Switch to DuckDuckGo Search
   March 7, 2020  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >


NOTICE: This channel is no longer actively logged.

Toggle Join/Part | bottom
[00:03:37] *** Asgaroth <Asgaroth!~Asgaroth@51.37.16.246> has quit IRC (Quit: Leaving)
[00:06:13] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[00:21:31] *** andy_js <andy_js!~andy@51.146.99.40> has quit IRC (Quit: andy_js)
[00:24:02] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[00:37:46] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has joined #illumos
[00:51:17] *** EisNerd <EisNerd!~manschwet@mail.cs-software-gmbh.de> has quit IRC (Remote host closed the connection)
[00:51:39] *** EisNerd <EisNerd!~manschwet@mail.cs-software-gmbh.de> has joined #illumos
[00:52:49] *** ngchk1 <ngchk1!~ngchk1@b2b-92-50-91-166.unitymedia.biz> has joined #illumos
[01:06:24] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[01:07:02] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[01:22:47] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has joined #illumos
[01:26:24] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 256 seconds)
[01:29:58] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[01:30:27] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has joined #illumos
[01:31:39] <gitomat> [illumos-gate] 12205 want generic NIC transceiver fault events -- Alex Wilson <alex at uq dot edu.au>
[01:31:40] <gitomat> [illumos-gate] 12203 FMA fault ASRUs missing FMRI annotations -- Alex Wilson <alex at uq dot edu.au>
[01:31:41] <gitomat> [illumos-gate] 12204 want driver for Mellanox ConnectX-4/5/6 NICs -- Alex Wilson <alex at uq dot edu.au>
[01:31:48] <rmustacc> Thanks arekinath!
[01:34:30] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has quit IRC (Client Quit)
[01:40:16] *** xzilla <xzilla!~robert@12.157.195.98> has joined #illumos
[01:42:34] <arekinath> woohoo
[01:43:24] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Remote host closed the connection)
[01:44:03] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[01:46:51] *** xzilla <xzilla!~robert@12.157.195.98> has quit IRC (Ping timeout: 260 seconds)
[03:33:54] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Ping timeout: 256 seconds)
[03:40:04] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has joined #illumos
[04:34:02] *** jcea <jcea!~Thunderbi@51.159.34.131> has quit IRC (Quit: jcea)
[05:00:10] <rzezeski> woot! Nice work arekinath!
[05:06:01] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[05:09:19] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has joined #illumos
[05:35:27] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 240 seconds)
[05:35:59] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[05:41:00] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:c5c9:29d7:bd7e:abc1> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[05:43:23] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 265 seconds)
[07:00:25] *** BOKALDO <BOKALDO!~BOKALDO@81.198.159.225> has joined #illumos
[07:14:33] *** MerlinDMC <MerlinDMC!~merlin@163.172.186.44> has quit IRC (Quit: ZNC 1.7.5+deb3+b1 - https://znc.in)
[07:18:17] *** MerlinDMC <MerlinDMC!~merlin@163.172.186.44> has joined #illumos
[07:50:38] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[07:58:58] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 258 seconds)
[08:13:31] <tsoome> :) nice one.
[09:01:51] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Read error: Connection reset by peer)
[09:02:40] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[09:08:47] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has quit IRC (Ping timeout: 268 seconds)
[09:14:23] *** jenelizabeth <jenelizabeth!~jenelizab@cpc155793-brmb11-2-0-cust474.1-3.cable.virginm.net> has joined #illumos
[09:19:01] *** jenelizabeth <jenelizabeth!~jenelizab@cpc155793-brmb11-2-0-cust474.1-3.cable.virginm.net> has quit IRC (Ping timeout: 255 seconds)
[09:21:17] *** jenelizabeth <jenelizabeth!~jenelizab@cpc155793-brmb11-2-0-cust474.1-3.cable.virginm.net> has joined #illumos
[09:25:52] *** jenelizabeth <jenelizabeth!~jenelizab@cpc155793-brmb11-2-0-cust474.1-3.cable.virginm.net> has quit IRC (Read error: Connection reset by peer)
[09:55:48] *** andy_js <andy_js!~andy@51.146.99.40> has joined #illumos
[10:07:28] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[10:16:50] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 240 seconds)
[11:19:18] <v_a_b> alanc Thanks for the OpenGroup link! It's... enlightening.
