From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> To: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Subject: [PATCH v5 1/6] drm/i915: Call i915_gem_evict_vm in vm_fault_gtt to prevent new ENOSPC errors, v2. Date: Thu, 13 Jan 2022 12:44:55 +0100 [thread overview] Message-ID: <20220113114500.2039439-2-maarten.lankhorst@linux.intel.com> (raw) In-Reply-To: <20220113114500.2039439-1-maarten.lankhorst@linux.intel.com> Now that we cannot unbind kill the currently locked object directly because we're removing short term pinning, we may have to unbind the object from gtt manually, using a i915_gem_evict_vm() call. Changes since v1: - Remove -ENOSPC warning, can still happen with concurrent mmaps where we can't unbind the other mmap because of the lock held. This fixes the gem_mmap_gtt@cpuset tests. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 5ac2506f4ee8..fafd158e5313 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -21,6 +21,7 @@ #include "i915_trace.h" #include "i915_user_extensions.h" #include "i915_gem_ttm.h" +#include "i915_gem_evict.h" #include "i915_vma.h" static inline bool @@ -358,8 +359,21 @@ static vm_fault_t vm_fault_gtt(struct vm_fault *vmf) vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags); } - /* The entire mappable GGTT is pinned? Unexpected! */ - GEM_BUG_ON(vma == ERR_PTR(-ENOSPC)); + /* + * The entire mappable GGTT is pinned? Unexpected! + * Try to evict the object we locked too, as normally we skip it + * due to lack of short term pinning inside execbuf. + */ + if (vma == ERR_PTR(-ENOSPC)) { + ret = mutex_lock_interruptible(&ggtt->vm.mutex); + if (!ret) { + ret = i915_gem_evict_vm(&ggtt->vm); + mutex_unlock(&ggtt->vm.mutex); + } + if (ret) + goto err_reset; + vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags); + } } if (IS_ERR(vma)) { ret = PTR_ERR(vma); -- 2.34.1
WARNING: multiple messages have this Message-ID (diff)
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> To: intel-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Subject: [Intel-gfx] [PATCH v5 1/6] drm/i915: Call i915_gem_evict_vm in vm_fault_gtt to prevent new ENOSPC errors, v2. Date: Thu, 13 Jan 2022 12:44:55 +0100 [thread overview] Message-ID: <20220113114500.2039439-2-maarten.lankhorst@linux.intel.com> (raw) In-Reply-To: <20220113114500.2039439-1-maarten.lankhorst@linux.intel.com> Now that we cannot unbind kill the currently locked object directly because we're removing short term pinning, we may have to unbind the object from gtt manually, using a i915_gem_evict_vm() call. Changes since v1: - Remove -ENOSPC warning, can still happen with concurrent mmaps where we can't unbind the other mmap because of the lock held. This fixes the gem_mmap_gtt@cpuset tests. Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> --- drivers/gpu/drm/i915/gem/i915_gem_mman.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 5ac2506f4ee8..fafd158e5313 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -21,6 +21,7 @@ #include "i915_trace.h" #include "i915_user_extensions.h" #include "i915_gem_ttm.h" +#include "i915_gem_evict.h" #include "i915_vma.h" static inline bool @@ -358,8 +359,21 @@ static vm_fault_t vm_fault_gtt(struct vm_fault *vmf) vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags); } - /* The entire mappable GGTT is pinned? Unexpected! */ - GEM_BUG_ON(vma == ERR_PTR(-ENOSPC)); + /* + * The entire mappable GGTT is pinned? Unexpected! + * Try to evict the object we locked too, as normally we skip it + * due to lack of short term pinning inside execbuf. + */ + if (vma == ERR_PTR(-ENOSPC)) { + ret = mutex_lock_interruptible(&ggtt->vm.mutex); + if (!ret) { + ret = i915_gem_evict_vm(&ggtt->vm); + mutex_unlock(&ggtt->vm.mutex); + } + if (ret) + goto err_reset; + vma = i915_gem_object_ggtt_pin_ww(obj, &ww, &view, 0, 0, flags); + } } if (IS_ERR(vma)) { ret = PTR_ERR(vma); -- 2.34.1
next prev parent reply other threads:[~2022-01-13 11:45 UTC|newest] Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-01-13 11:44 [PATCH v5 0/6] drm/i915: Remove short term pins from execbuf by requiring lock to unbind Maarten Lankhorst 2022-01-13 11:44 ` [Intel-gfx] " Maarten Lankhorst 2022-01-13 11:44 ` Maarten Lankhorst [this message] 2022-01-13 11:44 ` [Intel-gfx] [PATCH v5 1/6] drm/i915: Call i915_gem_evict_vm in vm_fault_gtt to prevent new ENOSPC errors, v2 Maarten Lankhorst 2022-01-13 13:30 ` Thomas Hellström 2022-01-13 11:44 ` [PATCH v5 2/6] drm/i915: Add locking to i915_gem_evict_vm() Maarten Lankhorst 2022-01-13 11:44 ` [Intel-gfx] " Maarten Lankhorst 2022-01-13 14:33 ` Thomas Hellström (Intel) 2022-01-13 14:37 ` Maarten Lankhorst 2022-01-13 11:44 ` [PATCH v5 3/6] drm/i915: Add object locking to i915_gem_evict_for_node and i915_gem_evict_something, v2 Maarten Lankhorst 2022-01-13 11:44 ` [Intel-gfx] " Maarten Lankhorst 2022-01-13 14:43 ` Thomas Hellström 2022-01-13 11:44 ` [PATCH v5 4/6] drm/i915: Add i915_vma_unbind_unlocked, and take obj lock for i915_vma_unbind, v2 Maarten Lankhorst 2022-01-13 11:44 ` [Intel-gfx] " Maarten Lankhorst 2022-01-13 14:53 ` Thomas Hellström 2022-01-13 11:44 ` [PATCH v5 5/6] drm/i915: Remove support for unlocked i915_vma unbind Maarten Lankhorst 2022-01-13 11:44 ` [Intel-gfx] " Maarten Lankhorst 2022-01-13 15:01 ` Thomas Hellström 2022-01-13 11:45 ` [PATCH v5 6/6] drm/i915: Remove short-term pins from execbuf, v6 Maarten Lankhorst 2022-01-13 11:45 ` [Intel-gfx] " Maarten Lankhorst 2022-01-13 12:48 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Remove short term pins from execbuf by requiring lock to unbind Patchwork 2022-01-13 12:50 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork 2022-01-13 12:53 ` [Intel-gfx] ✗ Fi.CI.DOCS: " Patchwork 2022-01-13 13:27 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork 2022-01-13 14:35 ` [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=20220113114500.2039439-2-maarten.lankhorst@linux.intel.com \ --to=maarten.lankhorst@linux.intel.com \ --cc=dri-devel@lists.freedesktop.org \ --cc=intel-gfx@lists.freedesktop.org \ /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.