Re: memory leak in scsi_cmd_cache 2.6.15

[Arjan van de Ven <arjan@infradead.org>]

On Sun, 2006-01-22 at 13:51 -0500, Ariel wrote:
> On Sun, 22 Jan 2006, Arjan van de Ven wrote:
> 
> > On Sun, 2006-01-22 at 09:16 +0100, Arjan van de Ven wrote:
> >> On Sat, 2006-01-21 at 21:13 -0500, Ariel wrote:
> >>> I have a memory leak in scsi_cmd_cache.
> 
> >> does this happen without the binary nvidia driver too? (it appears
> >> you're using that). That's a good datapoint to have if so...
> 
> I had the exact same nvidia driver with 2.6.12 (just recompiled) and it 
> didn't happen there.
> 
> But just in case I used slabtop to watch scsi_cmd_cache grow by 1.24KB per 
> second (104MB per day), then I rmmoded nvidia and watched it grow by 
> 1.16KB per 

please repeat this without nvidia ever being loaded. Just having a
module loaded before can already cause corruption that ripples through
later, so just unloading is not enough to get a clean result.

<lsmod snipped>

I see you also use vmware. The other person who reported this also uses
vmware. Could you please repeat the test without BOTH the nvidia and
vmware modules?



-
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/