Re: [PATCH v9 04/10] target/riscv/vector_helper.c: update tail with vext_set_tail_elems_1s()

On 2024/3/10 4:43, Daniel Henrique Barboza wrote: Change all code that updates tail elems to use vext_set_tail_elems_1s() instead of vext_set_elems_1s(). Hi Daniel, Notice vext_set_tail_elems_1s will use NF field, which is zero for most vector instructions. Thus it will do nothing. I think

Re: [PATCH v9 04/10] target/riscv/vector_helper.c: update tail with vext_set_tail_elems_1s()

On 3/10/24 04:41, Richard Henderson wrote: On 3/9/24 10:43, Daniel Henrique Barboza wrote: Change all code that updates tail elems to use vext_set_tail_elems_1s() instead of vext_set_elems_1s(). Setting 'env->vstart=0' needs to be the very last thing a helper does because env->vstart is bein

Re: [PATCH v9 04/10] target/riscv/vector_helper.c: update tail with vext_set_tail_elems_1s()

On 3/9/24 10:43, Daniel Henrique Barboza wrote: Change all code that updates tail elems to use vext_set_tail_elems_1s() instead of vext_set_elems_1s(). Setting 'env->vstart=0' needs to be the very last thing a helper does because env->vstart is being checked by vext_set_tail_elems_1s(). I did

[PATCH v9 04/10] target/riscv/vector_helper.c: update tail with vext_set_tail_elems_1s()

Change all code that updates tail elems to use vext_set_tail_elems_1s() instead of vext_set_elems_1s(). Setting 'env->vstart=0' needs to be the very last thing a helper does because env->vstart is being checked by vext_set_tail_elems_1s(). A side effect of this change is that a lot of 'vta' local