06:20:35 <gitomat> [illumos-gate] 17903 loader: expose addresses of firmware tables in loader env -- Toomas Soome <tsoome⊙mc>
10:50:52 <DptPirate> Hello. I can't seem to find a way to get Illumos (Tribblix) to work with KVM/qemu on a Linux host. It always panics after the welcome message after loading the kernel. I've tried all kinds of combinations of virt-install options (--os-variant openindiana,solaris10/11, i440fx, q35... ) but nothing works. Information on this is extremely sparse on search engines. Is there a magic virt-install incantation and best practices to get it to work?
10:51:08 <DptPirate> What goes furthest so far: https://privatebin.net/?521c6d0a57fcc3bc#D9cuz4xwpuBXMwCv8gPVyj7JP2srdjTwgjvGzXMt7rry
10:58:12 <denk> a better way is to share the panic info
11:01:00 <DptPirate> Here is the panic info: https://privatebin.net/?cdae4f8181580a97#HYN4n71ZsNnUKoUDraMFixbFwg3H5i2XiG7xnYSkuBj1
11:08:39 <denk> you can boot with -kv flags, it will stop in kmdb instead of rebooting
11:12:23 <denk> somebody broke src.illumos.org, I got 502 bad gateway
11:13:29 <DptPirate> denk: https://privatebin.net/?f5d740e028c4caa0#DtXkrgjhaZrq2fQywh3JSVJPQU14hQkqP6imRm1Bu4EY
11:15:09 <denk> Bad kernel fault at addr=0xfffffe0b8b006370
11:15:17 <denk> type 0xfffffe0b8b006370::whatis
11:16:00 <denk> or without 0x...
11:16:01 <DptPirate> fffffe0b8b006370 is unknown
11:17:45 <andyf> Can you show the output of "$C -vt" ?
11:17:49 <andyf> (quotes are mine)
11:18:37 <DptPirate> andyf: https://privatebin.net/?c3f300f88d996ef3#CS6Kr6oLT5yXKroycgorFRmk41zSpb44TtQituoFHHdE
11:19:53 <andyf> and "::regs"?
11:20:47 <DptPirate> https://privatebin.net/?44a0f14d375f5a37#65war9gmTuaWcmKMWPNaXFweXy4n3gXsKMpPNbSeNVUB
11:22:02 <andyf> startup_modules+10a::dis
11:22:02 <denk> I would disassemble startup_modules near the point 0x10a to locate it in the source code
11:22:15 <andyf> I think this is going to be the smbios_open() call
11:25:05 <denk> we don't see it in the backtrace, it is somewhere else
11:28:39 <denk> 90% that is an unexpected field for the parser
11:47:49 <DptPirate> Would you like me to type in more commands in the debugger?
11:48:02 <andyf> Yes, please do "startup_modules+10a::dis"
11:50:06 <DptPirate> https://privatebin.net/?c2824171dc9f625a#5pbE73JFshTjzuSaPCYYPKpQQnyRKLnjQM6mbMpkUPYU
11:54:13 <denk> indeed, the problem somewhere in smbios_open()
11:54:27 <denk> *is somewhere
11:58:12 <denk> there are 6 points with bcopy() inside the function
13:58:26 <andyf> DptPirate - if qemu has options to control the smbios information presented to the guest I suggest playing with a few of them to see if you can get past this, but it is a bug that it panics.
14:08:51 <tsoome> does 'show' from loader oK prompt list smbios data?
14:35:28 <DptPirate> tsoome: "kmdb: failed to dereference symbol: unknown symbol name"
14:35:50 <DptPirate> andyf: I'll have to investigate this.
14:36:09 <tsoome> not from mdb but from loader ok prompt (press esc in boot menu)
14:40:09 <DptPirate> tsoome: https://privatebin.net/?2e304b1880640bac#3UxeBPDZq3BwMoDukm5JG9kGDLLZezWAUBPMwhNWsJjj
14:40:58 <DptPirate> It looks like there is smbios info in that dump
14:41:32 <tsoome> yes, its bios boot and you do have smbios.
14:42:49 <DptPirate> Would using EFI instead help?
14:43:10 <tsoome> wont hurt to try.
14:43:31 <DptPirate> ok I'll try that in a bit.
15:03:10 <DptPirate> Unfortunately doesn't get to the boat loader (https://privatebin.net/?0726237e6e2439bd#37uszV9fgiaQKTbgSmkGqzuVfrhvtWAEq2KXL4RWh5Xa). And using EFI somehow hosed libvirtd.
17:24:06 <tsoome> 7 is device error - device did report an error condition.
18:31:09 <gitomat> [illumos-gate] 18027 smbd: dereferencing freed memory -- Toomas Soome <tsoome⊙mc>
18:34:16 <jbk> it wouldn't be the first time virtualized media doesn't behave correctly
18:35:48 <jbk> though i wonder if maybe the VM config is missing some parameters
18:40:19 <jbk> though it does remind me... anyone want to review https://code.illumos.org/c/illumos-gate/+/4679/2/usr/src/uts/common/io/sata/adapters/ahci/ahci.c ?
