src/sys/dev/e1000, branch upstream/11.1.0 FreeBSD source tree Direct commit to stable/11 to correctly setting the EIAC and IMS 2017-06-19T15:03:47+00:00 Sean Bruno sbruno@FreeBSD.org 2017-06-19T15:03:47+00:00 7088a1f12e4c049ce1356dbb7ca5d9cfd94e7890 registers to the same values when processing interrupts. This reverts a change made in r286831 that was not fully reverted in r311979 This resolves PR https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211219 PR: 211219 Submitted by: Franco Fitchner <franco@opnsense.org> Approved by: re (marius) 
registers to the same values when processing interrupts.  This reverts a
change made in r286831 that was not fully reverted in r311979

This resolves PR https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211219

PR:		211219
Submitted by:	Franco Fitchner <franco@opnsense.org>
Approved by:	re (marius)
Direct commit of fixes to stable/11, resolving PCI passthrough and initialization 2017-04-05T19:46:24+00:00 Sean Bruno sbruno@FreeBSD.org 2017-04-05T19:46:24+00:00 4fbced7be6d4edab9ec6ac6eb543801040ae03d3 issues when trying to passthrough a i340 (igb) to VMware. While here, cleanup some bits of em(4) to DTRT as well. PR: 218113 Submitted by: Kaho Toshikazu <kaho@elam.kais.kyoto-u.ac.jp> 
issues when trying to passthrough a i340 (igb) to VMware.

While here, cleanup some bits of em(4) to DTRT as well.

PR:		218113
Submitted by:	Kaho Toshikazu <kaho@elam.kais.kyoto-u.ac.jp>
Disable the driver managed queue for igb(4) when the legacy transmit 2017-02-25T20:21:39+00:00 Luiz Otavio O Souza loos@FreeBSD.org 2017-02-25T20:21:39+00:00 04fb023ad26897afe82ccccab47c7b99c7c6a7f7 interface is used. The legacy API (IGB_LEGACY_TX) is enabled when ALTQ is built into kernel. As noted in altq(9), it is responsibility of the caller to protect this queue against concurrent access and, in the igb case, the interface send queue is protected by tx queue mutex. This obviously cannot protect the driver managed queue against concurrent access from different tx queues and leads to numerous and quite strange panic traces (usually shown as packets disappearing into thin air). Improving the locking to cope with this means serialize all access to this (single) queue and produces no gain, it actually affects the performance quite noticeabily. The driver managed queue is already disabled when an ALTQ queue discipline is set on interface (in altq_enable()), because the driver managed queue can interfere with ALTQ timing (whence the reports that setting an ALTQ queue discipline on interface also fixes the issue). Disabling this additional queue keeps the ability to use if_start() to send packets to individual NIC queues while it simply eliminate the race. This is a direct commit to stable/11 as -head driver does not support ALTQ anymore. PR: 213257 PR: 212413 Discussed with: sbruno Tested by: Konstantin Kormashev <konstantin@netgate.com> Obtained from: pfSense Sponsored by: Rubicon Communications, LLC (Netgate) 
interface is used.

The legacy API (IGB_LEGACY_TX) is enabled when ALTQ is built into kernel.

As noted in altq(9), it is responsibility of the caller to protect this
queue against concurrent access and, in the igb case, the interface send
queue is protected by tx queue mutex.  This obviously cannot protect the
driver managed queue against concurrent access from different tx queues
and leads to numerous and quite strange panic traces (usually shown as
packets disappearing into thin air).

Improving the locking to cope with this means serialize all access to this
(single) queue and produces no gain, it actually affects the performance
quite noticeabily.

The driver managed queue is already disabled when an ALTQ queue discipline
is set on interface (in altq_enable()), because the driver managed queue
can interfere with ALTQ timing (whence the reports that setting an ALTQ
queue discipline on interface also fixes the issue).

Disabling this additional queue keeps the ability to use if_start() to
send packets to individual NIC queues while it simply eliminate the race.

