| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | PTX: add flag to disable mad/fma selection | Justin Holewinski | 2011-05-18 | 4 | -4/+19 | |
| | | | | | | | Patch by Dan Bailey llvm-svn: 131537 | |||||
| * | In r131488 I misunderstood how VREV works. It splits the vector in half and ↵ | Tanya Lattner | 2011-05-18 | 2 | -1343/+1350 | |
| | | | | | | | | | splits each half. Therefore, the real problem was that we were using a VREV64 for a 4xi16, when we should have been using a VREV32. Updated test case and reverted change to the PerfectShuffle Table. llvm-svn: 131529 | |||||
| * | Fix typo. | Cameron Zwarich | 2011-05-18 | 1 | -4/+4 | |
| | | | | | llvm-svn: 131519 | |||||
| * | Fix more of PR8825 by correctly using rGPR registers when lowering atomic | Cameron Zwarich | 2011-05-18 | 1 | -2/+11 | |
| | | | | | | | compare-and-swap intrinsics. llvm-svn: 131518 | |||||
| * | Actually, the address operand of the Thumb2 LDREX / STREX instructions *can* | Cameron Zwarich | 2011-05-17 | 1 | -1/+1 | |
| | | | | | | | take r13, so we can just make it a GPR. This fixes PR8825. llvm-svn: 131507 | |||||
| * | Correct a minor problem with the Thumb2 LDREX and STREX instruction ↵ | Cameron Zwarich | 2011-05-17 | 1 | -1/+1 | |
| | | | | | | | | | encodings. They were marked as taking a tGPR when in reality they take an rGPR. llvm-svn: 131506 | |||||
| * | vrev is incorrectly defined in the perfect shuffle table. The ordering is ↵ | Tanya Lattner | 2011-05-17 | 1 | -6565/+6566 | |
| | | | | | | | backwards (should be 0x3210 versus 0x1032) which exposed a bug when doing a shuffle on a 4xi16. I've attached a test case. llvm-svn: 131488 | |||||
| * | Enable autodetect of popcnt | Mon P Wang | 2011-05-17 | 1 | -0/+1 | |
| | | | | | llvm-svn: 131476 | |||||
| * | Add x86 fast-isel for calls returning first-class aggregates. rdar://9435872. | Eli Friedman | 2011-05-17 | 1 | -46/+51 | |
| | | | | | | | This is r131438 with a couple small fixes. llvm-svn: 131474 | |||||
| * | Clean up the mess created by r131467+r131469. | Eli Friedman | 2011-05-17 | 2 | -16/+0 | |
| | | | | | llvm-svn: 131471 | |||||
| * | Revert 131467 due to buildbot complaint. | Stuart Hastings | 2011-05-17 | 2 | -74/+28 | |
| | | | | | llvm-svn: 131469 | |||||
| * | Fix an obscure issue in X86_64 parameter passing: if a tiny byval is | Stuart Hastings | 2011-05-17 | 2 | -13/+75 | |
| | | | | | | | | passed as the fifth parameter, insure it's passed correctly (in R9). rdar://problem/6920088 llvm-svn: 131467 | |||||
| * | Fix a bug in PerformEXTRACT_VECTOR_ELTCombine. The code created an ADD SDNode | Nadav Rotem | 2011-05-17 | 1 | -2/+2 | |
| | | | | | | | | with two different types, in cases where the index and the ptr had different types. llvm-svn: 131461 | |||||
| * | Update comment. | Eric Christopher | 2011-05-17 | 1 | -4/+4 | |
| | | | | | llvm-svn: 131459 | |||||
| * | Support XOR and AND optimization with no return value. | Eric Christopher | 2011-05-17 | 2 | -5/+51 | |
| | | | | | | | Finishes off rdar://8470697 llvm-svn: 131458 | |||||
| * | Couple less magic numbers. | Eric Christopher | 2011-05-17 | 1 | -3/+5 | |
| | | | | | llvm-svn: 131457 | |||||
| * | Make this code a little less magic number laden. | Eric Christopher | 2011-05-17 | 1 | -12/+30 | |
