On Wed, Oct 18, 2006 at 03:31:53PM -0700, Jack Vogel wrote: > On 10/18/06, Kip Macy <[EMAIL PROTECTED]> wrote: > >I have a Sun T2000 that I generally run with the em driver from as of > >July in order to avoid watchdog timeouts. One trivial scenario that > >reproduces the problem with 100% consistency is running the ghc > >configure script (a 20kloc shell script) over NFS. As the T2000 > >doesn't exactly represent "typical" PC hardware it may not be the most > >desirable test platform. Nonetheless, let me know if you're > >interested. > >Thanks for looking into this issue. > > > > -Kip > > I'm a bit confused from the way you worded this, do you have watchdogs > with em, or you use em to avoid them? > > If you have problems when running NFS then thats a clue, is it TCP or > UDP based NFS? I am interested, give me details about the setup please. > > I have one of the engineers in our test organization trying to repro > symptoms > on a system installed with BETA2, it has shared interrupts between em and > usb. Any additional stuff he could run would be helpful.
I have been working with someone's system that has em shared with fxp, and a simple fetch over the em (e.g. of a 10 GB file of zeroes) is enough to produce watchdog timeouts after a few seconds. As previously mentioned, changing the INTR_FAST to INTR_MPSAFE in the driver avoids this problem. However, others are seeing sporadic watchdog timeouts at higher system load on non-shared em systems too. This is in addition to the hardware instances you already know about where the em driver has not worked going back to 5.x. Kris
pgpD0sIWlaYUK.pgp
Description: PGP signature
