| 78222e63630280b96488de0d51d1b2578e26f814 |
|
30-Jan-2013 |
Jerome Glisse <jglisse@redhat.com> |
r600g: add cs memory usage accounting and limit it v3 (backport for mesa 9.0)
We are now seing cs that can go over the vram+gtt size to avoid
failing flush early cs that goes over 70% (gtt+vram) usage. 70%
is use to allow some fragmentation.
The idea is to compute a gross estimate of memory requirement of
each draw call. After each draw call, memory will be precisely
accounted. So the uncertainty is only on the current draw call.
In practice this gave very good estimate (+/- 10% of the target
memory limit).
v2: Remove left over from testing version, remove useless NULL
checking. Improve commit message.
v3: Add comment to code on memory accounting precision
This version is a backport for mesa 9.0
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Marek Olšák <maraeo@gmail.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 8b2922ff5aa68b5dca90d5b83d1e31da48b069b3 |
|
23-Nov-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix broken streamout if streamout_begin caused a context flush
This fixes graphics corruption in the case where the DISCARD_RANGE flag
is used to map a buffer.
NOTE: This is a candidate for the stable branches.
(cherry picked from commit cff4c948ed2708a6eb4b090ae87443a707cbd67f)
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1106bbc7d6cd024d1e20573e281ea4c48c5eaf19 |
|
25-Sep-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix streamout on RS780 and RS880
The latest kernel from git is required. Transform feedback (along with GL3.0)
is turned off on older kernels.
(cherry picked from commit d063c7b1421a6e8ad4e2efc1bf913920766cc5ee)
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 8698a3b85dd89c5d2fa473e7942b7dc8d25f3c8f |
|
02-Aug-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: implement MSAA for r700
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| a01791add08fbcb5386e0e9209ba21ed58fbdc42 |
|
22-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: implement set_sample_mask
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 6517225078a6a56c9fb3c1ea9f310992e6400b77 |
|
22-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: implement alpha-to-coverage
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 2df399c34bb39122a45bdd5b430b48346542e1cb |
|
01-Aug-2012 |
Jerome Glisse <jglisse@redhat.com> |
r600g: atomize sampler state v2
Use atom for sampler state. Does not provide new functionality
or fix any bug. Just a step toward full atom base r600g.
v2: Split seamless on r6xx/r7xx into it's own atom. Make sure it's
emited after sampler and with a pipeline flush before otherwise
it does not take effect.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1cfb55c5098e17cd8e6188fc6382b393a9013e4c |
|
18-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove debug code for streamout
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| ff9a49328e5d8b647aeb7511a01e5a41592a2799 |
|
18-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: inline r600_context_draw_opaque_count
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bc2f5fc01e678ecd19f921e5bc7bd55c131dde66 |
|
18-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: atomize alphatest state
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| c26fadf195876271e559f844c1fc88effa6a60c1 |
|
15-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove unused code after conversion of sampler views
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 5d8d4252f2f6632fc455dcf1079c95495ef445ac |
|
14-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: convert sampler view emission into atoms
Vertex and constant buffers are emitted in the same way.
This is mainly a simplification of the code. The cleanup is in another patch.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 585baac652ffa172fb3fbbdd4c7559d03b7c27ef |
|
06-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: do fine-grained vertex buffer updates
If only some buffers are changed, the other ones don't have to re-emitted.
This uses bitmasks of enabled and dirty buffers just like
emit_constant_buffers does.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f4f2e8ebe1bad91b2c4339a8280d516738bd27d2 |
|
14-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't call inval_shader_cache in r600_context_flush twice
It's already called in r600_constant_buffers_dirty.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| c2f444c54db82a53f63fa7a103770ee5eedc3559 |
|
12-Jul-2012 |
Tom Stellard <tstellar@gmail.com> |
r600g: Emit vertex buffers using the same method as constant buffers
Signed-off-by: Marek Olšák <maraeo@gmail.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 4fe74412cfdba9af6ce878aebbb0c367f19cbb4b |
|
07-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: move CB_SHADER_MASK setup into cb_misc_state
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| a1a1ff5ec09acda0c4849c9e41a37ae82a80f000 |
|
07-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: move MULTIWRITE setup into cb_misc_state for r6xx-r7xx
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 0ea76916e63f8fc556f5e8f5a46c196d317cd5ad |
|
07-Jul-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: move CB_TARGET_MASK setup into new cb_misc_state
to remove some overhead from draw_vbo. This is a derived state.
