chao an
34d2cde8a8
net/l2/l3/l4: add support of iob offload
...
1. Add new config CONFIG_NET_LL_GUARDSIZE to isolation of l2 stack,
which will benefit l3(IP) layer for multi-MAC(l2) implementation,
especially in some NICs such as celluler net driver.
new configuration options: CONFIG_NET_LL_GUARDSIZE
CONFIG_NET_LL_GUARDSIZE will reserved l2 buffer header size of
network buffer to isolate the L2/L3 (MAC/IP) data on network layer,
which will be beneficial to L3 network layer protocol transparent
transmission and forwarding
------------------------------------------------------------
Layout of frist iob entry:
iob_data (aligned by CONFIG_IOB_ALIGNMENT)
|
| io_offset(CONFIG_NET_LL_GUARDSIZE)
| |
-------------------------------------------------
iob | Reserved | io_len |
-------------------------------------------------
-------------------------------------------------------------
Layout of different NICs implementation:
iob_data (aligned by CONFIG_IOB_ALIGNMENT)
|
| io_offset(CONFIG_NET_LL_GUARDSIZE)
| |
-------------------------------------------------
Ethernet | Reserved | ETH_HDRLEN | io_len |
---------------------------------|---------------
8021Q | Reserved | ETH_8021Q_HDRLEN | io_len |
---------------------------------|---------------
ipforward | Reserved | io_len |
-------------------------------------------------
--------------------------------------------------------------------
2. Support iob offload to l2 driver to avoid unnecessary memory copy
Support send/receive iob vectors directly between the NICs and l3/l4
stack to avoid unnecessary memory copies, especially on hardware that
supports Scatter/gather, which can greatly improve performance.
new interface to support iob offload:
------------------------------------------
| IOB version | original |
|----------------------------------------|
| devif_iob_poll() | devif_poll() |
| ... | ... |
------------------------------------------
--------------------------------------------------------------------
1> NIC hardware support Scatter/gather transfer
TX:
tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4)
/ \
/ \
devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...)
/ \ (copy:tcp)
/ \
devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll
|
dev->d_iob: |
--------------- ---------------
io_data iob1 | | | iob3 | | |
\ --------------- ---------------
--------------- | --------------- |
iob0 | | | | iob2 | | | |
--------------- | --------------- |
\ | / /
\ | / /
----------------------------------------------
NICs io vector | | | | | | | | | |
----------------------------------------------
RX:
[tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead)
|
|
[tcp|udp|icmp|...]_ipv[4|6]_in()/...
|
|
pkt/ipv[4/6]_input()/...
|
|
NICs io vector receive(iov_base to each iobs)
--------------------------------------------------------------------
2> CONFIG_IOB_BUFSIZE is greater than MTU:
TX:
"(CONFIG_IOB_BUFSIZE) > (MAX_NETDEV_PKTSIZE + CONFIG_NET_GUARDSIZE + CONFIG_NET_LL_GUARDSIZE)"
tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4)
/ \
/ \
devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...)
/ \ (copy:tcp)
/ \
devif_iob_poll("NIC"_txpoll) callback() // "NIC"_txpoll
|
"NIC"_send()
(dev->d_iob->io_data[CONFIG_NET_LL_GUARDSIZE - NET_LL_HDRLEN(dev)])
RX:
[tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead)
|
|
[tcp|udp|icmp|...]_ipv[4|6]_in()/...
|
|
pkt/ipv[4/6]_input()/...
|
|
NICs io vector receive(iov_base to io_data)
--------------------------------------------------------------------
3> Compatible with all old flat buffer NICs
TX:
tcp_poll()/udp_poll()/pkt_poll()/...(l3|l4)
/ \
/ \
devif_poll_[l3|l4]_connections() devif_iob_send() (nocopy:udp/icmp/...)
