Switch to DuckDuckGo Search
   May 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 | >

Toggle Join/Part | bottom
[00:00:02] <wilbury> uts/common/os/modctl.c
[00:02:22] <wilbury> ok, moduininstall() is not called if _init() returs nonzero
[00:03:53] <wilbury> i mean moduninstall()
[00:05:48] *** neirac <neirac!~neirac@pc-4-149-45-190.cm.vtr.net> has joined #illumos
[00:08:07] <wilbury> what's the preferred return for unknown ioctl()?
[00:08:26] <wilbury> EINVAL
[00:11:36] <rmustacc> ENOTTY
[00:14:12] <wilbury> ok, i triggered a panic
[00:15:13] <rmustacc> OK. If you need any help diagnosing, reach out.
[00:21:55] <wilbury> rmustacc: so far i'm progressing quite well. src.illumos.org is of a good help, ofc. i prefer to find out myself, only then ask (should i get stuck somewhere)
[00:22:20] <rmustacc> OK. I wasn't sure if it was a panic in the boader system or not.
[00:24:16] <wilbury> it was panic in my detach() (forgotten code, etc)
[00:24:24] <rmustacc> Ah, gotcha. OK.
[00:24:36] <rmustacc> I had thought it was related to the init/fini discussion.
[00:26:55] <wilbury> yes, detach() is called before _fino
[00:26:57] <wilbury> fini
[00:27:31] <rmustacc> Yes. Assuming attach has been successfuly called for an instance.
[00:28:33] <wilbury> yes, attach works. nevermind, after midnight, cu
[00:28:34] <wilbury> &
[00:50:43] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has quit IRC (Remote host closed the connection)
[00:53:12] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 272 seconds)
[00:54:22] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[00:57:09] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has joined #illumos
[01:06:30] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 272 seconds)
[01:07:54] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[01:49:08] *** ypankov <ypankov!~ypankov@85.175.17.187> has quit IRC (Ping timeout: 256 seconds)
[03:10:06] *** vgusev <vgusev!~vgusev@188.187.60.230> has quit IRC (Remote host closed the connection)
[03:10:52] *** vgusev <vgusev!~vgusev@188.187.60.230> has joined #illumos
[04:07:26] *** yuripv <yuripv!~yuripv@85.175.17.187> has joined #illumos
[04:27:13] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[04:36:28] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[04:48:49] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has quit IRC (Ping timeout: 264 seconds)
[04:53:32] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has joined #illumos
[04:55:26] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has joined #illumos
[05:02:23] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Ping timeout: 260 seconds)
[05:04:25] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has joined #illumos
[05:06:10] <gitomat> [illumos-gate] 12698 Fix typos in usr/src/man/man1m/vmstat.1m -- Nan Xiao <nan at chinadtrace dot org>
[05:18:39] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[05:35:54] *** baojg <baojg!~baojg@2604:a880:0:1010::16:1> has quit IRC ()
[06:11:17] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has quit IRC (Remote host closed the connection)
[06:11:44] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has joined #illumos
[06:28:47] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Ping timeout: 240 seconds)
[06:31:32] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[07:45:45] *** phyre <phyre!~phyre___@78.30.23.107> has joined #illumos
[08:18:52] *** tsoome_ <tsoome_!~tsoome@06c2-e8aa-94cd-4c50-2f80-4a40-07d0-2001.sta.estpak.ee> has quit IRC (Quit: This computer has gone to sleep)
[08:19:11] *** tsoome <tsoome!~tsoome@91.209.240.229> has quit IRC (Read error: Connection reset by peer)
[08:25:08] *** pwinder <pwinder!~pwinder@86.11.191.8> has joined #illumos
[08:26:01] *** neuroserve <neuroserve!~toens@ip-88-152-243-25.hsi03.unitymediagroup.de> has quit IRC (Ping timeout: 264 seconds)
[08:28:32] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has quit IRC (Ping timeout: 260 seconds)
[08:31:23] *** alanc <alanc!~alanc@inet-hqmc02-o.oracle.com> has joined #illumos
