Chandler Carruth 474be0d0f8 Teach the SLP vectorizer the correct way to check for consecutive access ... using GEPs. Previously, it used a number of different heuristics for analyzing the GEPs. Several of these were conservatively correct, but failed to fall back to SCEV even when SCEV might have given a reasonable answer. One was simply incorrect in how it was formulated. There was good code already to recursively evaluate the constant offsets in GEPs, look through pointer casts, etc. I gathered this into a form code like the SLP code can use in a previous commit, which allows all of this code to become quite simple. There is some performance (compile time) concern here at first glance as we're directly attempting to walk both pointers constant GEP chains. However, a couple of thoughts: 1) The very common cases where there is a dynamic pointer, and a second pointer at a constant offset (usually a stride) from it, this code will actually not do any unnecessary work. 2) InstCombine and other passes work very hard to collapse constant GEPs, so it will be rare that we iterate here for a long time. That said, if there remain performance problems here, there are some obvious things that can improve the situation immensely. Doing a vectorizer-pass-wide memoizer for each individual layer of pointer values, their base values, and the constant offset is likely to be able to completely remove redundant work and strictly limit the scaling of the work to scrape these GEPs. Since this optimization was not done on the prior version (which would still benefit from it), I've not done it here. But if folks have benchmarks that slow down it should be straight forward for them to add. I've added a test case, but I'm not really confident of the amount of testing done for different access patterns, strides, and pointer manipulation. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@189007 91177308-0d34-0410-b5e6-96231b3b80d8		2013-08-22 12:45:17 +00:00
..
Analysis	Reorder headers according to lint.	2013-08-21 21:14:19 +00:00
AsmParser
Bitcode
CodeGen	[stackprotector] When finding the split point to splice off the end of a parentmbb into a successmbb, include any DBG_VALUE MI.	2013-08-22 05:40:50 +00:00
DebugInfo	llvm-dwarfdump: Do not include address offsets for attributes, only for tags	2013-08-19 03:36:23 +00:00
ExecutionEngine	Reorder headers according to lint.	2013-08-21 21:14:19 +00:00
IR	Add a new helper method to Value to strip in-bounds constant offsets of	2013-08-22 11:25:11 +00:00
IRReader
Linker
MC	MC CFG: Remap enough for data too, analoguous to r188873.	2013-08-21 19:40:28 +00:00
Object	Move #includes from .h to .cpp file.	2013-08-21 01:20:11 +00:00
Option	Options: explicit handling of --	2013-08-13 22:23:05 +00:00
Support	MemoryBuffer.cpp: Consider if PageSize were not 4096 in shouldUseMmap(). Follow-up to r188903.	2013-08-22 10:23:52 +00:00
TableGen	Add an error check for a typo I accidentally made in a td file that caused an assert to fire.	2013-08-20 04:22:09 +00:00
Target	[ARM] Constrain some register classes in EmitAtomicBinary64 so that	2013-08-22 12:19:24 +00:00
Transforms	Teach the SLP vectorizer the correct way to check for consecutive access	2013-08-22 12:45:17 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile