Re: 2.1.120pre3 still has sound mem leak

Michael L. Galbraith (mikeg@weiden.de)
Thu, 3 Sep 1998 16:56:18 +0200 (CEST)


On Wed, 2 Sep 1998, Garst R. Reese wrote:

> Michael L. Galbraith wrote:
> >
> > On Tue, 1 Sep 1998, Garst R. Reese wrote:
> >
> > > Each use of sound followed by as many rmmod -a's as you choose,
> > > decreases total mem by 64K. This is with an old LT SB compatible ESS488
> > > 8 bit mono sound.
> > > On the first attempt, rmmod -a resulted in the msg.
> > > "sound module went missing!"
> > > This mem leak happened way back when, but I thought it had gone away. I
> > > don't let Biff bark very often :)
> > > --
> > > Garst
> >
> > Hi,
> >
> > Are you using the SB driver? According to memleak, it has a leak at
> > sb_mixer.c:456, but that's only 92 bytes. I've been using it to test
> > memleak with, but the results seem strange. Can anyone explain the
> > buffer/slab growth in the following? Is this normal?
> >
> > -Mike
> >
> > Script started on Wed Sep 2 15:45:57 1998
> Yes, I am using the SB driver.
> Oleg Drokin sent me the attached diff that fixes my problem. He also
> sent it to alan some time ago,but apparently it never got applied.
> I'll check the memleak also. Thanks
> --
> Garst

The patch led me to the answer to my question about buffer growth.
Thanks, and thanks again to Oleg who also sent it to me. My problem
was (is) remap_page_range(), quite a can of worms. My next boot is
going to be... done with eyes glued to CRT and finger on BRB :)

Cheers,

-Mike

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html