[08:41:49] *** BOKALDO <BOKALDO!~BOKALDO@87.110.146.209> has joined #illumos
[08:46:16] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has joined #illumos
[08:53:51] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[08:56:03] *** tsoome_ <tsoome_!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[09:01:32] *** trisk <trisk!~trisk@2601:196:4700:3a:8:20ff:fe3e:890e> has joined #illumos
[09:04:04] *** andy_js <andy_js!~andy@94.2.168.219> has joined #illumos
[09:31:53] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[09:34:57] *** jimklimov <jimklimov!~jimklimov@78.80.224.132> has joined #illumos
[09:35:04] *** neuroserve <neuroserve!~toens@ip-95-222-232-201.hsi15.unitymediagroup.de> has joined #illumos
[09:40:57] *** wacki <wacki!~wacki@i577B0946.versanet.de> has joined #illumos
[09:42:45] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has joined #illumos
[09:56:27] *** Smithx10 <Smithx10!sid243404@gateway/web/irccloud.com/x-ucmzivxneadmfnyn> has quit IRC (*.net *.split)
[10:01:28] *** Smithx10 <Smithx10!sid243404@gateway/web/irccloud.com/x-ucmzivxneadmfnyn> has joined #illumos
[10:05:22] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Quit: tsoome)
[10:16:22] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[10:19:57] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Client Quit)
[10:20:31] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[10:21:08] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Client Quit)
[10:25:03] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[10:45:04] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Quit: tsoome)
[10:48:05] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[10:55:24] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Quit: tsoome)
[10:59:19] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[11:12:31] *** tsoome_ <tsoome_!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: This computer has gone to sleep)
[11:27:38] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Quit: tsoome)
[11:30:21] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[11:38:19] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Quit: tsoome)
[11:38:49] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[11:39:48] *** Ericson2314 <Ericson2314!ericson231@gateway/shell/matrix.org/x-wbtgnuystlulourd> has quit IRC (Quit: killed)
[11:39:48] *** TelegramBridge[m <TelegramBridge[m!telegramt2@gateway/shell/matrix.org/x-ibwdtmvhjpravhyf> has quit IRC (Quit: killed)
[11:39:49] *** psydruid <psydruid!psydruidma@gateway/shell/matrix.org/x-nmweayxtemanflog> has quit IRC (Quit: killed)
[11:39:51] *** GrahamPerrin[m] <GrahamPerrin[m]!grahamperr@gateway/shell/matrix.org/x-boeyyhfoqprfldzu> has quit IRC (Quit: killed)
[11:40:06] *** teutat3s <teutat3s!jhotmatrix@gateway/shell/matrix.org/x-pgtomsrykoshofxf> has quit IRC (Quit: killed)
[11:55:56] *** psydruid <psydruid!psydruidma@gateway/shell/matrix.org/x-tzjqmgtxcbkdrnqn> has joined #illumos
[12:05:14] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Ping timeout: 240 seconds)
[12:07:57] *** tsoome <tsoome!~tsoome@89.219.128.66> has joined #illumos
[12:17:39] *** tsoome <tsoome!~tsoome@89.219.128.66> has quit IRC (Quit: tsoome)
[12:22:46] *** Ericson2314 <Ericson2314!ericson231@gateway/shell/matrix.org/x-jpdamnkuffgzoxdd> has joined #illumos
[12:22:46] *** teutat3s <teutat3s!jhotmatrix@gateway/shell/matrix.org/x-gpkajihoxogmxanp> has joined #illumos
[12:22:46] *** GrahamPerrin[m] <GrahamPerrin[m]!grahamperr@gateway/shell/matrix.org/x-bqxlrtmvrlgjeaoc> has joined #illumos
[12:22:46] *** TelegramBridge[m <TelegramBridge[m!telegramt2@gateway/shell/matrix.org/x-qeiunvppzlevfxen> has joined #illumos
[12:24:14] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has quit IRC (Ping timeout: 240 seconds)
[12:24:29] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has joined #illumos
[12:58:41] <EisNerd> someone here currently doing the mlx stuff?
[12:59:06] <EisNerd> do you know how to read the nv fw log?
