| Age | Commit message (Collapse) | Author |
|---|
|
|
|
rge can chain mbufs from multiple rx descriptors together to build
a jumbo packet. 1514 byte packets are still the most common, so
avoiding the waste of most of a 9k cluster if we can still makes
sense.
|
|
rge is a bit like intel nics where you can put small mbufs on the
rx ring and it will chain them together to make a jumbo packet.
except unlike intel you can use any sized mbufs for the rx descriptors.
this adds the handling of these packets to the rx path. we're still
putting 9k frames on the ring though.
|
|
this is very similar to the changes made for tx where we fill in
everything except the OWN flag on the rx descriptor, call bus_dmamap_sync
as a barrier for the hw, and then flip the ownership of the OWN
bit. this avoids the potential for the hw to see the own bit before
other things in the descriptor, such as the address and length of
the buffer.
while here, trim code that's not currently used. we don't currently
support rxing one packet by assembling buffers from multiple
descriptors, so stop fiddling with the queue mbuf head and tail
pointers. delete rge_discard_rxbuf() cos it looks like a leftover
from code which tries to reuse mbufs on the rx ring. we free mbufs
when there's an error and let the rxr stuff refill.
ok patrick@
no objections from kevlo@
|
|
Also, as used here, dup/dup2 will clear the close-on-exec flag, so
delete the superfluous fcntl(F_SETFD,0) calls
ok deraadt@
|
|
|
|
|
|
|
|
controller drivers.
|
|
(effectively) public dommain software is weird. put that info
into a better place.
|
|
|
|
|
|
OK tb@
|
|
`maxfiles' is atomically accessed integer which is lockless and
read-only accessed in file descriptors layer.
lim_startup() called during kernel bootstrap, no need to
atomic_load_int() within.
ok mpi
|
|
RFC 9632 introduced additional constraints and requirements for RPKI-based
Geofeed authenticators (at my request).
|
|
Speeds up tests by about 10% in the common case, hopefully more when
instrumented with something like valgrind.
|
|
With bounce buffers, we must pass a larger size to free()
ok bluhm@ hshoexer@
|
|
|
|
from bmercer@ and Bryan Vyhmeister
|
|
OK tb@
|
|
Also export the msgbuf queue len as a stats member so bgpctl does not need
to dig into the msgbuf structure inside struct peer.
This skips imsg related msgbuf since that will be handled by a imsgbuf
specific function.
OK tb@
|
|
There was an off-by-one bug when comparing the used pages for bounce
buffers with the available pages. As a result _bus_dmamap_load_buffer()
returned ENOMEM although there was one buffer left.
Also the _dm_nused field was updated and never reset in case of an
error. Use a local variable to count the used pages and update
global map->_dm_nused only if _bus_dmamap_load_buffer() was successful.
This fixes hanging network transmits if bounce buffers are enforced
for vio(4).
OK sf@ hshoexer@
|
|
constant (this makes no difference in practice because the length
is always the same); reported by martin AT nmkd.net
|
|
speeds overall test up by about 1%.
|
|
|
|
command-line flag, which simplifies some of the ssh command lines.
|
|
(otherwise the encrypted byte counts would not match) so this lets us
simplify the command lines.
|
|
`maxprocess' and `maxthread' are atomically accessed integers.
ok mpi
|
|
- ETHERIPCTL_ALLOW - atomically accessed integer;
- ETHERIPCTL_STATS - per-CPU counters
ok bluhm
|
|
the only IGMPCTL_STATS, PFSYNCCTL_STATS and RIPV6CTL_STATS per-CPU
counters.
sysctl_rdstruct() has "newp != NULL" check within and also returns
EPERM, no need for redundant check in igmp_sysctl().
ok bluhm
|
|
It is the only KERN_AUDIO_RECORD. `audio_record_enable' is atomically
accessed integer.
Reasonable from deraadt
|
|
is the pre-standardization name for the same thing, so remove it
as a duplicate. Speeds up test by a tiny amount.
|
|
parsing code, now has "-G" to dump its config and is slightly faster
to start up. This speeds up the test slightly (~5%) in the common case
but should help more during instrumented tests, eg under valgrind, where
startup costs are magnified.
|
|
If there is no device, don't write to any registers.
ok jan@
|
|
ok miod@
|
|
ok mglocker@
|
|
|
|
using mmap(3) with MAP_CONCEAL set. This prevents exposure of the
key material in coredumps, etc (this is in addition to other measures
we take in this area).
ok deraadt@
|
|
|
|
ok patrick@
thanks to kettenis@ for reminding me how computers work again.
|
|
In get_number(), reject values that are so large that they are
interpreted as negative numbers. In set_range(), step values smaller
than one or larger than the "stop" value are ignored. This prevents
bit_nset() from being called with out-of-range values.
Bug found by Dave G. of Supernetworks.
|
|
|
|
the MXT_T7_POWER_MODE_DEEP_SLEEP operation.
|
|
|
|
|
|
fixes panic when starting X on x40 (855GM gen 2)
problem with 945GM (gen 3) and G41 (gen 4) reported by semarie@
|
|
realistic option is to panic if we're doing smallish allocations and run
out of memory.
ok deraadt@
|
|
|
|
|
|
character of the terminator, so correct the size calculation to always
add one. GitHub issue 4082.
|
