| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | R600: Workaround for cayman loop bug | Vincent Lejeune | 2013-12-02 | 1 | -0/+10 |
| | | | | | llvm-svn: 196121 | ||||
| * | R600: Add IsExport bit to TableGen instruction definitions | Tom Stellard | 2013-08-16 | 1 | -9/+4 |
| | | | | | | Tested-by: Aaron Watry <awatry@gmail.com> llvm-svn: 188516 | ||||
| * | R600: Change the RAT instruction assembly names so they match the docs | Tom Stellard | 2013-08-16 | 1 | -2/+2 |
| | | | | | | Tested-by: Aaron Watry <awatry@gmail.com> llvm-svn: 188515 | ||||
| * | R600: Add 64-bit float load/store support | Tom Stellard | 2013-08-01 | 1 | -1/+3 |
| | | | | | | | | | | | | | | | | | | * Added R600_Reg64 class * Added T#Index#.XY registers definition * Added v2i32 register reads from parameter and global space * Added f32 and i32 elements extraction from v2f32 and v2i32 * Added v2i32 -> v2f32 conversions Tom Stellard: - Mark vec2 operations as expand. The addition of a vec2 register class made them all legal. Patch by: Dmitry Cherkassov Signed-off-by: Dmitry Cherkassov <dcherkassov@gmail.com> llvm-svn: 187582 | ||||
| * | R600: Remove predicated_break inst | Vincent Lejeune | 2013-07-31 | 1 | -9/+2 |
| | | | | | | | | | | | | | | We were using two instructions for similar purpose : break and predicated break. Only predicated_break was emitted and it was lowered at R600ControlFlowFinalizer to JUMP;CF_BREAK;POP. This commit simplify the situation by making AMDILCFGStructurizer emit IF_PREDICATE;BREAK;ENDIF; instead of predicated_break (which is now removed). There is no functionality change. llvm-svn: 187510 | ||||
| * | R600: Don't emit empty then clause and use alu_pop_after | Vincent Lejeune | 2013-07-19 | 1 | -9/+39 |
| | | | | | llvm-svn: 186725 | ||||
| * | Make some arrays 'static const' | Craig Topper | 2013-07-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 186307 | ||||
| * | R600: Do not predicated basic block with multiple alu clause | Vincent Lejeune | 2013-07-09 | 1 | -0/+2 |
| | | | | | | | | | | Test is not included as it is several 1000 lines long. To test this functionnality, a test case must generate at least 2 ALU clauses, where an ALU clause is ~110 instructions long. NOTE: This is a candidate for the stable branch. llvm-svn: 185943 | ||||
| * | R600: Use EXPORT_RAT_INST_STORE_DWORD for stores on Cayman | Tom Stellard | 2013-06-14 | 1 | -0/+1 |
| | | | | | | | | | | We were using RAT_INST_STORE_RAW, which seemed to work, but the docs say this instruction doesn't exist for Cayman, so it's probably safer to use a documented instruction instead. Reviewed-by: Vincent Lejeune<vljn at ovi.com> llvm-svn: 184015 | ||||
| * | R600: Anti dep better handled in tex clause | Vincent Lejeune | 2013-06-07 | 1 | -6/+4 |
| | | | | | llvm-svn: 183592 | ||||
| * | R600: Rework subtarget info and remove AMDILDevice classes | Tom Stellard | 2013-06-07 | 1 | -6/+8 |
| | | | | | | | | | This should simplify the subtarget definitions and make it easier to add new ones. Reviewed-by: Vincent Lejeune <vljn@ovi.com> llvm-svn: 183566 | ||||
| * | Don't cache the instruction and register info from the TargetMachine, because | Bill Wendling | 2013-06-07 | 1 | -7/+9 |
| | | | | | | | | | the internals of TargetMachine could change. No functionality change intended. llvm-svn: 183561 | ||||
| * | R600: CALL_FS consumes a stack size entry | Vincent Lejeune | 2013-06-03 | 1 | -0/+1 |
| | | | | | llvm-svn: 183108 | ||||
