summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
* Start generating arbitrary precision integer SCEVs. This removes the temporaryNick Lewycky2009-01-251-11/+1
| | | | | | code that rounded up and capped the size. llvm-svn: 62958
* The function that does nothing but call malloc is noalias return.Nick Lewycky2009-01-251-5/+6
| | | | llvm-svn: 62956
* Private linkage support for PPC / Darwin.Evan Cheng2009-01-251-0/+2
| | | | llvm-svn: 62955
* Teach 2addr pass to be do more commuting. If both uses of a two-address ↵Evan Cheng2009-01-251-6/+104
| | | | | | | | | | | | | | | | | | | | | instruction are killed, but the first operand has a use before and after the def, commute if the second operand does not suffer from the same issue. %reg1028<def> = EXTRACT_SUBREG %reg1027<kill>, 1 %reg1029<def> = MOV8rr %reg1028 %reg1029<def> = SHR8ri %reg1029, 7, %EFLAGS<imp-def,dead> insert => %reg1030<def> = MOV8rr %reg1028 %reg1030<def> = ADD8rr %reg1028<kill>, %reg1029<kill>, %EFLAGS<imp-def,dead> In this case, it might not be possible to coalesce the second MOV8rr instruction if the first one is coalesced. So it would be profitable to commute it: %reg1028<def> = EXTRACT_SUBREG %reg1027<kill>, 1 %reg1029<def> = MOV8rr %reg1028 %reg1029<def> = SHR8ri %reg1029, 7, %EFLAGS<imp-def,dead> insert => %reg1030<def> = MOV8rr %reg1029 %reg1030<def> = ADD8rr %reg1029<kill>, %reg1028<kill>, %EFLAGS<imp-def,dead> llvm-svn: 62954
* Fix an indent and a typo.Nate Begeman2009-01-242-2/+2
| | | | llvm-svn: 62940
* Revert previous change; even this mild and clearlyDale Johannesen2009-01-241-7/+4
| | | | | | | more accurate change loses more than it gains on benchmarks. llvm-svn: 62938
* add note about possible GEP improvement with fields of size 0.Torok Edwin2009-01-241-0/+6
| | | | llvm-svn: 62925
* testcase for PR3381.Torok Edwin2009-01-241-1/+1
| | | | | | Also it was an empty struct, not a void after all. llvm-svn: 62920
* void* is represented as pointer to empty struct {}.Torok Edwin2009-01-241-0/+2
| | | | | | | Thus we need to check whether the struct is empty before trying to index into it. This fixes PR3381. llvm-svn: 62918
* Some cleanups. No functional changes.Owen Anderson2009-01-241-26/+34
| | | | llvm-svn: 62917
* Improve the inlining cost function a bit.Dale Johannesen2009-01-241-3/+6
| | | | | | Little practical effect. llvm-svn: 62908
* Make InstCombineStoreToCast handle aggregates more aggressively,Chris Lattner2009-01-241-18/+48
| | | | | | | handling the case in Transforms/InstCombine/cast-store-gep.ll, which is a heavily reduced testcase from Clang on x86-64. llvm-svn: 62904
* Refactor code. No functionality change.Evan Cheng2009-01-231-20/+38
| | | | llvm-svn: 62893
* Introduce two DWARF attribute extentions DW_AT_APPLE_optimized, ↵Devang Patel2009-01-233-2/+13
| | | | | | | | | | | | DW_AT_APPLE_flags. DW_AT_APPLE_optimized flag is set when a compile_unit is optimized. The debugger takes advantage of this information some way. DW_AT_APPLE_flags encodes command line options when certain env. variable is set. This is used by build engineers to track various gcc command lines used by by a project, irrespective of whether the project used makefile, Xcode or something else. llvm-gcc patch is next. llvm-svn: 62888
* hopefully address PR3379 by making the P modifier work in x86 inline asm.Chris Lattner2009-01-231-0/+3
| | | | llvm-svn: 62887
* use CallSite::isCalle instead of slow getOperandNoGabor Greif2009-01-231-3/+3
| | | | llvm-svn: 62877
* Simplify the logic of getting hold of a PHI predecessor block.Gabor Greif2009-01-234-10/+5
| | | | | | | | | There is now a direct way from value-use-iterator to incoming block in PHINode's API. This way we avoid the iterator->index->iterator trip, and especially the costly getOperandNo() invocation. Additionally there is now an assertion that the iterator really refers to one of the PHI's Uses. llvm-svn: 62869