[11:22:06] <v_a_b> Both the S11.3 SRU 36.28 and the OmniOS CE 032 man page for isencrypt() say December 29, 1996 :-)
[11:22:22] <v_a_b> 36.18
[12:10:37] <igork> arekinath: thanks! i have tested it with mellanox 25g over aggr - works as well
[12:11:51] <igork> but your version was failed (panic) and with Paul changes on top - all fine. issue with rx groups
[12:24:40] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[12:27:33] *** rzezeski <rzezeski!uid151901@gateway/web/irccloud.com/x-cnepnobjhfbpayqo> has quit IRC (Quit: Connection closed for inactivity)
[12:35:40] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 255 seconds)
[12:47:08] *** wacki <wacki!~wacki@i577A5980.versanet.de> has joined #illumos
[13:07:47] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[13:10:48] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has joined #illumos
[13:13:09] <arekinath> igork: yeah I mentioned that in the testing notes in the ticket -- that c-x5 parts need Paul's patches as well
[13:13:28] <arekinath> hopefully he can get those ready for RTI soon now that the base one is in anywau
[13:15:51] <arekinath> the 25G c-x4 parts do work, just not the c-x5 ones since it tries to make too many groups by default
[13:15:52] <igork> arekinath: it was with x4 and x5
[13:16:35] <igork> one - it is x4 SIOM from supermicro
[13:16:42] <arekinath> oh? my 25G c-x4lx parts work fine as is...
[13:16:59] <arekinath> do you have x4s that aren't the Lx variant?
[13:17:07] <igork> dual 25g ports
[13:17:30] <igork> one moment - try to boot server and will provide more details
[13:17:34] <igork> 5 min
[13:18:00] <igork> right now x4 and x5 with Paul updates are fine
[13:18:36] <igork> i sent to you email some time ago about my panic - did you see it >
[13:18:44] <igork> ir was severl weeks ago
[13:19:00] <arekinath> there are dual 25G parts in the plain c-x4, the c-x4lx and the c-x5
[13:19:10] <arekinath> so saying that doesn't actually narrow it haha
[13:19:45] <igork> 1 in use PCI Exp. Gen 3 x16 CPU1 SLOT1, Mellanox Technologies MT27710 Family [ConnectX-4 Lx] (mlxcx)
[13:20:35] <igork> it is SIOM on supermicro server - dual 25g
[13:20:45] <arekinath> mm yeah that's the same chip I have here which works for me
[13:20:51] <arekinath> doesn't matter anyway if Paul's patches fix it
[13:20:57] <igork> yes
[13:20:59] <arekinath> since he'll be merging those in soon :)
[13:21:05] <igork> just wanted provide more info :)
[13:27:52] *** andy_js <andy_js!~andy@51.146.99.40> has quit IRC (Ping timeout: 255 seconds)
[13:28:10] *** andy_js <andy_js!~andy@51.146.99.40> has joined #illumos
[13:41:10] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[13:42:30] *** liv3010m <liv3010m!~liv3010m@77-72-245-190.fibertel.com.ar> has joined #illumos
[13:47:45] *** rzezeski <rzezeski!uid151901@gateway/web/irccloud.com/x-ntkywxrmudagwpur> has joined #illumos
[14:15:52] *** nde <nde!uid414739@gateway/web/irccloud.com/x-upitihovkojdiolo> has quit IRC (Quit: Connection closed for inactivity)
[14:29:23] *** BOKALDO <BOKALDO!~BOKALDO@81.198.159.225> has quit IRC (Quit: Leaving)
[14:43:40] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[14:51:58] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has quit IRC (Ping timeout: 265 seconds)
[14:55:36] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:40d0:87a3:8ecc:c806> has joined #illumos
[15:13:03] *** eki <eki!~eki@dsl-hkibng41-54f858-46.dhcp.inet.fi> has quit IRC (Quit: leaving)
[15:21:34] *** eki <eki!~eki@dsl-hkibng41-54f858-46.dhcp.inet.fi> has joined #illumos
[15:56:00] *** idodeclare <idodeclare!~textual@2600:1700:1101:17c0:40d0:87a3:8ecc:c806> has quit IRC (Ping timeout: 256 seconds)
[15:56:04] *** BOKALDO <BOKALDO!~BOKALDO@81.198.159.225> has joined #illumos
[15:57:16] <Smithx10> I wonder how fast those Mellanox cards will go. None of the offloads are being used right ?