/ \ (copy:tcp)
/ \
devif_iob_poll(devif_poll_callback()) devif_poll_callback() /* new interface, gather iobs to flat buffer */
/ \
/ \
devif_poll("NIC"_txpoll) "NIC"_send()(dev->d_buf)
RX:
[tcp|udp|icmp|...]ipv[4|6]_data_handler()(iob_concat/append to readahead)
|
|
[tcp|udp|icmp|...]_ipv[4|6]_in()/...
|
|
netdev_input() /* new interface, Scatter/gather flat/iob buffer */
|
|
pkt/ipv[4|6]_input()/...
|
|
NICs io vector receive(Orignal flat buffer)
3. Iperf passthrough on NuttX simulator:
-------------------------------------------------
| Protocol | Server | Client | |
|-----------------------------------------------|
| TCP | 813 | 834 | Mbits/sec |
| TCP(Offload) | 1720 | 1100 | Mbits/sec |
| UDP | 22 | 757 | Mbits/sec |
| UDP(Offload) | 25 | 1250 | Mbits/sec |
-------------------------------------------------
Signed-off-by: chao an <anchao@xiaomi.com>
2022-12-03 11:47:04 +08:00
Xiang Xiao
c562263205
net: Move if_nametoindex and if_indextoname to libc
...
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
2021-12-19 10:08:57 -06:00
Alin Jerpelea
37d5c1b0d9
net: Author Gregory Nutt: update licenses to Apache
...
Gregory Nutt has submitted the SGA and we can migrate the licenses
to Apache.
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
2021-02-20 00:38:18 -08:00
Gregory Nutt
32e3e51678
net/netdev: Add signal notification for the case where the network goes down.
2018-09-09 10:39:25 -06:00
Gregory Nutt
7c1394d814
Squashed commit of the following:
...
net/netdev: Add implementation of if_nametoindex() and if indextoname().
net/pkt: Raw AF_PACKET sockets now depend on CONFIG_NETDEV_IFINDEX.
net/procfs: Tweak to handle traversal of interfaces if CONFIG_NETDEV_IFINDEX is not defined.
net/netdev.h: Update netdev_findbyaddr() to use the assigned device index.
Trivial typo fix
net/netdev: Add support for assigned an interface index to a device when it is regisgtered.
2018-06-25 12:08:10 -06:00
Gregory Nutt
7af976c00a
Squashed commit of the following:
...
Networking: Add implementation of logic for SIOCGIFCONF and SIOCGLIFCOF IOCTL commnds.
Networking: Add definitions and stuctures to support SIOCGIFCONF and SIOCGLIFCONF IOCTL commands.
2017-09-19 14:17:05 -06:00
Gregory Nutt
969d7c5814
Networking: Remove driver based backlog support. This affects the entire network, but is used by only one driver. The only supported with of supported RX backlog is via common read-ahead buffering.
2017-08-21 06:28:59 -06:00
Gregory Nutt
b4a0ac53a8
Networking: Improve naming and simplify some logic of previous commit.
2017-07-07 06:51:52 -06:00
Gregory Nutt
70c6b52132
ICMPv6: Fix an address size calculation that was bungled in a recent commit.
2017-07-06 19:37:01 -06:00
Gregory Nutt
eb8668b0f7
net/: Add procfs support for network device status
2015-11-27 16:39:14 -06:00
Gregory Nutt
33085cb309
Networking: The network device list was protected by a re-entrant semaphore. With the recent change to support network device callback, the network stack needs to access the network device list too. Some drivers, however, run the network stack from the interrupt level -- this is bad but a fact in the current state. Of course,those drivers are unable to take the semaphore and will assert.
...
The solution here is to eliminate the device devices semaphore altogether. This eliminates netdev_semtake() and netdev_semgive() and replaces them with net_lock() and net_unlock() which have larger scope as needed for this purpose.
2015-05-31 08:34:03 -06:00
Gregory Nutt
f10fdf7ad4
Networking: Add a test to see a device pointer is still valid
2015-05-29 11:01:03 -06:00
Gregory Nutt
735aa4289a
Network: Add logic to get a default network device (for the case where are multiple devices). For INADDR_ANY
2014-11-21 14:14:39 -06:00
Gregory Nutt
79e098b20e
Clean missing function headers
2014-08-16 10:24:06 -06:00
Gregory Nutt
64c8dba8c4
NET: Create net/sockets directory and hook into the build/configuration system
2014-06-28 16:41:55 -06:00
Gregory Nutt
238d9d9694
Fix conflict in ChangeLog
2014-06-27 10:00:28 -06:00