[Arm-dev] Kernel problems on APM X-Gene

Fri Jan 26 20:17:49 UTC 2018
Jeremiah Rothschild <jeremiah at franz.com>

On Fri, Jan 26, 2018 at 10:07:03PM +0700, Phong Vo wrote:
> Jeremy,
> 
> Edit the grub command line and remove both or one of console= or earlycon=
> below
> 
> console=ttyS0,115200 earlycon=uart8250,mmio32,0x1c020000
> 
> to see if proceeds any further.

Wow, removing "console=" worked! Can you please explain?

> -Phong
> 
> -----Original Message-----
> From: Arm-dev [mailto:arm-dev-bounces at centos.org] On Behalf Of Jeremiah
> Rothschild
> Sent: Friday, January 26, 2018 5:04 PM
> To: Conversations around CentOS on ARM hardware
> Subject: Re: [Arm-dev] Kernel problems on APM X-Gene
> 
> On Fri, Jan 26, 2018 at 04:48:24PM +0700, Phong Vo wrote:
> > Jeremy,
> 
> Hi Phong. Thanks for the reply! Also, thank you for the tools &
> instructions to flash the Tianocore. That process worked great. :)
> 
> > Is this on which board?
> 
> Gigabyte MP30-AR0 (uart0).
> 
> > Unless you provide a log file, it is difficult to tell,
> 
> Understood. I have attached boot.log to this message. After this point, it
> freezes/hangs.
> 
> > but latest CentOS 7.4 (kernel 4.11) boots fine on X-Gene platforms.
> >
> > I am not sure how you built kernel, but this is what you can check. To
> > boot X-Gene using acpi, you need to enable
> > CONFIG_ARM64_ACPI_PARKING_PROTOCOL=y. CentOS (RPM) default
> > configuration should already have this enabled.
> 
> Interesting. It is enabled in the failing kernel:
> 
> [root at hammer ~]# grep CONFIG_ARM64_ACPI_PARKING_PROTOCOL
> /boot/config-4.11.0-45.el7.aarch64
> CONFIG_ARM64_ACPI_PARKING_PROTOCOL=y
> 
> but it is not enabled in the working kernel that Gordan built for me:
> 
> [root at hammer ~]# grep CONFIG_ARM64_ACPI_PARKING_PROTOCOL
> /boot/config-4.9.60-1.el7.centos.aarch64
> [root at hammer ]#
> 
> > -Phong
> >
> > +-----Original Message-----
> > +From: Arm-dev [mailto:arm-dev-bounces at centos.org] On Behalf Of
> > +Jeremiah Rothschild
> > +Sent: Friday, January 26, 2018 4:35 PM
> > +To: Conversations around CentOS on ARM hardware
> > +Subject: Re: [Arm-dev] Kernel problems on APM X-Gene
> > +
> > +On Sat, Nov 04, 2017 at 05:37:54PM +0000, Gordan Bobic wrote:
> > +> Apologies for taking so long to return to this thread, it took way
> > +> longer than expected to get to the machine and get it up and
> > +> running
> > +again.
> > +
> > +Thanks for the update & apologies as well on my delayed follow-up.
> > +
> > +Last time I wrote the mailing list about my kernel problems, I was
> > +told (by jperrin at centos.org) that my system was not in a supported
> > +state because I was daisy-chaining Tianocore EFI via U-Boot. I was
> > +directed to flash the Tianocore firmware and remove U-Boot from the
> equation.
> > +Although I was skeptical of this answer, I finally was able to do this.
> > +
> > +Unfortunately, however, this does not change behavior. I still cannot
> > +successfully boot into any kernels beyond 4.5 unless I use 'acpi=off'.
> > +
> > +I am, interestingly enough, able to run the 4.9.60 kernel that you
> > +supplied.
> > +
> > +So:
> > +
> > +(1) My experience leads me to believe that it is still a possibility
> > +that a kernel related bug exists.
> > +
> > +(2) I am confused as to why your kernel worked. I built kernel-alt-
> > +4.11.0-22.el7a from source and it fails like the others. Were there
> > +any special steps you took in building the rpm's you supplied me?
> > +
> > +(3) Is it true that you are able to boot into the (> 4.5) distro-
> > +supplied kernels? Or have you only tried/succeeded with your custom
> > +builds?
> > +
> > +Thanks again!
> > +
> > +j
> > +
> > +> I also just updated the build to the latest 4.9.60.
> > +>
> > +> Here is a download link to both binaries and src.rpm (download the
> > +> kernel tarball from kernel.org manually to build from src.rpm):
> > +> http://ftp.redsleeve.org/pub/misc/kernel/aarch64/RPMS/
> > +> http://ftp.redsleeve.org/pub/misc/kernel/aarch64/SRPMS/
> > +>
> > +> To recap - I am also running with Tianocore EFI chain-loaded from
> > +> u-boot, and mainling 4.9.x boots just fine on it.
> > +>
> > +> No need for disabling ACPI on the kernel command line, no need to
> > +> run EFI firmware as a 1st stage boot loader, it just works.
> > +> Do feel free to try it - if that works for you but the distro
> > +> supplied 4.5.x kernel doesn't, it seems reasonably conclusive that
> > +> it is the CentOS kernel that is broken for this board.
> > +>
> > +> I'd also be interested in learning whether you have any luck
> > +> getting PCIe cards to work with it without problems - I haven't
> > +> tried it since upgrading to 4.9.x, but certainly on 4.4.x mainline
> > +> the machine used to reliably lock up as soon as the driver for the
> PCIe card loads.
> > +>
> > +> Gordan
> > +>
> > +>
> > +> On Fri, Sep 22, 2017 at 12:06 PM, Jeremiah Rothschild
> > +> <jeremiah at franz.com>
> > +> wrote:
> > +>
> > +> > On Fri, Sep 22, 2017 at 11:59:00AM +0100, Gordan Bobic wrote:
> > +> > > On Fri, Sep 22, 2017 at 11:54 AM, Jeremiah Rothschild <
> > +> > jeremiah at franz.com>
> > +> > > wrote:
> > +> > >
> > +> > > > On Fri, Sep 22, 2017 at 11:39:19AM +0100, Gordan Bobic wrote:
> > +> > > > > If you are interested, I'm more than happy to share my
> > +> > > > > src.rpm for
> > +> > 4.9.x,
> > +> > > > > but won't be able to get to it before tomorrow morning as
> > +> > > > > the
> > +> > machine was
> > +> > > > > recently mothballed.
> > +> > > >
> > +> > > > Thanks. I actually need to test with as new of a version as I
> > +> > > > can
> > +> > because I
> > +> > > > have been experiencing an occasional "page allocation failure"
> > +> > > > kernel panic.
> > +> > > > No idea if/when that was fixed but I figure the newest
> > +> > > > version is my
> > +> > best
> > +> > > > hope.
> > +> > > >
> > +> > >
> > +> > > I've been on my own 4.9.x more or less since I got the machine,
> > +> > > it was in
> > +> > > 24/7 use, and I never experienced that issue. So it may be
> > +> > > worth a cross-check with the kernel that I'm running to see
> > +> > > whether the fault follows your machine or whether it is kernel
> dependent.
> > +> >
> > +> > You're right. It would be a good extra data point. Feel free to
> > +> > mail me directly once you're sorted and I'll gladly check out
> > +> > your 4.9
> > +build.
> > +> > Thanks
> > +> > again!
> > +> >
> > +> >
> > +> >
> > +> > > _______________________________________________
> > +> > > Arm-dev mailing list
> > +> > > Arm-dev at centos.org
> > +> > > https://lists.centos.org/mailman/listinfo/arm-dev
> > +> >
> > +> > _______________________________________________
> > +> > Arm-dev mailing list
> > +> > Arm-dev at centos.org
> > +> > https://lists.centos.org/mailman/listinfo/arm-dev
> > +> >
> > +
> > +> _______________________________________________
> > +> Arm-dev mailing list
> > +> Arm-dev at centos.org
> > +> https://lists.centos.org/mailman/listinfo/arm-dev
> > +
> > +_______________________________________________
> > +Arm-dev mailing list
> > +Arm-dev at centos.org
> > +https://lists.centos.org/mailman/listinfo/arm-dev
> > _______________________________________________
> > Arm-dev mailing list
> > Arm-dev at centos.org
> > https://lists.centos.org/mailman/listinfo/arm-dev
> _______________________________________________
> Arm-dev mailing list
> Arm-dev at centos.org
> https://lists.centos.org/mailman/listinfo/arm-dev