https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108270
--- Comment #2 from CVS Commits <cvs-commit at gcc dot gnu.org> --- The master branch has been updated by Kito Cheng <k...@gcc.gnu.org>: https://gcc.gnu.org/g:d06e9264b0192c2c77e07d7fb0fe090efcb510c0 commit r14-135-gd06e9264b0192c2c77e07d7fb0fe090efcb510c0 Author: Juzhe-Zhong <juzhe.zh...@rivai.ai> Date: Fri Apr 21 17:19:12 2023 +0800 RISC-V: Defer vsetvli insertion to later if possible [PR108270] Fix issue: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108270. Consider the following testcase: void f (void * restrict in, void * restrict out, int l, int n, int m) { for (int i = 0; i < l; i++){ for (int j = 0; j < m; j++){ for (int k = 0; k < n; k++) { vint8mf8_t v = __riscv_vle8_v_i8mf8 (in + i + j, 17); __riscv_vse8_v_i8mf8 (out + i + j, v, 17); } } } } Compile option: -O3 Before this patch: mv a7,a2 mv a6,a0 mv t1,a1 mv a2,a3 vsetivli zero,17,e8,mf8,ta,ma ble a7,zero,.L1 ble a4,zero,.L1 ble a3,zero,.L1 ... After this patch: mv a7,a2 mv a6,a0 mv t1,a1 mv a2,a3 ble a7,zero,.L1 ble a4,zero,.L1 ble a3,zero,.L1 add a1,a0,a4 li a0,0 vsetivli zero,17,e8,mf8,ta,ma ... This issue is a missed optmization produced by Phase 3 global backward demand fusion instead of LCM. This patch is fixing poor placement of the vsetvl. This point is seletected not because LCM but by Phase 3 (VL/VTYPE demand info backward fusion and propogation) which is I introduced into VSETVL PASS to enhance LCM && improve vsetvl instruction performance. This patch is to supress the Phase 3 too aggressive backward fusion and propagation to the top of the function program when there is no define instruction of AVL (AVL is 0 ~ 31 imm since vsetivli instruction allows imm value instead of reg). You may want to ask why we need Phase 3 to the job. Well, we have so many situations that pure LCM fails to optimize, here I can show you a simple case to demonstrate it: void f (void * restrict in, void * restrict out, int n, int m, int cond) { size_t vl = 101; for (size_t j = 0; j < m; j++){ if (cond) { for (size_t i = 0; i < n; i++) { vint8mf8_t v = __riscv_vle8_v_i8mf8 (in + i + j, vl); __riscv_vse8_v_i8mf8 (out + i, v, vl); } } else { for (size_t i = 0; i < n; i++) { vint32mf2_t v = __riscv_vle32_v_i32mf2 (in + i + j, vl); v = __riscv_vadd_vv_i32mf2 (v,v,vl); __riscv_vse32_v_i32mf2 (out + i, v, vl); } } } } You can see: The first inner loop needs vsetvli e8 mf8 for vle+vse. The second inner loop need vsetvli e32 mf2 for vle+vadd+vse. If we don't have Phase 3 (Only handled by LCM (Phase 4)), we will end up with : outerloop: ... vsetvli e8mf8 inner loop 1: .... vsetvli e32mf2 inner loop 2: .... However, if we have Phase 3, Phase 3 is going to fuse the vsetvli e32 mf2 of inner loop 2 into vsetvli e8 mf8, then we will end up with this result after phase 3: outerloop: ... inner loop 1: vsetvli e32mf2 .... inner loop 2: vsetvli e32mf2 .... Then, this demand information after phase 3 will be well optimized after phase 4 (LCM), after Phase 4 result is: vsetvli e32mf2 outerloop: ... inner loop 1: .... inner loop 2: .... You can see this is the optimal codegen after current VSETVL PASS (Phase 3: Demand backward fusion and propagation + Phase 4: LCM ). This is a known issue when I start to implement VSETVL PASS. gcc/ChangeLog: PR target/108270 * config/riscv/riscv-vsetvl.cc (vector_infos_manager::all_empty_predecessor_p): New function. (pass_vsetvl::backward_demand_fusion): Ditto. * config/riscv/riscv-vsetvl.h: Ditto. gcc/testsuite/ChangeLog: PR target/108270 * gcc.target/riscv/rvv/vsetvl/imm_bb_prop-1.c: Adapt testcase. * gcc.target/riscv/rvv/vsetvl/imm_conflict-3.c: Ditto. * gcc.target/riscv/rvv/vsetvl/pr108270.c: New test.