[13:00:05] <EisNerd> It looks like the esxi here has some trouble with this nics for totally unexplainable reasons
[13:01:27] <EisNerd> I'm thinking of putting one of them into one of the storage blades, to check if this solves the network issues
[13:01:46] <EisNerd> its all a pita here currently :(
[13:03:27] <jimklimov> someone here, but sorry no idea
[13:03:30] *** tsoome <tsoome!~tsoome@12d4-01e3-bff1-a8d5-2f80-4a40-07d0-2001.sta.estpak.ee> has joined #illumos
[13:03:39] *** tsoome_ <tsoome_!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[13:03:47] <sensille> EisNerd: did you narrow it down?
[13:07:05] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has quit IRC (Quit: nbhauke)
[13:08:07] *** tsoome_ <tsoome_!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Ping timeout: 246 seconds)
[13:15:21] <EisNerd> currently busy with working on the esx trouble, so the OI network problems are a bit potponed
[13:23:20] <tsoome> postponed is good term:)
[13:23:28] <tsoome> err, potponed*
[13:23:35] <tsoome> :D
[13:28:42] <EisNerd> the esx thing is ugly, as each time the esx refuses to boot, I have to spin up a live linux and install mlx tools to work with the stuff. Very time consuming. Maybe I can install a linux to a iSCSI on OI and boot it using the mlx, maybe I'll have to check this
[13:28:57] <EisNerd> someone exerience with such?
[13:30:45] <tsoome> why it is refusing to boot?
[13:30:59] <EisNerd> it boot but ends in a psod
[13:31:14] <tsoome> firmware update?
[13:31:33] <tsoome> or esx update?
[13:31:42] <EisNerd> since we applied fw updates to the system due to HPE alerts regarding high risk trouble with storage controller FW
[13:32:27] <tsoome> vmware support?
[13:32:35] <EisNerd> now sth bothers the box in a totaly obscure manor
[13:32:58] <EisNerd> yes and HPE, now I'm the one to collect all the stuff the ask for
[13:35:11] <EisNerd> it seems to be somehow related with the mlx cards, this is why I was asking if someone knows how to get the fw logs from nvram
[13:35:48] <EisNerd> best idea of HPE support was to replace the system board proactively
[13:41:03] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[13:42:07] <EisNerd> looks like I have a bunch of metal here currently relevant for illumos development
[13:42:19] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[13:43:32] *** igork1 <igork1!~igork@91.204.56.74> has joined #illumos
[13:45:13] *** igork <igork!~igork@91.204.56.74> has quit IRC (Ping timeout: 264 seconds)
[13:47:06] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[13:50:36] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[13:53:52] *** jcea <jcea!~Thunderbi@2001:bc8:2ecd:caed:7670:6e00:7670:6e00> has joined #illumos
[13:55:27] <EisNerd> high class dual port U2 NVMe, mlx cx4 VPI, intel xl710
[14:00:51] <jlevon> tsoome: ping
[14:18:16] *** tru_tru <tru_tru!~tru@157.99.90.140> has quit IRC (Ping timeout: 256 seconds)
[14:34:48] <pwinder> EisNerd: How uptodate is your mlx driver, and how much RAM have you got? By default it pre-allcoates a lot of ring buffers
[14:42:30] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:47:16] *** tru_tru <tru_tru!~tru@157.99.90.140> has joined #illumos
[14:49:11] <EisNerd> pwinder: we have latest and this esx box 512GB
[14:51:40] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has joined #illumos
[14:54:18] <jimklimov> EisNerd: tsoome: Recently read about UEFI state and plans for 2020 - that article included Intel dropping UEFI vs BIOS compat shims support for new CPUs (and firmwares?) from 2020 on, and so all addon cards have to carry UEFI-compatible firmwares.
[14:54:54] <jimklimov> There was no info in that article about AMD and card vendors' stance on this.
[14:55:27] <EisNerd> good point, but afaik they do, I'll try to disable one to see if it is one that is bad
[14:57:20] <jimklimov> EisNerd: likewise, I saw yet no info if VMWare ahs any say about this, in any manner.
