2008-05-31 |
Dan Gohman | Teach the DAGISelEmitter to not compute the variable_op... |
tree | commitdiff |
2008-05-30 |
Bill Wendling | Add the "AsCheapAsAMove" flag to some 64-bit xor instru... |
tree | commitdiff |
2008-05-29 |
Dan Gohman | Add patterns for CALL32m and CALL64m. They aren't match... |
tree | commitdiff |
2008-05-29 |
Dan Gohman | Fix a tblgen problem handling variable_ops in tblgen... |
tree | commitdiff |
2008-05-29 |
Dan Gohman | Expand small memmovs using inline code. Set the X86... |
tree | commitdiff |
2008-05-29 |
Evan Cheng | Implement vector shift up / down and insert zero with... |
tree | commitdiff |
2008-05-29 |
Bill Wendling | XOR?RI instructions aren't as cheap as moves. |
tree | commitdiff |
2008-05-29 |
Bill Wendling | Implement "AsCheapAsAMove" for some obviously cheap... |
tree | commitdiff |
2008-05-28 |
Dan Gohman | Fix the encoding for two more "rm" instructions that... |
tree | commitdiff |
2008-05-28 |
Mon P Wang | Fixed X86 encoding error CVTPS2PD and CVTPD2PS when... |
tree | commitdiff |
2008-05-28 |
Nate Begeman | Don't attempt to create VZEXT_LOAD out of an extload... |
tree | commitdiff |
2008-05-24 |
Evan Cheng | Eliminate x86.sse2.punpckh.qdq and x86.sse2.punpckl... |
tree | commitdiff |
2008-05-24 |
Evan Cheng | Eliminate x86.sse2.movs.d, x86.sse2.shuf.pd, x86.sse2... |
tree | commitdiff |
2008-05-24 |
Evan Cheng | This is done. |
tree | commitdiff |
2008-05-24 |
Evan Cheng | Remove x86.sse2.loadh.pd and x86.sse2.loadl.pd. These... |
tree | commitdiff |
2008-05-23 |
Evan Cheng | Use movlps / movhps to modify low / high half of 16... |
tree | commitdiff |
2008-05-23 |
Dan Gohman | Elaborate on the entry on integer vector multiplication... |
tree | commitdiff |
2008-05-23 |
Evan Cheng | Fix a duplicated pattern. |
tree | commitdiff |
2008-05-23 |
Dan Gohman | Use PMULDQ for v2i64 multiplies when SSE4.1 is availabl... |
tree | commitdiff |
2008-05-23 |
Evan Cheng | New entry. |
tree | commitdiff |
2008-05-23 |
Chris Lattner | we compile multiply-by-constant into horrible code... |
tree | commitdiff |
2008-05-23 |
Evan Cheng | Bug: rcpps can only folds a load if the address is... |
tree | commitdiff |
2008-05-23 |
Dale Johannesen | Put const weak stuff in appropriate section on Darwin. |
tree | commitdiff |
2008-05-22 |
Evan Cheng | X86CodeEmitter should not set PIC style to None at... |
tree | commitdiff |
2008-05-22 |
Evan Cheng | Add missing patterns. |
tree | commitdiff |
2008-05-20 |
Evan Cheng | movsd and movq do not require 16-byte alignment. This... |
tree | commitdiff |
2008-05-20 |
Evan Cheng | runOnMachineFunction should set IsPIC because relocatio... |
tree | commitdiff |
2008-05-19 |
Dale Johannesen | Handle quoted names when constructing $stub's, |
tree | commitdiff |
2008-05-16 |
Dale Johannesen | Treat common as distinct from weak global on Darwin... |
tree | commitdiff |
2008-05-15 |
Evan Cheng | Fix typos and comments. |
tree | commitdiff |
2008-05-15 |
Evan Cheng | Make use of vector load and store operations to impleme... |
tree | commitdiff |
2008-05-14 |
Dale Johannesen | Add CommonLinkage; currently tentative definitions |
tree | commitdiff |
2008-05-14 |
Evan Cheng | Doh. Alignment is in bytes, not in bits. |
tree | commitdiff |
2008-05-14 |
Dan Gohman | Change target-specific classes to use more precise... |
tree | commitdiff |
2008-05-13 |
Chris Lattner | add a note |
tree | commitdiff |
2008-05-13 |
Evan Cheng | - Fix the pasto in the fix for a previous pasto. |
tree | commitdiff |
2008-05-13 |
Chris Lattner | add a note |
tree | commitdiff |
2008-05-13 |
Nate Begeman | Fix one more encoding bug. |
tree | commitdiff |
2008-05-13 |
Evan Cheng | - Don't treat anyext 16-bit load as a 32-bit load if... |
tree | commitdiff |
2008-05-13 |
Evan Cheng | Instead of a vector load, shuffle and then extract... |
tree | commitdiff |
2008-05-13 |
Nate Begeman | Fix and encoding error in the psrad xmm, imm8 instruction. |
tree | commitdiff |
2008-05-13 |
Evan Cheng | On x86, it's safe to treat i32 load anyext as a normal... |
tree | commitdiff |
2008-05-13 |
Dan Gohman | Clean up the use of static and anonymous namespaces... |
tree | commitdiff |
2008-05-12 |
Nate Begeman | Teach Legalize how to scalarize VSETCC |
tree | commitdiff |
2008-05-12 |
Evan Cheng | Xform bitconvert(build_pair(load a, load b)) to a singl... |
tree | commitdiff |
2008-05-12 |
Bill Wendling | Constify the machine instruction passed into the |
tree | commitdiff |
2008-05-12 |
Nate Begeman | Initial X86 codegen support for VSETCC. |
tree | commitdiff |
2008-05-12 |
Dan Gohman | Fix a copy+paste bug; pseudo-instructions shouldn't... |
tree | commitdiff |
2008-05-12 |
Evan Cheng | Refactor isConsecutiveLoad from X86 to TargetLowering... |
tree | commitdiff |
2008-05-12 |
Dan Gohman | Fix a compile error on compilers that still want a... |
