Christopher Li wrote:
> On Tue, Feb 13, 2007 at 12:15:56AM +0000, Anton Altaparmakov wrote:
>> I applied this patch to the current spares git code and reran my test and 
>> now we have two warnings:
>>
>>   CHECK   fs/ntfs/file.c
>> fs/ntfs/file.c:2241:5: warning: incorrect type in argument 8 (different 
>> signedness)
>> fs/ntfs/file.c:2241:5:    expected int [signed] ( [signed] [usertype] 
>> get_block )( ... )
>> fs/ntfs/file.c:2241:5:    got int [signed] ( static [toplevel] *<noident> 
>> )( ... )
>> include/linux/fs.h:1791:14: warning: incorrect type in argument 8 
>> (different signedness)
>> include/linux/fs.h:1791:14:    expected int [signed] ( [signed] [usertype] 
>> get_block )( ... )
>> include/linux/fs.h:1791:14:    got int [signed] ( *get_block )( ... )
>>
>> Best regards,
>>
> 
> I see. In evaluate_call().
> evaluate_arguments() is called before target function arguments
> are converted into pointers.
> 
> Can you please try this patch instead?

Anton reported success with this revision of your patch, so I'd like to apply
it to the sparse tree; could you please supply a Signed-off-by so I can do so?

The same goes for several of the patches in your -cl2 tree, which I would
eagerly like to apply.  Your patches look quite excellent, and greatly advance
the capabilities of Sparse in directions I've wanted to see it progress in.
I'd prefer to pull your patches in as a block, to avoid inconveniencing you by
pulling them in piecemeal and forcing you to rebase the remaining patches.

- Josh Triplett