- 01 Nov, 2011 9 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r143290 | d0k | 2011-10-29 12:43:38 -0700 (Sat, 29 Oct 2011) | 3 lines PPC: Disable moves for all CR subregisters. Should fix assertion failures on ppc buildbots. ``` --------------------------------------------------------------------- llvm-svn: 143451
-
Bill Wendling authored
------------------------------------------------------------------------ llvm-svn: 143450
-
Bill Wendling authored
```--------------------------------------------------------------------- r143194 | chapuni | 2011-10-28 07:12:22 -0700 (Fri, 28 Oct 2011) | 7 lines Dwarf: [PR11022] Fix emitting DW_AT_const_value(>i64), to be host-endian-neutral. Don't assume APInt::getRawData() would hold target-aware endianness nor host-compliant endianness. rawdata[0] holds most lower i64, even on big endian host. FIXME: Add a testcase for big endian target. FIXME: Ditto on CompileUnit::addConstantFPValue() ? ``` --------------------------------------------------------------------- llvm-svn: 143449
-
Bill Wendling authored
llvm-svn: 143448
-
Bill Wendling authored
llvm-svn: 143447
-
Bill Wendling authored
llvm-svn: 143439
-
Bill Wendling authored
```--------------------------------------------------------------------- r143437 | efriedma | 2011-10-31 21:49:29 -0700 (Mon, 31 Oct 2011) | 3 lines Make sure we use the right insertion point when instcombine replaces a PHI with another instruction. (Specifically, don't insert an arbitrary instruction before a PHI.) Fixes PR11275. ``` --------------------------------------------------------------------- llvm-svn: 143438
-
Bill Wendling authored
```--------------------------------------------------------------------- r143345 | chandlerc | 2011-10-31 01:42:27 -0700 (Mon, 31 Oct 2011) | 7 lines Add "just one more" include path to the monstrosity that is our header search logic. The Debian multiarch seems to have completely changed from when these were originally added, and I'd like to remove a bunch of them, but I'll be lazy and delay that until this logic is hoisted into the driver where it belongs. This should resolve PR11223. ``` --------------------------------------------------------------------- llvm-svn: 143433
-
Bill Wendling authored
```--------------------------------------------------------------------- r143344 | chandlerc | 2011-10-31 01:42:24 -0700 (Mon, 31 Oct 2011) | 14 lines Fix part of PR11223 and probably a few dups as well. This teaches the library search logic to "properly" handle multiarch installations. I've tested this on both Debian unstable and the latest Ubuntu which both use this setup, and this appears to work largely the same way as GCC does. It isn't exactly the same, but it is close enough and more principled in its behavior where it differs. This should resolve any failures to find 'crt1.o' etc on Debian-based Linux distributions. If folks find more cases where we fail, please file bugs and CC me. Test cases for all of the debian silliness are waiting both to simplify the process of merging these down into the 3.0 release, and because they're so crazy I haven't yet been able to really produce a fake tree that represents what we need to test for. I'll eventually add them though. ``` --------------------------------------------------------------------- llvm-svn: 143432
-
- 29 Oct, 2011 4 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r143302 | void | 2011-10-29 16:49:52 -0700 (Sat, 29 Oct 2011) | 1 line Do a relative path ln command instead of an absolute path one. Some people strangely enough have different directory layouts... ``` --------------------------------------------------------------------- llvm-svn: 143303
-
Bill Wendling authored
llvm-svn: 143301
-
Bill Wendling authored
```--------------------------------------------------------------------- r143247 | chapuni | 2011-10-28 16:11:03 -0700 (Fri, 28 Oct 2011) | 1 line test/CodeGen/PowerPC/2008-10-17-AsmMatchingOperands.ll: [PR11218] Mark "REQUIRES: asserts" for now. ``` --------------------------------------------------------------------- llvm-svn: 143272
-
Bill Wendling authored
Patch by David Chisnall. llvm-svn: 143270
-
- 28 Oct, 2011 1 commit
-
-
Bill Wendling authored
------------------------------------------------------------------------ llvm-svn: 143226
-
- 26 Oct, 2011 4 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r142998 | efriedma | 2011-10-25 17:55:23 -0700 (Tue, 25 Oct 2011) | 3 lines Add a note about the removal of the atomic intrinsics to the "Major Changes" section of the release notes. ``` --------------------------------------------------------------------- llvm-svn: 143070
