summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
* use 64-bit types instead of off_t/size_t to avoid the issue whenIvan Krasin2011-09-153-8/+8
| | | | | | | | gold plugin is built with Large File Support (sizeof(off_t) == 64 on i686) and the rest of LLVM is built w/o Large File Support (sizeof(off_t) == 32 on i686) which corrupts the stack. llvm-svn: 139873
* Add a fixme note!Bruno Cardoso Lopes2011-09-151-0/+2
| | | | llvm-svn: 139872
* Thumb2 assembly parsing and encoding for SHADD16/SHADD8.Jim Grosbach2011-09-151-0/+16
| | | | llvm-svn: 139871
* Thumb2 assembly parsing and encoding for SHASX/SHSAX.Jim Grosbach2011-09-152-0/+40
| | | | llvm-svn: 139870
* Minor cleanup.Eli Friedman2011-09-151-3/+2
| | | | llvm-svn: 139869
* Unbreak the build for compilers that don't include cstdint everywhere.Benjamin Kramer2011-09-151-0/+1
| | | | llvm-svn: 139868
* Thumb2 assembly parsing and encoding for SEV.W.Jim Grosbach2011-09-151-0/+12
| | | | llvm-svn: 139866
* Use a more efficient lowering for Unordered/Monotonic atomic load/store on ↵Eli Friedman2011-09-152-2/+43
| | | | | | Thumb1. llvm-svn: 139865
* Add the remaining AVX versions of instructions to X86InstrInfo, thisBruno Cardoso Lopes2011-09-151-4/+33
| | | | | | | time for describing high latency ones and for recognizting loads from the same base pointer llvm-svn: 139864
* Thumb2 assembly parsing and encoding for SEL.Jim Grosbach2011-09-151-0/+12
| | | | llvm-svn: 139861
* DWARF: Reset the state after parsing a line table prologue and remove an ↵Benjamin Kramer2011-09-153-5/+5
| | | | | | unnecessary lookup. llvm-svn: 139859
* Thumb2 assembly parsing and encoding for SBFX.Jim Grosbach2011-09-151-0/+12
| | | | llvm-svn: 139858
* Factor out partial register update checks for some SSE instructions.Bruno Cardoso Lopes2011-09-151-32/+53
| | | | | | Also add the AVX versions and add comments! llvm-svn: 139854
* Some legalization fixes for atomic load and store.Eli Friedman2011-09-154-1/+38
| | | | llvm-svn: 139851
* llvm-dwarfdump: Add an option to print out line info for a specific addressBenjamin Kramer2011-09-151-5/+15
| | | | | | | | Usage: $ llvm-dwarfdump -address=0x0000000100000ed4 a.out.dSYM/Contents/Resources/DWARF/a.out xxx.c:6:0 llvm-svn: 139850
* Add some missing 'CHECK' lines and tidy up others.Jim Grosbach2011-09-152-84/+48
| | | | llvm-svn: 139849
* CommandLine: Add support for 64 bit unsigned integer options.Benjamin Kramer2011-09-152-0/+35
| | | | llvm-svn: 139848
* DWARF: Don't crash when looking up an invalid address.Benjamin Kramer2011-09-151-0/+6
| | | | llvm-svn: 139846
* Add an option to disable spill hoisting.Jakob Stoklund Olesen2011-09-151-1/+5
| | | | | | | | When -split-spill-mode is enabled, spill hoisting is performed by SplitKit instead of by InlineSpiller. This hidden command line option is for testing the splitter spill mode. llvm-svn: 139845
* Thumb2 assembly parsing and encoding for SBC.Jim Grosbach2011-09-151-0/+46
| | | | llvm-svn: 139844
* Thumb2 assembly parsing and encoding for SASX.Jim Grosbach2011-09-152-0/+20
| | | | llvm-svn: 139843
* Reapply r139759. Disable IV rewriting by default. See PR10916.Andrew Trick2011-09-153-3/+3
| | | | llvm-svn: 139842
* Thumb2 assembly parsing and encoding for SADD16/SADD8.Jim Grosbach2011-09-151-0/+24
| | | | llvm-svn: 139841
* Thumb2 assembly parsing and encoding for RSB.Jim Grosbach2011-09-152-0/+40
| | | | llvm-svn: 139839
* DWARF: Put all the pieces we have together and provide a single accessor to ↵Benjamin Kramer2011-09-154-2/+74
| | | | | | DIContext that provides line information when given an address. llvm-svn: 139836
* DWARF: Remove accessors that parse the whole line table section in one go, ↵Benjamin Kramer2011-09-154-58/+39
| | | | | | | | this can't possibly work. The address size is specified by the compile unit associated with a line table, there is no global address size. llvm-svn: 139835
* Thumb2 assembly parsing and encoding for RRX.Jim Grosbach2011-09-151-0/+16
| | | | llvm-svn: 139831
* Thumb2 assembly parsing and encoding for ROR.Jim Grosbach2011-09-151-0/+38
| | | | llvm-svn: 139830
* Thumb2 assembly parsing and encoding for REV16/REVSH.Jim Grosbach2011-09-152-1/+35
| | | | llvm-svn: 139828
* The the MC disassembler C API to print in verbose mode. Perhaps there ↵Owen Anderson2011-09-152-1/+31
| | | | | | should be a parameter to request verbose mode? llvm-svn: 139821
* Add support for stored annotations to MCInst, and provide facilities for ↵Owen Anderson2011-09-157-4/+70
| | | | | | MC-based InstPrinters to print them out. Enhance the ARM and X86 InstPrinter's to do so in verbose mode. llvm-svn: 139820
* VirtRegMap is counting spill slots, not register spills.Jakob Stoklund Olesen2011-09-152-4/+4
| | | | | | Fix the stats counters to reflect that. llvm-svn: 139819
* Change all checks regarding the presence of any SSE level to alwaysBruno Cardoso Lopes2011-09-152-63/+76
| | | | | | | | | | take into consideration the presence of AVX. This change, together with the SSEDomainFix enabled for AVX, makes AVX codegen to always (hopefully) emit the same code as SSE for 128-bit vector ops. I don't have a testcase for this, but AVX now beats SSE in performance for 128-bit ops in the majority of programas in the llvm testsuite llvm-svn: 139817
* Enable SSEDomainFix pass for AVX mode.Bruno Cardoso Lopes2011-09-151-4/+7
| | | | llvm-svn: 139816
* Count correctly when a COPY turns into a spill or reload.Jakob Stoklund Olesen2011-09-151-1/+7
| | | | | | | The number of spills could go negative since a folded COPY is just a spill, and it may be eliminated. llvm-svn: 139815
* Thumb2 assembly parsing and encoding for REV.Jim Grosbach2011-09-152-0/+18
| | | | llvm-svn: 139813
* Thumb2 assembly parsing and encoding for RBIT.Jim Grosbach2011-09-151-0/+12
| | | | llvm-svn: 139811
* Thumb2 assembly parsing and encoding for signed saturating arithmetic insns.Jim Grosbach2011-09-151-0/+68
| | | | llvm-svn: 139810
* DWARF: Print line tables per compile unit, so they get the right address size.Benjamin Kramer2011-09-153-5/+23
| | | | llvm-svn: 139808
* Assmebler symbol attribute directives don't work on temporary symbols.Jim Grosbach2011-09-151-1/+6
| | | | | | | | | Assembler private local symbols aren't legal targets of symbol attributes, so issue a diagnostic for them. Based on patch by Stepan Dyatkovskiy. llvm-svn: 139807
* Count inserted spills and reloads more accurately.Jakob Stoklund Olesen2011-09-151-14/+22
| | | | | | | | Adjust counters when removing spill and reload instructions. We still don't account for reloads being removed by eliminateDeadDefs(). llvm-svn: 139806
* DWARF: wire up .debug_str dumping.Benjamin Kramer2011-09-155-18/+30
| | | | llvm-svn: 139799
* Handle missing newline at EOF more gracefully in MC AsmLexer.Jim Grosbach2011-09-151-1/+8
| | | | | | | | | If we see an EOF w/o a preceding end-of-line, return an EndOfStatement token before returning the Eof token. Based on patch by Stepan Dyatkovskiy. llvm-svn: 139798
* Trace through sibling PHIs in bulk.Jakob Stoklund Olesen2011-09-151-23/+50
| | | | | | | | | | | | | | | When traceSiblingValue() encounters a PHI-def value created by live range splitting, don't look at all the predecessor blocks. That can be very expensive in a complicated CFG. Instead, consider that all the non-PHI defs jointly dominate all the PHI-defs. Tracing directly to all the non-PHI defs is much faster that zipping around in the CFG when there are many PHIs with many predecessors. This significantly improves compile time for indirectbr interpreters. llvm-svn: 139797
* ARM support the pre-UAL mnemonic 'qsubaddx' for 'qsax.'Jim Grosbach2011-09-151-0/+3
| | | | llvm-svn: 139796
* Re-order test.Jim Grosbach2011-09-151-17/+18
| | | | llvm-svn: 139795
* Thumb2 push/pop mnemonic recognition.Jim Grosbach2011-09-151-0/+6
| | | | llvm-svn: 139794
* Use getPrevSlot() instead of getPrevIndex().Jakob Stoklund Olesen2011-09-151-1/+1
| | | | | | | | The getPrevIndex() function moves to the same slot in the previous instruction. For getVNInfoBefore(), we just need the previous slot in the same instruction. llvm-svn: 139793
* Speed up LiveIntervals::shrinkToUse with some caching.Jakob Stoklund Olesen2011-09-151-6/+8
| | | | | | | | | | | | | Blocks with multiple PHI successors only need to go on the worklist once. Use a SmallPtrSet to track the live-out blocks that have already been handled. This is a lot faster than the two live range check we would otherwise do. Also stop recomputing hasPHIKill flags. Like RenumberValues(), it is conservatively correct to leave them in, and they are not used for anything important. llvm-svn: 139792
* Revert r139782, "RemoveCopyByCommutingDef doesn't need hasPHIKill()."Jakob Stoklund Olesen2011-09-151-8/+8
| | | | | | | | | | It does, after all. RemoveCopyByCommutingDef rewrites the uses of one particular value number in A. It doesn't know how to rewrite phi uses, so there can't be any. llvm-svn: 139787
OpenPOWER on IntegriCloud