Re: [PATCH 6/6] hw/ppc: Enable Fadump for PSeries

["Nicholas Piggin" <npiggin@gmail.com>]

On Thu Feb 27, 2025 at 5:07 PM AEST, Aditya Gupta wrote:
> On 27/02/25 09:03, Nicholas Piggin wrote:
>
>> On Mon Feb 17, 2025 at 5:17 PM AEST, Aditya Gupta wrote:
>>> With all support in place, enable fadump by exporting the
>>> "ibm,configure-kernel-dump" RTAS call in the device tree.
>>>
>>> Presence of "ibm,configure-kernel-dump" tells the kernel that the
>>> platform (QEMU) supports fadump.
>>>
>>> Pass "fadump=on" to enable Linux to use firmware assisted dump.
>>>
>>> Logs of a linux boot with firmware assisted dump:
>>>
>>>      ./build/qemu-system-ppc64 -M pseries,x-vof=on --cpu power10 --smp 4 -m 4G -kernel some-vmlinux -initrd some-initrd -append "debug fadump=on crashkernel=1G" -nographic
>>>      [    0.000000] random: crng init done
>>>      [    0.000000] fadump: Reserved 1024MB of memory at 0x00000040000000 (System RAM: 4096MB)
>>>      ...
>>>      [    1.084686] rtas fadump: Registration is successful!
>>>      ...
>>>      # cat /sys/kernel/debug/powerpc/fadump_region
>>>      CPU :[0x00000040000000-0x000000400013d3] 0x13d4 bytes, Dumped: 0x0
>>>      HPTE:[0x000000400013d4-0x000000400013d3] 0x0 bytes, Dumped: 0x0
>>>      DUMP: Src: 0x00000000000000, Dest: 0x00000040010000, Size: 0x40000000, Dumped: 0x0 bytes
>>>
>>>      [0x000000fffff800-0x000000ffffffff]: cmdline append: ''
>>>      # echo c > /proc/sysrq-trigger
>>>
>>> The fadump boot after crash:
>>>
>>>      [    0.000000] rtas fadump: Firmware-assisted dump is active.
>>>      [    0.000000] fadump: Updated cmdline: debug fadump=on crashkernel=1G
>>>      [    0.000000] fadump: Firmware-assisted dump is active.
>>>      [    0.000000] fadump: Reserving 3072MB of memory at 0x00000040000000 for preserving crash data
>>>      ....
>>>      # file /proc/vmcore
>>>      /proc/vmcore: ELF 64-bit LSB core file, 64-bit PowerPC or cisco 7500, OpenPOWER ELF V2 ABI, version 1 (SYSV), SVR4-style
>>>
>>> Analysing the vmcore with crash-utility:
>>>
>>>            KERNEL: vmlinux-6.14-rc2
>>>          DUMPFILE: vmcore-a64dcfb451e2-nocma
>>>              CPUS: 4
>>>              DATE: Thu Jan  1 05:30:00 IST 1970
>>>            UPTIME: 00:00:30
>>>      LOAD AVERAGE: 0.74, 0.21, 0.07
>>>             TASKS: 94
>>>          NODENAME: buildroot
>>>           RELEASE: 6.14.0-rc2+
>>>           VERSION: #1 SMP Wed Feb 12 06:49:59 CST 2025
>>>           MACHINE: ppc64le  (1000 Mhz)
>>>            MEMORY: 4 GB
>>>             PANIC: "Kernel panic - not syncing: sysrq triggered crash"
>>>               PID: 270
>>>           COMMAND: "sh"
>>>              TASK: c000000009e7cc00  [THREAD_INFO: c000000009e7cc00]
>>>               CPU: 3
>>>             STATE: TASK_RUNNING (PANIC)
>>>
>>> Signed-off-by: Aditya Gupta <adityag@linux.ibm.com>
>> This is very cool, nice work. Does it work with KVM? I think... probably
>> it could?
>
> Yes it does, atleast for crashing CPU :)
>
> But there are problems with reading the CPU regs, regs don't seem 
> correct for non-crashing CPUs.
>
> Crash is able to work perfectly for the crashing CPU as of now (as the 
> registers are stored by the kernel in that case).

You may need to call cpu_synchronize_state() in the CPU_FOREACH loop
before you read out the register state. Does that fix it?

>>
>> Are you able to add a functional test case for it? This is something
>> that people (including me) will forget to test...
>
> Sure, I will add a test case.
>
>
> Thanks for your reviews Nick.
>
> It might take few weeks for me to post another version, will see into 
> the tests in qemu and arrange the code bit more nicely.

Yeah that's okay, I'm way behind with reviews and merging unfortunately
so may have to wait for next release, but I'm keen to get it merged when
we can. Sorry for the late review.

Thanks,
Nick