| Summary: | qemuarm cannot shutdown entirely through UNFS | ||
|---|---|---|---|
| Product: | [Runtime] System Startup | Reporter: | Yi Zhao <yi.zhao> |
| Component: | system-startup | Assignee: | Paul Eggleton <bluelightning> |
| Status: | VERIFIED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | Medium | CC: | dvhart, jiajun.xu, kevin.tian, liang.li2, liangliang.wang, poky.bs.watcher, poky.watcher, sgw, yi.zhao, yp.kernel.watcher, yp.watcher |
| Version: | 1.0 | ||
| Target Milestone: | 1.1 | ||
| Hardware: | x86 | ||
| OS: | arm | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | --- | |
|
Description
Yi Zhao
2011-04-07 00:40:26 UTC
Hi Yi, Is the bug a regression or a new one? (In reply to comment #0) > Testing Env: > Host: Ubuntu 10.10 x86_64 > Kernel: 2.6.35-28-generic > Image location: > http://autobuilder.yoctoproject.org/nightly/20110404-1/machines/qemu/qemuarm/ > Commit information: f897723207b00b4c9dcdb43107891b7b82ce22f3 > Boot up qemuarm through UNFS > ############### > $ poky-extract-sdk poky-image-sato-qemuarm-20110405024447.rootfs.tar.bz2 rootfs > $ poky-qemu nfs > zImage-2.6.37+git1+727fc4769aa920fc5bb3eb9a81bf92d0e6340903_1+e5ca41def834db9d18b28393a45d53a8d18f3d05-r16-qemuarm-20110405234815.bin > rootfs > ############### > when I click the shutdown icon on the desktop, system goes into shutdown > process. But qemu window and qemu process doesn't disappear after that. When I > run command 'reboot', the issue still happens. > When I boot up qemuarm image directly, > ############# > $ poky-qemu qemuarm > zImage-2.6.37+git1+727fc4769aa920fc5bb3eb9a81bf92d0e6340903_1+e5ca41def834db9d18b28393a45d53a8d18f3d05-r16-qemuarm-20110405234815.bin > poky-image-sato-qemuarm-20110405024447.rootfs.ext3 ext3 > ############# > This issue doesn't happen. > You can refer to bug 100 for more information. It is a new bug. But this issue also happens on 1.0 release build. Suggested next step: try with linux-yocto-stable and see if this is due to the switch to linux-yocto. Bruce, will the same folks that worked on Bug 100 be able to have a look at this as they seem to have experience and context here? I presume we are still agreed that "reboot" shutting down qemu is an acceptable solution? reboot is still the right option. the actual hardware shows the same behaviour as qemuarm, it's a property of the bootloader, so we'll live with it. As for why it hangs through nfs, it's likely not a kernel issue. I've seen this plenty of times. check to see if the ethernet / networking is being shut down too early in the process. If it shuts down, NFS hangs and you can't reboot no matter what you do. Saul, can we move this to "Runtime Distribution" to check on the shutdown scripts and the networking shutdown order? Also needs an assignee. changing the product designation, since this really isn't kernel. please update it again if I've got the wrong item. With 1.0.1 build 20110513, this issue does *not* happened. Branch: Bernard git commit: 1b159ff35d9f484e72afb8b19a9e69a2770d8379 With 1.1 20110613 build, this issue does *not* happened. Image location: http://autobuilder.yoctoproject.org/nightly/20110613-1/ Tree/Branch: Poky/master Commit: e5f3cc34df74c08c8cab3e1cf9207fafaec6cb9a I can confirm, self-built image running over nfs (with unfs) shuts down fine here. (Of master, that is) This is apparently now fixed, marking as resolved. Verified it with 20110715 build Image location: http://autobuilder.pokylinux.org/nightly/20110715-1/ Tree/Branch: Poky/1.1_M2 Poky Commit: 93a2ca70482f09e93469481936302e7b6847156e |