| Age | Commit message (Collapse) | Author |
|---|
|
length instead of the packet instruction length to free the old
instruction.
ok reyk@
|
|
|
|
|
|
This factors out the code configuring General Purpose Interrupts into
a separate function and provides LASI (Link Alarm Status Interrupt)
handler used by controllers in the X550 family lacking integrated PHY.
|
|
no binary change.
|
|
- separate functions for delay value calculation and figuring out
whether or not we're doing SFP.
- MAC type detection is now done by ixgbe_set_mac_type;
- call {enable,disable}_tx_laser conditionally;
- unused TSO code bites the dust;
- default to "IFM_ETHER | IFM_AUTO" when we can't select any other
media type.
|
|
|
|
|
|
|
|
|
|
indirectly documented in the specification. Add the states to the
header as it is important for all implementations. Both switch(4) and
switchd(8) have to be updated to follow the fsm instead of answering
any requests at any time.
OK rzalamena@
|
|
- Use the right amount of zeros to match the header bit length;
- Use hex instead of bit shifting;
|
|
|
|
the Yeelong Lemote mips-based netbook. Kernel bits only for now, needs
polishing; most of this work done during h2k9 last month, although the
porting effort started earlier this year.
|
|
each counter is identified by an enum value which correspond to the
original members of the udpstat struct.
udpstat_inc(udps_foo) replaces udpstat.udps_foo++ for the actual
updates. udpstat_inc is a thin wrapper around counters_inc.
counters are still returned to userland via the udpstat struct for
now.
ok mpi@ mikeb@ deraadt@
|
|
|
|
|
|
|
|
|
|
|
|
Suggested by reyk@ as a quick fix to unbreak ports/net/wireshark.
|
|
The work on the platform was made possible by a hardware donation
from Loongson Technology.
|
|
and put the pieces together by platform glue.
Feedback from miod@
|
|
ok mpi@
|
|
Since soft-interrupt for network processing are no more, the
KERNEL_LOCK() is what protect two concurrent threads from
accessing data structures.
ok goda@
|
|
after fragment headers. Add an extra check that the hop-by-hop
header is always the first extension header after the IPv6 header.
Found by Antonios Atlasis; OK sthen@ mpi@
|
|
A custom method to enable RX DMA is introduced because of the
upcoming API change in the common code. Otherwise it's just
a copy of the ixgbe_enable_rx_dma_generic.
|
|
|
|
|
|
Obtained from the Intel ixgbe driver in FreeBSD.
|
|
|
|
|
|
There's not much point in adding an additional condition to the
ifq_restart call so just get rid of those bits. Also while here,
remove a debugging message and make sure to cleanup the whole RNDIS
packet before sending it out to the hypervisor.
|
|
|
|
it just schedule a task.
|
|
|
|
While here properly account for used reference in bpfwrite().
ok bluhm@
|
|
Prodded by and ok bluhm@
|
|
eg. tcpdump -y openflow -i switch0
Includes a minor bump for libpcap.
Feedback and OK rzalamena@
|
|
IPv4 dynamic route inherits the priority. Only clone from a valid
IPv6 route. Do not use splsoftnet() in IPv6. Some stylistic changes
to make the functions similar.
OK mpi@
|
|
From Rivo Nurges; OK claudio@ mpi@ phessler@
|
|
This will help trading the KERNEL_LOCK for a mutex.
ok bluhm@
|
|
While here keep local definitions local.
ok bluhm@
|
|
ok bluhm@
|
|
|
|
from Bryan Vyhmeister, ok sthen@
|
|
as source and destination never overlap in this case. From
Michael W. Bombardieri <mb at ii.net>.
|
|
the important change is to no longer account for packets and
descriptors separately, instead we just deal with available descriptors
on the ring.
this allows us to guarantee that oactive will only be set after
re_start has queued at least one packet, which in turn guarantees
that the chip will interrupt in the future and run re_txeof to clear
oactive.
without this change it was possible for a call of re_start to almost
fill the descriptor ring, followed by a second call that could fail
to post a packet because it had run out of descriptors and would
then set oactive. if re_txeof ran concurrently it could complete
all the previously posted packets without noticing that oactive had
been set and therefore needed to be cleared. in that situation
oactive would be set, but nothing would be set up to clear it
in the immediate future.
this fixes a lockup reported by haesbaert.
|
|
previously there was a limit of 64 packets on the tx ring. soon we
will use as many descriptors as we want to fill the ring.
|
|
|