-
Bill Wendling authored
```--------------------------------------------------------------------- r142994 | efriedma | 2011-10-25 17:36:41 -0700 (Tue, 25 Oct 2011) | 3 lines Remove dead atomic intrinsics from LangRef. ``` --------------------------------------------------------------------- llvm-svn: 143069
-
Bill Wendling authored
llvm-svn: 143014
-
Bill Wendling authored
```--------------------------------------------------------------------- r143006 | stoklund | 2011-10-25 18:47:48 -0700 (Tue, 25 Oct 2011) | 8 lines Don't use floating point to do an integer's job. This code makes different decisions when compiled into x87 instructions because of different rounding behavior. That caused phase 2/3 miscompares on 32-bit Linux when the phase 1 compiler was built with gcc (using x87), and the phase 2 compiler was built with clang (using SSE). This fixes PR11200. ``` --------------------------------------------------------------------- llvm-svn: 143013
-
- 25 Oct, 2011 2 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r142956 | grosbach | 2011-10-25 13:30:48 -0700 (Tue, 25 Oct 2011) | 4 lines Make assert() message more informative. PR11217. ``` --------------------------------------------------------------------- llvm-svn: 142967
-
Bill Wendling authored
```--------------------------------------------------------------------- r142918 | theraven | 2011-10-25 03:12:21 -0700 (Tue, 25 Oct 2011) | 3 lines Change an int64_t to an intptr_t so that we don't end up with crashes in the back end on large classes on 32-bit. ``` --------------------------------------------------------------------- llvm-svn: 142953
-
- 24 Oct, 2011 5 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r142869 | void | 2011-10-24 16:05:43 -0700 (Mon, 24 Oct 2011) | 4 lines Check the visibility of the global variable before placing it into the stubs table. A hidden variable could potentially end up in both lists. <rdar://problem/10336715> ``` --------------------------------------------------------------------- llvm-svn: 142870
-
Bill Wendling authored
```--------------------------------------------------------------------- r142841 | efriedma | 2011-10-24 13:24:21 -0700 (Mon, 24 Oct 2011) | 3 lines Add support to the old JIT for acquire/release loads and stores on x86. PR11207. ``` --------------------------------------------------------------------- llvm-svn: 142842
-
Bill Wendling authored
x86-64. We (GHC team) would like this patch included as we've recently added support to GHC for the ARM platform. Patch by David Terei! llvm-svn: 142820
-
Bill Wendling authored
```--------------------------------------------------------------------- r142806 | grosbach | 2011-10-24 10:26:26 -0700 (Mon, 24 Oct 2011) | 1 line Update test for r142801. ``` --------------------------------------------------------------------- llvm-svn: 142809
-
Bill Wendling authored
```--------------------------------------------------------------------- r142801 | grosbach | 2011-10-24 10:16:24 -0700 (Mon, 24 Oct 2011) | 4 lines Thumb2 LDM instructions can target PC. Make sure to encode it. PR11220 ``` --------------------------------------------------------------------- llvm-svn: 142808
-
- 22 Oct, 2011 2 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r142476 | dgregor | 2011-10-18 22:50:34 -0700 (Tue, 18 Oct 2011) | 2 lines Add TypeKind.CONSTANTARRAY, from Anders Waldenborg! ``` --------------------------------------------------------------------- llvm-svn: 142723
-
Bill Wendling authored
```--------------------------------------------------------------------- r142474 | dgregor | 2011-10-18 22:47:46 -0700 (Tue, 18 Oct 2011) | 5 lines This new field was introduced in clang-c in r141277 Python needs this in its structure definition so it allocates enough memory. From Anders Waldenborg! ``` --------------------------------------------------------------------- llvm-svn: 142722
-
- 20 Oct, 2011 5 commits
-
-
Bill Wendling authored
llvm-svn: 142614
-
Bill Wendling authored
```--------------------------------------------------------------------- r142574 | baldrick | 2011-10-20 01:57:54 -0700 (Thu, 20 Oct 2011) | 3 lines Comment out or remove unused parameter names so as to avoid a slew of compiler warnings. ``` --------------------------------------------------------------------- llvm-svn: 142611
-
Bill Wendling authored
```--------------------------------------------------------------------- r142573 | baldrick | 2011-10-20 01:56:41 -0700 (Thu, 20 Oct 2011) | 2 lines Avoid warnings about the parameter G being unused. ``` --------------------------------------------------------------------- llvm-svn: 142610
-
Bill Wendling authored
```--------------------------------------------------------------------- r142350 | baldrick | 2011-10-18 05:44:00 -0700 (Tue, 18 Oct 2011) | 3 lines Fix a bunch of unused variable warnings when doing a release build with gcc-4.6. ``` --------------------------------------------------------------------- llvm-svn: 142609
-
Bill Wendling authored
```--------------------------------------------------------------------- r142559 | void | 2011-10-19 17:45:46 -0700 (Wed, 19 Oct 2011) | 1 line Revamp the script to handle the 'TEST=simple' output. ``` --------------------------------------------------------------------- llvm-svn: 142560
-
- 19 Oct, 2011 8 commits
-
-
Bill Wendling authored
```--------------------------------------------------------------------- r142550 | evancheng | 2011-10-19 15:22:54 -0700 (Wed, 19 Oct 2011) | 1 line Fix TLS lowering bug. The CopyFromReg must be glued to the TLSCALL. rdar://10291355 ``` --------------------------------------------------------------------- llvm-svn: 142555
-
Bill Wendling authored
```--------------------------------------------------------------------- r142493 | chandlerc | 2011-10-19 03:32:19 -0700 (Wed, 19 Oct 2011) | 5 lines Generalize the reading of probability metadata to work for both branches and switches, with arbitrary numbers of successors. Still optimized for the common case of 2 successors for a conditional branch. Add a test case for switch metadata showing up in the BlockFrequencyInfo pass. ``` --------------------------------------------------------------------- llvm-svn: 142548
-
Bill Wendling authored
```--------------------------------------------------------------------- r142492 | chandlerc | 2011-10-19 03:30:30 -0700 (Wed, 19 Oct 2011) | 7 lines Teach the BranchProbabilityInfo analysis pass to read any metadata encoding of probabilities. In the absense of metadata, it continues to fall back on static heuristics. This allows __builtin_expect, after lowering through llvm.expect a branch instruction's metadata, to actually enter the branch probability model. This is one component of resolving PR2577. ``` --------------------------------------------------------------------- llvm-svn: 142547
-
Bill Wendling authored
```--------------------------------------------------------------------- r142491 | chandlerc | 2011-10-19 03:12:41 -0700 (Wed, 19 Oct 2011) | 9 lines Add pass printing support to BlockFrequencyInfo pass. The implementation layer already had support for printing the results of this analysis, but the wiring was missing. Now that printing the analysis works, actually bring some of this analysis, and the BranchProbabilityInfo analysis that it wraps, under test! I'm planning on fixing some bugs and doing other work here, so having a nice place to add regression tests and a way to observe the results is really useful. ``` --------------------------------------------------------------------- llvm-svn: 142546
-
Bill Wendling authored
```--------------------------------------------------------------------- r142168 | chandlerc | 2011-10-16 18:11:57 -0700 (Sun, 16 Oct 2011) | 8 lines Add a routine to swap branch instruction operands, and update any profile metadata at the same time. Use it to preserve metadata attached to a branch when re-writing it in InstCombine. Add metadata to the canonicalize_branch InstCombine test, and check that it is tranformed correctly. Reviewed by Nick Lewycky! ``` --------------------------------------------------------------------- llvm-svn: 142545
-
Bill Wendling authored
```--------------------------------------------------------------------- r142537 | rafael | 2011-10-19 11:48:52 -0700 (Wed, 19 Oct 2011) | 1 line Fix parsing of a line with only a # in it. ``` --------------------------------------------------------------------- llvm-svn: 142539
-
Bill Wendling authored
```--------------------------------------------------------------------- r142489 | void | 2011-10-19 02:47:00 -0700 (Wed, 19 Oct 2011) | 1 line Duncan pointed out that sometimes CC and CXX are used to specify the compiler. Also that the configure script takes care of finding an appropriate compiler if one's not specified. ``` --------------------------------------------------------------------- llvm-svn: 142490
-
Bill Wendling authored
```--------------------------------------------------------------------- r142486 | void | 2011-10-19 02:25:49 -0700 (Wed, 19 Oct 2011) | 1 line Use bash instead. ``` --------------------------------------------------------------------- llvm-svn: 142487
-