Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Fix a typo in a comment. | Dan Gohman | 2008-05-02 | 1 | -1/+1 |
| | | | | llvm-svn: 50562 | ||||
* | Use push_back(...) instead of resize(1, ...), per review feedback. | Dan Gohman | 2008-05-02 | 1 | -2/+2 |
| | | | | llvm-svn: 50561 | ||||
* | Update old-style syntax in some "not grep" tests. | Dan Gohman | 2008-05-01 | 16 | -16/+16 |
| | | | | llvm-svn: 50560 | ||||
* | Fix uninitialized uses of the FPC variable. | Dan Gohman | 2008-05-01 | 1 | -2/+2 |
| | | | | llvm-svn: 50558 | ||||
* | New test for bug fixed in 50545. | Dale Johannesen | 2008-05-01 | 1 | -0/+26 |
| | | | | llvm-svn: 50548 | ||||
* | Don't try to create PHIs of struct types. Fallout | Dale Johannesen | 2008-05-01 | 1 | -8/+31 |
| | | | | | | from x86-64 calling convention work. llvm-svn: 50545 | ||||
* | Fix an overaggressive SimplifyDemandedBits optimization on urem. This | Dan Gohman | 2008-05-01 | 2 | -4/+37 |
| | | | | | | fixes the 254.gap regression on x86 and the 403.gcc regression on x86-64. llvm-svn: 50537 | ||||
* | Adding testcase. | Bill Wendling | 2008-05-01 | 1 | -0/+15 |
| | | | | llvm-svn: 50536 | ||||
* | Initialize a local variable. | Ted Kremenek | 2008-05-01 | 1 | -1/+1 |
| | | | | llvm-svn: 50527 | ||||
* | don't randomly miscompile seto/setuo just because we are in | Chris Lattner | 2008-05-01 | 2 | -4/+19 |
| | | | | | | | | | ffastmath mode. This fixes rdar://5902801, a miscompilation of gcc.dg/builtins-8.c. Bill, please pull this into Tak. llvm-svn: 50523 | ||||
* | 1) add '-debug' output | Chris Lattner | 2008-05-01 | 1 | -11/+19 |
| | | | | | | | 2) Return NULL instead of false in several places for tidiness. 3) fix a bug optimizing sprintf(p, "%c", x) llvm-svn: 50521 | ||||
* | Delete the IPO simplify-libcalls and completely reimplement it as | Chris Lattner | 2008-05-01 | 4 | -2219/+1443 |
| | | | | | | | | a FunctionPass. This makes it simpler, fixes dozens of bugs, adds a couple of minor features, and shrinks is considerably: from 2214 to 1437 lines. llvm-svn: 50520 | ||||
* | fix typo | Chris Lattner | 2008-05-01 | 1 | -1/+1 |
| | | | | llvm-svn: 50519 | ||||
* | instcombine does memset optzns. | Chris Lattner | 2008-05-01 | 1 | -18/+0 |
| | | | | llvm-svn: 50518 | ||||
* | simplifylibcalls doesn't optimize llvm.memmove, instcombine does. | Chris Lattner | 2008-05-01 | 1 | -22/+0 |
| | | | | llvm-svn: 50517 | ||||
* | move some tests from libcall optimizer suite. | Chris Lattner | 2008-05-01 | 1 | -0/+17 |
| | | | | llvm-svn: 50516 | ||||
* | Add CreateCall3/CreateCall4 at Eric's request. | Chris Lattner | 2008-05-01 | 1 | -0/+10 |
| | | | | llvm-svn: 50515 | ||||
* | Add a spiffy little "CreateCall2" method, which can be used to make | Chris Lattner | 2008-05-01 | 1 | -3/+8 |
| | | | | | | a function call that takes two Value*'s as arguments. llvm-svn: 50514 | ||||
* | fix a bug in my previous patch, a classic =/== bug. | Chris Lattner | 2008-04-30 | 1 | -1/+1 |
| | | | | llvm-svn: 50483 | ||||
* | Really commit the test checking the argument lowering behaviour on x86-64 :). | Arnold Schwaighofer | 2008-04-30 | 1 | -0/+21 |
| | | | | llvm-svn: 50478 | ||||
* | Tail call optimization improvements: | Arnold Schwaighofer | 2008-04-30 | 16 | -294/+985 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Move platform independent code (lowering of possibly overwritten arguments, check for tail call optimization eligibility) from target X86ISelectionLowering.cpp to TargetLowering.h and SelectionDAGISel.cpp. Initial PowerPC tail call implementation: Support ppc32 implemented and tested (passes my tests and test-suite llvm-test). Support ppc64 implemented and half tested (passes my tests). On ppc tail call optimization is performed if caller and callee are fastcc call is a tail call (in tail call position, call followed by ret) no variable argument lists or byval arguments option -tailcallopt is enabled Supported: * non pic tail calls on linux/darwin * module-local tail calls on linux(PIC/GOT)/darwin(PIC) * inter-module tail calls on darwin(PIC) If constraints are not met a normal call will be emitted. A test checking the argument lowering behaviour on x86-64 was added. llvm-svn: 50477 | ||||
* | fcntl.h is pretty standard on unix (without the sys/) | Gabor Greif | 2008-04-30 | 1 | -2/+1 |
| | | | | llvm-svn: 50475 | ||||
* | Move this test to LoopDeletion, where it now passes. | Owen Anderson | 2008-04-30 | 2 | -2/+4 |
| | | | | llvm-svn: 50474 | ||||
* | This condition got inverted accidentally. | Owen Anderson | 2008-04-30 | 1 | -1/+1 |
| | | | | llvm-svn: 50473 | ||||
* | move lowering of llvm.memset -> store from simplify libcalls | Chris Lattner | 2008-04-30 | 2 | -7/+64 |
| | | | | | | to instcombine. llvm-svn: 50472 | ||||
* | no reason for simplifylibcalls to simplify intrinsics, instcombine does | Chris Lattner | 2008-04-30 | 1 | -3/+1 |
| | | | | | | a fine job. llvm-svn: 50470 | ||||
* | remove redundant check. | Chris Lattner | 2008-04-30 | 1 | -3/+1 |
| | | | | llvm-svn: 50469 | ||||
* | add missing #include | Chris Lattner | 2008-04-30 | 1 | -0/+2 |
| | | | | llvm-svn: 50468 | ||||
* | add a method for comparing to see if a value has a specified name. | Chris Lattner | 2008-04-30 | 2 | -0/+11 |
| | | | | llvm-svn: 50465 | ||||
* | use string length computation to generalize several xforms. | Chris Lattner | 2008-04-30 | 1 | -36/+39 |
| | | | | llvm-svn: 50464 | ||||
* | Add comments for previous patch as requested. | Dale Johannesen | 2008-04-30 | 1 | -0/+6 |
| | | | | llvm-svn: 50463 | ||||
* | Bug fixes and updates for CellSPU, syncing up with trunk. Most notable | Scott Michel | 2008-04-30 | 7 | -53/+133 |
| | | | | | | | | fixes are target-specific lowering of frame indices, fix constants generated for the FSMBI instruction, and fixing SPUTargetLowering::computeMaskedBitsFor- TargetNode(). llvm-svn: 50462 | ||||
* | Fix custom target lowering for zero/any/sign_extend: make sure that | Scott Michel | 2008-04-30 | 1 | -5/+3 |
| | | | | | | | DAG.UpdateNodeOperands() is called before (not after) the call to TLI.LowerOperation(). llvm-svn: 50461 | ||||
* | Make eh_frame objects by 8-byte aligned on 64-bit | Dale Johannesen | 2008-04-29 | 1 | -2/+4 |
| | | | | | | targets. llvm-svn: 50451 | ||||
* | Minor spelling and typo fixes. | John Criswell | 2008-04-29 | 1 | -4/+4 |
| | | | | llvm-svn: 50448 | ||||
* | Revert r50441. The original code was correct. Add some more comments so ↵ | Owen Anderson | 2008-04-29 | 1 | -3/+9 |
| | | | | | | that I don't make the same mistake in the future. llvm-svn: 50446 | ||||
* | Fix a bug in memcpyopt where the memcpy-memcpy transform was never being ↵ | Owen Anderson | 2008-04-29 | 2 | -4/+23 |
| | | | | | | | | | applied because we were checking for it in the wrong order. This caused a miscompilation because the return slot optimization assumes that the call it is dealing with is NOT a memcpy. llvm-svn: 50444 | ||||
* | We should be returning true here since we've changed the function. | Owen Anderson | 2008-04-29 | 1 | -1/+1 |
| | | | | llvm-svn: 50442 | ||||
* | A lot of cleanups and documentation improvements, as well as a few corner ↵ | Owen Anderson | 2008-04-29 | 1 | -59/+76 |
| | | | | | | | | case fixes. Most of this was suggested by Chris. llvm-svn: 50441 | ||||
* | Rename DeadLoopElimination to LoopDeletion, part 2. | Owen Anderson | 2008-04-29 | 3 | -17/+15 |
| | | | | llvm-svn: 50437 | ||||
* | Rename DeadLoopElimination to LoopDeletion, part one. | Owen Anderson | 2008-04-29 | 1 | -0/+0 |
| | | | | llvm-svn: 50436 | ||||
* | Don't do stupid things: doInitialization(Module&) is not applicable to ↵ | Anton Korobeynikov | 2008-04-29 | 1 | -17/+2 |
| | | | | | | ModulePass :) llvm-svn: 50433 | ||||
* | don't eliminate load from volatile value on paths where the load is dead. | Chris Lattner | 2008-04-29 | 2 | -0/+34 |
| | | | | | | This fixes the second half of PR2262 llvm-svn: 50430 | ||||
* | make this test reduced and *valid* | Chris Lattner | 2008-04-29 | 1 | -15/+6 |
| | | | | llvm-svn: 50429 | ||||
* | fix a subtle volatile handling bug. | Chris Lattner | 2008-04-29 | 2 | -7/+42 |
| | | | | llvm-svn: 50428 | ||||
* | Use std::set instead of std::priority_queue for the RegReductionPriorityQueue. | Roman Levenstein | 2008-04-29 | 2 | -45/+31 |
| | | | | | | | | | | | | | | This removes the existing bottleneck related to the removal of elements from the middle of the queue. Also fixes a subtle bug in ScheduleDAGRRList::CapturePred: It was updating the state of the SUnit before removing it. As a result, the comparison operators were working incorrectly and this SUnit could not be removed from the queue properly. Reviewed by Evan and Dan. Approved by Dan. llvm-svn: 50412 | ||||
* | Implement more aggressive support for analyzing string length. This | Chris Lattner | 2008-04-29 | 1 | -10/+136 |
| | | | | | | | | | | generalizes the previous code to handle the case when the string is not an immediate to the strlen call (for example, crazy stuff like strlen(c ? "foo" : "bart"+1) -> 3). This implements gcc.c-torture/execute/builtins/strlen-2.c. I will generalize other cases in simplifylibcalls to use the same routine later. llvm-svn: 50408 | ||||
* | Clarify what we mean by a dead loop. | Owen Anderson | 2008-04-29 | 1 | -0/+4 |
| | | | | llvm-svn: 50406 | ||||
* | new testcase for PR2094. The inline asms should not pin allocas to the | Chris Lattner | 2008-04-29 | 1 | -0/+12 |
| | | | | | | stack anymore. llvm-svn: 50397 | ||||
* | don't delete the last store to an alloca if the store is volatile. | Chris Lattner | 2008-04-29 | 2 | -1/+9 |
| | | | | llvm-svn: 50390 |