svn commit: r275121 - in head/sys: compat/linux compat/svr4 fs/procfs kern sys

Konstantin Belousov kostikbel at gmail.com
Thu Jan 7 15:37:10 UTC 2016


On Thu, Jan 07, 2016 at 04:39:53PM +0300, Chagin Dmitry wrote:
> On Wed, Nov 26, 2014 at 02:10:00PM +0000, Konstantin Belousov wrote:
> > Author: kib
> > Date: Wed Nov 26 14:10:00 2014
> > New Revision: 275121
> > URL: https://svnweb.freebsd.org/changeset/base/275121
> > 
> > Log:
> >   The process spin lock currently has the following distinct uses:
> >   
> >   - Threads lifetime cycle, in particular, counting of the threads in
> >     the process, and interlocking with process mutex and thread lock.
> >     The main reason of this is that turnstile locks are after thread
> >     locks, so you e.g. cannot unlock blockable mutex (think process
> >     mutex) while owning thread lock.
> >   
> >   - Virtual and profiling itimers, since the timers activation is done
> >     from the clock interrupt context.  Replace the p_slock by p_itimmtx
> >     and PROC_ITIMLOCK().
> >   
> >   - Profiling code (profil(2)), for similar reason.  Replace the p_slock
> >     by p_profmtx and PROC_PROFLOCK().
> >   
> >   - Resource usage accounting.  Need for the spinlock there is subtle,
> >     my understanding is that spinlock blocks context switching for the
> >     current thread, which prevents td_runtime and similar fields from
> >     changing (updates are done at the mi_switch()).  Replace the p_slock
> >     by p_statmtx and PROC_STATLOCK().
> >   
> >   The split is done mostly for code clarity, and should not affect
> >   scalability.
> >   
> >   Tested by:	pho
> >   Sponsored by:	The FreeBSD Foundation
> >   MFC after:	1 week
> hi, any chance to merge it?

I do not want to merge the split, for many reasons.
I suppose the merge conflicts due to the replace of PROC_SLOCK() by
PROC_*LOCK() are what you are looking after, am I right ?

What could be done, is to merge the syntax changes from the patch.  I mean,
provide the PROC_*LOCK() macros and merge all changes to use new macros,
but internal implementation would still lock the same proc spinlock.



More information about the svn-src-all mailing list