Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Patches to build EFI with Clang/LLVM. By Carl Norum. | Evan Cheng | 2011-02-01 | 8 | -17/+41 |
| | | | | llvm-svn: 124639 | ||||
* | Remove stale test that has never worked, afaik. | Devang Patel | 2011-02-01 | 1 | -60/+0 |
| | | | | llvm-svn: 124635 | ||||
* | Don't tell the linker about available_externally definitions. If we do, it will | Rafael Espindola | 2011-02-01 | 1 | -0/+4 |
| | | | | | | complain about duplicated definitions. llvm-svn: 124634 | ||||
* | Keep track of incoming argument's location while emitting LiveIns. | Devang Patel | 2011-01-31 | 13 | -32/+144 |
| | | | | llvm-svn: 124611 | ||||
* | While printing "interesting" breakpoint locations for debug info quality ↵ | Devang Patel | 2011-01-31 | 1 | -27/+12 |
| | | | | | | test harness, focus only on entry block's terminator for now. llvm-svn: 124610 | ||||
* | Enumerate .code16/32/64 instead of checking .code prefix. This | Roman Divacky | 2011-01-31 | 1 | -1/+1 |
| | | | | | | unbreaks some ARM tests. llvm-svn: 124608 | ||||
* | Tidy up. | Devang Patel | 2011-01-31 | 1 | -35/+132 |
| | | | | llvm-svn: 124605 | ||||
* | Focus on arguments for now. | Devang Patel | 2011-01-31 | 1 | -2/+2 |
| | | | | llvm-svn: 124604 | ||||
* | Error on all .code* directives instead of just .code16 as they | Roman Divacky | 2011-01-31 | 1 | -2/+2 |
| | | | | | | all lead to a silent miscompilation of code. llvm-svn: 124603 | ||||
* | Fix vector sign extend to put the source and destination types in the | David Greene | 2011-01-31 | 1 | -3/+3 |
| | | | | | | correct places. llvm-svn: 124601 | ||||
* | add a note, progress unblocked by PR8575 being fixed. | Chris Lattner | 2011-01-31 | 1 | -0/+48 |
| | | | | llvm-svn: 124599 | ||||
* | Adds some platform checks to cmake/config-ix.cmake and fixes checking | Oscar Fuentes | 2011-01-31 | 2 | -12/+36 |
| | | | | | | | | for dlopen/dlerror. Patch by arrowdodger! llvm-svn: 124590 | ||||
* | Fix bug where ReduceLoadWidth was creating illegal ZEXTLOAD instructions. | Richard Osborne | 2011-01-31 | 2 | -2/+12 |
| | | | | llvm-svn: 124587 | ||||
* | CMake: Fix a few definition to suppress warnings. | NAKAMURA Takumi | 2011-01-31 | 1 | -3/+3 |
| | | | | llvm-svn: 124583 | ||||
* | Save a mapping between original and cloned constpool entries. | Anton Korobeynikov | 2011-01-30 | 2 | -0/+25 |
| | | | | llvm-svn: 124570 | ||||
* | Clarify the LSDASection NULL check | Anton Korobeynikov | 2011-01-30 | 1 | -0/+2 |
| | | | | llvm-svn: 124569 | ||||
* | Recognize and simplify | Anders Carlsson | 2011-01-30 | 2 | -1/+27 |
| | | | | | | | (A+B) == A -> B == 0 A == (A+B) -> B == 0 llvm-svn: 124567 | ||||
* | Respect the -tail-dup-size command line option even when optimizing for size. | Jakob Stoklund Olesen | 2011-01-30 | 1 | -1/+2 |
| | | | | | | | This is similar to the -unroll-threshold option. There should be no change in behavior when -tail-dup-size is not explicit on the llc command line. llvm-svn: 124564 | ||||
* | Commit 124487 broke 254.gap. See if disabling the part that might be triggered | Duncan Sands | 2011-01-30 | 2 | -9/+5 |
| | | | | | | by PR9088 fixes things. llvm-svn: 124561 | ||||
* | Transform (X/Y)*Y into X if the division is exact. Instcombine already ↵ | Duncan Sands | 2011-01-30 | 2 | -0/+27 |
| | | | | | | | | | | | | knows how to do this and more, but would only do it if X/Y had only one use. Spotted as the most common missed simplification in SPEC by my auto-simplifier, now that it knows about nuw/nsw/exact flags. This removes a bunch of multiplications from 447.dealII and 483.xalancbmk. It also removes a lot from tramp3d-v4, which results in much more inlining. llvm-svn: 124560 | ||||
* | Teach DAGCombine to fold fold (sra (trunc (sr x, c1)), c2) -> (trunc (sra x, ↵ | Benjamin Kramer | 2011-01-30 | 3 | -21/+32 |
| | | | | | | | | | | | | | | | | | | c1+c2) when c1 equals the amount of bits that are truncated off. This happens all the time when a smul is promoted to a larger type. On x86-64 we now compile "int test(int x) { return x/10; }" into movslq %edi, %rax imulq $1717986919, %rax, %rax movq %rax, %rcx shrq $63, %rcx sarq $34, %rax <- used to be "shrq $32, %rax; sarl $2, %eax" addl %ecx, %eax This fires 96 times in gcc.c on x86-64. llvm-svn: 124559 | ||||
* | Fix 'fcmp one' constant folding. Noticed by inspection. | Nick Lewycky | 2011-01-30 | 1 | -1/+1 |
| | | | | llvm-svn: 124557 | ||||
* | Fix some formatting and upgrade comments from llvm 1.x to 2.x syntax. | Nick Lewycky | 2011-01-30 | 1 | -6/+6 |
| | | | | llvm-svn: 124556 | ||||
* | This is #included by .c files. Remove C++-style comments. | Bill Wendling | 2011-01-29 | 1 | -1/+1 |
| | | | | llvm-svn: 124552 | ||||
* | Add the select optimization recently added to instcombine to constant folding. | Nick Lewycky | 2011-01-29 | 2 | -0/+21 |
| | | | | | | | This is the one where one of the branches of the select is another select on the same condition. llvm-svn: 124547 | ||||
* | Unbreak the MSVC build. | Francois Pichet | 2011-01-29 | 1 | -0/+1 |
| | | | | | | The DEBUG() call at line 606 demands to see raw_ostream's definition. I have no idea why this seems to only break MSVC. llvm-svn: 124545 | ||||
* | Fix comment. | Nick Lewycky | 2011-01-29 | 1 | -1/+1 |
| | | | | llvm-svn: 124544 | ||||
* | Call SimplifyFDivInst() in InstCombiner::visitFDiv(). | Frits van Bommel | 2011-01-29 | 2 | -0/+10 |
| | | | | llvm-svn: 124535 | ||||
* | Move InstCombine's knowledge of fdiv to SimplifyInstruction(). | Frits van Bommel | 2011-01-29 | 5 | -18/+47 |
| | | | | llvm-svn: 124534 | ||||
* | Fix typo: should have been testing that X was odd, not V. | Duncan Sands | 2011-01-29 | 2 | -3/+27 |
| | | | | llvm-svn: 124533 | ||||
* | Add the missing sub identity "A-(A-B) -> B" to DAGCombine. | Benjamin Kramer | 2011-01-29 | 1 | -0/+3 |
| | | | | | | | | This happens e.g. for code like "X - X%10" where we lower the modulo operation to a series of multiplies and shifts that are then subtracted from X, leading to this missed optimization. llvm-svn: 124532 | ||||
* | Add a test for TCE return duplication. | Evan Cheng | 2011-01-29 | 2 | -0/+25 |
| | | | | llvm-svn: 124527 | ||||
* | Re-apply r124518 with fix. Watch out for invalidated iterator. | Evan Cheng | 2011-01-29 | 13 | -73/+226 |
| | | | | llvm-svn: 124526 | ||||
* | Revert r124518. It broke Linux self-host. | Evan Cheng | 2011-01-29 | 13 | -223/+73 |
| | | | | llvm-svn: 124522 | ||||
* | Re-commit r124462 with fixes. Tail recursion elim will now dup ret into ↵ | Evan Cheng | 2011-01-29 | 13 | -73/+223 |
| | | | | | | unconditional predecessor to enable TCE on demand. llvm-svn: 124518 | ||||
* | Implementation of path profiling. | Andrew Trick | 2011-01-29 | 23 | -51/+3423 |
| | | | | | | | | | | Modified patch by Adam Preuss. This builds on the existing framework for block tracing, edge profiling and optimal edge profiling. See -help-hidden for new flags. For documentation, see the technical report "Implementation of Path Profiling..." in llvm.org/pubs. llvm-svn: 124515 | ||||
* | Most browsers eliminate whitespace between anchor tags. Force whitespace with | Nick Lewycky | 2011-01-29 | 1 | -1/+1 |
| | | | | | | so that the code reads properly. llvm-svn: 124514 | ||||
* | Error on .code16 instead of producing wrong (32bit) code. | Roman Divacky | 2011-01-28 | 1 | -0/+3 |
| | | | | llvm-svn: 124498 | ||||
* | This dyn_cast should be a cast. Pointed out by Frits van Bommel. | Duncan Sands | 2011-01-28 | 1 | -1/+1 |
| | | | | llvm-svn: 124497 | ||||
* | Thread divisions over selects and phis. This doesn't fire much and has ↵ | Duncan Sands | 2011-01-28 | 1 | -9/+19 |
| | | | | | | | | basically zero effect on the testsuite (it improves two Ada testcases). llvm-svn: 124496 | ||||
* | PR9030: Fix disassembly of ARM "mov pc, lr" instruction. | Bob Wilson | 2011-01-28 | 2 | -2/+5 |
| | | | | | | Patch by Jyun-Yan You. llvm-svn: 124492 | ||||
* | My auto-simplifier noticed that ((X/Y)*Y)/Y occurs several times in SPEC | Duncan Sands | 2011-01-28 | 5 | -61/+235 |
| | | | | | | | | | | | | | | | | | benchmarks, and that it can be simplified to X/Y. (In general you can only simplify (Z*Y)/Y to Z if the multiplication did not overflow; if Z has the form "X/Y" then this is the case). This patch implements that transform and moves some Div logic out of instcombine and into InstructionSimplify. Unfortunately instcombine gets in the way somewhat, since it likes to change (X/Y)*Y into X-(X rem Y), so I had to teach instcombine about this too. Finally, thanks to the NSW/NUW flags, sometimes we know directly that "Z*Y" does not overflow, because the flag says so, so I added that logic too. This eliminates a bunch of divisions and subtractions in 447.dealII, and has good effects on some other benchmarks too. It seems to have quite an effect on tramp3d-v4 but it's hard to say if it's good or bad because inlining decisions changed, resulting in massive changes all over. llvm-svn: 124487 | ||||
* | Fix libffi usage when it is on a custom path. | Oscar Fuentes | 2011-01-28 | 1 | -9/+1 |
| | | | | llvm-svn: 124486 | ||||
* | Add support for parsing .float | Roman Divacky | 2011-01-28 | 2 | -1/+4 |
| | | | | llvm-svn: 124485 | ||||
* | Rename functions to follow coding standard. Also rejiggers comments. No | Nick Lewycky | 2011-01-28 | 1 | -89/+88 |
| | | | | | | functionality change. llvm-svn: 124482 | ||||
* | Add a doxygen comment for this class. | Nick Lewycky | 2011-01-28 | 1 | -0/+2 |
| | | | | llvm-svn: 124480 | ||||
* | Reorder for readability. (Chris, is this what you meant?) | Nick Lewycky | 2011-01-28 | 1 | -148/+150 |
| | | | | llvm-svn: 124479 | ||||
* | Revert r124462. There are a few big regressions that I need to fix first. | Evan Cheng | 2011-01-28 | 10 | -78/+32 |
| | | | | llvm-svn: 124478 | ||||
* | Reduce the number of functions we look at in the first pass, and preallocate | Nick Lewycky | 2011-01-28 | 1 | -1/+3 |
| | | | | | | the function equality set. llvm-svn: 124475 | ||||
* | Clean up the tests a little, make sure we match an instruction in the right | Nick Lewycky | 2011-01-28 | 1 | -5/+4 |
| | | | | | | test. llvm-svn: 124473 |