- 20 Dec, 2011 40 commits
-
-
Jerome Glisse authored
Somewhat specializaed sub-allocator designed to perform sub-allocation for command buffer not only for current cs ioctl but for future command submission ioctl as well. Patch also convert current ib pool to use the sub allocator. Idea is that ib poll buffer can be share with other command buffer submission not having 64K granularity. v2 Harmonize pool handling and add suspend/resume callback to pin/unpin sa bo (tested on rv280, rv370, r420, rv515, rv610, rv710, redwood, cayman, rs480, rs690, rs880) v3 Simplify allocator v4 Fix radeon_ib_get error path to properly free fence Signed-off-by:
Jerome Glisse <jglisse@redhat.com> Signed-off-by:
Dave Airlie <airlied@redhat.com>
-
Alex Deucher authored
Signed-off-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
Christian König <deathsimple@vodafone.de> Signed-off-by:
-
Alex Deucher authored
cayman is wb only and doesn't have a VC. Signed-off-by:
-
Alex Deucher authored
Some rptr/wptrs fields have different offsets and not all rings are pm4 so add a new nop field. Signed-off-by:
-
Christian König authored
Also test multiple waits on the same semaphore. Signed-off-by:
-
Jerome Glisse authored
Add a start fence driver helper function which will be call once for each ring and will compute cpu/gpu addr for fence depending on wether to use wb buffer or scratch reg. This patch replace initialize fence driver separately which was broken in regard of GPU lockup. The fence list for created, emited, signaled must be initialize once and only from the asic init callback not from the startup call back which is call from the gpu reset. v2: With this in place we no longer need to know the number of rings in fence_driver_init, also writing to the scratch reg before knowing its offset is a bad idea. v3: rebase on top of change to previous patch in the serie Signed-off-by: -
Christian König authored
Those debugfs files aren't r600 specific, so they shouldn't be in r600.c. Move them to radeon_ring.c and also add functionality to dump CP1 & CP2 ring informations. Signed-off-by:
-
Christian König authored
That naming seems to make more sense, since we not only want to run PM4 rings with it. Signed-off-by:
-
Christian König authored
Disable the additional compute rings on cayman until their setup is fully implemented. Signed-off-by:
-
Christian König authored
Split counting of emited fences out of power management into a seperate function. Signed-off-by:
-
Christian König authored
Emitting fences, semaphores and ib works differently on different ring, so its is easier to maintain separate functions for each ring. Signed-off-by:
-
Christian König authored
Tests syncing between all rings by using semaphores and fences. v2: use radeon_testing as a bit flag rather than on/off switch this allow to test for one thing at a time (bo_move or semaphore test). It kind of break the usage if user wheren't using 1 for bo move test but as it's a test feature i believe it's ok. Signed-off-by: -
Christian König authored
Replace cp, cp1 and cp2 members with just an array of radeon_cp structs. Signed-off-by:
-
Christian König authored
Every ring seems to have the concept of read and write pointers. Make the register offset variable so we can use the functions for different types of rings. Signed-off-by:
-
Christian König authored
Give all asic and radeon_ring_* functions a radeon_cp parameter, so they know the ring to work with. Signed-off-by:
-
Christian König authored
They are used to sync between rings, while fences sync between a ring and the cpu. v2 Fix radeon_semaphore_driver_fini when no semaphore were allocated. v3 Initialize list early on to avoid issue in case or early error Signed-off-by:
-
Alex Deucher authored
For supporting multiple CP ring buffers, async DMA engines and UVD. We still need a way to synchronize between engines. v2 initialize unused fence driver ring to avoid issue in suspend/unload Signed-off-by:
-
Christian König authored
Better fix it before this obvious typo spreads even more. Signed-off-by:
-
Christian König authored
Only check the previously checked relocs for duplicates. Also leaving the handle uninitialized isn't such a good idea. Signed-off-by:
-
Christian König authored
Having registered debugfs files globally causes the files to not show up on the second, third etc.. card in the system. v2: fix crash on module unloading v3: fix space indentation Signed-off-by:
-
Alan Cox authored
The GCT setup was used on Moorestown. The Oaktrail version uses a normal PC interface. That means we must also honour the dither info from the BIOS data. Signed-off-by:
Alan Cox <alan@linux.intel.com> Signed-off-by:
-
Alan Cox authored
The move to connectors breaks Oaktrail again if we have memory poisoning enabled. Fix this. Signed-off-by:
-
Dave Airlie authored
Merge in the upstream tree to bring in the mainline fixes. Conflicts: drivers/gpu/drm/exynos/exynos_drm_fbdev.c drivers/gpu/drm/nouveau/nouveau_sgdma.c
-
Alan Cox authored
This switches the ifdef to match the Kconfig so that Oaktrail probing occurs and adds some additional minor bulletproofing. Tested on a Fujtisu Stylistic Q550 internal display. HDMI might work but that remains to be seen. Signed-off-by:
-
Patrik Jakobsson authored
We currently don't have support for parsing SDVO mappings from BIOS so we're guessing the bus switch parameter. This isn't working so hardcode it to a configuration known to work on most poulsbo hardware. Signed-off-by:
Patrik Jakobsson <patrik.r.jakobsson@gmail.com> Signed-off-by:
-
Patrik Jakobsson authored
Replace psb_intel_output with psb_intel_encoder and psb_intel_connector Signed-off-by:
-
Patrik Jakobsson authored
Replace psb_intel_output with psb_intel_encoder and psb_intel_connector. Things will need to be cleaned up and tested so consider this an initial patch for Cedarview. Signed-off-by:
-
Patrik Jakobsson authored
Our current SDVO implementation is not working properly, so replace it with a modified version of the i915. Further testing and debugging is needed to make sure we can handle the different SDVO setups and wiring. Signed-off-by:
-
Patrik Jakobsson authored
Before we integrate the new SDVO code we need GMBUS support Signed-off-by:
-
Patrik Jakobsson authored
LVDS for PSB now uses psb_intel_encoder and psb_intel_connectors instead of psb_intel_output. i2c_bus and ddc_bus are moved to lvds_priv. There was also a pointer to mode_dev (for no obvious reason) that we now get directly from dev_priv. Signed-off-by:
-
Patrik Jakobsson authored
Fix cases where we need to know what encoder type is behind a given connector. Signed-off-by:
-
Patrik Jakobsson authored
Signed-off-by:
-
Patrik Jakobsson authored
First step towards adding i915 alike encoder and connector abstractions. This will make life easier when adding i915 output code into our driver. It also removes the old psb_intel_output struct. Signed-off-by:
-
Dave Airlie authored
This just fixes up the staging code to keep building. Signed-off-by: -
Ville Syrjälä authored
Otherwise each driver would need to keep the information inside their own framebuffer object structure. Also add offsets[]. BOs on the other hand are driver specific, so those can be kept in driver specific structures. Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:
-
Ville Syrjälä authored
Signed-off-by:
-
Ville Syrjälä authored
Signed-off-by:
-
Ville Syrjälä authored
Signed-off-by:
-
Ville Syrjälä authored
Help drivers a little by guaranteeing that crtc_x+crtc_w and crtc_y+crtc_h don't overflow. Signed-off-by:
-
Ville Syrjälä authored
Make sure the source coordinates stay within the buffer. Signed-off-by:
-
