summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
* white space cleanupsNadav Rotem2011-09-181-5/+4
| | | | llvm-svn: 139994
* Fix typo by changing Lower256IntVETCC to Lower256IntVSETCC.Craig Topper2011-09-181-3/+3
| | | | llvm-svn: 139993
* Apply Duncan's test fix from r139986 to the avx version of that test too.Benjamin Kramer2011-09-181-4/+4
| | | | llvm-svn: 139992
* Synthesize x86 max/min instructions also for vectors (i.e. produceDuncan Sands2011-09-173-11/+46
| | | | | | | | | maxps and maxpd). This broke the sse41-blend.ll testcase by causing maxpd to be produced rather than a cmp+blend pair, which is the reason I tweaked it. Gives a small speedup on doduc with dragonegg when the GCC vectorizer is used. llvm-svn: 139986
* Describe more AVX 128-bit convert instructions without patterns to haveBruno Cardoso Lopes2011-09-161-4/+15
| | | | | | mayLoad = 1 llvm-svn: 139973
* Bitfield mask instructions are unpredictable if the encoded LSB is higher ↵Owen Anderson2011-09-161-1/+4
| | | | | | than the encoded MSB. llvm-svn: 139972
* Add a testcase for another corner-case decoding.Owen Anderson2011-09-161-0/+2
| | | | llvm-svn: 139970
* Fix bitfield decoding based on Eli's feedback.Owen Anderson2011-09-161-4/+3
| | | | llvm-svn: 139969
* Add ImmutableMapRef and ImmutableSetRef, which consolidate ↵Ted Kremenek2011-09-162-0/+286
| | | | | | Immutable[Map,Set] and its Factory. This may eventually replace Immtuable[Map,Set]. llvm-svn: 139967
* Thumb2 assembly parsing and encoding for SUB(immediate).Jim Grosbach2011-09-164-7/+57
| | | | llvm-svn: 139966
* Thumb2 pre-indexed loads/stores use the restricted GPR set for Rt.Owen Anderson2011-09-161-1/+1
| | | | llvm-svn: 139965
* Fix disassembly of Thumb2 BFI instructions with bit range of [0, 32).Owen Anderson2011-09-162-0/+6
| | | | llvm-svn: 139964
* Thumb2 assembly parsing and encoding for STRT.Jim Grosbach2011-09-161-0/+14
| | | | llvm-svn: 139963
* Thumb2 assembly parsing and encoding for LDRHT/STRHT.Jim Grosbach2011-09-161-0/+28
| | | | llvm-svn: 139962
* Thumb2 assembly parsing and encoding for STREX/STREXB/STREXH/STREXD.Jim Grosbach2011-09-161-0/+18
| | | | llvm-svn: 139961
* Thumb2 assembly parsing and encoding for STRD.Jim Grosbach2011-09-161-0/+18
| | | | llvm-svn: 139960
* Simplify comment. There's no Thumb LDRD(register) encoding. That's ARM only.Jim Grosbach2011-09-161-1/+1
| | | | llvm-svn: 139959
* Add fixed bits to correctly distinguish Thumb2 SSAT/SSAT16's.Owen Anderson2011-09-162-0/+8
| | | | llvm-svn: 139958
* Thumb2 assembly parsing and encoding for STRBT.Jim Grosbach2011-09-161-0/+14
| | | | llvm-svn: 139957
* Thumb2 assembly parsing and encoding for STRH.Jim Grosbach2011-09-161-0/+46
| | | | llvm-svn: 139956
* Remove test of undocumented format.Jim Grosbach2011-09-161-9/+0
| | | | llvm-svn: 139955
* Thumb2 assembly parsing and encoding for STRB.Jim Grosbach2011-09-161-0/+46
| | | | llvm-svn: 139954
* Add mayLoad attribute to AVX convert instructions, since non of themBruno Cardoso Lopes2011-09-161-0/+1
| | | | | | | | are declared with load patterns. This fix the crash in PR10941. No testcases, since a fold is triggered and then converted back to the register form afterwards. llvm-svn: 139953
* Shuffle a few more thumb2 tests to match the comment headings.Jim Grosbach2011-09-161-24/+24
| | | | llvm-svn: 139952
* Thumb2 tests for STR(literal), STR(register) and STR pre/post indexed immediate.Jim Grosbach2011-09-161-0/+39
| | | | llvm-svn: 139951
* Shuffle a few tests around.Jim Grosbach2011-09-161-12/+12
| | | | llvm-svn: 139950
* Thumb2 assembly parsing and encoding for STR.Jim Grosbach2011-09-163-21/+74
| | | | | | | More addressing mode encoding bits. Handle pre increment for STR/STRB/STRH and STR(register). llvm-svn: 139949
* Add -rfunc and -rglob options to llvm-extract to support regularChad Rosier2011-09-162-6/+83
