[4498] in linux-net channel archive
Re: seeing crashes? read this... YUP... on intel also
daemon@ATHENA.MIT.EDU (Ingo Molnar)
Sat Sep 21 14:03:40 1996
Date: Sat, 21 Sep 1996 19:25:09 +0200 (MET DST)
From: Ingo Molnar <mingo@pc5829.hil.siemens.at>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
cc: submit-linux-dev-net@ratatosk.yggdrasil.com
In-Reply-To: <5215u5$2v4@lightning.swansea.linux.org.uk>
On 21 Sep 1996, Alan Cox wrote:
> AMD5x86-90, 64MB memory,
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
> This odd divide by 0 error is always only seen on the AMD5x86 chips. You
> need to disable the use of the pentium fast gettimeofday for these
> clone processors it seems.
ppl can do this by turning APM support on. The K5 does some internal clock
regulation which completely disturbs the fast gettimeofday logic? Time to
make that code APM aware ...
maybe we should detect the K5 or add a CONFIG_K5 ? Running the K5 with the
fast pentium stuff is lethal for sure it seems so. Or isnt there an APM
flag in the CPUID mask [dont have the docs here] ... thus we can printk a
warning that gettimeofday isnt safe in this case.
-- mingo
