** Changed in: powerd
Status: Fix Committed = Fix Released
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen; exception thrown from
This bug was fixed in the package powerd -
0.13+14.04.20140108.1-0ubuntu1
---
powerd (0.13+14.04.20140108.1-0ubuntu1) trusty; urgency=low
[ Ricardo Salveti de Araujo ]
* Using pkg-config files from hybris, and depending on latest version
.
[ Seth Forshee ]
*
If I understand, we have a fix, so lets move with that.
We can always improve mir to be more robust, but the heart of this
problem was that mir/unity-mir was not given the guarantee that it needs
(the system on enough to start operations).
--
You received this bug notification because you are a
** Changed in: powerd
Assignee: Kevin DuBois (kdub) = Seth Forshee (sforshee)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen; exception
Fix committed into lp:powerd at revision None, scheduled for release in
powerd, milestone ubuntu-13.05
** Changed in: powerd
Status: In Progress = Fix Committed
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
I disagree. Mir shouldn't behave badly because the display driver is
behaving badly. And I don't think crashing really helps anyway. Won't
unity just be respawned and again try to use the display device without
doing anything to fix the unrecoverable state?
If the display driver still hangs after
** Branch linked: lp:~sforshee/powerd/fix-display-timing
** Changed in: mir
Milestone: 0.1.4 = None
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking
On Fri, Dec 20, 2013 at 01:59:20AM -, Daniel van Vugt wrote:
Forgive my silly question, but:
Why can't we just be more fault-tolerant and avoid the offending throw from
Mir? Log a warning at most, but don't intentionally crash on it.
+1. However ...
Would that result in an interaction
With the way things are now, I'm pretty sure if we simply retried in mir
after experiencing an error, we would hang the driver. Once we get these
throws, the driver has essentially been put into an unrecoverable state.
It seems Seth's investigation aligned with my intuition about the
driver.
--
I've been poking at this a bit today, and I think it comes down to a
race between powerd and the kernel's earlysuspend code. Here's the
sequence of events I see in the logs for one case (there may also be
other sequences that are problematic).
1. The initial state is that the screen is on. The
Forgive my silly question, but:
Why can't we just be more fault-tolerant and avoid the offending throw from
Mir? Log a warning at most, but don't intentionally crash on it.
Would that result in an interaction with powerd rendering the system
unusable? Or would it be sufficient to allow some
** Changed in: mir
Milestone: 0.1.3 = 0.1.4
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen; exception thrown from
** Also affects: powerd
Importance: Undecided
Status: New
** Changed in: powerd
Status: New = In Progress
** Changed in: powerd
Assignee: (unassigned) = Kevin DuBois (kdub)
** Changed in: mir (Ubuntu)
Status: Triaged = Invalid
--
You received this bug notification
I've been able to avert the problem on mako by patching powerd.
Currently cooking up a patch to propose to powerd that works in the
generic case. The root of the problem was that devices with early
suspend were turning mir on and off when they did not guarantee the
framebuffer was in a usable
this problem would only affect devices that use 'earlysuspend'. Other
devices do not have a second thread messing with the display state
tracking, only earlysuspend devices would spawn the thread that was
causing problems.
** Changed in: mir
Milestone: 0.1.3 = None
--
You received this bug
** Branch linked: lp:~kdub/powerd/fix-1258655
** Changed in: mir
Status: In Progress = Invalid
** Changed in: powerd
Importance: Undecided = Critical
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
regardless of whether we throw or not on unblank, the problem at the
root of this bug is that mir and powerd are not synced up about the
power state of the system (clocks, suspend, fb available, etc). This
problem sometimes manifests in failure in the blank/unblank request,
sometimes a hang
i've identified some problems in powerd that cause double requests to be
called. currently working to eliminate those in that code base.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
mir needs to be robustified against the double on and double off
calls. Even when I account for this though, the system will still run
into problems.
I still have reason to suspect that when the dbus message to unity mir
to turn the display on is received, the system is still in the suspend
** Changed in: mir (Ubuntu)
Status: In Progress = Triaged
** Changed in: mir (Ubuntu)
Importance: Undecided = Critical
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort
There is only one possible exception thrown from the offending
mga::HWCCommonDevice::mode()
if (err)
{
std::string blanking_status_msg = Could not +
((mode_request == mir_power_mode_off) ? std::string(blank) :
std::string(unblank)) + display;
Gerry collected this helpful dbus log
http://pastebin.ubuntu.com/6567567/ (also attached) that shows we get an
on, on message from dbus.
the error shown:
terminate called after throwing an instance of
I also did a mir-only test to make sure mir can perform a display on/off
in quick succession in a loop (which it can), as well as on/display some
frames/off in a loop (which it also can). So it doesn't look like the
drivers are breaking on us, just something between mir/unity-
mir/unity8/powerd is
** Changed in: mir
Importance: High = Critical
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen; exception thrown from
From mir's perspective, i see start, stop, start, stop, ..., start,
start being called on the compositor. The double start, start causes
an error in the compositor. (specifically, an egl context is current in
two threads, causing an error BAD_ACCESS in the second thread).
powerd-cli listen just
** Tags removed: regression
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen
To manage notifications about this bug go to:
** Changed in: mir (Ubuntu)
Status: Confirmed = In Progress
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen
To manage notifications about
** Summary changed:
- Abort when blanking/unblanking screen
+ Abort when blanking/unblanking screen; exception thrown from
mir::graphics::android::HWCCommonDevice::mode(MirPowerMode)
** Project changed: unity8 = mir
** Changed in: mir
Assignee: (unassigned) = Kevin DuBois (kdub)
**
Not a regression but something happened with latest Unity8 that made it
way easier to be reproduced. Just used latest image for a few minutes
and was able to reproduce the crash a few times already.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is
** Changed in: mir (Ubuntu)
Assignee: (unassigned) = Kevin DuBois (kdub)
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen
To manage
This is not a regression, I've reported this in October (bug #1238287),
it just got lost in the slew of other raise() bugs, which LP
unfortunately marks duplicate over-eagerly - and then I've mistakenly
marked it Incomplete...
** Summary changed:
- [regression] Abort when blanking/unblanking
** Attachment added: Crash file.
https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1258655/+attachment/3924517/+files/_usr_bin_unity8.32011.crash
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
** Changed in: unity8 (Ubuntu)
Status: New = Confirmed
** Also affects: unity8
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort
Status changed to 'Confirmed' because the bug affects multiple users.
** Changed in: mir (Ubuntu)
Status: New = Confirmed
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort
** Changed in: unity8
Status: New = Triaged
** Changed in: unity8
Importance: Undecided = High
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking
This happens on #51 on mako.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1258655
Title:
Abort when blanking/unblanking screen
To manage notifications about this bug go to:
** Tags added: regression
** Tags added: r51
** Summary changed:
- Abort when blanking/unblanking screen
+ [regression] Abort when blanking/unblanking screen
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
37 matches
Mail list logo
