Re: [Toshiba IDE DMA BUG] Pinpointed version

From: Marc Duponcheel (mduponch@cisco.com)
Date: Mon May 15 2000 - 18:21:39 EST


> > > BTW, I use "mem=63M" and it's fine.
> >
> > If it is a detection issue, what about putting the correct amount of memory
> > "mem=64M" (or "mem=128M" in my case) ... I think I am going to try that too
>
> No, "mem=64M" is dangerous and leads to occasional filesystem
> corruption. In fact, mem=64M is more or less what the kernel memory
> detection chooses.
>
> The problem is that it's wrong. Some of that 64M isn't available for
> kernel use. I forget the details -- either it was ACPI "reclaimable"
> (but not really) memory or it was behaving as some kind of ROM.

on 2.3.99-pre8 "mem=128M" reports 126884 with 'free' (whereas it reports 128880
with 2.2.15 without any "mem=") and seems to be stable during huge rsync's during
which it used to fail guarantied. Of course there is about 2 meg difference which
is maybe enough to avoid trouble ...

-- Marc Duponcheel.
        [work] mduponch@cisco.com
        [home] marc@offline.be

-
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.tux.org/lkml/



This archive was generated by hypermail 2b29 : Mon May 15 2000 - 21:00:27 EST