aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* r600g/sb: improve alu packing on caymanVadim Girlin2013-07-172-15/+89
| | | | | | | | | | | | | | | | Scheduler/register allocator in r600-sb was developed and optimized on evergreen (VLIW-5) hardware, so currently it's not optimal for VLIW-4 chips. This patch should improve performance on cayman gpus due to better alu packing, but also it tends to increase register usage, so overall positive effect on performance has to be proven by real benchmarks yet. Some results with bfgminer kernel on cayman: source bytecode: 60 gprs, 3905 alu groups, sbcl before the patch: 45 gprs, 4088 alu groups, sbcl with this patch: 55 gprs, 3474 alu groups. Signed-off-by: Vadim Girlin <[email protected]>
* r600g/sb: fix handling of new multislot instructions on caymanVadim Girlin2013-07-173-5/+6
| | | | | | | Ex-scalar instructions that became multislot on cayman do replicate result to all channels - handle them similar to DOT4. Signed-off-by: Vadim Girlin <[email protected]>
* r600g/sb: fix debug dump code in schedulerVadim Girlin2013-07-171-4/+5
| | | | | | Update the stale debug code for other changes related to debug output. Signed-off-by: Vadim Girlin <[email protected]>
* r600g/sb: fix initial register allocationVadim Girlin2013-07-171-0/+1
| | | | | | | | | | Mark values that are members of the 'same register' constraint as preallocated in ra_init pass, this will prevent incorrect reallocation in scheduler in some cases. Should fix https://bugs.freedesktop.org/show_bug.cgi?id=66713 Signed-off-by: Vadim Girlin <[email protected]>
* r600g/sb: move chip & class name functions to sb_contextVadim Girlin2013-07-174-53/+55
| | | | Signed-off-by: Vadim Girlin <[email protected]>
* r600g/sb: fix handling of PS in source bytecode on caymanVadim Girlin2013-07-171-0/+5
| | | | | | | | | Actually PS doesn't make sense for cayman and isn't even mentioned in cayman docs, but llvm backend currently uses it in bytecode and, assuming that hw seems to be mostly ok with it, this will allow sb to parse such source bytecode correctly. Signed-off-by: Vadim Girlin <[email protected]>
* r600g/sb: Initialize ra_checker member variables.Vinson Lee2013-07-171-1/+1
| | | | | | Fixes "Uninitialized scalar field" defect reported by Coverity. Signed-off-by: Vinson Lee <[email protected]>
* gallium/util: use explicily sized types for {un, }pack_rgba_{s, u}intEmil Velikov2013-07-172-8/+8
| | | | | | | | Every function but the above four uses explicitly sized types for their src and dst arguments. Even fetch_rgba_{s,u}int follows the convention. Signed-off-by: Emil Velikov <[email protected]> Signed-off-by: Marek Olšák <[email protected]>
* llvmpipe: use MCJIT on ARM and AArch64Kyle McMartin2013-07-172-2/+9
| | | | | | | | MCJIT is the only supported LLVM JIT on AArch64 and ARM (the regular JIT has bit-rotted badly on ARM and doesn't exist on AArch64.) Signed-off-by: Kyle McMartin <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
* llvmpipe: support sRGB framebuffersRoland Scheidegger2013-07-164-18/+111
| | | | | | | | | | | | | | | Just use the new conversion functions to do the work. The way it's plugged in into the blend code is quite hacktastic but follows all the same hacks as used by packed float format already. Only support 4x8bit srgb formats (rgba/rgbx plus swizzle), 24bit formats never worked anyway in the blend code and are thus disabled, and I don't think anyone is interested in L8/L8A8. Would need even more hacks otherwise. Unless I'm missing something, this is the last feature except MSAA needed for OpenGL 3.0, and for OpenGL 3.1 as well I believe. v2: prettify a bit, use separate function for packing. Reviewed-by: Jose Fonseca <[email protected]>
* Revert "r300g: allow HiZ with a 16-bit zbuffer"Marek Olšák2013-07-151-0/+1
| | | | | | | | This reverts commit 631c631cbf5b7e84e42a7cfffa1c206d63143370. https://bugs.freedesktop.org/show_bug.cgi?id=66921 Cc: [email protected]
* r300g/swtcl: fix a lockup in MSAA resolveMarek Olšák2013-07-151-0/+7
| | | | Cc: [email protected]
* r300g/swtcl: fix geometry corruption by uploading indices to a bufferMarek Olšák2013-07-153-45/+31
| | | | | | | | | | | | | The splitting of a draw call into several draw commands was broken, because the split sometimes took place in the middle of a primitive. The splitting was supposed to be dealing with the case when there are more indices than the maximum size of a CS. This commit throws that code away and uses a real index buffer instead. https://bugs.freedesktop.org/show_bug.cgi?id=66558 Cc: [email protected]
* gallivm: (trivial) use constant instead of exp2f() functionRoland Scheidegger2013-07-141-2/+3
| | | | | | | Some lame compilers can't do exp2f() and as far as I can tell they can't do exp2() (with doubles) neither so instead of providing some workaround for that (wouldn't actually be too bad just replace with pow) and since it is used with a constant only just use the precalculated constant.
* ilo: skip 3DSTATE_INDEX_BUFFER when possibleChia-I Wu2013-07-144-59/+77
| | | | | | When only the offset to the index buffer is changed, we can skip the 3DSTATE_INDEX_BUFFER if we always use 0 for the offset, and add (offset / index_size) to Start Vertex Location in 3DPRIMITIVE.
* gallivm: handle srgb-to-linear and linear-to-srgb conversionsRoland Scheidegger2013-07-136-7/+332
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | srgb-to-linear is using 3rd degree polynomial for now which should be _just_ good enough. Reverse is using some rational polynomials and is quite accurate, though not hooked into llvmpipe's blend code yet and hence unused (untested). Using a table might also be an option (for srgb-to-linear especially). This does not enable any new features yet because EXT_texture_srgb was already supported via util_format fallbacks, but performance was lacking probably due to the external function call (the table used by the util_format_srgb code may not be all that much slower on its own). Some performance figures (taken from modified gloss, replaced both base and sphere texture to use GL_SRGB instead of GL_RGB, measured on 1Ghz Sandy Bridge, the numbers aren't terribly accurate): normal gloss, aos, 8-wide: 47 fps normal gloss, aos, 4-wide: 48 fps normal gloss, forced to soa, 8-wide: 48 fps normal gloss, forced to soa, 4-wide: 47 fps patched gloss, old code, soa, 8-wide: 21 fps patched gloss, old code, soa, 4-wide: 24 fps patched gloss, new code, soa, 8-wide: 41 fps patched gloss, new code, soa, 4-wide: 38 fps So there's a performance hit but it seems acceptable, certainly better than using the fallback. Note the new code only works for 4x8bit srgb formats, others (L8/L8A8) will continue to use the old util_format fallback, because I can't be bothered to write code for formats noone uses anyway (as decoding is done as part of lp_build_unpack_rgba_soa which can only handle block type width of 32). Compressed srgb formats should get their own path though eventually (it is going to be expensive in any case, first decompress, then convert). No piglit regressions. v2: use lp_build_polynomial instead of ad-hoc polynomial construction, also since keeping both linear to srgb functions for now make sure both are compiled (since they share quite some code just integrate into the same function). v3: formatting fixes and bugfix in the complicated (disabled) linear-to-srgb path. Reviewed-by: Jose Fonseca <[email protected]>
* gallivm: better support for fast rsqrtRoland Scheidegger2013-07-132-16/+63
| | | | | | | | | | | | | | | We had to disable fast rsqrt before because it wasn't precise enough etc. However in situations when we know we're not going to need more precision we can still use a fast rsqrt (which can be several times faster than the quite expensive sqrt). Hence introduce a new helper which does exactly that - it is probably not useful calling it in some situations if there's no fast rsqrt available so make it queryable if it's available too. v2: use fast_rsqrt consistently instead of rsqrt_fast, fix indentation, let rsqrt use fast_rsqrt. Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Jose Fonseca <[email protected]>
* r600g/sb: Initialize ra_constraint::cost.Vinson Lee2013-07-131-1/+1
| | | | | | Fixes "Uninitialized scalar field" reported by Coverity. Signed-off-by: Vinson Lee <[email protected]>
* winsys/radeon: allow a NULL cs pointer in radeon_bo_map to fix a segfaultMarek Olšák2013-07-131-9/+11
| | | | | The original idea was that cs=NULL should be allowed here, but we never used NULL until 862f69fbe1e54e0e9a3c439450a14f. This fixes a segfault in CoreBreach.
* ilo: move a santiy check into its assert()Chia-I Wu2013-07-131-5/+2
| | | | | | The compiler does not know that ilo_3d_pipeline_estimate_size() is pure and can be eliminated in a release build in gen6_pipeline_end(). Move the call into the assert().
* ilo: mark some states dirty when they are really changedChia-I Wu2013-07-131-0/+16
| | | | | The checks may seem redundant because cso_context handles them, but util_blitter does not have access to cso_context.
* ilo: clean up ilo_blitter_pipe_begin()Chia-I Wu2013-07-133-27/+39
| | | | | Document why certain states need to be saved, and fix a bug when blitting with scissor enabled.
* r600g: don't use the CB/DB CP COHER logic on r6xxAlex Deucher2013-07-121-2/+10
| | | | | | | | | There are hw bugs. Flush and inv event is sufficient. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=66837 Signed-off-by: Alex Deucher <[email protected]>
* nv30: fix KILL_IF breakageBrian Paul2013-07-121-1/+1
| | | | Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66858
* gallium: fixup definitions of the rsq and sqrtZack Rusin2013-07-113-16/+11
| | | | | | | | | | | | GLSL spec says that rsq is undefined for src<=0, but the D3D10 spec says it needs to be a NaN, so lets stop taking an absolute value of the source which completely breaks that behavior. For the gl program we can simply insert an extra abs instrunction which produces the desired behavior there. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]> Reviewed-by: Brian Paul <[email protected]>
* util/u_format: Comment out half float denormal test case.José Fonseca2013-07-121-0/+5
| | | | So that lp_test_format doesn't fail until we decide what should be done.
* gallivm: Eliminate redundant lp_build_select calls.José Fonseca2013-07-121-12/+2
| | | | | | | lp_build_cmp already returns 0 / ~0, so the lp_build_select call is unnecessary. Reviewed-by: Roland Scheidegger <[email protected]>
* tgsi: rename the TGSI fragment kill opcodesBrian Paul2013-07-1230-105/+103
| | | | | | | | | | | | | | | | | | | | | TGSI_OPCODE_KIL and KILP had confusing names. The former was conditional kill (if any src component < 0). The later was unconditional kill. At one time KILP was supposed to work with NV-style condition codes/predicates but we never had that in TGSI. This patch renames both opcodes: TGSI_OPCODE_KIL -> KILL_IF (kill if src.xyzw < 0) TGSI_OPCODE_KILP -> KILL (unconditional kill) Note: I didn't just transpose the opcode names to help ensure that I didn't miss updating any code anywhere. I believe I've updated all the relevant code and comments but I'm not 100% sure that some drivers had this right in the first place. For example, the radeon driver might have llvm.AMDGPU.kill and llvm.AMDGPU.kilp mixed up. Driver authors should review their code. Reviewed-by: Jose Fonseca <[email protected]>
* tgsi: fix-up KILP commentsBrian Paul2013-07-123-10/+8
| | | | | | | | KILP is really unconditional fragment kill. We've had KIL and KILP transposed forever. I'll fix that next. Reviewed-by: Jose Fonseca <[email protected]>
* tgsi: exec TGSI_OPCODE_SQRT as a scalar instruction, not vectorBrian Paul2013-07-121-1/+1
| | | | | | To align with the docs and the state tracker. Reviewed-by: Jose Fonseca <[email protected]>
* tgsi: use X component of the second operand in exec_scalar_binary()Brian Paul2013-07-121-1/+1
| | | | | | | | | The code happened to work in the past since the (scalar) src args effectively always have a swizzle of .xxxx, .yyyy, .zzzz, or .wwww so whether you grab the X or Y component doesn't really matter. Just fixing the code to make it look right. Reviewed-by: Roland Scheidegger <[email protected]>
* os: add os_get_process_name() functionBrian Paul2013-07-123-0/+133
| | | | | v2: explicitly test for BSD/APPLE, #warning for unexpected environments.
* softpipe: silence some MSVC warningsBrian Paul2013-07-122-14/+14
|
* hud: silence some MSVC warningsBrian Paul2013-07-121-8/+8
|
* util: add casts to silence MSVC warnings in u_blit.cBrian Paul2013-07-121-14/+14
|
* tgsi: s/unsigned/int/ to silence MSVC warningBrian Paul2013-07-121-1/+1
|
* radeon/uvd: fall back to shader based decoding for MPEG2 on UVD 2.x v2Christian König2013-07-122-5/+19
| | | | | | | | | | | UVD 2.x doesn't support hardware decoding of MPEG2, just use shader based decoding for those chipsets. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=66450 v2: fix interlacing as well Signed-off-by: Christian König <[email protected]>
* r600g: x/y coordinates must be divided by block dim in dma blitChristoph Bumiller2013-07-112-4/+16
| | | | | | | Note: this is a candidate for the 9.1 branch. Reviewed-by: Marek Olšák <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
* r600g/sb: Fix Android build v2Chih-Wei Huang2013-07-124-7/+8
| | | | | | | Add the sb CXX files to the Android Makefile and also stop using some c++11 features. v2 (Vadim Girlin): use &bc[0] instead of bc.begin()
* r600g/sb: improve math optimizations v2Vadim Girlin2013-07-1111-47/+435
| | | | | | | | | | | | | | | | This patch adds support for some math optimizations that are generally considered unsafe, that's why they are currently disabled for compute shaders. GL requirements are less strict, so they are enabled for for GL shaders by default. In case of any issues with applications that rely on higher precision than guaranteed by GL, 'sbsafemath' option in R600_DEBUG allows to disable them. v2 - always set proper src vector size for transformed instructions - check for clamp modifier in the expr_handler::fold_assoc Signed-off-by: Vadim Girlin <[email protected]>
* st/xvmc/tests: avoid non portable error.h functionsJonathan Gray2013-07-111-5/+8
| | | | | Signed-off-by: Jonathan Gray <[email protected]> Reviewed-by: Christian König <[email protected]>
* winsys/intel: build with VISIBILITY_CFLAGSChia-I Wu2013-07-111-1/+2
| | | | There is no public symbol in this winsys.
* ilo: reduce PIPE_CAP_MAX_TEXTURE_CUBE_LEVELS to 12Chia-I Wu2013-07-111-2/+3
| | | | So that there are at most (2^22 * 6) texels, lower than the 2^26 limit.
* ilo: correctly initialize undefined registers in fsChia-I Wu2013-07-111-5/+15
| | | | | Initialize all 4 channels of undefined registers (that is, TEMPs that are used before being assigned) in FS.
* radeonsi: Handle TGSI_OPCODE_DDX/Y using local memoryMichel Dänzer2013-07-104-2/+103
| | | | | | 16 more little piglits. Reviewed-by: Tom Stellard <[email protected]>
* radeonsi: Handle TGSI_OPCODE_TXDMichel Dänzer2013-07-101-2/+25
| | | | | | One more little piglit. Reviewed-by: Tom Stellard <[email protected]>
* util/u_math: Use xmmintrin.h whenever possible.José Fonseca2013-07-101-9/+17
| | | | | | | | | | | | | It seems __builtin_ia32_ldmxcsr is only available on gcc and only when -msse is used. xmmintrin.h/pmmintrin.h provide portable intrinsics, but these too are only available with gcc when -msse/-msse3 are set. scons build always sets -msse on x86 builds, but autotools doesn't seem to. We could try to get this working on gcc x86 without -msse by emitting assembly, but I believe that in this day and age we really should be building Mesa with -msse and -msse2.
* ilo: honor surface padding requirementsChia-I Wu2013-07-101-0/+53
| | | | The PRM specifies several padding requirements that we failed to honor.
* util: treat denorm'ed floats like zeroZack Rusin2013-07-096-0/+83
| | | | | | | | | | | | | The D3D10 spec is very explicit about treatment of denorm floats and the behavior is exactly the same for them as it would be for -0 or +0. This makes our shading code match that behavior, since OpenGL doesn't care and on a few cpu's it's faster (worst case the same). Float16 conversions will likely break but we'll fix them in a follow up commit. Signed-off-by: Zack Rusin <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* r600g: improve the mechanism for recognizing an empty CSMarek Olšák2013-07-083-3/+8
| | | | Reviewed-by: Alex Deucher <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-14 18:56:45 +0000