[15:57:21] *** idodeclare <idodeclare!~textual@104-189-74-222.lightspeed.snantx.sbcglobal.net> has joined #illumos
[15:58:46] <igork> Smithx10: i have tested by iperf - 22-23Gbit
[15:59:12] <igork> mtu=9k
[15:59:15] <Smithx10> What card ?
[15:59:28] <igork> connectX-4 and X-5
[16:00:01] <igork> direct connect card-to-card
[16:00:26] <Smithx10> 25 gb cards?
[16:00:32] <igork> yes
[16:00:57] <Smithx10> Wonder how it will scale up
[16:04:08] <rzezeski> Smithx10: It's my understanding that Alex made use of mac address + vlan hardware steering. This is one reason why there was a renewed effort to get my vlan/aggr work upstreamed, as his work depended on it.
[16:04:56] <Smithx10> Dope
[16:06:18] <rmustacc> The main missing thing right now that'll help is tso support. However checksum offload is used at least.
[16:06:29] <Smithx10> Should get Alex some 100gbps cards :P
[16:06:45] <rmustacc> Then the other thing is taking the sw lro work I started and finishing that up.
[16:06:59] <rmustacc> Those combined should give a pretty good pop if you care about single stream on 100 GbE.
[16:07:12] <rmustacc> Though that's not usually the workload over 100 GbE, tbh.
[16:07:52] <rmustacc> Usually it's a larger number of smaller throughput conns.
[16:10:10] <Smithx10> Understood, At the moment I believe the hdfs people at work claim they want it. Not sure if they actually get anywhere close
[16:11:21] <rmustacc> There are a bunch of reasons to go to 100 GbE still. The thing to understand is just how it's going to be used.
[16:11:25] <Smithx10> rmustacc: I’d imagine tho at some point nvme over fabrics will use that bw ... but obviously that’s not tcp
[16:11:30] <rmustacc> For example, I want 4x25 GbE.
[16:11:54] <rmustacc> Smithx10: Yes, though I think for a number of folks, rather than trying to use an Ethernet based fabric, doing that over tcp will become more common.
[16:12:02] <rmustacc> Certainly it's easier to get going than RoCE and iWARP.
[16:13:17] <Smithx10> What’s the reasoning for the 4 25gb?
[16:13:39] <rmustacc> I just used that as a density example.
[16:13:48] <Smithx10> Ahhh 😱
[16:13:50] <rmustacc> As in you want 4 logical 25 GbE streams over 1 link.
[16:14:16] <rmustacc> e.g. you have 4 VMs at 25, or 10 VMs at 10, rather than 1 at 100 GbE.
[16:34:42] *** andy_js <andy_js!~andy@51.146.99.40> has quit IRC (Read error: Connection reset by peer)
[16:35:06] *** andy_js <andy_js!~andy@51.146.99.40> has joined #illumos
[17:06:44] <Smithx10> Not a fan of the name iWARP
[17:07:11] <Smithx10> Not an acronym? Come on?!
[17:10:27] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has joined #illumos
[17:39:42] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[17:52:15] *** nde <nde!uid414739@gateway/web/irccloud.com/x-sqvgbrjlvyjtzdwc> has joined #illumos
[18:15:05] <gitomat> [illumos-gate] 12364 mdb trips assertion related to autowrap -- John Levon <john.levon at joyent dot com>
[19:53:40] *** ngchk1 <ngchk1!~ngchk1@b2b-92-50-91-166.unitymedia.biz> has quit IRC (Quit: Leaving)
[20:04:26] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[20:16:31] *** Bdragon <Bdragon!~bdragon@drupal.org/user/53081/view> has joined #illumos
[20:18:02] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has quit IRC (Ping timeout: 258 seconds)
[20:33:14] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has joined #illumos
[21:11:57] *** andy_js <andy_js!~andy@51.146.99.40> has quit IRC (Read error: Connection reset by peer)
[21:12:25] *** andy_js <andy_js!~andy@51.146.99.40> has joined #illumos
[21:23:18] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has quit IRC (Remote host closed the connection)
[21:23:45] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has joined #illumos
[21:25:17] <tsoome> jlevon we did introduce new bug with 12195 acpidump failed under EFI, sorry I did notice it in review:)
[21:25:26] <jlevon> oh?