[15:01:09] <EisNerd> hm looks like one of the mlx cards is the bad guy
[15:01:41] <EisNerd> sadly the more important one
[15:02:42] <EisNerd> so I'll try to clean it and reload the fw with all defaults
[15:03:35] <EisNerd> pwinder: btw still would like to have some prebuild bind to check nvmens management
[15:03:38] *** phyre <phyre!~phyre___@78.30.23.107> has quit IRC (Remote host closed the connection)
[15:03:57] <EisNerd> bins, sorry new keyboard
[15:05:40] <pwinder> EisNerd: ns management was never tested on NVMe which allowed it - my drives didn't permit it. I can give you a binary on v. latest illumos source. But no warranty :-)
[15:07:41] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has quit IRC (Quit: nbhauke)
[15:07:49] <EisNerd> pwinder: If you can give me the binary, I can try to test it I think it have some readonly operations, so I can start with them
[15:08:14] <EisNerd> pwinder: btw how stable is the mlx driver so far?
[15:09:00] <EisNerd> As I think about putting one of the mlx cards in the OI box to replace the xl710 temporarily to see if this is the problem
[15:10:58] <EisNerd> but first I would have to dig into the mlx firmware mgmt manuals, to se how I can turn my mlx back into a virgin
[15:11:39] <EisNerd> I didi this before, but I forgot how and with which tools, there are so much, with so much options
[15:12:30] <EisNerd> smart nics are great as long as they work, but the hell if they start making trouble
[15:12:44] <EisNerd> ok see you, have to get my car from the garage
[15:16:11] <pwinder> EisNerd: Since this commit "12481 attempting to change MTU ...." it has been good. We are on connectx5
[15:20:20] *** xtuh_ <xtuh_!~xtuh@88.198.230.182> has quit IRC (Quit: Leaving)
[15:25:25] <igork1> pwinder: did you test 100g mellanox connctX5 ?
[15:25:33] *** igork1 is now known as igork
[15:26:54] <pwinder> Yes .... no performance testing though. I'm experience (eg cxgbe 100G) to get close to line rate need to increase the number of interruptt vectors
[15:27:14] <pwinder> s/I'm/In my/
[15:27:37] <igork> i'm interested in iperf between 2 mellanox 100g
[15:28:00] <igork> i have tested 25g - i can see 20-23Gbit
[15:28:30] <igork> direct cable link and over mellanox switch - all fine
[15:28:44] <pwinder> That's similar to my experience
[15:28:45] <igork> also, i have tested aggr
[15:29:22] <igork> also i have tested connect between connectX4 and X5 - all fine too
[15:34:25] <tsoome> jimklimov yes, bios, and that is including vis bios, is doomed.
[15:34:41] <tsoome> s/vis/vga/
[15:41:32] *** leons <leons!leonsiscur@gateway/shell/matrix.org/x-rwpixsoffxfriyfc> has joined #illumos
[15:50:09] *** vgusev_ <vgusev_!~vgusev@188.187.60.230> has joined #illumos
[15:52:35] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Remote host closed the connection)
[15:53:09] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[15:53:37] *** vgusev <vgusev!~vgusev@188.187.60.230> has quit IRC (Ping timeout: 260 seconds)
[15:59:37] *** ibenn <ibenn!~benn@HSI-KBW-095-208-237-062.hsi5.kabel-badenwuerttemberg.de> has joined #illumos
[16:08:38] *** tsoome <tsoome!~tsoome@12d4-01e3-bff1-a8d5-2f80-4a40-07d0-2001.sta.estpak.ee> has quit IRC (Quit: Leaving)
[16:08:47] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[16:10:20] *** ldepandis <ldepandis!~ldepandis@unaffiliated/ldepandis> has quit IRC (Remote host closed the connection)
[16:18:03] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[16:23:07] *** jzu_ <jzu_!~jussi@pupu.jus.si> has quit IRC (Quit: leaving)
[16:23:34] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[16:28:37] *** rzezeski <rzezeski!uid151901@gateway/web/irccloud.com/x-wqqocphzlrhjbrys> has quit IRC (Quit: Connection closed for inactivity)
[16:28:43] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[16:53:28] <gitomat> [illumos-gate] 12696 libpri: leftover bits from 12454 -- Toomas Soome <tsoome at me dot com>
[16:53:46] *** jzu_ <jzu_!~jussi@pupu.jus.si> has joined #illumos
[17:00:00] <gitomat> [illumos-gate] 12642 n2rng: NULL pointer errors -- Toomas Soome <tsoome at me dot com>
[17:16:28] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[17:28:15] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Remote host closed the connection)
[17:28:43] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[17:28:50] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[17:29:34] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Remote host closed the connection)
[17:30:18] <gitomat> [illumos-gate] 12701 segspt_minfree needs right-sizing -- John Levon <john.levon at joyent dot com>
[17:31:05] <neirac> toastersonerson thanks for pushing the changes for nomad on docker term, now they are in upstream!, less work , now Smithx10 going to push gopsutil :)
[17:33:50] <Smithx10> probably next week
[17:34:02] <toastersonerson> neirac (IRC): It's still going to take a while until clients have rebased and updated their vendor deps. but yeah its in the pipeline.
[17:42:17] *** neuroserve <neuroserve!~toens@ip-95-222-232-201.hsi15.unitymediagroup.de> has quit IRC (Ping timeout: 258 seconds)
[17:42:28] *** neirac <neirac!~neirac@pc-4-149-45-190.cm.vtr.net> has quit IRC (Quit: ZNC 1.8.0 - https://znc.in)
[17:47:40] <LeftWing> https://www.illumos.org/issues/12706
[17:47:42] *** neuroserve <neuroserve!~toens@ip-88-152-243-25.hsi03.unitymediagroup.de> has joined #illumos
[17:48:09] <LeftWing> (Thoughts?)
[17:48:33] *** jimklimov <jimklimov!~jimklimov@78.80.224.132> has quit IRC (Quit: Leaving.)
[17:50:43] <jlevon> LeftWing: sounds fine though I'd kind of like to burn ttydefs with fire altogether :)
[17:50:44] <tsoome> I have some hip around it too
[17:50:54] <tsoome> wip*
[17:51:11] <tsoome> stupid autocorrect:D
[17:52:36] <tsoome> but mine is in *very* early state
[17:52:55] <toastersonerson> Linux assumes something even higher by default on modenr devices.
[17:53:37] *** pwinder <pwinder!~pwinder@86.11.191.8> has quit IRC (Quit: This computer has gone to sleep)
[17:54:04] <tsoome> LeftWing, https://github.com/tsoome/illumos-gate/commit/d2c0b48dbdff73a55c3a9829b409e4fb3f917c1a
[17:55:10] <tsoome> basically, my though was, if we land in with console on serial, we actually can trust the tty*-mode line, so the etymon should pick it up and use it. but my idea is not quite finished.
[17:55:43] <tsoome> s/etymon/ttymon/
[17:57:35] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[17:58:25] <wilbury> regadring my kernel coding exercise: i'm trying to a) get a grasp of kernel, b) port nmdm driver for usage with bhyve
[18:01:14] *** neirac <neirac!~neirac@pc-4-149-45-190.cm.vtr.net> has joined #illumos
[18:02:12] <toastersonerson> Question is it nasty if I would to expose mount(2) to golang?
[18:02:23] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has quit IRC (Remote host closed the connection)
[18:02:59] *** jollyd <jollyd!~alarcher@2a01:cb1d:611:bb00:8820:7cbd:8965:af56> has joined #illumos
[18:03:11] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has joined #illumos
[18:08:22] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[18:08:58] *** Kurlon <Kurlon!~Kurlon@167-48.gwi.net> has joined #illumos
[18:09:07] <LeftWing> tsoome: Yes, I thought about that too -- but I'm not sure we wouldn't want to make it so that you could override the label anyway
[18:10:09] <tsoome> yes, the ttymon command line (in smf) should have final word, but if not set, just pick property value from boot.
[18:10:23] <LeftWing> Yeah I think that would be fine
[18:10:37] <LeftWing> In the interim I think 115200 is much more likely than 9600 for now haha
[18:10:49] <tsoome> yes
[18:13:52] <tsoome> https://paste.ec/paste/YSQiLNCc#qiS0RAkguNUNThRQf3nfej3RosoVHK2SZsUNTkLlyUr
[18:14:12] <toastersonerson> Could we actually expose mount(2) to golang? It's a systemcall so I assume it is not present in libc?
