On 3/7/25 4:05 AM, Robin Dapp wrote:
Hi,
in PR119115 we end up with an orphaned
vsetvli zero,t1,e16,m1,ta,ma.
t1 originally came from another vsetvl that was fused from
vsetvli a4,a3,e8,mf2,ta,ma
vsetvli t1,a3,e8,mf2,ta,ma (1)
to
vsetvli zero,a3,e16,m1,ta,ma.
This patch checks if t1, the VL operand of (1), has AVL uses and does
not delete the vsetvl if so. While doing so, it also wraps the search
for VL uses into two new functions reg_used and reg_single_use_in_avl.
Regtested on rv64gcv_zvl512b and rv64gcv_zvl256b. Let's see what the CI
says. For my last patch it showed an execution failure that I cannot
reproduce locally.
Regards
Robin
PR target/119115
gcc/ChangeLog:
* config/riscv/riscv-vsetvl.cc (reg_used): New function.
(reg_single_use_in_avl): Ditto.
(pre_vsetvl::fuse_local_vsetvl_info): Use reg_single_use_in_avl
when checking if vsetvl can be deleted.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/base/pr119115.c: New test.
OK. Sorry about the delay.
jeff
