| Commit message (Expand) | Author | Age | Files | Lines |
... | |
* | Avoid using lossy load / stores for memcpy / memset expansion. e.g. | Evan Cheng | 2012-12-12 | 1 | -2/+2 |
* | Add a triple to this test. | Chad Rosier | 2012-12-11 | 1 | -1/+1 |
* | Fix a miscompile in the DAG combiner. Previously, we would incorrectly | Chandler Carruth | 2012-12-11 | 1 | -2/+23 |
* | move X86-specific test | Paul Redmond | 2012-12-11 | 1 | -0/+8 |
* | Fall back to the selection dag isel to select tail calls. | Chad Rosier | 2012-12-11 | 1 | -3/+2 |
* | Some enhancements for memcpy / memset inline expansion. | Evan Cheng | 2012-12-10 | 2 | -9/+10 |
* | Teach DAG combine to handle vector add/sub with vectors of all 0s. | Craig Topper | 2012-12-10 | 2 | -5/+5 |
* | Teach DAG combine to handle vector logical operations with vectors of all 1s ... | Craig Topper | 2012-12-08 | 3 | -23/+21 |
* | When we use the BLEND instruction that uses the MSB as a mask, we can remove | Nadav Rotem | 2012-12-07 | 2 | -2/+2 |
* | X86: Prefer using VPSHUFD over VPERMIL because it has better throughput. | Nadav Rotem | 2012-12-07 | 3 | -5/+5 |
* | Fix a bug in the code that merges consecutive stores. Previously we did not | Nadav Rotem | 2012-12-06 | 1 | -0/+23 |
* | Remove intrinsic specific instructions for (V)MOVQUmr with patterns pointing ... | Craig Topper | 2012-12-06 | 1 | -1/+4 |
* | RegisterPressureTracker: fix findUseBetween to handle DebugValue | Andrew Trick | 2012-12-05 | 1 | -0/+49 |
* | RegisterPresssureTracker: Track live physical register by unit. | Andrew Trick | 2012-12-05 | 1 | -0/+30 |
* | Simplified BLEND pattern matching for shuffles. | Elena Demikhovsky | 2012-12-05 | 2 | -6/+53 |
* | Add x86 isel lowering logic to form bit test with inverted condition. e.g. | Evan Cheng | 2012-12-05 | 1 | -3/+97 |
* | Use the 'count' attribute to calculate the upper bound of an array. | Bill Wendling | 2012-12-04 | 6 | -6/+6 |
* | Add a 'count' field to the DWARF subrange. | Bill Wendling | 2012-12-04 | 5 | -5/+5 |
* | Allow merging multiple store sequences on the same chain. | Nadav Rotem | 2012-12-02 | 1 | -0/+31 |
* | Fix an invalid regex in the test | Eli Bendersky | 2012-12-02 | 1 | -1/+1 |
* | misched: Fix RegisterPressureTracker handling of DebugVals. | Andrew Trick | 2012-12-01 | 1 | -0/+43 |
* | misched: Fix the DAG builder to handle an undef operand at ExitSU. | Andrew Trick | 2012-12-01 | 1 | -0/+26 |
* | misched: Fix LiveInterval update to better handle DebugVal. | Andrew Trick | 2012-12-01 | 1 | -0/+50 |
* | misched: fix RegionBegin when DebugValues get shuffled to the top. | Andrew Trick | 2012-12-01 | 1 | -0/+85 |
* | When combining consecutive stores allow loads in between the stores, if the l... | Nadav Rotem | 2012-11-29 | 1 | -0/+52 |
* | misched: Analysis that partitions the DAG into subtrees. | Andrew Trick | 2012-11-28 | 1 | -0/+68 |
* | misched: better alias analysis. | Andrew Trick | 2012-11-28 | 1 | -0/+127 |
* | X86: do not fold load instructions such as [V]MOVS[S|D] to other instructions | Manman Ren | 2012-11-27 | 1 | -0/+39 |
* | Revert accidental commit. | Craig Topper | 2012-11-27 | 1 | -0/+42 |
* | Make PrintReg constructor explicit to prevent weird implicit conversions from... | Craig Topper | 2012-11-27 | 1 | -42/+0 |
* | Add test cases for r168417. | Craig Topper | 2012-11-27 | 1 | -0/+20 |
* | llvm/test/CodeGen/X86/2012-07-15-broadcastfold.ll: Loosen expression correspo... | NAKAMURA Takumi | 2012-11-27 | 1 | -1/+1 |
* | Remove the X86 Maximal Stack Alignment Check pass as it is no longer necessary. | Chad Rosier | 2012-11-26 | 2 | -25/+19 |
* | Normalize splat 256bit vectors with 8 elements. | Jakub Staszak | 2012-11-26 | 1 | -7/+7 |
* | Intel OCL built-ins calling conventions now support MacOS 32-bit. | Elena Demikhovsky | 2012-11-20 | 1 | -36/+58 |
* | Handle mixed normal and early-clobber defs on inline asm. | Jakob Stoklund Olesen | 2012-11-19 | 1 | -0/+7 |
* | llvm/test/CodeGen/X86/hipe-cc*.ll: Add explicit -mcpu, or they don't expect t... | NAKAMURA Takumi | 2012-11-16 | 2 | -2/+2 |
* | Add the Erlang/HiPE calling convention, patch by Yiannis Tsiouris. | Duncan Sands | 2012-11-16 | 2 | -0/+164 |
* | Use roundps/pd for llvm.ceil, llvm.trunc, llvm.rint, and llvm.nearbyint of ve... | Craig Topper | 2012-11-16 | 1 | -0/+144 |
* | Make sure to not get AVX code on an AVX-capable host. Revealed in r167967. | Jakub Staszak | 2012-11-14 | 4 | -8/+8 |
* | llvm/test/CodeGen/X86/memset.ll: FileCheck-ize, and add another case on +avx. | NAKAMURA Takumi | 2012-11-14 | 1 | -2/+23 |
* | Force CPU in test so we don't accidentally get AVX code on an AVX-capable host. | Benjamin Kramer | 2012-11-14 | 1 | -2/+2 |
* | X86: Enable SSE memory intrinsics even when stack alignment is less than 16 b... | Benjamin Kramer | 2012-11-14 | 4 | -22/+79 |
* | Handle DAG CSE adding new uses during ReplaceAllUsesWith. Fixes PR14333. | Rafael Espindola | 2012-11-14 | 1 | -0/+12 |
* | Revert "Use the 'count' attribute instead of the 'upper_bound' attribute." | Eric Christopher | 2012-11-13 | 1 | -2/+2 |
* | X86: when constructing VZEXT_LOAD from other loads, makes sure its output | Manman Ren | 2012-11-13 | 1 | -0/+51 |
* | Use the 'count' attribute instead of the 'upper_bound' attribute. | Bill Wendling | 2012-11-13 | 1 | -2/+2 |
* | Fix test case added in patch fixing PR14314 | Michael Liao | 2012-11-12 | 1 | -4/+4 |
* | Fix PR14314 | Michael Liao | 2012-11-12 | 2 | -4/+17 |
* | Cleanup pcmp(e/i)str(m/i) instruction definitions and load folding support. | Craig Topper | 2012-11-10 | 1 | -5/+47 |