[Trac #602] evdev startup is taking too long

Zarro Boogs per Child bugtracker at laptop.org
Wed Dec 27 23:41:20 EST 2006


#602: evdev startup is taking too long
-----------------------------+----------------------------------------------
 Reporter:  ajax             |        Owner:  warp   
     Type:  defect           |       Status:  new    
 Priority:  normal           |    Milestone:  BTest-2
Component:  x window system  |   Resolution:         
 Keywords:                   |  
-----------------------------+----------------------------------------------
Old description:

> I tweaked the X server to print a timestamp before each log write.
> Overall we're taking ~1.9 seconds to start up (well, between first and
> last prints), about 1.5 of which is in input device init.
>
> In the attached log, the number after STAMP: is in milliseconds since the
> epoch.  The code prints the timestamp, then the message; each stamp
> corresponds to the text after it, in other words.
>
> Note the following sequence at the end:
>
> STAMP: 2650347848
> (II) PT-isa0060/serio1/input0: Init
> STAMP: 2650348611
> (II) ATKbd-isa0060/serio0/input0: Init
> STAMP: 2650349360
> (II) Keyboard-geode_pmc/input0: Init

New description:

 I tweaked the X server to print a timestamp before each log write.
 Overall we're taking ~1.9 seconds to start up (well, between first and
 last prints), about 1.5 of which is in input device init.

 In the attached log, the number after STAMP: is in milliseconds since the
 epoch.  The code prints the timestamp, then the message; each stamp
 corresponds to the text after it, in other words.

 Note the following sequence at the end:

 STAMP: 2650347848[[BR]]
 (II) PT-isa0060/serio1/input0: Init[[BR]]
 STAMP: 2650348611[[BR]]
 (II) ATKbd-isa0060/serio0/input0: Init[[BR]]
 STAMP: 2650349360[[BR]]
 (II) Keyboard-geode_pmc/input0: Init[[BR]]

-- 
Ticket URL: <http://dev.laptop.org/ticket/602#comment:2>
One Laptop Per Child <http://laptop.org/>



More information about the Bugs mailing list