Re: [PATCH 5/5] linux-user/s390x: Save and restore psw.mask properly

[David Hildenbrand]

On 15.06.21 05:07, Richard Henderson wrote:
> At present, we're referencing env->psw.mask directly, which
> fails to ensure that env->cc_op is incorporated or updated.
> Use s390_cpu_{set_psw,get_psw_mask} to fix this.
>
> Mirror the kernel's cleaning of the psw.mask in save_sigregs
> and restore_sigregs.  Ignore PSW_MASK_RI for now, as qemu does
> not support that.

s/qemu/qemu\/tcg/ ? because qemu/kvm supports it .

Maybe reference the bug reports this fixes?

> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> ---
>   linux-user/s390x/signal.c | 37 ++++++++++++++++++++++++++++++++-----
>   1 file changed, 32 insertions(+), 5 deletions(-)
>
> diff --git a/linux-user/s390x/signal.c b/linux-user/s390x/signal.c
> index ef136dae33..bf8a8fbfe9 100644
> --- a/linux-user/s390x/signal.c
> +++ b/linux-user/s390x/signal.c
> @@ -112,15 +112,23 @@ get_sigframe(struct target_sigaction *ka, CPUS390XState 
> *env, size_t frame_size)
>       return (sp - frame_size) & -8ul;
>   }
>   
> +#define PSW_USER_BITS   (PSW_MASK_DAT | PSW_MASK_IO | PSW_MASK_EXT | \
> +                         PSW_MASK_MCHECK | PSW_MASK_PSTATE | PSW_ASC_PRIMARY)
> +#define PSW_MASK_USER   (PSW_MASK_ASC | PSW_MASK_CC | PSW_MASK_PM | \
> +                         PSW_MASK_64 | PSW_MASK_32)
> +
>   static void save_sigregs(CPUS390XState *env, target_sigregs *sregs)
>   {
> +    uint64_t psw_mask = s390_cpu_get_psw_mask(env);
>       int i;
>   
>       /*
>        * Copy a 'clean' PSW mask to the user to avoid leaking
>        * information about whether PER is currently on.
> +     * TODO: qemu does not support PSW_MASK_RI; it will never be set.
>        */
> -    __put_user(env->psw.mask, &sregs->regs.psw.mask);
> +    psw_mask = PSW_USER_BITS | (psw_mask & PSW_MASK_USER);
> +    __put_user(psw_mask, &sregs->regs.psw.mask);
>       __put_user(env->psw.addr, &sregs->regs.psw.addr);
>   
>       for (i = 0; i < 16; i++) {
> @@ -289,7 +297,7 @@ void setup_rt_frame(int sig, struct target_sigaction *ka,
>   
>   static void restore_sigregs(CPUS390XState *env, target_sigregs *sc)
>   {
> -    target_ulong prev_addr;
> +    uint64_t prev_addr, prev_mask, mask, addr;
>       int i;
>   
>       for (i = 0; i < 16; i++) {
> @@ -297,9 +305,28 @@ static void restore_sigregs(CPUS390XState *env, 
> target_sigregs *sc)
>       }
>   
>       prev_addr = env->psw.addr;
> -    __get_user(env->psw.mask, &sc->regs.psw.mask);
> -    __get_user(env->psw.addr, &sc->regs.psw.addr);
> -    trace_user_s390x_restore_sigregs(env, env->psw.addr, prev_addr);
> +    __get_user(mask, &sc->regs.psw.mask);
> +    __get_user(addr, &sc->regs.psw.addr);
> +    trace_user_s390x_restore_sigregs(env, addr, prev_addr);
> +
> +    /*
> +     * Use current psw.mask to preserve PER bit.
> +     * TODO:
> +     *  if (!is_ri_task(current) && (user_sregs.regs.psw.mask & PSW_MASK_RI))
> +     *          return -EINVAL;
> +     * Simply do not allow it to be set in mask.
> +     */
> +    prev_mask = s390_cpu_get_psw_mask(env);
> +    mask = (prev_mask & ~PSW_MASK_USER) | (mask & PSW_MASK_USER);
> +    /* Check for invalid user address space control. */
> +    if ((mask & PSW_MASK_ASC) == PSW_ASC_HOME) {
> +        mask = (mask & ~PSW_MASK_ASC) | PSW_ASC_PRIMARY;
> +    }
> +    /* Check for invalid amode. */
> +    if (mask & PSW_MASK_64) {
> +        mask |= PSW_MASK_32;
> +    }
> +    s390_cpu_set_psw(env, mask, addr);
>   
>       for (i = 0; i < 16; i++) {
>           __get_user(env->aregs[i], &sc->regs.acrs[i]);

LGTM

Reviewed-by: David Hildenbrand <david@redhat.com>

--
Thanks,

David / dhildenb