* Add support for deleting a module provider from a JIT in such a way that it ↵Nate Begeman2009-01-233-1/+41
| | | | | | does not cause the owned module to be fully materialized. llvm-svn: 62864
* Empty DIType represents void. In this case no need to construct any type DIE.Devang Patel2009-01-231-3/+1
| | | | llvm-svn: 62861
* Fold x-0 to x in unsafe-fp-math mode. This comes up in theDan Gohman2009-01-232-9/+19
| | | | | | | | | | | | | | testcase from PR3376, and in fact is sufficient to completely avoid the problem in that testcase. There's an underlying problem though; TLI.isOperationLegal considers Custom to be Legal, which might be ok in some cases, but that's what DAGCombiner is using in many places to test if something is legal when LegalOperations is true. When DAGCombiner is running after legalize, this isn't sufficient. I'll address this in a separate commit. llvm-svn: 62860
* fix two more cases where we could let the NLPDI cache get unsorted.Chris Lattner2009-01-231-5/+14
| | | | | | With this, sqlite3 now passes. llvm-svn: 62839
* Unconditionally reset 'cache' to zero, even if we don't need to resort it.Chris Lattner2009-01-231-5/+4
| | | | | | | | | This avoids using a dangling pointer. Reset NumSortedEntries after restoring Cache to avoid extraneous sorts. This fixes the reduced sqlite3 testcase, but apparently not the whole app. llvm-svn: 62838
* Only check if coalescing is worthwhile when the result is targeting a more ↵Evan Cheng2009-01-231-1/+6
| | | | | | restrictive register class. llvm-svn: 62837
* Stage two of fixing pre-alloc-splitting's code size issues: filter out ↵Owen Anderson2009-01-231-9/+37
| | | | | | | | | | restores that are just going to be re-spilled again. This also helps performance. Pre-alloc-splitting now seems to be an overall win on SPEC. llvm-svn: 62834
* Cross register class coalescing. Not yet enabled.Evan Cheng2009-01-232-109/+123
| | | | llvm-svn: 62832
* Code did not follow associated comment. not a good idea.Devang Patel2009-01-231-1/+1
| | | | llvm-svn: 62828
* Set appropriate tag for the composite type.Devang Patel2009-01-231-0/+1
| | | | llvm-svn: 62827
* a minor tweak to my previous patch, handle the invalidation caseChris Lattner2009-01-231-3/+4
| | | | | | when there are multiple iterations of the loop. This fixes PR3375. llvm-svn: 62822
* Add an initial pass at dead spill/restore removal for pre alloc splitting.Owen Anderson2009-01-231-4/+56
| | | | llvm-svn: 62821
* do not sign extend characters input to isprint. This improvesChris Lattner2009-01-221-1/+1
| | | | | | compatibility with VC++. Patch by Max Burke! llvm-svn: 62813
* Fix a minor bug in DAGCombiner's folding of SELECT. Folding "select C, 0, 1"Bob Wilson2009-01-221-3/+5
| | | | | | | | | | | | | | | | | | | | | | to "C ^ 1" is only valid when C is known to be either 0 or 1. Most of the similar foldings in this function only handle "i1" types, but this one appears intentionally written to handle larger integer types. If C has an integer type larger than "i1", this needs to check if the high bits of a boolean are known to be zero. I also changed the comment to describe this folding as "C ^ 1" instead of "~C", since that is what the code does and since the latter would only be valid for "i1" types. The good news is that most LLVM targets use TargetLowering::ZeroOrOneBooleanContent so this change will not disable the optimization; the bad news is that I've been unable to come up with a testcase to demonstrate the problem. I have also removed a "FIXME" comment for folding "select C, X, 0" to "C & X", since the code looks correct to me. It could be made more aggressive by not limiting the type to "i1", but that would then require checking for TargetLowering::ZeroOrNegativeOneBooleanContent. Similar changes could be done for the other SELECT foldings, but it was decided to be not worth the trouble and complexity (see e.g., r44663). llvm-svn: 62790
