[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v3 03/15] target/arm: Do not use aarch64_sve_zcr_get_valid_le
From: |
Peter Maydell |
Subject: |
Re: [PATCH v3 03/15] target/arm: Do not use aarch64_sve_zcr_get_valid_len in reset |
Date: |
Tue, 31 May 2022 15:55:42 +0100 |
On Tue, 31 May 2022 at 15:28, Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> On 5/31/22 05:15, Peter Maydell wrote:
> > On Fri, 27 May 2022 at 19:07, Richard Henderson
> > <richard.henderson@linaro.org> wrote:
> >>
> >> We don't need to constrain the value set in zcr_el[1],
> >> because it will be done by sve_zcr_len_for_el.
> >>
> >> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> >> ---
> >> target/arm/cpu.c | 3 +--
> >> 1 file changed, 1 insertion(+), 2 deletions(-)
> >>
> >> diff --git a/target/arm/cpu.c b/target/arm/cpu.c
> >> index d2bd74c2ed..0621944167 100644
> >> --- a/target/arm/cpu.c
> >> +++ b/target/arm/cpu.c
> >> @@ -208,8 +208,7 @@ static void arm_cpu_reset(DeviceState *dev)
> >> CPACR_EL1, ZEN, 3);
> >> /* with reasonable vector length */
> >> if (cpu_isar_feature(aa64_sve, cpu)) {
> >> - env->vfp.zcr_el[1] =
> >> - aarch64_sve_zcr_get_valid_len(cpu, cpu->sve_default_vq -
> >> 1);
> >> + env->vfp.zcr_el[1] = cpu->sve_default_vq - 1;
> >> }
> >
> > I'm still not a fan of the zcr_el[] value not actually being
> > a valid one. I'd rather we constrained it when we write the
> > value into the field.
>
> It is an architecturally valid value, exactly like the kernel might write
> while probing
> for supported vector lengths. It will result in this, or the next smaller
> supported
> vector size, being selected.
Mmm, I guess so (having re-read the ZCR_EL1 docs).
-- PMM
- [PATCH v3 00/15] target/arm: SME prep patches, Richard Henderson, 2022/05/27
- [PATCH v3 06/15] target/arm: Rename sve_zcr_len_for_el to sve_vqm1_for_el, Richard Henderson, 2022/05/27
- [PATCH v3 08/15] target/arm: Add el_is_in_host, Richard Henderson, 2022/05/27
- [PATCH v3 09/15] target/arm: Use el_is_in_host for sve_vqm1_for_el, Richard Henderson, 2022/05/27
- [PATCH v3 10/15] target/arm: Split out load/store primitives to sve_ldst_internal.h, Richard Henderson, 2022/05/27
- [PATCH v3 12/15] target/arm: Move expand_pred_b to vec_internal.h, Richard Henderson, 2022/05/27