Re: [PATCH v5 1/2] target/s390x: Implement the MVPG condition-code-option bit

[Richard Henderson <richard.henderson@linaro.org>]

On 3/11/21 12:16 PM, David Hildenbrand wrote:
> On 11.03.21 18:52, Richard Henderson wrote:
>> On 3/11/21 10:17 AM, David Hildenbrand wrote:
>>> +#if !defined(CONFIG_USER_ONLY)
>>> +    /*
>>> +     * For !CONFIG_USER_ONLY, we cannot rely on TLB_INVALID_MASK or 
>>> haddr==NULL
>>> +     * to detect if there was an exception during tlb_fill().
>>> +     */
>>> +    env->tlb_fill_exc = 0;
>>> +#endif
>>> +    flags = probe_access_flags(env, vaddr1, access_type, mmu_idx,
>>> +                               nofault, &haddr1, ra);
>>> +#if !defined(CONFIG_USER_ONLY)
>>> +    if (env->tlb_fill_exc) {
>>> +        return env->tlb_fill_exc;
>>> +    }
>>> +#else
>>> +    if (!haddr1) {
>>> +        env->__excp_addr = vaddr1;
>>> +        return PGM_ADDRESSING;
>>> +    }
>>> +#endif
>>
>> The assumption of PGM_ADDRESSING is incorrect here -- it could still be
>> PGM_PROTECTION, depending on how the page is mapped.
>>
> 
> Interesting, I was only looking at the s390x tlb_fill() implementation. But I 
> assume these checks are performed in common code.

Actually, no.  It's a common bug in our linux-user targets, where we don't fill 
in the SIGSEGV si_code correctly.

See e.g. 8db94ab4e5d.

> Thanks, maybe I can factor that out in a nice way. I guess we could do the 
> access via probe_access_flags() and only on error do the page_get_flags()?

Yes, we could do that.  It's certainly better for !nofault, which is the 
common-case user of this function.


r~