summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* i965: Add Gen7+ tessellation engine state (3DSTATE_TE).Kenneth Graunke2015-12-111-8/+28
| | | | | Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* i965: Add Gen8+ tessellation evaluation shader state (3DSTATE_DS).Kenneth Graunke2015-12-111-7/+59
| | | | | Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* i965: Add tessellation shader push constant support.Kenneth Graunke2015-12-117-29/+63
| | | | | | | Based on a patch by Chris Forbes. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* i965: Add tessellation shader sampler support.Kenneth Graunke2015-12-114-1/+51
| | | | | | | Based on code by Chris Forbes and Fabian Bieler. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* i965: Add tessellation shader surface support.Kenneth Graunke2015-12-1110-11/+389
| | | | | | | | | | | | This is brw_gs_surface_state.c copy and pasted twice with search and replace. brw_binding_table.c code is similarly copy and pasted. v2: Drop dword_pitch related fields. Signed-off-by: Kenneth Graunke <[email protected]> Acked-by: Jason Ekstrand <[email protected]>
* i965: Make fs_visitor::emit_urb_writes set EOT for TES as well.Kenneth Graunke2015-12-111-1/+1
| | | | | | | | | | | | | Tessellation evaluation shaders work almost identically to vertex shaders - we have a set of URB writes at the end of the program, and the last one should terminate it. Geometry shaders really are the special case, where multiple EmitVertex() calls trigger URB writes in the middle of the program. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]>
* i965: Don't hardcode g1 for URB handles in fs_visitor::emit_urb_writes().Kenneth Graunke2015-12-111-4/+5
| | | | | | | | | Tessellation evaluation shaders will use g4 instead. For now, make an fs_reg called urb_handle and use that in place of hardcoding g1. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]>
* i965: Make brw_set_message_descriptor() non-static.Kenneth Graunke2015-12-112-1/+9
| | | | | | | | I want to use this directly from brw_vec4_generator.cpp. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]> Reviewed-by: Kristian Høgsberg <[email protected]>
* i965: Move brw_cs_fill_local_id_payload() to libi965_compilerKristian Høgsberg Kristensen2015-12-114-40/+43
| | | | | | This is a helper function for setting up the local invocation ID payload according to the cs_prog_data generated by the compiler. It's intended to be available to users of libi965_compiler so move it there.
* vc4: Add quick algebraic optimization for clamping of unpacked values.Eric Anholt2015-12-111-0/+18
| | | | | | | | | | | | | | | | GL likes to saturate your incoming color, but if that color's coming from unpacking from unorms, there's no point. Ideally we'd have a range propagation pass that cleans these up in NIR, but that doesn't seem to be going to land soon. It seems like we could do a one-off optimization in nir_opt_algebraic, except that doesn't want to operate on expressions involving unpack_unorm_4x8, since it's sized. total instructions in shared programs: 87879 -> 87761 (-0.13%) instructions in affected programs: 6044 -> 5926 (-1.95%) total estimated cycles in shared programs: 349457 -> 349252 (-0.06%) estimated cycles in affected programs: 6172 -> 5967 (-3.32%) No SSPD on openarena (which had the biggest gains, in its VS/CSes), n=15.
* vc4: When doing algebraic optimization into a MOV, use the right MOV.Eric Anholt2015-12-111-1/+6
| | | | | If there were src unpacks, changing to the integer MOV instead of float (for example) would change the unpack operation.
* vc4: Fix handling of src packs on in qir_follow_movs().Eric Anholt2015-12-111-2/+8
| | | | | | The caller isn't going to expect it from a return, so it would probably get misinterpreted. If the caller had an unpack in its reg, that's fine, but don't lose track of it.
* vc4: Add missing progress note in opt_algebraic.Eric Anholt2015-12-111-0/+1
|
* vc4: Add debugging of the estimated time to run the shader to shader-db.Eric Anholt2015-12-113-17/+50
|
* vc4: Fix handling of sample_mask output.Eric Anholt2015-12-112-6/+6
| | | | | | | I apparently broke this in a late refactor, in such a way that I decided its tests were some of those interminable ones that I should just blacklist from my testing. As a result, the refactors related to it were totally wrong.
