Re: [PATCH v2 4/9] x86, memcpy_mcsafe: add write-protection-fault handling

[Dan Williams]

On Wed, May 2, 2018 at 10:29 PM, Mika PenttilÃ
<mika.penttila@xxxxxxxxxxxx> wrote:
> On 05/03/2018 07:59 AM, Dan Williams wrote:
>> In preparation for using memcpy_mcsafe() to handle user copies it needs
>> to be to handle write-protection faults while writing user pages. Add
>> MMU-fault handlers alongside the machine-check exception handlers.
>>
>> Note that the machine check fault exception handling makes assumptions
>> about source buffer alignment and poison alignment. In the write fault
>> case, given the destination buffer is arbitrarily aligned, it needs a
>> separate / additional fault handling approach. The mcsafe_handle_tail()
>> helper is reused. The @limit argument is set to @len since there is no
>> safety concern about retriggering an MMU fault, and this simplifies the
>> assembly.
>>
>
>> diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c
>> index 75d3776123cc..9787f5ee0cf9 100644
>> --- a/arch/x86/lib/usercopy_64.c
>> +++ b/arch/x86/lib/usercopy_64.c
>> @@ -75,6 +75,23 @@ copy_user_handle_tail(char *to, char *from, unsigned len)
>>       return len;
>>  }
>>
>> +/*
>> + * Similar to copy_user_handle_tail, probe for the write fault point,
>> + * but reuse __memcpy_mcsafe in case a new read error is encountered.
>> + * clac() is handled in _copy_to_iter_mcsafe().
>> + */
>> +__visible unsigned long
>> +mcsafe_handle_tail(char *to, char *from, unsigned len)
>> +{
>> +     for (; len; --len, to++) {
>> +             unsigned long rem = memcpy_mcsafe(to, from, 1);
>> +
>
>
> Hmm why not
>         for (; len; --len, from++, to++)
>
>

Why not indeed. Great catch, I'll fix this and extend the unit test to
verify the data transfer as well.