel-gfx] [PATCH v9 04/70] drm/i915: Add missing -EDEADLK handling to execbuf pinning, v2.
Date: Tue, 23 Mar 2021 16:49:53 +0100 [thread overview]
Message-ID: <20210323155059.628690-5-maarten.lankhorst@linux.intel.com> (raw)
In-Reply-To: <20210323155059.628690-1-maarten.lankhorst@linux.intel.com>
i915_vma_pin may fail with -EDEADLK when we start locking page tables,
so ensure we handle this correctly.
Changes since v1:
- Drop -EDEADLK todo, this commit handles it.
- Change eb_pin_vma from sort-of-bool + -EDEADLK to a proper int. (Matt)
Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
.../gpu/drm/i915/gem/i915_gem_execbuffer.c | 35 +++++++++++++------
1 file changed, 24 insertions(+), 11 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 309d661c4c71..1c6d0d92701f 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -420,13 +420,14 @@ static u64 eb_pin_flags(const struct drm_i915_gem_exec_object2 *entry,
return pin_flags;
}
-static inline bool
+static inline int
eb_pin_vma(struct i915_execbuffer *eb,
const struct drm_i915_gem_exec_object2 *entry,
struct eb_vma *ev)
{
struct i915_vma *vma = ev->vma;
u64 pin_flags;
+ int err;
if (vma->node.size)
pin_flags = vma->node.start;
@@ -438,24 +439,29 @@ eb_pin_vma(struct i915_execbuffer *eb,
pin_flags |= PIN_GLOBAL;
/* Attempt to reuse the current location if available */
- /* TODO: Add -EDEADLK handling here */
- if (unlikely(i915_vma_pin_ww(vma, &eb->ww, 0, 0, pin_flags))) {
+ err = i915_vma_pin_ww(vma, &eb->ww, 0, 0, pin_flags);
+ if (err == -EDEADLK)
+ return err;
+
+ if (unlikely(err)) {
if (entry->flags & EXEC_OBJECT_PINNED)
- return false;
+ return err;
/* Failing that pick any _free_ space if suitable */
- if (unlikely(i915_vma_pin_ww(vma, &eb->ww,
+ err = i915_vma_pin_ww(vma, &eb->ww,
entry->pad_to_size,
entry->alignment,
eb_pin_flags(entry, ev->flags) |
- PIN_USER | PIN_NOEVICT)))
- return false;
+ PIN_USER | PIN_NOEVICT);
+ if (unlikely(err))
+ return err;
}
if (unlikely(ev->flags & EXEC_OBJECT_NEEDS_FENCE)) {
- if (unlikely(i915_vma_pin_fence(vma))) {
+ err = i915_vma_pin_fence(vma);
+ if (unlikely(err)) {
i915_vma_unpin(vma);
- return false;
+ return err;
}
if (vma->fence)
@@ -463,7 +469,10 @@ eb_pin_vma(struct i915_execbuffer *eb,
}
ev->flags |= __EXEC_OBJECT_HAS_PIN;
- return !eb_vma_misplaced(entry, vma, ev->flags);
+ if (eb_vma_misplaced(entry, vma, ev->flags))
+ return -EBADSLT;
+
+ return 0;
}
static inline void
@@ -906,7 +915,11 @@ static int eb_validate_vmas(struct i915_execbuffer *eb)
if (err)
return err;
- if (eb_pin_vma(eb, entry, ev)) {
+ err = eb_pin_vma(eb, entry, ev);
+ if (err == -EDEADLK)
+ return err;
+
+ if (!err) {
if (entry->offset != vma->node.start) {
entry->offset = vma->node.start | UPDATE;
eb->args->flags |= __EXEC_HAS_RELOC;
--
2.31.0
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2021-03-23 15:51 UTC|newest]
Thread overview: 108+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-23 15:49 [Intel-gfx] [PATCH v9 00/70] drm/i915: Remove obj->mm.lock! Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 01/70] drm/i915: Do not share hwsp across contexts any more, v8 Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 02/70] drm/i915: Pin timeline map after first timeline pin, v4 Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 03/70] drm/i915: Move cmd parser pinning to execbuffer Maarten Lankhorst
2021-03-23 15:49 ` Maarten Lankhorst [this message]
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 05/70] drm/i915: Ensure we hold the object mutex in pin correctly Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 06/70] drm/i915: Add gem object locking to madvise Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 07/70] drm/i915: Move HAS_STRUCT_PAGE to obj->flags Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 08/70] drm/i915: Rework struct phys attachment handling Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 09/70] drm/i915: Convert i915_gem_object_attach_phys() to ww locking, v2 Maarten Lankhorst
2021-03-23 15:49 ` [Intel-gfx] [PATCH v9 10/70] drm/i915: make lockdep slightly happier about execbuf Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 11/70] drm/i915: Disable userptr pread/pwrite support Maarten Lankhorst
2021-03-24 13:57 ` Jason Ekstrand
