Commits

Show all
Author Commit Message Labels Comments Date
Tanya Lattner (Tanya Brethour)
2.5 release notes.
Tags
RELEASE_25
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64066 from mainline. Document the 'llvm.OP.with.overflow' intrinsics.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Remove test.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Update tests to work on ppc.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Revert 62987. This is already reverted in mainline.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. Complete 64076 merge from mainline.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64582 from mainline Fix PR3522. It's not safe to sink into landing pad BB's.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64579 from mainline. Fix pr3571: If stride is a value defined by an instruction, make sure it dominates the loop preheader. When IV users are strength reduced, the stride is inserted into the preheader. It could create a use before def situation.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge all changes to SimpleRegisterCoalescing to the release.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64314 from mainline. Make a transformation added in 63266 a bit less aggressive. It was transforming (x&y)==y to (x&y)!=0 in the case where y is variable and known to have at most one bit set (e.g. z&1). This is not correct; the expressions are not equivalent when y==0. I believe this patch salvages what can be salvaged, including all the cases in bt.ll.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge r64316 from mainline. Don't try to set an EFLAGS operand to dead if no instruction was created. This fixes a bug introduced by r61215.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64412 from mainline. Avoid order files for now, as they aren't supported in 3.79.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64367 from mainline. Fix MingW build: define GTEST_OS_WINDOWS if OS is MingW, but disable exceptions.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64364 from mainline. Fix a nasty bug (PR3550) where the inline pass could incorrectly mark calls with the tail marker when inlining them through an invoke. Patch, testcase, and perfect analysis by Jay Foad!
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64124 frm mainline. Turns out AnalyzeBranch can modify the mbb being analyzed. This is a nasty suprise to some callers, e.g. register coalescer. For now, add an parameter that tells AnalyzeBranch whether it's safe to modify the mbb. A better solution is out there, but I don't have time to deal with it right now.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge 64177 from mainline. Fix PR 3471, and some cleanups.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Revert disabling of building unittests.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. FIx spelling.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. Probe for flags before using them to try to help compiling with compilers that don't support those flags. This hopefully will help gcc 3.X compile this code. http://llvm.org/PR3487
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. Fix PR3486. Fix a bug in code that manually patch physical register live interval after its sub-register is coalesced with a virtual register.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. Pre-alloc splitting needs to be more careful to avoid inserting spills/restores between call frame setup/restore points. Unfortunately, this regresses code size a bit, but at least it's correct now!
Branches
release_25
Tanya Lattner (Tanya Brethour)
Disable building unittests unless PR3487 can be fixed.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. First initialize DAG otherwise dwarf writer is used uninitialized. Duncan spotted this. Thanks!
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. Tevert part of the x86 subtarget logic changes: when -march=x86-64 is given, override the subtarget settings and enable 64-bit support. This restores the earlier behavior, and fixes regressions on Non-64-bit-capable x86-32 hosts. This isn't necessarily the best approach, but the most obvious alternative is to require -mcpu=x86-64 or -mattr=+64bit to be used with -march=x86-64 when the host doesn't have 64-bit support. This makes things little more consistent, but it's less convenient, and it has the practical drawback of requiring lots of test changes, so I opted for the above approach for now.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Back out r62958 from the release branch.
Branches
release_25
Tanya Lattner (Tanya Brethour)
SROA CBE Fix.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Disable support for vicmp/vfcmp from the LLVM 2.5 release branch, like we did for LLVM 2.4. These are slated to be removed (PR3370) and we don't want to have to be backwards compatible with them in the future.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Merge from mainline. Revert r63600.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Regenerate configure.
Branches
release_25
Tanya Lattner (Tanya Brethour)
Update version number to 2.5.
Branches
release_25
  1. Prev
  2. Next