| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Make a std::vector a SmallVector<*, 32> like the other vectors in the same | Cameron Zwarich | 2011-01-18 | 1 | -1/+1 | |
| | | | | | | | | function. This seems to be about a 1.5% speedup of -scalarrepl on test-suite with SPEC2000 and SPEC2006. llvm-svn: 123731 | |||||
| * | Reduce indentation and remove commented out code. | Rafael Espindola | 2011-01-18 | 1 | -122/+101 | |
| | | | | | llvm-svn: 123729 | |||||
| * | Remove some now-unused DominanceFrontier methods. | Cameron Zwarich | 2011-01-18 | 2 | -151/+0 | |
| | | | | | llvm-svn: 123726 | |||||
| * | Remove code for updating dominance frontiers and some outdated references to | Cameron Zwarich | 2011-01-18 | 7 | -105/+21 | |
| | | | | | | | dominance and post-dominance frontiers. llvm-svn: 123725 | |||||
| * | Remove outdated references to dominance frontiers. | Cameron Zwarich | 2011-01-18 | 6 | -31/+29 | |
| | | | | | llvm-svn: 123724 | |||||
| * | McARM: Start marking T2 address operands as such, for the benefit of the parser. | Daniel Dunbar | 2011-01-18 | 2 | -0/+17 | |
| | | | | | llvm-svn: 123722 | |||||
| * | Formatting tweak. | Daniel Dunbar | 2011-01-18 | 1 | -1/+1 | |
| | | | | | llvm-svn: 123718 | |||||
| * | Support/CommandLine: Add "Did you mean" print for mismatched operands. | Daniel Dunbar | 2011-01-18 | 1 | -0/+39 | |
| | | | | | llvm-svn: 123717 | |||||
| * | The stub routine that we're calling uses test and so clobbers | Eric Christopher | 2011-01-18 | 1 | -2/+2 | |
| | | | | | | | the flags. llvm-svn: 123712 | |||||
| * | minor change to rafael's recent patches: if something is | Chris Lattner | 2011-01-18 | 1 | -1/+7 | |
| | | | | | | | | constant but requires a unique address, we can still put it in a readonly section, just not a mergable one. llvm-svn: 123711 | |||||
| * | Remove unused variables found by gcc-4.6's -Wunused-but-set-variable. | Jeffrey Yasskin | 2011-01-18 | 5 | -17/+0 | |
| | | | | | llvm-svn: 123707 | |||||
| * | Remove checking that prevented overlapping CALLSEQ_START/CALLSEQ_END | Stuart Hastings | 2011-01-18 | 1 | -15/+9 | |
| | | | | | | | | ranges, add legalizer support for nested calls. Necessary for ARM byval support. Radar 7662569. llvm-svn: 123704 | |||||
| * | Windows/PathV2.inc: For CryptAcquireContext(), CRYPT_VERIFYCONTEXT may be ↵ | NAKAMURA Takumi | 2011-01-17 | 1 | -1/+1 | |
| | | | | | | | specified for easy use. llvm-svn: 123687 | |||||
| * | Windows/PathV2.inc: MoveFileEx() can behave like Posix's mv(1) to specify ↵ | NAKAMURA Takumi | 2011-01-17 | 1 | -1/+2 | |
| | | | | | | | MOVEFILE_COPY_ALLOWED | MOVEFILE_REPLACE_EXISTING. llvm-svn: 123686 | |||||
| * | lib/Support/Windows/Signals.inc: "Showstopper" dialogs may be suppressed ↵ | NAKAMURA Takumi | 2011-01-17 | 1 | -0/+3 | |
| | | | | | | | with SetErrorMode() on Windows 7. llvm-svn: 123685 | |||||
| * | Remove dead code, that I apparently wrote a while back. We seem to be doing ↵ | Owen Anderson | 2011-01-17 | 1 | -15/+0 | |
| | | | | | | | | | | | well enough without whatever this was trying to do. When/if someone has the time to do some empirical evaluations, it might be worth it to figure out what this code was trying to do and see if it's worth resurrecting/fixing. llvm-svn: 123684 | |||||
| * | Add a missing <cctype> include, from Joerg Sonnenberger! | Douglas Gregor | 2011-01-17 | 1 | -0/+1 | |
| | | | | | llvm-svn: 123670 | |||||
| * | Trailing whitespace. | Jim Grosbach | 2011-01-17 | 1 | -17/+17 | |
| | | | | | llvm-svn: 123665 | |||||
| * | Fix an off-by-one error in ctpop combining. | Benjamin Kramer | 2011-01-17 | 2 | -3/+12 | |
| | | | | | llvm-svn: 123664 | |||||
| * | Update tests to accomodate unnamed_addr introduction. | Devang Patel | 2011-01-17 | 2 | -3/+3 | |
| | | | | | llvm-svn: 123663 | |||||
| * | Roll r123609 back in with two changes that fix test failures with expensive | Cameron Zwarich | 2011-01-17 | 4 | -63/+123 | |
| | | | | | | | | | | | | | | | checks enabled: 1) Use '<' to compare integers in a comparison function rather than '<='. 2) Use the uniqued set DefBlocks rather than Info.DefiningBlocks to initialize the priority queue. The speedup of scalarrepl on test-suite + SPEC2000 + SPEC2006 is a bit less, at just under 16% rather than 17%. llvm-svn: 123662 | |||||
| * | Revert rr123550. It causes clang build failure on darwin9. | Devang Patel | 2011-01-17 | 1 | -12/+22 | |
| | | | | | llvm-svn: 123661 | |||||
| * | Archive: Fix temp path names. | Michael J. Spencer | 2011-01-17 | 1 | -4/+6 | |
| | | | | | llvm-svn: 123660 | |||||
| * | Add some platform checks. Also fix a typo on a Makefile. | Oscar Fuentes | 2011-01-17 | 3 | -13/+27 | |
| | | | | | | | Patch by arrowdodger! llvm-svn: 123659 | |||||
| * | Support/raw_ostream: Fix uninitalized variable in raw_fd_ostream constructor. | Michael J. Spencer | 2011-01-17 | 1 | -0/+7 | |
| | | | | | llvm-svn: 123643 | |||||
| * | Remove useless Tag enumeration. | Jay Foad | 2011-01-17 | 3 | -14/+10 | |
| | | | | | llvm-svn: 123623 | |||||
| * | Split up RotateShift itinerary in SPU. | Kalle Raiskila | 2011-01-17 | 2 | -38/+40 | |
| | | | | | | | | | 'rotq*' and 'shlq*' instructions go to the odd pipeline, wheras the inter-vector equivalents 'rot*', 'shl*' go to the even. llvm-svn: 123622 | |||||
| * | Add a DAGCombine to turn (ctpop x) u< 2 into (x & x-1) == 0. | Benjamin Kramer | 2011-01-17 | 2 | -0/+55 | |
| | | | | | | | | | | This shaves off 4 popcounts from the hacked 186.crafty source. This is enabled even when a native popcount instruction is available. The combined code is one operation longer but it should be faster nevertheless. llvm-svn: 123621 | |||||
| * | Don't crash SPU BE with memory accesses with big alignmnet. | Kalle Raiskila | 2011-01-17 | 2 | -4/+13 | |
| | | | | | llvm-svn: 123620 | |||||
| * | Materialize GA addresses with movw + movt pairs for Darwin in PIC mode. e.g. | Evan Cheng | 2011-01-17 | 18 | -85/+248 | |
| | | | | | | | | | | | | | movw r0, :lower16:(L_foo$non_lazy_ptr-(LPC0_0+4)) movt r0, :upper16:(L_foo$non_lazy_ptr-(LPC0_0+4)) LPC0_0: add r0, pc, r0 It's not yet enabled by default as some tests are failing. I suspect bugs in down stream tools. llvm-svn: 123619 | |||||
| * | Roll out r123609 due to failures on the llvm-x86_64-linux-checks bot. | Cameron Zwarich | 2011-01-17 | 4 | -122/+62 | |
| | | | | | llvm-svn: 123618 | |||||
| * | Suppress warning "switch statement contains 'default' but no 'case' labels" ↵ | Francois Pichet | 2011-01-17 | 1 | -0/+1 | |
| | | | | | | | on MSVC. llvm-svn: 123610 | |||||
| * | Eliminate the use of dominance frontiers in PromoteMemToReg. In addition to | Cameron Zwarich | 2011-01-17 | 4 | -62/+122 | |
| | | | | | | | | | | | | | | eliminating a potentially quadratic data structure, this also gives a 17% speedup when running -scalarrepl on test-suite + SPEC2000 + SPEC2006. My initial experiment gave a greater speedup around 25%, but I moved the dominator tree level computation from dominator tree construction to PromoteMemToReg. Since this approach to computing IDFs has a much lower overhead than the old code using precomputed DFs, it is worth looking at using this new code for the second scalarrepl pass as well. llvm-svn: 123609 | |||||
| * | UnRevert "Revert "Archive: Replace all internal uses of PathV1 with PathV2. ↵ | Michael J. Spencer | 2011-01-16 | 2 | -38/+39 | |
| | | | | | | | The external API still uses PathV1."" llvm-svn: 123605 | |||||
| * | Fix rename. | Michael J. Spencer | 2011-01-16 | 1 | -2/+11 | |
| | | | | | llvm-svn: 123604 | |||||
| * | Test for lazy value info's ability to prove the absense of NULLs in pointers. | Nick Lewycky | 2011-01-16 | 1 | -0/+103 | |
| | | | | | llvm-svn: 123601 | |||||
| * | Make everyone happy this time. | Michael J. Spencer | 2011-01-16 | 1 | -1/+1 | |
| | | | | | llvm-svn: 123599 | |||||
| * | Provide instruction sizes for ARMv5 variants of MUL instructions. | Anton Korobeynikov | 2011-01-16 | 1 | -29/+30 | |
| | | | | | | | This fixes PR8987 llvm-svn: 123598 | |||||
| * | Update README.txt to remove the DAE enhancement. | Anders Carlsson | 2011-01-16 | 1 | -23/+0 | |
| | | | | | llvm-svn: 123597 | |||||
| * | Teach DAE to look for functions whose arguments are unused, and change all ↵ | Anders Carlsson | 2011-01-16 | 2 | -2/+88 | |
| | | | | | | | callers to pass in an undefvalue instead. llvm-svn: 123596 | |||||
| * | UnRevert "Revert the archive part of "Support/PathV2: Add identify_magic."" | Michael J. Spencer | 2011-01-16 | 2 | -6/+7 | |
| | | | | | | | This reverts commit dd103021a889a986a181ce36ed7b0e8dc9b645e1. llvm-svn: 123595 | |||||
| * | Try and fix this test. For some reason llvm-ar thinks that the file exists when | Michael J. Spencer | 2011-01-16 | 1 | -0/+1 | |
| | | | | | | | | it shouldn't, but I have no way to verify that it doesn't actually exist on the buildbot. llvm-svn: 123594 | |||||
| * | Revert the archive part of "Support/PathV2: Add identify_magic." | Michael J. Spencer | 2011-01-16 | 2 | -7/+6 | |
| | | | | | llvm-svn: 123593 | |||||
| * | Update tests. | Rafael Espindola | 2011-01-16 | 7 | -26/+26 | |
| | | | | | llvm-svn: 123591 | |||||
| * | tidy up a comment, as suggested by duncan | Chris Lattner | 2011-01-16 | 1 | -2/+2 | |
| | | | | | llvm-svn: 123590 | |||||
| * | Only put unnamed_addr constants in mergeable sections. Fixes PR8297. | Rafael Espindola | 2011-01-16 | 1 | -1/+1 | |
| | | | | | llvm-svn: 123585 | |||||
| * | Don't merge two constants if we care about the address of both. | Rafael Espindola | 2011-01-16 | 4 | -23/+105 | |
| | | | | | | | | | | | | | | | This fixes the original testcase in PR8927. It also causes a clang binary built with a patched clang to increase in size by 0.21%. We can probably get some of the size back by writing a pass that detects that a global never has its pointer compared and adds unnamed_addr to it (maybe extend global opt). It is also possible that there are some other cases clang could add unnamed_addr to. I will investigate extending globalopt next. llvm-svn: 123584 | |||||
| * | Simplify the construction and destruction of Uses. Simplify | Jay Foad | 2011-01-16 | 4 | -35/+26 | |
| | | | | | | | User::dropHungOffUses(). llvm-svn: 123580 | |||||
| * | Reduce and merge testcases. | Owen Anderson | 2011-01-16 | 2 | -60/+16 | |
| | | | | | llvm-svn: 123579 | |||||
| * | fix PR8514, a bug where the "heroic" transformation of shift/and | Chris Lattner | 2011-01-16 | 2 | -13/+45 | |
| | | | | | | | | | into and/shift would cause nodes to move around and a dangling pointer to happen. The code tried to avoid this with a HandleSDNode, but got the details wrong. llvm-svn: 123578 | |||||

