| Age | Commit message (Collapse) | Author |
|---|
|
Instead, silently ignore the fact and instead let the underlying
ssh (or $RSH) command handle it.
ok millert@
|
|
Avoids a NULL dereference encountered with a device I have.
ok mpi@
|
|
ok miod@ millert@ krw@ guenther@
|
|
PIC sometimes loses isp^Wqlw interrupts.
|
|
way by mpi@. Bump version.
ok mpi@
|
|
we use this term.
ok jmc@
|
|
termios.h so we don't need the code for when it is not defined.
ok reyk@
|
|
> obfuscate the monotonic clock values we put on the wire by offsetting
> them with a random value.
|
|
> bump the size of the time types on the wire to 64bit, and use nsec
> instead of usec so its easier to translate to/from timespecs instead
> of timevals.
this raises the size of the default ping payload from 16 to 24bytes,
but claudio@ thinks we'll be fine with this.
|
|
them with a random value.
picking between ^ and +- as the operator for the obfuscation has
turned out to be a great bikeshed. im picking + and - cos it prevents
observers seeing high bits roll over which can indicate the magnitude
of the current clock value.
discussed with many.
|
|
|
|
ssh v.1 keys when compiled without SSH1 support
RSA/DSA/ECDSA keys when compiled without OpenSSL
based on patch by Mike Frysinger; bz#2369
|
|
and scanning conditional on the 5GHz support bit in the nvm.
Problem reported and fix tested by Mattieu Baptiste.
ok stsp@ kettenis@
|
|
end-of-file, returning 0, in order not to print an unrelated
strerror(errno) in the latter case
ok millert@
|
|
|
|
by Kaspars Bankovskis. ok jmc@
|
|
|
|
`ranges' property, i.e. sun4c, so that child devices can attach correctly.
This makes be(4) work on sun4c.
|
|
functions, and ocsp and s_time need networking enabled too, this just moves
BIO_sock_init() up into main() as a catch-all for all of the commands.
Of course, it is a no-op on any other platform.
ok @guenther
|
|
|
|
|
|
|
|
Adjust the conditions to correct that.
ok millert@
|
|
and NPROCESSORS_ONLN. OK guenther@
|
|
ok millert@ jung@
|
|
thanks sobrado for verifying (and okaying this)
|
|
|
|
section;
|
|
|
|
in the one place where it is mentioned, use an Xr rather than Fn;
|
|
|
|
and remove the update_firmware() function all together.
OK halex@ krw@
|
|
support to the X87, SSE and AVX state.
This gives us (almost) full AVX support. The AVX state isn't saved by
signal handlers yet, and ptrace(2) support is still missing.
ok guenther@
|
|
reachable with the default slot aperture (such as hme(4) devices, which have
their registers at offset 0x08c00000 from the slot base), force the attachment
of the device to fail.
While hme would fail early due to being unable to talk on the mdio bus, hme+esp
boards would have the esp part of the board get nuts and would spin during
autoconf. Of course, you are not supposed to use such boards on sun4c, because
(as mentioned above) they can't run without an iommu.
|
|
|
|
In the name of simplicity, remove the interface FSM that was inherited
from ospfd. In ldpd interfaces are just up or down, so keeping a
FSM for that is an overkill. Now instead of calling if_fsm(), just
call if_update() whenever a relevant event occurs (status change,
address addition/removal).
Additional notes:
1 - s/if_act_/if_/
2 - Remove the IMSG_IFUP and IMSG_IFDOWN events. Now whenever an
interface changes its state a IMSG_IFSTATUS event will be generated
with the new status.
kroute.c ldpd.h ldpe.c ldpe.h CVS:
----------------------------------------------------------------------
|
|
after an address removal in the system.
|
|
to the bottom of the file to make it easier to see what
code is actually executed.
OK krw@ halex@
|
|
for whatever reason.
If one interface is disabled, the holdtimer of the attached adjacencies
will eventually timeout after a few seconds. But there's no need
to wait when we know that the interface is disabled. In these cases,
remove the attached adjacencies to speedup the convergence process.
|
|
Although RFC 5036 is not explicit about this, LDP should not assign
labels for BGP routes. Doing that would be very resource consuming
in some scenarios and unnecessary. The goal is generally only to
establish LSPs among all PEs in the AS since LDP is not used as an
end in itself but as a means to implement advanced solutions like
MPLS L2/L3 VPNs. Some implementations (e.g. JunOS) go further and
only assign labels for /32 loopback routes advertised in the IGP.
If Inter-AS LSPs are necessary, BGP itself should be used for
distributing IPv4 labeled routes (e.g. option C. of section 10 in
RFC 4364).
|
|
Our reset function did not reset the ring and LDC map and our stop function
did not reset the LDC channel state and the some of the ring state. Also
make sure we clear IFF_OACTIVE whenever we re-establish a connection on the
LDC channel. Finally, initialize the link state to LINK_STATE_DOWN such
that interfaces that have not been configured yet, don't show up as active.
This should improve the reliability of re-establishing network connections
between domains after some sort of hickup considerably.
|
|
This makes sure the other side correctly sees the channel being reset.
|
|
|
|
fall back to /usr/share/man:/usr/X11R6/man:/usr/local/man
|
|
OK krw@
|
|
repeated use of tls_connect. ok jsing
|
|
|
|
ok brynet@, miod@
|
|
The actual control flow is intentional while the indenting is incorrect.
This is intended to be a cosmetic change.
Verified that each of these was part of a KNF commit that wasn't intending
to change behavior. Also, double checked against the history of changes in
OpenSSL and BoringSSL.
Addresses Coverity CIDs: 78842, 78859, 78863.
ok tedu@
|
|
memory instead of sysconf(_SC_PHYS_PAGES) (which also counts pages
wired by the kernel). Don't try to use a memory buffer larger than
the datasize hard resource limit. Adapted from a diff by Andre Smagin.
|
