linux/drivers/gpu/drm/msm/msm_ringbuffer.c, branch v6.4

drm/msm/gem: Avoid obj lock in job_run()

2023-03-25T23:31:44Z

Now that everything that controls which LRU an obj lives in *except* the backing pages is protected by the LRU lock, add a special path to unpin in the job_run() path, where we are assured that we already have backing pages and will not be racing against eviction (because the GEM object's dma_resv contains the fence that will be signaled when the submit/job completes). Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/527845/ Link: https://lore.kernel.org/r/20230320144356.803762-10-robdclark@gmail.com

drm/msm: Decouple vma tracking from obj lock

2023-03-25T23:31:44Z

We need to use the inuse count to track that a BO is pinned until we have the hw_fence. But we want to remove the obj lock from the job_run() path as this could deadlock against reclaim/shrinker (because it is blocking the hw_fence from eventually being signaled). So split that tracking out into a per-vma lock with narrower scope. Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/527839/ Link: https://lore.kernel.org/r/20230320144356.803762-5-robdclark@gmail.com

drm/msm/gem: Tidy up VMA API

2023-03-25T23:31:44Z

Stop open coding VMA construction, which will be needed in the next commit. And since the VMA already has a ptr to the adress space, stop passing that around everywhere. (Also, an aspace always has an mmu so we can drop a couple pointless NULL checks.) Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/527833/ Link: https://lore.kernel.org/r/20230320144356.803762-4-robdclark@gmail.com

drm/msm: Move submit bo flags update from obj lock

2023-03-25T23:31:43Z

The flags are only accessed (1) when submit is constructed, before enqueuing to gpu sched (ie. when still visible to only the task calling the submit ioctl), (2) here, where we own a reference to the submit and are serialized on the gpu sched thread, and (3) after the submit is retired and last reference is dropped, which is serialized on the submit's reference count. Hence locking is unneeded here. Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/527830/ Link: https://lore.kernel.org/r/20230320144356.803762-3-robdclark@gmail.com

drm/msm: Pre-allocate hw_fence

2023-03-25T23:31:43Z

Avoid allocating memory in job_run() by pre-allocating the hw_fence. Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/527832/ Link: https://lore.kernel.org/r/20230320144356.803762-2-robdclark@gmail.com

drm/msm/gem: Unpin objects slightly later

2022-09-30T16:01:33Z

The introduction of "drm/msm/gem: Evict active GEM objects when necessary" exposes a problem with "drm/msm/gem: Unpin buffers earlier", in that we need to keep the object pinned in the time the submit is queued up in the gpu scheduler. Otherwise the shrinker will see it as a thing that can be evicted if we wait for it to be signaled. But if the shrinker path is waiting on it with the obj lock held, the job cannot be scheduled, as that also requires briefly grabbing the obj lock, leading to deadlock. (Not to mention, we don't want the shrinker to evict an obj queued up in gpu scheduler.) Fixes: f371bcc0c2ac ("drm/msm/gem: Unpin buffers earlier") Fixes: 025d27239a2f ("drm/msm/gem: Evict active GEM objects when necessary") Closes: https://gitlab.freedesktop.org/drm/msm/-/issues/19 Signed-off-by: Rob Clark Tested-by: Chia-I Wu Patchwork: https://patchwork.freedesktop.org/patch/504528/ Link: https://lore.kernel.org/r/20220923224043.2449152-1-robdclark@gmail.com

drm/msm: Remove unnecessary pm_runtime_get/put

2022-08-28T16:29:27Z

We already enable gpu power from msm_gpu_submit(), so avoid a duplicate pm_runtime_get/put from msm_job_run(). Signed-off-by: Akhil P Oommen Patchwork: https://patchwork.freedesktop.org/patch/498390/ Link: https://lore.kernel.org/r/20220819015030.v5.1.Icf1e8f0c9b3e7e9933c3b48c70477d0582f3243f@changeid Signed-off-by: Rob Clark

drm/msm/gem: Separate object and vma unpin

2022-06-15T20:06:54Z

Previously the BO_PINNED state in the submit was tracking two related but different things: (1) that the buffer object was pinned, and (2) that the vma (mapping within a set of pagetables) was pinned. But with fenced vma unpin (needed so that userspace couldn't race with retire path for releasing a vma) these two were decoupled. The fact that the BO_PINNED flag was already cleared meant that we leaked the bo pin count which should have been dropped when the submit was retired. So split this state into BO_OBJ_PINNED and BO_VMA_PINNED, so they can be dropped independently. Fixes: 95d1deb02a9c ("drm/msm/gem: Add fenced vma unpin") Signed-off-by: Rob Clark Patchwork: https://patchwork.freedesktop.org/patch/487559/ Link: https://lore.kernel.org/r/20220527172341.2151005-1-robdclark@gmail.com

drm/msm: change msm_sched_ops from global to static

2022-04-27T07:50:22Z

Smatch reports this issue msm_ringbuffer.c:43:36: warning: symbol 'msm_sched_ops' was not declared. Should it be static? msm_sched_ops is only used in msm_ringbuffer.c so change its storage-class specifier to static. Signed-off-by: Tom Rix Reviewed-by: Dmitry Baryshkov Patchwork: https://patchwork.freedesktop.org/patch/482883/ Link: https://lore.kernel.org/r/20220421131507.1557667-1-trix@redhat.com Signed-off-by: Dmitry Baryshkov

drm/msm/gem: Add fenced vma unpin

2022-04-21T22:03:12Z

With userspace allocated iova (next patch), we can have a race condition where userspace observes the fence completion and deletes the vma before retire_submit() gets around to unpinning the vma. To handle this, add a fenced unpin which drops the refcount but tracks the fence, and update msm_gem_vma_inuse() to check any previously unsignaled fences. v2: Fix inuse underflow (duplicate unpin) v3: Fix msm_job_run() vs submit_cleanup() race condition Signed-off-by: Rob Clark Link: https://lore.kernel.org/r/20220411215849.297838-10-robdclark@gmail.com Signed-off-by: Rob Clark