[18:18:39] <LeftWing> toastersonerson: System calls documented in section 2 are nonetheless exposed with at least a stub in libc
[18:19:12] <toastersonerson> ah that should work with the libc binding generator. thanks
[18:19:28] <LeftWing> $ /usr/bin/nm /usr/lib/libc.so.1 | grep '|mount$'
[18:19:30] <LeftWing> [9499] | 964784| 30|FUNC |GLOB |3 |15 |mount
[18:20:33] <LeftWing> A GLOBally visible FUNCtion symbol
[18:20:59] <toastersonerson> ah thats something nice to add to my toolbelt then :)
[18:25:30] <jbk> LeftWing: (I just added to the ticket) I actually disagree about 115200 -- I've only seen 1 device in the past 20 years that doesn't default to 9600
[18:25:40] <jbk> and linux and fbsd all seem to default to 9600 as well
[18:25:43] <LeftWing> Whaaat
[18:26:34] <LeftWing> When you say "device" though what do you mean
[18:26:47] <LeftWing> I don't think it matters what random embedded systems default to
[18:26:52] <LeftWing> for their UARTs
[18:27:30] <jbk> basically anything and everything found in a datacenter
[18:27:39] <jbk> whether it be servers, switches, firewalls, routers
[18:28:11] <jbk> caching web proxies/filters
[18:28:34] <jollyd> tsoome: gcc 10 coming to you tomorrow
[18:29:14] <andyf> That's true.. 9600 is the rule for all that stuff.
[18:29:43] <andyf> I did an NTP server recently that defaulted to 115200, and some McAfee Web Gateway boxes that were at 19200, but they are the exception in a DC
[18:29:54] <LeftWing> I guess the real question, though, is: what are people actually trying to have their equipment do
[18:30:08] <toastersonerson> And mount requires 8 arguments acording to it's man page.... so much to adding that sycall now. I seem to need to shell out for the moment.
[18:30:10] <andyf> but I usually use SOL to access illumos boxes, and for them I want a higher speed
[18:30:15] <LeftWing> Right
[18:30:34] <LeftWing> 9600 is peril; are people not generally trying to use something better for their system console?
[18:30:50] <LeftWing> We defaulted to 115200 in SmartOS
[18:30:52] <andyf> tbh, system console is for recovery only really
[18:31:01] <LeftWing> andyf: Or the debugger
[18:31:20] <andyf> Good point - I was still thinking about the switches, routers, proxies...
[18:32:20] <Smithx10> Or..... Packer!!!!
[18:32:21] <Smithx10> :P
[18:32:31] <Smithx10> Building images over Serial baby! oh yea!
[18:33:01] *** rzezeski <rzezeski!uid151901@gateway/web/irccloud.com/x-mgbqpvaswpcmghyw> has joined #illumos
[18:33:07] <LeftWing> for e.g., https://omniosce.org/setup/serial_console.html has you set the console to 115200 using sttydefs
[18:33:38] <tsoome> if we pick tty*-mode, it defaults to 9600 and hence the default is fine. but also, if someone is setting serial to something else, it should be possible to accept that change without having PhD in SunOS:D
[18:34:18] <andyf> sed -i -e '/^console:/s/9600/115200/g' $ALTROOT/etc/ttydefs
[18:34:29] <LeftWing> lol
[18:34:32] <LeftWing> Where is that
[18:34:33] <andyf> That seems to be what I had kayak do during installation of my newer boxes
[18:34:36] <LeftWing> heh
[18:35:08] <LeftWing> andyf: Did you also do tty*-mode stuff in the loader config?
[18:35:45] <andyf> yes, in /boot/conf.d/serial
[18:35:52] <LeftWing> Right
[18:36:27] <andyf> I have to head out, back later
[18:36:51] <LeftWing> Thanks for your input andyf
[18:41:46] <tsoome> jollyd, thats cool.
