Issue |
91369
|
Summary |
[LoopVectorize] Sign-extension miscompile
|
Labels |
new issue
|
Assignees |
|
Reporter |
patrick-rivos
|
C Testcase:
```c
long a, d;
short b[20];
short c[20];
_Bool e[20];
int f(_Bool g) {
for (long h = 0; h < 16; h += g)
b[h] = (short) g;
for (int i = 0; i < 12; i++)
for (int j = d; j; j++)
c[i] = (short) a < e[i];
return (int) b[0];
}
int main() {
__builtin_printf("%X\n", f(1));
}
```
Godbolt: https://godbolt.org/z/3E8frs1Me
With loop-vectorize: `FFFFFFFF`
Without loop-vectorize: `1`
Reduced LLVM IR:
```llvm ir
; ModuleID = 'red.ll'
source_filename = "red.c"
target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"
target triple = "riscv64-unknown-linux-gnu"
@b = global [20 x i16] zeroinitializer, align 2
@.str = constant [4 x i8] c"%X\0A\00"
; Function Attrs: nofree norecurse nosync nounwind memory(readwrite, argmem: none, inaccessiblemem: none) uwtable vscale_range(2,1024)
define i32 @f(i1 %g) #0 {
entry:
%conv = zext i1 %g to i16
%zext.g.i64 = zext i1 %g to i64
br label %for.body
for.cond3.preheader: ; preds = %for.body
%0 = load i64, ptr @b, align 2
%zero = and i64 %0, 0
%true = icmp eq i64 %zero, 0
%2 = load i16, ptr @b, align 2
%conv26 = sext i16 %2 to i32
ret i32 %conv26
for.body: ; preds = %for.body, %entry
%phi.0 = phi i64 [ 0, %entry ], [ %add, %for.body ]
%idx.0 = getelementptr inbounds [20 x i16], ptr @b, i64 0, i64 %phi.0
store i16 %conv, ptr %idx.0, align 2
%add = add nuw nsw i64 %phi.0, %zext.g.i64
%cmp = icmp ult i64 %add, 16
br i1 %cmp, label %for.body, label %for.cond3.preheader, !llvm.loop !0
}
; Function Attrs: nofree nounwind uwtable vscale_range(2,1024)
define noundef signext i32 @main() #1 {
entry:
%call = tail call signext i32 @f(i1 noundef zeroext true)
%call1 = tail call signext i32 (ptr, ...) @printf(ptr noundef nonnull @.str, i32 noundef signext %call) #3
ret i32 0
}
; Function Attrs: nofree nounwind
declare noundef signext i32 @printf(ptr nocapture noundef readonly, ...) #2
attributes #0 = { nofree norecurse nosync nounwind memory(readwrite, argmem: none, inaccessiblemem: none) uwtable vscale_range(2,1024) "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="generic-rv64" "target-features"="+64bit,+a,+c,+d,+f,+m,+relax,+v,+zicsr,+zifencei,+zve32f,+zve32x,+zve64d,+zve64f,+zve64x,+zvl128b,+zvl32b,+zvl64b,-e,-experimental-smmpm,-experimental-smnpm,-experimental-ssnpm,-experimental-sspm,-experimental-ssqosid,-experimental-supm,-experimental-zaamo,-experimental-zabha,-experimental-