summaryrefslogtreecommitdiffstats
path: root/llvm/test
Commit message (Collapse)AuthorAgeFilesLines
* MC-COFF: Fix Simple and Complex type. Fixes PR8320.Michael J. Spencer2010-10-081-11/+24
| | | | llvm-svn: 116037
* Revert "reimplement the second half of the or/add optimization. We should now",Daniel Dunbar2010-10-081-36/+2
| | | | | | which depends on r116007, which I am about to revert. llvm-svn: 116031
* reimplement the second half of the or/add optimization. We should nowChris Lattner2010-10-081-2/+36
| | | | | | | | | | only end up emitting LEA instead of OR. If we aren't able to promote something into an LEA, we should never be emitting it as an ADD. Add some testcases that we emit "or" in cases where we used to produce an "add". llvm-svn: 116026
* Add test file for simple ARM binary encodings with MCJim Grosbach2010-10-081-0/+18
| | | | llvm-svn: 116024
* MC-COFF: Add test for my last commit.Michael J. Spencer2010-10-081-0/+29
| | | | llvm-svn: 116015
* Remove LoopIndexSplit pass. It is neither maintained nor used by anyone.Devang Patel2010-10-0738-2377/+0
| | | | llvm-svn: 116004
* convert cmp to use a multipatternChris Lattner2010-10-072-8/+8
| | | | llvm-svn: 115978
* Canonicalize X86ISD::MOVDDUP nodes to v2f64 to make sure all cases match. ↵Evan Cheng2010-10-071-1/+16
| | | | | | Also eliminate unneeded isel patterns. rdar://8520311 llvm-svn: 115977
* MC-COFF: Fix symbol aliases. Fixes PR8251.Michael J. Spencer2010-10-071-0/+48
| | | | llvm-svn: 115909
* test: Fix binary stdin issues with coff-dump on Windows.Michael J. Spencer2010-10-071-1/+4
| | | | llvm-svn: 115908
* Allow use of the 16-bit literal move instruction in CMOVs for ARM mode.Jim Grosbach2010-10-071-2/+2
| | | | llvm-svn: 115884
* Get binding and visibility info from the the alias, but Type from the symbolRafael Espindola2010-10-061-9/+34
| | | | | | being aliased. llvm-svn: 115836
* If a symbol is global, reloc against it even if it is in a mergeable section.Rafael Espindola2010-10-061-0/+15
| | | | llvm-svn: 115817
* Make sure weak symbols are listed after the local ones.Rafael Espindola2010-10-061-0/+16
| | | | llvm-svn: 115795
* Correctly handle GOTPCREL relocations.Rafael Espindola2010-10-061-0/+9
| | | | llvm-svn: 115793
* - Add TargetInstrInfo::getOperandLatency() to compute operand latencies. ThisEvan Cheng2010-10-062-3/+3
| | | | | | | | | | | | | allow target to correctly compute latency for cases where static scheduling itineraries isn't sufficient. e.g. variable_ops instructions such as ARM::ldm. This also allows target without scheduling itineraries to compute operand latencies. e.g. X86 can return (approximated) latencies for high latency instructions such as division. - Compute operand latencies for those defined by load multiple instructions, e.g. ldm and those used by store multiple instructions, e.g. stm. llvm-svn: 115755
* Generalize tblgen's dag parsing logic to handle arbitrary expressionsChris Lattner2010-10-063-2/+38
| | | | | | | | | | | | | | as the operator of the dag. Specifically, this allows parsing things like (F.x 4) in addition to just (a 4). Unfortunately, this runs afoul of an idiom being used by llvmc. It is using dags like (foo [1,2,3]) to represent a list of stuff being passed into foo. With this change, this is parsed as a [1,2,3] subscript on foo instead of being the first argument to the dag. Cope with this in the short term by requiring a "-llvmc-temp-hack" argument to tblgen to get the old parsing behavior. llvm-svn: 115742
* rename add some comments.Chris Lattner2010-10-061-0/+5
| | | | llvm-svn: 115741
* filecheckizeChris Lattner2010-10-062-21/+24
| | | | llvm-svn: 115740
* remove the !nameconcat tblgen feature. It "shorthand" and only used in 4 placesChris Lattner2010-10-061-91/+0
| | | | | | where !cast is just as short. llvm-svn: 115722
* Use a relocation against the symbol if it is a PLT and the symbol is in anotherRafael Espindola2010-10-051-4/+20
| | | | | | section. Common because of linkonce sections. llvm-svn: 115718