* softpipe: enable GL_ARB_viewport_array support, update GL3.txt docEdward O'Callaghan2015-12-112-2/+2
| | | | | Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* softpipe: implement some support for multiple viewportsEdward O'Callaghan2015-12-118-52/+112
| | | | | | | | Mostly related to making sure the rasterizer can correctly pick out the correct scissor box for the current viewport. Signed-off-by: Edward O'Callaghan <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* draw: don't assume fixed offset for data in struct vertex_infoRoland Scheidegger2015-12-111-5/+3
| | | | | | Otherwise, if struct vertex_info is changed, you're in for some surprises... Reviewed-by: Jose Fonseca <[email protected]>
* i965/gen9: Don't do fast clears when GL_FRAMEBUFFER_SRGB is enabledNeil Roberts2015-12-111-0/+11
| | | | | | | | | When GL_FRAMEBUFFER_SRGB is enabled any single-sampled renderbuffers are resolved in intel_update_state because the hardware can't cope with fast clears on SRGB buffers. In that case it's pointless to do a fast clear because it will just be immediately resolved. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/gen9: Allow fast clears for non-MSRT SRGB buffersNeil Roberts2015-12-111-1/+2
| | | | | | | | | | | | | | | SRGB buffers are not marked as losslessly compressible so previously they would not be used for fast clears. However in practice the hardware will never actually see that we are using SRGB buffers for fast clears if we use the linear equivalent format when clearing and make sure to resolve the buffer as a linear format before sampling from it. This is an important use case because by default the window system framebuffers are created as SRGB so without this fast clears won't be used there. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/gen9: Resolve SRGB color buffers when GL_FRAMEBUFFER_SRGB enabledNeil Roberts2015-12-111-0/+27
| | | | | | | | | | SKL can't cope with the CCS buffer for SRGB buffers. Normally the hardware won't see the SRGB formats because when GL_FRAMEBUFFER_SRGB is disabled these get mapped to their linear equivalents. In order to avoid relying on the CCS buffer when it is enabled this patch now makes it flush the renderbuffers. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/gen8+: Don't upload the MCS buffer for single-sampled texturesNeil Roberts2015-12-111-1/+5
| | | | | | | | | | | For single-sampled textures the MCS buffer is only used to implement fast clears. However the surface always needs to be resolved before being used as a texture anyway so the the MCS buffer doesn't actually achieve anything. This is important for Gen9 because in that case SRGB surfaces are not supported for fast clears and we don't want the hardware to see the MCS buffer in that case. Reviewed-by: Topi Pohjolainen <[email protected]>
* i965/meta-fast-clear: Disable GL_FRAMEBUFFER_SRGB during clearNeil Roberts2015-12-111-0/+16
| | | | | | | | | | | | | | | Adds MESA_META_FRAMEBUFFER_SRGB to the meta save state so that GL_FRAMEBUFFER_SRGB will be disabled when performing the fast clear. That way the render surface state will be programmed with the linear equivalent format during the clear. This is important for Gen9 because the SRGB formats are not marked as losslessly compressible so in theory they aren't support for fast clears. It shouldn't make any difference whether GL_FRAMEBUFFER_SRGB is enabled for the fast clear operation because the color is not actually written to the framebuffer so there is no chance for the hardware to apply the SRGB conversion on it anyway. Reviewed-by: Topi Pohjolainen <[email protected]>
* winsys/amdgpu: clear the buffer cache on mmap failure and try againMarek Olšák2015-12-111-0/+5
| | | | Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: clear the buffer cache on mmap failure and try againMarek Olšák2015-12-111-3/+10
| | | | Reviewed-by: Michel Dänzer <[email protected]>
* winsys/amdgpu: clear the buffer cache on allocation failure and try againMarek Olšák2015-12-111-2/+7
| | | | Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: clear the buffer cache on allocation failure and try againMarek Olšák2015-12-111-2/+7
| | | | Reviewed-by: Michel Dänzer <[email protected]>
* gallium/radeon: remove radeon_winsys_cs_handleMarek Olšák2015-12-1139-173/+130
| | | | | | | | "radeon_winsys_cs_handle *cs_buf" is now equivalent to "pb_buffer *buf". Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: use pb_cache instead of pb_cache_managerMarek Olšák2015-12-114-178/+75
