search for: rbuild1

Displaying 3 results from an estimated 3 matches for "rbuild1".

Did you mean: rbuild
2017 Aug 02
3
[InstCombine] Simplification sometimes only transforms but doesn't simplify instruction, causing side effect in other pass
...ressInstruction pass for it. After instcombine, %conv.i's live range is extended, both %and7.i and %conv.i are alive after the "and" instruction, so TwoAddressInstruction pass will generate an extra mov for %and7.i = and i32 %conv.i. *** After instcombine: *** ~/workarea/llvm-r309240/rbuild1/bin/opt -instcombine -S < a.ll > b.ll; cat b.ll @a = global i64 0, align 8 @b = global i32 0, align 8 define i32 @_Z25InternalUncompressAllTagsv(i16* %arrayidx) local_unnamed_addr { entry: %t1 = load i16, i16* %arrayidx, align 2 %conv.i = zext i16 %t1 to i32 %and.i = and i32 %conv.i, 2...
2017 Aug 02
3
[InstCombine] Simplification sometimes only transforms but doesn't simplify instruction, causing side effect in other pass
....i's live >> range is extended, both %and7.i and %conv.i are alive after the "and" >> instruction, so TwoAddressInstruction pass will generate an extra mov >> for %and7.i = and i32 %conv.i. >> >> *** After instcombine: *** >> ~/workarea/llvm-r309240/rbuild1/bin/opt -instcombine -S < a.ll > b.ll; cat b.ll >> @a = global i64 0, align 8 >> @b = global i32 0, align 8 >> >> define i32 @_Z25InternalUncompressAllTagsv(i16* %arrayidx) local_unnamed_addr { >> entry: >> %t1 = load i16, i16* %arrayidx, align 2 >>...
2017 Aug 02
2
[InstCombine] Simplification sometimes only transforms but doesn't simplify instruction, causing side effect in other pass
...After instcombine, %conv.i's live > range is extended, both %and7.i and %conv.i are alive after the "and" > instruction, so TwoAddressInstruction pass will generate an extra mov > for %and7.i = and i32 %conv.i. > > *** After instcombine: *** > ~/workarea/llvm-r309240/rbuild1/bin/opt -instcombine -S < a.ll > b.ll; cat > b.ll > @a = global i64 0, align 8 > @b = global i32 0, align 8 > > define i32 @_Z25InternalUncompressAllTagsv(i16* %arrayidx) > local_unnamed_addr { > entry: > %t1 = load i16, i16* %arrayidx, align 2 > %conv.i = zext i16...