On Wed, 25 Feb 2009 06:57:42 -0500, "Paul O'Keefe" <paul@xxxxxxxxxxxxx> wrote:
> strace attached
> lsusb attached
>
> still "running" hasn't returned.
You need to interrupt it to get strace to write out the output.
The attachment was truncated:
gettimeofday({1235547424, 341658}, NULL) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfd69bf8) = 0
futex(0x8106964, FUTEX_WAIT_PRIVATE, 5, NULL) = 0
futex(0x8106948, FUTEX_WAIT_PRIVATE, 2, NULL) = 0
futex(0x8106948, FUTEX_WAKE_PRIVATE, 1) = 0
gettimeofday({1235547424, 342853}, NULL) = 0
ioctl(3, USBDEVFS_SUBMITURB, 0xbfd699a4) = 0
ioctl(3, USBDEVFS_REAPURBNDELAY, 0xbfd699e8) = 0
futex(0x8106964, FUTEX_WAIT_PRIVATE, 7, NULL <------------ here
What survived was not consistent with the previously posted usbmon
log in any way. It would be best to get both from the same run.
Also, strace -tt should help to corellate the events (usbmon does
not print the absolute time as strace, but we can subtract stamps
and see which event is which).
I'm wondering why Chris needs to use REAPURBNDELAY for
a non-interactive tool, especially by one which also uses futex!
Usually it's called by single-threaded GUI applications that want
GUI response across long operations. But there's nothing illegal
about doing it.
-- Pete
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
