https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123301
Jeffrey A. Law <law at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|[Trunk] RISC-V rv64gcv: ICE |[16 Regression] RISC-V
|at -O3 during RTL pass: |rv64gcv: ICE at -O3 during
|expand |RTL pass: expand
Ever confirmed|0 |1
CC| |kito at gcc dot gnu.org
Status|UNCONFIRMED |NEW
Last reconfirmed| |2025-12-30
--- Comment #1 from Jeffrey A. Law <law at gcc dot gnu.org> ---
Bisected to:
commit 76eb9c49275941e15cb598af5d9b58361dd65a98
Author: Kito Cheng <[email protected]>
Date: Mon Sep 1 18:29:53 2025 +0800
RISC-V: Allow VLS types using up to LMUL 8
We used to apply -mrvv-max-lmul= to limit VLS code gen, auto vectorizer,
and builtin string function expansion. But I think the VLS code gen part
doesn't
need this limit, since it only happens when the user explicitly writes
vector
types.
For example, int32x8_t under -mrvv-max-lmul=m1 with VLEN=128 would be split
into
two int32x4_t, which generate more instructions and runs slower.
In this patch, I changed -mrvv-max-lmul= to only affect auto vectorization
and
builtin string function expansion. Actually, the option's help text already
says it only controls the LMUL used by auto-vectorization, so I believe
this
change is makes sense :)
Changes since v1:
- Add testase to make sure auto vectorizer is still constrained by
-mrvv-max-lmul=.
gcc/ChangeLog:
* config/riscv/riscv-protos.h (vls_mode_valid_p): New argument
allow_up_to_lmul_8.
* config/riscv/riscv-v.cc (autovectorize_vector_modes): Set
allow_up_to_lmul_8 to false.
(vls_mode_valid_p): Add new argument allow_up_to_lmul_8, and use
it to determine whether to allow LMUL 8.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/vls-type-rvv-max-lmul.c: New test.
* gcc.target/riscv/rvv/vls-type-rvv-max-lmul-autovec.c: New
test.
I haven't done any additional debugging beyond bisection.