[18:41:57] <wilbury> so i'm now able to create and delete minor devices
[19:20:32] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 256 seconds)
[19:22:22] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[19:39:31] *** yuripv <yuripv!~yuripv@85.175.17.187> has quit IRC (Remote host closed the connection)
[19:41:10] *** Teknix <Teknix!~pds@69.41.134.110> has quit IRC (Ping timeout: 272 seconds)
[19:41:42] *** Teknix <Teknix!~pds@69.41.134.110> has joined #illumos
[19:47:27] <igork> question about snprintf() - does it include +1 end of line terminated symbol to length parameter or i need allocate (+1) of buffer by myself ?
[19:48:54] <jbk> you pass the size of the buffer, and it will truncate the output at size-1 so the result can always be NUL-terminated
[19:49:08] <igork> https://paste.dilos.org/?7a47236b3b4e4b32#FVMKrehYXy5oxYCMyUb5tJIToxN6tOb2+mXKxoN3q1M=
[19:49:10] <jbk> or rather, it will always NUL-terminate the result
[19:49:15] <igork> it is from man page
[19:49:44] <jbk> (unless you pass NULL and 0)
[19:49:45] <jbk> yes
[19:50:05] <jbk> n is the size of the buffer
[19:50:34] <jbk> it will write at most n-1 bytes from processing the parameters, and then always add NUL at the end
[19:51:43] <jbk> if you're using x = snprintf(NULL, 0, ...) to determine the size
[19:51:51] <jbk> you need to add 1 to the results (or just use asprintf)
[19:53:34] *** pwinder <pwinder!~pwinder@86.11.191.8> has joined #illumos
[19:53:59] *** pwinder <pwinder!~pwinder@86.11.191.8> has quit IRC (Client Quit)
[19:58:30] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[19:59:10] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Client Quit)
[20:11:21] *** ibenn <ibenn!~benn@HSI-KBW-095-208-237-062.hsi5.kabel-badenwuerttemberg.de> has quit IRC (Quit: Leaving)
[20:19:51] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[20:23:36] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has joined #illumos
[20:26:17] <LeftWing> asprintf() is almost always the right answer
[20:27:25] <jbk> yes
[20:44:57] <igork> jbk: thanks
[20:45:16] <igork> LeftWing too
[20:54:44] <jbk> rmustacc: or rzezeski: (I'm guessing either one of you would know) when calling a ring tx (I suspect mc_tx would be the same) -- is it permissible if asserting flow control (i.e. returning an mblk_t) that it could be a different mblk_t than what was given (as long as the content is the same)
[20:55:57] <jbk> e.g. you do a msgpullup, but for some other reason can't tx, but either can return the original mblk_t (and have to do the msgpullup() again when mac tries to tx again), or return the pulled up mblk_t so that is the one that is retried
[20:56:19] <jbk> (i suspect 'yes', but it's not clear about that that I could find in the man pages)
[20:56:39] <rzezeski> hmmm, I'm not sure if we actually say anything in that regard, but seems like a good question, my memory says we typically return the original mp, but not sure that is necessary
[21:07:34] <rmustacc> I would be hesitant to assert one way or the other.
[21:08:42] <rmustacc> jbk: One of the major things there is that most folks aren't correctly going to copy any requests for TSO or checksum offload, etc. when doing that.
[21:09:14] <rmustacc> msgpullup, for example, doesn't try to copy (nor really should it) things like TSO or checksums.
[21:09:37] <jollyd> tsoome: I need to check again the illumos patchset but if you want you can push sparc fixes to the same PR https://github.com/OpenIndiana/oi-userland/pull/5712/ or as a follow up. I tried to preserve the i386 variant directives from gcc-9, let me know if I missed any. I'll add the testsuite results as soon as they are ready.
[21:09:39] <rmustacc> So if you returned something different without actually copying the datablk fields and more, when that packet ends up going back to the driver you'll probably end up doing the rong thing.
[21:10:20] <rmustacc> So based on that, I'd say, no you probably should not return a modified block and should return the original where possible. But obviously, details and context matter.
[21:11:14] <rmustacc> Other than being confusing, is there an actual bug in ixgbe that we're trying to track down that's motivating rewriting things?
[21:11:56] <jbk> basically just anyone someone looks at it, they lose days trying to figure out what the hell is going on
[21:12:12] <jbk> err anytime
[21:12:41] <jbk> almost all due to completely unnecessary work
[21:13:29] <rmustacc> OK. Just please get a really good regression suite put together then.
