Created attachment 1482 [details] dmesg and lsmod outputs System hangs on boot process on stage when Yocto progress bar almost on the end. Sometimes it's possible to login in text terminal (after switching with Ctrl+alt+f1), but usually system hangs and doesn't respond to anything. Reproducibility: 1-10% Occurs with: - released Yocto Project 1.4.2 - Poky 9.0.2 with released meta-crownbay layer (9.0.0 "Dylan") - Yocto Project 1.5_M4.rc3 with meta-crownbay layer 1.5_M4.rc3 Test machine - Intel® Atom™ Processor E680T (512K Cache, 1.60 GHz) Steps to reproduce: 1. Build core-image-sato image 2. Copy it on machine's SSD 3. Boot system Please see in attached files output of lsmod and dmesg commands (case when it was possible to login)
Can you try booting with psplash=false on the kernel command line? This will prevent the graphical splash screen from appearing. Then please provide the final lines of output at the time of the hang.
(In reply to comment #1) > Can you try booting with psplash=false on the kernel command line? This will > prevent the graphical splash screen from appearing. Then please provide the > final lines of output at the time of the hang. Below output which you asked about: X.Org X Server 1.9.3 Release Date: 2010-12-13 X Protocol Version 11, Revision 0 Build Operating System: Linux 3.0.0-32-generic-pae i686 Current Operating System: Linux crownbay 3.8.13-yocto-standard #1 SMP PREEMPT Fri Sep 6 17:32:08 EEST 2013 i686 Kernel command line: ro root=/dev/sda2 video=vesafb vga=0x318 psplash=false BOOT_IMAGE=/boot/bzImage Build Date 06 September 2013 05:33:40PM Current version of pitman: 0.30.2 Before reporting problems, check http://wiki.x.org to make sure that you have latest version. Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II)informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (==) Log file: "/var/log/Xorg.0.log", Time: Fri Sep 6 16:20:00 2013 (==) Using config file: "etc/X11/xorg.conf" (==) Using system config directory "use/share/X11/xorg.conf.d" Public key portion is: ssh-rsa XXXXXXXXXXXXXX…….. root@crownbay Fingerprint: md5 XX:XX:XX….. dropbear. Starting Advanced Configuration and Power Interface daemon: INFO: rcu_preempt detected stalls on CPUs/tasks: {} (detected by 0, t=21002, jiffies, g=749, c=748, q=277) INFO: Stall ended before state dump start INFO: rcu_preempt detected stalls on CPUs/tasks: {} (detected by 0, t=84007, jiffies, g=749, c=748, q=529) INFO: Stall ended before state dump start INFO: rcu_preempt detected stalls on CPUs/tasks: {} (detected by 0, t=147012, jiffies, g=749, c=748, q=781) INFO: Stall ended before state dump start INFO: rcu_preempt detected stalls on CPUs/tasks: {} (detected by 0, t=210017, jiffies, g=749, c=748, q=1033) Last lines are repeated permanently with different values of 't' and 'q'. Keyboard doesn't response to anything (even num or caps buttons).
(In reply to comment #1) > Can you try booting with psplash=false on the kernel command line? This will > prevent the graphical splash screen from appearing. Then please provide the > final lines of output at the time of the hang. One more case: ... INIT: Entering runlevel: 5 Starting system message bus: dbus. Starting Connection Manager Starting Xserver Netfilter messages via NETLINK v0.30. ip_tables: (C) 2000-2006 Netfilter Core Team Starting Dropbear SSH server: nf_conntrack version 0.5.0 (15890 buckets, 63560 max) Will output1024 bit rsa secret key to '/varlib/dropbear/dropbear_rsa_host_key' Generating key, this may take a while... IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready Public key portion is: ssh-rsa XXXXXXXX.... root-crownbay Fingerprint: md5 XX:XX:XX... dropbear. Starting Advanced Configuration and Power Interface daemon: acpid. acpid: starting up hwclock: can't open '/dev/misc/rtc': No such file or directory acpid: 1 rule loaded acpid: waiting for events: event logging is off Starting syslogd/klogd: done * Starting Avahi mDNS/DNS-SD Daemon: avahi-daemon [fail] Starting Telephony daemon Starting Linux NFC daemon Stopping Bootlog daemon: bootlogd. Poky Next (Yocto Project Reference Distro) 1.4+snapshot-20130906 crownbay /dev/tty1 ===== And then a prompt to login into tty1. In this case I'm able to login in system as I mentioned before
I tried to reproduce this issue on my crownbay hardware. And After 10 tries I could not reproduce the issue. What hardware are you using? It is possible that your hardware got damaged, especially the RAM.
(In reply to comment #4) > I tried to reproduce this issue on my crownbay hardware. And After 10 tries > I could not reproduce the issue. It can take big number of attempts to catch this issue. To make it easier I've added script which reboots machine in loop. Even in this case machine can work without the fail for 2-3 hours. > What hardware are you using? I use conga-QA6/E680-1G Qseven module with Intel® Atom™ E680 processor with 1.6GHz, 512kB L2 cache and 1GB DDR2 onboard memory. Based on Intel® Platform Controller Hub EG20. 8GB MLC SSD http://www.congatec.com/en/products/qseven/conga-qa6.html > It is possible that your hardware got damaged, especially the RAM. I think that possibility that hardware got damaged is low because I reproduce this issue on all machines that available for me (4 units)
humm, so you are not using crownbay hardware. This BSP was specifically made for crownbay platform. I think in your case the BIOS or VBIOS may be different from what crownbay has.
(In reply to comment #6) > humm, so you are not using crownbay hardware. This BSP was specifically made > for crownbay platform. I think in your case the BIOS or VBIOS may be > different from what crownbay has. Here is information about ours BIOS: BIOS Version 2.14.1219 American Megatrends. Ver. QTOPR111 IGB VBIOS Version 1904 if you need additional information let me know please
We have 3.10 kernel working with EMGD 1.18 driver now. Once that is upstream, it can be checked if that helps with this issue.
Created attachment 1592 [details] boot logs
Created attachment 1593 [details] boot logs
Created attachment 1594 [details] boot logs
Created attachment 1595 [details] boot logs
I've verified the system (core-image-sato) with Kernel 3.10 and EMGD v1.18. Issue is still reproduced. I've attached 4 screenshots with logs (in kernel command line were set parameters: loglevel=7 and psplash=false). Sometimes during boot the string about bad swap file is printed: swap_dup: Bad swap file entry 0000000N
This involves the binary EMGD X driver from ISG. As I do not have sources for the driver I am passing this bug to ISG.
Hi, The EMGD for ATOM Processor E600 series already in maintenance. Is there still a need to investigate this issue? I may not be able to find someone to help to look reproduce the issue. Set "Target Milestone" to Future
The EMGD for ATOM Processor E600 series already in maintenance. There will be no schedule release anymore for EMGD on Intel Atom E600. However, EMGD still support EMGD for Atom E600 through our PAE for critical issues. If you are working on a project with Intel's PAEs support, please contact them and filed the bug through them.