llvm-project/llvm/lib/Transforms
Chris Lattner 98457101fc Enhance the "compare with shift" and "compare with div"
optimizations to be much more aggressive in the face of
exact/nsw/nuw div and shifts.  For example, these (which
are the same except the first is 'exact' sdiv:

define i1 @sdiv_icmp4_exact(i64 %X) nounwind {
  %A = sdiv exact i64 %X, -5   ; X/-5 == 0 --> x == 0
  %B = icmp eq i64 %A, 0
  ret i1 %B
}

define i1 @sdiv_icmp4(i64 %X) nounwind {
  %A = sdiv i64 %X, -5   ; X/-5 == 0 --> x == 0
  %B = icmp eq i64 %A, 0
  ret i1 %B
}

compile down to:

define i1 @sdiv_icmp4_exact(i64 %X) nounwind {
  %1 = icmp eq i64 %X, 0
  ret i1 %1
}

define i1 @sdiv_icmp4(i64 %X) nounwind {
  %X.off = add i64 %X, 4
  %1 = icmp ult i64 %X.off, 9
  ret i1 %1
}

This happens when you do something like:
  (ptr1-ptr2) == 42

where the pointers are pointers to non-unit types.

llvm-svn: 125266
2011-02-10 05:23:05 +00:00
..
Hello Since the Hello pass is built as a loadable dynamic library, don't try to convert it to new-style registration yet. 2010-10-07 00:31:16 +00:00
IPO When removing a function from the function set and adding it to deferred, we 2011-02-09 06:32:02 +00:00
InstCombine Enhance the "compare with shift" and "compare with div" 2011-02-10 05:23:05 +00:00
Instrumentation Implementation of path profiling. 2011-01-29 01:09:53 +00:00
Scalar Revert this in an attempt to bring the builders back. 2011-02-10 01:48:24 +00:00
Utils SimplifyCFG: Track the number of used icmps when turning a icmp chain into a switch. If we used only one icmp, don't turn it into a switch. 2011-02-07 22:37:28 +00:00
Makefile Revert r100896 and around - this breaks the only mingw32 buildbot we have. 2010-04-15 19:51:42 +00:00