[21:14:22] *** jimklimov <jimklimov!~jimklimov@ip-86-49-254-26.net.upcbroadband.cz> has joined #illumos
[21:16:21] <rzezeski> Oh yea, great point about flags/cksum values, if those are not properly copied then things break and it's a pain to debug (ask me how I know).
[21:16:39] <rzezeski> I would definitely look at what the drivers already do and stick to that probably
[21:16:55] <rzezeski> Also, I'm not sure rewriting ixgbe is going to make things any clearer
[21:17:08] <rzezeski> But I could be wrong
[21:35:52] *** KeiraT- <KeiraT-!~k4ra@gateway/tor-sasl/k4ra> has joined #illumos
[21:36:43] *** KeiraT <KeiraT!~k4ra@gateway/tor-sasl/k4ra> has quit IRC (Ping timeout: 240 seconds)
[21:39:08] *** KeiraT- is now known as KeiraT
[21:40:25] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 264 seconds)
[21:41:57] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[21:43:42] *** BOKALDO <BOKALDO!~BOKALDO@87.110.146.209> has quit IRC (Quit: Leaving)
[22:10:39] <KungFuJesus> LeftWing: The ALIX and APU boards from PC engines are at 115200 baud
[22:11:24] <KungFuJesus> and you need a decent cable with CTS pins to not get gobbly guck on your serial terminal
[22:11:37] <KungFuJesus> or RTS or whatever they call it
[22:13:56] <LeftWing> I don't think that's really true
[22:14:42] <LeftWing> Well, rather I believe the quality cable part
[22:14:54] <LeftWing> Higher speeds for single ended serial can get harder over longer cables
[22:15:07] <LeftWing> But I don't imagine you would need hardware flow control to avoid a mess per se
[22:20:02] *** phyre <phyre!~phyre___@78.30.23.107> has joined #illumos
[22:31:16] *** blackwood821 <blackwood821!~blackwood@c-174-49-16-176.hsd1.tn.comcast.net> has joined #illumos
[22:32:07] <blackwood821> Is it necessary to run `devfsadm -i iscsi` after adding an iSCSI target and enabling discovery? I've never done that and just now noticed it in the oracle docs (step 4): https://docs.oracle.com/cd/E23823_01/html/817-5093/fncpi.html#fqnlk. But it seems to work fine without it.
[22:33:06] *** nbhauke <nbhauke!~hauke@55d41097.access.ecotel.net> has quit IRC (Quit: nbhauke)
[22:38:39] <LeftWing> I don't think it should be
[22:42:23] <blackwood821> Ok, is that something illumos handles automatically?
[22:44:36] *** wacki <wacki!~wacki@i577B0946.versanet.de> has left #illumos
[22:46:16] <LeftWing> I think so, but I don't recall all the details myself. I'm sure I've also added iSCSI targets and just had them show up without other poking, but it was a while ago.
[22:47:37] <blackwood821> Ok thanks
[22:58:42] *** Kurlon_ <Kurlon_!~Kurlon@cpe-67-253-136-97.rochester.res.rr.com> has joined #illumos
[23:01:42] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[23:02:34] *** Kurlon <Kurlon!~Kurlon@167-48.gwi.net> has quit IRC (Ping timeout: 272 seconds)
[23:18:55] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[23:42:45] <LeftWing> Good grief, I wondered how "autobaud" worked and... now I know
[23:43:56] <LeftWing> https://github.com/illumos/illumos-gate/blob/master/usr/src/cmd/ttymon/tmautobaud.c
[23:44:50] <LeftWing> It appears to put the port in 2400 baud mode and then look for we believe a CR looks like at some other speed
[23:53:03] <v_a_b> LeftWing I like that other proposal to start at 115200 much better ;-)
[23:53:19] <LeftWing> Ha!
[23:53:46] <LeftWing> I'm about half way to plumbing up looking at "ttyX-mode" for the speed at the moment
[23:54:07] <v_a_b> I'll test it as soon as it is in OmniOS bloody!
[23:58:50] *** phyre <phyre!~phyre___@78.30.23.107> has quit IRC (Remote host closed the connection)
top

   May 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 | >