https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98917
Bug ID: 98917 Summary: SVE: wrong code with -O -ftree-vectorize -msve-vector-bits=128 --param=aarch64-autovec-preference=2 Product: gcc Version: 11.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: acoplan at gcc dot gnu.org Target Milestone: --- Created attachment 50109 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=50109&action=edit reduced testcase The attached C++ testcase appears to be miscompiled with -O -ftree-vectorize -march=armv8.2-a+sve -msve-vector-bits=128 --param=aarch64-autovec-preference=2. The testcase comes from YARPGen and is hard to reduce. It runs fine with/without sanitizers at various optimisation levels under GCC/clang for both x86 and aarch64 but segfaults with the above options on trunk.