zalasr,-experimental-zalrsc,-experimental-zfbfmin,-experimental-zicfilp,-experimental-zicfiss,-experimental-ztso,-experimental-zvfbfmin,-experimental-zvfbfwma,-h,-shcounterenw,-shgatpa,-shtvala,-shvsatpa,-shvstvala,-shvstvecd,-smaia,-smepmp,-smstateen,-ssaia,-ssccptr,-sscofpmf,-sscounterenw,-ssstateen,-ssstrict,-sstc,-sstvala,-sstvecd,-ssu64xl,-svade,-svadu,-svbare,-svinval,-svnapot,-svpbmt,-xcvalu,-xcvbi,-xcvbitmanip,-xcvelw,-xcvmac,-xcvmem,-xcvsimd,-xsfcease,-xsfvcp,-xsfvfnrclipxfqf,-xsfvfwmaccqqq,-xsfvqmaccdod,-xsfvqmaccqoq,-xsifivecdiscarddlone,-xsifivecflushdlone,-xtheadba,-xtheadbb,-xtheadbs,-xtheadcmo,-xtheadcondmov,-xtheadfmemidx,-xtheadmac,-xtheadmemidx,-xtheadmempair,-xtheadsync,-xtheadvdot,-xventanacondops,-za128rs,-za64rs,-zacas,-zama16b,-zawrs,-zba,-zbb,-zbc,-zbkb,-zbkc,-zbkx,-zbs,-zca,-zcb,-zcd,-zce,-zcf,-zcmop,-zcmp,-zcmt,-zdinx,-zfa,-zfh,-zfhmin,-zfinx,-zhinx,-zhinxmin,-zic64b,-zicbom,-zicbop,-zicboz,-ziccamoa,-ziccif,-zicclsm,-ziccrse,-zicntr,-zicond,-zihintntl,-zihintpause,-zihpm,-zimop,-zk,-zkn,-zknd,-zkne,-zknh,-zkr,-zks,-zksed,-zksh,-zkt,-zmmul,-zvbb,-zvbc,-zvfh,-zvfhmin,-zvkb,-zvkg,-zvkn,-zvknc,-zvkned,-zvkng,-zvknha,-zvknhb,-zvks,-zvksc,-zvksed,-zvksg,-zvksh,-zvkt,-zvl1024b,-zvl16384b,-zvl2048b,-zvl256b,-zvl32768b,-zvl4096b,-zvl512b,-zvl65536b,-zvl8192b" }
attributes #1 = { nofree nounwind uwtable vscale_range(2,1024) "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="generic-rv64" "target-features"="+64bit,+a,+c,+d,+f,+m,+relax,+v,+zicsr,+zifencei,+zve32f,+zve32x,+zve64d,+zve64f,+zve64x,+zvl128b,+zvl32b,+zvl64b,-e,-experimental-smmpm,-experimental-smnpm,-experimental-ssnpm,-experimental-sspm,-experimental-ssqosid,-experimental-supm,-experimental-zaamo,-experimental-zabha,-experimental-zalasr,-experimental-zalrsc,-experimental-zfbfmin,-experimental-zicfilp,-experimental-zicfiss,-experimental-ztso,-experimental-zvfbfmin,-experimental-zvfbfwma,-h,-shcounterenw,-shgatpa,-shtvala,-shvsatpa,-shvstvala,-shvstvecd,-smaia,-smepmp,-smstateen,-ssaia,-ssccptr,-sscofpmf,-sscounterenw,-ssstateen,-ssstrict,-sstc,-sstvala,-sstvecd,-ssu64xl,-svade,-svadu,-svbare,-svinval,-svnapot,-svpbmt,-xcvalu,-xcvbi,-xcvbitmanip,-xcvelw,-xcvmac,-xcvmem,-xcvsimd,-xsfcease,-xsfvcp,-xsfvfnrclipxfqf,-xsfvfwmaccqqq,-xsfvqmaccdod,-xsfvqmaccqoq,-xsifivecdiscarddlone,-xsifivecflushdlone,-xtheadba,-xtheadbb,-xtheadbs,-xtheadcmo,-xtheadcondmov,-xtheadfmemidx,-xtheadmac,-xtheadmemidx,-xtheadmempair,-xtheadsync,-xtheadvdot,-xventanacondops,-za128rs,-za64