en Copyright 2015 Java http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#efcc0423d80e4048e11e997abc46bd0343537e1b em: remove M_HASHTYPE when RSS is not enabledSummary:Since "73fe85e486d2 tcp: store flowid info in syncache", inp_flowid can be setif the incoming packet is not M_HASHTYPE_NONE. But this can introduce dummyand duplicated flowid when a virtual interface set M_HASHTYPE_OPAQUE. Thischange will let the upper layer know how to deal with software hash, withbenefits like inp_flowid can be set correctly and m_pkthdr.flowid can be setcorrectly in output path.This fix is similar to "20285cad7a55"Reviewed by: kbowlingDifferential Revision: https://reviews.freebsd.org/D55137 List of files: /freebsd/sys/dev/e1000/em_txrx.c Thu, 05 Feb 2026 22:12:15 +0100 Cheng Cui <cc@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#e07b4a2e9bc4f6b2e129f3a101dfce2027694bd3 e1000: use newly exposed RSS hash key API rather than ad-hoc hashingDifferential Revision: https://reviews.freebsd.org/D53097Reviewed by: kbowlingSponsored by: Netflix List of files: /freebsd/sys/dev/e1000/em_txrx.c Sat, 22 Nov 2025 15:29:33 +0100 Andrew Gallatin <gallatin@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#4b29599fbbe33b75b7b58cfc5deea7a881e9a10e e1000: Fix vlan PCP/DEI on lem(4)The vlan PCP and CFI/DEI were discarded when receiving vlan taggedpackets on lem(4) interfaces with vlanhwtag.According to the 82540 SDM[1] (pg. 24), vlan tag is in the standardformat, so there's no reason to discard PCP/DEI.[1]: http://iommu.com/datasheets/ethernet/controllers-nics/intel/e1000/pci-pci-x-family-gbe-controllers-software-dev-manual.pdfMFC after: 3 daysSponsored by: Stormshield (author)Differential Revision: https://reviews.freebsd.org/D48987 List of files: /freebsd/sys/dev/e1000/em_txrx.c Thu, 13 Feb 2025 19:08:42 +0100 Aurelien Cazuc <aurelien.cazuc.external@stormshield.eu> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#7390daf87c53073da5e74b2d73ac7320c1a44475 e1000: Style txrxFix up indentation and reflow long lines.MFC after: 3 daysSponsored by: BBOX.io List of files: /freebsd/sys/dev/e1000/em_txrx.c Sun, 24 Nov 2024 08:39:43 +0100 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#9dc452b983f042aa9d2a2562517d44c1928ff42a e1000: txrx function prototype cleanupDrop variable names of function prototypes since the file is mixed inlisting them or not and they fall out of sync.MFC after: 1 weekSponsored by: BBOX.io List of files: /freebsd/sys/dev/e1000/em_txrx.c Mon, 14 Oct 2024 18:07:29 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#3e501ef896671cb190e8c40c6258b8f27d136f07 e1000: Re-add AIMWe originally left this out because iflib modulates interrupts andaccomplishes some level of batching versus the custom queues in theolder driver. Upon more detailed study of the Linux driver which has anewer implementation, it finally became clear to me this is actually aholdoff timer and not an interrupt limit as it is conventionally(statically) programmed and displayed as an interrupt rate. The datasheets also make this somewhat clear.Thus, AIM accomplishes two beneficial things for a wide variety ofworkloads[1]:1. At low throughput/packet rates, it will significantly lower latency(by counter-intuitively "increasing" the interrupt rate.. betterthought of as decreasing the holdoff timer because you will modulatedown before coming anywhere near these interrupt rates).2. At bulk data rates, it is tuned to achieve a lower interrupt rate(by increasing the holdoff timer) than the current static 8000/s. Thisdecreases processing overhead and yields more headroom for other worksuch as packet filters or userland.For a single NIC this might be worth a few sys% on common CPUs, but maybe meaningful when multiplied such as if_lagg, if_bridge and forwardingsetups.The AIM algorithm was re-introduced from the older igb or out of treedriver, and then modernized with permission to use Intel code from otherdrivers.I have retroactively added it to lem(4) and em(4) where the same conceptapplies, albeit to a single ITR register.[1]: http://iommu.com/datasheets/ethernet/controllers-nics/intel/e1000/gbe-controllers-interrupt-moderation-appl-note.pdfTested by: cc (https://wiki.freebsd.org/chengcui/testD46768)MFC after: 1 weekRelnotes: yesSponsored by: Rubicon Communications, LLC ("Netgate")Sponsored by: BBOX.ioDifferential Revision: https://reviews.freebsd.org/D46768 List of files: /freebsd/sys/dev/e1000/em_txrx.c Sun, 22 Sep 2024 11:26:05 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#71625ec9ad2a9bc8c09784fbd23b759830e0ee5f sys: Remove $FreeBSD$: one-line .c comment patternRemove /^/[*/]\s*\$FreeBSD\$.*\n/ List of files: /freebsd/sys/dev/e1000/em_txrx.c Wed, 16 Aug 2023 19:54:24 +0200 Warner Losh <imp@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#eac761e967b34066aa7183474249df87d79a2f32 e1000: Fix off by one ipcseThis has been off by one in the FreeBSD drivers as far back as I'velooked. Emperically HW and SW emulations I have available don't seem tomind. Noticed while debugging other issues.MFC after: 3 days List of files: /freebsd/sys/dev/e1000/em_txrx.c Tue, 15 Aug 2023 23:37:43 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#e1353dcb63bbdb33be6a907ed6236c120518129e e1000: Fix lem(4)/em(4) TSO6* Fix TSO6 by specializing IP checksum insertion and following Intel SDM values for IPv6.* Remove unnecessary 82544 IP-bit handling* Remove TSO6 from lem(4) capabilititiesReviewed by: erj (earlier version)MFC after: 1 weekDifferential Revision: https://reviews.freebsd.org/D41170 List of files: /freebsd/sys/dev/e1000/em_txrx.c Mon, 31 Jul 2023 17:16:24 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#cbcab907f8ad1a4ac38dbc574c747ac2901faa54 e1000: Corrections for lem(4)/em(4) txcsum offloadExplicitly set ipcss/ipcse/ipcso for IPv6 per intel SDM as indicated ininline comments.Fix and consolidate 82543/82547 hwcsum exemption.While here rearrange and expand some commentary. List of files: /freebsd/sys/dev/e1000/em_txrx.c Fri, 28 Jul 2023 00:50:32 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#92fd2f39e5b1fd588fdc49a07d626273eb8f7539 e1000: add missing parens in csum setupReported by: rscheffFixes: 4f9a44a215f8 e1000: Fix/enable IPv6 transmit checksum offloadMFC after: 2 weeks List of files: /freebsd/sys/dev/e1000/em_txrx.c Sun, 23 Jul 2023 02:03:26 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#4f9a44a215f873c6842a693f2e81e0abbe1531cf e1000: Fix/enable IPv6 transmit checksum offloadFixes and enables txcsum6 offload for lem(4) and em(4).MFC after: 2 weeks List of files: /freebsd/sys/dev/e1000/em_txrx.c Sun, 23 Jul 2023 00:58:34 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#918c25677d882a901696672bd4d39b62faa56dfa e1000: lem(4)/em(4) ifcaps, TSO and hwcsum fixes* em(4) obey administrative ifcaps for using hwcsum offload* em(4) obey administrative ifcaps for hw vlan receive tagging* em(4) add additional TSO6 ifcap, but disabled by default as is TSO4* lem(4) obey administrative ifcaps for using hwcsum offload* lem(4) add support for hw vlan receive tagging* lem(4) Add ifcaps for TSO offload experimentation, but disabled by default due to errata and possibly missing txrx code.* lem(4) disable HWCSUM ifcaps by default on 82547 due to errata around full duplex links. It may still be administratively enabled.Reviewed by: markj (previous version)MFC after: 2 weeksDifferential Revision: https://reviews.freebsd.org/D30072 List of files: /freebsd/sys/dev/e1000/em_txrx.c Sat, 22 Jul 2023 20:33:27 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#5d3c982549278c72d59debae7227366453d3c4a8 Revert "e1000: lem(4)/em(4) ifcaps, TSO and hwcsum fixes"Seems to cause a panic when booting under VitrualBox.Reported by: yasuThis reverts commit 95f7b36e8fac45092b9a4eea5e32732e979989f0. List of files: /freebsd/sys/dev/e1000/em_txrx.c Sat, 22 Jul 2023 06:42:23 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#95f7b36e8fac45092b9a4eea5e32732e979989f0 e1000: lem(4)/em(4) ifcaps, TSO and hwcsum fixes* em(4) obey administrative ifcaps for using hwcsum offload* em(4) obey administrative ifcaps for hw vlan receive tagging* em(4) add additional TSO6 ifcap, but disabled by default as is TSO4* lem(4) obey administrative ifcaps for using hwcsum offload* lem(4) add support for hw vlan receive tagging* lem(4) Add ifcaps for TSO offload experimentation, but disabled by default due to errata and possibly missing txrx code.* lem(4) disable HWCSUM ifcaps by default on 82547 due to errata around full duplex links. It may still be administratively enabled.Reviewed by: markj (previous version)MFC after: 2 weeksDifferential Revision: https://reviews.freebsd.org/D30072 List of files: /freebsd/sys/dev/e1000/em_txrx.c Fri, 21 Jul 2023 05:30:00 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#201c4b7c29da44500ccb9e47b854dc1207df8b0a e1000: Some fixes for em(4) TSO setupAlways set TXD_CMD_IP for 82544Otherwise set TXD_CMD_IP for IPv4, not IPv6Reviewed by: markj (previous version)MFC after: 2 weeksDifferential Revision: https://reviews.freebsd.org/D30072 List of files: /freebsd/sys/dev/e1000/em_txrx.c Fri, 21 Jul 2023 03:51:02 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#5253d74e4650300815af351cc510a1c392a16584 e1000: Format em_txrxNo functional changes intended.Reviewed by: markj (prior version)MFC after: 1 weekDifferential Revision: https://reviews.freebsd.org/D30072 List of files: /freebsd/sys/dev/e1000/em_txrx.c Fri, 09 Jun 2023 04:06:04 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#034f38cdf864e6794fd63b67bf16c59627576699 e1000: Add SPDX tag to em_txrxReviewed by: markjMFC after: 1 weekDifferential Revision: https://reviews.freebsd.org/D30072 List of files: /freebsd/sys/dev/e1000/em_txrx.c Fri, 09 Jun 2023 03:42:20 +0200 Kevin Bowling <kbowling@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#f7926a6d0c1029c8da265769e7c57b4065faa2df net: iflib: fix vlan processing in the driversThe logic that sets iri_vtag and M_VLANTAG does not handle thecase where the 802.11q VLAN tag is 0. Fix this issue acrossthe iflib drivers. While there, also improve and align theVLAN tag check extraction, by moving it outside the RX descriptorloop, eliminating a local variable and additional checks.PR: 260068Reviewed by: kbowling, gallatinReported by: erjMFC after: 1 monthDifferential Revision: https://reviews.freebsd.org/D33156 List of files: /freebsd/sys/dev/e1000/em_txrx.c Tue, 28 Dec 2021 12:05:31 +0100 Vincenzo Maffione <vmaffione@FreeBSD.org> http://kernelsources.org:8080/source/history/freebsd/sys/dev/e1000/em_txrx.c#52f45d8acee95199159b65a33c94142492c38e41 net: iflib: let the drivers use isc_capenableSince isc_capenable (private copy of ifp->if_capenable) isnow synchronized to if_capenable, use it in the driverswhen checking the IFCAP_* bits.This results in better cache usage and avoids indirectionthrough the ifp pointer.PR: 260068Reviewed by: kbowling, gallatinMFC after: 1 weekDifferential Revision: https://reviews.freebsd.org/D33156 List of files: /freebsd/sys/dev/e1000/em_txrx.c Tue, 28 Dec 2021 12:00:32 +0100 Vincenzo Maffione <vmaffione@FreeBSD.org>