summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
* Add MMX calling conventions.Bill Wendling2007-03-301-4/+14
| | | | llvm-svn: 35489
* New entry.Evan Cheng2007-03-291-0/+5
| | | | llvm-svn: 35480
* Can't re-materialize mov r, imm in thumb since mov would clobber the ↵Evan Cheng2007-03-291-1/+1
| | | | | | condition code. llvm-svn: 35479
* Use the stat information in the Path object, if it is already obtained. ThisReid Spencer2007-03-291-4/+3
| | | | | | | avoids a call to ::fstat by MappedFile when the file size information was already obtained by the Path object. llvm-svn: 35477
* For PR789:Reid Spencer2007-03-298-60/+75
| | | | | | | | Make the sys::Path::getFileStatus function more efficient by having it return a pointer to the FileStatus structure rather than copy it. Adjust uses of the function accordingly. Also, fix some memory issues in sys::Path. llvm-svn: 35476
* the bytecode reader supports dematerializeFunctionChris Lattner2007-03-292-7/+22
| | | | llvm-svn: 35475
* Regenerate.Reid Spencer2007-03-295-448/+451
| | | | llvm-svn: 35473
* Give users a grace period on the implementation keyword. This *will* getReid Spencer2007-03-292-3/+3
| | | | | | removed in a few days. llvm-svn: 35472
* MS C does have inlining after all, just uses _inline instead of inline.Jeff Cohen2007-03-291-1/+1
| | | | llvm-svn: 35467
* Determine absolute paths the correct way :)Jeff Cohen2007-03-291-3/+9
| | | | llvm-svn: 35464
* Add a uniqueID field to the FileStatus structure for Paths. This will mapReid Spencer2007-03-292-0/+8
| | | | | | | to the inode number on Unix and something far less unique on Windows. The windows case needs to be improved. llvm-svn: 35461
* For PR789:Reid Spencer2007-03-293-4/+4
| | | | | | Updates for change in interface of getFileStatus method of sys::Path class. llvm-svn: 35458
* For PR789:Reid Spencer2007-03-292-26/+49
| | | | | | | | | * Add a method: bool isAbsolute() const, which determines if the path name is absolute or not. * Implement caching of file status information in the Path object. Allow it to be updated forcefully or lazily re-fetched from the cached value. llvm-svn: 35456
* 1. Make more use of APInt::getHighBitsSet/getLowBitsSet.Zhou Sheng2007-03-291-24/+28
| | | | | | | 2. Let APInt variable do the binary operation stuff instead of using ConstantExpr::getXXX. llvm-svn: 35450
* Disable load width reduction xform of variant (zext (truncate load x)) forEvan Cheng2007-03-291-1/+3
| | | | | | big endian targets until llvm-gcc build issue has been resolved. llvm-svn: 35449
* Add support for hidden visibility to darwin/arm.Evan Cheng2007-03-291-0/+4
| | | | llvm-svn: 35448
* Clean up some codes in InstCombiner::SimplifyDemandedBits().Zhou Sheng2007-03-291-12/+6
| | | | llvm-svn: 35446
* New entries.Evan Cheng2007-03-291-0/+21
| | | | llvm-svn: 35445
* Clean up codes in InstCombiner::SimplifyDemandedBits():Zhou Sheng2007-03-291-32/+35
| | | | | | | | 1. Line out nested call of APInt::zext/trunc. 2. Make more use of APInt::getHighBitsSet/getLowBitsSet. 3. Use APInt[] operator instead of expression like "APIntVal & SignBit". llvm-svn: 35444
* 1. Make the APInt variable do the binary operation stuff if possibleZhou Sheng2007-03-291-3/+5
| | | | | | | | instead of using ConstantExpr::getXX. 2. Use constant reference to APInt if possible instead of expensive APInt copy. llvm-svn: 35443
* Fix C Backend to generate code that works with Microsoft C for the benefit ofJeff Cohen2007-03-281-1/+4
| | | | | | front ends that do not depend on the GCC runtime (someday...). llvm-svn: 35441
* Made this into a bug report: PR1286Bill Wendling2007-03-281-40/+0
| | | | llvm-svn: 35439
* Oops :)Anton Korobeynikov2007-03-281-1/+1
| | | | llvm-svn: 35438
* Don't allow MatchAddress recurse too much. This trims exponentialAnton Korobeynikov2007-03-281-6/+14
| | | | | | behaviour in some cases. llvm-svn: 35437
