For what its worth, if the task is running as "real time" nanosleep will
use udelay() instead of the clock (I would like to see some rational for
this). This would cause the sluggishness. As to what the key click
does, I am lost.
George
brian@worldcontrol.com wrote:
>
> I posted earlier that my daughter found that gphoto could no longer
> download pictures from her digital camera after I upgraded
> her machine to 2.2.16.
>
> I tried 2.2.17pre6 to see if it made any difference.
>
> I also did an strace against gphoto 0.2, 0.3, and 0.4.3.
>
> Basically all versions do the same thing. After I click
> on the button to fetch the picture index from the camera
> the code does:
>
> nanosleep({0, 1000000}, NULL) = 0
> write(6, "\0", 1) = 1
> select(7, [6], [], [6], {1, 0}) = 1 (in [6], left {0, 610000})
> read(6, "\25", 1) = 1
> select(7, [6], [], [6], {0, 0}) = 0 (Timeout)
> nanosleep({0, 250000}, NULL) = 0
> write(6, "\33", 1) = 1
> nanosleep({0, 250000}, NULL) = 0
> write(6, "S\6\0", 3) = 3
> nanosleep({0, 500000}, NULL) = 0
> write(6, "\0\21\5\0\0\0\26\0", 8) = 8
> select(7, [6], [], [6], {0, 400000}) = 1 (in [6], left {0, 400000})
> read(6, "\25", 1) = 1
> nanosleep({0, 250000}, NULL) = 0
> write(6, "\33", 1) = 1
> nanosleep({0, 250000}, NULL) = 0
> write(6, "S\6\0", 3) = 3
> nanosleep({0, 500000}, NULL) = 0
> write(6, "\0\21\5\0\0\0\26\0", 8) = 8
>
> This code calls nanosleep a total of 58 times and takes 15 seconds of
> real time if left to itself. During this time the X pointer is stuck in
> molasses. After which it reports the timeout on the screen.
>
> If I press a key on the keyboard at any time during the 15 seconds, the
> code zips by to the point where it reports the timeout.
>
> I.E. click (wait 15 second) for timeout
> or
> click (press key) timeout shows up more or less instantly.
>
> The machine runs XFree86 4.0, has an AMD K6-3/400, and 128MB of RAM.
>
> --
> Brian Litzinger <brian@litzinger.com>
>
> Copyright (c) 2000 By Brian Litzinger, All Rights Reserved
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jun 26 2000 - 21:00:09 EST