llvm-project/llvm/lib/Transforms
Sanjay Patel 57fd1dc5db transform fmin/fmax calls when possible (PR24314)
If we can ignore NaNs, fmin/fmax libcalls can become compare and select
(this is what we turn std::min / std::max into).

This IR should then be optimized in the backend to whatever is best for
any given target. Eg, x86 can use minss/maxss instructions.

This should solve PR24314:
https://llvm.org/bugs/show_bug.cgi?id=24314

Differential Revision: http://reviews.llvm.org/D11866

llvm-svn: 245187
2015-08-16 20:18:19 +00:00
..
Hello Revert r240137 (Fixed/added namespace ending comments using clang-tidy. NFC) 2015-06-23 09:49:53 +00:00
IPO MergeFunc: Quick fix for r245140, Ignore second, aka Function*, in sorting. 2015-08-16 02:41:23 +00:00
InstCombine [InstCombine] Replace an and+icmp with a trunc+icmp 2015-08-16 07:09:17 +00:00
Instrumentation Try to appease VS 2015 warnings from http://reviews.llvm.org/D11890 2015-08-15 19:06:14 +00:00
ObjCARC [PM/AA] Run clang-format over the ObjCARC Alias Analysis code to 2015-08-14 03:57:00 +00:00
Scalar [LSR][NFC] Don’t duplicate entity name at the beginning of the comment. 2015-08-16 18:22:46 +00:00
Utils transform fmin/fmax calls when possible (PR24314) 2015-08-16 20:18:19 +00:00
Vectorize [PM/AA] Explicitly depend on TLI rather than getting it out of the 2015-08-12 18:06:08 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile