| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Implement "superhero" strength reduction, or full strength | Dan Gohman | 2009-02-20 | 2 | -1/+36 |
| * | GV with null value initializer shouldn't go to BSS if it's meant for a mergea... | Evan Cheng | 2009-02-18 | 1 | -0/+4 |
| * | Change these tests to use regular loads instead of llvm.x86.sse2.loadu.dq. | Dan Gohman | 2009-02-16 | 2 | -18/+14 |
| * | Fix PR3522. It's not safe to sink into landing pad BB's. | Evan Cheng | 2009-02-15 | 1 | -0/+30 |
| * | Teach x86 target -soft-float. | Evan Cheng | 2009-02-13 | 1 | -0/+27 |
| * | Fix the code that checked if a SCEVAddRecExpr Start contains an | Dan Gohman | 2009-02-13 | 1 | -0/+292 |
| * | Fix LSR's IV sorting function to explicitly sort by bitwidth | Dan Gohman | 2009-02-13 | 1 | -0/+22 |
| * | Arrange to print constants that match "n" and "i" constraints | Dale Johannesen | 2009-02-12 | 1 | -0/+18 |
| * | fix the X86 backend to just drop llvm.declare nodes for VLAs instead of | Chris Lattner | 2009-02-12 | 1 | -0/+76 |
| * | add PR | Chris Lattner | 2009-02-12 | 1 | -0/+1 |
| * | It's (currently) not safe to keep certain physical registers live across basi... | Evan Cheng | 2009-02-12 | 1 | -0/+28 |
| * | Replace one of burr scheduling heuristic with something more sensible. Now ca... | Evan Cheng | 2009-02-12 | 1 | -1/+1 |
| * | fix PR3537: if resetting bbi back to the start of a block, we need to | Chris Lattner | 2009-02-12 | 1 | -0/+35 |
| * | rename test to avoid messing with tab completion of dates. | Chris Lattner | 2009-02-12 | 1 | -0/+0 |
| * | Remove a bogus assertion. It's possible a live-in available value is used by ... | Evan Cheng | 2009-02-11 | 1 | -0/+49 |
| * | Don't use special heuristics for nodes with no data predecessors | Dan Gohman | 2009-02-11 | 4 | -25/+26 |
| * | Make a transformation added in 63266 a bit less aggressive. | Dale Johannesen | 2009-02-11 | 1 | -0/+70 |
| * | Implement PR3495: local spiller optimization. The local spiller can now keep ... | Evan Cheng | 2009-02-11 | 1 | -0/+78 |
| * | Handle llvm.x86.sse2.maskmov.dqu in 64-bit. | Evan Cheng | 2009-02-10 | 1 | -0/+11 |
| * | Fix PR3457: Ignore control successors when looking for closest scheduled succ... | Evan Cheng | 2009-02-10 | 2 | -1/+19 |
| * | Implement FpSET_ST1_*. | Evan Cheng | 2009-02-09 | 1 | -0/+7 |
| * | Make sure constant subscript is truncated to ptr size if it may not fit. | Evan Cheng | 2009-02-09 | 1 | -0/+10 |
| * | Re-enable machine sinking pass now that the coalescer bugs and the AnalyzeBrn... | Evan Cheng | 2009-02-09 | 2 | -2/+2 |
| * | Fix PR3486. Fix a bug in code that manually patch physical register live inte... | Evan Cheng | 2009-02-08 | 1 | -0/+22 |
| * | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64073 91177308-0d34-... | Evan Cheng | 2009-02-08 | 1 | -0/+491 |
| * | Revert r63999. It was breaking self-hosting builds. | Bill Wendling | 2009-02-08 | 1 | -1/+1 |
| * | Enable machine sinking pass in non-fast mode. | Evan Cheng | 2009-02-07 | 1 | -1/+1 |
| * | Fix test. It produces unexpected code if sse4.1 is on. | Evan Cheng | 2009-02-06 | 1 | -3/+3 |
| * | isAsCheapAsMove instructions can have register src operands. Check if they ar... | Evan Cheng | 2009-02-05 | 1 | -0/+12 |
| * | Turn on machine LICM in non-fast mode. | Evan Cheng | 2009-02-05 | 1 | -3/+3 |
| * | if we have a large GEP offset on a 32-bit or other target, make | Chris Lattner | 2009-02-05 | 1 | -0/+9 |
| * | Add test case for r63760. | Mon P Wang | 2009-02-04 | 1 | -0/+15 |
| * | Fixes a case where we generate an incorrect mask for pshfhw in the presence | Mon P Wang | 2009-02-04 | 1 | -0/+23 |
| * | Fix PR3411. When replacing values, nodes are analyzed | Duncan Sands | 2009-02-03 | 1 | -0/+30 |
| * | rearrange how SRoA handles promotion of allocas to vectors. | Chris Lattner | 2009-02-03 | 1 | -1/+0 |
| * | Add explicit -march=x86 to these tests so that they don't | Dan Gohman | 2009-02-03 | 1 | -2/+2 |
| * | Fix another test to not use -mcpu=yonah with 64-bit code. | Dan Gohman | 2009-02-02 | 1 | -2/+2 |
| * | Yonah does not support x86-64. Change the -mcpu value to one that does. | Dan Gohman | 2009-02-02 | 1 | -1/+1 |
| * | xfail this for now, will fix shortly. | Chris Lattner | 2009-02-02 | 1 | -0/+1 |
| * | Teach LowerBRCOND to recognize (xor (setcc x), 1). The xor inverts the condit... | Evan Cheng | 2009-02-02 | 1 | -0/+23 |
| * | add 2 more testcases for -mattr=-sse (r63495). | Torok Edwin | 2009-02-01 | 2 | -0/+66 |
| * | Implement -mno-sse: if SSE is disabled on x86-64, don't store XMM on stack for | Torok Edwin | 2009-02-01 | 2 | -2/+46 |
| * | Fix PR3453 and probably a bunch of other potential | Duncan Sands | 2009-02-01 | 1 | -0/+32 |
| * | Fix PR3401: when using large integers, the type | Duncan Sands | 2009-01-31 | 3 | -0/+51 |
| * | If unsafe FP optimization is not set, don't allow -(A-B) => B-A because | Mon P Wang | 2009-01-31 | 1 | -0/+12 |
| * | XFAIL this test. It only worked before because of a bug in the spill point s... | Owen Anderson | 2009-01-29 | 1 | -0/+1 |
| * | Local register allocator shouldn't assume only the entry and landing pad basi... | Evan Cheng | 2009-01-29 | 1 | -0/+38 |
| * | In the case of an extractelement on an insertelement value, | Dan Gohman | 2009-01-29 | 1 | -0/+25 |
| * | Exit with nice warnings when register allocator run out of registers. | Evan Cheng | 2009-01-29 | 1 | -0/+32 |
| * | Make x86's BT instruction matching more thorough, and add some | Dan Gohman | 2009-01-29 | 2 | -4/+433 |
