| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Refactor all the addressing mode selection stuff into the isel lowering | Chris Lattner | 2006-11-08 | 3 | -229/+325 |
| | | | | | | | class, where it can be used for preinc formation. llvm-svn: 31536 | ||||
| * | correct the (currently unused) pattern for lwzu. | Chris Lattner | 2006-11-08 | 1 | -2/+8 |
| | | | | | llvm-svn: 31535 | ||||
| * | preincrement case we miss. xfail until we catch it | Chris Lattner | 2006-11-08 | 1 | -0/+11 |
| | | | | | llvm-svn: 31534 | ||||
| * | Beautify. | Devang Patel | 2006-11-08 | 1 | -19/+17 |
| | | | | | llvm-svn: 31533 | ||||
| * | Fix the section headings. | Reid Spencer | 2006-11-08 | 1 | -29/+38 |
| | | | | | llvm-svn: 31532 | ||||
| * | For PR950: | Reid Spencer | 2006-11-08 | 1 | -32/+404 |
| | | | | | | | | Early commit of the documentation for new conversion instructions for easier review. llvm-svn: 31531 | ||||
| * | optimize single MBB loops better. In particular, produce: | Chris Lattner | 2006-11-08 | 1 | -0/+17 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | LBB1_57: #bb207.i movl 72(%esp), %ecx movb (%ecx,%eax), %cl movl 80(%esp), %edx movb %cl, 1(%edx,%eax) incl %eax cmpl $143, %eax jne LBB1_57 #bb207.i jmp LBB1_64 #cond_next255.i intead of: LBB1_57: #bb207.i movl 72(%esp), %ecx movb (%ecx,%eax), %cl movl 80(%esp), %edx movb %cl, 1(%edx,%eax) incl %eax cmpl $143, %eax je LBB1_64 #cond_next255.i jmp LBB1_57 #bb207.i This eliminates a branch per iteration of the loop. This hurted PPC particularly, because the extra branch meant another dispatch group for each iteration of the loop. llvm-svn: 31530 | ||||
| * | Beautify. | Devang Patel | 2006-11-08 | 1 | -16/+17 |
| | | | | | | | Clarify comments. llvm-svn: 31529 | ||||
| * | Update new pass managers to use PassManagerAnalysisHelper API. | Devang Patel | 2006-11-07 | 1 | -6/+21 |
| | | | | | llvm-svn: 31526 | ||||
| * | Derive new pass managers from PassManagerAnalysisHelper. | Devang Patel | 2006-11-07 | 1 | -4/+8 |
| | | | | | llvm-svn: 31525 | ||||
| * | scalarrepl should not split the two elements of the vsiidx array: | Chris Lattner | 2006-11-07 | 1 | -3/+7 |
| | | | | | | | | | | | | | | | | int func(vFloat v0, vFloat v1) { int ii; vSInt32 vsiidx[2]; vsiidx[0] = _mm_cvttps_epi32(v0); vsiidx[1] = _mm_cvttps_epi32(v1); ii = ((int *) vsiidx)[4]; return ii; } This fixes Transforms/ScalarRepl/2006-11-07-InvalidArrayPromote.ll llvm-svn: 31524 | ||||
| * | scalarrepl should not split the two elements of the vsiidx array: | Chris Lattner | 2006-11-07 | 1 | -0/+19 |
| | | | | | | | | | | | | | | int func(vFloat v0, vFloat v1) { int ii; vSInt32 vsiidx[2]; vsiidx[0] = _mm_cvttps_epi32(v0); vsiidx[1] = _mm_cvttps_epi32(v1); ii = ((int *) vsiidx)[4]; return ii; } llvm-svn: 31523 | ||||
| * | Introduce PassManagerAnalysisHelper. | Devang Patel | 2006-11-07 | 2 | -0/+70 |
| | | | | | llvm-svn: 31522 | ||||
| * | Add PassManager_New. | Devang Patel | 2006-11-07 | 2 | -0/+75 |
| | | | | | llvm-svn: 31521 | ||||
| * | New test case. | Evan Cheng | 2006-11-07 | 1 | -2/+9 |
| | | | | | llvm-svn: 31520 | ||||
| * | Fixed a bug which causes x86 be to incorrectly match | Evan Cheng | 2006-11-07 | 3 | -5/+30 |
| | | | | | | | | | | | | shuffle v, undef, <2, ?, 3, ?> to movhlps It should match to unpckhps instead. Added proper matching code for shuffle v, undef, <2, 3, 2, 3> llvm-svn: 31519 | ||||
| * | Fix comment. | Devang Patel | 2006-11-07 | 1 | -1/+1 |
| | | | | | llvm-svn: 31518 | ||||
| * | Add ModulePassManager_New. | Devang Patel | 2006-11-07 | 2 | -1/+77 |
| | | | | | llvm-svn: 31517 | ||||
| * | Accidently reran commit. | Jim Laskey | 2006-11-07 | 1 | -0/+3 |
| | | | | | llvm-svn: 31516 | ||||
| * | Add FunctionPassManager_New. | Devang Patel | 2006-11-07 | 2 | -0/+91 |
| | | | | | llvm-svn: 31515 | ||||
| * | Missed a label map, reverting till covered. | Jim Laskey | 2006-11-07 | 1 | -3/+0 |
| | | | | | llvm-svn: 31514 | ||||
| * | Add BasicBlockPassManager_New. | Devang Patel | 2006-11-07 | 2 | -0/+79 |
| | | | | | llvm-svn: 31513 | ||||
| * | Missed a label map, reverting till covered. | Jim Laskey | 2006-11-07 | 1 | -0/+3 |
| | | | | | llvm-svn: 31512 | ||||
| * | Optionally allow comparison operations from affect DSGraphs | Andrew Lenharth | 2006-11-07 | 1 | -1/+6 |
| | | | | | llvm-svn: 31511 | ||||
| * | Allow loop detection during debug in forwarding nodes, and revert auxcall ↵ | Andrew Lenharth | 2006-11-07 | 1 | -8/+20 |
| | | | | | | | patch as it make 176.gcc untenable llvm-svn: 31510 | ||||
| * | debug type for DSA TD | Andrew Lenharth | 2006-11-07 | 1 | -1/+1 |
| | | | | | llvm-svn: 31509 | ||||
| * | 1. Add a pass to fold debug label instructions so a debug info client can detect | Jim Laskey | 2006-11-07 | 5 | -59/+124 |
| | | | | | | | | | | | | | | empty ranges. 2. Reorg how MachineDebugInfo maintains changes to debug labels. 3. Have dwarf writer use debug label info to simplify scopes and source line coorespondence. 4. Revert the merging of compile units until I can get the bugs ironed out. llvm-svn: 31507 | ||||
| * | add a note from viterbi | Chris Lattner | 2006-11-07 | 1 | -0/+29 |
| | | | | | llvm-svn: 31506 | ||||
| * | Enable improved spilling costs by default. This speeds up viterbi on x86 | Chris Lattner | 2006-11-07 | 1 | -12/+7 |
| | | | | | | | | | by 40%, FreeBench/fourinarow by 20%, and many other programs 10-25%. On PPC, this speeds up fourinarow by 18%, and probably other things as well. llvm-svn: 31504 | ||||
| * | Got a little ahead of myself here | Andrew Lenharth | 2006-11-07 | 1 | -0/+0 |
| | | | | | llvm-svn: 31502 | ||||
| * | Fix this test | Andrew Lenharth | 2006-11-07 | 1 | -14/+14 |
| | | | | | llvm-svn: 31501 | ||||
| * | Use correct value for float HUGH_VAL. | Jim Laskey | 2006-11-07 | 2 | -6/+6 |
| | | | | | llvm-svn: 31500 | ||||
| * | Added target hook for post-indexed memory ops transformation. | Evan Cheng | 2006-11-07 | 1 | -0/+10 |
| | | | | | llvm-svn: 31499 | ||||
| * | Add post-indexed load / store transformations. | Evan Cheng | 2006-11-07 | 1 | -49/+198 |
| | | | | | llvm-svn: 31498 | ||||
| * | Make some corrections to this test case. | Reid Spencer | 2006-11-07 | 1 | -4/+6 |
| | | | | | llvm-svn: 31497 | ||||
| * | Ignore the Output dir. | Reid Spencer | 2006-11-07 | 1 | -0/+1 |
| | | | | | llvm-svn: 31496 | ||||
| * | Add a test case for making sure gdb can get a stack trace with our debug | Reid Spencer | 2006-11-07 | 2 | -0/+32 |
| | | | | | | | information. llvm-svn: 31495 | ||||
| * | Add a new llcbeta option. This speeds up viterbi from 12.34 to 8.76s on | Chris Lattner | 2006-11-07 | 1 | -4/+19 |
| | | | | | | | X86. If happy, I'll enable this by default. llvm-svn: 31493 | ||||
| * | Unbreak X86/ELF Debugging. Somehow this line got lost in Jim's cleanup. | Reid Spencer | 2006-11-07 | 1 | -0/+2 |
| | | | | | llvm-svn: 31492 | ||||
| * | Bumping version number | Tanya Lattner | 2006-11-07 | 1 | -1/+1 |
| | | | | | llvm-svn: 31491 | ||||
| * | wrong bug # | Chris Lattner | 2006-11-07 | 1 | -1/+1 |
| | | | | | llvm-svn: 31488 | ||||
| * | Fix PR988 and CodeGen/Generic/2006-11-06-MemIntrinsicExpand.ll. | Chris Lattner | 2006-11-07 | 1 | -1/+1 |
| | | | | | | | The low part goes in the first operand of expandop, not the second one. llvm-svn: 31487 | ||||
| * | new testcase | Chris Lattner | 2006-11-07 | 1 | -0/+12 |
| | | | | | llvm-svn: 31486 | ||||
| * | fix encoding of BLR | Chris Lattner | 2006-11-07 | 1 | -2/+2 |
| | | | | | llvm-svn: 31485 | ||||
| * | emit TIED_TO correctly | Chris Lattner | 2006-11-07 | 1 | -2/+2 |
| | | | | | llvm-svn: 31484 | ||||
| * | Mark predicate operands as such in operand info. | Chris Lattner | 2006-11-06 | 1 | -0/+4 |
| | | | | | llvm-svn: 31483 | ||||
| * | Add a new operand flag to mark which operand is the first predicate operand | Chris Lattner | 2006-11-06 | 1 | -0/+5 |
| | | | | | | | of an M_PREDICATED instruction. llvm-svn: 31482 | ||||
| * | simplify the way operand flags and constraints are handled, making it easier | Chris Lattner | 2006-11-06 | 4 | -70/+79 |
| | | | | | | | to extend. llvm-svn: 31481 | ||||
| * | recognize ppc's blr instruction as predicated | Chris Lattner | 2006-11-06 | 3 | -0/+4 |
| | | | | | llvm-svn: 31480 | ||||
| * | add a flag so that predicated instructions can be recognized by branch | Chris Lattner | 2006-11-06 | 1 | -0/+8 |
| | | | | | | | folding llvm-svn: 31479 | ||||

