mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-16 11:30:51 +00:00
10bb82e54f
Patch by Michele Scandale! Rewrite of the functions used to compute the backedge taken count of a loop on LT and GT comparisons. I decided to split the handling of LT and GT cases becasue the trick "a > b == -a < -b" in some cases prevents the trip count computation due to the multiplication by -1 on the two operands of the comparison. This issue comes from the conservative computation of value range of SCEVs: taking the negative SCEV of an expression that have a small positive range (e.g. [0,31]), we would have a SCEV with a fullset as value range. Indeed, in the new rewritten function I tried to better handle the maximum backedge taken count computation when MAX/MIN expression are used to handle the cases where no entry guard is found. Some test have been modified in order to check the new value correctly (I manually check them and reasoning on possible overflow the new values seem correct). I finally added a new test case related to the multiplication by -1 issue on GT comparisons. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194116 91177308-0d34-0410-b5e6-96231b3b80d8
39 lines
1.2 KiB
LLVM
39 lines
1.2 KiB
LLVM
; RUN: opt < %s -analyze -scalar-evolution | FileCheck %s
|
|
|
|
; CHECK: Loop %bb: backedge-taken count is ((-5 + %x) /u 3)
|
|
; CHECK: Loop %bb: max backedge-taken count is 1431655764
|
|
|
|
|
|
; ScalarEvolution can't compute a trip count because it doesn't know if
|
|
; dividing by the stride will have a remainder. This could theoretically
|
|
; be teaching it how to use a more elaborate trip count computation.
|
|
|
|
define i32 @f(i32 %x) nounwind readnone {
|
|
entry:
|
|
%0 = icmp ugt i32 %x, 4 ; <i1> [#uses=1]
|
|
br i1 %0, label %bb.nph, label %bb2
|
|
|
|
bb.nph: ; preds = %entry
|
|
br label %bb
|
|
|
|
bb: ; preds = %bb.nph, %bb1
|
|
%indvar = phi i32 [ 0, %bb.nph ], [ %indvar.next, %bb1 ] ; <i32> [#uses=2]
|
|
%tmp = mul i32 %indvar, -3 ; <i32> [#uses=1]
|
|
%x_addr.04 = add i32 %tmp, %x ; <i32> [#uses=1]
|
|
%1 = add i32 %x_addr.04, -3 ; <i32> [#uses=2]
|
|
br label %bb1
|
|
|
|
bb1: ; preds = %bb
|
|
%2 = icmp ugt i32 %1, 4 ; <i1> [#uses=1]
|
|
%indvar.next = add i32 %indvar, 1 ; <i32> [#uses=1]
|
|
br i1 %2, label %bb, label %bb1.bb2_crit_edge
|
|
|
|
bb1.bb2_crit_edge: ; preds = %bb1
|
|
%.lcssa = phi i32 [ %1, %bb1 ] ; <i32> [#uses=1]
|
|
br label %bb2
|
|
|
|
bb2: ; preds = %bb1.bb2_crit_edge, %entry
|
|
%x_addr.0.lcssa = phi i32 [ %.lcssa, %bb1.bb2_crit_edge ], [ %x, %entry ] ; <i32> [#uses=1]
|
|
ret i32 %x_addr.0.lcssa
|
|
}
|