From: Richard Henderson <richard.henderson@linaro.org>
To: Cornelia Huck <cohuck@redhat.com>,
qemu-devel@nongnu.org, qemu-arm@nongnu.org
Subject: Re: [PATCH v2 02/14] arm/kvm: add accessors for storing host features into idregs
Date: Wed, 5 Mar 2025 17:15:11 -0800 [thread overview]
Message-ID: <8f4b598e-d09d-4d2e-afcc-317a6997ad02@linaro.org> (raw)
In-Reply-To: <20250305163819.2477553-3-cohuck@redhat.com>
On 3/5/25 08:38, Cornelia Huck wrote:
> Signed-off-by: Cornelia Huck <cohuck@redhat.com>
> ---
> target/arm/cpu-sysregs.h | 3 +++
> target/arm/cpu64.c | 25 +++++++++++++++++++++++++
> target/arm/kvm.c | 12 ++++++++++++
> 3 files changed, 40 insertions(+)
>
> diff --git a/target/arm/cpu-sysregs.h b/target/arm/cpu-sysregs.h
> index de09ebae91a5..54a4fadbf0c1 100644
> --- a/target/arm/cpu-sysregs.h
> +++ b/target/arm/cpu-sysregs.h
> @@ -128,4 +128,7 @@ static const uint32_t id_register_sysreg[NUM_ID_IDX] = {
> [CTR_EL0_IDX] = SYS_CTR_EL0,
> };
>
> +int get_sysreg_idx(ARMSysRegs sysreg);
> +uint64_t idregs_sysreg_to_kvm_reg(ARMSysRegs sysreg);
> +
> #endif /* ARM_CPU_SYSREGS_H */
> diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c
> index 8188ede5cc8a..9ae78253cb34 100644
> --- a/target/arm/cpu64.c
> +++ b/target/arm/cpu64.c
> @@ -736,6 +736,31 @@ static void aarch64_a53_initfn(Object *obj)
> define_cortex_a72_a57_a53_cp_reginfo(cpu);
> }
>
> +#ifdef CONFIG_KVM
> +
> +int get_sysreg_idx(ARMSysRegs sysreg)
> +{
> + int i;
> +
> + for (i = 0; i < NUM_ID_IDX; i++) {
> + if (id_register_sysreg[i] == sysreg) {
> + return i;
> + }
> + }
> + return -1;
> +}
> +
> +uint64_t idregs_sysreg_to_kvm_reg(ARMSysRegs sysreg)
> +{
> + return ARM64_SYS_REG((sysreg & CP_REG_ARM64_SYSREG_OP0_MASK) >> CP_REG_ARM64_SYSREG_OP0_SHIFT,
> + (sysreg & CP_REG_ARM64_SYSREG_OP1_MASK) >> CP_REG_ARM64_SYSREG_OP1_SHIFT,
> + (sysreg & CP_REG_ARM64_SYSREG_CRN_MASK) >> CP_REG_ARM64_SYSREG_CRN_SHIFT,
> + (sysreg & CP_REG_ARM64_SYSREG_CRM_MASK) >> CP_REG_ARM64_SYSREG_CRM_SHIFT,
> + (sysreg & CP_REG_ARM64_SYSREG_OP2_MASK) >> CP_REG_ARM64_SYSREG_OP2_SHIFT);
> +}
> +
> +#endif
Why are these here, with an ifdef, instead of in kvm.c?
Rather than a loop over an array, you could do
#define DEF(NAME, OP0, OP1, CRN, CRM, OP2) \
case SYS_##NAME: return NAME##_IDX;
int get_sysreg_idx(ARMSysRegs sysreg)
{
switch (sysreg) {
#include "cpu-sysregs.h.inc"
}
g_assert_not_reached();
}
#undef DEF
> diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> index da30bdbb2349..2381c87e4ba1 100644
> --- a/target/arm/kvm.c
> +++ b/target/arm/kvm.c
> @@ -246,6 +246,18 @@ static bool kvm_arm_pauth_supported(void)
> kvm_check_extension(kvm_state, KVM_CAP_ARM_PTRAUTH_GENERIC));
> }
>
> +/* read a sysreg value and store it in the idregs */
> +static int get_host_cpu_reg(int fd, ARMHostCPUFeatures *ahcf, ARMIDRegisterIdx index)
> +{
> + uint64_t *reg;
> + int ret;
> +
> + reg = &ahcf->isar.idregs[index];
> + ret = read_sys_reg64(fd, reg,
> + idregs_sysreg_to_kvm_reg(id_register_sysreg[index]));
> + return ret;
> +}
Surely this patch doesn't compile by itself,
because this will Werror for the unused function.
r~
next prev parent reply other threads:[~2025-03-06 1:15 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-05 16:38 [PATCH v2 00/14] arm: rework id register storage Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 01/14] arm/cpu: Add sysreg definitions in cpu-sysregs.h Cornelia Huck
2025-03-06 1:14 ` Richard Henderson
2025-03-07 15:32 ` Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 02/14] arm/kvm: add accessors for storing host features into idregs Cornelia Huck
2025-03-06 1:15 ` Richard Henderson [this message]
2025-03-07 15:35 ` Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 03/14] arm/cpu: Store aa64isar0/aa64zfr0 into the idregs arrays Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 04/14] arm/cpu: Store aa64isar1/2 into the idregs array Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 05/14] arm/cpu: Store aa64pfr0/1 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 06/14] arm/cpu: Store aa64mmfr0-3 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 07/14] arm/cpu: Store aa64dfr0/1 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 08/14] arm/cpu: Store aa64smfr0 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 09/14] arm/cpu: Store id_isar0-7 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 10/14] arm/cpu: Store id_pfr0/1/2 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 11/14] arm/cpu: Store id_dfr0/1 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 12/14] arm/cpu: Store id_mmfr0-5 " Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 13/14] arm/cpu: Add sysreg generation scripts Cornelia Huck
2025-03-05 16:38 ` [PATCH v2 14/14] arm/cpu: Add generated files Cornelia Huck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8f4b598e-d09d-4d2e-afcc-317a6997ad02@linaro.org \
--to=richard.henderson@linaro.org \
--cc=cohuck@redhat.com \
--cc=qemu-arm@nongnu.org \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).