18:40:20 <fenix> → CODE REVIEW 4679: 18011 ahci could handle buggy emulators better (NEW) | https://www.illumos.org/issues/18011
18:55:02 <danmcd> ^^^ done
19:04:19 <jbk> thanks
19:38:18 <richlowe> so I'm more stupid even than usual lately, and I have a question:  Why is `ctrun nightly ...` insufficient to 1) catch the children of make(1) in nightly's contract, 2) keep the contract up until they all quit?
19:41:26 <gitomat> [illumos-gate] 18040 vtfontcvt: struct file_header needs __nonstring attribute -- Toomas Soome <tsoome⊙mc>
20:30:04 <gitomat> [illumos-gate] 17992 libdevinfo: enable smatch and fix issues -- Andy Fiddaman <illumos⊙fn>
21:06:18 <gitomat> [illumos-gate] 13157 maxbw can kill link with hires tick -- Kyle Simpson <kyle⊙oc>
21:06:18 <gitomat> [illumos-gate] 18078 snoop should report drops when capturing to file -- Ryan Zezeski <ryan⊙zc>
21:16:33 <richlowe> I guess the person who approved 18078 didn't copy the list?
21:16:39 <richlowe> or is my mail fucking up?
21:17:04 <danmcd> No... turns out MY mail is fucking up.  Engaging with my local mail-runners now.  I'm super-pissed.
21:17:16 <danmcd> Four mails got gagged, the fourth was a test email.
21:57:23 <sommerfeld> richlowe: the ctrun thing is mystifying (as I don't see how children wouldn't get attached barring something in an intermediate process detaching them).   Unless the newtask invocation does something weird to the contract..
22:06:23 <jbk> if there's a long enough running child, could you use ps to look at the contract ids of the children?
22:07:43 <jbk> or maybe ptree -c[g] ?
22:13:39 <iximeow> i tripped over https://www.illumos.org/issues/18083 just now and i'm wondering what a good way to improve things would be. maybe just changing pbind -u output when a pid has >1 (bound?) lwp?
22:13:40 <fenix> → BUG 18083: pbind -u is confusing with LWPs (New)
22:14:21 <iximeow> i struggle to imagine processor_bind() returning One Of The Previously Bound Processors to be useful if LWPs are bound in interesting ways, but going and just changing *that* feels pretty reckless :D
22:14:59 <rmustacc> Specifically you didn't specify an lwp right?
22:15:05 <iximeow> right, just unbound the whole pid
22:15:36 <rmustacc> I think it's probably impractical for it to report all unbindings.
22:15:47 <rmustacc> Which doesn't answer the question.
22:15:54 <rmustacc> But maybe reduces the state?
22:15:59 <rmustacc> s/state/space
22:16:58 <iximeow> why do you think it's impractical? that's actually one of the options i was considering. it's just annoying to go through each lwp when we really want to unbind them all. that's what we'd do with something like pid/0-
22:17:28 <iximeow> that's just clumsy when we know we want P_PID and the old bindings aren't necessarily that interesting
22:17:53 <rmustacc> Well, I assume you could have probably at least 10k+ threads with bindings.
22:18:00 <rmustacc> And so that's a lot of data to copy out.
22:18:14 <iximeow> oh, yeah
22:18:17 <rmustacc> And you don't even know how much memory to hold to get it all.
22:18:48 <rmustacc> The other thing and the reason it's differnt with walking all threads is that you're able to do it in one atomic-ish action.
22:19:03 <rmustacc> Where as the other mechanism allows the process reign to move and have others intervene.
22:19:05 <iximeow> yeah
22:19:24 <iximeow> i was wondering what happens if you unbind `pid/0-` while the process spawns more and binds them. don't love it
22:19:28 <richlowe> it's almost like being less specific there would at least let you not be confused
22:19:38 <richlowe> if it wasn't trying to help, you'd have been fine
22:19:41 <rmustacc> I wonder if we could change add a return flag to basically say there was a lot.
22:20:08 <rmustacc> Yeah. Basically it seems that what's more useful is being less helpful when there was more than one unbinding.
22:20:52 <iximeow> yeah. i'm happy to go have pbind just say "process id $PID: was bound, now not bound" when processor_bind() succeeds and returned an old binding
22:21:14 <iximeow> rmustacc: are you imagining a sentinel obind to say there was more than one processor unbound in the action?
22:22:08 <rmustacc> Kind of yeah.
22:22:17 <rmustacc> Some variant of none.
22:22:28 <rmustacc> Dunno exactly. Just thinking out loud.
22:22:37 <sommerfeld> something like "threads were bound to multiple cpus, now not bound"
22:22:49 <iximeow> seems reasonable
22:23:22 <rmustacc> In this case it'd be threads, but it'd be whatever the constinuents of the id type were.