Comment 1 for bug 2021930

Revision history for this message
William Tu (wtu) wrote (last edit ): Re: kdump cannot generate coredump file on bluefield with 5.4 kernel

I also tested it on 5.15.0-1031-bluefield and it also fails.

Configurations:

root@bu-oob:~# kdump-config show
DUMP_MODE: kdump
USE_KDUMP: 1
KDUMP_COREDIR: /var/crash
crashkernel addr: 0xbd000000
   /var/lib/kdump/vmlinuz: symbolic link to /boot/vmlinuz-5.15.0-1031-bluefield
kdump initrd:
   /var/lib/kdump/initrd.img: symbolic link to /var/lib/kdump/initrd.img-5.15.0-1031-bluefield
current state: ready to kdump

kexec command:
  /sbin/kexec -p --command-line="BOOT_IMAGE=/boot/vmlinuz-5.15.0-1031-bluefield root=UUID=8e8b38a6-7d3d-4a29-b7a0-99761624f941 ro console=hvc0 console=ttyAMA0 earlycon=pl011,0x13010000 fixrtc net.ifnames=0 biosdevname=0 iommu.passthrough=1 console=tty1 console=ttyS0 reset_devices systemd.unit=kdump-tools-dump.service nr_cpus=1" --initrd=/var/lib/kdump/initrd.img /var/lib/kdump/vmlinuz
root@bu-lab60v3-oob:~#

###################################
root@bu-oob:~# dmesg |grep -i crash
[ 0.000000] crashkernel reserved: 0x00000000bd000000 - 0x00000000fd000000 (1024 MB)
[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-5.15.0-1031-bluefield root=UUID=8e8b38a6-7d3d-4a29-b7a0-99761624f941 ro console=hvc0 console=ttyAMA0 earlycon=pl011,0x13010000 fixrtc net.ifnames=0 biosdevname=0 iommu.passthrough=1 console=tty1 console=ttyS0 crashkernel=2G-4G:320M,4G-32G:1024M,32G-64G:1536M,64G-128G:2048M,128G-:4096M
[ 5.230439] pstore: Using crash dump compression: deflate
root@bu-oob:~#

################
root@bu-oob:~# cat /etc/default/grub.d/kdump-tools.cfg
GRUB_CMDLINE_LINUX_DEFAULT="$GRUB_CMDLINE_LINUX_DEFAULT crashkernel=2G-4G:320M,4G-32G:1024M,32G-64G:1536M,64G-128G:2048M,128G-:4096M"

root@bu-lab60v3-oob:~# grep -e "CRASH" -e "KEXEC" /boot/config-5.15.0-1031-bluefield
CONFIG_KEXEC=y
CONFIG_KEXEC_FILE=y
CONFIG_KEXEC_SIG=y
CONFIG_KEXEC_IMAGE_VERIFY_SIG=y
CONFIG_CRASH_DUMP=y
CONFIG_CRASH_CORE=y
CONFIG_KEXEC_CORE=y
CONFIG_HAVE_IMA_KEXEC=y
CONFIG_IMA_KEXEC=y

*** How to reproduce ***
When manually triggers the crash "echo c > /proc/sysrq-trigger"
the system just hangs without showing any message/log.