Bug 5121 - Starting apmd failed at system startup
Summary: Starting apmd failed at system startup
Status: RESOLVED FIXED
Alias: None
Product: System Startup
Classification: Runtime
Component: system-startup (show other bugs)
Version: 1.5
Hardware: x86 Multiple
: Medium normal
Target Milestone: 1.5 M5
Assignee: Saul Wold
QA Contact:
URL:
Whiteboard:
Depends on:
Blocks:
 
Reported: 2013-09-05 06:22 UTC by Chen Qi
Modified: 2013-09-09 20:10 UTC (History)
4 users (show)

See Also:
OS type for building Yocto: ---
Type of Regression: ---
Verified:
Documentation change: No (bug/feature does not impact docs)

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Chen Qi 2013-09-05 06:22:11 UTC 
The error message is:
Starting advanced power management daemon: No APM support in kernel
(failed.)

The problem is that CONFIG_APM is not enabled in kernel, but the MACHINE_FEATURES contains 'apm'. 

meta/conf/machine/include/qemu.inc:MACHINE_FEATURES = "apm alsa pcmcia bluetooth irda usbgadget screen"
meta-yocto-bsp/conf/machine/beagleboard.conf:MACHINE_FEATURES = "apm usbgadget usbhost vfat alsa"

The apmd package is installed because MACHINE_FEATURES contains 'apm', but CONFIG_APM is not enabled, and there's no corresponding kernel configuration fragment.

//Chen Qi
Comment 1 Bruce Ashfield 2013-09-05 12:45:19 UTC 
Let's just update the fragments to contain the right modules, and not over design
a solution to this.

We already have bugzilla entries around mapping machine features -> fragments, and
the balancing of the BSPs need to specify what it actually supports, versus the abstraction
of a build system turning on functionality that can't actually work on the board .. simply
because a recipe or high level function asks for it. That's 1.6 and beyond work.

For now, let's just tweak the config appropriately.
Comment 2 Saul Wold 2013-09-05 14:45:30 UTC 
Any reason not to drop apm from the qemu MACHINE_FEATURES, instead of adding it into the kernel config?

Richard believes that it's only useful on older x86 and older arm, and mostly dead at point, it not useful in the virtual environments and questionable in beagleboard.

No not adding it to kernel would be better.

I am going to do a review of apmd and it's general usefulness.