Made it a bit further with this, I undid this kernel commit: https://bugzilla.redhat.com/show_bug.cgi?id=1627816#c37 When I connect to the port now it comes up as 10Mb/s, stays like that for a few seconds, the link goes down and then comes back up at 1000Mb/s. Perhaps it's like that for people on newer kernels and they don't notice, or they've fixed it in a better manner.
-- Paul On Thu, 14 Feb 2019 at 13:08, Paul Gildea <gild...@tcd.ie> wrote: > Thanks, same driver as me, so maybe the problem is buried somewhere in the > kernel, or if it is a problem with the e1000e driver they may have made > changes in Ubuntu that mitigate issues such as this. > > On Thu, 14 Feb 2019 at 12:56, Andre Ruiz <andre.r...@gmail.com> wrote: > >> >> I can't remember on fedora, but now on ubuntu cosmic it's: >> >> andre@thinkpad:~/temp$ uname -a >> Linux thinkpad 4.18.0-15-generic #16-Ubuntu SMP Thu Feb 7 10:56:39 UTC >> 2019 x86_64 x86_64 x86_64 GNU/Linux >> andre@thinkpad:~/temp$ modinfo e1000e | head -n 6 >> filename: >> >> /lib/modules/4.18.0-15-generic/kernel/drivers/net/ethernet/intel/e1000e/e1000e.ko >> version: 3.2.6-k >> license: GPL >> description: Intel(R) PRO/1000 Network Driver >> author: Intel Corporation, <linux.n...@intel.com> >> srcversion: 523CF030A04777C2DBD2CDC >> >> Andre >> >> >> On Thu, Feb 14, 2019 at 10:10 AM Paul Gildea <gild...@tcd.ie> wrote: >> >>> Thanks for the info Andre, what driver e1000e driver does your Ubuntu or >>> newer Fedora use if you know, have they upgraded from 3.2.6-k? >>> From the redhat forum above the problem also appears with: >>> >>> *i219*-LM rev 21 (8086:15d7), openSUSE Leap 15.0, 5.0.0-rc1, with (IIRC) a >>> HP 1080-24 switch. >>> *i219*-LM rev 31 (8086:15b7), Debian 9, 4.9.0-8-amd64, with an unknown type >>> of switch/router. >>> >>> I've seen people mention my exact problem when searching, just unplugging >>> for a few seconds and it comes back up as 10Mb/s. Unsure now of where the >>> problem lies. >>> >>> -- >>> >>> Paul >>> >>> >>> >>> On Thu, 14 Feb 2019 at 12:01, Andre Ruiz <andre.r...@gmail.com> wrote: >>> >>>> >>>> Paul, >>>> >>>> Indeed I did not get any replies (other than other similar reports of >>>> problems). In my case it's certainly not the hardware, the port worked fine >>>> in windows and I have recently switched from fedora to ubuntu where it >>>> works fine too (both on bionic and cosmic) so I kinda "fixed" it. >>>> >>>> In my case, once it works, it works all the time, you can unplug/plug >>>> with no problems. Once it went to sleep (suspend to ram), it would only get >>>> 10Mbit/s after coming back, until next reboot (a warm boot would suffice). >>>> >>>> The trouble reproducing came when fedora released some kernels (stable >>>> release updates) after my report, and with some of them the problem was a >>>> little more random, sometimes it would work nice all the time. >>>> >>>> Andre >>>> >>>> >>>> On Thu, Feb 14, 2019 at 9:49 AM Paul Gildea <gild...@tcd.ie> wrote: >>>> >>>>> Thanks Andre, I see your report here, did it have no replies or am I >>>>> reading that wrong? And your report here >>>>> https://bugzilla.redhat.com/show_bug.cgi?id=1038156 >>>>> Also another report by Michael Yartys where he concluded his >>>>> hardware was faulty. Is this still happening for you? You seemed to have >>>>> some trouble reproducing after upgrading? >>>>> In my case it happens reliably 100% of the time, over multiple similar >>>>> hardware setups (ruling out hardware failure). My system doesn't have a >>>>> sleep mode, this just happens after any cable unplug after boot. >>>>> >>>>> The kernel fix by Kamil reported on the the redhat thread seemed >>>>> promising but that commit was not in place on kernel 4.7.8 which I already >>>>> tested. >>>>> I will try and upgrade to a newer version of the driver than 3.2.6-k >>>>> and see is the issue still there, I imagine it is. >>>>> >>>>> -- >>>>> Paul >>>>> >>>>> >>>>> >>>>> On Wed, 13 Feb 2019 at 19:38, Andre Ruiz <andre.r...