[haiku-commits] Re: r36559 - haiku/trunk/src/system/kernel/arch/x86
- From: Ingo Weinhold <ingo_weinhold@xxxxxx>
- To: haiku-commits@xxxxxxxxxxxxx
- Date: Fri, 30 Apr 2010 23:06:50 +0200
On 2010-04-30 at 22:45:17 [+0200], François Revol <revol@xxxxxxx> wrote:
> Le Fri, 30 Apr 2010 21:37:19 +0200 (CEST), ingo_weinhold@xxxxxx a écrit
> :
> > Author: bonefish
> > Date: 2010-04-30 21:37:19 +0200 (Fri, 30 Apr 2010)
> > New Revision: 36559
> > Changeset: http://dev.haiku-os.org/changeset/36559/haiku
> >
> > Modified:
> > haiku/trunk/src/system/kernel/arch/x86/arch_cpu.cpp
> > Log:
> > Added a third rendez-vous point for the call_all_cpus() MTRR
> > functions. This
> > fixes the problem that the CPU initiating the call could make the
> > next call
> > and reset sCpuRendezvous2 before the other CPUs have returned from
> > their
> > smp_cpu_rendezvous(). Probably virtually impossible on real hardware,
> > but I
> > could almost reliably reproduce it with qemu -smp 2 (would hang the
> > late boot
> > process without ability to enter KDL).
>
> Couldn't it be interesting to have a synchronized version of
> call_all_cpus() that'd do this on its own ? It might be useful in more
> than 1 place maybe ?
There already is a call_all_cpus_sync() which ensures that the call has
finished on all CPUs when it returns. It is implemented by first messaging
the other CPUs synchronously (i.e. waiting for them to be done) and
afterwards calling the function itself. This makes is unusable for this
case, since setting the MTRRs itself is supposed to happen synchronously on
all CPUs.
CU, Ingo
Other related posts:
