Hi Dave:
For which kernel versions is the fix: "Race conditions in timer handling"
relevent? Would all versions of 2.4 kernels benefit from this improvement,
only those after a specific release, or only 2.6 kernels? I can not correlate
the abstract fix description to the explanation regarding latest RH beta. The
RH beta explanation talks about compilation issues and nothing else related
to a race condition referenced in the abstract.
Thanks for the clarification - Matt Gierlach
On Mon, 18 Aug 2003, David Grothe wrote:
> Now on the FTP site. This version fixes a few recently reported problems.
> * Deadly embrace between backenable and getpmsg
> * Race conditions in timer handling
> * Truncation to 16 bits in message length calculation
> The multi-threaded open problem remains unfixed.
>
> The timer fixes eliminate the reference to the linked list fields of the
> kernel timer structure and thus give some hope of successful compilation
> against the latest Red Hat beta.
>
> The inet driver has an unresolved external tcp_openreq_cachep in some
> versions of Linux. Report this problem to Brian at [EMAIL PROTECTED],
> or via the LiS mailing list.
>
> -- Dave
_______________________________________________
Linux-streams mailing list
[EMAIL PROTECTED]
http://gsyc.escet.urjc.es/mailman/listinfo/linux-streams
