6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zack Rusin zack.rusin@broadcom.com
commit aba07b9a0587f50e5d3346eaa19019cf3f86c0ea upstream.
The kms paths keep a persistent map active to read and compare the cursor buffer. These maps can race with each other in simple scenario where: a) buffer "a" mapped for update b) buffer "a" mapped for compare c) do the compare d) unmap "a" for compare e) update the cursor f) unmap "a" for update At step "e" the buffer has been unmapped and the read contents is bogus.
Prevent unmapping of active read buffers by simply keeping a count of how many paths have currently active maps and unmap only when the count reaches 0.
Fixes: 485d98d472d5 ("drm/vmwgfx: Add support for CursorMob and CursorBypass 4") Cc: Broadcom internal kernel review list bcm-kernel-feedback-list@broadcom.com Cc: dri-devel@lists.freedesktop.org Cc: stable@vger.kernel.org # v5.19+ Signed-off-by: Zack Rusin zack.rusin@broadcom.com Link: https://patchwork.freedesktop.org/patch/msgid/20240816183332.31961-2-zack.ru... Reviewed-by: Martin Krastev martin.krastev@broadcom.com Reviewed-by: Maaz Mombasawala maaz.mombasawala@broadcom.com [Shivani: Modified to apply on v6.6.y] Signed-off-by: Shivani Agarwal shivani.agarwal@broadcom.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/gpu/drm/vmwgfx/vmwgfx_bo.c | 13 +++++++++++-- drivers/gpu/drm/vmwgfx/vmwgfx_bo.h | 3 +++ 2 files changed, 14 insertions(+), 2 deletions(-)
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c @@ -331,6 +331,8 @@ void *vmw_bo_map_and_cache(struct vmw_bo void *virtual; int ret;
+ atomic_inc(&vbo->map_count); + virtual = ttm_kmap_obj_virtual(&vbo->map, ¬_used); if (virtual) return virtual; @@ -353,11 +355,17 @@ void *vmw_bo_map_and_cache(struct vmw_bo */ void vmw_bo_unmap(struct vmw_bo *vbo) { + int map_count; + if (vbo->map.bo == NULL) return;
- ttm_bo_kunmap(&vbo->map); - vbo->map.bo = NULL; + map_count = atomic_dec_return(&vbo->map_count); + + if (!map_count) { + ttm_bo_kunmap(&vbo->map); + vbo->map.bo = NULL; + } }
@@ -390,6 +398,7 @@ static int vmw_bo_init(struct vmw_privat BUILD_BUG_ON(TTM_MAX_BO_PRIORITY <= 3); vmw_bo->tbo.priority = 3; vmw_bo->res_tree = RB_ROOT; + atomic_set(&vmw_bo->map_count, 0);
params->size = ALIGN(params->size, PAGE_SIZE); drm_gem_private_object_init(vdev, &vmw_bo->tbo.base, params->size); --- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h @@ -68,6 +68,8 @@ struct vmw_bo_params { * @map: Kmap object for semi-persistent mappings * @res_tree: RB tree of resources using this buffer object as a backing MOB * @res_prios: Eviction priority counts for attached resources + * @map_count: The number of currently active maps. Will differ from the + * cpu_writers because it includes kernel maps. * @cpu_writers: Number of synccpu write grabs. Protected by reservation when * increased. May be decreased without reservation. * @dx_query_ctx: DX context if this buffer object is used as a DX query MOB @@ -86,6 +88,7 @@ struct vmw_bo { struct rb_root res_tree; u32 res_prios[TTM_MAX_BO_PRIORITY];
+ atomic_t map_count; atomic_t cpu_writers; /* Not ref-counted. Protected by binding_mutex */ struct vmw_resource *dx_query_ctx;