| * | R600: Fix R600ControlFlowFinalizer not considering VTX_READ 128 bit dst reg | Tom Stellard | 2013-05-23 | 1 | -2/+9 |
| | | | | | | | | | | Patch by: Vincent Lejeune https://bugs.freedesktop.org/show_bug.cgi?id=64877 NOTE: This is a candidate for the 3.3 branch. llvm-svn: 182600 | ||||
| * | Move passes from namespace llvm into anonymous namespaces. Sort includes ↵ | Benjamin Kramer | 2013-05-23 | 1 | -4/+5 |
| | | | | | | | while there. llvm-svn: 182594 | ||||
| * | Setting the default value (fixes CRT assertions about uninitialized variable ↵ | Aaron Ballman | 2013-05-23 | 1 | -3/+3 |
| | | | | | | | use when doing debug MSVC builds), and fixing coding style. llvm-svn: 182585 | ||||
| * | R600: Some factorization | Vincent Lejeune | 2013-05-17 | 1 | -9/+7 |
| | | | | | llvm-svn: 182123 | ||||
| * | R600: Factorize Fetch size limit inside AMDGPUSubTarget | Vincent Lejeune | 2013-05-17 | 1 | -5/+2 |
| | | | | | llvm-svn: 182122 | ||||
| * | R600: Signed literals are 64bits wide | Vincent Lejeune | 2013-05-02 | 1 | -3/+3 |
| | | | | | llvm-svn: 180960 | ||||
| * | R600: use native for alu | Vincent Lejeune | 2013-04-30 | 1 | -1/+109 |
| | | | | | llvm-svn: 180761 | ||||
| * | R600: Take inner dependency into tex/vtx clauses | Vincent Lejeune | 2013-04-30 | 1 | -0/+34 |
| | | | | | llvm-svn: 180757 | ||||
| * | R600: Turn TEX/VTX into native instructions | Vincent Lejeune | 2013-04-30 | 1 | -10/+31 |
| | | | | | llvm-svn: 180756 | ||||
| * | R600: Add FetchInst bit to instruction defs to denote vertex/tex instructions | Vincent Lejeune | 2013-04-30 | 1 | -36/+9 |
| | | | | | | | v2[Vincent Lejeune]: Split FetchInst into usesTextureCache/usesVertexCache llvm-svn: 180755 | ||||
| * | R600: Use correct CF_END instruction on Northern Island GPUs | Tom Stellard | 2013-04-29 | 1 | -1/+1 |
| | | | | | llvm-svn: 180735 | ||||
| * | R600: Use .AMDGPU.config section to emit stacksize | Vincent Lejeune | 2013-04-23 | 1 | -5/+21 |
| | | | | | llvm-svn: 180124 | ||||
| * | R600: Add CF_END | Vincent Lejeune | 2013-04-23 | 1 | -42/+49 |
| | | | | | llvm-svn: 180123 | ||||
| * | R600ControlFlowFinalizer.cpp: Fix a warning. [-Wunused-variable] | NAKAMURA Takumi | 2013-04-11 | 1 | -0/+1 |
| | | | | | llvm-svn: 179263 | ||||
| * | Whitespace. | NAKAMURA Takumi | 2013-04-11 | 1 | -2/+1 |
| | | | | | llvm-svn: 179262 | ||||
| * | R600: Add VTX_READ_* and RAT_WRITE_CACHELESS_* when computing cf addr | Vincent Lejeune | 2013-04-10 | 1 | -1/+10 |
| | | | | | llvm-svn: 179174 | ||||
| * | R600: Control Flow support for pre EG gen | Vincent Lejeune | 2013-04-08 | 1 | -12/+72 |
| | | | | | llvm-svn: 179020 | ||||
| * | R600: Fix wrong address when substituting ENDIF | Vincent Lejeune | 2013-04-04 | 1 | -1/+1 |
| | | | | | llvm-svn: 178762 | ||||
| * | R600: Take export into account when computing cf address | Vincent Lejeune | 2013-04-04 | 1 | -0/+4 |
| | | | | | llvm-svn: 178761 | ||||
| * | R600: Simplify data structure and add DEBUG to R600ControlFlowFinalizer | Vincent Lejeune | 2013-04-03 | 1 | -20/+21 |
| | | | | | llvm-svn: 178665 | ||||
| * | R600: Consider KILLGT as an ALU instruction | Vincent Lejeune | 2013-04-03 | 1 | -1/+0 |
| | | | | | | | | Mesa does not override llvm behavior wrt KILLGT anymore so llvm has to handle KILLGT on its own. llvm-svn: 178664 | ||||
| * | R600: Add support for native control flow | Vincent Lejeune | 2013-04-01 | 1 | -0/+264 |
| llvm-svn: 178505 | |||||