| | | | | | llvm-svn: 131456 | |||||
| * | add a note | Chris Lattner | 2011-05-17 | 1 | -0/+26 | |
| | | | | | llvm-svn: 131455 | |||||
| * | Back out r131444 and r131438; they're breaking nightly tests. I'll look into | Eli Friedman | 2011-05-17 | 1 | -43/+48 | |
| | | | | | | | it more tomorrow. llvm-svn: 131451 | |||||
| * | Add x86 fast-isel for calls returning first-class aggregates. rdar://9435872. | Eli Friedman | 2011-05-17 | 1 | -48/+43 | |
| | | | | | llvm-svn: 131438 | |||||
| * | Kill some dead code. | Jim Grosbach | 2011-05-16 | 1 | -3/+1 | |
| | | | | | llvm-svn: 131431 | |||||
| * | Remove dead code. Fix associated test to use FileCheck. | Eli Friedman | 2011-05-16 | 1 | -14/+0 | |
| | | | | | llvm-svn: 131424 | |||||
| * | Make fast-isel work correctly s/uadd.with.overflow intrinsics. | Eli Friedman | 2011-05-16 | 1 | -76/+7 | |
| | | | | | llvm-svn: 131420 | |||||
| * | sets bit 0 of the function address of thumb function in .symtab | Rafael Espindola | 2011-05-16 | 2 | -6/+21 | |
| | | | | | | | | | | | ("T is 1 if the target symbol S has type STT_FUNC and the symbol addresses a Thumb instruction ;it is 0 otherwise." from "ELF for the ARM Architecture" 4.7.1.2) Patch by Koan-Sin Tan! llvm-svn: 131406 | |||||
| * | Fix a FIXME by moving the fast-isel implementation of the objectsize ↵ | Eli Friedman | 2011-05-14 | 1 | -23/+0 | |
| | | | | | | | intrinsic from the x86 code to the generic code. llvm-svn: 131332 | |||||
| * | Don't produce a vmovntdq if we don't have AVX support. | Rafael Espindola | 2011-05-14 | 1 | -1/+1 | |
| | | | | | llvm-svn: 131330 | |||||
| * | Zap useless code; this hasn't done anything useful since fast-isel switched ↵ | Eli Friedman | 2011-05-14 | 1 | -29/+0 | |
| | | | | | | | to being bottom-up (a very long time ago). llvm-svn: 131329 | |||||
| * | Fix a source of non determinism in FindUsedTypes, use a SetVector instead of a | Julien Lerouge | 2011-05-13 | 1 | -5/+4 | |
| | | | | | | | | | set. rdar://9423996 llvm-svn: 131283 | |||||
| * | Fix setting of isCommutable flag. | Akira Hatanaka | 2011-05-12 | 2 | -16/+25 | |
| | | | | | llvm-svn: 131233 | |||||
| * | Turn this into a table, this will make more sense shortly. | Eric Christopher | 2011-05-11 | 1 | -11/+29 | |
| | | | | | | | Part of rdar://8470697 llvm-svn: 131200 | |||||
| * | Fix encoding of Thumb BLX register instructions. Patch by Koan-Sin Tan. | Owen Anderson | 2011-05-11 | 1 | -4/+8 | |
| | | | | | llvm-svn: 131189 | |||||
| * | Add custom lowering of X86 vector SRA/SRL/SHL when the shift amount is a ↵ | Nadav Rotem | 2011-05-11 | 2 | -11/+80 | |
| | | | | | | | splat vector. llvm-svn: 131179 | |||||
| * | Give the 'eh.sjlj.dispatchsetup' intrinsic call the value coming from the setjmp | Bill Wendling | 2011-05-11 | 2 | -4/+4 | |
| | | | | | | | | | intrinsic call. This prevents it from being reordered so that it appears *before* the setjmp intrinsic (thus making it completely useless). <rdar://problem/9409683> llvm-svn: 131174 | |||||
| * | Optimize atomic lock or that doesn't use the result value. | Eric Christopher | 2011-05-10 | 2 | -1/+83 | |
| | | | | | | | | | Next up: xor and and. Part of rdar://8470697 llvm-svn: 131171 | |||||
| * | Refactor lock versions of binary operators to be a little less | Eric Christopher | 2011-05-10 | 1 | -73/+83 | |
