Switch to DuckDuckGo Search
   November 10, 2019  
< | 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 | >

NOTICE: This channel is no longer actively logged.

Toggle Join/Part | bottom
[00:04:57] <despair86> basically the graphics adapter is causing the whole system to fall over
[00:05:11] <despair86> and i can't even disconnect it because the system has no other GPU
[00:18:22] *** ramcneal <ramcneal!~ramcneal@> has quit IRC (Ping timeout: 245 seconds)
[00:27:38] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 240 seconds)
[00:30:22] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #illumos
[00:48:43] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[00:59:24] <gitomat> [illumos-gate] 11818 IPMI topo plugin shouldn't return data from unavailable sensors -- Matthias Scheler <matthias.scheler at wdc dot com>
[00:59:25] <gitomat> [illumos-gate] 11906 Add line drawing characters to ptree(1) -- Jason King <jason.king at joyent dot com>
[01:09:44] *** Kruppt <Kruppt!~Kruppt@> has quit IRC (Quit: Leaving)
[01:52:58] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has quit IRC (Quit: Leaving)
[01:53:38] *** sarvet <sarvet!~o.o@i59F63875.versanet.de> has quit IRC (Ping timeout: 240 seconds)
[01:58:16] <jbk> *sigh* so that smatch fix didn't end up working -- or rather, once you applied it w/ other changes, it's enough to trigger the same error
[02:01:04] <Reinhilde> wheeww...
[02:05:50] *** fanta1 <fanta1!~fanta1@p200300F76BCAF500406563D1C0771889.dip0.t-ipconnect.de> has quit IRC (Quit: fanta1)
[02:06:23] <LeftWing> Alright I think I have this xhci disk DMA problem sorted out
[02:20:00] *** ed209 <ed209!~ed209@> has quit IRC (Remote host closed the connection)
[02:20:07] *** ed209 <ed209!~ed209@> has joined #illumos
[02:21:42] <LeftWing> https://www.illumos.org/issues/11952
[02:27:07] <Reinhilde> LeftWing: could you remind me the keyword i needed to find|grep for in gate to find my pet SIGBUS ENOMEM problem?
[02:27:35] <LeftWing> err
[02:27:40] <LeftWing> FC_MAKE_ERR() I think?
[02:27:47] <LeftWing> Also BUS_OBJERR / FC_OBJERR etc?
[02:28:08] <LeftWing> I'd start from the place in trap() that we assign BUS_OBJERR to the place you then see it in the ELF note
[02:28:12] <LeftWing> And work backwards
[02:29:01] <Reinhilde> LeftWing: where is trap()?
[02:29:43] <LeftWing> I think it's in trap.c
[02:29:50] <LeftWing> Under uts/i86pc
[02:30:17] <LeftWing> I recommend "cscope"!
[02:30:41] <Reinhilde> and do I need to be doing this in a copy of omnios' fork of gate or can I compile a kernel in a copy of the highest upstream and expect it to Just Work™®© on omnios?
[02:31:39] <LeftWing> I'd probably start with a kernel that's close to the one you have; e.g., start from whatever illumos-omnios.git is used to build the thing you're already running
[02:33:07] <Reinhilde> >.>
[03:05:56] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)
[03:23:55] <LeftWing> https://code.illumos.org/c/illumos-gate/+/163
[03:24:40] <Reinhilde> LeftWing: ?
[03:24:58] <LeftWing> That is unrelated to your thing -- just a change I am putting out for review
[03:30:55] <Reinhilde> LeftWing: oh so you have a largefile bug
[03:31:49] <LeftWing> Not really
[03:31:58] <Reinhilde> LeftWing: it looks largefile to mee
[03:32:03] <Reinhilde> but i'm just a novice
[03:32:07] <Reinhilde> an end user as they call them
[03:32:28] <LeftWing> Yes but it's not files, per se; it's the SCSI stack in the kernel interacting with our USB mass storage layer
[03:36:36] <Reinhilde> 926 bzero(&siginfo, sizeof (siginfo));
[03:36:38] <Reinhilde> 927 siginfo.si_addr = addr;
[03:36:40] <Reinhilde> 928 switch (FC_CODE(res)) {
[03:36:42] <Reinhilde> ...
