src/sys/dev/gem/if_gem.c, branch releng/12.2 FreeBSD source tree https://cgit-dev.freebsd.org/src/atom?h=releng%2F12.2 2018-05-18T20:13:34Z ifnet: Replace if_addr_lock rwlock with epoch + mutex 2018-05-18T20:13:34Z Matt Macy mmacy@FreeBSD.org 2018-05-18T20:13:34Z urn:sha1:d7c5a620e2b99e914f1770abde956cf0d0a970b7 Run on LLNW canaries and tested by pho@ gallatin: Using a 14-core, 28-HTT single socket E5-2697 v3 with a 40GbE MLX5 based ConnectX 4-LX NIC, I see an almost 12% improvement in received packet rate, and a larger improvement in bytes delivered all the way to userspace. When the host receiving 64 streams of netperf -H $DUT -t UDP_STREAM -- -m 1, I see, using nstat -I mce0 1 before the patch: InMpps OMpps InGbs OGbs err TCP Est %CPU syscalls csw irq GBfree 4.98 0.00 4.42 0.00 4235592 33 83.80 4720653 2149771 1235 247.32 4.73 0.00 4.20 0.00 4025260 33 82.99 4724900 2139833 1204 247.32 4.72 0.00 4.20 0.00 4035252 33 82.14 4719162 2132023 1264 247.32 4.71 0.00 4.21 0.00 4073206 33 83.68 4744973 2123317 1347 247.32 4.72 0.00 4.21 0.00 4061118 33 80.82 4713615 2188091 1490 247.32 4.72 0.00 4.21 0.00 4051675 33 85.29 4727399 2109011 1205 247.32 4.73 0.00 4.21 0.00 4039056 33 84.65 4724735 2102603 1053 247.32 After the patch InMpps OMpps InGbs OGbs err TCP Est %CPU syscalls csw irq GBfree 5.43 0.00 4.20 0.00 3313143 33 84.96 5434214 1900162 2656 245.51 5.43 0.00 4.20 0.00 3308527 33 85.24 5439695 1809382 2521 245.51 5.42 0.00 4.19 0.00 3316778 33 87.54 5416028 1805835 2256 245.51 5.42 0.00 4.19 0.00 3317673 33 90.44 5426044 1763056 2332 245.51 5.42 0.00 4.19 0.00 3314839 33 88.11 5435732 1792218 2499 245.52 5.44 0.00 4.19 0.00 3293228 33 91.84 5426301 1668597 2121 245.52 Similarly, netperf reports 230Mb/s before the patch, and 270Mb/s after the patch Reviewed by: gallatin Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D15366 sys/dev: further adoption of SPDX licensing ID tags. 2017-11-27T14:52:40Z Pedro F. Giffuni pfg@FreeBSD.org 2017-11-27T14:52:40Z urn:sha1:718cf2ccb9956613756ab15d7a0e28f2c8e91cab Mainly focus on files that use BSD 2-Clause license, however the tool I was using misidentified many licenses so this was mostly a manual - error prone - task. The Software Package Data Exchange (SPDX) group provides a specification to make it easier for automated tools to detect and summarize well known opensource licenses. We are gradually adopting the specification, noting that the tags are considered only advisory and do not, in any way, superceed or replace the license texts. Mechanically convert to if_inc_counter(). 2014-09-18T20:21:46Z Gleb Smirnoff glebius@FreeBSD.org 2014-09-18T20:21:46Z urn:sha1:8da56a6fec0849725d99dded113d344ea994282e Use define from if_var.h to access a field inside struct if_data, 2014-08-30T19:55:54Z Gleb Smirnoff glebius@FreeBSD.org 2014-08-30T19:55:54Z urn:sha1:1bffa9511fe94125820858b0ead796db54ac019d that resides in struct ifnet. Sponsored by: Nginx, Inc. The r48589 promised to remove implicit inclusion of if_var.h soon. Prepare 2013-10-26T17:58:36Z Gleb Smirnoff glebius@FreeBSD.org 2013-10-26T17:58:36Z urn:sha1:76039bc84fae9915788b54ff28fe0cc4876952d2 to this event, adding if_var.h to files that do need it. Also, include all includes that now are included due to implicit pollution via if_var.h Sponsored by: Netflix Sponsored by: Nginx, Inc. Mechanically substitute flags from historic mbuf allocator with 2012-12-04T09:32:43Z Gleb Smirnoff glebius@FreeBSD.org 2012-12-04T09:32:43Z urn:sha1:c6499eccad497913a5025fbde8ae76da70e08043 malloc(9) flags in sys/dev. Revert r224157, re-enabling r222135. The underlying problem keeping the 2011-08-19T19:13:31Z Marius Strobl marius@FreeBSD.org 2011-08-19T19:13:31Z urn:sha1:5ca7dfffdf925aec3238b53ca7034f7be4d37388 latter from working as expected was fixed in r225014. MFC after: 3 days Approved by: re (kib) Revert r222135 by allowing controller reinitialization. Due to 2011-07-17T21:54:51Z Pyun YongHyeon yongari@FreeBSD.org 2011-07-17T21:54:51Z urn:sha1:23b179f5d3b14f6a6164e0a2a988bbe558a1eb2e unknown reason Apple UniNorth2 gem(4) device required manual interface down/up operation after r222135. Even though this is not correct thing and I don't like to revert it but it would be better than breaking gem(4) on PPC. This should be revisited. PR: kern/157405 Correct r223648; as gem_init_locked() was calling gem_setladrf(), which 2011-07-12T13:06:44Z Marius Strobl marius@FreeBSD.org 2011-07-12T13:06:44Z urn:sha1:5ed0b954175cd816fe4cb10024e472a38701d343 sets GEM_MAC_RX_CONFIG based on sc_mac_rxcfg which in turn is initialized to zero, before reading the supposedly default configuration we were effectively not basing sc_mac_rxcfg and thus GEM_MAC_RX_CONFIG on the default configuration. Solve this by calling gem_setladrf() after reading in the default configuration of GEM_MAC_RX_CONFIG. This also avoids the need to distinguish whether gem_setladrf() should enable the RX MAC again and should be slightly more correct as we're now doing all of the RX MAC configuration in the intended step. According to the OpenSolaris source the PCI latency and the cache line size 2011-07-12T08:20:15Z Marius Strobl marius@FreeBSD.org 2011-07-12T08:20:15Z urn:sha1:9f9cc2edd8f36f11c8d2fe5657e890cbbc44a7df should be set for Sun ERI. Tested by: yongari