diff options
author | Maciej W. Rozycki <macro@linux-mips.org> | 2007-07-18 09:49:11 +0200 |
---|---|---|
committer | Linus Torvalds <torvalds@woody.linux-foundation.org> | 2007-07-18 17:38:22 +0200 |
commit | 8b4a40809e5330c9da5d20107d693d92d73b31dc (patch) | |
tree | 14de7320d32a7e72dc2cddf4833405db0f49a7ba /fs | |
parent | serial: add early_serial_setup() back to header file (diff) | |
download | linux-8b4a40809e5330c9da5d20107d693d92d73b31dc.tar.xz linux-8b4a40809e5330c9da5d20107d693d92d73b31dc.zip |
zs: move to the serial subsystem
This is a reimplementation of the zs driver for the serial subsystem. Any
resemblance to the old driver is purely coincidential. ;-) I do hope I got
the handling of modem lines right -- better do not tackle me about the
issue unless you feel too good...
Any users of the old driver: please note the numbers of the serial lines
have now been swapped, i.e. ttyS0 <-> ttyS1 and ttyS2 <-> ttyS3. It has
to do with the modem lines mentioned above; basically the port A in a given
chip has to be initialised before the port B if you want to use the latter
as the serial console (which is usually the case), as operations on modem
lines of the serial line associated with the port B access both ports (see
the comment at the top of the driver for the details of wiring used).
Please update your scripts.
This is also the reason each SCC now requests an IRQ once only (as seen in
"/proc/interrupts") -- the handler takes care of both ports at once as the
line associated with the port B has to take status update interrupts from
both ports (and yet the line of the port A takes its own for itself too).
The old driver never got it right...
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'fs')
0 files changed, 0 insertions, 0 deletions