[03:36:50] <Reinhilde> 940 case FC_OBJERR:
[03:36:52] <Reinhilde> 941 if ((siginfo.si_errno = FC_ERRNO(res)) != EINTR) {
[03:36:54] <Reinhilde> 942 siginfo.si_signo = SIGBUS;
[03:36:56] <Reinhilde> 943 siginfo.si_code = BUS_OBJERR;
[03:37:01] <Reinhilde> this is not the bug, but it is where the signal is being set
[03:37:13] <LeftWing> Yes
[03:37:41] <Reinhilde> it's neighbors with opteron erratum 91
[03:38:07] <LeftWing> Well you're deep in the messy innards
[03:38:34] <Reinhilde> I can see that.
[03:39:24] <Reinhilde> i assume this is a T_PGFLT+USER
[03:39:28] <Reinhilde> though i may be wrong
[03:42:42] <Reinhilde> i mean i'm probably blind
[03:43:09] <Reinhilde> how do i set faultdebug to 1?
[03:43:58] <Reinhilde> not that i really should
[03:44:45] <Reinhilde> what's FTRACE_2?
[04:02:12] <jbk> electric bogaloo?
[04:30:59] <Reinhilde> jbk: ...
[04:32:34] <rmustacc> An old tracing framework in the kernel. From a pre-DTrace day.
[04:33:41] <Reinhilde> rmustacc: oh dear
[04:34:48] <rmustacc> Don't worry about it, it's not in your way or a problem.
[04:36:49] <Reinhilde> rmustacc: the issue is I want to get something out of it
[04:36:53] <Reinhilde> does it feed into dtrace nowadays?
[04:36:59] *** trisk <trisk!~trisk@2601:196:4700:3f0b:8:20ff:fe3e:890e> has quit IRC (Ping timeout: 276 seconds)
[04:39:41] <rmustacc> No.
[04:57:11] <LeftWing> You could adjust it so that it does, I imagine!
[04:59:11] <rmustacc> They could easily all become sdt probes.
[05:00:52] <rmustacc> Which is probably better than trying to use uadmin.
[05:25:25] <Reinhilde> ow
[06:17:22] *** trisk <trisk!~trisk@2601:196:4700:3f0b:8:20ff:fe3e:890e> has joined #illumos
[06:22:38] <despair86> triggered a stupid bug in a brand-new ryzen board
[06:23:02] <despair86> > update EFI ROM and monitor program to latest image
[06:23:12] <despair86> > update completes normally
[06:23:19] <despair86> > does not boot again
[06:33:51] <despair86> https://cdn.discordapp.com/attachments/638594330403274763/642958392800837642/image.png
[08:09:39] *** fanta1 <fanta1!~fanta1@p200300F76BCC3E00C424931F9AF5734E.dip0.t-ipconnect.de> has joined #illumos
[09:05:09] <gitomat> [illumos-gate] 11923 loader: zfs.c cstyle cleanup -- Toomas Soome <tsoome at me dot com>
[10:15:40] *** arnoldoree <arnoldoree!~arnoldore@ranoldoree.plus.com> has joined #illumos
[11:20:00] *** ed209 <ed209!~ed209@> has quit IRC (Remote host closed the connection)
[11:20:07] *** ed209 <ed209!~ed209@> has joined #illumos
[11:53:24] *** sengir <sengir!~sh42@fsf/member/zeroSignal> has quit IRC (Ping timeout: 268 seconds)
[12:09:39] *** sengir <sengir!~sh42@> has joined #illumos
[12:19:46] *** arnold_oree <arnold_oree!~arnoldore@ranoldoree.plus.com> has joined #illumos
[12:32:25] *** Yogurt <Yogurt!~Yogurt@c-73-189-45-147.hsd1.ca.comcast.net> has joined #illumos
[12:36:36] *** Yogurt <Yogurt!~Yogurt@c-73-189-45-147.hsd1.ca.comcast.net> has quit IRC (Ping timeout: 240 seconds)
[12:43:39] *** neirac <neirac!~cneira@pc-184-104-160-190.cm.vtr.net> has quit IRC (Quit: Leaving)
[12:51:52] *** amrmesh <amrmesh!~Thunderbi@> has joined #illumos
[12:52:26] *** amrfrsh <amrfrsh!~Thunderbi@> has quit IRC (Ping timeout: 240 seconds)
[12:52:27] *** amrmesh is now known as amrfrsh
[13:16:00] *** amrmesh <amrmesh!~Thunderbi@> has joined #illumos
[13:17:29] *** amrfrsh <amrfrsh!~Thunderbi@> has quit IRC (Ping timeout: 252 seconds)
[13:17:29] *** amrmesh is now known as amrfrsh
