Navdeep Parhar 298d969c53 cxgbe(4): netmap support for Terminator 5 (T5) based 10G/40G cards. ... Netmap gets its own hardware-assisted virtual interface and won't take over or disrupt the "normal" interface in any way. You can use both simultaneously. For kernels with DEV_NETMAP, cxgbe(4) carves out an ncxl<N> interface (note the 'n' prefix) in the hardware to accompany each cxl<N> interface. These two ifnet's per port share the same wire but really are separate interfaces in the hardware and software. Each gets its own L2 MAC addresses (unicast and multicast), MTU, checksum caps, etc. You should run netmap on the 'n' interfaces only, that's what they are for. With this, pkt-gen is able to transmit > 45Mpps out of a single 40G port of a T580 card. 2 port tx is at ~56Mpps total (28M + 28M) as of now. Single port receive is at 33Mpps but this is very much a work in progress. I expect it to be closer to 40Mpps once done. In any case the current effort can already saturate multiple 10G ports of a T5 card at the smallest legal packet size. T4 gear is totally untested. trantor:~# ./pkt-gen -i ncxl0 -f tx -D 00:07:43🆎cd:ef 881.952141 main [1621] interface is ncxl0 881.952250 extract_ip_range [275] range is 10.0.0.1:0 to 10.0.0.1:0 881.952253 extract_ip_range [275] range is 10.1.0.1:0 to 10.1.0.1:0 881.962540 main [1804] mapped 334980KB at 0x801dff000 Sending on netmap:ncxl0: 4 queues, 1 threads and 1 cpus. 10.0.0.1 -> 10.1.0.1 (00:00:00:00:00:00 -> 00:07:43🆎cd:ef) 881.962562 main [1882] Sending 512 packets every 0.000000000 s 881.962563 main [1884] Wait 2 secs for phy reset 884.088516 main [1886] Ready... 884.088535 nm_open [457] overriding ifname ncxl0 ringid 0x0 flags 0x1 884.088607 sender_body [996] start 884.093246 sender_body [1064] drop copy 885.090435 main_thread [1418] 45206353 pps (45289533 pkts in 1001840 usec) 886.091600 main_thread [1418] 45322792 pps (45375593 pkts in 1001165 usec) 887.092435 main_thread [1418] 45313992 pps (45351784 pkts in 1000834 usec) 888.094434 main_thread [1418] 45315765 pps (45406397 pkts in 1002000 usec) 889.095434 main_thread [1418] 45333218 pps (45378551 pkts in 1001000 usec) 890.097434 main_thread [1418] 45315247 pps (45405877 pkts in 1002000 usec) 891.099434 main_thread [1418] 45326515 pps (45417168 pkts in 1002000 usec) 892.101434 main_thread [1418] 45333039 pps (45423705 pkts in 1002000 usec) 893.103434 main_thread [1418] 45324105 pps (45414708 pkts in 1001999 usec) 894.105434 main_thread [1418] 45318042 pps (45408723 pkts in 1002001 usec) 895.106434 main_thread [1418] 45332430 pps (45377762 pkts in 1001000 usec) 896.107434 main_thread [1418] 45338072 pps (45383410 pkts in 1001000 usec) ... Relnotes: Yes Sponsored by: Chelsio Communications.		2014-05-27 18:18:41 +00:00
..
