https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115375
Bug ID: 115375
Summary: [15 Regression] RISCV scan failures since 2024-05-04
Product: gcc
Version: 15.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: patrick at rivosinc dot com
Target Milestone: ---
Commit range:
https://github.com/gcc-mirror/gcc/compare/4d207044195b97ecb27c72a7dc987eb8b86644a0...00fb385a25a7fbaa9c7060ddd5f41a8c3b1548d1
Good: r15-1003-g4d207044195
Bad: r15-1010-g00fb385a25a
Regressions:
FAIL: gcc.dg/vect/costmodel/riscv/rvv/dynamic-lmul8-8.c scan-assembler-not csrr
FAIL: gcc.dg/vect/costmodel/riscv/rvv/dynamic-lmul8-9.c scan-assembler-not csrr
FAIL: gcc.dg/vect/costmodel/riscv/rvv/dynamic-lmul8-9.c scan-tree-dump-times
vect "Maximum lmul = 8" 2
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvfwredusum\\.vs\\tv[0-9]+,v[0-9]+,v[0-9]+,v0\\.t 3
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvsext\\.vf4\\tv[0-9]+,v[0-9]+,v0\\.t 2
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvsext\\.vf8\\tv[0-9]+,v[0-9]+,v0\\.t 1
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvwredsum\\.vs\\tv[0-9]+,v[0-9]+,v[0-9]+,v0\\.t 3
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvwredsumu\\.vs\\tv[0-9]+,v[0-9]+,v[0-9]+,v0\\.t 3
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvzext\\.vf4\\tv[0-9]+,v[0-9]+,v0\\.t 2
FAIL: gcc.target/riscv/rvv/autovec/cond/cond_widen_reduc-2.c
scan-assembler-times \\tvzext\\.vf8\\tv[0-9]+,v[0-9]+,v0\\.t 1
FAIL: gcc.target/riscv/rvv/autovec/pr112399.c -O3 -ftree-vectorize
check-function-bodies foo
FAIL: gcc.target/riscv/rvv/base/pr112431-42.c scan-assembler-not csrr
Relevant postcommit issue:
https://github.com/patrick-rivos/gcc-postcommit-ci/issues/1056
>From the precommit: https://github.com/ewlu/gcc-precommit-ci/issues/1669
it looks like this was probably caused by this series:
r15-1005-g0592000aeed Avoid inserting after a GIMPLE_COND with SLP and early
break
r15-1006-gd93353e6423 Do single-lane SLP discovery for reductions
Might just need to update the scans since these are all riscv/* testcases.
