summaryrefslogtreecommitdiff
path: root/bin/ls
diff options
context:
space:
mode:
authorDavid Gwynne <dlg@cvs.openbsd.org>2005-01-10 12:05:09 +0000
committerDavid Gwynne <dlg@cvs.openbsd.org>2005-01-10 12:05:09 +0000
commit4f1b83eb275a9bc760466bc4acf3d006a464cff7 (patch)
tree694d953e45810f113d20361bc2dda9ae1b41c046 /bin/ls
parent2ba2e9c431f5064e237b2e54f6f2a6b86b2ac07c (diff)
from netbsd uplcom.c revision 1.37:
Make sure that we call uplcom_set_line_state() to get consistent state between our idea of flow controll settings and the chips one. In uplcom_set_line_state use the same value as the manufacturer suggests, since the previously used (umodem based) values do not seem to work. It is not completely clear what happened here - either the "not calling uplcom_set_line_state" due to luck prevented the non working state to be discovered or there are variants of the chip that need the old values. During testing none such chips have been found - if they ever apear we might have to make this function more intelligent, depending on the chip in use.
Diffstat (limited to 'bin/ls')
0 files changed, 0 insertions, 0 deletions