kernel/linux.git/include/drm/drm_mm.h, branch v6.18.22

drm/mm: Remove unused drm_mm_replace_node

2024-06-06T14:24:13+00:00

Last caller was removed with commit 078a5b498d6a ("drm/tests: Remove slow tests"). Cc: Maxime Ripard Acked-by: Maxime Ripard Link: https://patchwork.freedesktop.org/patch/msgid/20240604175438.48125-1-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi

drm: Replace kernel.h with the necessary inclusions

2021-12-09T12:46:13+00:00

When kernel.h is used in the headers it adds a lot into dependency hell, especially when there are circular dependencies are involved. Replace kernel.h inclusion with the list of what is really being used. Signed-off-by: Andy Shevchenko Signed-off-by: Thomas Zimmermann Link: https://patchwork.freedesktop.org/patch/msgid/20211110102423.54282-1-andriy.shevchenko@linux.intel.com

drm: fix spelling error in comments

2020-09-17T11:39:44+00:00

Change the comment typo: "manger" -> "manager". Signed-off-by: Wang Qing Signed-off-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/1600308275-32094-1-git-send-email-wangqing@vivo.com

drm/mm: optimize rb_hole_addr rbtree search

2020-05-05T11:39:38+00:00

Userspace can severely fragment rb_hole_addr rbtree by manipulating alignment while allocating buffers. Fragmented rb_hole_addr rbtree would result in large delays while allocating buffer object for a userspace application. It takes long time to find suitable hole because if we fail to find a suitable hole in the first attempt then we look for neighbouring nodes using rb_prev()/rb_next(). Traversing rbtree using rb_prev()/rb_next() can take really long time if the tree is fragmented. This patch improves searches in fragmented rb_hole_addr rbtree by modifying it to an augmented rbtree which will store an extra field in drm_mm_node, subtree_max_hole. Each drm_mm_node now stores maximum hole size for its subtree in drm_mm_node->subtree_max_hole. Using drm_mm_node->subtree_max_hole, it is possible to eliminate a complete subtree if that subtree is unable to serve a request hence reducing number of rb_prev()/rb_next() used. With this patch applied, 1 million bo allocs on amdgpu took ~8 sec, compared to 50k bo allocs which took 28 sec without it. partial test code: int test_fragmentation(void) { int i = 0; uint32_t minor_version; uint32_t major_version; struct amdgpu_bo_alloc_request request = {}; amdgpu_bo_handle vram_handle[MAX_ALLOC] = {}; amdgpu_device_handle device_handle; request.alloc_size = 4096; request.phys_alignment = 8192; request.preferred_heap = AMDGPU_GEM_DOMAIN_VRAM; int fd = open("/dev/dri/card0", O_RDWR | O_CLOEXEC); amdgpu_device_initialize(fd, &major_version, &minor_version, &device_handle); for (i = 0; i < MAX_ALLOC; i++) { amdgpu_bo_alloc(device_handle, &request, &vram_handle[i]); } for (i = 0; i < MAX_ALLOC; i++) amdgpu_bo_free(vram_handle[i]); return 0; } v2: Use RB_DECLARE_CALLBACKS_MAX to maintain subtree_max_hole v3: insert_hole_addr() should be static a function fix return value of next_hole_high_addr()/next_hole_low_addr() Reported-by: kbuild test robot v4: Fix commit message. Signed-off-by: Nirmoy Das Reviewed-by: Chris Wilson Acked-by: Christian König Link: https://patchwork.freedesktop.org/patch/364341/ Signed-off-by: Christian König

drm/mm: Allow drm_mm_initialized() to be used outside of the locks

2020-03-16T10:31:21+00:00

Mark up the potential racy read in drm_mm_initialized(), as we want a cheap and cheerful check: [ 121.098731] BUG: KCSAN: data-race in _i915_gem_object_create_stolen [i915] / rm_hole [ 121.098766] [ 121.098789] write (marked) to 0xffff8881f01ed330 of 8 bytes by task 3568 on cpu 3: [ 121.098831] rm_hole+0x64/0x140 [ 121.098860] drm_mm_insert_node_in_range+0x3d3/0x6c0 [ 121.099254] i915_gem_stolen_insert_node_in_range+0x91/0xe0 [i915] [ 121.099646] _i915_gem_object_create_stolen+0x9d/0x100 [i915] [ 121.100047] i915_gem_object_create_region+0x7a/0xa0 [i915] [ 121.100451] i915_gem_object_create_stolen+0x33/0x50 [i915] [ 121.100849] intel_engine_create_ring+0x1af/0x280 [i915] [ 121.101242] __execlists_context_alloc+0xce/0x3d0 [i915] [ 121.101635] execlists_context_alloc+0x25/0x40 [i915] [ 121.102030] intel_context_alloc_state+0xb6/0xf0 [i915] [ 121.102420] __intel_context_do_pin+0x1ff/0x220 [i915] [ 121.102815] i915_gem_do_execbuffer+0x46b4/0x4c20 [i915] [ 121.103211] i915_gem_execbuffer2_ioctl+0x2c3/0x580 [i915] [ 121.103244] drm_ioctl_kernel+0xe4/0x120 [ 121.103269] drm_ioctl+0x297/0x4c7 [ 121.103296] ksys_ioctl+0x89/0xb0 [ 121.103321] __x64_sys_ioctl+0x42/0x60 [ 121.103349] do_syscall_64+0x6e/0x2c0 [ 121.103377] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 121.103403] [ 121.103426] read to 0xffff8881f01ed330 of 8 bytes by task 3109 on cpu 1: [ 121.103819] _i915_gem_object_create_stolen+0x30/0x100 [i915] [ 121.104228] i915_gem_object_create_region+0x7a/0xa0 [i915] [ 121.104631] i915_gem_object_create_stolen+0x33/0x50 [i915] [ 121.105025] intel_engine_create_ring+0x1af/0x280 [i915] [ 121.105420] __execlists_context_alloc+0xce/0x3d0 [i915] [ 121.105818] execlists_context_alloc+0x25/0x40 [i915] [ 121.106202] intel_context_alloc_state+0xb6/0xf0 [i915] [ 121.106595] __intel_context_do_pin+0x1ff/0x220 [i915] [ 121.106985] i915_gem_do_execbuffer+0x46b4/0x4c20 [i915] [ 121.107375] i915_gem_execbuffer2_ioctl+0x2c3/0x580 [i915] [ 121.107409] drm_ioctl_kernel+0xe4/0x120 [ 121.107437] drm_ioctl+0x297/0x4c7 [ 121.107464] ksys_ioctl+0x89/0xb0 [ 121.107489] __x64_sys_ioctl+0x42/0x60 [ 121.107511] do_syscall_64+0x6e/0x2c0 [ 121.107535] entry_SYSCALL_64_after_hwframe+0x44/0xa9 Signed-off-by: Chris Wilson Reviewed-by: Daniel Vetter Link: https://patchwork.freedesktop.org/patch/msgid/20200309121529.16497-1-chris@chris-wilson.co.uk

