Please correct this code

[Sri Ram Vemulpali]

thanks for all explanation, it really helped to refine my code

Sri

On Mon, Jan 31, 2011 at 12:58 PM, Rajat Sharma <fs.rajat at gmail.com> wrote:
>> Is is not that Code A is wrong, for not mentioning cobbler list
>> 'memory' to indicate memory contents are getting updated.
>> Please rectify me, if I am wrong.
>
> From Code B, it looks like instruction to compiler for not using extra
> memory location other than ones specified for input param and output
> param. Is it some sort of memory synchronization code?
>
>> Also, what does '+m' constraint mean, if I use in inline assembly.
>
> It means the memory location for param is both Readable and Writable
>
> Rajat
>
> On Mon, Jan 31, 2011 at 10:07 PM, Sri Ram Vemulpali
> <sri.ram.gmu06 at gmail.com> wrote:
>> Hi,
>>
>> I have following snippet of code.
>>
>> Code A
>>
>> __asm__ __volatile__(
>>                     "   lock       ;\n"
>>                     "   addl %1,%0 ;\n"
>>                     : "=m"  (my_var)
>>                     : "ir"  (my_int), "m" (my_var)
>>                     :                                 /* no clobber-list */
>>                     );
>>
>> Code B
>>
>>  __asm__ __volatile__(  "decl %0; sete %1"
>>                     : "=m" (my_var), "=q" (cond)
>>                     : "m" (my_var)
>>                     : "memory"
>>                     );
>>
>> In Code A memory contents are getting updated. And in Code B memory
>> contents and 'cond' var are getting updated.
>> In Code B in cobbler column 'memory' is used to indicate that this
>> code modifies memory contents. But where as in Code A it was not
>> mentioned
>>
>> Is is not that Code A is wrong, for not mentioning cobbler list
>> 'memory' to indicate memory contents are getting updated.
>> Please rectify me, if I am wrong.
>>
>> Also, what does '+m' constraint mean, if I use in inline assembly.
>> --
>> Regards,
>> Sri.
>>
>> _______________________________________________
>> Kernelnewbies mailing list
>> Kernelnewbies at kernelnewbies.org
>> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
>>
>



-- 
Regards,
Sri.