* Don't create ISD::FNEG nodes after legalize if they aren't legal.Dan Gohman2009-01-222-8/+30
| | | | | | | | | | | | | | Simplify x+0 to x in unsafe-fp-math mode. This avoids a bunch of redundant work in many cases, because in unsafe-fp-math mode, ISD::FADD with a constant is considered free to negate, so the DAGCombiner often negates x+0 to -0-x thinking it's free, when in reality the end result is -x, which is more expensive than x. Also, combine x*0 to 0. This fixes PR3374. llvm-svn: 62789
* introduce a useful abstraction to find out if a Use is in the call position ↵Gabor Greif2009-01-224-9/+8
| | | | | | of an instruction llvm-svn: 62788
* Revert r62553 and r62616 due to issues with portability.Tanya Lattner2009-01-221-206/+496
| | | | llvm-svn: 62777
* rename methods in System/Host to be more consistent.Chris Lattner2009-01-223-8/+8
| | | | llvm-svn: 62776
* Add SelectionDAG::getNOT method to construct bitwise NOT operations,Bob Wilson2009-01-225-34/+41
| | | | | | | corresponding to the "not" and "vnot" PatFrags. Use the new method in some places where it seems appropriate. llvm-svn: 62768
* Few targets like the tiny little PIC16 have only 16-bit pointers.Sanjiv Gupta2009-01-221-0/+4
| | | | llvm-svn: 62763
* Eliminate a couple of fields from TargetRegisterClass: SubRegClasses and ↵Evan Cheng2009-01-223-37/+7
| | | | | | SuperRegClasses. These are not necessary. Also eliminate getSubRegisterRegClass and getSuperRegisterRegClass. These are slow and their results can change if register file names change. Just use TargetLowering::getRegClassFor() to get the right TargetRegisterClass instead. llvm-svn: 62762
* fix a typoChris Lattner2009-01-221-1/+1
| | | | llvm-svn: 62761
* add a noteChris Lattner2009-01-221-0/+10
| | | | llvm-svn: 62760
* Fix PR3358, a really nasty bug where recursive phi translated Chris Lattner2009-01-221-3/+36
| | | | | | | | analyses could be run without the caches properly sorted. This can fix all sorts of weirdness. Many thanks to Bill for coming up with the 'issorted' verification idea. llvm-svn: 62757
* Recognize inline asm for bswap on x86-64 GLIBC. This allows itDan Gohman2009-01-211-1/+11
| | | | | | to be supported in the JIT. llvm-svn: 62730
* Do not use host floating point types when emittingDale Johannesen2009-01-213-9/+14
| | | | | | | | | ASCII IR; loading and storing these can change the bits of NaNs on some hosts. Remove or add warnings at a few other places using host floating point; this is a bad thing to do in general. llvm-svn: 62712
* Also favors NOT64r.Evan Cheng2009-01-211-1/+1
| | | | llvm-svn: 62710
* fix warning in release-asserts mode and spelling of assert.Chris Lattner2009-01-211-3/+2
| | | | llvm-svn: 62699
* add getPointerToGlobal to the C bindings, patch by Lennart Augustsson!Chris Lattner2009-01-211-0/+4
| | | | | | PR3364 llvm-svn: 62697
* Remove uses of uint32_t in favor of 'unsigned' for better Chris Lattner2009-01-212-194/+194
| | | | | | compatibility with cygwin. Patch by Jay Foad! llvm-svn: 62695
* Simplify ReduceLoadWidth's logic: it doesn't need several differentDan Gohman2009-01-211-19/+10
| | | | | | | | special cases after producing the new reduced-width load, because the new load already has the needed adjustments built into it. This fixes several bugs due to the special cases, including PR3317. llvm-svn: 62692
* Fix a recent regression. ClrOpcode is not set for i8; for i8, ifDan Gohman2009-01-211-2/+3
| | | | | | | we want to clear %ah to zero before a division, just use a zero-extending mov to %al. This fixes PR3366. llvm-svn: 62691
* Mimic gcc behaviour with regard to response files.Mikhail Glushenkov2009-01-211-16/+15
| | | | llvm-svn: 62688
OpenPOWER on IntegriCloud