Date: Sun, 22 Dec 2013 11:53:47 +0100 From: Wojciech Macek <wma@semihalf.com> To: Ruslan Bukin <br@freebsd.org> Cc: freebsd-arm@freebsd.org Subject: Re: arm SMP on Cortex-A15 Message-ID: <CANsEV8fSoygoSUyQqKoEQ7tRxjqDOwrPD8dU7O2V2PXRj35j4A@mail.gmail.com> In-Reply-To: <20131222092913.GA89153@mail.bsdpad.com> References: <CANsEV8euHTsfviiCMP_aet3qYiK2T-oK%2B-37eay7zAPH2S2vaA@mail.gmail.com> <20131220125638.GA5132@mail.bsdpad.com> <20131222092913.GA89153@mail.bsdpad.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Thanks, so it seems that there is still something wrong... >From what I observed, the place where you got the panic was the most likely to fail if there are issues with TLB cache. I guess your case can also have the same root case. Nevertheless, I'll try to reproduce your setup and debug it futher. Regards, Wojtek 2013/12/22 Ruslan Bukin <br@freebsd.org> > On Fri, Dec 20, 2013 at 04:56:39PM +0400, Ruslan Bukin wrote: > > On Thu, Dec 19, 2013 at 12:41:59PM +0100, Wojciech Macek wrote: > > > Hi, > > > > > > Finally, I'm able to run FreeBSD stable on Cortex-A15. The TLB issue > which > > > was observed, was caused by an aggressive A15 feature called "L2 TLB > > > prefetch". > > > > > > > Great!? I successfully run buildworld with these patches on SMP-enabled > > dual-core Cortex-A15 machine (Exynos5250) with no problems at all. > > > > My timings: > > kernel-toolchain - ~3h > > buildkernel - ~1h > > buildworld - ~16h > > > > Note I used NFS for mount source and obj using USB-based ethernet. > > The next test I will try to produce is to adding -j2 option. > > > > With -j2 it always goes panic, not immediately, > but after a few minutes of compilation, like this. > > -Ruslan >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANsEV8fSoygoSUyQqKoEQ7tRxjqDOwrPD8dU7O2V2PXRj35j4A>