summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Target
Commit message (Collapse)AuthorAgeFilesLines
...
* Add complete assembler support for FMA3 instructions, with descriptions and ↵Bruno Cardoso Lopes2010-07-233-0/+72
| | | | | | encodings taken from the AVX manual llvm-svn: 109204
* The only supported calling convention for X86-64 usesDale Johannesen2010-07-231-1/+14
| | | | | | | | | | | | | | | SSE, so we can't return floating point values if this is disabled. Detect this error for clang. With SSE1 only, f64 is a problem; it can be done, but neither llvm-gcc nor clang has ever generated correct code for it. Since nobody noticed this I think it's OK to treat it as an error for now. This also handles SSE-sized vectors of floating point. 8207686, 8204109. llvm-svn: 109201
* Fix some AVX instructions which didnt had HasAVX prefix. And also a problem ↵Bruno Cardoso Lopes2010-07-231-6/+8
| | | | | | with PINSRW, which was totally wrong because of a typo I introduced previously llvm-svn: 109198
* eliminate the TargetInstrInfo::GetInstSizeInBytes hook. Chris Lattner2010-07-224-5/+5
| | | | | | | | ARM/PPC/MSP430-specific code (which are the only targets that implement the hook) can directly reference their target-specific instrinfo classes. llvm-svn: 109171
* Add remaining AVX instructions (most of them dealing with GR64 destinations. ↵Bruno Cardoso Lopes2010-07-221-5/+30
| | | | | | This complete the assembler support for the general AVX ISA. But we still miss instructions from FMA3 and CLMUL specific feature flags, which are now the next step llvm-svn: 109168
* remove the JIT "NeedsExactSize" feature and supporting logic.Chris Lattner2010-07-222-9/+0
| | | | llvm-svn: 109167
* switch a private implementation of GetFunctionSizeInBytes.Chris Lattner2010-07-221-1/+14
| | | | | | | | | This is probably not the best way to implement "Force LR to be spilled if the Thumb function size is > 2048." do this, it should use the branch shortening infrastructure, but I'm just preserving functionality here. llvm-svn: 109165
* X86MCInstLower now depends on AsmPrinter being around.Chris Lattner2010-07-222-32/+11
| | | | llvm-svn: 109154
* instead of migrating it to the MC instruction encoder, justChris Lattner2010-07-223-553/+100
| | | | | | | | | | | | | | rip out the implementation of X86InstrInfo::GetInstSizeInBytes. The code being ripped out just implemented a copy and hacked up version of the (old) instruction encoder, and is buggy and terrible in other ways. Since "GetInstSizeInBytes" is really only there to support the JIT's "NeedsExactSize" hook (which noone is using), just rip out the code. I will rip out the NeedsExactSize hook next. This resolves rdar://7617809 - switch X86InstrInfo::GetInstSizeInBytes to use X86MCCodeEmitter llvm-svn: 109149
* ARMv4 JIT forgets to set the lr register when making a indirect function ↵Xerxes Ranby2010-07-222-1/+14
| | | | | | call. Fixes PR7608 llvm-svn: 109125
* undo 80 column trespassing I causedGabor Greif2010-07-221-1/+2
| | | | llvm-svn: 109092
* Mark an assert-only variable as used.Chandler Carruth2010-07-221-0/+1
| | | | llvm-svn: 109091
* Fix the generated file name for CMake.Chandler Carruth2010-07-221-1/+1
| | | | llvm-svn: 109090
* Attempt to fix linking issues with CMake. Please review other CMake users,Chandler Carruth2010-07-222-2/+2
| | | | | | especially on other platforms. Is there a better way to fix this. llvm-svn: 109084
* Update CMake files.Owen Anderson2010-07-221-0/+1
| | | | llvm-svn: 109081
* Custom lower the memory barrier instructions and add supportEric Christopher2010-07-225-18/+75
| | | | | | | | for lowering without sse2. Add a couple of new testcases. Fixes a few libgomp tests and latent bugs. Remove a few todos. llvm-svn: 109078
* Fix constant island pass's handling of tBR_JTr. The offset of the ↵Evan Cheng2010-07-221-2/+10
| | | | | | | | | | | | | instruction does not have to be 4-byte aligned. Rather, it's the offset + 2 that must be aligned since the instruction expands into: mov pc, r1 .align 2 LJTI0_0_0: .long LBB0_14 This fixes rdar://8213383. No test case since it's not possible to come up with a suitable small one. llvm-svn: 109076
* 80-columns.Eric Christopher2010-07-221-9/+12
| | | | llvm-svn: 109070
* Make fast isel win64-aware w.r.t. call-clobbered regsNate Begeman2010-07-221-3/+14
| | | | llvm-svn: 109069
* More register pressure aware scheduling work.Evan Cheng2010-07-211-14/+11
| | | | llvm-svn: 109064
* Add more 256-bit forms for a bunch of regular AVX instructionsBruno Cardoso Lopes2010-07-212-76/+160
| | | | | | | Add 64-bit (GR64) versions of some instructions (which are not described in their SSE forms, but are described in AVX) llvm-svn: 109063
* Fixes win64. It was broken by a previous patch where I missed the !isWin64Rafael Espindola2010-07-211-4/+8
| | | | | | and then forced every register to be a vr128 on win64. llvm-svn: 109060
* For ARM/Darwin, add a dwarf entry indicating whether a function is arm or thumbJim Grosbach2010-07-211-0/+17
| | | | | | rdar://8202967 llvm-svn: 109057
* add some rough support for making mcinst lowering work without anChris Lattner2010-07-211-5/+23
| | | | | | | | asmprinter or mangler around. This is option #B for killing off X86InstrInfo::GetInstSizeInBytes. Option #A (killing "needsexactsize") was sent for consideration to llvmdev. llvm-svn: 109056
* Baby steps towards ARM fast-isel.Eric Christopher2010-07-215-1/+89
| | | | llvm-svn: 109047
* Fix batch of converting RegisterPass<> to INTIALIZE_PASS().Owen Anderson2010-07-211-2/+1
| | | | llvm-svn: 109045
* Add missing AVX convert instructions. Those instructions are not described ↵Bruno Cardoso Lopes2010-07-211-22/+56
| | | | | | in their SSE forms (although they exist), but add the AVX forms anyway, so the assembler can benefit from it llvm-svn: 109039
* Fix a couple issues with Win64 ABINate Begeman2010-07-212-6/+3
| | | | | | | | | | | 1) all registers were spilled as xmm, regardless of actual size 2) win64 abi doesn't do the varargs-size-in-%al thing Still to look into: xmm6-15 are marked as clobbered by call instructions on win64 even though they aren't. llvm-svn: 109035
* Avoid AVX instructions to be selected instead of its SSE formBruno Cardoso Lopes2010-07-211-1/+1
| | | | llvm-svn: 109032
* Fix calling convention on ARM if vfp2+ is enabled.Rafael Espindola2010-07-211-1/+5
| | | | llvm-svn: 109009
* Pulling out previous patch, must've run the tests inEric Christopher2010-07-213-40/+3
| | | | | | the wrong directory. llvm-svn: 109005
* Lower MEMBARRIER on x86 and support processors without SSE2.Eric Christopher2010-07-213-3/+40
| | | | | | | Fixes a pile of libgomp failures in the llvm-gcc testsuite due to the libcall not existing. llvm-svn: 109004
* Add AVX only vzeroall and vzeroupper instructionsBruno Cardoso Lopes2010-07-212-2/+10
| | | | llvm-svn: 109002
* Teach bottom up pre-ra scheduler to track register pressure. Work in progress.Evan Cheng2010-07-214-14/+45
| | | | llvm-svn: 108991
* Add new AVX vpermilps, vpermilpd and vperm2f128 instructionsBruno Cardoso Lopes2010-07-211-0/+35
| | | | llvm-svn: 108984
* Add new AVX vmaskmov instructions, and also fix the VEX encoding bits to ↵Bruno Cardoso Lopes2010-07-212-11/+45
| | | | | | support it llvm-svn: 108983
* Add new AVX vextractf128 instructionsBruno Cardoso Lopes2010-07-201-0/+10
| | | | llvm-svn: 108964
* make asmprinter optional, even though passing in null will cause things to ↵Chris Lattner2010-07-202-13/+14
| | | | | | explode right now. llvm-svn: 108955
* continue pushing dependencies around.Chris Lattner2010-07-202-8/+9
| | | | llvm-svn: 108952
* reduce X86MCInstLower dependencies on asmprinter.Chris Lattner2010-07-202-8/+14
| | | | llvm-svn: 108950
* pass around MF, not MMI.Chris Lattner2010-07-202-5/+5
| | | | llvm-svn: 108949
* cleanups.Chris Lattner2010-07-202-14/+10
| | | | llvm-svn: 108947
* move two asmprinter methods into the asmprinter .cpp file.Chris Lattner2010-07-202-38/+39
| | | | llvm-svn: 108945
* prune #includes a little.Chris Lattner2010-07-205-6/+8
| | | | llvm-svn: 108929
* Add new AVX instruction vinsertf128Bruno Cardoso Lopes2010-07-202-0/+15
| | | | llvm-svn: 108892
* Using BIC for immediates needs an extra bump for its complexity to getJim Grosbach2010-07-201-0/+1
| | | | | | instruction selection to prefer it when possible. rdar://7903972 llvm-svn: 108844
* Removed un-used code.Jim Grosbach2010-07-201-49/+0
| | | | llvm-svn: 108841
* Fix PR7174, a couple o Mips fixes:Bruno Cardoso Lopes2010-07-202-1/+21
| | | | | | | | | | | - Fix a typo for PIC check during jmp table lowering - Also fix the "first jump table basic block is not considered only reachable by fall through" problem, use this ad-hoc solution until I come up with something better. Patch by stetorvs@gmail.com llvm-svn: 108820
* Fix Mips PR7473. Patch by stetorvs@gmail.comBruno Cardoso Lopes2010-07-201-3/+3
| | | | llvm-svn: 108816
* Constify some arguments.Eric Christopher2010-07-209-14/+14
| | | | llvm-svn: 108812
OpenPOWER on IntegriCloud