BTW, I've got no idea how compute interacts with 3D here, but it should
use cb_misc_state, so that 3D and compute don't conflict.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 3d3194e93cbfe0fc42dd96218e8e999f5ccfe726 |
|
26-Jun-2012 |
Tom Stellard <thomas.stellard@amd.com> |
r600g: Add r600_context_pipe_state_emit()
This function is used when dispatching compute shader in order to avoid
mixing compute and 3D registers in the context's dirty list. This
allows the compute code to resuse 3D functions like evergreen_cb, which
return a struct r600_pipe_state and still have control over when and how
the register writes are emitted.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| e00e1586dd94c1bbcd1ecfd9649fde9281be1977 |
|
25-Jun-2012 |
Tom Stellard <thomas.stellard@amd.com> |
r600g: Add pkt_flag parameter to r600_context_block_emit_dirty()
This allows the shader type bit to be set in the pm4 header when
emitting registers for compute shaders.
Reviewed-by: Marek Olšák <maraeo@gmail.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| cb4d1d377d7a1143db48feb48ef54dfa3b29487d |
|
14-Jun-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix lockups with streamout on r7xx
This requires the latest streamout kernel patches.
Streamout is disabled by default on r7xx, so this patch is safe for regular
users.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f01594be0ee34758fc62274b9e967a2a4c3595df |
|
14-Jun-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: compute CS space for streamout correctly, add comments
SET_CONTEXT_REG was not counted in.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bb07e251318d1b224fb78fc0bd4200216c42a5cf |
|
14-Jun-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: set SMX_ACTION_ENA to fix streamout cache flushes on some chipsets
It helps on R7xx.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 0a6120244e66494db070ce875c0a464fbc5b15a1 |
|
26-Apr-2012 |
Marek Olšák <maraeo@gmail.com> |
winsys/radeon: simplify buffer map/unmap functions
The idea is not to use pb_map and pb_unmap wrappers, calling straight
into the winsys.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| a52b3338c6e51421e3836ae210cd98d9c1ec337b |
|
10-Apr-2012 |
Marek Olšák <maraeo@gmail.com> |
u_vbuf: remove u_vbuf_resource
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| c7d0e9ec32edfbb6c77715d51170608deb41f1b7 |
|
02-Apr-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: invalidate caches at the beginning of CS
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| a633d82ce352a69fd54a2b0e262a91f27f478119 |
|
02-Apr-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove dead code after the rework
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 68bbfc1afe210d82acfb14a78b0fd8c436a8f78c |
|
01-Apr-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rework state emission of constant buffers
Framerate in ipers:
before: 43.6 FPS
after: 46.6 FPS
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| c76462b45f1e3a0aa2ee7971191e30e8a5f52015 |
|
30-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rework state emission of vertex buffers
This reduces a little of CPU overhead.
The idea is to translate pipe vertex buffers directly into the CS
and not using any intermediate representations.
Framerate in Torcs:
before: 32.2
after: 34.6
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 42bc0b9b9dc31a15c08d409d14d25ccf19501255 |
|
01-Apr-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: unduplicate code for PS partial flush
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bc95152f1d3bf0dbf79d81e6b9956fd60198e3df |
|
31-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: determine in advance if hw has vertex cache
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| b4082f492b4b55df4c636445e47b97d1f1e4b5b2 |
|
21-Mar-2012 |
Alex Deucher <alexander.deucher@amd.com> |
r600g: add support for TN (trinity) APUs
Note: this is a candidate for the stable branches.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 91ddfa72b4384d04e4d84e60b4cdf3efee7ef49b |
|
19-Mar-2012 |
Alex Deucher <alexander.deucher@amd.com> |
r600g: fix r600_get_backend_mask() for virtual addresses
I don't think we should hit this path for virtual addresses,
but just in case.
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f6546535c9c8ce001e081487b8cd30b6703c0f4d |
|
07-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: use SX_MISC to implement rasterizer discard
Yeah I am reworking it again. This is way simpler than the other methods.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| c7eaf274a9b7e3e5f4b060be2320eb605464ba5c |
|
08-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
Revert "r600g: fix and improve rasterizer discard for r600-r700"
I will use SX_MISC instead.
This reverts commit 597fd6dc8c424a35f8442d5fd9de708013a69830.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 80e4d18f84748f903cea07b9bba2d519cfa4163a |
|
08-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
Revert "r600g: rework rasterizer discard for evergreen"
I will use SX_MISC instead.
This reverts commit 734792e83fdc526623d8fe0a60479648c936bd53.
