code would hit the interface multiple times. Now, all media and media
options are gathered and processed at once, hitting the interface only
once (except to grab the currently selected media exactly once; there is
no way to avoid that).
display. The display now looks like this:
bishop:thorpej 35$ obj.alpha/ifconfig -m fxp0
fxp0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
media: Ethernet autoselect (100baseTX full-duplex)
status: active
supported Ethernet media:
media autoselect
media 100baseTX mediaopt full-duplex
media 100baseTX
media 10baseT mediaopt full-duplex
media 10baseT
media 100baseTX mediaopt loopback
inet 129.99.50.41 netmask 0xffffff00 broadcast 129.99.50.255
bishop:thorpej 36$
The intent here is to allow cut'n'pasting of the output to a command
line, or allow easier automatic extraction from scripts, etc.
XXX Still at issue is what to do about non-0 media instances.
data structure is used, but a much saner matching mechanism is used, one
which allows greater ease in adding new types.
Also, rewrite the way media words are displayed. The status display now
looks like this:
bishop:thorpej 137$ ifconfig -m fxp0
fxp0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
media: Ethernet autoselect (10baseT)
status: active
supported media:
Ethernet autoselect
Ethernet 100baseTX <full-duplex>
Ethernet 100baseTX
Ethernet 10baseT <full-duplex>
Ethernet 10baseT
Ethernet 100baseTX <loopback>
inet 129.99.50.41 netmask 0xffffff00 broadcast 129.99.50.255
bishop:thorpej 138$
This makes it more clear which media type (e.g. Ethernet) is currently
in-use, handy for devices such as the TI ThunderLAN which can also have
both Ethernet and Token Ring PHYs attached.
-fix overflow conditions (PR bin/5534, Zdenek Salvet <salvet@ics.muni.cz>)
(+ one more: can VAX SMD drives be >4G?)
-fix output of "*" at odd end cylinder number (not odd size)
-break some lines >80 cols
commit message from Bill Fenner:
Turn on TCP_NODELAY on the remote socket, to turn off sender silly window
syndrome avoidance. The combination of SWS avoidance and ack-every-other
causes low throughput if the block size divided by the MSS is odd (which
is true with the default block size and MSS).
Turning on TCP_NODELAY disables the Nagle algorithm and sender SWS avoidance.
The rdump request/response protocol can not invoke Nagle and cannot cause
SWS, so this has no negative effects.
Also, put back the code that sets the TOS to "throughput", which seems to
have been erroneously removed during the Lite-2 merge.
and the top, and then set the effective gid back to kmem around the call to
kvm_openfiles(). this reduces the time group kmem is available.
- for those above that also allow this, setgid(getgid()) after the call to
kvm_openfiles() to fully revoke priviledges.
- some KNF
- use err(3) over fprintf(3) in some places
- ensure hostname from gethostname() is nul-terminated in all cases
- minor KNF
- use MAXHOSTNAMELEN over various other values/defines
- be safe will buffers that hold hostnames
check at 3 second intervals if any are left.
This will help slow machines to cleanly shut down X servers (to make the
console visible), databases, or Usenet news servers.
checksums are calculated paying attention to the fact that the way
the checksum works a sectors filled with a single byte value will
always checksum correctly.
of the old boot area so that 'disklabel -r -w' won't clobber anything,
be sure to seek to the offset of the boot area (like readlabel() does).
Otherwise, if the machine has code which looks up machine-dependent label
bits, the offset read from will be incorrect, and disklabel will end up
secretly replacing the boot area with other data (which probably doesn't
look like a boot block). The only port this currently affects is
NetBSD/arm32, because it's the only one of the three ports that use
NUMBOOT > 0 that has MD label grovelling code.
mark