| | | | | | expression matching. llvm-svn: 139945
* Tidy up. 80 columns.Jim Grosbach2011-09-161-6/+12
| | | | llvm-svn: 139944
* Fix disassembly of Thumb2 LDRSH with a #-0 offset.Owen Anderson2011-09-163-3/+9
| | | | llvm-svn: 139943
* Thumb2 assembly parsing and encoding for STR(immediate).Jim Grosbach2011-09-162-0/+32
| | | | | | Add aliases for STRB/STRH while there. Tests forthcoming for those. llvm-svn: 139942
* Thumb2 assembly parsing and encoding for STMDB.Jim Grosbach2011-09-161-0/+14
| | | | llvm-svn: 139940
* Fix PR10884.Bruno Cardoso Lopes2011-09-162-9/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This PR basically reports a problem where a crash in generated code happened due to %rbp being clobbered: pushq %rbp movq %rsp, %rbp .... vmovmskps %ymm12, %ebp .... movq %rbp, %rsp popq %rbp ret Since Eric's r123367 commit, the default stack alignment for x86 32-bit has changed to be 16-bytes. Since then, the MaxStackAlignmentHeuristicPass hasn't been really used, but with AVX it becomes useful again, since per ABI compliance we don't always align the stack to 256-bit, but only when there are 256-bit incoming arguments. ReserveFP was only used by this pass, but there's no RA target hook that uses getReserveFP() to check for the presence of FP (since nothing was triggering the pass to run, the uses of getReserveFP() were removed through time without being noticed). Change this pass to use setForceFramePointer, which is properly called by MachineFunction hasFP method. The testcase is very big and dependent on RA, not sure if it's worth adding to test/CodeGen/X86. llvm-svn: 139939
* Thumb2 assembly parsing and encoding for STMIA.Jim Grosbach2011-09-163-1/+55
| | | | llvm-svn: 139938
* Move the AC_PROG_CC checks earlier in the configure file. The latenessEric Christopher2011-09-162-1949/+932
| | | | | | | | | | | | of the original check meant that configure was caching the default CC check and using that instead of the result of AC_PROG_CC in both configure checks and during compilation. This wasn't affecting C++ so it was hard to notice. Regenerate configure. llvm-svn: 139937
* Migrate this to use clang by default as well.Eric Christopher2011-09-161-6/+3
| | | | llvm-svn: 139936
* We now look for clang, then llvm-gcc, then gcc as our compiler. We don't needEric Christopher2011-09-161-12/+0
| | | | | | this anymore. llvm-svn: 139935
* Have the llvm configure process look for clang, then llvm-gcc, and then gccEric Christopher2011-09-162-6/+6
| | | | | | on all platforms. llvm-svn: 139934
* Thumb2 assembly parsing and encoding for SSUB16/SSUB8.Jim Grosbach2011-09-161-0/+16
| | | | llvm-svn: 139931
* Thumb2 assembly parsing and encoding for SSAX.Jim Grosbach2011-09-162-0/+20
| | | | llvm-svn: 139929
* Fix a minor bug in fs::create_directories. Patch by Albert Wong.Eli Friedman2011-09-161-1/+1
| | | | llvm-svn: 139928
* Thumb2 assembly parsing and encoding for SSAT16.Jim Grosbach2011-09-161-0/+10
| | | | llvm-svn: 139927
* Thumb2 assembly parsing and encoding for SSAT.Jim Grosbach2011-09-162-2/+25
| | | | llvm-svn: 139926
* Thumb2 assembly parsing and encoding for SRS.Jim Grosbach2011-09-162-20/+48
| | | | llvm-svn: 139925
* Thumb2 assembly parsing and encoding for SMMUSD/SMUSDX.Jim Grosbach2011-09-161-0/+16
| | | | llvm-svn: 139923
* Thumb2 assembly parsing and encoding for SMMULWB/SMULWT.Jim Grosbach2011-09-161-0/+16
| | | | llvm-svn: 139922
* Thumb2 assembly parsing and encoding for SMMULL.Jim Grosbach2011-09-162-3/+15
| | | | llvm-svn: 139921
* Fix comment.Jim Grosbach2011-09-161-1/+1
| | | | llvm-svn: 139919
* Thumb2 assembly parsing and encoding for SMULBB/SMULBT/SMULTB/SMULTT.Jim Grosbach2011-09-161-0/+24
| | | | llvm-svn: 139918
* Thumb2 assembly parsing and encoding for SMMUAD'dib.Jim Grosbach2011-09-161-0/+16
| | | | llvm-svn: 139917
OpenPOWER on IntegriCloud