Conflicts:
src/gallium/drivers/r600/evergreen_hw_context.c
src/gallium/drivers/r600/evergreen_state.c
src/gallium/drivers/r600/r600_hw_context.c
src/gallium/drivers/r600/r600_pipe.h
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| e363dd5c7d8ba40984d937ad7487abbb5be439bc |
|
05-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: replace atom_ prefix with _state and _cmd suffixes
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 330b6c85c961b32f704ce8ec7dbf8cb7fc0b80a8 |
|
05-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: cleanup includes
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 45d731fdf9e469011a071a101b510e5b1578215f |
|
05-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove unused function r600_context_reg_bo
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 8a56ed8d6dc37c23d4695d0fc292d7bcc2976f08 |
|
05-Mar-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove unused definitions
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 428146c14fed5b41e3da9e72488396f214b23b0d |
|
27-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: initialize VS_GS_OUT_PRIM_TYPE
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 597fd6dc8c424a35f8442d5fd9de708013a69830 |
|
26-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix and improve rasterizer discard for r600-r700
For polygons, we have been using face culling with success, but that doesn't
work for points and lines.
Setting the point size and line width to 0 fixes it.
Also improve it even more by setting SCREEN_SCISSOR to a zero area.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 734792e83fdc526623d8fe0a60479648c936bd53 |
|
24-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rework rasterizer discard for evergreen
Implement it right using STRMOUT_CONFIG.RAST_STREAM. This fixes rasterizer
discard with points and lines.
This also adds another derived state. It's a combination of rasterizer discard
and streamout enable.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| ca78a47b056b5a4589933492db364f6e418abb0f |
|
26-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: make some scissor regs invariant on r600-r700
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| aacd653834264b70a2537aff7914762b39f0683f |
|
26-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rework scissor for r600-r700
We must use VPORT_SCISSOR, because that's the only one we can use for multiple
scissor rectangles in ARB_viewport_array.
R700 can use the VPORT_SCISSOR_ENABLE bit, but R600 doesn't have that and must
emit a 8192x8192 rectangle if scissor is disabled.
This commit also cleanups magic numbers in create_rs_state.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 09ec30f02830b3cbf9d6ac04d9497634dca86846 |
|
23-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't suspend timer queries for u_blitter
Timer queries should be able to measure the time spent in u_blitter as well.
Queries are split into two groups: the timer ones and the others (streamout,
occlusion), because we should only suspend non-timer queries for u_blitter,
and later if the non-timer queries are suspended, the context flush should
only suspend and resume the timer queries.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 83667acfd9feed932f6864092382e752466975ed |
|
23-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: move all query code into r600_query.c
And rename or inline functions where appropriate.
There is no reason to keep this stuff in r600_hw_context.c.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 4c02baade9d590787acae157ce9b2d09c6a9bf55 |
|
26-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix streamout cache flush for r600
Figured out by trial and error.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| df00dc3c817771ac5034a44dff2b14cd7759b207 |
|
22-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rework queries
We always mapped the query buffer in begin_query, causing stalls
if the buffer was busy.
This commit reworks it such that the query buffer is only mapped
in get_query_result as it's supposed to be.
The query buffer is no longer treated as a ring buffer. Instead, the results
are just appended and when the buffer is full, we create a new one. One query
can have more than one query buffer, though that's a very rare case.
Begin_query releases all query buffers.
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 5b84a8c3c3f85cd6b109861c6eebe3ece29d664e |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: cleanup magic numbers in set_xx_sampler
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 89293287b8f1a24c1405750007ca20bd32e02eab |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove duplicated evergreen_context_ps_partial_flush
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 84db8f9a2289f3ae7eb0af4825f4370e31e4ae11 |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove duplicated evergreen_context_pipe_state_set_sampler
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 985bd5487edc974f1ed7a9c50456478d959b51e2 |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove duplicated function r600_state_sampler_init
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 6a94c9dde0d62b5ccfcbe3dbafdadeab37fd48a6 |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix possible crashes in destroy_context when failing in create_context
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 9670e72f6ef85f0e6f6c9afe136dc67c017e3fae |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: consolidate common context init code
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f841b8a833f32074c79724c66c85cdc7b48087e2 |
|
21-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: define GROUP_FORCE_NEW_BLOCK in common header
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| fbebd431ec4e2e461a0cbcd5f3a04a000b8f6bbf |
|
03-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: move invariant register updates into start_cs for r6xx-r7xx
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| e2809849ecac69615ece294a55ee355afaac33d3 |
|
02-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: add a depth misc state which depends on occlusion queries
This is a state which is derived from other states and is actually the first
state which doesn't correspond to any gallium state.
