Bug 10456

Yi, did this work on 2.1 or 2.2? If so then it's a Medium+ bug but if not then it's an enhancement and likley something that we would not do unless there's a clear motivation.

(In reply to comment #1)
> Yi, did this work on 2.1 or 2.2? If so then it's a Medium+ bug but if not
> then it's an enhancement and likley something that we would not do unless
> there's a clear motivation.

Hi Randy,

This is a regression bug, used to work on 2.1 and in previous testing for 2.2 it also works.

Thanks,
Yi

This issue still happens on 2.3 M1 rc1 build

I'm 99% sure X can in fact launch correctly: you can even start apps if you guess where to click. The issue almost certainly arrived at b9d90ace0055 "poky: Update to linux-yocto 4.8 for qemu* machines".

(In reply to comment #4)
> I'm 99% sure X can in fact launch correctly: you can even start apps if you
> guess where to click. The issue almost certainly arrived at b9d90ace0055
> "poky: Update to linux-yocto 4.8 for qemu* machines".

I tested this: That commit does introduce the issue.

(In reply to comment #5)
> (In reply to comment #4)
> > I'm 99% sure X can in fact launch correctly: you can even start apps if you
> > guess where to click. The issue almost certainly arrived at b9d90ace0055
> > "poky: Update to linux-yocto 4.8 for qemu* machines".
> 
> I tested this: That commit does introduce the issue.

Yes, I change PREFERRED_VERSION_linux-yocto_qemumips64 in poky.conf to "4.4%" on the latest oe-core upstream, it is ok for the core-image-sato, I will work on this continuely.

I tested it on kernel 4.9, that is ok.

This is because the Upstream commit in 4.8: 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
commit d03516df837587368fc6e75591f6329c072b9eb5
Author: Matthew Leach <matthew.leach@codethink.co.uk>
Date:   Tue Jun 21 16:19:49 2016 +0100

    tty: serial: 8250: add CON_CONSDEV to flags
    
    When using the 8250 as a boot console and the main console results in
    messages being printed twice.  The console framework will only
    unregister boot consoles if a new console is registered with the
    CON_CONSDEV flag set.
    
    Set this flag for the univ8250 console to prevent double-registration.
    
    Signed-off-by: Matthew Leach <matthew.leach@codethink.co.uk>
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The version 4.4 kernel didn't contain, and the 4.9 kernel have been reverted it. I
have backport the reverted patch to yocto-2.2.1

The fixed commit was:

  http://git.yoctoproject.org/cgit/cgit.cgi/linux-yocto-4.8/commit/?h=standard/mti-malta64&id=57061c1d3c2e35b4676cc431098899a3f3d8eb19

This issue still happens on 2.2.2 rc1.

Git rev: morty/d05941ae4567def4a288894717e5f550da246107

The patch had been merged into linux-yocto-4.8 repo, but seems the SRCREV_machine_qemumips64 did not update.

Bruce, can you do an upgrade for morty?

(In reply to comment #10)
> This issue still happens on 2.2.2 rc1.
> 
> Git rev: morty/d05941ae4567def4a288894717e5f550da246107
> 
> The patch had been merged into linux-yocto-4.8 repo, but seems the
> SRCREV_machine_qemumips64 did not update.

The issue is the fix went in just after the 4.8.20 sync. Our current version is 4.8.18. The offending commit went in @ v4.10-rc3-5.



the mip32 fix went in earlier than other machines.
http://git.yoctoproject.org/cgit/cgit.cgi/linux-yocto-4.8/commit/?h=standard/mti-malta32&id=ecc76d8ac3c3d03fac582de1013ff02847b0d86d

ok. digging a bit more in linux-yocto-4.8, the offending commit went in with via


d03516d tty: serial: 8250: add CON_CONSDEV to flags

back in 4.7 so we should have seen this issue since day one for Morty.

Do we know if this was tested with morty when it was released?

(In reply to comment #14)
> Do we know if this was tested with morty when it was released?

According to Comment 2 this was tested on 2.2 and was working correctly.

(In reply to comment #15)
> (In reply to comment #14)
> > Do we know if this was tested with morty when it was released?
> 
> According to Comment 2 this was tested on 2.2 and was working correctly.

ah, missed that. Then the commit noted should not fix this as 2.2 release with

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
commit d03516df837587368fc6e75591f6329c072b9eb5
Author: Matthew Leach <matthew.leach@codethink.co.uk>
Date:   Tue Jun 21 16:19:49 2016 +0100

    tty: serial: 8250: add CON_CONSDEV to flags

if I am not mistaken Morty released mips64 with
SRCREV_machine_qemumips64 ?= "64f96ba530e58456070f26b0f3fcce3f64988b72"

git log --oneline -4 drivers/tty/serial/8250/8250_core.c
5db4f7f Revert "tty/serial/8250: use mctrl_gpio helpers"
4ef03d3 tty/serial/8250: use mctrl_gpio helpers
d03516d tty: serial: 8250: add CON_CONSDEV to flags <-- the fixed noted reverts this.

issue may be else were

errors I am seeing:

INIT: version 2.88 booting
Error, no support currently for 8 bpp frame buffers
Trying to change pixel format...
Error, failed to switch to a 32 bpp 8,8,8 frame buffer

later

bluetoothd
hwclock: can't open '/dev/misc/rtc': No such file or directory
[   39.369758] 
[   39.369758] do_page_fault(): sending SIGSEGV to bluetoothd for invalid read access from 0000000000000030
[   39.369882] epc = 0000007db56dfc88 in libc-2.24.so[7db5650000+192000]
[   39.369915] ra  = 0000007db56ed744 in libc-2.24.so[7db5650000+192000]
[   39.369919] 
Starting syslogd/klogd: done
 * Starting Avahi mDNS/DNS-SD Daemon: avahi-daemon                       [ ok ]
Starting Telephony daemon

Poky (Yocto Project Reference Distro) 2.2.2 qemumips64 /dev/ttyS0

qemumips64 login: D-BUS per-session daemon address is: unix:abstract=/tmp/dbus-ZAGNgQtPr1,guid=a3ad8bf3051e42c54ec8f96759372d3e
Using calibration data stored in /etc/pointercal.xinput
Using calibration data stored in /etc/pointercal.xinput
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed

** (process:467): WARNING **: Failed to register client: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.gnome.SessionManager was not provided by any .service files
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
alloc color failed
matchbox: Cant find a keycode for keysym 269025056
matchbox: ignoring key shortcut XF86Calendar=!$contacts

matchbox: Cant find a keycode for keysym 2809
matchbox: ignoring key shortcut telephone=!$dates

matchbox: Cant find a keycode for keysym 269025050
matchbox: ignoring key shortcut XF86Start=!matchbox-remote -desktop

libGL error: No matching fbConfigs or visuals found
libGL error: failed to load driver: swrast
[settings daemon] Forking. run with -n to prevent fork
libGL error: No matching fbConfigs or visuals found
libGL error: failed to load driver: swrast
libGL error: No matching fbConfigs or visuals found
libGL error: failed to load driver: swrast
libGL error: No matching fbConfigs or visuals found
libGL error: failed to load driver: swrast
Activating service name='org.a11y.atspi.Registry'
Successfully activated service 'org.a11y.atspi.Registry'
SpiRegistry daemon is running with well-known name - org.a11y.atspi.Registry

** (matchbox-panel:489): WARNING **: Failed to load applet "battery" (/usr/lib/matchbox-panel/libbattery.so: cannot open shared object file: No such file or directory).
matchbox-wm: X error warning (0xa00002): BadWindow (invalid Window parameter) (opcode: 12)

I manually updated the 4.8 kernel to 4.8.20 which includes the revert patch and I still have the issue.

I am going back to 2.2.1 to ensure my env will work and will march forward from there to see what change is doing this.

I forgot how bloody slow this machine is.

I went back to the original release point and I got the calibration screen, then  now Icons. I started clickin in the black space and I got the system to shutdown.

I saw the "Black" screen with the 2.2.1 and assumed there was a problem. I am going to retry 2.2.1 with a bit more patients this them to see if I the the calibration screen.

the linux-yocto-4.8 was updated to 4.8.24. I used that version and I get the full graphic display with the sato image.

Here is my take. This issue existed since Morty was shipped so its not a regressions for Morty. More of a known issue. It will be fixed once the kernel version moved past 4.8.19.

In order to fix this all machines will have new kernel versions. Not sure what it does to the QA testing process ( ie restarts things).

If you really want this fixed and don't mind slipping schedule then updating the kernel is fine. If schedule is important, then ship with this issue and it will be fixed in kernel sync.

I pinged Bruce about the 4.8 kernel cache version being 22 and the linux-yocto sources having been synced to .24 and he sent a patchset to move both to latest for the 4.8 kernel . This fixed the qemumips64 X issue. I pushed these changes to my contrib branch.

this works with the updated kernel

code in morty

Verified with 2.2.2 rc2
git rev: morty/60402978fe3648bf560b3386c6e9dd661cdf2083