aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/radeon/r600_pipe_common.c
Commit message (Collapse)AuthorAgeFilesLines
* radeonsi: make const and stream uploaders allocate read-only memoryMarek Olšák2017-12-061-2/+5
| | | | | | | and anything that clones these uploaders, like u_threaded_context. Tested-by: Dieter Nützel <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: use a separate allocator for fine fencesMarek Olšák2017-12-061-0/+7
| | | | | Tested-by: Dieter Nützel <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium/u_upload_mgr: allow drivers to specify pipe_resource::flagsMarek Olšák2017-12-051-2/+2
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* r300,r600,radeonsi: replace RADEON_FLUSH_* with PIPE_FLUSH_*Marek Olšák2017-11-291-4/+4
| | | | | | and handle PIPE_FLUSH_HINT_FINISH in r300. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: remove r600_common_screenMarek Olšák2017-11-291-20/+20
| | | | | | | | | | Most files in gallium/radeon now include si_pipe.h. chip_class and family are now here: sscreen->info.family sscreen->info.chip_class Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move shader debug helpers out of r600_pipe_common.cMarek Olšák2017-11-291-11/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: dismantle si_common_screen_init/destroyMarek Olšák2017-11-291-154/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: set all pipe buffer functions in r600_buffer_common.cMarek Olšák2017-11-291-19/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move video queries into si_get.cMarek Olšák2017-11-291-6/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: remove more functions from r600_pipe_common.cMarek Olšák2017-11-291-31/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move/remove ac_shader_binary helpersMarek Olšák2017-11-291-22/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move all get functions to si_get.c; disk_cache_create to si_pipe.cMarek Olšák2017-11-291-410/+1
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: remove r600_common_context::clear_bufferMarek Olšák2017-11-291-11/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: enable DCC with MSAA for VIMarek Olšák2017-11-291-0/+2
| | | | | Tested-by: Dieter Nützel <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium/radeon: disable the cache when nir backend enabledTimothy Arceri2017-11-151-0/+4
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: remove DBG_NO_DISCARD_RANGEMarek Olšák2017-11-091-2/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move fence functions to si_fence.cNicolai Hähnle2017-11-091-267/+0
| | | | Reviewed-by: Marek Olšák <[email protected]>
* util: move os_time.[ch] to src/utilNicolai Hähnle2017-11-091-1/+1
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: move pipe debug callback to si_contextNicolai Hähnle2017-11-091-12/+0
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: use ac_get_llvm_processor_nameMarek Olšák2017-11-071-35/+2
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* winsys/amdgpu: Add R600_DEBUG flag to reserve VMID per ctx.Andrey Grodzovsky2017-11-031-0/+1
| | | | | | | | | | Fixes reverted patch f03b7c9 by doing VMID reservation per process and not per context. Also updates required amdgpu libdrm version since the change involved interface updates in amdgpu libdrm. Signed-off-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Marek Olšák <[email protected]>
* radeonsi: remove 'Authors:' commentsMarek Olšák2017-11-021-3/+0
| | | | | | | It's inaccurate. Instead, see the copyright and use "git log" and "git blame" to know the authorship. Acked-by: Nicolai Hähnle <[email protected]>
* Revert "winsys/amdgpu: Add R600_DEBUG flag to reserve VMID per ctx."Marek Olšák2017-11-011-1/+0
| | | | | | This reverts commit f03b7c9ad92c1656a221297819fbc6d065cc0af7. The libdrm interface is wrong.
