Re: [Qemu-devel] [PATCH] coverity-model: model address_space_read/write

[Paolo Bonzini <pbonzini@redhat.com>]

On 15/03/2017 08:56, Markus Armbruster wrote:
> Paolo Bonzini <pbonzini@redhat.com> writes:
> 
>> When more complexity was added to facilitate inlining, Coverity
>> lost the ability to use the address_space_rw model for
>> address_space_read/write.
> 
> Got the commit hash of the change?
> 
>> This causes false positive defects when Coverity sees
>> a length-8 write in address_space_read and a length-4
>> (e.g. int*) buffer to read into.  As long as the size of
>> the buffer is okay, this is a false positive.
>>
>> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
>> ---
>>  scripts/coverity-model.c | 19 +++++++++++++++++++
>>  1 file changed, 19 insertions(+)
>>
>> diff --git a/scripts/coverity-model.c b/scripts/coverity-model.c
>> index ee5bf9d..2e30150 100644
>> --- a/scripts/coverity-model.c
>> +++ b/scripts/coverity-model.c
>> @@ -79,6 +79,25 @@ MemTxResult address_space_rw(AddressSpace *as, hwaddr addr, MemTxAttrs attrs,
>>      return result;
>>  }
>>  
>> +MemTxResult address_space_read(AddressSpace *as, hwaddr addr,
>> +                               MemTxAttrs attrs,
>> +                               uint8_t *buf, int len)
>> +{
>> +    MemTxResult result;
>> +    __bufwrite(buf, len);
>> +    return result;
>> +}
>> +
>> +MemTxResult address_space_write(AddressSpace *as, hwaddr addr,
>> +                                MemTxAttrs attrs,
>> +                                const uint8_t *buf, int len)
>> +{
>> +    MemTxResult result;
>> +    __bufread(buf, len);
>> +    return result;
>> +}
>> +
>> +
>>  /* Tainting */
>>  
>>  typedef struct {} name2keysym_t;
> 
> Preferably with the commit hash of the change that necessitates this
> update mentioned in the commit message:
> Reviewed-by: Markus Armbruster <armbru@redhat.com>
> 
> I can stick this in a miscellaneous pull request due later today.

Commit eb7eeb8 ("memory: split address_space_read and
address_space_write", 2015-12-17).  On the other hand, the model for
address_space_rw is not needed anymore, so I'll send v2.

Paolo