| | | | | | | | This is a prerequisite for the removal of radeon_winsys_cs_handle. Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: use radeon_bomgr lessMarek Olšák2015-12-112-16/+8
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: rename radeon_bomgr_init_functionsMarek Olšák2015-12-113-3/+3
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: move variables from radeon_bomgr to radeon_drm_winsysMarek Olšák2015-12-113-126/+129
| | | | | | | | radeon_bomgr is going away. Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/radeon: remove redundant radeon_bomgr::vaMarek Olšák2015-12-111-7/+4
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/amdgpu: don't use the "rws" abbreviation for amdgpu_winsysMarek Olšák2015-12-114-20/+20
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* winsys/amdgpu: use pb_cache instead of pb_cache_managerMarek Olšák2015-12-114-173/+78
| | | | | | | | This is a prerequisite for the removal of radeon_winsys_cs_handle. Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* gallium/pb_bufmgr_cache: use the new pb_cache moduleMarek Olšák2015-12-111-198/+34
| | | | | | Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Acked-by: Michel Dänzer <[email protected]>
* gallium/pb_cache: add a copy of cache bufmgr independent of pb_managerMarek Olšák2015-12-113-0/+362
| | | | | | | | | | | | | | | This simplified (basically duplicated) version of pb_cache_manager will allow removing some ugly hacks from radeon and amdgpu winsyses and flatten simplify their design. The difference is that winsyses must manually add buffers to the cache in "destroy" functions and the cache doesn't know about the buffers before that. The integration is therefore trivial and the impact on the winsys design is negligible. Reviewed-by: Nicolai Hähnle <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]> Acked-by: Michel Dänzer <[email protected]>
* radeonsi: implement fast stencil clearMarek Olšák2015-12-114-23/+53
| | | | Reviewed-by: Alex Deucher <[email protected]>
* radeonsi: re-enable Hyper-Z for stencilMarek Olšák2015-12-111-9/+3
| | | | Reviewed-by: Alex Deucher <[email protected]>
* r600g: remove a Hyper-Z workaround that's likely not needed anymoreMarek Olšák2015-12-111-19/+7
| | | | | | FORCE_OFF == 0, no need to set that Reviewed-by: Alex Deucher <[email protected]>
* r600g: re-enable Hyper-Z for stencil on Evergreen & CaymanMarek Olšák2015-12-111-4/+1
| | | | Reviewed-by: Alex Deucher <[email protected]>
* gallium/radeon: fix Hyper-Z hangs by programming PA_SC_MODE_CNTL_1 correctlyMarek Olšák2015-12-113-5/+18
| | | | | | | | | | This is the recommended setting according to hw people and it makes Hyper-Z stable. Just the two magic states. This fixes Evergreen, Cayman, SI, CI, VI (using the Cayman code). Cc: 11.0 11.1 <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
* radeonsi: don't use the CP DMA workaround on Fiji and newerMarek Olšák2015-12-111-16/+20
| | | | Reviewed-by: Alex Deucher <[email protected]>
* radeonsi: apply the streamout workaround to Fiji as wellMarek Olšák2015-12-111-1/+3
| | | | | Cc: 11.0 11.1 <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
* radeonsi: also print hexadecimal values for register fields in the IB parserMarek Olšák2015-12-111-4/+7
| | | | | Reviewed-by: Michel Dänzer <[email protected] Reviewed-by: Edward O'Callaghan <[email protected]>
* radeonsi: implement RB+ for Stoney (v2)Marek Olšák2015-12-115-2/+170
| | | | | | v2: fix dual source blending Reviewed-by: Alex Deucher <[email protected]>
* radeonsi: don't call of u_prims_for_vertices for patches and rectanglesMarek Olšák2015-12-111-1/+13
| | | | | | | | | Both caused a crash due to a division by zero in that function. This is an alternative fix. Cc: 11.0 11.1 <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Edward O'Callaghan <[email protected]>
* radeonsi: use tgsi_shader_info::colors_writtenMarek Olšák2015-12-113-11/+1
| | | | Reviewed-by: Alex Deucher <[email protected]>
* r600g: write all MRTs only if there is exactly one output (fixes a hang)Marek Olšák2015-12-111-1/+2
| | | | | | | | This fixes a hang in piglit/arb_blend_func_extended-fbo-extended-blend-pattern_gles2 on REDWOOD. Cc: 11.0 11.1 <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
* tgsi/scan: add flag colors_writtenMarek Olšák2015-12-112-0/+4
| | | | | | | This is a prerequisite for the following r600g fix. Cc: 11.0 11.1 <[email protected]> Reviewed-by: Alex Deucher <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-27 14:48:51 +0000