@gmail.com> wrote: >>>>> >>>>>> >>>>>> Paul, >>>>>> >>>>>> I reported something very similar some time ago, take a look on the >>>>>> list history (can't search for you right now, sorry). >>>>>> >>>>>> Andre >>>>>> >>>>>> On Wed, Feb 13, 2019 at 5:25 PM Paul Gildea <gild...@tcd.ie> wrote: >>>>>> >>>>>>> I am having an issue where my link comes up as 10Mb/s if I plug in >>>>>>> the >>>>>>> cable after the system has booted, no matter what I connect to. >>>>>>> Tested with >>>>>>> kernel 4.7.8 and 4.14.73. If the cable is connected during boot the >>>>>>> correct >>>>>>> speed of 1000Mb/s is observed. I have multiple instances of this >>>>>>> hardware >>>>>>> setup and it happens with all of them reliably. >>>>>>> >>>>>>> Here is some info: >>>>>>> >>>>>>> driver: e1000e >>>>>>> version: 3.2.6-k >>>>>>> firmware-version: 0.8-4 >>>>>>> expansion-rom-version: >>>>>>> bus-info: 0000:00:1f.6 >>>>>>> supports-statistics: yes >>>>>>> supports-test: yes >>>>>>> supports-eeprom-access: yes >>>>>>> supports-register-dump: yes >>>>>>> supports-priv-flags: no >>>>>>> >>>>>>> >>>>>>> -------------------------------------------------------- >>>>>>> >>>>>>> >>>>>>> MAC Registers >>>>>>> ------------- >>>>>>> 0x00000: CTRL (Device control register) 0x40180240 >>>>>>> Endian mode (buffers): little >>>>>>> Link reset: normal >>>>>>> Set link up: 1 >>>>>>> Invert Loss-Of-Signal: no >>>>>>> Receive flow control: disabled >>>>>>> Transmit flow control: disabled >>>>>>> VLAN mode: enabled >>>>>>> Auto speed detect: disabled >>>>>>> Speed select: 1000Mb/s >>>>>>> Force speed: no >>>>>>> Force duplex: no >>>>>>> 0x00008: STATUS (Device status register) 0x40080000 >>>>>>> Duplex: half >>>>>>> Link up: no link config >>>>>>> TBI mode: disabled >>>>>>> Link speed: 10Mb/s >>>>>>> Bus type: PCI >>>>>>> Bus speed: 33MHz >>>>>>> Bus width: 32-bit >>>>>>> 0x00100: RCTL (Receive control register) 0x04008002 >>>>>>> Receiver: enabled >>>>>>> Store bad packets: disabled >>>>>>> Unicast promiscuous: disabled >>>>>>> Multicast promiscuous: disabled >>>>>>> Long packet: disabled >>>>>>> Descriptor minimum threshold size: 1/2 >>>>>>> Broadcast accept mode: accept >>>>>>> VLAN filter: disabled >>>>>>> Canonical form indicator: disabled >>>>>>> Discard pause frames: filtered >>>>>>> Pass MAC control frames: don't pass >>>>>>> Receive buffer size: 2048 >>>>>>> 0x02808: RDLEN (Receive desc length) 0x00001000 >>>>>>> 0x02810: RDH (Receive desc head) 0x00000000 >>>>>>> 0x02818: RDT (Receive desc tail) 0x000000F0 >>>>>>> 0x02820: RDTR (Receive delay timer) 0x00000000 >>>>>>> 0x00400: TCTL (Transmit ctrl register) 0x3103F0F8 >>>>>>> Transmitter: disabled >>>>>>> Pad short packets: enabled >>>>>>> Software XOFF Transmission: disabled >>>>>>> Re-transmit on late collision: enabled >>>>>>> 0x03808: TDLEN (Transmit desc length) 0x00001000 >>>>>>> 0x03810: TDH (Transmit desc head) 0x00000000 >>>>>>> 0x03818: TDT (Transmit desc tail) 0x00000000 >>>>>>> 0x03820: TIDV (Transmit delay timer) 0x00000008 >>>>>>> PHY type: unknown >>>>>>> >>>>>>> >>>>>>> -------------------------------------------------------- >>>>>>> >>>>>>> Supported ports: [ TP ] >>>>>>> Supported link modes: 10baseT/Half 