That's clearly not right - I'm not sure quite what undefined behaviour
assumption convinces GCC to optimise the whole thing away>
While the pointer casting is a bit ghastly, I don't actually think that
GCC is taking advantage of undefined behaviour here, rather it looks
like
you have a simpl
On Fri, Mar 17, 2017 at 12:00:42PM +, Robin Murphy wrote:
> On 17/03/17 04:42, Subash Abhinov Kasiviswanathan wrote:
> > Android devices use multiple ip[6]tables for statistics, UID matching
> > and other functionality. Perf output indicated that ip6_do_table
> > was taking a considerable amoun
On 17/03/17 04:42, Subash Abhinov Kasiviswanathan wrote:
> Android devices use multiple ip[6]tables for statistics, UID matching
> and other functionality. Perf output indicated that ip6_do_table
> was taking a considerable amount of CPU and more that ip_do_table
> for an equivalent rate. ipv6_mask
Android devices use multiple ip[6]tables for statistics, UID matching
and other functionality. Perf output indicated that ip6_do_table
was taking a considerable amount of CPU and more that ip_do_table
for an equivalent rate. ipv6_masked_addr_cmp was chosen for
optimization as there are more instruc