[parisc-linux] 9000/819/K210

Thibaut VARENE varenet@esiee.fr
Mon, 8 Jul 2002 20:14:06 +0200


Le lundi 8 juillet 2002, =E0 06:32 , Grant Grundler a =E9crit :

> Stephan Trajkoff wrote:
>> Hello,
>> TEAM from esiee.fr told me that it is common SMP problem...
>> after heavy compilation, 'w' and 'ps ax' don't work
>
> My a500 has been running -pa52 and iteratively built 88 linux kernels
pa52 ? Well, if you're one step further than everyone, that's kinda=20
cheating :^)
> since yesterday. Both w and ps ax still work. Something else must be
> triggering the problem.
Do you mean that the hangs Ryan and I are studying should be solved by=20=

your changes ?
Anyway, I have noticed that on J5k and A500, using the 'normal IO' mode=20=

for the SYM53C8XX driver
seems to decrease risk of such hangs (running 2 setis on both machines=20=

and building ISOs on the
A500 for about 4 days, pa46 on both, without hangs, where the A500 could=20=

only run for about
3 hours is the same conditions with the MMIO mode, 1 day in the best=20
case.)

But I had not enough time to investigate further, so it should only be=20=

considered as a
*interesting coincidence* :)

I have now installed pa51 on these boxes, so I'll keep checking for=20
hangs.


Thibaut VARENE
PA/Linux ESIEE Team
http://pateam.esiee.fr/