Re: memory leak in scsi_cmd_cache 2.6.15

[Pasi Kärkkäinen <pasik@iki.fi>]

On Sat, Jan 28, 2006 at 08:27:14PM +0100, Jens Axboe wrote:
> On Fri, Jan 27 2006, Nix wrote:
> > On 26 Jan 2006, Ariel noted:
> > > Is this good or bad? I'm guessing it means it's still leaking. So it's
> > > really starting to look like ata_piix is the problem. But we need someone
> > > who has the leak to remove that and see if it helps. I can't, since my
> > > drives are connected to it.
> > 
> > FWIW, a bit of negative-confirmatory evidence: I have a sym53c875
> > and non-SATA IDE drive on this 2.6.15.1, and there is no leak,
> > nor was there in 2.6.15:
> > 
> >     11     11 100%    0.34K      1       11         4K scsi_cmd_cache
> 
> It's not a bad data point, it just confirms that setting ->ordered_flush
> to 0 in the SATA drivers will fix the leak. So really, it's as expected.
> So far apparently nobody tried it, suggested it twice.
> 

Are all sata drivers affected by this bug in 2.6.15?

Any 'official' patch available?

Or is the recommended workaround to set ordered_flush to 0 to fix this..
does that have any downsides?

-- Pasi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/