You mentioned that the console changes were not forward ported from
lowlatency-2.2.10-N6B to lowlatency-2.2.13-A1. After a few diffs,
it appears that all console changes _are_ ported with the exception
of the /drivers/char/pc_keyb.c changes. I am now exercising X
on the console and I notice a 5msec latency when I switch to a
virtual terminal then back to X, the kernel functions responsible
are the delays in pc_keyb.c with respect to psaux mouse init.
I applied the 2.2.10-N6B pc_keyb.c changes and the problem shifted
to a new group of functions:
3887204.62 0.15 detect_auxiliary_port pid(113)
3887204.78 0.13 kb_wait pid(113)
3887204.90 3.33 pckbd_unexpected_up pid(113)
3887208.23 0.13 kb_wait pid(113)
3887208.36 1.62 pckbd_unexpected_up pid(113)
3887209.98 0.13 __down_failed_trylock pid(113)
3887210.10 996.07 csum_partial_copy pid(113)
3888206.18 1.51 pckbd_unexpected_up pid(113)
3888207.69 0.13 __down_failed_trylock pid(113)
3888207.82 996.07 csum_partial_copy pid(113)
3889203.89 1.60 pckbd_unexpected_up pid(113)
3889205.50 0.13 __down_failed_trylock pid(113)
3889205.63 996.07 csum_partial_copy pid(113)
The "csum_partial_copy" now looks suspect. I'm not sure if this happened
in 2.2.10? Did I break something?
I am using the vga console instead of the frame buffer stuff so
I didnt have to do anything for the fbcon drivers.
Wm
-
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/