* enhance tblgen to support anonymous defm's, use this toChris Lattner2010-10-051-0/+12
| | | | | | simplify the X86 CMOVmr's. llvm-svn: 115702
* Implement more alias cases.Rafael Espindola2010-10-051-0/+37
| | | | llvm-svn: 115699
* PSHUFW is in SSE, not SSSE3.Bill Wendling2010-10-051-2/+2
| | | | llvm-svn: 115691
* Replace a gross hack (the MOV64ri_alt instruction) with a slightly less Chris Lattner2010-10-052-3/+3
| | | | | | gross hack (having the asmmatcher handle the alias). llvm-svn: 115685
* Don't crash in a strange .size directive.Rafael Espindola2010-10-051-0/+9
| | | | llvm-svn: 115684
* MC-COFF: Fix (PR8278) temporary symbol relocations.Michael J. Spencer2010-10-051-30/+37
| | | | llvm-svn: 115656
* Add support for a fill value in the .zero directive.Rafael Espindola2010-10-051-2/+3
| | | | llvm-svn: 115655
* Implement a simple alias case and refactor the code a bit so that theRafael Espindola2010-10-051-0/+22
| | | | | | isInSymtab and isLocal logic in the two loops don't get easily out of sync. llvm-svn: 115643
* test/COFF: Fix symbol indexes and names. Update tests to match.Michael J. Spencer2010-10-053-20/+24
| | | | llvm-svn: 115642
* test/COFF: Remove temp file usage.Michael J. Spencer2010-10-053-7/+4
| | | | llvm-svn: 115641
* test/coff-dump: Support reading from stdin.Michael J. Spencer2010-10-051-1/+6
| | | | llvm-svn: 115640
* Cleanup Whitespace.Michael J. Spencer2010-10-051-2/+2
| | | | llvm-svn: 115639
* Use a more efficient lowering of uint64_t --> float that can take advantage ↵Owen Anderson2010-10-051-0/+21
| | | | | | | | | of hardware signed integer conversion without having to do a double cast (uint64_t --> double --> float). This is based on the algorithm from compiler_rt's __floatundisf for X86-64. llvm-svn: 115634
* Produce a undefined reference to _GLOBAL_OFFSET_TABLE_ when needed.Rafael Espindola2010-10-051-1/+5
| | | | llvm-svn: 115623
* Tests that now pass.Rafael Espindola2010-10-052-19/+80
| | | | llvm-svn: 115622
* test/CodeGen/X86/atomic_op.ll: Rename @main to @func. Extra sequences will ↵NAKAMURA Takumi2010-10-051-1/+1
| | | | | | be inserted to @main as prologue on cygming, to fail. llvm-svn: 115611
* Added a testcase for the ENTER instruction.Sean Callanan2010-10-051-1/+4
| | | | llvm-svn: 115580
* The pshufw instruction came about in MMX2 when SSE was introduced. Don't placeBill Wendling2010-10-041-0/+3
| | | | | | | | | it in with the SSSE3 instructions. Steward! Could you place this chair by the aft sun deck? I'm trying to get away from the Astors. They are such boors! llvm-svn: 115552
* Implement ELF::R_X86_64_GOTPCREL.Rafael Espindola2010-10-041-1/+8
| | | | llvm-svn: 115547
* Produce a R_X86_64_PLT32 when needed.Rafael Espindola2010-10-041-0/+14
| | | | llvm-svn: 115541
* Produce a R_X86_64_GOT32 when needed.Rafael Espindola2010-10-041-0/+14
| | | | llvm-svn: 115537
* Add hook in MCSection to decide when to use "optimized nops", for eachJan Wen Voung2010-10-043-1/+86
| | | | | | | section kind. Previously, optimized nops were only used for MachO. Also added tests for ELF and COFF. llvm-svn: 115523
* Include the section address in the computation of the relocation.Rafael Espindola2010-10-041-0/+3
| | | | llvm-svn: 115509
* Correctly compute the relocation when it is not in the first fragment.Rafael Espindola2010-10-041-1/+2
| | | | llvm-svn: 115506
* va_args support for Win64.Anton Korobeynikov2010-10-031-0/+20
| | | | | | Patch by Cameron! llvm-svn: 115480
* Properly emit stack probe on win64 (for non-mingw targets).Anton Korobeynikov2010-10-031-2/+3
| | | | | | Based on the patch by Cameron Esfahani! llvm-svn: 115479
* Add 3DNowA instructions.Eli Friedman2010-10-031-0/+16
| | | | llvm-svn: 115477
* unbreak buildbotChris Lattner2010-10-031-2/+2
| | | | llvm-svn: 115476
* the immediate field of pshufw is actually an 8-bit field, not a 8-bit field ↵Chris Lattner2010-10-031-0/+5
| | | | | | that is sign extended. This fixes PR8288 llvm-svn: 115473
OpenPOWER on IntegriCloud