https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116161
Bug ID: 116161
Summary: -fcompare-debug -gno-statement-frontiers failure with
-O3 -march=native -fno-vect-cost-model when building
rapidfuzz
Product: gcc
Version: 15.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: middle-end
Assignee: unassigned at gcc dot gnu.org
Reporter: sjames at gcc dot gnu.org
Target Milestone: ---
Created attachment 58787
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58787&action=edit
fuzz_cupp_avx2.ii.xz
The testcase is a bit weird and sensitive, e.g. -march=znver2 isn't OK,
-march=native is required, and with no optimisation, I get 'error: the last
argument must be an 8-bit immediate'. It's also very slow to compile.
```
$ g++ -c fuzz_cupp_avx2.ii -fcompare-debug -gno-statement-frontiers
-march=native -fno-vect-cost-model -O2
$ g++ -c fuzz_cupp_avx2.ii -fcompare-debug -gno-statement-frontiers
-march=native -fno-vect-cost-model -O3
g++: error: fuzz_cupp_avx2.ii: ‘-fcompare-debug’ failure
```