* add a noteChris Lattner2007-03-281-0/+8
| | | | llvm-svn: 35435
* Compile CodeGen/X86/lea-3.ll:test2 to:Chris Lattner2007-03-281-0/+13
| | | | | | | | | | | | | | | | _test3: leaq (,%rdi,4), %rax orq %rdi, %rax ret instead of: _test2: movq %rdi, %rax shlq $2, %rax orq %rdi, %rax ret llvm-svn: 35434
* Avoid unnecessary APInt construction.Zhou Sheng2007-03-281-3/+2
| | | | llvm-svn: 35431
* Dan pointed out that this is done, remove it!Chris Lattner2007-03-281-21/+0
| | | | llvm-svn: 35430
* First test check-in.Scott Michel2007-03-281-0/+10
| | | | llvm-svn: 35429
* 1. Make more use of getLowBitsSet/getHighBitsSet.Zhou Sheng2007-03-281-13/+16
| | | | | | | | 2. Use APInt[] instead of "X & SignBit". 3. Clean up some codes. 4. Make the expression like "ShiftAmt = ShiftAmtC->getZExtValue()" safe. llvm-svn: 35424
* It is not possible to determine dominance between two PHI nodesDevang Patel2007-03-281-0/+5
| | | | | | based on their ordering. This is applicable to ETForest::dominates() also. llvm-svn: 35423
* 1. Make more use of getLowBitsSet/getHighBitsSet.Zhou Sheng2007-03-281-5/+3
| | | | | | | 2. Make the APInt value do the zext/trunc stuff instead of using ConstantExpr::getZExt(). llvm-svn: 35422
* Notes on re-materialization.Evan Cheng2007-03-281-0/+40
| | | | llvm-svn: 35420
* Move rematerialization out of beta.Evan Cheng2007-03-281-7/+1
| | | | llvm-svn: 35419
* Use UnknownBIts[BitWidth-1] instead of UnknownBIts & SignBits.Zhou Sheng2007-03-281-1/+1
| | | | llvm-svn: 35418
* Remove unused APInt variable.Zhou Sheng2007-03-281-3/+0
| | | | llvm-svn: 35414
* Regenerate.Reid Spencer2007-03-285-451/+448
| | | | llvm-svn: 35411
* Replace the "implementation" keyword with "checkpoint". This more accuratelyReid Spencer2007-03-282-3/+3
| | | | | | | | describes its function as check against unresolved types and definitions. This is handy for forcing error messages in assembly test cases or otherwise ensuring that everything is resolved at that point in the assembly. llvm-svn: 35410
* Clean up codes in ComputeMaskedBits():Zhou Sheng2007-03-281-29/+27
| | | | | | | | 1. Line out nested use of zext/trunc. 2. Make more use of getHighBitsSet/getLowBitsSet. 3. Use APInt[] != 0 instead of "(APInt & SignBit) != 0". llvm-svn: 35408
* Scale 1 is always ok.Evan Cheng2007-03-281-1/+1
| | | | llvm-svn: 35407
* Remove isLegalAddressImmediate.Evan Cheng2007-03-283-28/+0
| | | | llvm-svn: 35406
* No need to generate the implementation keyword any more. Its frivolous.Reid Spencer2007-03-281-2/+0
| | | | llvm-svn: 35405
* GEP index sinking fixes:Evan Cheng2007-03-281-40/+35
| | | | | | | | | 1) Take address scale into consideration. e.g. i32* -> scale 4. 2) Examine all the users of GEP. 3) Generalize to inter-block GEP's (no longer uses loopinfo). 4) Don't do xform if GEP has other variable index(es). llvm-svn: 35403
* For PR1280:Reid Spencer2007-03-281-3/+13
| | | | | | | When converting an add/xor/and triplet into a trunc/sext, only do so if the intermediate integer type is a bitwidth that the targets can handle. llvm-svn: 35400
* Fix for PR1279. Dead def has a live interval of length 1. Copy coalescing shouldEvan Cheng2007-03-281-3/+5
| | | | | | not violate that. llvm-svn: 35396
* Remove cruft I put in there...Bill Wendling2007-03-281-72/+0
| | | | llvm-svn: 35394
* Fix a problem building llvm-gcc on amd64-unknown-freebsd6.2, due to theChris Lattner2007-03-281-1/+3
| | | | | | system assembler not groking legal instructions like "leal (,%esi,8), %ecx". llvm-svn: 35393
* Unbreak mmx arithmetic. It was barfing trying to do v8i8 arithmetic.Bill Wendling2007-03-282-3/+78
| | | | llvm-svn: 35392
* Add the "unpack low packed data" instructions. This should be the last ofBill Wendling2007-03-271-73/+130
| | | | | | the MMX instructions that are needed... llvm-svn: 35389
* Use SmallPtrSet.Devang Patel2007-03-271-1/+2
| | | | llvm-svn: 35388
OpenPOWER on IntegriCloud