maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

This has been implemented. 

llvm-svn: 125738
This commit is contained in:
Duncan Sands 2011-02-17 08:16:56 +00:00
parent 5ee06f02f9
commit 491eb276a7
1 changed files with 0 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
llvm/lib/Target/README.txt
View File

@ -2225,16 +2225,6 @@ avoids partial register stalls in some important cases.
//===---------------------------------------------------------------------===//
We should fold compares like this:
%1266 = add nsw i32 %.84.i.i.i, 1
%560 = add nsw i32 %556, 1
%1267 = icmp slt i32 %1266, %560
to a single 'icmp slt' when the add's have a single use, since they are NSW.
//===---------------------------------------------------------------------===//
We don't fold (icmp (add) (add)) unless the two adds only have a single use.
There are a lot of cases that we're refusing to fold in (e.g.) 256.bzip2, for
example: