NaNs in multiple screens

Bug #608802 reported by Jodie Cunningham
30
This bug affects 8 people
Affects Status Importance Assigned to Milestone
g15stats (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

Binary package hint: g15stats

After much debugging, I have determined that when a meter displays "-nan" in the meter, it causes the lower information to cycle erratically. This obviously only happens when the machine is extremely idle.
Under certain conditions if the LCD keys are pressed while a "-nan" is displayed, the entire display will cycle erratically as quickly as the system allows, until g15stats is killed.
Under typical conditions, the lower portion of the screen should only cycle every second, per line 1852: g15stats_wait(1);

From looking at the sprintfs it does not look like NaN handling is taking place, eg: lines 777, 779, 781
I know little about C but I will see if I can come up with a patch. This is driving me nuts.

Package: g15stats 1.9.2-1
Description: Ubuntu 10.04 LTS

summary: - NaNs in mutliple screens
+ NaNs in multiple screens
Revision history for this message
Jodie Cunningham (jodie-cunningham) wrote :

Having made some modifications to prevent the -NaN from happening, I notice that the lower "info" portion of the display still cycles erratically.

So this is happening somewhere else, possibly in the graphing functions.
When the CPU is at 100%, the lower "info" portion of the screen cycles erratically as well. I suspect a >0 exit in the graphing function.

Revision history for this message
Jodie Cunningham (jodie-cunningham) wrote :

With a fresh install of Precise and g15stats 1.9.2-2 this still happens in the exact same manner.

Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in g15stats (Ubuntu):
status: New → Confirmed
Revision history for this message
zapman (zapman) wrote :

I'm using Precise and this bug affects me too. This is apparently fixed upstream but the latest version needs to be packaged. A Debian bug report was submitted (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=677446). Bug 749834 which is a duplicate of this one has additional information.

Revision history for this message
Ben Shadwick (benshadwick) wrote :

No response on the upstream bug report after a year and a half. I just sent an email to the package maintainer listed here: http://packages.debian.org/sid/g15stats

Note that the sourceforge page now has a .deb package for amd64, which installed fine for me on Kubuntu 13.10 x64 except for the fact that g15stats is not configured to start up by default.

What is the process for getting this updated at the Ubuntu level if the Debian package maintainers won't update it upstream?

Revision history for this message
Ben Shadwick (benshadwick) wrote :

Update: Got a promising response from the upstream package maintainer:

"OK, I'll try to update them shortly.
BTW if you want to help, become to maintainer, co-maintainer, or just to
learn how to pack packages, you are welcome."

Revision history for this message
Ben Shadwick (benshadwick) wrote :

Update: It appears that Debian has dropped the ball on this. Are there any other end-user options for relief?

Revision history for this message
dlancer (dlancer) wrote :

Only my old PPA repo (for Precise, Oneiric and Quantal):
https://launchpad.net/~dlancer/+archive/g15stats

I moved to 14.04 now, so I probably update this ppa soon.

Revision history for this message
Ben Shadwick (benshadwick) wrote :

@dlancer: Thanks. I saw your PPA but noticed that it was almost 2 years and 3 Ubuntu releases out of date. I'll keep an eye on it.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.