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

Fix some typos in a comment. 

llvm-svn: 76249
This commit is contained in:
Dan Gohman 2009-07-17 22:16:21 +00:00
parent 18fe458103
commit a565d4f937
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
llvm/lib/Transforms/Scalar/InstructionCombining.cpp
View File

@ -5456,13 +5456,13 @@ static Value *EmitGEPOffset(User *GEP, Instruction &I, InstCombiner &IC) {
}
/// EvaluateGEPOffsetExpression - Return an value that can be used to compare of
/// the *offset* implied by GEP to zero. For example, if we have &A[i], we want
/// to return 'i' for "icmp ne i, 0". Note that, in general, indices can be
/// complex, and scales are involved. The above expression would also be legal
/// to codegen as "icmp ne (i*4), 0" (assuming A is a pointer to i32). This
/// later form is less amenable to optimization though, and we are allowed to
/// generate the first by knowing that pointer arithmetic doesn't overflow.
/// EvaluateGEPOffsetExpression - Return a value that can be used to compare
/// the *offset* implied by a GEP to zero. For example, if we have &A[i], we
/// want to return 'i' for "icmp ne i, 0". Note that, in general, indices can
/// be complex, and scales are involved. The above expression would also be
/// legal to codegen as "icmp ne (i*4), 0" (assuming A is a pointer to i32).
/// This later form is less amenable to optimization though, and we are allowed
/// to generate the first by knowing that pointer arithmetic doesn't overflow.
///
/// If we can't emit an optimized form for this expression, this returns null.
///