amd64	Add segment protection and limits violation checks in vie_calculate_gla()	2014-05-27 04:26:22 +00:00
arm	Rework the Ti GPIO driver to work on multiple SoCs. At the moment it could	2014-05-26 18:02:36 +00:00
boot	"%p" formatting already includes "0x" prefix in printout.	2014-05-21 18:02:19 +00:00
bsm	IFp4 @1192291:	2014-04-07 20:44:00 +00:00
cam	Provide better descriptions for 'struct ctl_scsiio' fields; based mostly	2014-05-04 15:35:04 +00:00
cddl	Improve sysctl descriptions for new ZFS sysctls:	2014-05-22 05:30:38 +00:00
compat	In freebsd32_sendmsg(), replace the call to sockargs() followed by a	2014-04-05 18:56:01 +00:00
conf	cxgbe(4): netmap support for Terminator 5 (T5) based 10G/40G cards.	2014-05-27 18:18:41 +00:00
contrib	Move mutex creation from ipf_log_soft_init() to ipf_log_soft_create()	2014-05-24 06:05:21 +00:00
crypto	NO_MAN= has been deprecated in favor of MAN= for some time, go ahead	2014-04-13 05:21:56 +00:00
ddb	ddb: Minor style cleanups.	2014-03-31 16:37:41 +00:00
dev	cxgbe(4): netmap support for Terminator 5 (T5) based 10G/40G cards.	2014-05-27 18:18:41 +00:00
fs	Initial import of character device in userspace support for FreeBSD.	2014-05-23 08:46:28 +00:00
gdb
geom	Add a topology trace to the g_spoil_event.	2014-05-19 16:08:15 +00:00
gnu	Symbolic bindings for the dts files...	2014-02-28 16:51:33 +00:00
i386	imagact_binmisc builds for all supported architectures, so enable it for all.	2014-05-22 05:04:40 +00:00
ia64	Bring in the mpr(4) driver for LSI's MPT3 12Gb SAS controllers.	2014-05-02 20:25:09 +00:00
isa	Remove vestiges of knowing the ISA bus, which we gave up on around 20	2014-03-19 21:03:04 +00:00
kern	Add a new taskqueue setup method that takes a cpuid to pin the	2014-05-24 20:37:15 +00:00
kgssapi
libkern	Fix typo	2013-11-29 20:14:26 +00:00
mips	Remove duplicate:	2014-05-21 21:30:00 +00:00
modules	cxgbe(4): netmap support for Terminator 5 (T5) based 10G/40G cards.	2014-05-27 18:18:41 +00:00
net	Bump bhyve allocation up to 20 bits to avoid	2014-05-20 02:59:13 +00:00
net80211	Use logical rather than bitwise OR in if() expression.	2014-04-26 23:22:49 +00:00
netgraph	Make ng_mppc to not disable the node in case of multiple packet loss.	2014-05-22 07:27:04 +00:00
netinet	The users of RSS shouldn't be directly concerned about hash -> CPU ID	2014-05-27 08:06:20 +00:00
netinet6	Cancel DAD for an ifa when the ifp has ND6_IFF_IFDISABLED as early as	2014-05-16 15:53:31 +00:00
netipsec	Only do a ports check if this is a NAT-T SA. Otherwise other	2014-05-24 09:29:23 +00:00
netnatm	Provide includes that are needed in these files, and before were read	2013-10-26 18:18:50 +00:00
netpfil	Since ipfw nat configures all options in one step, we should set all bits	2014-05-18 14:25:19 +00:00
netsmb	Remove redundant unlock.	2014-04-17 12:22:08 +00:00
nfs	- Remove rt_metrics_lite and simply put its members into rtentry.	2014-03-05 01:17:47 +00:00
nfsclient	For software builds, the NFS client does many small	2013-12-07 23:05:59 +00:00
nfsserver	The PR reported that the old NFS server did not set uio_td == NULL	2014-04-24 20:47:58 +00:00
nlm	Correct a typo in nlm_find_host_by_addr(): the intention of the	2014-03-11 00:25:55 +00:00
ofed	Use src.opts.mk in preference to bsd.own.mk except where we need stuff	2014-05-06 04:22:01 +00:00
opencrypto	allow the kern.cryptodevallowsoft sysctl to enable symetric/hashes too...	2014-03-11 01:45:46 +00:00
pc98	lindev(4): finish the partial commit in r265212	2014-05-02 07:14:22 +00:00
pci	Increase the number of TX DMA segments from 32 to 35. It turned	2014-03-31 01:54:59 +00:00
powerpc	oea64 uses 4k pages, too.	2014-05-15 15:17:44 +00:00
rpc	Properly free resources in case of error.	2014-05-02 20:45:55 +00:00
security	audit: plug FILEDESC_LOCK leak in audit_canon_path.	2014-03-21 01:30:33 +00:00
sparc64	Bring in the mpr(4) driver for LSI's MPT3 12Gb SAS controllers.	2014-05-02 20:25:09 +00:00
sys	Garbage-collect a couple of unused identifiers.	2014-05-27 02:00:43 +00:00
teken	Fix typo.	2014-02-06 13:28:06 +00:00
tools	Fix logic error. <blush>	2014-04-30 20:52:38 +00:00
ufs	Due to reasons unknown at this time, the system can be forced to write	2014-05-06 20:40:16 +00:00
vm	There is no reason to perform the pmap_remove() on the kernel pmap while	2014-05-23 16:22:36 +00:00
x86	Now that there are separate back-end implementations of busdma, the bounce	2014-05-27 14:18:57 +00:00
xdr
xen	Make this compile with gcc.	2014-04-05 22:43:18 +00:00
Makefile	Add a new target cscope-hook.	2014-05-15 03:47:52 +00:00