* winsys/amdgpu: Add R600_DEBUG flag to reserve VMID per ctx.Andrey Grodzovsky2017-10-311-0/+1
| | | | | Signed-off-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Marek Olšák <[email protected]>
* radeonsi: implement sync_file import/exportMarek Olšák2017-10-121-1/+76
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* ac: add radeon_info::has_sync_filecros-mesa-17.2.3-vanillachadv/cros-mesa-17.2.3-vanillaMarek Olšák2017-10-121-0/+1
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: shrink r600d_common.h and stop using itMarek Olšák2017-10-091-0/+1
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: import cayman_msaa.c from drivers/radeonMarek Olšák2017-10-091-1/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: remove r600_emit_relocMarek Olšák2017-10-091-3/+4
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: import r600_streamout from drivers/radeonMarek Olšák2017-10-091-12/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: disable primitive binning on Vega10 (v2)Marek Olšák2017-10-091-0/+2
| | | | | | | | | | | | | | | Our driver implementation is known to decrease performance for some tests, but we don't know if any apps and benchmarks (e.g. those tested by Phoronix) are affected. This disables the feature just to be safe. Set this to enable partial primitive binning: R600_DEBUG=dpbb Set this to enable full primitive binning: R600_DEBUG=dpbb,dfsm v2: add new debug options Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: enumerize DBG flagsMarek Olšák2017-10-091-50/+49
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move si_draw_rectangle into si_state_draw.cMarek Olšák2017-10-071-83/+0
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium/radeon: remove old r600g code checking chip_class and familyMarek Olšák2017-10-071-89/+10
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium: Remove util_format_s3tc_init()Matt Turner2017-10-021-1/+0
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Emil Velikov <[email protected]>
* radeonsi: move current_rast_prim into si_contextNicolai Hähnle2017-10-021-4/+0
| | | | | | v2: rebase fixes Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: move and rename scissor and viewport state and functionsNicolai Hähnle2017-10-021-1/+0
| | | | | | v2: change GET_MAX_SCISSOR to SI_MAX_SCISSOR Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: fix maximum advertised point size / line widthNicolai Hähnle2017-10-021-6/+1
| | | | | | | | | | The hardware registers store the half-size/width in 12.4 fixed point format, so 8192 is the maximum. Fixes dEQP-GLES3.functional.rasterization.* Cc: [email protected] Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: move current_rast_prim to r600_common_contextNicolai Hähnle2017-10-021-0/+4
| | | | | | | | | | We'll use it in the scissors / clip / guardband state. v2: avoid a performance regression on r600 when applied to (pre-fork) stable branches Cc: [email protected] Reviewed-by: Marek Olšák <[email protected]>
* r600: fork and import gallium/radeonMarek Olšák2017-09-261-100/+72
| | | | | | | | | | | This marks the end of code sharing between r600 and radeonsi. It's getting difficult to work on radeonsi without breaking r600. A lot of functions had to be renamed to prevent linker conflicts. There are also minor cleanups. Acked-by: Dave Airlie <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: enable out-of-order rasterization when possible on VI and GFX9 dGPUsNicolai Hähnle2017-09-181-0/+1
| | | | | | | | | This does not take commutative blending into account yet. R600_DEBUG=nooutoforder disables it. Reviewed-by: Marek Olšák <[email protected]> Tested-by: Dieter Nützel <[email protected]>
* gallium: add PIPE_QUERY_OCCLUSION_PREDICATE_CONSERVATIVENicolai Hähnle2017-09-181-1/+2
| | | | | | | | | | | | | | | | | To be able to properly distinguish between GL_ANY_SAMPLES_PASSED and GL_ANY_SAMPLES_PASSED_CONSERVATIVE. This patch goes through all drivers, having them treat the two query types identically, except: 1. radeon incorrectly enabled conservative mode on PIPE_QUERY_OCCLUSION_PREDICATE. We now do it correctly, only on PIPE_QUERY_OCCLUSION_PREDICATE_CONSERVATIVE. 2. st/mesa uses the new query type. Fixes dEQP-GLES31.functional.fbo.no_attachments.* Reviewed-by: Marek Olšák <[email protected]>
* amd/common: get ME/PFP/CE firmware feature versions as wellNicolai Hähnle2017-09-131-0/+3
| | | | Reviewed-by: Marek Olšák <[email protected]>
* radeonsi: implement pipe_context::fence_server_syncMarek Olšák2017-09-111-0/+45
| | | | | | This will be more useful once we have sync_file support. Reviewed-by: Nicolai Hähnle <[email protected]>
* gallium/radeon: add a proper fail path for calloc in r600_flush_from_stMarek Olšák2017-09-111-3/+6
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* r600: handle the non-TXF_LZ support path.Dave Airlie2017-09-111-1/+1
| | | | | | | it appears that texcoord.z/w will be 0 in all cases already, so just put them into the vbo always. Signed-off-by: Dave Airlie <[email protected]>
* gallium/u_blitter: use draw_rectangle for all blits except cubemapsMarek Olšák2017-09-111-1/+5
| | | | | | | Add ZW coordinates to the draw_rectangle callback and use it. Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Brian Paul <[email protected]>
* gallium/u_blitter: use draw_rectangle callback for layered clearsMarek Olšák2017-09-111-3/+10
| | | | | | | They are done with instancing. Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Brian Paul <[email protected]>
* gallium/u_blitter: add new union blitter_attrib to replace pipe_color_unionMarek Olšák2017-09-111-10/+10
| | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Tested-by: Brian Paul <[email protected]>
* gallium/radeon: use rectangles for 1D and 2D texture blitsMarek Olšák2017-09-111-7/+13
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>