linux/drivers/gpu/drm/ttm
Konrad Rzeszutek Wilk 36d7c537c3 drm/ttm/dma: Only call set_pages_array_wb when the page is not in WB pool.
Otherwise we are doing redundant work. Especially since the 'unbind'
and 'unpopulate' have been merged and nouveau driver ends up calling
it quite excessivly. On a GeForce 8600 GT with Gnome Shell (GNOME 3)
we end up spending about 54% CPU time in __change_page_attr_set_clr
checking the page flags.

The callgraph (annotated) looks as so before this patch:

    53.29%  gnome-shell  [kernel.kallsyms]                   [k] static_protections
            |
            --- static_protections
               |
               |--91.80%-- __change_page_attr_set_clr
               |          change_page_attr_set_clr
               |          set_pages_array_wb
               |          |
               |          |--96.55%-- ttm_dma_unpopulate
               |          |          nouveau_ttm_tt_unpopulate
               |          |          ttm_tt_destroy
               |          |          ttm_bo_cleanup_memtype_use
               |          |          ttm_bo_release
               |          |          kref_put
               |          |          ttm_bo_unref
               |          |          nouveau_gem_object_del
               |          |          drm_gem_object_free
               |          |          kref_put
               |          |          drm_gem_object_unreference_unlocked
               |          |          drm_gem_object_handle_unreference_unlocked.part.1
               |          |          drm_gem_handle_delete
               |          |          drm_gem_close_ioctl
               |          |          drm_ioctl
               |          |          do_vfs_ioctl
               |          |          sys_ioctl
               |          |          system_call_fastpath
               |          |          __GI___ioctl
               |          |
               |           --3.45%-- ttm_dma_pages_put
               |                     ttm_dma_page_pool_free
               |                     ttm_dma_unpopulate
               |                     nouveau_ttm_tt_unpopulate
               |                     ttm_tt_destroy
               |                     ttm_bo_cleanup_memtype_use
               |                     ttm_bo_release
               |                     kref_put
               |                     ttm_bo_unref
               |                     nouveau_gem_object_del
               |                     drm_gem_object_free
               |                     kref_put
               |                     drm_gem_object_unreference_unlocked
               |                     drm_gem_object_handle_unreference_unlocked.part.1
               |                     drm_gem_handle_delete
               |                     drm_gem_close_ioctl
               |                     drm_ioctl
               |                     do_vfs_ioctl
               |                     sys_ioctl
               |                     system_call_fastpath
               |                     __GI___ioctl
               |
                --8.20%-- change_page_attr_set_clr
                          set_pages_array_wb
                          |
                          |--93.76%-- ttm_dma_unpopulate
                          |          nouveau_ttm_tt_unpopulate
                          |          ttm_tt_destroy
                          |          ttm_bo_cleanup_memtype_use
                          |          ttm_bo_release
                          |          kref_put
                          |          ttm_bo_unref
                          |          nouveau_gem_object_del
                          |          drm_gem_object_free
                          |          kref_put
                          |          drm_gem_object_unreference_unlocked
                          |          drm_gem_object_handle_unreference_unlocked.part.1
                          |          drm_gem_handle_delete
                          |          drm_gem_close_ioctl
                          |          drm_ioctl
                          |          do_vfs_ioctl
                          |          sys_ioctl
                          |          system_call_fastpath
                          |          __GI___ioctl
                          |
                           --6.24%-- ttm_dma_pages_put
                                     ttm_dma_page_pool_free
                                     ttm_dma_unpopulate
                                     nouveau_ttm_tt_unpopulate
                                     ttm_tt_destroy
                                     ttm_bo_cleanup_memtype_use
                                     ttm_bo_release
                                     kref_put
                                     ttm_bo_unref
                                     nouveau_gem_object_del
                                     drm_gem_object_free
                                     kref_put
                                     drm_gem_object_unreference_unlocked
                                     drm_gem_object_handle_unreference_unlocked.part.1
                                     drm_gem_handle_delete
                                     drm_gem_close_ioctl
                                     drm_ioctl
                                     do_vfs_ioctl
                                     sys_ioctl
                                     system_call_fastpath
                                     __GI___ioctl

and after this patch all of that disappears.

Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2012-01-06 09:29:43 +00:00
..
Makefile drm/ttm: provide dma aware ttm page pool code V9 2011-12-06 10:39:33 +00:00
ttm_agp_backend.c drm/ttm: merge ttm_backend and ttm_tt V5 2011-12-06 10:39:17 +00:00
ttm_bo_manager.c drm/ttm: Use private locks for the default bo range manager 2010-11-09 13:33:24 +10:00
ttm_bo_util.c drm/ttm: simplify memory accounting for ttm user v2 2011-12-06 10:40:11 +00:00
ttm_bo_vm.c drm/ttm: introduce callback for ttm_tt populate & unpopulate V4 2011-12-06 10:39:24 +00:00
ttm_bo.c Merge tag 'v3.2-rc6' of /home/airlied/devel/kernel/linux-2.6 into drm-core-next 2011-12-20 14:43:53 +00:00
ttm_execbuf_util.c Revert "drm/ttm: add a way to bo_wait for either the last read or last write" 2011-10-27 18:28:37 +02:00
ttm_lock.c atomic: use <linux/atomic.h> 2011-07-26 16:49:47 -07:00
ttm_memory.c drm/ttm: provide dma aware ttm page pool code V9 2011-12-06 10:39:33 +00:00
ttm_module.c drm: move ttm global code to core drm 2010-08-04 09:46:06 +10:00
ttm_object.c atomic: use <linux/atomic.h> 2011-07-26 16:49:47 -07:00
ttm_page_alloc_dma.c drm/ttm/dma: Only call set_pages_array_wb when the page is not in WB pool. 2012-01-06 09:29:43 +00:00
ttm_page_alloc.c drm/ttm: isolate dma data from ttm_tt V4 2011-12-06 10:40:02 +00:00
ttm_tt.c drm/ttm: isolate dma data from ttm_tt V4 2011-12-06 10:40:02 +00:00