There are two state flags:
bool occlusion_query_enabled
bool flush_depthstencil_enabled
Additional flags can be added later if needed, e.g. bool hiz_enabled.
The emit function will have to figure out the register values by itself.
It basically just emits the registers when the state changes.
This commit also adds a few helper functions for writing registers directly
into a command stream.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 8f5c172c854a52b2a69b8383a55c76f1faa5f704 |
|
02-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: consolidate the main draw code
The code was almost the same for r600 and eg. What can't be consolidated is
in the *_prepare functions.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 182fd4c54476ced92b61412b0cac5be32e257b53 |
|
02-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: move all invariant state from draw_vbo into start_cs
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f126253040654d52db134063a69ebaf0c417d410 |
|
31-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: turn init_config into a command buffer for starting a CS
This is the first pure command buffer. It contains CS initialization
packets and emits invariant state (i.e. the registers which never or rarely
change).
The affected registers are removed from *_hw_context.c, so that both ways
of emitting commands can co-exist.
v2: emit context_control in cayman's start_cs too
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 73191026b4198a3dce86dac015f973616f76363a |
|
02-Feb-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove redundant r600_context::vs_so_stride_in_dw
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| c0c979eebc076b95cc8d18a013ce2968fe6311ad |
|
30-Jan-2012 |
Jerome Glisse <jglisse@redhat.com> |
r600g: add support for common surface allocator for tiling v13
Tiled surface have all kind of alignment constraint that needs to
be met. Instead of having all this code duplicated btw ddx and
mesa use common code in libdrm_radeon this also ensure that both
ddx and mesa compute those alignment in the same way.
v2 fix evergreen
v3 fix compressed texture and workaround cube texture issue by
disabling 2D array mode for cubemap (need to check if r7xx and
newer are also affected by the issue)
v4 fix texture array
v5 fix evergreen and newer, split surface values computation from
mipmap tree generation so that we can get them directly from the
ddx
v6 final fix to evergreen tile split value
v7 fix mipmap offset to avoid to use random value, use color view
depth view to address different layer as hardware is doing some
magic rotation depending on the layer
v8 fix COLOR_VIEW on r6xx for linear array mode, use COLOR_VIEW on
evergreen, align bytes per pixel to a multiple of a dword
v9 fix handling of stencil on evergreen, half fix for compressed
texture
v10 fix evergreen compressed texture proper support for stencil
tile split. Fix stencil issue when array mode was clear by
the kernel, always program stencil bo. On evergreen depth
buffer bo need to be big enough to hold depth buffer + stencil
buffer as even with stencil disabled things get written there.
v11 rebase on top of mesa, fix pitch issue with 1d surface on evergreen,
old ddx overestimate those. Fix linear case when pitch*height < 64.
Fix r300g.
v12 Fix linear case when pitch*height < 64 for old path, adapt to
libdrm API change
v13 add libdrm check
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| b0337b679ad4c2feae59215104cfa60b58a619d5 |
|
31-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: shorten expressions accessing family and chip_class
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f661405637bba32c2cfbeecf6e2e56e414e9521e |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: use the new code for streamout flush as well
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 70d40fc6755a2e4b23c831143cb02133f30fbb02 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rename r600_reg::flush_flags -> sbu_flags
There is no other use for that.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 6d0a011acbea90537d142c86b8dd5aecd1e14390 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix computation of how many dwords is needed for a flush at the end of CS
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1c5625cdb96332a746c9eebe042a3a9dbe844351 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove unused r600_reg::flush_mask
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 0671400dcc72553b5f9719639e86069a1734dbe5 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove more dead code
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 74027a9ba250f07a77ee40e105c50c27e4fc3bc0 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove dead code for tracking relocations
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 4fd97b46a071a8dcec6ec6658dfc835e08f4e2c6 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove unused flush code
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 0813e58a3e41faf6f2072d034dfdc6198a3a1fee |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: rework cache flushing
This also significantly improves the RV670 flush by using the CB1 flush
*always* and also DEST_BASE_0_ENA, which appears to magically fix some tests.