This is a direct commit to stable/11 as -head driver does not support ALTQ
anymore.

PR:		213257
PR:		212413
Discussed with:	sbruno
Tested by:	Konstantin Kormashev <konstantin@netgate.com>
Obtained from:	pfSense
Sponsored by:	Rubicon Communications, LLC (Netgate)
MFC: r311979 2017-02-22T17:57:19+00:00 Marius Strobl marius@FreeBSD.org 2017-02-22T17:57:19+00:00 b8d6207cd1a7dbee152818b86086eab95de76dab Reset the EIAC register to include the LINK status bit and restore link up/down notifications. 
Reset the EIAC register to include the LINK status bit and restore
link up/down notifications.
MFC r307649: 2016-11-02T08:41:01+00:00 Konstantin Belousov kib@FreeBSD.org 2016-11-02T08:41:01+00:00 8f58ba6adc0312c5082a674942e899605da9db0b Partial workaround for Intel PCI adapters reading past the end of the host-programmed DMA regions. 
Partial workaround for Intel PCI adapters reading past the end of the
host-programmed DMA regions.
MFC r304149 2016-08-18T07:31:19+00:00 Sean Bruno sbruno@FreeBSD.org 2016-08-18T07:31:19+00:00 b81277c6c5c3ce1e39682f5b06dc9732fe06a0eb e1000: Add support for Kaby Lake IDs Fixup some errors when transitioning to/from low power states. 
e1000:  Add support for Kaby Lake IDs

Fixup some errors when transitioning to/from low power states.
MFC r303638 2016-08-02T15:43:18+00:00 Sean Bruno sbruno@FreeBSD.org 2016-08-02T15:43:18+00:00 5dd6aeda913171a080b7f11058f84ecbb598142d Unbreak NETMAP usage with em(4) broken in r293331. Approved by: re (gjb) 
Unbreak NETMAP usage with em(4) broken in r293331.

Approved by:	re (gjb)
Do not initialize the adapter on MTU change when adapter status is down. 2016-07-07T03:39:18+00:00 Sean Bruno sbruno@FreeBSD.org 2016-07-07T03:39:18+00:00 761e52618557a81decca704074edcf3330da8359 This fixes long-standing problems when changing settings of the adapter. Discussed in: https://lists.freebsd.org/pipermail/freebsd-net/2016-June/045509.html Submitted by: arnaud.ysmal@stormshield.eu Reviewed by: erj@freebsd.org Approved by: re (gjb) Differential Revision: https://reviews.freebsd.org/D7030 
This fixes long-standing problems when changing settings of the adapter.

Discussed in:
https://lists.freebsd.org/pipermail/freebsd-net/2016-June/045509.html

Submitted by:	arnaud.ysmal@stormshield.eu
Reviewed by:	erj@freebsd.org
Approved by:	re (gjb)
Differential Revision:	https://reviews.freebsd.org/D7030
net: Use M_HASHTYPE_OPAQUE_HASH if the mbuf flowid has hash properties 2016-06-07T04:51:50+00:00 Sepherosa Ziehau sephe@FreeBSD.org 2016-06-07T04:51:50+00:00 36ad8372d499c26f47569da1e76b358d225d969f Reviewed by: hps, erj, tuexen Sponsored by: Microsoft OSTC Differential Revision: https://reviews.freebsd.org/D6688 
Reviewed by:	hps, erj, tuexen
Sponsored by:	Microsoft OSTC
Differential Revision:	https://reviews.freebsd.org/D6688
Don't repeat the the word 'the' 2016-05-17T12:52:31+00:00 Eitan Adler eadler@FreeBSD.org 2016-05-17T12:52:31+00:00 cef367e6a15b6e3e2aca88d453c74962cf1f94e7 (one manual change to fix grammar) Confirmed With: db Approved by: secteam (not really, but this is a comment typo fix) 
(one manual change to fix grammar)

Confirmed With: db
Approved by: secteam (not really, but this is a comment typo fix)