search for: conv15

...BBV: fusing loop #1 for for.body10 in main... BBV: found 22 instructions with candidate pairs BBV: found 82 pair connections. BBV: selected pairs in the best tree for: %0 = load i8* %r.063, align 1, !tbaa !0 BBV: selected pair: %mul23 = mul nsw i32 %conv14, 234 <-> %mul35 = mul nsw i32 %conv15, 543 BBV: selected pair: %0 = load i8* %r.063, align 1, !tbaa !0 <-> %1 = load i8* %incdec.ptr11, align 1, !tbaa !0 BBV: selected pair: %conv14 = zext i8 %0 to i32 <-> %conv15 = zext i8 %1 to i32 BBV: selected pair: %add26 = add i32 %mul25, %mul23 <-> %add36 = add i32...

On Thu, 2012-01-26 at 15:12 -0600, Sebastian Pop wrote: > On Thu, Jan 26, 2012 at 2:49 PM, Hal Finkel <hfinkel at anl.gov> wrote: > > Thanks! Did you compile with any non-default flags other than -mllvm > > -vectorize? > > I used -O3 and -vectorize, no other non-default flags. If I run clang -O3 -mllvm -vectorize -S -emit-llvm -o test.ll test.c then I get no

...;getValueName()<<" | " <<previous->getOperand(0)->getValueID()<<" | " <<previous->getOperand(0)->getNumUses()<<" TTTTT\n"; The results with getOperand(1) are similar. The output is: *PREVIOUS: store i32 %conv15, i32* %i, align 4 TTTTT conv15 | 0x9b69090 | 59 | 1 TTTTT ... *PREVIOUS: store i32 %inc13, i32* %i, align 4 TTTTT inc13 | 0x9b76478 | 30 | 1 TTTTT ... *PREVIOUS: %cmp11 = icmp sgt i32 %8, 3 TTTTT | 0x0 | 49 | 1 TTTTT ... *PREVIOUS: store i32 %dec, i32* %i, align 4 TTTTT dec | 0x9b69130 | 30...

...> <<previous->getOperand(0)->getValueID()<<" | " > <<previous->getOperand(0)->getNumUses()<<" TTTTT\n"; > | > > The results with getOperand(1) are similar. The output is: > > |*PREVIOUS: store i32 %conv15, i32* %i, align 4 > TTTTT conv15 | 0x9b69090 | 59 | 1 TTTTT > ... > *PREVIOUS: store i32 %inc13, i32* %i, align 4 > TTTTT inc13 | 0x9b76478 | 30 | 1 TTTTT > ... > *PREVIOUS: %cmp11 = icmp sgt i32 %8, 3 > TTTTT | 0x0 | 49 | 1 TTTTT > ... > *PREVIOUS: store i32 %dec,...

On Thu, Jan 26, 2012 at 3:41 PM, Hal Finkel <hfinkel at anl.gov> wrote: > On Thu, 2012-01-26 at 15:36 -0600, Sebastian Pop wrote: >> arm-none-linux-gnueabi > > Indeed, adding -ccc-host-triple arm-none-linux-gnueabi I also get Minor remark: please use -target instead of -ccc-host-triple that is now deprecated. Thanks for looking at this testcase. Sebastian -- Qualcomm

On Thu, 2012-01-26 at 15:36 -0600, Sebastian Pop wrote: > arm-none-linux-gnueabi Indeed, adding -ccc-host-triple arm-none-linux-gnueabi I also get vectorization (even though I don't get vectorization when targeting x86_64). I'll let you know what I find. -Hal -- Hal Finkel Postdoctoral Appointee Leadership Computing Facility Argonne National Laboratory

...vious->getOperand(0)->**getValueID()<<" | " >> <<previous->getOperand(0)->**getNumUses()<<" TTTTT\n"; >> | >> >> The results with getOperand(1) are similar. The output is: >> >> |*PREVIOUS: store i32 %conv15, i32* %i, align 4 >> TTTTT conv15 | 0x9b69090 | 59 | 1 TTTTT >> ... >> *PREVIOUS: store i32 %inc13, i32* %i, align 4 >> TTTTT inc13 | 0x9b76478 | 30 | 1 TTTTT >> ... >> *PREVIOUS: %cmp11 = icmp sgt i32 %8, 3 >> TTTTT | 0x0 | 49 | 1 TTTTT >> ... &...

...cay.i.i.i.i, align 2, !tbaa !5 %conv.i.i.i.i = zext i16 %144 to i32 %neg.i.i.i.i = xor i32 %conv.i.i.i.i, 65535 %conv11.i.i.i.i = trunc i32 %neg.i.i.i.i to i16 store i16 %conv11.i.i.i.i, i16* %arraydecay.i.i.i.i, align 2, !tbaa !5 %145 = load i16* %arrayidx14.i.i.i.i, align 2, !tbaa !5 %conv15.i.i.i.i = zext i16 %145 to i32 %146 = load i16* %arrayidx16.i.i.i.i, align 2, !tbaa !5 %conv17.i.i.i.i = zext i16 %146 to i32 %sub.i.i.i.i = sub nsw i32 %conv15.i.i.i.i, %conv17.i.i.i.i %cmp.i3.i.i.i = icmp sgt i32 %sub.i.i.i.i, 0 br i1 %cmp.i3.i.i.i, label %if.then19.i.i.i.i, label %if.e...