tree | commitdiff |
2008-05-11 |
Anton Korobeynikov | Add note |
tree | commitdiff |
2008-05-10 |
Evan Cheng | When transforming a vector_shuffle to a load, the base... |
tree | commitdiff |
2008-05-10 |
Dan Gohman | For now, abort when an ISD::VAARG is encountered on... |
tree | commitdiff |
2008-05-10 |
Evan Cheng | Some clean up. |
tree | commitdiff |
2008-05-10 |
Evan Cheng | If movl top bits are undef, let it be selected to movlp... |
tree | commitdiff |
2008-05-09 |
Evan Cheng | Add a pattern to do move the low element of a v4f32... |
tree | commitdiff |
2008-05-09 |
Evan Cheng | Handle a few more cases of folding load i64 into xmm... |
tree | commitdiff |
2008-05-08 |
Evan Cheng | Use movq to move low half of XMM register and zero... |
tree | commitdiff |
2008-05-08 |
Evan Cheng | Handle vector move / load which zero the destination... |
tree | commitdiff |
2008-05-07 |
Duncan Sands | Output correct exception handling and frame info |
tree | commitdiff |
2008-05-05 |
Chris Lattner | add a micro optzn. |
tree | commitdiff |
2008-05-05 |
Mon P Wang | Improved generated code for atomic operators |
tree | commitdiff |
2008-05-05 |
Evan Cheng | Code clean up. No functionality change. |
tree | commitdiff |
2008-05-05 |
Mon P Wang | Added addition atomic instrinsics and, or, xor, min... |
tree | commitdiff |
2008-05-05 |
Dan Gohman | Fix IsLinux being uninitialized on non-Linux targets. |
tree | commitdiff |
2008-05-05 |
Anton Korobeynikov | Fix 80col violation |
tree | commitdiff |
2008-05-05 |
Dan Gohman | Use a dedicated IsLinux flag instead of an ELFLinux... |
tree | commitdiff |
2008-05-05 |
Dan Gohman | Add AsmPrinter support for emitting a directive to... |
tree | commitdiff |
2008-05-04 |
Anton Korobeynikov | Add General Dynamic TLS model for X86-64. Some parts... |
tree | commitdiff |
2008-05-04 |
Evan Cheng | Select vector shift with non-immediate i32 shift amount... |
tree | commitdiff |
2008-05-03 |
Evan Cheng | Add separate intrinsics for MMX / SSE shifts with i32... |
tree | commitdiff |
2008-05-02 |
Evan Cheng | Undo r50574. We are already ensuring the folded load... |
tree | commitdiff |
2008-05-02 |
Evan Cheng | 80 column violation. |
tree | commitdiff |
2008-05-02 |
Evan Cheng | Not safe folding a load + FsXORPSrr into FsXORPSrm... |
tree | commitdiff |
2008-04-30 |
Arnold Schwaighofer | Tail call optimization improvements: |
tree | commitdiff |
2008-04-28 |
Dan Gohman | Fix the SVOffset values for loads and stores produced by |
tree | commitdiff |
2008-04-28 |
Anton Korobeynikov | Fix FP return for Win64 ABI |
tree | commitdiff |
2008-04-27 |
Anton Korobeynikov | Properly lower vararg's FORMAL_ARGUMENTS node on win64 |
tree | commitdiff |
2008-04-27 |
Anton Korobeynikov | Handle fp80 for win64 |
tree | commitdiff |
2008-04-26 |
Chris Lattner | A few inline asm cleanups: |
tree | commitdiff |
2008-04-25 |
Evan Cheng | Extract the lower 64-bit if a MMX value is passed in... |
tree | commitdiff |
2008-04-25 |
Evan Cheng | Fix illegal MMX_MOVDQ2Qrr pattern. vector_extract resul... |
tree | commitdiff |
2008-04-25 |
Evan Cheng | Special handling for MMX values being passed in either... |
tree | commitdiff |
2008-04-25 |
Evan Cheng | Fix MMX_MOVQ2DQrr pattern. It's illegal to do a bitconv... |
tree | commitdiff |
2008-04-25 |
Evan Cheng | Not checking for intrinsics which do not have a chain... |
tree | commitdiff |
2008-04-25 |
Evan Cheng | - Switch from std::set to SmallPtrSet. |
tree | commitdiff |
2008-04-25 |
Evan Cheng | MMX argument passing fixes: |
tree | commitdiff |
2008-04-25 |
Chris Lattner | Loosen up an assertion to allow intrinsics. I really... |
tree | commitdiff |
2008-04-25 |
Evan Cheng | Fix bug in x86 memcpy / memset lowering. If there are... |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Fix typo |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Only allow increase of max alignment value |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Be over-conservative: scan for all used virtual registe... |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Add X86 Maximal Stack Alignment Calculator Pass before RA |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Do proper book-keeping of offsets and prologue/epilogue... |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | If stack realignment is used - incoming args will use... |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Eastimate required stack alignment early, so we can... |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Cleanup |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Simplify |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Make stack alignment options global for all targets |
tree | commitdiff |
2008-04-23 |
Anton Korobeynikov | Provide option for enabling-disabling stack realignment |
tree | commitdiff |
next |