Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | radeon/llvm: Use a custom inserter to lower STORE_OUTPUT | Tom Stellard | 2012-05-08 | 4 | -34/+23 |
| | |||||
* | radeon/llvm: Remove AMDGPULowerShaderInstructions class | Tom Stellard | 2012-05-08 | 6 | -86/+4 |
| | | | | It is no longer used. | ||||
* | radeon/llvm: Use a custom inserter to lower LOAD_INPUT | Tom Stellard | 2012-05-08 | 4 | -39/+15 |
| | |||||
* | radeon/llvm: Remove the ReorderPreloadInstructions pass | Tom Stellard | 2012-05-08 | 9 | -100/+4 |
| | |||||
* | radeon/llvm: Remove old comment from AMDIL.h | Tom Stellard | 2012-05-08 | 1 | -5/+0 |
| | |||||
* | radeon/llvm: add suport for cube textures | Vadim Girlin | 2012-05-08 | 2 | -23/+91 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add support for CUBE ALU instruction | Vadim Girlin | 2012-05-08 | 5 | -21/+63 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add support for some ALU instructions | Vadim Girlin | 2012-05-08 | 4 | -13/+293 |
| | | | | | | | | Add support for IABS, NOT, AND, XOR, OR, UADD, UDIV, IDIV, MOD, UMOD, INEG, I2F, U2F, F2U, F2I, USEQ, USGE, USLT, USNE, ISGE, ISLT, ROUND, MIN, MAX, IMIN, IMAX, UMIN, UMAX Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add missing cases for BREAK/CONTINUE | Vadim Girlin | 2012-05-08 | 2 | -0/+3 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add support for AHSR/LSHR/LSHL instructions | Vadim Girlin | 2012-05-08 | 4 | -0/+53 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add support for TXQ/TXF/DDX/DDY instructions | Vadim Girlin | 2012-05-08 | 5 | -4/+43 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add support for VertexID, InstanceID | Vadim Girlin | 2012-05-08 | 3 | -0/+50 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: fix live-in handling for inputs | Vadim Girlin | 2012-05-08 | 2 | -2/+3 |
| | | | | | | Set the input registers as live-in for entry basic block. Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: add support for v4i32 | Vadim Girlin | 2012-05-08 | 4 | -5/+20 |
| | | | | Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: fix ABS_i32 instruction lowering | Vadim Girlin | 2012-05-08 | 1 | -2/+2 |
| | | | | | | Swap source operands. Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: use integer comparison for IF | Vadim Girlin | 2012-05-08 | 1 | -2/+4 |
| | | | | | | | Replacing "float equal to 1.0f" with "int not equal to 0". This should help for further optimization of boolean computations. Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | radeon/llvm: use bitcasts for integers | Vadim Girlin | 2012-05-08 | 3 | -5/+73 |
| | | | | | | | | | We're using float as default type, so basically for every instruction that wants other types for dst/src operands we need to perform the bitcast to/from default float. Currently bitcast produces no-op MOV instruction, will be eliminated later. Signed-off-by: Vadim Girlin <[email protected]> | ||||
* | r600g: Fix out of tree builds that use the LLVM backend | Tom Stellard | 2012-05-07 | 1 | -1/+1 |
| | | | | https://bugs.freedesktop.org/show_bug.cgi?id=49567 | ||||
* | radeon/llvm: Remove references to DebugFlag and isCurrentDebugType() | Tom Stellard | 2012-05-07 | 4 | -22/+3 |
| | | | | | | | These weren't being used at all and they were causing build failures when LLVM was built with NDEBUG defined and mesa was not. https://bugs.freedesktop.org/show_bug.cgi?id=49110 | ||||
* | nv50: handle VP without inputs | Marcin Slusarz | 2012-05-07 | 1 | -0/+11 |
| | |||||
* | nvc0/ir: allow abs,neg source modifiers with ceil,floor,trunc | Christoph Bumiller | 2012-05-06 | 1 | -0/+3 |
| | |||||
* | nv50/ir/opt: don't lose saturation in tryCollapseChainedMULs | Christoph Bumiller | 2012-05-06 | 1 | -2/+3 |
| | |||||
* | nvc0/ir: fix lowering of textureGrad | Christoph Bumiller | 2012-05-06 | 3 | -12/+13 |
| | |||||
* | nouveau: fix nouveau_scratch_runout_release bo count underflow | Christoph Bumiller | 2012-05-06 | 1 | -1/+3 |
| | |||||
* | nv50: fix typo causing NULL-deref in nv50_resource_resolve | Christoph Bumiller | 2012-05-04 | 1 | -1/+1 |
| | | | | Introduced in b328949a37fee7b0f68ed3e068ffc4426c083042. | ||||
* | nv50/ir: move expansion of IMUL to later stage and handle memory operands | Christoph Bumiller | 2012-05-04 | 4 | -17/+51 |
| | |||||
* | nv50: implement stream output | Christoph Bumiller | 2012-05-04 | 12 | -33/+468 |
| | |||||
* | nv50: enable array textures | Christoph Bumiller | 2012-05-04 | 2 | -3/+4 |
| | |||||
* | svga: specify 4-byte aligned vertex elements | Brian Paul | 2012-05-04 | 1 | -1/+2 |
| | | | | | | | We haven't found a case where this is needed, but it would be prudent for some hosts, per Jose. Reviewed-by: José Fonseca <[email protected]> | ||||
* | r600g/llvm: Mask write of pred_inst in llvm_if() | Tom Stellard | 2012-05-03 | 1 | -0/+1 |
| | |||||
* | r600g/llvm: Lower ULT A, B, C to SETGT_UINT A, C, B | Tom Stellard | 2012-05-03 | 1 | -0/+7 |
| | |||||
* | r600g: Print integer values of literal constants in shader dumps | Tom Stellard | 2012-05-03 | 1 | -1/+2 |
| | |||||
* | r600g: Add support for reading BREAK_LOGICALZ_i32 from bytestream | Tom Stellard | 2012-05-03 | 2 | -0/+5 |
| | |||||
* | r600g/llvm: Don't duplicate R600 intrinsics installed by LLVM | Tom Stellard | 2012-05-03 | 4 | -0/+26 |
| | | | | | | | At this point, in order for OpenCL to work correctly with r600g, OpenCL specific intrinsics need to be defined in the LLVM tree. So, we need to check for these intrinsics in the LLVM include directory to make sure not to re-define them. | ||||
* | r600g: Fix the evergreen offset/end register definitions | Tom Stellard | 2012-05-02 | 1 | -9/+5 |
| | |||||
* | radeon/llvm: Fix MachineInstr dump | Tom Stellard | 2012-05-02 | 2 | -8/+9 |
| | |||||
* | r600g: fixed the bug with VTX fetches in TEX clauses for evergreen | Adam Rak | 2012-05-02 | 1 | -7/+6 |
| | | | | Signed-off-by: Tom Stellard <[email protected]> | ||||
* | r600g: Add support for reading vertex fetches from bytestream | Tom Stellard | 2012-05-02 | 1 | -0/+37 |
| | | | | Signed-off-by: Tom Stellard <[email protected]> | ||||
* | r600g: Add support for reading native instructions from the LLVM bytestream | Tom Stellard | 2012-05-02 | 1 | -0/+10 |
| | | | | Signed-off-by: Tom Stellard <[email protected]> | ||||
* | r600g: Add FC_NATIVE instruction | Tom Stellard | 2012-05-02 | 3 | -0/+20 |
| | | | | | | | This is a pseudo instruction that enables the LLVM backend to encode instructions and pass it through r600_bytecode_build() Signed-off-by: Tom Stellard <[email protected]> | ||||
* | r600g: bypass alpha for integer types (v2) | Dave Airlie | 2012-05-02 | 4 | -2/+18 |
| | | | | | | | | | | | | This moves the alpha test control to derived state and disables alpha testing for integer fbs. fbo-blending test in piglit gets further when we do this (not a pass but less fail). v2: drop the fb_sx_alpha_test_control Signed-off-by: Dave Airlie <[email protected]> | ||||
* | gallivm: Added lp_build_const_mask_aos_swizzled | James Benton | 2012-05-02 | 1 | -9/+1 |
| | | | | | | | | | | Allows the creation of const aos masks which have the mask swizzled to match the correct format. Updated existing mask creation code to use the swizzled version where necessary (tgsi register masks and llvmpipe aos blending). Signed-off-by: José Fonseca <[email protected]> | ||||
* | llvmpipe: add masking support to aos blend | James Benton | 2012-05-02 | 3 | -6/+35 |
| | | | | Signed-off-by: José Fonseca <[email protected]> | ||||
* | llvmpipe: Added support for color masks in AoS blending. | James Benton | 2012-05-02 | 4 | -33/+58 |
| | | | | Signed-off-by: José Fonseca <[email protected]> | ||||
* | radeon/llvm: Fix build for updated LLVM 3.1 release branch | Tom Stellard | 2012-05-01 | 2 | -18/+18 |
| | |||||
* | softpipe: use any_swizzle() helper in sp_tex_sample.c | Brian Paul | 2012-05-01 | 1 | -8/+15 |
| | | | | Reviewed-by: Jose Fonseca <[email protected]> | ||||
* | softpipe: whitespace, comment clean-ups in sp_tex_sample.c | Brian Paul | 2012-05-01 | 1 | -26/+32 |
| | | | | Reviewed-by: Jose Fonseca <[email protected]> | ||||
* | softpipe: implement coord clamping for texel fetches (TXF) | Brian Paul | 2012-05-01 | 1 | -14/+31 |
| | | | | | | | | | | The GL spec says out of bounds fetches produce undefined results. Use clamping to avoid failed assertions or crashes. Fixes failed assertion in https://bugs.freedesktop.org/show_bug.cgi?id=49125 but the test still fails. Reviewed-by: Jose Fonseca <[email protected]> | ||||
* | radeon/llvm: Add subtarget feature: DumpCode | Tom Stellard | 2012-05-01 | 5 | -6/+9 |
| | | | | | | With this feature enabled, the LLVM backend will dump the MachineIntrs prior to emitting code. The mesa env variable R600_DUMP_SHADERS will enable this feature in the backend. | ||||
* | r600g/llvm: Remove unnecessary dynamic casts | Dragomir Ivanov | 2012-04-30 | 1 | -5/+5 |
| | | | | | | | When the result of dynamic_cast is not checked, it can be replaced with static_cast Signed-off-by: Tom Stellard <[email protected]> |