[13:59:49] *** arnold_oree <arnold_oree!~arnoldore@ranoldoree.plus.com> has quit IRC (Ping timeout: 268 seconds)
[14:29:12] *** Kruppt <Kruppt!~Kruppt@> has joined #illumos
[14:46:47] *** andy_js <andy_js!~andy@97e29e78.skybroadband.com> has joined #illumos
[14:54:32] *** eki <eki!~eki@dsl-hkibng41-567327-143.dhcp.inet.fi> has quit IRC (Quit: leaving)
[15:26:03] *** eki <eki!~eki@dsl-hkibng41-567327-143.dhcp.inet.fi> has joined #illumos
[15:59:29] *** Teknix <Teknix!~pds@> has quit IRC (Ping timeout: 250 seconds)
[15:59:49] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[16:01:31] *** Teknix <Teknix!~pds@> has joined #illumos
[17:22:30] <despair86> has anyone else had problems running solaris or illumos on AMD B450 chipset platforms?
[17:23:20] <despair86> at best the kernel triggers a PCI fatal error on the PCI bridge where an nvidia gpu is located
[17:25:45] <despair86> have been struggling with a new build for the past four days now
[17:31:48] *** Kruppt <Kruppt!~Kruppt@> has quit IRC (Quit: Leaving)
[17:45:31] <v_a_b> igork I haven't forgotten you, but I need to do some other things before I can set up the Solaris 11.4 SPARC machine...
[18:08:07] <igork> v_a_b: thanks :) still waiting your report
[18:08:32] <igork> oracle dropped support of T5220 from 11.4
[18:27:04] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Read error: Connection reset by peer)
[18:27:24] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[18:35:14] *** Teknix <Teknix!~pds@> has quit IRC (Ping timeout: 240 seconds)
[18:39:49] *** Qatz <Qatz!~db@2601:187:8400:5::42d> has quit IRC (Read error: Connection reset by peer)
[18:40:01] *** Qatz <Qatz!~db@2601:187:8400:5::42d> has joined #illumos
[18:45:47] *** Teknix <Teknix!~pds@> has joined #illumos
[18:46:01] *** fanta1 <fanta1!~fanta1@p200300F76BCC3E00C424931F9AF5734E.dip0.t-ipconnect.de> has quit IRC (Quit: fanta1)
[18:53:12] <v_a_b> Yes, unfortunately. I have a T4-2 that needs to be set up. It has power, and I am currently configuring the SP :-)
[18:56:24] <igork> ok, will wait your report
[19:19:18] *** fanta1 <fanta1!~fanta1@p200300F76BCC3E00C424931F9AF5734E.dip0.t-ipconnect.de> has joined #illumos
[19:21:02] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has joined #illumos
[19:29:19] *** pwinder <pwinder!~pwinder@> has quit IRC (Quit: This computer has gone to sleep)
[19:48:00] *** fanta1 <fanta1!~fanta1@p200300F76BCC3E00C424931F9AF5734E.dip0.t-ipconnect.de> has quit IRC (Quit: fanta1)
[20:23:52] *** despair86_ <despair86_!~despair86@75-60-137-0.lightspeed.rcsntx.sbcglobal.net> has joined #illumos
[20:25:34] *** despair86 <despair86!~despair86@> has quit IRC (Ping timeout: 265 seconds)
[20:25:37] *** despair86_ is now known as despair86
[20:45:14] *** Teknix <Teknix!~pds@> has quit IRC (Ping timeout: 240 seconds)
[20:47:10] *** Teknix <Teknix!~pds@> has joined #illumos
[21:03:41] *** DrLou <DrLou!~DrLou@c-73-33-92-169.hsd1.nj.comcast.net> has quit IRC (Ping timeout: 276 seconds)
[21:10:27] *** arnold_oree <arnold_oree!~arnoldore@ranoldoree.plus.com> has joined #illumos
[21:11:11] *** arnold_oree <arnold_oree!~arnoldore@ranoldoree.plus.com> has quit IRC (Remote host closed the connection)
[21:20:00] *** ed209 <ed209!~ed209@> has quit IRC (Remote host closed the connection)
[21:20:06] *** ed209 <ed209!~ed209@> has joined #illumos
[21:27:16] <v_a_b> igork OK, so now I have the T4-2 up and running with a test installation of Solaris 11.4.
[21:27:51] <v_a_b> I have a choice of gcc 7.3.0 and 9.1.0 -- no gcc 6 available. I'll try both versions. :-)
[21:30:43] <igork> gcc7
[21:33:20] <v_a_b> Here's gcc9:
[21:33:23] <v_a_b> <t4-2-cd:/tmp,28# g++ -g -m64 -o t6g t.cpp
[21:33:23] <v_a_b> <t4-2-cd:/tmp,29# ./t6g
[21:33:23] <v_a_b> Division by zero condition!
[21:33:31] <v_a_b> gcc 7 coming up :-)
[21:33:32] <igork> cool
[21:33:39] <igork> interest
[21:33:52] <igork> we have issue with illumos ld on sparc
[21:35:55] *** amrmesh <amrmesh!~Thunderbi@> has joined #illumos
[21:36:02] *** amrfrsh <amrfrsh!~Thunderbi@> has quit IRC (Ping timeout: 240 seconds)
[21:36:02] *** amrmesh is now known as amrfrsh
[21:38:13] <Reinhilde> igork: does gnu ld work
[21:38:26] <Reinhilde> kinda curiousn
[21:38:37] <Reinhilde> oh wait this is solaris you're talking not illumos
[21:38:41] <igork> yes, gnu ld produce correct binary on sparc
[21:38:48] <Reinhilde> igork: wha, woah
[21:39:12] <Reinhilde> then why are you using SLE
[21:39:20] <igork> ?
[21:39:45] <Reinhilde> illumos ld calls itself "solaris link editor"
[21:39:52] <Reinhilde> i guess I should call it ILE
[21:40:18] <v_a_b> Same result for gcc 7.3.0, output is "Division by zero condition!"
[21:40:33] <igork> v_a_b: thanks for your testing!
[21:40:36] <v_a_b> NP
[21:41:06] <v_a_b> I will power down the machine now if you don't need any more tests.
[21:41:43] <v_a_b> For completeness:
[21:41:48] <v_a_b> <t4-2-cd:/tmp,45# uname -a
[21:41:48] <v_a_b> SunOS t4-2-cd 5.11 sun4v sparc sun4v
[21:41:48] <v_a_b> <t4-2-cd:/tmp,46# ld --version
[21:41:48] <v_a_b> ld: Software Generation Utilities - Solaris Link Editors: 5.11-1.3196
[21:42:21] <Reinhilde> wew
[21:42:52] <Reinhilde> v_a_b: have you found a sun4v version of illumos to install on the same machine? if so, have you tested that
[21:44:08] <igork> v_a_b: thanks for testing, i just need confirmation - i'm not crazy :)
[21:44:42] <v_a_b> Not yet. I will set up a control domain and then I'll try a fully virtualized LDOM with v9os. I have a v9os LDOM running on a T5120 just fine. Also, I promised Peter at FOSDEM that I would try and run Tribblix in an LDOM.
[21:44:58] <v_a_b> igork you're welcome. It forced me to finally set up my T4-2 which I had wanted to do for some weeks.
[21:45:15] <v_a_b> Only 512 GB memory ;-)
[21:45:26] <Reinhilde> so v9os would be a sun4v version of illumos?
[21:45:27] <igork> v_a_b: give me dilos LDOM on it :)
[21:45:47] <v_a_b> igork Why not? But we have to set it up another day.
[21:45:55] <igork> np "_
[21:46:13] <Reinhilde> did any of the companies ever come out with a laptop whose brain was SPARC
[21:46:22] <v_a_b> Yes, I have one sitting here.
[21:46:26] <igork> v_a_b: i'll try to finish latest iso for sparc and we can test it
[21:46:30] <Reinhilde> v_a_b: what?!
