Are you sure that your function is being parsed successfully? I get a
reasonable answer for:
int main ()
{
for (i = 0; i != 10; ++i) {
switch (q) {
case FOO:
for (i = 0; i != 20; ++i) {
return 12;
}
case XXX:
return 12;
}
}
}
and
@@
expression a;
@@
* for ( ... ; ... ; a ) {
...
* a
...
}
julia
On Wed, 9 Mar 2011, Nicolas Kaiser wrote:
> Hi there!
>
> I recently noticed a couple of patches by Roel Kluin on lkml
> that looked interesting to me, and I tried to write a semantic
> patch to find such occurrences. One of the lkml patches:
>
> can: wrong index used in inner loop
> http://groups.google.com/group/linux.kernel/browse_thread/thread/6d1e353cedcf5862
>
> Here are nested 'for' loops using the same iterator, and the
> inner one will overwrite the outer one. In this case, the inner
> loop is within an 'if' clause. Simplified:
>
> for ( i = 0; ... ; ++i ) {
> ...
> if ( ... ) {
> ...
> for ( i = 0; ... ; ++i ) {
> ...
> }
> ...
> }
> }
>
> A very simple semantic patch I came up with to point to this
> pattern would look like:
>
> @@
> expression a;
> @@
>
> * for ( ... ; ... ; a ) {
> ...
> * a
> ...
> }
>
> However, this semantic patch fails when 'if' is replaced with
> 'switch', like here:
>
> nfsd: wrong index used in inner loop
> http://groups.google.com/group/linux.kernel/browse_thread/thread/0366edcfe5497481
>
> Simplified:
>
> for ( i = 0; ... ; ++i ) {
> ...
> switch ( ... ) {
> ...
> case ...:
> ...
> for ( i = 0; ... ; ++i ) {
> ...
> }
> ...
> }
> }
>
> I'm afraid I'm stuck trying to find the expression within the 'switch'.
> Could you possibly give me a hint what I'm doing wrong?
>
> Best regards,
> Nicolas Kaiser
> _______________________________________________
> Cocci mailing list
> [email protected]
> http://lists.diku.dk/mailman/listinfo/cocci
> (Web access from inside DIKUs LAN only)
>
_______________________________________________
Cocci mailing list
[email protected]
http://lists.diku.dk/mailman/listinfo/cocci
(Web access from inside DIKUs LAN only)