[21:25:39] <tsoome> in fakebop.c: ../../i86pc/os/fakebop.c:2932:8: error: 'rsdp' may be used uninitialized in this
[21:25:39] <tsoome> function [-Werror=maybe-uninitialized]
[21:26:08] <tsoome> xpc case does not call find_rsdp()
[21:26:14] <tsoome> xpv*
[21:26:53] <jlevon> how come that doesn't break the build?
[21:27:11] <tsoome> only gcc9 does catch those
[21:27:33] <tsoome> I just got gcc9 test build done:)
[21:27:36] <jlevon> also I don't understand the error
[21:28:23] <jlevon> oh
[21:28:23] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[21:28:30] <jlevon> it's build_firmware_properties()
[21:28:31] <tsoome> tp = find_fw_table(rsdp, ACPI_SIG_MCFG); is using rsdp, but it is not set in this #if #else #endif branch
[21:28:36] <tsoome> yes
[21:30:23] <jlevon> so it's a simple fix, are you going to or would you like me to?
[21:30:31] <jlevon> we should just add find_rsdp() to the xpv case too
[21:31:16] <tsoome> I do not mind either way really, atm looking on other case we recently got:)
[21:32:06] <jlevon> I don't have a gcc9 set up, so if you could?
[21:32:08] <jlevon> thanks
[21:32:30] <tsoome> yes, I'll post it.
[21:33:32] <tsoome> ../../common/io/aggr/aggr_grp.c:743:22: error: 'i' may be used uninitialized and ../../common/io/aggr/aggr_grp.c:792:2: error: 'addr' may be used uninitialized
[21:34:05] <tsoome> those are other 2 new ones and apparently actual bugs:)
[21:35:49] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[21:37:11] <tsoome> anyhow, those have to wait tomorrow, quite tired of this week. I had whole week fight with HDS GAD setup at client site:D at least I got the POC up...
[21:42:38] *** BOKALDO <BOKALDO!~BOKALDO@81.198.159.225> has quit IRC (Quit: Leaving)
[21:45:14] <jlevon> hah I know zero
[22:28:16] *** wacki <wacki!~wacki@i577A5980.versanet.de> has left #illumos
[22:52:41] *** gitomat <gitomat!~nodebot@165.225.148.18> has quit IRC (Remote host closed the connection)
[22:53:27] *** gitomat <gitomat!~nodebot@165.225.148.18> has joined #illumos
[23:26:33] <arekinath> Smithx10: I have a 100g c-x5 card in a box now that I just haven't had time to test yet
[23:26:53] <arekinath> dual 100g in fact, and it's even plugged into a switch haha
[23:27:05] <Smithx10> arekinath: hell yeah !!!!
[23:28:11] <arekinath> start of semester is busy with random other crap, hopefully it calms down in a week or three and I can get back to it
[23:28:30] <Smithx10> Should make a big bug bounty :P
[23:30:07] <Smithx10> Hold 100g throughput ransom lol
[23:30:20] <Smithx10> Sell it to Samsung
[23:31:54] <Smithx10> Does eating maple Cheerios make you Canadian ?
[23:32:42] <rmustacc> arekinath: Is it in a box with pcieg4?
[23:33:56] <igork> i'll have 100g mellanox in dgx on next week
[23:34:12] <igork> 4 cards with singe port
[23:35:07] <arekinath> rmustacc: uh
[23:35:31] <rmustacc> arekinath: Is in a system with AMD Rome?
[23:35:36] <rmustacc> *Is it
[23:35:45] <arekinath> it's an old r630 so nope
[23:35:46] <arekinath> I mostly just wanted to test that it worked as was faster than a 25G haha
[23:36:01] <rmustacc> Yeah, that's fine. Just wanted to point out the pci bw limit.
[23:36:06] <rmustacc> So you don't go too crazy.
[23:36:10] <arekinath> nah, the Rome box is separate
[23:36:25] <rmustacc> With gen 3 you sohuld be able to saturate one port, iirc.
[23:36:58] <arekinath> haha thanks
[23:36:59] <arekinath> yeah that sounds about right
[23:37:18] <arekinath> might use it for getting LSO going
[23:41:33] <rmustacc> I should use that as an incentive to pick back up lor.
[23:41:34] <rmustacc> *lro
top

   March 7, 2020  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >