[bug] aarch64 host no longer boots after 767507654c22 ("arch_numa: switch over to numa_memblks")

[Mike Rapoport]

On Tue, Oct 29, 2024 at 04:43:39PM +0100, Jan Stancek wrote:
> On Tue, Oct 29, 2024 at 4:07 PM Zi Yan <ziy at nvidia.com> wrote:
> >
> > +tegra mailing list and maintainers
> >
> > On 29 Oct 2024, at 8:47, Jan Stancek wrote:
> >
> > > Hi,
> > >
> > > I'm seeing a regression on Nvidia IGX system, which no longer boots.
> > >
> > > bisect points at commit 767507654c22 ("arch_numa: switch over to numa_memblks").
> > > It hangs very early, with 4k or 64k pages, with no kernel messages printed:
> > >
> > > EFI stub: Booting Linux Kernel...
> > > EFI stub: Using DTB from configuration table
> > > EFI stub: Exiting boot services...
> > > <hangs here>
> > >
> >
> > Is it possible to have earlycon output? It is hard to debug without any
> > information except kernel fails to boot.
> 
> I know it was a long shot, so far I haven't had luck getting it to work.

Does it boot with numa=off and numa=fake?

In the log from successful boot it seems there is no NUMA information in
the device tree, can you send the device tree as well please?

> > Since the previous commit boots and I assume both kernels are compiled
> > with the same gcc toolchain, this should not be caused by the binuils
> > bug in 2.42[1]. Is your binutils version 2.42?
> 
> Yes, both are compiled locally, with binutils 2.41
> 
> >
> > Thanks.
> >
> >
> > [1] https://sourceware.org/bugzilla/show_bug.cgi?id=31924
> >
> > > Here's a log from successful boot with previous commit:
> > > https://people.redhat.com/jstancek/aarch64_numa_boot/console-log-good.txt
> > > and config: https://people.redhat.com/jstancek/aarch64_numa_boot/config
> > >
> > > # lscpu
> > > Architecture:             aarch64
> > >   CPU op-mode(s):         32-bit, 64-bit
> > >   Byte Order:             Little Endian
> > > CPU(s):                   12
> > >   On-line CPU(s) list:    0-11
> > > Vendor ID:                ARM
> > >   BIOS Vendor ID:         NVIDIA
> > >   Model name:             Cortex-A78AE
> > >     BIOS Model name:      Not Specified Not Specified CPU @ 0.0GHz
> > >     BIOS CPU family:      257
> > >     Model:                1
> > >     Thread(s) per core:   1
> > >     Core(s) per cluster:  12
> > >     Socket(s):            1
> > >     Cluster(s):           1
> > >     Stepping:             r0p1
> > >     CPU(s) scaling MHz:   100%
> > >     CPU max MHz:          1971.2000
> > >     CPU min MHz:          115.2000
> > >     BogoMIPS:             62.50
> > >     Flags:                fp asimd evtstrm aes pmull sha1 sha2 crc32
> > > atomics fphp asimdhp cpuid asimdrdm lrcpc dcpop asimddp uscat ilrcpc
> > > flagm paca pacg
> > > Caches (sum of all):
> > >   L1d:                    768 KiB (12 instances)
> > >   L1i:                    768 KiB (12 instances)
> > >   L2:                     3 MiB (12 instances)
> > >   L3:                     6 MiB (3 instances)
> > > NUMA:
> > >   NUMA node(s):           1
> > >   NUMA node0 CPU(s):      0-11
> > > Vulnerabilities:
> > >   Gather data sampling:   Not affected
> > >   Itlb multihit:          Not affected
> > >   L1tf:                   Not affected
> > >   Mds:                    Not affected
> > >   Meltdown:               Not affected
> > >   Mmio stale data:        Not affected
> > >   Reg file data sampling: Not affected
> > >   Retbleed:               Not affected
> > >   Spec rstack overflow:   Not affected
> > >   Spec store bypass:      Mitigation; Speculative Store Bypass
> > > disabled via prctl
> > >   Spectre v1:             Mitigation; __user pointer sanitization
> > >   Spectre v2:             Mitigation; CSV2, BHB
> > >   Srbds:                  Not affected
> > >   Tsx async abort:        Not affected
> > >
> > > Regards,
> > > Jan
> >
> >
> > Best Regards,
> > Yan, Zi
> >
> 

-- 
Sincerely yours,
Mike.