On Wed, Jun 19, 2013 at 14:19, Marc Espie wrote:
> On Wed, Jun 19, 2013 at 01:40:19PM +0200, Martin Pelikan wrote:
>> > If you're right that atomic_{clear,set}bits_int is correct and
>> > sufficient and actually faster, then all dynamic executables would
>> > benefit from this speedup (sigprocmask is used in ld.so(1)).
>>
>> Since on i386 GENERIC these atomic_* things don't emit the LOCK prefix,
>> performance shouldn't be an issue; I'm actually more worried about this
> bit:
>>
>> > > - p->p_sigmask = mask &~ sigcantmask;
>> > > + p->p_sigmask = mask;
>>
>> On the right architecture where a word store isn't atomic enough and
>> with the right compiler that decides to put p_sigmask on an address
>> ending with 0xFFF with 4k-sized pages, we have two problems already.
>
> If I read the code correctly, just adding an alignment constraint
> to pool_init for the proc_pool would fix that 2nd part.
pool will never give you an object that spans pages (unless the object
itself is larger than a page). p_sigmask will always be aligned
nicely. i mean, if it weren't, that code already would crash.
