kernel.git/drivers/gpu/drm/msm/msm_gem.h, branch linux-rolling-stable

drm/msm: Fix 32b size truncation

2025-11-13T20:37:16Z

[ Upstream commit 6d6a29a19b232e29b61a14d58b71fefc2e003fa4 ] Somehow we never noticed this when arm64 became a thing, many years ago. v2: also fix npages Signed-off-by: Rob Clark Tested-by: Connor Abbott Patchwork: https://patchwork.freedesktop.org/patch/669785/ Signed-off-by: Sasha Levin

drm/msm: Fix a few comments

2025-08-07T16:22:28Z

Fix a couple comments which had become (partially) obsolete or incorrect with the gpuvm conversion. Signed-off-by: Rob Clark Reviewed-by: Dmitry Baryshkov Patchwork: https://patchwork.freedesktop.org/patch/667237/

drm/msm: Add VM_BIND throttling

2025-07-05T00:48:39Z

A large number of (unsorted or separate) small (<2MB) mappings can cause a lot of, probably unnecessary, prealloc pages. Ie. a single 4k page size mapping will pre-allocate 3 pages (for levels 2-4) for the pagetable. Which can chew up a large amount of unneeded memory. So add a mechanism to put an upper bound on the # of pre-alloc pages. Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661529/

drm/msm: Defer VMA unmap for fb unpins

2025-07-05T00:48:38Z

With the conversion to drm_gpuvm, we lost the lazy VMA cleanup, which means that fb cleanup/unpin when pageflipping to new scanout buffers immediately unmaps the scanout buffer. This is costly (with tlbinv, it can be 4-6ms for a 1080p scanout buffer, and more for higher resolutions)! To avoid this, introduce a vma_ref, which is incremented whenever userspace has a GEM handle or dma-buf fd. When unpinning if the vm is the kms->vm we defer tearing down the VMA until the vma_ref drops to zero. If the buffer is still part of a flip-chain then userspace will be holding some sort of reference to the BO, either via a GEM handle and/or dma-buf fd. So this avoids unmapping the VMA when there is a strong possibility that it will be needed again. Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661538/

drm/msm: use trylock for debugfs

2025-07-05T00:48:38Z

This resolves a potential deadlock vs msm_gem_vm_close(). Otherwise for _NO_SHARE buffers msm_gem_describe() could be trying to acquire the shared vm resv, while already holding priv->obj_lock. But _vm_close() might drop the last reference to a GEM obj while already holding the vm resv, and msm_gem_free_object() needs to grab priv->obj_lock, a locking inversion. OTOH this is only for debugfs and it isn't critical if we undercount by skipping a locked obj. So just use trylock() and move along if we can't get the lock. Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661525/

drm/msm: Add VMA unmap reason

2025-07-05T00:48:38Z

Make the VM log a bit more useful by providing a reason for the unmap (ie. closing VM vs evict/purge, etc) Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661527/

drm/msm: Add VM logging for VM_BIND updates

2025-07-05T00:48:38Z

When userspace opts in to VM_BIND, the submit no longer holds references keeping the VMA alive. This makes it difficult to distinguish between UMD/KMD/app bugs. So add a debug option for logging the most recent VM updates and capturing these in GPU devcoredumps. The submitqueue id is also captured, a value of zero means the operation did not go via a submitqueue (ie. comes from msm_gem_vm_close() tearing down the remaining mappings when the device file is closed. Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661518/

drm/msm: Add VM_BIND ioctl

2025-07-05T00:48:38Z

Add a VM_BIND ioctl for binding/unbinding buffers into a VM. This is only supported if userspace has opted in to MSM_PARAM_EN_VM_BIND. Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661524/

drm/msm: Support pgtable preallocation

2025-07-05T00:48:37Z

Introduce a mechanism to count the worst case # of pages required in a VM_BIND op. Note that previously we would have had to somehow account for allocations in unmap, when splitting a block. This behavior was removed in commit 33729a5fc0ca ("iommu/io-pgtable-arm: Remove split on unmap behavior)" Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661515/

drm/msm: Add VM_BIND submitqueue

2025-07-05T00:48:37Z

This submitqueue type isn't tied to a hw ringbuffer, but instead executes on the CPU for performing async VM_BIND ops. Signed-off-by: Rob Clark Signed-off-by: Rob Clark Tested-by: Antonino Maniscalco Reviewed-by: Antonino Maniscalco Patchwork: https://patchwork.freedesktop.org/patch/661517/