Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 26 Nov 2013 14:17:15 +0100
From:      Wojciech Macek <wma@semihalf.com>
To:        Olivier Houchard <cognet@ci0.org>
Cc:        arm@freebsd.org
Subject:   Re: arm SMP fix
Message-ID:  <CANsEV8csbSf%2BVS6m89ZXZJpwhR=UNs0d-FzLnu=Ffu3hWiuR2A@mail.gmail.com>
In-Reply-To: <20131126121034.GA53847@ci0.org>
References:  <20131122000457.GA44039@freebsd.org> <CAFHCsPUfXhDW-UL1iXUe5ez5QbdFC=HCWmbisuf0fFASOF=LuQ@mail.gmail.com> <CANsEV8e_6oNUA0sD3_5_zDabKxE84z15zFAUTgt70UGnnx6S8Q@mail.gmail.com> <20131126121034.GA53847@ci0.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Not realy... I'm only able to boot the 4-core Corex-A15, run world from
SATA and perform some basic operations, including multithreaded
compilation, if only takes less than ~5 minutes. Any more intensive stuff
(creating/destroying a lot of processes at the same time, to be precise)
results in kernel panic in pmap. We are trying to debug this issue, but
have not found anything interesting so far.

Regards,
Wojtek


2013/11/26 Olivier Houchard <cognet@ci0.org>

> On Tue, Nov 26, 2013 at 10:36:02AM +0100, Wojciech Macek wrote:
> > Hi,
> >
> > Applying the patch is causing a failure on systems with more than 2
> cores.
> > get_pcpu() macro is masking cpuid with incorrect value, since the only
> last
> > two bits of cp15 register contain the right value. Simple fix attached
> > below.
> >
> > Regards,
> > Wojtek
>
> Hi Wojtek,
>
> Oops that's true, there I meant 0xf and not 0x15. Does that mean that with
> your
> fix, you get it to work ?
>
> Regards,
>
> Olivier
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANsEV8csbSf%2BVS6m89ZXZJpwhR=UNs0d-FzLnu=Ffu3hWiuR2A>