tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Fstrans rewrite without rwlock



"J. Hannken-Illjes" <hannken%eis.cs.tu-bs.de@localhost> writes:

> On Jan 13, 2013, at 7:24 PM, Greg Troxel <gdt%ir.bbn.com@localhost> wrote:
>
>> 
>> J. Hannken-Illjes <hannken%eis.cs.tu-bs.de@localhost> writes:
>> 
>>> Attached is a complete rewrite based on passive serialization and
>>> condvars.  The fast paths (fstrans_start/fstrans_done on a file system
>>> not suspended or suspending and fscow_run) now run without locks or
>>> other atomic operations.  Suspension and cow handler insertion and
>>> removal is done with mutex / condvars.
>> 
>> How many ports have you run the regression test on with the change?
>> (Presumably there are no new failures.)
>
>
> Did functional and stress tests on i386 only  --  my atf testbed is
> currently not working.

I think any change of this magnitude should have atf runs on multiple
architectures, including a few that are outliers relative to passive
seralization implementations.

Attachment: pgpg5opzSo0Ff.pgp
Description: PGP signature



Home | Main Index | Thread Index | Old Index