drm/mm: Convert drm_mm_node booleans to bitops

2019-10-04T12:43:13+00:00

A straightforward conversion of assignment and checking of the boolean state flags (allocated, scanned) into non-atomic bitops. The caller remains responsible for all locking around the drm_mm and its nodes. Signed-off-by: Chris Wilson Reviewed-by: Tvrtko Ursulin Link: https://patchwork.freedesktop.org/patch/msgid/20191003210100.22250-4-chris@chris-wilson.co.uk

drm/mm: Add a search-by-address variant to only inspect a single hole

2018-05-24T14:04:30+00:00

Searching for an available hole by address is slow, as there no guarantee that a hole will be available and so we must walk over all nodes in the rbtree before we determine the search was futile. In many cases, the caller doesn't strictly care for the highest available hole and was just opportunistically laying out the address space in a preferred order. In such cases, the caller can accept any address and would rather do so then do a slow walk. To be able to mix search strategies, the caller wants to tell the drm_mm how long to spend on the search. Without a good guide for what should be the best split, start with a request to try once at most. That is return the top-most (or lowest) hole if it fulfils the alignment and size requirements. v2: Documentation, by why of example (selftests) and kerneldoc. Signed-off-by: Chris Wilson Cc: Joonas Lahtinen Reviewed-by: Joonas Lahtinen Link: https://patchwork.freedesktop.org/patch/msgid/20180521082131.13744-2-chris@chris-wilson.co.uk

drm/mm: Reject over-sized allocation requests early

2018-05-24T14:04:16+00:00

As we keep an rbtree of available holes sorted by their size, we can very easily determine if there is any hole large enough that might satisfy the allocation request. This helps when dealing with a highly fragmented address space and a request for a search by address. To cache the largest size, we convert into the cached rbtree variant which tracks the leftmost node for us. However, currently we sorted into ascending size order so the leftmost node is the smallest, and so to make it the largest hole we need to invert our sorting. Signed-off-by: Chris Wilson Cc: Joonas Lahtinen Reviewed-by: Joonas Lahtinen Link: https://patchwork.freedesktop.org/patch/msgid/20180521082131.13744-1-chris@chris-wilson.co.uk

drm/drm_mm.h: Fix the name of the referenced function in comment

2017-11-02T13:10:47+00:00

drm_mm_insert_node_generic() is a simplified version of drm_mm_insert_node_in_range(), update comment to reflect correct function name. Signed-off-by: Liviu Dudau Reviewed-by: Alex Deucher Reviewed-by: Christian König Signed-off-by: Gustavo Padovan Link: https://patchwork.freedesktop.org/patch/msgid/20171101140445.2798-1-Liviu.Dudau@arm.com

lib/interval_tree: fast overlap detection

2017-09-09T01:26:49+00:00

Allow interval trees to quickly check for overlaps to avoid unnecesary tree lookups in interval_tree_iter_first(). As of this patch, all interval tree flavors will require using a 'rb_root_cached' such that we can have the leftmost node easily available. While most users will make use of this feature, those with special functions (in addition to the generic insert, delete, search calls) will avoid using the cached option as they can do funky things with insertions -- for example, vma_interval_tree_insert_after(). [jglisse@redhat.com: fix deadlock from typo vm_lock_anon_vma()] Link: http://lkml.kernel.org/r/20170808225719.20723-1-jglisse@redhat.com Link: http://lkml.kernel.org/r/20170719014603.19029-12-dave@stgolabs.net Signed-off-by: Davidlohr Bueso Signed-off-by: Jérôme Glisse Acked-by: Christian König Acked-by: Peter Zijlstra (Intel) Acked-by: Doug Ledford Acked-by: Michael S. Tsirkin Cc: David Airlie Cc: Jason Wang Cc: Christian Benvenuti Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds