src/sys/dev/igc, branch releng/14.3 FreeBSD source tree dev: Use recently added improvements to PME# support to simplify drivers 2025-04-29T14:45:07+00:00 John Baldwin jhb@FreeBSD.org 2025-03-27T20:55:12+00:00 6f27822152e46c1890db0a17737517c1beba8ea3 Depend on the PCI bus driver clearing PME# after resume to remove the need for clearing PME# from DEVICE_RESUME methods. Use pci_has_pm and pci_enable_pme. Reviewed by: Krzysztof Galazka <krzysztof.galazka@intel.com> Differential Revision: https://reviews.freebsd.org/D49251 (cherry picked from commit ddaf6524682b3ab9e50f7575db319814dbbd053a) 
Depend on the PCI bus driver clearing PME# after resume to remove the
need for clearing PME# from DEVICE_RESUME methods.

Use pci_has_pm and pci_enable_pme.

Reviewed by:	Krzysztof Galazka <krzysztof.galazka@intel.com>
Differential Revision:	https://reviews.freebsd.org/D49251

(cherry picked from commit ddaf6524682b3ab9e50f7575db319814dbbd053a)
igc(4): Fix attach for I226-K and LMVP devices 2025-04-03T19:59:47+00:00 Eric Joyner erj@FreeBSD.org 2025-02-26T17:19:38+00:00 f034ddd2fa38cb2b0e3d16059cb059626a770e19 Summary: The device IDs for these were in the driver's list of PCI ids to attach to, but igc_set_mac_type() had never been setup to set the correct mac type for these devices. Fix this by adding these IDs to the switch block in order for them to be recognized by the driver instead of returning an error. This fixes the igc(4) attach for the I226-K LOM on the ASRock Z790 PG-ITX/TB4 motherboard, allowing it to be recognized and used. Signed-off-by: Eric Joyner <erj@FreeBSD.org> Reviewed by: kbowling@ Relnotes: yes Differential Revision: https://reviews.freebsd.org/D49147 (cherry picked from commit 7ee310c80ea7b336972f53cc48b8c3d03029941e) 
Summary:
The device IDs for these were in the driver's list of PCI ids to attach
to, but igc_set_mac_type() had never been setup to set the correct mac
type for these devices. Fix this by adding these IDs to the switch block
in order for them to be recognized by the driver instead of returning an
error.

This fixes the igc(4) attach for the I226-K LOM on the ASRock Z790
PG-ITX/TB4 motherboard, allowing it to be recognized and used.

Signed-off-by: Eric Joyner <erj@FreeBSD.org>

Reviewed by:	kbowling@
Relnotes:	yes
Differential Revision: https://reviews.freebsd.org/D49147

(cherry picked from commit 7ee310c80ea7b336972f53cc48b8c3d03029941e)
igc: Remove unused register IGC_RXD_SPC_VLAN_MASK 2025-02-16T00:25:41+00:00 Kevin Bowling kbowling@FreeBSD.org 2025-02-13T18:22:50+00:00 07e712e4d1f0e0227c774a9621035b949deb5aae We don't use legacy receive descriptors and masking out the vlan ID isn't necessary since the tag is in the standard format, so remove it. (cherry picked from commit 124b7722aad7d4cf12d96c030659aef78175aa9c) 
We don't use legacy receive descriptors and masking out the vlan ID
isn't necessary since the tag is in the standard format, so remove it.

(cherry picked from commit 124b7722aad7d4cf12d96c030659aef78175aa9c)
igc.4: Add I226 and other additions to supported list 2024-12-23T22:00:51+00:00 Kevin Bowling kbowling@FreeBSD.org 2024-12-03T00:43:17+00:00 6cfff795e5aae0ec1ba834666c55fe02ff586aac (cherry picked from commit 850f78d5a191d0058fcd20beb8b1309b795ecdbd) 
(cherry picked from commit 850f78d5a191d0058fcd20beb8b1309b795ecdbd)
igc: disable hw.igc.sbp 2024-11-27T00:28:44+00:00 Kevin Bowling kbowling@FreeBSD.org 2024-11-24T21:08:54+00:00 9b7f8ee3a1a6e216092d735f54878d41bf323c18 Similar to 548d8a131d536d5f in e1000, disable this by default. Sponsored by: BBOX.io (cherry picked from commit e80419da6cef3fb6ea1902bd9a250ca21072a42c) 
Similar to 548d8a131d536d5f in e1000, disable this by default.

Sponsored by:	BBOX.io

(cherry picked from commit e80419da6cef3fb6ea1902bd9a250ca21072a42c)
igc: Style pass igc_txrx 2024-11-27T00:28:21+00:00 Kevin Bowling kbowling@FreeBSD.org 2024-11-24T07:27:12+00:00 cb9c9b0c35a03915d235dfd4d1232db857a267ae Fix up indentation and reflow long lines. Sponsored by: BBOX.io (cherry picked from commit c7fb7b5d9fcd1fc67794228820aa54df388d47b4) 
Fix up indentation and reflow long lines.

Sponsored by:	BBOX.io

(cherry picked from commit c7fb7b5d9fcd1fc67794228820aa54df388d47b4)
igc: Reflow long lines 2024-11-27T00:27:57+00:00 Kevin Bowling kbowling@FreeBSD.org 2024-11-24T05:58:13+00:00 bfabd58a9b58c88a3df6036b47d2812666b941a6 Sponsored by: BBOX.io (cherry picked from commit 9efc7325f1acb85ac8772096cc2b54fd0c6a60ff) 
Sponsored by:	BBOX.io

(cherry picked from commit 9efc7325f1acb85ac8772096cc2b54fd0c6a60ff)
igc: Normalize indentation a bit 2024-11-27T00:27:32+00:00 Kevin Bowling kbowling@FreeBSD.org 2024-11-24T03:17:27+00:00 23a4b5497a379fd417b19453107a8490384d875d Sponsored by: BBOX.io (cherry picked from commit d1bb1a501118851b02a182afdd9dd1413e9f4ce0) 
Sponsored by:	BBOX.io

(cherry picked from commit d1bb1a501118851b02a182afdd9dd1413e9f4ce0)
igc: sysctl for TCP flag handling during TSO 2024-11-24T03:21:30+00:00 Kevin Bowling kbowling@FreeBSD.org 2024-11-21T02:38:01+00:00 6890e1e79341b52665792978899b94a1bbe311dd Add tso_tcp_flags_mask_first_segment, tso_tcp_flags_mask_middle_segment, and tso_tcp_flags_mask_last_segment sysctl-variables to control the handling of TCP flags during TSO. This allows to change the masks appropriate for classical ECN and to configure appropriate masks for accurate ECN. Sponsored by: Netflix (cherry picked from commit ab540d44ba3201ff8313b90ba0096004603b2e34) 
Add tso_tcp_flags_mask_first_segment, tso_tcp_flags_mask_middle_segment,
and tso_tcp_flags_mask_last_segment sysctl-variables to control the
handling of TCP flags during TSO.

This allows to change the masks appropriate for classical ECN and to
configure appropriate masks for accurate ECN.

Sponsored by:	Netflix

(cherry picked from commit ab540d44ba3201ff8313b90ba0096004603b2e34)
igc: Remove a bogus register write in igc_if_queues_free() 2024-11-02T14:29:26+00:00 Mark Johnston markj@FreeBSD.org 2024-10-26T21:04:59+00:00 16b24ba8722735ee2657f07e32e5a5bf7ee62a2e As explained in PR 277038, iflib calls IFDI_DETACH() and then IFDI_QUEUES_FREE(). With igc, the latter writes to a register after it has been unmapped. igc_if_detach() already calls igc_release_hw_control(), and looking at callers of igc_if_queues_free(), that appears to be sufficient. So, just remove the igc_release_hw_control() call. PR: 277038 Reported by: Mike Belanger <mibelanger@qnx.com> Reviewed by: kbowling Tested by: kbowling MFC after: 1 week Differential Revision: https://reviews.freebsd.org/D47293 (cherry picked from commit 35d05a14ed7e9935be1ed0fe965b91aaaa4c92ef) 
As explained in PR 277038, iflib calls IFDI_DETACH() and then
IFDI_QUEUES_FREE().  With igc, the latter writes to a register after it
has been unmapped.

igc_if_detach() already calls igc_release_hw_control(), and looking at
callers of igc_if_queues_free(), that appears to be sufficient.  So,
just remove the igc_release_hw_control() call.

PR:		277038
Reported by:	Mike Belanger <mibelanger@qnx.com>
Reviewed by:	kbowling
Tested by:	kbowling
MFC after:	1 week
Differential Revision:	https://reviews.freebsd.org/D47293

(cherry picked from commit 35d05a14ed7e9935be1ed0fe965b91aaaa4c92ef)