Re: OOPS (kswapd) in 2.4.5 and 2.4.6

[Andrew Morton]

Henry wrote:
> 
> >
> > I wonder why it only affects you.  Is the drive which holds
> > your swap partition running in PIO mode?  `hdparm' will tell
> > you.  If it is, then that could easily cause the page to come
> > unlocked before brw_page() has finished touching the buffer
> > ring.  Then all it takes is a parallel try_to_free_buffers
> > on the other CPU.
> 
> Here's output from htparm:
> 
> /dev/hda:
>  multcount    =  0 (off)
>  I/O support  =  0 (default 16-bit)
>  unmaskirq    =  0 (off)
>  using_dma    =  0 (off)
>  keepsettings =  0 (off)
>  nowerr       =  0 (off)
>  readonly     =  0 (off)
>  readahead    =  8 (on)
>  geometry     = 2494/255/63, sectors = 40079088, start = 0
> 
> Does this provide the info you need?

Bingo.  PIO mode -> synchronous writes in submit_bh().  Thanks.

> I believe another chap responded to my post with a similar issue (also
> SMP machine).

No, his oops was a bad inode state while trying to
release unused NFS client inodes.  Different bug :)

-