* 'drm-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6: (52 commits) drm/kms: Init the CRTC info fields for modes forced from the command line. drm/radeon/r600: CS parser updates drm/radeon/kms: add debugfs for power management for AtomBIOS devices drm/radeon/kms: initial mode validation support drm/radeon/kms/atom/dce3: call transmitter init on mode set drm/radeon/kms: store detailed connector info drm/radeon/kms/atom/dce3: fix up usPixelClock calculation for Transmitter tables drm/radeon/kms/r600: fix rs880 support v2 drm/radeon/kms/r700: fix some typos in chip init drm/radeon/kms: remove some misleading debugging output drm/radeon/kms: stop putting VRAM at 0 in MC space on r600s. drm/radeon/kms: disable D1VGA and D2VGA if enabled drm/radeon/kms: Don't RMW CP_RB_CNTL drm/radeon/kms: fix coherency issues on AGP cards. drm/radeon/kms: fix rc410 suspend/resume. drm/radeon/kms: add quirk for hp dc5750 drm/radeon/kms/atom: fix potential oops in spread spectrum code drm/kms: typo fix drm/radeon/kms/atom: Make card_info per device drm/radeon/kms/atom: Fix DVO support ...
When we are evicting from VRAM->RAM we allocate the ttm object, but we don't set the caching policy on it before blitting into it. This means on AGP we end up blitting into cached pages, and the CPU later flushes out on top of them. This was mostly seen as font corruption. The other question is why we don't evict VRAM->GTT in a lot of cases, this would save us some cache transitions since a lot of objects that are evicted from VRAM will probably end up being pulled back in a few operations later, and evicting them to system memory involves 2 unnecessary cache transitions. Signed-off-by: Dave Airlie <airlied@redhat.com>
conflict in radeon since new init path merged with vga arb code. Conflicts: drivers/gpu/drm/radeon/radeon.h drivers/gpu/drm/radeon/radeon_asic.h drivers/gpu/drm/radeon/radeon_device.c
the global refcount wasn't being increased after the first reference. this caused an oops on unload on a multi-gpu card. Signed-off-by: Dave Airlie <airlied@redhat.com>
* mark struct vm_area_struct::vm_ops as const * mark vm_ops in AGP code But leave TTM code alone, something is fishy there with global vm_ops being used. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This merges the TTM and drm cache flushing into one file in the drm core. Signed-off-by: Dave Airlie <airlied@redhat.com>
ttm: Fix error paths when kobject_add returns an error. Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
ttm: Remove a stray debug printout. Remove a re-init of the lru spinlock at device init. radeon: Fix the size of the bo_global allocation. Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
linux-next conflict reported needed resolution. Conflicts: drivers/gpu/drm/drm_crtc.c drivers/gpu/drm/drm_edid.c drivers/gpu/drm/i915/intel_sdvo.c drivers/gpu/drm/radeon/radeon_ttm.c drivers/gpu/drm/ttm/ttm_bo.c
Common resources, like memory accounting and swap lists should be global and not per device. Introduce a struct ttm_bo_global to accomodate this, and register it with sysfs. Add a small sysfs interface to return the number of active buffer objects. Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com> Signed-off-by: Dave Airlie <airlied@linux.ie>