| | | | | | | | cut and paste. llvm-svn: 131139 | |||||
| * | First cut at getting debugging support for ARM/MC/ELF/.o | Jason W Kim | 2011-05-10 | 1 | -1/+0 | |
| | | | | | | | | DWARF stuff also gets fixed up by ELFARMAsmBackend::ApplyFixup(), but the offset is not guaranteed to be mod 4 == 0 as in text/data. llvm-svn: 131137 | |||||
| * | PTX: add PTX 2.3 setting in PTX sub-target. | Justin Holewinski | 2011-05-10 | 2 | -0/+4 | |
| | | | | | | | Patch by Wei-Ren Chen llvm-svn: 131123 | |||||
| * | Fix td file comments for Mips. | Eric Christopher | 2011-05-09 | 3 | -3/+7 | |
| | | | | | | | Patch by Liu <proljc@gmail.com>! llvm-svn: 131086 | |||||
| * | Fixed MC encoding for index_align for VLD1/VST1 (single element from one ↵ | Mon P Wang | 2011-05-09 | 4 | -2/+60 | |
| | | | | | | | lane) for size 32 llvm-svn: 131085 | |||||
| * | X86: Add a bunch of peeps for add and sub of SETB. | Benjamin Kramer | 2011-05-08 | 1 | -0/+24 | |
| | | | | | | | | | | | | | | | | | "b + ((a < b) ? 1 : 0)" compiles into cmpl %esi, %edi adcl $0, %esi instead of cmpl %esi, %edi sbbl %eax, %eax andl $1, %eax addl %esi, %eax This saves a register, a false dependency on %eax (Intel's CPUs still don't ignore it) and it's shorter. llvm-svn: 131070 | |||||
| * | Eliminate the ARM sub-register indexes that are not needed by the sources. | Jakob Stoklund Olesen | 2011-05-07 | 1 | -33/+6 | |
| | | | | | | | | Tablegen will invent its own names for these indexes, and the register file is a bit simpler. llvm-svn: 131059 | |||||
| * | Fix the non-MC encoding of pkhbt and pkhtb. | Eric Christopher | 2011-05-07 | 1 | -0/+6 | |
| | | | | | | | Patch by Stephen Hines. llvm-svn: 131045 | |||||
| * | 1. Keep lines in 80 columns. | Akira Hatanaka | 2011-05-06 | 2 | -6/+6 | |
| | | | | | | | | 2. Remove unused function. 3. Correct indentation. llvm-svn: 131028 | |||||
| * | Make the logic for determining function alignment more explicit. No ↵ | Eli Friedman | 2011-05-06 | 26 | -97/+30 | |
| | | | | | | | functionality change. llvm-svn: 131012 | |||||
| * | Dead code elimination. | Rafael Espindola | 2011-05-06 | 2 | -4/+0 | |
| | | | | | llvm-svn: 130984 | |||||
| * | PTX: add PTX 2.3 language target | Justin Holewinski | 2011-05-06 | 3 | -1/+7 | |
| | | | | | | | Patch by Wei-Ren Chen llvm-svn: 130980 | |||||
| * | Move PPC Linux to CFI. | Rafael Espindola | 2011-05-05 | 1 | -1/+1 | |
| | | | | | llvm-svn: 130951 | |||||
| * | PR9848: pandn is not commutative. | Eli Friedman | 2011-05-05 | 1 | -1/+1 | |
| | | | | | | | No test because I can't think of any way to write one that won't break quickly. llvm-svn: 130932 | |||||
| * | Remove a flag that would set the ".eh" symbol as .globl. MachO was the only one | Bill Wendling | 2011-05-05 | 1 | -1/+0 | |
| | | | | | | | | | who used this flag, and it now emits CFI and doesn't emit this anymore. All other targets left this flag "false". <rdar://problem/8486371> llvm-svn: 130918 | |||||
| * | Fix X86RegisterInfo::getMatchingSuperRegClass for sub_8bit_hi. | Jakob Stoklund Olesen | 2011-05-04 | 1 | -13/+8 | |
| | | | | | | | | It is OK for B to be any GR8_ABCD_H superclass, the returned register class doesn't have to map surjectively onto B. llvm-svn: 130892 | |||||

