[OpenAFS-devel] Very simple patch for libafs CPU hog on signal

[Derek Atkins]

Matt Peterson <matt@caldera.com> writes:

> I agree that it would be good to allow the app in user space to
> handle the syscall.  The problem is that there are several syscall
> that NEVER return.

That doesn't matter.  So long as the signal is:
        a) saved across the sleep, but
        b) allows the sleep to complete normally,
it will solve the spinning problem.

> The work to make sure that these syscalls return gracefully with the
> appropriate errorcodes is going to take some work.  The patch I
> propose is an interim solution that at least keeps the CPU from
> being hogged.

Which syscalls do you mean?  The syscalls from afsd should never
return (except at shutdown).  All other syscalls do return gracefully.

-derek

-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord@MIT.EDU                        PGP key available