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 12/70] drm/i915: No longer allow exporting userptr through dma-buf Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 13/70] drm/i915: Reject more ioctls for userptr, v2 Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 14/70] drm/i915: Reject UNSYNCHRONIZED " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 15/70] drm/i915: Make compilation of userptr code depend on MMU_NOTIFIER Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 16/70] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v7 Maarten Lankhorst
2021-03-24 11:28 ` Daniel Vetter
2021-03-24 11:34 ` Thomas Hellström (Intel)
2021-03-25 9:23 ` [Intel-gfx] [PATCH] drm/i915: Fix userptr so we do not have to worry about obj->mm.lock, v8 Maarten Lankhorst
2021-03-25 9:55 ` Thomas Hellström (Intel)
2021-03-25 10:27 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 17/70] drm/i915: Flatten obj->mm.lock Maarten Lankhorst
2021-03-24 11:13 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 18/70] drm/i915: Populate logical context during first pin Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 19/70] drm/i915: Make ring submission compatible with obj->mm.lock removal, v2 Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 20/70] drm/i915: Handle ww locking in init_status_page Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 21/70] drm/i915: Rework clflush to work correctly without obj->mm.lock Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 22/70] drm/i915: Pass ww ctx to intel_pin_to_display_plane Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 23/70] drm/i915: Add object locking to vm_fault_cpu Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 24/70] drm/i915: Move pinning to inside engine_wa_list_verify() Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 25/70] drm/i915: Take reservation lock around i915_vma_pin Maarten Lankhorst
2021-03-24 12:35 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 26/70] drm/i915: Make lrc_init_wa_ctx compatible with ww locking, v3 Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 27/70] drm/i915: Make __engine_unpark() compatible with ww locking Maarten Lankhorst
2021-03-24 12:37 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 28/70] drm/i915: Take obj lock around set_domain ioctl Maarten Lankhorst
2021-03-24 14:12 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 29/70] drm/i915: Defer pin calls in buffer pool until first use by caller Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 30/70] drm/i915: Fix pread/pwrite to work with new locking rules Maarten Lankhorst
2021-03-24 14:45 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 31/70] drm/i915: Fix workarounds selftest, part 1 Maarten Lankhorst
2021-03-24 16:16 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 32/70] drm/i915: Prepare for obj->mm.lock removal, v2 Maarten Lankhorst
2021-03-23 16:18 ` Matthew Auld
2021-03-23 20:25 ` Thomas Hellström
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 33/70] drm/i915: Add igt_spinner_pin() to allow for ww locking around spinner Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 34/70] drm/i915: Add ww locking around vm_access() Maarten Lankhorst
2021-03-24 16:21 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 35/70] drm/i915: Increase ww locking for perf Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 36/70] drm/i915: Lock ww in ucode objects correctly Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 37/70] drm/i915: Add ww locking to dma-buf ops, v2 Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 38/70] drm/i915: Add missing ww lock in intel_dsb_prepare Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 39/70] drm/i915: Fix ww locking in shmem_create_from_object Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 40/70] drm/i915: Use a single page table lock for each gtt Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 41/70] drm/i915/selftests: Prepare huge_pages testcases for obj->mm.lock removal Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 42/70] drm/i915/selftests: Prepare client blit " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 43/70] drm/i915/selftests: Prepare coherency tests " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 44/70] drm/i915/selftests: Prepare context " Maarten Lankhorst
2021-03-24 16:40 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 45/70] drm/i915/selftests: Prepare dma-buf " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 46/70] drm/i915/selftests: Prepare execbuf " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 47/70] drm/i915/selftests: Prepare mman testcases " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 48/70] drm/i915/selftests: Prepare object tests " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 49/70] drm/i915/selftests: Prepare object blit " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 50/70] drm/i915/selftests: Prepare igt_gem_utils " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 51/70] drm/i915/selftests: Prepare context selftest " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 52/70] drm/i915/selftests: Prepare hangcheck " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 53/70] drm/i915/selftests: Prepare execlists and lrc selftests " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 54/70] drm/i915/selftests: Prepare mocs tests " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 55/70] drm/i915/selftests: Prepare ring submission " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 56/70] drm/i915/selftests: Prepare timeline tests " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 57/70] drm/i915/selftests: Prepare i915_request " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 58/70] drm/i915/selftests: Prepare memory region " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 59/70] drm/i915/selftests: Prepare cs engine " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 60/70] drm/i915/selftests: Prepare gtt " Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 61/70] drm/i915: Finally remove obj->mm.lock Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 62/70] drm/i915: Keep userpointer bindings if seqcount is unchanged, v2 Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 63/70] drm/i915: Move gt_revoke() slightly Maarten Lankhorst
2021-03-24 17:00 ` Daniel Vetter
2021-03-24 17:15 ` Ville Syrjälä
2021-03-24 17:16 ` Daniel Vetter
2021-03-24 17:58 ` Ville Syrjälä
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 64/70] drm/i915: Add missing -EDEADLK path in execbuffer ggtt pinning Maarten Lankhorst
2021-03-24 17:05 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 65/70] drm/i915: Fix pin_map in scheduler selftests Maarten Lankhorst
2021-03-24 17:14 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 66/70] drm/i915: Add ww parameter to get_pages() callback Maarten Lankhorst
2021-03-24 17:20 ` Daniel Vetter
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 67/70] drm/i915: Add ww context to prepare_(read/write) Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 68/70] drm/i915: Pass ww ctx to pin_map Maarten Lankhorst
2021-03-23 17:30 ` Matthew Auld
2021-03-24 9:31 ` Maarten Lankhorst
2021-03-24 10:11 ` Daniel Vetter
2021-03-24 11:54 ` [Intel-gfx] [PATCH] drm/i915: Pass ww ctx to pin_map, v2 Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 69/70] drm/i915: Pass ww ctx to i915_gem_object_pin_pages Maarten Lankhorst
2021-03-23 15:50 ` [Intel-gfx] [PATCH v9 70/70] drm/i915: Remove asynchronous vma binding Maarten Lankhorst
2021-03-24 17:19 ` Daniel Vetter
2021-03-23 16:07 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Remove obj->mm.lock! (rev18) Patchwork
2021-03-23 16:09 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-03-23 16:12 ` [Intel-gfx] ✗ Fi.CI.DOCS: " Patchwork
2021-03-23 16:35 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
2021-03-24 12:44 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Remove obj->mm.lock! (rev19) Patchwork
2021-03-24 12:46 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-03-24 12:49 ` [Intel-gfx] ✗ Fi.CI.DOCS: " Patchwork
2021-03-24 13:13 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
2021-03-25 21:28 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for drm/i915: Remove obj->mm.lock! (rev20) 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=20210323155059.628690-5-maarten.lankhorst@linux.intel.com \
--to=maarten.lankhorst@linux.intel.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: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox