| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | new testcase | Chris Lattner | 2006-09-13 | 1 | -0/+65 | |
| | | | | | llvm-svn: 30302 | |||||
| * | Skip over first operand when determining REX prefix for two-address code. | Evan Cheng | 2006-09-13 | 1 | -7/+14 | |
| | | | | | llvm-svn: 30300 | |||||
| * | Erase dead instructions. | Nick Lewycky | 2006-09-13 | 1 | -2/+3 | |
| | | | | | llvm-svn: 30298 | |||||
| * | The sense of this branch was backwards | Chris Lattner | 2006-09-13 | 1 | -1/+1 | |
| | | | | | llvm-svn: 30296 | |||||
| * | Turn X < 0 -> TEST X,X js | Chris Lattner | 2006-09-13 | 1 | -8/+11 | |
| | | | | | llvm-svn: 30294 | |||||
| * | The sense of this branch was inverted :( | Chris Lattner | 2006-09-13 | 1 | -1/+1 | |
| | | | | | llvm-svn: 30293 | |||||
| * | Fix a ton of jit failures | Chris Lattner | 2006-09-13 | 1 | -1/+1 | |
| | | | | | llvm-svn: 30292 | |||||
| * | add shifts to addressing mode 1 | Rafael Espindola | 2006-09-13 | 6 | -32/+81 | |
| | | | | | llvm-svn: 30291 | |||||
| * | If LSR went through a lot of trouble to put constants (e.g. the addr of a global | Chris Lattner | 2006-09-13 | 1 | -0/+9 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | in a specific BB, don't undo this!). This allows us to compile CodeGen/X86/loop-hoist.ll into: _foo: xorl %eax, %eax *** movl L_Arr$non_lazy_ptr, %ecx movl 4(%esp), %edx LBB1_1: #cond_true movl %eax, (%ecx,%eax,4) incl %eax cmpl %edx, %eax jne LBB1_1 #cond_true LBB1_2: #return ret instead of: _foo: xorl %eax, %eax movl 4(%esp), %ecx LBB1_1: #cond_true *** movl L_Arr$non_lazy_ptr, %edx movl %eax, (%edx,%eax,4) incl %eax cmpl %ecx, %eax jne LBB1_1 #cond_true LBB1_2: #return ret This was noticed in 464.h264ref. This doesn't usually affect PPC, but strikes X86 all the time. llvm-svn: 30290 | |||||
| * | new testcase | Chris Lattner | 2006-09-13 | 1 | -0/+25 | |
| | | | | | llvm-svn: 30289 | |||||
| * | Fix a regression in the 32-bit port from the 64-bit port landing. | Chris Lattner | 2006-09-13 | 1 | -2/+2 | |
| | | | | | | | | | | | | | | | | | | | | | | We now compile CodeGen/X86/lea-2.ll into: _test: movl 4(%esp), %eax movl 8(%esp), %ecx leal -5(%ecx,%eax,4), %eax ret instead of: _test: movl 4(%esp), %eax leal (,%eax,4), %eax addl 8(%esp), %eax addl $4294967291, %eax ret llvm-svn: 30288 | |||||
| * | New testcase | Chris Lattner | 2006-09-13 | 1 | -0/+10 | |
| | | | | | llvm-svn: 30287 | |||||
| * | new note | Chris Lattner | 2006-09-13 | 1 | -0/+32 | |
| | | | | | llvm-svn: 30286 | |||||
| * | new note | Chris Lattner | 2006-09-13 | 1 | -0/+13 | |
| | | | | | llvm-svn: 30285 | |||||
| * | Compile X << 1 (where X is a long-long) to: | Chris Lattner | 2006-09-13 | 1 | -7/+24 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | addl %ecx, %ecx adcl %eax, %eax instead of: movl %ecx, %edx addl %edx, %edx shrl $31, %ecx addl %eax, %eax orl %ecx, %eax and to: addc r5, r5, r5 adde r4, r4, r4 instead of: slwi r2,r9,1 srwi r0,r11,31 slwi r3,r11,1 or r2,r0,r2 on PPC. llvm-svn: 30284 | |||||
| * | Compile X > -1 -> text X,X; js dest | Chris Lattner | 2006-09-13 | 2 | -28/+23 | |
| | | | | | | | This implements CodeGen/X86/jump_sign.ll. llvm-svn: 30283 | |||||
| * | new testcase | Chris Lattner | 2006-09-13 | 1 | -0/+20 | |
| | | | | | llvm-svn: 30282 | |||||
| * | Initialize DontInternalize. | Devang Patel | 2006-09-13 | 1 | -1/+2 | |
| | | | | | llvm-svn: 30281 | |||||
| * | Reflects MachineConstantPoolEntry changes. | Evan Cheng | 2006-09-12 | 8 | -11/+12 | |
| | | | | | llvm-svn: 30279 | |||||
| * | Added support for machine specific constantpool values. These are useful for | Evan Cheng | 2006-09-12 | 11 | -61/+275 | |
| | | | | | | | representing expressions that can only be resolved at link time, etc. llvm-svn: 30278 | |||||
| * | Reflect MachineConstantPoolEntry changes. | Evan Cheng | 2006-09-12 | 1 | -3/+12 | |
| | | | | | llvm-svn: 30277 | |||||
| * | Correct the URL to the attrib.incl file. | John Criswell | 2006-09-12 | 1 | -1/+1 | |
| | | | | | llvm-svn: 30276 | |||||
| * | An sinkable instruction may exist with uses, if those uses are in dead blocks. | Chris Lattner | 2006-09-12 | 1 | -0/+4 | |
| | | | | | | | Handle this. This fixes PR908 and Transforms/LICM/2006-09-12-DeadUserOfSunkInstr.ll llvm-svn: 30275 | |||||
| * | testcase for PR908 | Chris Lattner | 2006-09-12 | 1 | -0/+215 | |
| | | | | | llvm-svn: 30274 | |||||
| * | Testcase that crashes the C++ FE. | Chris Lattner | 2006-09-12 | 1 | -0/+28 | |
| | | | | | llvm-svn: 30273 | |||||
| * | add a note | Chris Lattner | 2006-09-12 | 1 | -0/+14 | |
| | | | | | llvm-svn: 30271 | |||||
| * | Add ability to remove nodes from DominatorTree, for when a BasicBlock | Nick Lewycky | 2006-09-12 | 1 | -0/+8 | |
| | | | | | | | is being removed. llvm-svn: 30270 | |||||
| * | Testcase noticed from PR906 | Chris Lattner | 2006-09-11 | 1 | -0/+34 | |
| | | | | | llvm-svn: 30269 | |||||
| * | add compilable testcase | Chris Lattner | 2006-09-11 | 1 | -1/+6 | |
| | | | | | llvm-svn: 30268 | |||||
| * | new testcase for PR906 | Chris Lattner | 2006-09-11 | 1 | -0/+12 | |
| | | | | | llvm-svn: 30267 | |||||
| * | Fix PR905 and InstCombine/2006-09-11-EmptyStructCrash.ll | Chris Lattner | 2006-09-11 | 1 | -1/+2 | |
| | | | | | llvm-svn: 30266 | |||||
| * | new testcase for PR905 | Chris Lattner | 2006-09-11 | 1 | -0/+47 | |
| | | | | | llvm-svn: 30265 | |||||
| * | implement SRL and MUL | Rafael Espindola | 2006-09-11 | 1 | -0/+9 | |
| | | | | | llvm-svn: 30262 | |||||
| * | add the correct fallback for ARMDAGToDAGISel::SelectAddrMode1 | Rafael Espindola | 2006-09-11 | 1 | -6/+3 | |
| | | | | | llvm-svn: 30261 | |||||
| * | burg is gone | Chris Lattner | 2006-09-11 | 1 | -1/+0 | |
| | | | | | llvm-svn: 30253 | |||||
| * | partial implementation of the ARM Addressing Mode 1 | Rafael Espindola | 2006-09-11 | 5 | -37/+74 | |
| | | | | | llvm-svn: 30252 | |||||
| * | Skip the linear search if the answer is already known. | Nick Lewycky | 2006-09-11 | 1 | -20/+22 | |
| | | | | | llvm-svn: 30251 | |||||
| * | No longer used, and confuses maintainers | Jim Laskey | 2006-09-11 | 1 | -876/+0 | |
| | | | | | llvm-svn: 30248 | |||||
| * | Include the UIUC CS department attribution on the main doxygen page. | John Criswell | 2006-09-11 | 1 | -0/+4 | |
| | | | | | llvm-svn: 30247 | |||||
| * | call AsmPrinter::doInitialization in ARMAsmPrinter::doInitialization | Rafael Espindola | 2006-09-11 | 1 | -0/+1 | |
| | | | | | llvm-svn: 30246 | |||||
| * | Updates. | Evan Cheng | 2006-09-11 | 2 | -154/+0 | |
| | | | | | llvm-svn: 30245 | |||||
| * | Update README file. | Evan Cheng | 2006-09-11 | 1 | -98/+3 | |
| | | | | | llvm-svn: 30244 | |||||
| * | Rename a test-not-cmp.ll to cmp-test.ll; add a test. | Evan Cheng | 2006-09-11 | 2 | -14/+28 | |
| | | | | | llvm-svn: 30243 | |||||
| * | 1) With X86 lowering change, the following can no longer happen since | Evan Cheng | 2006-09-11 | 1 | -19/+16 | |
| | | | | | | | | | | | | | | | | | | | | | | the branch's chain is also produced by cmp. [ch, r : ld] ^ ^ | | [XX]--/ \- [flag : cmp] ^ ^ | | \---[br flag]- Remove an isel check which prevents loads from being folded into cmp / test instructions. 2) Whenever possible, delete a selected node to allow more load folding opportunities. Note not all nodes can be deleted after it has been selected. Some may have simply morphed; some have not changed at all (e.g. EntryToken). llvm-svn: 30242 | |||||
| * | X86ISD::CMP now produces a chain as well as a flag. Make that the chain | Evan Cheng | 2006-09-11 | 5 | -157/+131 | |
| | | | | | | | | operand of a conditional branch to allow load folding into CMP / TEST instructions. llvm-svn: 30241 | |||||
| * | Behold, more work on relocations. Things are looking pretty good now. | Nate Begeman | 2006-09-10 | 5 | -90/+213 | |
| | | | | | llvm-svn: 30240 | |||||
| * | Removed unnecessary Mangler creation. | Anton Korobeynikov | 2006-09-10 | 1 | -1/+0 | |
| | | | | | llvm-svn: 30239 | |||||
| * | add a testcase for this commit: ↵ | Chris Lattner | 2006-09-10 | 1 | -0/+25 | |
| | | | | | | | http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20050606/026673.html llvm-svn: 30238 | |||||
| * | Allow tail duplication in more cases, relaxing the previous restriction a | Chris Lattner | 2006-09-10 | 1 | -1/+12 | |
| | | | | | | | bit. This fixes Regression/Transforms/TailDup/MergeTest.ll llvm-svn: 30237 | |||||
| * | This code was trying too hard. By eliminating redundant edges in the CFG | Chris Lattner | 2006-09-10 | 1 | -6/+3 | |
| | | | | | | | | | | due to switch cases going to the same place, it make #pred != #phi entries, breaking live interval analysis. This fixes 458.sjeng on x86 with llc. llvm-svn: 30236 | |||||

