Re: pre-patch-2.0.31 try_to_free_page() messages
Wolfgang Wander (wwc@lars.desy.de)
Sun, 1 Jun 1997 20:02:09 +0200
Leonard N. Zubkoff writes:
> My 2.0.30 SMP deadlock patch has solved the problem 100% for my systems but
> some other people still reported problems. However, despite Ingo's posting a
> clever new deadlock detection patch, no one has reported tracking down any of
> the remaining deadlocks.
>
> If we ever expect to find any deadlocks that remain, people who actually run
> into them are going to have to assist in tracking them down, or provide
> reproducible test cases so others can.
I pretty much agree here - I'm now running 2.0.31-pre (including the
deadlock-patch) on our 10 machines cluster and see a much more stable
environment (uptimes of 7 days and more under heavy load). However I
still have some deadlocks and am willing to contribute using Ingos
Patch.
This patch however freezes the machines at boot time and I did so far
get no reply from Ingo on how to fix this...
Wolfgang