Sanjay Patel dc18ebc4b1 Shrinkify libcalls: use float versions of double libm functions with fast-math (bug 17850) ... When a call to a double-precision libm function has fast-math semantics (via function attribute for now because there is no IR-level FMF on calls), we can avoid fpext/fptrunc operations and use the float version of the call if the input and output are both float. We already do this optimization using a command-line option; this patch just adds the ability for fast-math to use the existing functionality. I moved the cl::opt from InstructionCombining into SimplifyLibCalls because it's only ever used internally to that class. Modified the existing test cases to use the unsafe-fp-math attribute rather than repeating all tests. This patch should solve: http://llvm.org/bugs/show_bug.cgi?id=17850 Differential Revision: http://reviews.llvm.org/D5893 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220390 91177308-0d34-0410-b5e6-96231b3b80d8		2014-10-22 15:29:23 +00:00
..
IPO	Add control of function merging to the PMBuilder.	2014-09-13 21:46:00 +00:00
Utils	Shrinkify libcalls: use float versions of double libm functions with fast-math (bug 17850)	2014-10-22 15:29:23 +00:00
Instrumentation.h	[ASan] Completely remove sanitizer blacklist file from instrumentation pass.	2014-07-08 00:50:49 +00:00
IPO.h	Move -verify-use-list-order into llvm-uselistorder	2014-07-25 17:13:03 +00:00
ObjCARC.h	Reformat linefeeds.	2014-05-26 00:25:26 +00:00
Scalar.h	[SimplifyCFG] threshold for folding branches with common destination	2014-09-30 22:23:38 +00:00
Vectorize.h	Allow vectorization of bit intrinsics in BB Vectorizer.	2014-04-25 03:33:48 +00:00