Chandler Carruth 7fcb422bb2 [x86] Revert r212324 which was too aggressive w.r.t. allowing undef ... lanes in vector splats. The core problem here is that undef lanes can't *unilaterally* be considered to contribute to splats. Their handling needs to be more cautious. There is also a reported failure of the nightly testers (thanks Tobias!) that may well stem from the same core issue. I'm going to fix this theoretical issue, factor the APIs a bit better, and then verify that I don't see anything bad with Tobias's reduction from the test suite before recommitting. Original commit message for r212324: [x86] Generalize BuildVectorSDNode::getConstantSplatValue to work for any constant, constant FP, or undef splat and to tolerate any undef lanes in a splat, then replace all uses of isSplatVector in X86's lowering with it. This fixes issues where undef lanes in an otherwise splat vector would prevent the splat logic from firing. It is a touch more awkward to use this interface, but it is much more accurate. Suggestions for better interface structuring welcome. With this fix, the code generated with the widening legalization strategy for widen_cast-4.ll is *dramatically* improved as the special lowering strategies for a v16i8 SRA kick in even though the high lanes are undef. We also get a slightly different choice for broadcasting an aligned memory location, and use vpshufd instead of vbroadcastss. This looks like a minor win for pipelining and domain crossing, but a minor loss for the number of micro-ops. I suspect its a wash, but folks can easily tweak the lowering if they want. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@212475 91177308-0d34-0410-b5e6-96231b3b80d8		2014-07-07 19:03:32 +00:00
..
Analysis	[CostModel][x86] Improved cost model for alternate shuffles.	2014-07-03 22:24:18 +00:00
Assembler	IR: Add COMDATs to the IR	2014-06-27 18:19:56 +00:00
Bindings	Reduce verbiage of lit.local.cfg files	2014-06-09 22:42:55 +00:00
Bitcode	Add some test files for r211710.	2014-06-25 15:41:39 +00:00
BugPoint
CodeGen	[x86] Revert r212324 which was too aggressive w.r.t. allowing undef	2014-07-07 19:03:32 +00:00
DebugInfo	Temporarily revert "Don't try to construct debug LexicalScopes hierarchy for functions that do not have top level debug information." as it appears to be breaking some LTO constructs.	2014-07-03 22:24:54 +00:00
ExecutionEngine	Fix this test to not write to the source tree, and instead to write to	2014-06-28 05:18:49 +00:00
Feature	IR: Add COMDATs to the IR	2014-06-27 18:19:56 +00:00
FileCheck
Instrumentation	[asan] Generate asm instrumentation in MC.	2014-07-07 13:57:37 +00:00
Integer
JitListener
Linker	IR: Add COMDATs to the IR	2014-06-27 18:19:56 +00:00
LTO	Change the default input for llvm-nm to be a.out instead of standard input	2014-06-23 20:27:53 +00:00
MC	Revert r212375 because of test failures	2014-07-05 19:46:10 +00:00
Object	Ignore llvm.* globals.	2014-07-04 19:08:22 +00:00
Other	IR: Fold away compares between GV GEPs and GVs	2014-07-04 22:05:26 +00:00
TableGen	Fix error in tablegen when either operand of !if is an empty list.	2014-06-10 20:10:08 +00:00
tools	[llvm-readobj] Fix output of MIPS GOT without local and global entries.	2014-07-05 19:28:49 +00:00
Transforms	CodeGen: it turns out that NAND is not the same thing as BIC. At all.	2014-07-07 09:06:35 +00:00
Unit	Let test/Unit/lit.cfg add config.shlibdir to $PATH on DLL platforms like cygming.	2014-07-04 05:11:55 +00:00
Verifier	IR: Add COMDATs to the IR	2014-06-27 18:19:56 +00:00
YAMLParser
.clang-format
CMakeLists.txt	OK, NAKAMURA Takumi beat me to this change. So backing out my addition of	2014-05-19 23:26:51 +00:00
lit.cfg	Delete utils/FileUpdate.	2014-06-23 17:58:39 +00:00
lit.site.cfg.in
Makefile	lld test for configure & make	2014-06-06 09:06:25 +00:00
Makefile.tests
TestRunner.sh