10baseT/Full >>>>>>> 100baseT/Half 100baseT/Full >>>>>>> 1000baseT/Full >>>>>>> Supported pause frame use: No >>>>>>> Supports auto-negotiation: Yes >>>>>>> Advertised link modes: 10baseT/Half 10baseT/Full >>>>>>> 100baseT/Half 100baseT/Full >>>>>>> 1000baseT/Full >>>>>>> Advertised pause frame use: No >>>>>>> Advertised auto-negotiation: Yes >>>>>>> Speed: 10Mb/s >>>>>>> Duplex: Full >>>>>>> Port: Twisted Pair >>>>>>> PHYAD: 1 >>>>>>> Transceiver: internal >>>>>>> Auto-negotiation: on >>>>>>> MDI-X: on (auto) >>>>>>> Supports Wake-on: pumbg >>>>>>> Wake-on: g >>>>>>> Current message level: 0x00000007 (7) >>>>>>> drv probe link >>>>>>> Link detected: yes >>>>>>> >>>>>>> >>>>>>> >>>>>>> Running a command like "ethtool -s eth5 advertise 20" or "ethtool -s >>>>>>> eth5 >>>>>>> advertise 2c" causes the link to go down and come back up with the >>>>>>> right >>>>>>> speed. If I then remove the cable for a few seconds and put it back >>>>>>> in the >>>>>>> speed will go back to 10Mb/s. Quicker than that and it seems to stay >>>>>>> at >>>>>>> 1000Mb/s. >>>>>>> >>>>>>> >>>>>>> Here is a dmesg with the cable out at boot and then me plugging it >>>>>>> in after: >>>>>>> >>>>>>> >>>>>>> [root@trx-r6 /]# grep -ri 1000e a >>>>>>> [ 12.836934] e1000e: Intel(R) PRO/1000 Network Driver - 3.2.6-k >>>>>>> [ 12.836936] e1000e: Copyright(c) 1999 - 2015 Intel Corporation. >>>>>>> [ 12.841135] e1000e 0000:00:1f.6: Interrupt Throttling Rate >>>>>>> (ints/sec) >>>>>>> set to dynamic conservative mode >>>>>>> [ 13.029402] e1000e 0000:00:1f.6 0000:00:1f.6 (uninitialized): >>>>>>> registered >>>>>>> PHC clock >>>>>>> [ 13.107349] e1000e 0000:00:1f.6 eth5: (PCI Express:2.5GT/s:Width >>>>>>> x1) >>>>>>> 00:13:f2:a0:01:85 >>>>>>> [ 13.107352] e1000e 0000:00:1f.6 eth5: Intel(R) PRO/1000 Network >>>>>>> Connection >>>>>>> [ 13.107408] e1000e 0000:00:1f.6 eth5: MAC: 12, PHY: 12, PBA No: >>>>>>> FFFFFF-0FF >>>>>>> [ 2957.722516] e1000e: eth5 NIC Link is Up 10 Mbps Full Duplex, Flow >>>>>>> Control: Rx/Tx >>>>>>> [ 2957.722521] e1000e 0000:00:1f.6 eth5: 10/100 speed: disabling TSO >>>>>>> >>>>>>> >>>>>>> Cable in the whole time during boot: >>>>>>> >>>>>>> [ 12.273114] e1000e: Intel(R) PRO/1000 Network Driver - 3.2.6-k >>>>>>> [ 12.273116] e1000e: Copyright(c) 1999 - 2015 Intel Corporation. >>>>>>> [ 12.276161] e1000e 0000:00:1f.6: Interrupt Throttling Rate >>>>>>> (ints/sec) >>>>>>> set to dynamic conservative mode >>>>>>> [ 12.375477] e1000e 0000:00:1f.6 0000:00:1f.6 (uninitialized): >>>>>>> registered >>>>>>> PHC clock >>>>>>> [ 12.458574] e1000e 0000:00:1f.6 eth5: (PCI Express:2.5GT/s:Width >>>>>>> x1) >>>>>>> 00:13:f2:a0:01:85 >>>>>>> [ 12.458578] e1000e 0000:00:1f.6 eth5: Intel(R) PRO/1000 Network >>>>>>> Connection >>>>>>> [ 12.458631] e1000e 0000:00:1f.6 eth5: MAC: 12, PHY: 12, PBA No: >>>>>>> FFFFFF-0FF >>>>>>> [ 133.384512] e1000e: eth5 NIC Link is Up 1000 Mbps Full Duplex, >>>>>>> Flow >>>>>>> Control: Rx/Tx >>>>>>> >>>>>>> Any ideas? >>>>>>> >>>>>>> Regards, >>>>>>> >>>>>>> -- >>>>>>> Paul >>>>>>> >>>>>>> _______________________________________________ >>>>>>> E1000-devel mailing list >>>>>>> E1000-devel@lists.sourceforge.net >>>>>>> https://lists.sourceforge.net/lists/listinfo/e1000-devel >>>>>>> To learn more about Intel® Ethernet, visit >>>>>>> http://communities.intel.com/community/wired >>>>>>> >>>>>> _______________________________________________ E1000-devel mailing list E1000-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/e1000-devel To learn more about Intel® Ethernet, visit http://communities.intel.com/community/wired