I am not entirely sure, but it's possible that RV670 flushing is fixed
completely.
v2: fix cayman by flushing texture cache instead of vertex cache
Thanks to Dave Airlie for testing Cayman.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 2fe521fa419ee153afb6295289dc5e95d3229858 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: add a new simple API for state emission
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 621e0db71c5ddcb379171064a4f720c9cf01e888 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove redundant pm4 and pm4_cdwords in r600_context
These only mirrored radeon_winsys_cs.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 3b0467192057cc561b6d495d6445b2ed04fc3ebf |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: consolidate some context_draw code
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 302862defa61b2cee1ae24159aca306f090ca854 |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove the now-useless internal flush callback
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bc70bcfc6915883bd5c60712b70e9ed7e1d80ea8 |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't initialize the screen and winsys pointer twice
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| e4340c1908a6a3b09e1a15d5195f6da7d00494d0 |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: merge r600_context with r600_pipe_context
The split made no sense.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 78293b99b23268e6698f1267aaf40647c17d95a5 |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove u8,u16,u32,u64 types
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bed20f02a771f43e1c5092254705701c228cfa7f |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: get rid of r600_context_reg
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 85db52025189c88437f82369716ffe88429dc08b |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't use r600_context_reg on r6xx-r7xx
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 9344ab382a1765c1a7c2560e771485edf4954fe2 |
|
29-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: get rid of the mask in r600_pipe_reg
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| f9f8ce3ead04b5334bb323c37ec1f905ca5f58b2 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove unused streamout code
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| ce816dd831ba4e57b024631ff39591efae50ce14 |
|
30-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
r600g: disable printing debug info for streamout
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 3d993aff258378e0ca3246577198066317e82a5b |
|
27-Jan-2012 |
Marek Olšák <maraeo@gmail.com> |
Revert "r600g: use S_028B20_BUFFER_0_EN macro for streamout buffer enable"
This reverts commit d6cd514edbeca0de38561f66189748078d0dc602.
It broke multiple streamout buffers, because it only enabled the first one.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 54e8dcaad65cbe3603730414fd8d76ac53f89a86 |
|
20-Jan-2012 |
Vadim Girlin <vadimgirlin@gmail.com> |
r600g: implement clip vertex v2
Clip planes are uploaded as a constant buffer and used by the vertex
shader to produce corresponding clip distances for hw clipping.
Signed-off-by: Vadim Girlin <vadimgirlin@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| d6cd514edbeca0de38561f66189748078d0dc602 |
|
21-Jan-2012 |
Alex Deucher <alexander.deucher@amd.com> |
r600g: use S_028B20_BUFFER_0_EN macro for streamout buffer enable
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 90b36b50f51fd2ea906b609ae0270b9eff7a1848 |
|
21-Jan-2012 |
Alex Deucher <alexander.deucher@amd.com> |
r600g: update streamout support for virtual addresses
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 2449695e822421fdcaf1c66dffc12d7d705ea69d |
|
17-Dec-2011 |
Marek Olšák <maraeo@gmail.com> |
gallium: improve the pipe_stream_output_info struct (v2)
There are 3 changes:
1) stride is specified for each buffer, not just one, so that drivers don't
have to derive it from the outputs
2) new per-output property dst_offset, which specifies the offset
into the buffer in dwords where the output should be stored,
so that drivers don't have to compute the offsets manually;
this will also be useful for gl_SkipComponents
from ARB_transform_feedback3
3) register_mask is removed, instead, there is start_component
and num_components; register_mask with non-consecutive 1s
doesn't make much sense (some hardware cannot do packing of components)
Christoph Bumiller: fixed nvc0.
v2: resolve merge conflicts in Draw and clean it up
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bb1f0cf3508630a9a93512c79badf8c493c46743 |
|
02-Dec-2011 |
Jerome Glisse <jglisse@redhat.com> |
r600g: add support for virtual address space on cayman v11
Virtual address space put the userspace in charge of their GPU
address space. It's up to userspace to bind bo into the virtual
address space. Command stream can them be executed using the
IB_VM chunck.
This patch add support for this configuration. It doesn't remove
the 64K ib size limit thought this limit can be extanded up to
1M for IB_VM chunk.
v2: fix rendering
v3: fix rendering when using index buffer
v4: make vm conditional on kernel support add basic va management
v5: catch the case when we already have va for a bo
v6: agd5f: update on top of ioctl changes
v7: agd5f: further ioctl updates
v8: indentation cleanup + fix non cayman
v9: rebase against lastest mesa + improvement from Marek & Michel
v10: fix cut/paste bug
v11: don't rely on updated radeon_drm.h
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 543b2331d7b45a29ccd3530daa2389e87e65d89b |
|
08-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: implement transform feedback
r600: DONE.
r700: MOSTLY (done but locks up).