[21:46:36] <v_a_b> It is a Tadpole SPARCbook.
[21:46:36] <Reinhilde> you've got to be joking
[21:46:48] <v_a_b> It is basically an SS5 in a Thinkpad chassis.
[21:46:54] <Reinhilde> SS5=?
[21:47:07] <tsoome> sparcstation 5
[21:47:27] <Reinhilde> ah
[21:47:32] <Reinhilde> v_a_b: what's the battery life like on it?
[21:47:46] <jbk> tsoome: btw, I should be done updating https://illumos.org/rb/r/2446/ if you want to look again :)
[21:47:55] <v_a_b> http://www.computinghistory.org.uk/det/32324/Tadpole-SPARCbook-3/
[21:48:06] <Reinhilde> tsoome: i'm like a fat kid in a candy store when it comes to weird Unix workstations for which open source OSes are available
[21:48:06] <v_a_b> I haven't powered it on for 1+ yr.
[21:48:13] <Reinhilde> v_a_b: You should solve that.
[21:48:39] <jbk> ... of course smatch then decided to go all non-determinstic afterwards... :)
[21:48:43] <Reinhilde> so it's a good thing I'm broke or I'd probably have a house full of Sun Blades
[21:48:51] <v_a_b> I was given an original SparcStation Voyager when the Sun museum at the Sun office in Frankfurt was closed, but it was stolen from my car the same night. :-(
[21:48:53] <jbk> I thought i had a workaround, but then testing it with other changes, it freaks out again...
[21:49:06] <v_a_b> igork The only limiting factor is my time :-(
[21:49:16] <tsoome> ya, sorry, I forgot to press the button:D
[21:49:22] <igork> v_a_b: np
[21:49:36] <Reinhilde> v_a_b: if you figure out who stole it you COULD ask tem if they still have it and if they'll charge a ransom
[21:49:45] <jbk> tsoome: np, didn't want to RTI it if you hadn't had a chance ot make sure you were happy w/ the updates
[21:50:20] <Reinhilde> Shame, the sparcbook only goes up to SunOS 5.6
[21:50:21] <v_a_b> I did check ebay the next few weeks but never saw anything. Frankfurt police closed the case after a month b/c they couldn't do anything.
[21:50:21] <jbk> heh i'd have to imagine whoever stole it was probably disappointed considering it has somewhat limited resale value
[21:51:00] <v_a_b> ISTR that there is Solaris 5.6 on it, yes. I think it was actually in "suspend to disk" mode so the uptime is probably 10 years...
[21:51:04] <jbk> i'd imagine someone stealing it from a car isn't thinking about its value as a collector's item
[21:51:12] <Reinhilde> duh what
[21:51:32] <v_a_b> Exactly. I was stupid to leave it in there because it was in something that looked like a laptop bag.
[21:52:14] <Reinhilde> v_a_b: IF I AM NOT MISTAKEN, there is no solaris 5, only 2.5 or SunOS5.x (which is all Solaris except Solaris 1)
[21:52:36] <v_a_b> Solaris 5.6 == SunOS 2.6
[21:52:37] <Reinhilde> all of the versions from 3 to, what, 7? were skipped
[21:53:06] <v_a_b> Marketing, gotta love it...
[21:53:08] <Reinhilde> SunOS 2 was BSD based, not SVR4 as Solaris 2 and up are
[21:53:23] <Reinhilde> you probably have a better grasp than me
[21:53:30] <v_a_b> 2.4 2.5 2.6 err no 6 7 8 9 10 11 12 err no 11.4 :-)
[21:53:42] <Reinhilde> wew
[21:54:09] <jbk> IIRC it was 2.x up through 2.6, then marketing decided to call it 'Solaris 7' instead of 2.7
[21:54:14] <v_a_b> yep
[21:54:37] *** MarcelT <MarcelT!~marcel@tortuga.telka.sk> has quit IRC (Ping timeout: 240 seconds)
[21:54:44] <v_a_b> Not sure if they ever shipped something called Solaris 6.
[21:54:49] <Reinhilde> but here's how I comprehend it. SunOS 1-3 and most of 4 were only SunOS, and were SPARC or 68k BSDs
[21:54:58] <Reinhilde> the last SunOS4 was solaris 1
[21:55:02] <v_a_b> They renamed SunOS to Solaris 1.x after the fact as well.
[21:55:19] <Reinhilde> Solaris 2.x, or SunOS 5.x, is the version we're still on after all these years
[21:55:34] <v_a_b> The machine I am typing this on thinks it is "SunOS 5.11".
[21:55:39] <despair86> heh
[21:55:49] <despair86> filing a bug rn
[21:56:16] <despair86> i can boot pretty much anything but illumos on this box
[21:56:23] <Reinhilde> that's the major suck
[21:56:24] <v_a_b> It's even worse when you look at the SRU: 0.5.11-
[21:56:25] <despair86> haven't even tried solaris
[21:56:32] <v_a_b> What box is that`
[21:56:33] <v_a_b> ?
[21:56:47] <despair86> a new ryzen 7 on amd b450 chipset
[21:57:30] <Reinhilde> what the fuck, did my C++ compiler just stack trace again?
[21:57:37] <Reinhilde> please excuse my language I know there are children in here x_x
[21:57:38] <despair86> lul
[21:57:41] <v_a_b> Ah yes. The German mag "c't" just published their annual "DIY" systems, and I am planning to do the €750 Ryzen 3000 system.
[21:57:51] <v_a_b> Children == anyone under 50
[21:57:55] <despair86> > c't still exists
[21:57:56] <despair86> nice
[21:58:12] <Reinhilde> v_a_b: Go home, graybeard.
[21:58:28] <jbk> heh.. i think c++ is best dealt with using a generous amount of profanity, but that's just my opinion :)
[21:58:30] <Reinhilde> The fresh faces are running this shindig now.
[21:58:36] <Reinhilde> jbk: agreed, but it happens with C compilers too
[21:59:06] <Reinhilde> jbk: overall, my experience with illumos has been a mix of "pretty good" and "Making Debuggers Get Debuggers"
[21:59:45] <v_a_b> I am home (in my home office, hearing the T4-2 going on in the boiler room errr datacenter.
[21:59:48] <v_a_b> )
[22:00:26] <v_a_b> (gdb) fix-bug
[22:00:26] <v_a_b> Undefined command: "fix-bug". Try "help".
[22:00:26] <despair86> for comparison, illumos worked fine without video and wi-fi on a i3-7100U NUC
[22:01:27] <v_a_b> Illumos works great in server contexts, I have both OmniOS and SmartOS here. I am ashamed to say that my OpenIndiana on a Thinkpad T420s hasn't been on for months.
[22:01:48] <Reinhilde> v_a_b: no this is having to gdb /usr/bin/gdb gdb.core
[22:01:55] <Reinhilde> (i know it is called core and not gdb.core but w/e)
[22:02:13] <Reinhilde> debuggers are the only thing I haven't actually had core dump on me yet
[22:03:12] <despair86> ah yes
[22:03:18] <despair86> dbx - core
[22:04:51] <Reinhilde> despair86: ?
[22:05:27] <despair86> the command used to get gdb and dbx to make them discover the binary on its own
[22:06:07] <Reinhilde> k!
[22:10:07] <v_a_b> Gary Mills just posted his OpenIndiana SPARC stuff so there's another OS to try in an LDOM on that T4-2 :-)
[22:11:20] <Reinhilde> If someone releases an automotive ECU whose processor is a SPARC, would it be called SparcPlug?
[22:13:39] <v_a_b> Sounds like something Sun might have used as a code name.
[22:21:16] <Reinhilde> v_a_b: Nah, I don't think Sun was gawdy enough
[22:21:48] <v_a_b> They had hw components called "Big Mac" and "Happy Meal". They had systems codenamed "Enchilada".
[22:22:05] <Reinhilde> A car ECU named Sparcplug would not be out of character for Sun, then?
[22:22:16] <Reinhilde> (if they were in the business of keeping engines running)
[22:22:51] <v_a_b> I dimly remember that someone somewhere used "sparcplug" as a name for something but as we have already established I am old so I forgot what it was.
[22:23:54] <Reinhilde> It'd be weird if it was a car part
[22:24:11] <v_a_b> Nah, some computery thing having to do w/SPARC hence the pun.
[22:24:35] <v_a_b> Did some searching but results are drowned in car-related links.
[22:47:38] <despair86> https://www.illumos.org/issues/11953 - i've stumbled upon a nasty kernel bug
[22:50:10] <v_a_b> The B450 chipset is also on the board that c't proposes for their Ryzen 5 3600 project. *sigh*
[22:51:25] <v_a_b> One must assume that Ubuntu 19.10 would also boot fine.
[22:54:34] <despair86> yes
[23:04:01] *** wl_ <wl_!~wl_@2605:6000:1b0c:6060::87c> has joined #illumos
[23:04:07] <Reinhilde> what does c't stand for
[23:04:39] <Reinhilde> and can someone in canada buy a copy
[23:05:30] <Reinhilde> despair86: will boot -kd drop you into mdb
[23:08:20] <Reinhilde> ... curious. What exactly is "SIGJVM1"?
[23:09:07] <despair86> reinhilde: yes
[23:09:43] <Reinhilde> despair86: ok so the kernel is not COMPLETELY munted.
[23:09:46] <Reinhilde> it's just mostly munted
[23:09:53] <Reinhilde> do you have an external ATI card?
[23:09:58] <despair86> no
[23:10:29] <Reinhilde> i wondered because you said you had a nvidia
[23:11:17] <Reinhilde> do you have some way of getting a serial (possibly firewire? I don't know if illumos supports firewire, let alone a firewire mdb console) console on the system, like a PCI serial card?
[23:12:26] <Reinhilde> does the same hard drive boot in another MB?
[23:18:23] <despair86> > different MB
[23:18:24] <despair86> yes
[23:18:34] <despair86> my other pc is a i3 NUC
[23:18:51] <despair86> works fine but i915 doesn't support Intel HD 620
[23:19:11] <despair86> and i did manage to take pictures of the debug dump this time
[23:19:20] <despair86> i'll post them in a bit
[23:23:21] <Reinhilde> that one time I booted, I think OmniOS on this computer I'm running HBSD on now, I had normal video but no Wi-Fi, which is understandable because the atheros driver doesn't support my chip
[23:28:12] <despair86> https://snowlight.net/files/bug/mdb_msg1.jpg
[23:28:18] <despair86> https://snowlight.net/files/bug/mdb_msg2.jpg
[23:28:21] <despair86> https://snowlight.net/files/bug/mdb_msg3.jpg
[23:28:24] <despair86> https://snowlight.net/files/bug/mdb_msg4.jpg
[23:32:34] <LeftWing> despair86: That's unfortunate!
[23:33:37] <LeftWing> I guess the system is telling us about some kind of PCI-e error that we don't know what to do with
[23:35:03] <LeftWing> despair86: https://github.com/illumos/illumos-gate/blob/master/usr/src/uts/common/io/pciex/pcieb.c#L1700-L1701
[23:35:09] <LeftWing> This is where that message is emitted
[23:36:06] <LeftWing> If you do $C you should get arguments for the functions
[23:37:49] <LeftWing> If you get the first argument to pcieb_intr_handler() you should be able to do something like
[23:38:12] <LeftWing> fffff8848484::print pcieb_devstate_t pcieb_dip | ::devinfo
[23:38:19] <LeftWing> And find out a bit more about which PCI-e bridge it is
[23:38:47] <LeftWing> The second argument to pcieb_intr_handler() looks like an interrupt number which might be useful too
[23:40:40] <LeftWing> It looks like you could also try and set pcieb_die to 0
[23:40:51] <LeftWing> And we might at least not panic at that moment
[23:42:17] <LeftWing> (like, boot -kd so that we drop into kmdb before getting started, so you can set it before boot)
[23:46:47] <despair86> yeah trying that. typing on a usb keyboard in mdb is somehow extremely painful for some reason D:
[23:50:01] *** DrLou <DrLou!~DrLou@c-73-33-92-169.hsd1.nj.comcast.net> has joined #illumos
[23:51:22] <despair86> what mdb command sets that? 'pcieb_die/W 0' does nothing
[23:55:13] *** andy_js <andy_js!~andy@97e29e78.skybroadband.com> has quit IRC (Quit: andy_js)

   November 10, 2019  
< | 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 | >