运行失败

[Hoblovski]

lkp qemu 在 oscenter 与个人 PC 上, 运行 hackbench 和 boot 均失败.

错误以及复现

在 f1f87a26fc3e3cfc1e3607d9d08ad158c1b32c23 上执行如下命令复现

lkp split-job lkp-tests/jobs/boot.yaml
lkp install boot-1.yaml
lkp job2sh boot-1.yaml -o boot-1.sh
lkp qemu boot-1.sh

错误显示为, 卡死在

dzy@oslab-i9:/data/dzy/linux/kis/run-lkp % sudo lkp qemu boot-1.sh
find: ‘boot-1.yaml/*’: No such file or directory
1 block
dirname: missing operand
Try 'dirname --help' for more information.
result_root: /home/dzy/.lkp//2
downloading kernel image ...
/usr/bin/wget -q --timeout=1800 --tries=1 --local-encoding=UTF-8 https://download.01.org/0day-ci/lkp-qemu/ -N -P /home/dzy/.lkp
downloading initrds ...

事实上这一步是 (不知为何) 等待 stdin 输入, Ctrl-D 直接输入 EOF 又出现错误

exec command: qemu-system-x86_64 -enable-kvm -fsdev local,id=test_dev,path=/home/dzy/.lkp//2,security_model=none -device virtio-9p-pci,fsdev=test_dev,mount_tag=9p/virtfs_mount -kernel /home/dzy/.lkp/cache/ -append root=/dev/ram0 job=boot-1.yaml user=  ARCH=x86_64 kconfig=x86_64-rhel commit=051d101ddcd268a7429d6892c089c1c0858df20b branch=linux-devel/devel-hourly-2015033109 max_uptime=1247 RESULT_ROOT= earlyprintk=ttyS0,115200 rd.udev.log-priority=err systemd.log_target=journal systemd.log_level=warning debug apic=debug sysrq_always_enabled rcupdate.rcu_cpu_stall_timeout=100 panic=-1 softlockup_panic=1 nmi_watchdog=panic oops=panic load_ramdisk=2 prompt_ramdisk=0 console=ttyS0,115200 console=tty0 vga=normal rw ip=dhcp result_service=9p/virtfs_mount -initrd /home/dzy/.lkp/cache/final_initrd -smp 32 -m 63173M -no-reboot -watchdog i6300esb -rtc base=localtime -device e1000,netdev=net0 -netdev user,id=net0 -display none -monitor null -serial stdio              
......
qemu-system-x86_64: warning: host doesn't support requested feature: CPUID.80000001H:ECX.svm [bit 2]                                                                                             
qemu-system-x86_64: warning: host doesn't support requested feature: 
.......
qemu: could not load kernel '/home/dzy/.lkp/cache/': Is a directory

因为根本没有下载 kernel. (注意上面的 -kernel /home/dzy/.lkp/cache/)

[elliott10]

调试发现卡住的于“ cat > $concatenate_initrd”；
原因应该是$initrds变量为空状态；

提供了一个该bug的修复patch如下：

diff --git a/lib/kexec.sh b/lib/kexec.sh
index c34b12d9086a..214c40c9306c 100755
--- a/lib/kexec.sh
+++ b/lib/kexec.sh
@@ -134,7 +134,7 @@ download_initrd()
        # modules can not be the first, must be behind initrd
        initrds="${initrds} $local_modules_initrd"
 
-       [ -n "$initrds" ] && {
+       [ -n "${initrds%% *}" ] && {
                [ $# != 0 ] && initrds="${initrds}$*"
 
                concatenate_initrd="$CACHE_DIR/initrd-concatenated"

[elliott10]

请尝试：
lkp split -s 'user: lkp' -s 'kernel: /fbc/5b51ae969e3d8ab0134ee3c98a769ad6d2cc2e24/vmlinuz-5.2.0-rc3-00004-g5b51ae969e3d8a' -s 'initrd: /osimage/debian/debian-x86_64-2019-09-23.cgz' jobs/boot.yaml