summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
...
* Add MBB successors and physreg Uses in the same order thatDan Gohman2008-10-071-7/+4
| | | | | | | SDISel typically adds them in. This makes it a little easier to compare FastISel output with SDISel output. llvm-svn: 57266
* Avoid emitting redundant materializations of integer constantsDan Gohman2008-10-071-3/+4
| | | | | | | for things like null pointers, which at this level aren't different from regular integer constants. llvm-svn: 57265
* Instead of emitting an implicit use for the super-register ofDan Gohman2008-10-071-6/+10
| | | | | | | | X86::CL that was used, emit an EXTRACT_SUBREG from the CL super-register to CL. This more precisely describes how the CL register is being used. llvm-svn: 57264
* Unconditional branch instruction encoding fix. Needs to use ABI, not AXI, to ↵Jim Grosbach2008-10-071-1/+1
| | | | | | get the proper opcode bits. llvm-svn: 57262
* need ARM.h for ARMCC definitionJim Grosbach2008-10-071-0/+1
| | | | llvm-svn: 57261
* Add a testcase for i256 add. i256 isn't fully supported inDan Gohman2008-10-071-0/+18
| | | | | | codegen right now, but add and subtract work. llvm-svn: 57260
* Add an option to enable StrongPHIElimination, for ease of testing.Owen Anderson2008-10-076-4/+31
| | | | llvm-svn: 57259
* Encode the conditional execution predicate when JITing.Jim Grosbach2008-10-072-10/+16
| | | | llvm-svn: 57258
* Model hardwired inputs & outputs of x86 8-bit divides correctly.Dale Johannesen2008-10-071-4/+4
| | | | | | Fixes local RA miscompilation of gcc.c-torture/execute/20020904-1.c -O0. llvm-svn: 57257
* Use Dan's supperior checkAndrew Lenharth2008-10-071-13/+9
| | | | llvm-svn: 57255
* Correctly set attributes when removing args during cloning. Fixes PR2765Andrew Lenharth2008-10-071-1/+17
| | | | llvm-svn: 57254
* Improve function definition, call and invoke instruction docs.Devang Patel2008-10-071-5/+12
| | | | llvm-svn: 57253
* Clarify naming and correct conditional so that CMP and CMN instructions get ↵Jim Grosbach2008-10-071-8/+5
| | | | | | the Rn operand encoded properly llvm-svn: 57252
* Fix Opcode values of CMP and CMNJim Grosbach2008-10-071-4/+4
| | | | llvm-svn: 57251
* 128 mul test, xfailedAndrew Lenharth2008-10-071-0/+8
| | | | llvm-svn: 57250
* No need for |=Andrew Lenharth2008-10-071-3/+3
| | | | llvm-svn: 57249
* Use ADDC if it is valid at any smaller size. Do it right this timeAndrew Lenharth2008-10-071-3/+6
| | | | llvm-svn: 57248
* Use ADDC if it is valid at any smaller size. fixes ↵Andrew Lenharth2008-10-071-3/+18
| | | | | | test/Codegen/Generic/i128-addsub.ll on x86 llvm-svn: 57247
* Certain patterns involving the "movss" instruction were marked as requiring ↵Anders Carlsson2008-10-072-3/+25
| | | | | | SSE2, when in reality movss is an SSE1 instruction. llvm-svn: 57246
* add support for running the test suite with valgrind. to run it just type ↵Nuno Lopes2008-10-072-3/+33
| | | | | | | | 'make VG=1', as in clang beware of the 42000 leaks reported by valgrind in the Constant.cpp + Type.cpp files. it needs fixing IMHO llvm-svn: 57245
* Expand arith on machines without carry flagsAndrew Lenharth2008-10-073-10/+42
| | | | llvm-svn: 57243
* Add two forgotten </i>'s.Matthijs Kooijman2008-10-071-2/+2
| | | | llvm-svn: 57239
* fix filetype suffixGabor Greif2008-10-071-0/+0
| | | | llvm-svn: 57237
* Advance version to 2.5Tanya Lattner2008-10-072-37/+41
| | | | llvm-svn: 57233
* no need to write the output to the diskChris Lattner2008-10-071-1/+1
| | | | llvm-svn: 57232
* Add test case for ADDC ADDE expansionAndrew Lenharth2008-10-071-0/+10
| | | | llvm-svn: 57228
* Note that ADDC and company don't actually expand yet (missing in legalizeAndrew Lenharth2008-10-074-2/+65
| | | | llvm-svn: 57226
* Mark negative-zero-to-int conversion as Inexact,Dale Johannesen2008-10-071-1/+2
| | | | | | | | | since -0.0 can't be represented as an int. This prevents llvm from reducing -0.0 to a load-and-convert of int 0. Fixes ieee.exp/mzero[2356].c in gcc testsuite. llvm-svn: 57224
* Correctly handle calls with no return values. This fixesDan Gohman2008-10-071-1/+1
| | | | | | 2006-01-23-UnionInit on x86-64 when inlining is not enabled. llvm-svn: 57223
* Fix typo, fix PR 2865.Devang Patel2008-10-062-2/+33
| | | | llvm-svn: 57221
* Be more precise about which conversions of NaNsDale Johannesen2008-10-063-3/+49
| | | | | | | | are Inexact. (These are not Inexact as defined by IEEE754, but that seems like a reasonable way to abstract what happens: information is lost.) llvm-svn: 57218
* Set UNIVERSAL_ARCH value.Devang Patel2008-10-061-1/+2
| | | | | | Fixes build failure when target is i386. llvm-svn: 57211
* Fix PR2850 and PR2863. Only generate movddup for 128-bit SSE vector shuffles.Evan Cheng2008-10-062-0/+13
| | | | llvm-svn: 57210
* Temporarily revert functionality change from my previous patch;Dale Johannesen2008-10-061-2/+1
| | | | | | it is too aggressive. llvm-svn: 57203
* Remove interfaces implemented by dead pass from the list of available passes.Devang Patel2008-10-062-7/+28
| | | | | | Patch By Matthijs Kooijman. llvm-svn: 57202
* Cosmetic.Evan Cheng2008-10-061-1/+1
| | | | llvm-svn: 57200
* Update function attributes docs.Devang Patel2008-10-061-9/+37
| | | | llvm-svn: 57197
* This is an objective-c test, not an objective-c++ one.Evan Cheng2008-10-061-1/+1
| | | | llvm-svn: 57193
* Mark shortening NaN conversions as Inexact. PR 2856.Dale Johannesen2008-10-061-6/+9
| | | | | | Improve description of unsupported formats. llvm-svn: 57185
* It is possible that all functions in one module are not being Devang Patel2008-10-061-5/+2
| | | | | | optimized for size. Set OptForSize for each function separately. llvm-svn: 57182
* Don't dereference the end() iterator. Thanks toDan Gohman2008-10-061-1/+1
| | | | | | ENABLE_EXPENSIVE_CHECKS for finding this. llvm-svn: 57181
* Remove unncessary isDeclaration() checks.Devang Patel2008-10-063-3/+3
| | | | llvm-svn: 57179
* Allow scalarrepl to treat an all-zero GEP just as bitcast.Matthijs Kooijman2008-10-062-14/+52
| | | | | | | This includes not marking a GEP involving a vector as unsafe, but only when it has all zero indices. This allows scalarrepl to work in a few more cases. llvm-svn: 57177
* Actually run Obj-C++ tests if llvm-gcc supports.Duncan Sands2008-10-063-3/+3
| | | | | | | | | | | Before there were two problems: (1) configure turned "obj-c++" into "obj" in the langs line; (2) the dejagnu library called it objc++ not obj-c++. Now the problem is that some of these tests don't pass! llvm-svn: 57167
* Clarify the relationship between byval and readonly/Duncan Sands2008-10-061-15/+20
| | | | | | | | | | readnone. Make clearer that readnone functions do not dereference pointer arguments. Do not use the highly ambiguous "side-effects" in the readonly description (since such functions can have control flow side-effects, such as throwing an exception, or looping for ever). llvm-svn: 57166
* reorder #include order, patch by Kenneth Boyd!Chris Lattner2008-10-061-4/+2
| | | | llvm-svn: 57148
* Add #include to get alloca, patch by Kenneth Boyd!Chris Lattner2008-10-061-0/+1
| | | | llvm-svn: 57147
* Fix shift overflow bug that would occur when a field was a full 32-bits Chris Lattner2008-10-051-1/+1
| | | | | | | in tblgen. This is PR2827, thanks to Waldemar Knorr for tracking this down. llvm-svn: 57124
* fix an incorrect and extremely confusing error messageChris Lattner2008-10-051-1/+1
| | | | llvm-svn: 57123
* make the autoupgrade code for ret attributes dramatically simpler Chris Lattner2008-10-051-30/+22
| | | | | | and actually work. We can now read the llvm 2.3 bc file from PR2849 llvm-svn: 57122
OpenPOWER on IntegriCloud