Evergreen: MOSTLY (done but doesn't work for an unknown reason).
The kernel support will come soon.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| b8428e63353373e769a9ba787875b49fdd25a1ee |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't set the query result to 0 after a call to get_query_result
get_query_result doesn't reset the result. Only begin_query does.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 72c1658554b2d1e90e687340c5db4fec939b3901 |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: the type of OCCLUSION_PREDICATE result should be boolean
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| b999be83744c9419fa3134201006451db119b81d |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: implement PIPE_QUERY_OCCLUSION_PREDICATE
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| ec3d2c7f089e0a4750dc42f10da914b58858499e |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: flush the whole context when CS is full, not just hw_context
Because we disable render condition in r600_flush, but not in r600_context_flush.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 6a490149d9b8b77ead81b4227f2860ec22095217 |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: turn some CS overflow checks into assertions
They're not really exhaustive and not so useful either.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 5222e16e2b611064c0add9eef7f56488c9b8fbe6 |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: put the rest of CS overflow checks in r600_need_cs_space
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 30a570e450da30fa73546c0ff54b4757678a5663 |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: flush caches regardless of render condition
What if somebody enables render condition just before we flush...
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 578b211be6a1466eea00fc4a8fc5d4efa4bfd62c |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: disable render condition at the end of CS, re-enable at the beginning
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 9564185b0a1ac54e7188f88e57ed1a8d09d4e7bb |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't suspend queries if they have already been suspended by blitter
And add some assertions.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| cb7c6c30d02ab46efc4ed711b2a632de50157347 |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: properly reserve CS space for queries_suspend
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1d09831253b82374774daee772e90a83de17517a |
|
10-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: reserve CS space for a draw command in begin_query and render_condition
There's no point in emitting those if you can't emit a draw command too.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 11bdd28796e303eda70f1d73bd2469597bd561bc |
|
09-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: consolidate checking whether CS is full
This adds a new function r600_need_cs_space. Currently, it's easy to overflow
the CS - queries are not counted in. I guess that's not the only case where
the driver may crap out.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1b375f8413d898a18c989c20d0dc16d3b1be33bc |
|
08-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: cosmetic changes in query code
Mainly updating comments and removing one use of a magic number.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 8187ad07546cae0ae91255c665315f86b03367a9 |
|
08-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: use modulo for computing index into query (ring) buffers
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 552e90bd6ae315ba7e0a331c6f66f8b5f1542631 |
|
08-Nov-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: make r600_query_result more generic
We'll soon start adding new query types, maybe even querying more than
one value per query.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| b40edc63d7d272ac132fded52f3119f4780c7e6b |
|
29-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: shortcut ctx->screen->ws to ctx->ws
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 7684808f9ce2da944be3ca9cc1ac4661d9b167fb |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove one pointless flush
It's not useful for anything.
The rest of the patch is just a cleanup resulting
from some of the variables being no longer used.
There are no piglit regressions.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bbad5103e61806ef6d6a0d5e53abc4b9e77a7fe9 |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: get backend mask after the context is fully set up
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1e1a7011ee8f7e954b9f77985665495e53efae22 |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't flush before reading query results
Taken care of by the winsys.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| e9b6f21a5054b2639a1dfc6401e4e9053dce5394 |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: only maintain the list of active queries
And not all existing queries. The only reason we have that list is to be able
to suspend and resume the active ones.
This reduces looping over queries when suspending and resuming.
The queries no longer have to track some of their states.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| b70a51318a43f489611f4fc2f93887a362f5c99f |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: cleanup r600_query_result
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 1de02d7364fe7a2ea51848e67b65144fe4a8c7f4 |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: don't flush in r600_get_backend_mask
The winsys does the flush in buffer_map.
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| 6e79127e79e6ff80e3715a2190e437cfa92b4e96 |
|
28-Oct-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: remove redundant variable r600_query::buffer_size
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| aae342cff098034b6f685a81ecb4cd5e3fa92162 |
|
30-Sep-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: fix a compiler warning
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|
| bf0baa771709e1076613cf9f9bed6ac004dd0439 |
|
27-Sep-2011 |
Marek Olšák <maraeo@gmail.com> |
r600g: move all files from winsys/r600 into drivers/r600
Be sure to reconfigure after this commit.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
/external/mesa3d/src/gallium/drivers/r600/r600_hw_context.c
|