rs,-zacas,-zama16b,-zawrs,-zba,-zbb,-zbc,-zbkb,-zbkc,-zbkx,-zbs,-zca,-zcb,-zcd,-zce,-zcf,-zcmop,-zcmp,-zcmt,-zdinx,-zfa,-zfh,-zfhmin,-zfinx,-zhinx,-zhinxmin,-zic64b,-zicbom,-zicbop,-zicboz,-ziccamoa,-ziccif,-zicclsm,-ziccrse,-zicntr,-zicond,-zihintntl,-zihintpause,-zihpm,-zimop,-zk,-zkn,-zknd,-zkne,-zknh,-zkr,-zks,-zksed,-zksh,-zkt,-zmmul,-zvbb,-zvbc,-zvfh,-zvfhmin,-zvkb,-zvkg,-zvkn,-zvknc,-zvkned,-zvkng,-zvknha,-zvknhb,-zvks,-zvksc,-zvksed,-zvksg,-zvksh,-zvkt,-zvl1024b,-zvl16384b,-zvl2048b,-zvl256b,-zvl32768b,-zvl4096b,-zvl512b,-zvl65536b,-zvl8192b" }
attributes #2 = { nofree nounwind "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="generic-rv64" "target-features"="+64bit,+a,+c,+d,+f,+m,+relax,+v,+zicsr,+zifencei,+zve32f,+zve32x,+zve64d,+zve64f,+zve64x,+zvl128b,+zvl32b,+zvl64b,-e,-experimental-smmpm,-experimental-smnpm,-experimental-ssnpm,-experimental-sspm,-experimental-ssqosid,-experimental-supm,-experimental-zaamo,-experimental-zabha,-experimental-zalasr,-experimental-zalrsc,-experimental-zfbfmin,-experimental-zicfilp,-experimental-zicfiss,-experimental-ztso,-experimental-zvfbfmin,-experimental-zvfbfwma,-h,-shcounterenw,-shgatpa,-shtvala,-shvsatpa,-shvstvala,-shvstvecd,-smaia,-smepmp,-smstateen,-ssaia,-ssccptr,-sscofpmf,-sscounterenw,-ssstateen,-ssstrict,-sstc,-sstvala,-sstvecd,-ssu64xl,-svade,-svadu,-svbare,-svinval,-svnapot,-svpbmt,-xcvalu,-xcvbi,-xcvbitmanip,-xcvelw,-xcvmac,-xcvmem,-xcvsimd,-xsfcease,-xsfvcp,-xsfvfnrclipxfqf,-xsfvfwmaccqqq,-xsfvqmaccdod,-xsfvqmaccqoq,-xsifivecdiscarddlone,-xsifivecflushdlone,-xtheadba,-xtheadbb,-xtheadbs,-xtheadcmo,-xtheadcondmov,-xtheadfmemidx,-xtheadmac,-xtheadmemidx,-xtheadmempair,-xtheadsync,-xtheadvdot,-xventanacondops,-za128rs,-za64rs,-zacas,-zama16b,-zawrs,-zba,-zbb,-zbc,-zbkb,-zbkc,-zbkx,-zbs,-zca,-zcb,-zcd,-zce,-zcf,-zcmop,-zcmp,-zcmt,-zdinx,-zfa,-zfh,-zfhmin,-zfinx,-zhinx,-zhinxmin,-zic64b,-zicbom,-zicbop,-zicboz,-ziccamoa,-ziccif,-zicclsm,-ziccrse,-zicntr,-zicond,-zihintntl,-zihintpause,-zihpm,-zimop,-zk,-zkn,-zknd,-zkne,-zknh,-zkr,-zks,-zksed,-zksh,-zkt,-zmmul,-zvbb,-zvbc,-zvfh,-zvfhmin,-zvkb,-zvkg,-zvkn,-zvknc,-zvkned,-zvkng,-zvknha,-zvknhb,-zvks,-zvksc,-zvksed,-zvksg,-zvksh,-zvkt,-zvl1024b,-zvl16384b,-zvl2048b,-zvl256b,-zvl32768b,-zvl4096b,-zvl512b,-zvl65536b,-zvl8192b" }
attributes #3 = { nounwind }
!0 = distinct !{!0, !1}
!1 = !{!"llvm.loop.mustprogress"}
```
LoopVectorize pass.
Looks like a sext/zext related issue.
_______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs