Switch to DuckDuckGo Search
   August 29, 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 | 31 | >

Toggle Join/Part | bottom
[00:08:36] <jlevon> I guess that'd be fine?
[00:09:51] <jlevon> would people prefer I do that instead?
[00:10:33] <andyf> No preference here, the module seems fine
[00:11:28] *** andy_js <andy_js!~andy@94.12.192.123> has quit IRC (Quit: andy_js)
[00:13:04] <LeftWing> It saves a build step, which is cool
[00:13:44] <LeftWing> unix could just do it if the file exists
[00:13:48] <richlowe> well, I wasn't being serious, but if you _want_ me to be, it being constructed at runtime lets other people put stuff in there.
[00:14:09] <richlowe> so in OI
[00:14:20] <richlowe> or I guess probably omnios
[00:14:25] <richlowe> that'd likely help with things such as kvm
[00:14:32] <richlowe> since they're not building one big wad of platform image
[00:15:02] <andyf> /etc/versions.d/ or something?
[00:15:35] <richlowe> andyf: what alan said was already a directory.
[00:16:10] <andyf> ah, ok
[00:16:48] <LeftWing> So your package with unbundled modules could include a file in there and bam
[00:17:18] <alanc> yes, my system currently has /etc/versions/osnet, /etc/versions/uts_version, /etc/versions/uts_version_internal
[00:17:49] <LeftWing> Is there structure to the files, or are they just a string? You mentioned "uname -v" uses these somehow?
[00:17:59] <alanc> just strings
[00:18:13] <LeftWing> So "uname -v" emits whatever is in "uts_version" then?
[00:19:16] <alanc> yes - looks like we have an SMF service that calls utssys() to set the name
[00:19:33] <LeftWing> Huh.
[00:19:44] <LeftWing> What does it return before that runs?
[00:19:46] <alanc> % cat /etc/versions/uts_version
[00:19:46] <alanc> 11.5.0.52.0
[00:19:46] <alanc> % uname -v
[00:19:47] <alanc> 11.5.0.52.0
[00:19:49] *** jellydonut <jellydonut!~quassel@s91904425.blix.com> has quit IRC (Quit: jellydonut)
[00:20:30] <alanc> I'm not sure
[00:20:36] <LeftWing> Fair enough!
[00:20:37] <richlowe> Well, I'm also not endorsing whatever Oracle happen to have done. :)
[00:20:45] <LeftWing> Sure, I'm just curious
[00:20:54] <richlowe> I was just thinking it seems to fit non-smartos systems better that way.
[00:21:02] <richlowe> since I got provoked into having opinions other than that john was slack
[00:21:18] <LeftWing> Honestly it fits SmartOS systems pretty well too -- our module started outside our illumos, rather it was built in smartos-live where we have the full git picture
[00:21:27] <LeftWing> We could just chuck this file into place in the ramdisk instead
[00:22:12] <alanc> our /etc/versions/osnet has the detailed info from hg: src:38889:636f8db59982:st_053+25;closed:2479:c32c7a0ad284:st_053:st_053.1+3
[00:22:32] <alanc> still available via ::status in mdb
[00:22:48] <LeftWing> Yeah that's the sort of thing we have in ours
[00:24:00] <LeftWing> https://gist.github.com/jclulow/89eb13a1ffdac60c374e2a4c02830c36
[00:24:37] <jlevon> so I'd prefer a name other than 'osnet' given what illumos-joyent does with it.
[00:24:48] <jlevon> /etc/versions/build ?
[00:24:55] <LeftWing> Yeah that sounds good
[00:25:10] <jlevon> OK. I'll give that a go.
[00:25:30] <richlowe> jlevon: wouldn't you have like /etc/versions/illumos-joyent /etc/versions/smartos-live and such?
[00:25:58] <jlevon> richlowe: and a bunch of kernel mods to read each file? seems a bit tedious there. and also they can vary
[00:26:08] <jlevon> mods meaning modifications
[00:26:22] <LeftWing> Can we readdir?
[00:26:34] <richlowe> well, that gets us back to omnios etc, having a harder time dropping new bits in
[00:26:58] <andyf> We read /etc/system.d/ so I assume there's a mechanism
[00:26:58] <jlevon> I don't think the presumption is that everything in /etc/versions/ ends up kernelside?
[00:27:16] <LeftWing> andyf: Unfortunately _that_ mechanism is, I believe, "cat"
[00:27:16] <andyf> Ideally, everything in /etc/versions/ would get built into a JSON string or something inside the kernel
[00:27:25] <andyf> LeftWing - oh, so it is
[00:28:02] <jlevon> it's starting to sound a bit over-engineered ...
[00:28:51] <jlevon> omnios can construct that json string if they like for that file. that's what the joyent build does
[00:29:08] <LeftWing> I think /etc/versions/build is a great first cut, and if we want more later we can do more.
[00:29:31] <LeftWing> Other things can start delivering their versions in there alongside, even, and if they do and it's good we can figure out loading or whatever later.
[00:29:38] <jlevon> ok
[00:29:49] <Smithx10> hahahaha LeftWing
[00:29:53] <Smithx10> "cat"
[00:29:59] <LeftWing> Smithx10: You laugh, but it so is
[00:30:03] <LeftWing> I recall reviewing it
[00:30:05] <Smithx10> I know....
[00:30:16] <Smithx10> Im just thinking about the immediate sigh
[00:32:01] <richlowe> jlevon, LeftWing: sounds good
[00:32:07] <jlevon> thanks all.
[00:33:20] <LeftWing> Thanks John!
[00:33:35] <andyf> It might make sense (later) to have an SMF service that assembles /etc/versions/build from the other files in there, like we do for system.d
[00:33:56] <jlevon> yeah, let's see if we end up needing it. it's not something smartos would/could do
[00:36:21] *** merzo <merzo!~merzo@172-1-133-95.pool.ukrtel.net> has joined #illumos
[00:37:44] <yuripv> andyf: I don't think we use SMF service for that?
[00:37:58] *** phyre__ <phyre__!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 245 seconds)
[00:38:11] <andyf> no, bootadm assembles system.d, but I means assemble in the same way
[00:38:22] <yuripv> oh ok
[00:38:24] <andyf> we use SMF services to assemble lots of other things like exec_attr
[00:39:04] <LeftWing> That's partially so that they can be updated live by pkg(5) I believe, by kicking the service that assembles
[00:39:09] <andyf> yep
[00:39:14] <andyf> so that fits for this too..
[00:39:22] <andyf> a package that updates /etc/versions/XXX can kick the service
[00:39:23] <LeftWing> Yeah it could load any that haven't been loaded
[00:39:47] <LeftWing> You could even include such a service just in OmniOS to start until we're happy with it
[00:40:30] <andyf> Similar to our makewhatis service that triggers whenever a man page is installed or removed
[00:40:35] <richlowe> yeah, driving that bit from somewhere that actually needs it makes sense
[00:40:48] <richlowe> andyf: if yours actually works, you should sell OI on it.
[00:40:52] <richlowe> andyf: and get them to fix their info one, too
[00:40:55] <richlowe> says I, info's only user.
[00:40:58] <LeftWing> lol
[00:41:06] <LeftWing> I used info recently! Admittedly, it was on Ubuntu.
[00:53:48] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has quit IRC ()
[00:58:24] <yuripv> andyf: mind upstreaming it?
[01:00:48] <andyf> no, it's on the list
[01:00:55] <yuripv> good, thanks
[01:02:21] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has joined #illumos
[01:03:12] * LeftWing inches towards 7119
[01:03:28] <LeftWing> It's sooo good to be able to boot an rpool image on different disks
[01:05:34] <richlowe> LeftWing: I regret giving up on my issue now, I could have brought it up in your RTI
[01:05:46] <LeftWing> Which issue
[01:06:12] <richlowe> LeftWing: the one BE that said it couldn't read the pool label, despite the others being able to.
[01:06:15] <andyf> yuripv - it's at https://illumos.org/rb/r/2270/ if it's of interest, I can open a bug etc. in the morning
[01:06:26] <LeftWing> richlowe: Yeah that was odd
[01:10:03] <yuripv> service_bundle type='manifest' name='SUNWcsr:whatis'>
[01:10:11] <yuripv> are those SUNW names still needed?
[01:10:20] <yuripv> (general question)
[01:11:02] <richlowe> yuripv: totally just added that as a comment, in fact.
[01:14:04] <andyf> I'm pretty sure it's arbitrary, SUNWxxx is just the pattern for 98% of the manifests in gate
[01:14:15] <andyf> the dtd says it should be named after the package that delivers
[01:15:32] <richlowe> right, because they came from when that really was the package name
[01:15:47] <andyf> so 'system/man:whatis' would probably be better, although there is not much prior art to look at
[01:15:56] <andyf> one is system/zones-monitoring
[01:16:17] <richlowe> you might be the first prior art where someone looked at it and went "SUNWcsr? what year is this?"
[01:16:19] <richlowe> let alone two people
[01:16:23] <richlowe> s/prior //
[01:29:48] *** sarvet <sarvet!~sarvet@i59F4518D.versanet.de> has quit IRC (Quit: sarvet)
[01:35:13] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has quit IRC (Quit: Leaving)
[01:51:05] <gitomat> [illumos-gate] 11577 blown assert in i40e_intr_io_clear_cause() -- Robert Mustacchi <rm at joyent dot com>
[01:58:58] <gitomat> [illumos-gate] 10622 ZFS should still check paths for devices that have no devid -- Joshua M. Clulow <josh at sysmgr dot org>
[02:44:57] *** gwr <gwr!~gwr@151.203.114.175> has joined #illumos
[03:11:11] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has quit IRC (Remote host closed the connection)
[03:11:32] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has joined #illumos
[03:11:34] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has quit IRC (Remote host closed the connection)
[03:12:13] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has joined #illumos
[03:16:43] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has quit IRC (Ping timeout: 245 seconds)
[03:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[03:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[03:22:05] *** gwr <gwr!~gwr@151.203.114.175> has quit IRC (Quit: Leaving)
[03:36:33] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[04:01:37] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[04:11:23] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[04:13:28] *** Yogurt <Yogurt!~Yogurt@64-71-13-30.static.wiline.com> has joined #illumos
[04:38:36] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has quit IRC (Quit: sleep(2))
[04:44:29] *** Yogurt <Yogurt!~Yogurt@64-71-13-30.static.wiline.com> has quit IRC (Remote host closed the connection)
[04:45:10] *** Yogurt <Yogurt!~Yogurt@64-71-13-30.static.wiline.com> has joined #illumos
[04:46:18] *** Yogurt <Yogurt!~Yogurt@64-71-13-30.static.wiline.com> has quit IRC (Remote host closed the connection)
[04:58:56] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has quit IRC (Remote host closed the connection)
[05:00:30] *** mnrmnaugh <mnrmnaugh!~mnrmnaugh@unaffiliated/mnrmnaugh> has joined #illumos
[05:01:58] *** Yogurt <Yogurt!~Yogurt@64-71-13-30.static.wiline.com> has joined #illumos
[05:06:18] *** Yogurt <Yogurt!~Yogurt@64-71-13-30.static.wiline.com> has quit IRC (Ping timeout: 244 seconds)
[05:17:07] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[05:17:49] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 258 seconds)
[05:22:58] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[05:23:10] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[05:31:14] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 258 seconds)
[05:31:37] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 258 seconds)
[05:32:55] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[05:33:11] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[05:41:26] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 244 seconds)
[05:41:26] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 244 seconds)
[05:41:38] *** ryaeng <ryaeng!sid364071@gateway/web/irccloud.com/x-ddjkglquiwoqkcwp> has left #illumos
[05:41:48] *** ryaeng <ryaeng!sid364071@gateway/web/irccloud.com/x-ddjkglquiwoqkcwp> has joined #illumos
[05:43:33] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[05:44:06] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[05:51:15] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 244 seconds)
[05:51:46] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 244 seconds)
[05:52:51] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[05:53:16] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[05:53:28] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[06:01:17] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[06:01:52] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 272 seconds)
[06:03:01] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[06:03:52] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[06:10:53] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[06:11:43] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[06:12:46] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[06:13:02] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[06:21:24] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 248 seconds)
[06:21:50] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 258 seconds)
[06:23:24] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[06:24:49] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[06:25:55] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Remote host closed the connection)
[06:31:00] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 248 seconds)
[06:33:06] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[06:33:44] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[06:41:28] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 268 seconds)
[06:42:05] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 268 seconds)
[06:42:54] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[06:43:22] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[06:50:43] *** papertigers <papertigers!~papertige@pool-108-55-27-43.bflony.fios.verizon.net> has quit IRC (Ping timeout: 268 seconds)
[06:50:53] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[06:51:43] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[06:53:03] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[06:53:22] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[07:00:58] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 246 seconds)
[07:01:19] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 246 seconds)
[07:02:56] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[07:04:02] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[07:10:53] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[07:11:40] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 258 seconds)
[07:11:43] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[07:13:01] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[07:13:20] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[07:15:21] *** papertigers <papertigers!~papertige@pool-72-75-249-69.bflony.fios.verizon.net> has joined #illumos
[07:15:38] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[07:21:40] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 272 seconds)
[07:22:18] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 272 seconds)
[07:23:12] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[07:23:30] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[07:31:48] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 272 seconds)
[07:31:48] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 272 seconds)
[07:33:01] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[07:33:15] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has joined #illumos
[07:41:42] *** ruid_ <ruid_!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[07:41:42] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 245 seconds)
[07:42:36] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[07:51:38] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 244 seconds)
[07:53:06] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[07:55:16] *** jimklimov <jimklimov!~jimklimov@ip-86-49-243-242.net.upcbroadband.cz> has quit IRC (Quit: Leaving.)
[08:00:20] *** phyre__ <phyre__!~phyre___@31.211.191.164> has joined #illumos
[08:02:12] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 272 seconds)
[08:02:54] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[08:11:19] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 246 seconds)
[08:13:33] *** ruid <ruid!~ruid@unaffiliated/ruid> has joined #illumos
[08:14:13] *** phyre__ <phyre__!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 245 seconds)
[08:17:29] *** amrfrsh <amrfrsh!~Thunderbi@190.2.145.106> has quit IRC (Quit: amrfrsh)
[08:20:05] *** amrfrsh <amrfrsh!~Thunderbi@190.2.145.106> has joined #illumos
[08:20:08] *** amrfrsh <amrfrsh!~Thunderbi@190.2.145.106> has quit IRC (Client Quit)
[08:20:28] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: tsoome)
[08:20:55] *** Lirion <Lirion!~m00se@wikimedia-commons/Lirion> has quit IRC (Quit: echo 16i[q]sa[ln0=aln100%Pln100/snlbx]sbA0D4D465452snlb xq |dc)
[08:21:24] *** ruid <ruid!~ruid@unaffiliated/ruid> has quit IRC (Ping timeout: 248 seconds)
[08:36:06] *** alanc <alanc!~alanc@129.157.69.38> has quit IRC (Remote host closed the connection)
[08:36:33] *** alanc <alanc!~alanc@129.157.69.38> has joined #illumos
[08:50:29] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has joined #illumos
[08:57:41] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[09:01:00] <tsoome> any advocate around to check on 11635 ?:)
[09:11:19] <LeftWing> tsoome: I assume it wsdiffs clean?
[09:13:11] <gitomat> [illumos-gate] 11621 fmadm and fmstat document privileges incorrectly -- Peter Tribble <peter.tribble at gmail dot com>
[09:13:13] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[09:18:05] <LeftWing> Well, I'm going to bed -- but assuming it's clean, you should push it
[09:18:32] <LeftWing> (Is this the last fix before you can push the NULL definition patch?)
[09:18:41] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Remote host closed the connection)
[09:19:08] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[09:21:20] <gitomat> [illumos-gate] 11628 typos detected by mandoc style -- Peter Tribble <peter.tribble at gmail dot com>
[09:21:53] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[09:22:23] <tsoome> yes
[09:22:24] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[09:22:33] <tsoome> and eys
[09:22:35] <tsoome> yes
[09:23:00] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[09:23:17] <LeftWing> That's exciting!
[09:24:39] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[09:37:19] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: tsoome)
[09:56:18] *** merzo <merzo!~merzo@172-1-133-95.pool.ukrtel.net> has quit IRC (Ping timeout: 245 seconds)
[10:05:21] *** andy_js <andy_js!~andy@94.12.192.123> has joined #illumos
[10:08:35] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[10:15:09] <gitomat> [illumos-gate] 11635 ip: passing argument 2 of 'cc_cong_signal' makes integer from pointer without a cast -- Toomas Soome <tsoome at me dot com>
[10:23:27] *** atrius <atrius!~atrius@2600:1700:22d1:518:5836:8d3c:8c23:6cf5> has quit IRC (Ping timeout: 264 seconds)
[10:25:25] *** atrius <atrius!~atrius@2600:1700:22d1:518:9cae:a179:351c:4dfc> has joined #illumos
[10:26:52] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[10:28:17] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[10:28:19] *** amrfrsh <amrfrsh!~Thunderbi@eduroamerw3-46-16.uni-paderborn.de> has joined #illumos
[10:33:11] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[10:33:51] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[10:36:49] *** man_u <man_u!~manu@manu2.gandi.net> has joined #illumos
[10:36:55] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has quit IRC (Ping timeout: 246 seconds)
[11:11:50] *** merzo <merzo!~merzo@185.39.197.205> has joined #illumos
[11:13:37] <gitomat> [illumos-gate] 11112 sys/null.h: make NULL void * -- Toomas Soome <tsoome at me dot com>
[11:14:14] <jlevon> tsoome: woo!
[11:14:38] <tsoome> just had to perform additional build to be sure:D
[11:16:37] *** amrfrsh <amrfrsh!~Thunderbi@eduroamerw3-46-16.uni-paderborn.de> has quit IRC (Quit: amrfrsh)
[11:17:02] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has joined #illumos
[11:18:04] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[11:18:57] <tsoome> jlevon: about https://illumos.org/rb/r/2268/ - in first take I did focus on getting ESP updates; then we can start to build followups to make it possible to operate on disk image files.
[11:18:58] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:19:15] <jlevon> tsoome: yeah I'm meaning to review for you
[11:19:46] <tsoome> I mean, I have not forgotten the need to support images:)
[11:20:15] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[11:20:28] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:20:42] <jlevon> tsoome: appreciated :)
[11:21:52] <tsoome> we also have a bit confusing set of support libraries — libefi, libfdisk, libadm to handle gpt, mbr and vtoc respectively…
[11:22:06] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has quit IRC (Quit: sleep(2))
[11:22:30] <jlevon> it's a right old mess
[11:25:24] <tsoome> ye, in perfect world, we could have one ring^H^H^H^Hlibrary to provide the API to handle both live (synced with kernel) and image based storage, and that could make it easier. Still there is also concern about how to address the file system creation…
[11:36:10] *** merzo <merzo!~merzo@185.39.197.205> has quit IRC (Ping timeout: 258 seconds)
[11:38:43] <andyf> tsoome, congratulations and thanks for the null fix.
[11:38:59] <andyf> I hope to get around to testing the ESP stuff today
[11:40:26] <tsoome> thanks:)
[11:41:53] <tsoome> darn, libbe is leaking snapshots
[11:42:54] <tsoome> I have one BE left, but ton of snapshots..
[11:45:26] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC ()
[11:47:08] *** amrfrsh <amrfrsh!~Thunderbi@190.2.145.106> has joined #illumos
[11:47:43] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has joined #illumos
[11:53:30] <tsoome> I guess those should be nuked by beadm destroy -s
[11:55:17] <tsoome> nope, still there.
[11:55:48] *** nexgen2 <nexgen2!~nexgen@144.172.68.147> has joined #illumos
[11:57:59] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has quit IRC (Quit: sleep(2))
[12:01:16] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has joined #illumos
[12:02:12] *** nexgen2 <nexgen2!~nexgen@144.172.68.147> has quit IRC (Ping timeout: 248 seconds)
[12:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[12:20:07] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[12:28:12] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[12:28:50] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[12:37:10] <tsoome> and now it did destroy snapshot… grr.
[12:40:17] <igork> tsoome: BE = snapshot + clone to new dataset. and it remove another snapshorts not every time
[12:41:42] <tsoome> the problem is that apparently it is leaving origin around sometimes.
[12:57:30] *** nexgen2 <nexgen2!~nexgen@144.172.68.147> has joined #illumos
[13:01:17] *** merzo <merzo!~merzo@185.39.197.205> has joined #illumos
[13:08:05] *** phyre__ <phyre__!~phyre___@31.211.191.164> has joined #illumos
[13:27:59] *** nexgen2 <nexgen2!~nexgen@144.172.68.147> has quit IRC (Ping timeout: 244 seconds)
[13:36:06] *** jellydonut <jellydonut!~quassel@s91904424.blix.com> has joined #illumos
[13:45:15] <jimklimov> Hi all, short question: did I miss something, or is there no pool/dataset option to avoid automatic mounts of "this (and child)" filesystems?
[13:45:30] <jimklimov> not "canmount" I guess, for the usecase of backups
[13:46:03] <jimklimov> so I want to replicate my pools to another disk; I do not need those filesystems mounted while on it
[13:46:15] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: tsoome)
[13:46:30] <andyf> mountpoint=legacy ?
[13:46:49] <jimklimov> but I also do not want to pollute the ordinary dataset attributes like "canmount" because their values are functional part of rpool/zone/... setup
[13:47:09] <jimklimov> so when I restore, I want the original values applied on the "live" copy
[13:47:42] <jimklimov> there is a "zpool import -N" that avoids mounting just during import, but does not preclude automounts to happen later
[13:48:07] <jimklimov> I wondered if there is a more persistent toggle for such effect
[13:49:31] <jimklimov> in practical terms, I've mounted the backup disk with an altroot so at least it does not pollute the common OS namespace, but as it goes through zfs recv'ing my rpool and lots of BEs, it keeps complaining that it can not mount "/backup/var" and "/backup/usr" as they are already busy (from the first copied BE)
[13:52:30] <jimklimov> hm... maybe might abuse "zoned=on"? :)
[14:07:46] <andyf> that would probably work
[14:25:10] *** nexgen2 <nexgen2!~nexgen@144.172.68.147> has joined #illumos
[14:30:23] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:31:25] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[14:35:19] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Remote host closed the connection)
[14:35:46] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[14:36:10] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Remote host closed the connection)
[14:36:42] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[14:38:04] *** nexgen2 <nexgen2!~nexgen@144.172.68.147> has quit IRC (Ping timeout: 246 seconds)
[14:38:43] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[14:46:10] <andyf> tsoome - this ESP patch is working pretty much perfectly..
[14:46:23] <andyf> some of the things I've thrown at it have caused problems, but they need manual intervention to fix anyway
[14:46:32] <andyf> (like formatting the ESP as ufs and seeing what happens...)
[14:49:49] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has quit IRC (Quit: sleep(2))
[14:50:27] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has joined #illumos
[14:52:58] <tsoome> yes, well, I did discuss that a bit with _alhazred and from one hand we did figure that in case of errors, if we just mkfs -F pcfs on it, we will make life easier for appliances. But that would bite multiboot setups… so other option would be to have defaults file to determine the behavior - error out or mkfs:)
[14:55:20] <andyf> The problem here is actually that `mkfs -F pcfs` still leaves the UFS signature in place
[14:56:09] <tsoome> ah, in that case, it is mkfs_pcfs bug.
[14:56:24] <andyf> or something - I don't expect installboot to deal with that one
[14:56:35] *** _i4k_ <_i4k_!~i4k@78.17.52.69> has quit IRC (Quit: unplugging)
[14:56:39] <tsoome> (we need to address mkfs_pcfs anyhow for other reasons)
[14:57:52] <andyf> I just need to get the right bits into the installer miniroot now and I can test an installation
[14:58:04] <tsoome> :)
[14:59:41] <tsoome> hm, essentially all mkfs_* (and zpool create) should perform unmap-like action :)
[15:07:11] <yuripv> we also need a fat32 mkfs that doesn't go looking for diskettes
[15:16:12] <tsoome> yes, and autodetect fat12/16/32 + large sector size support
[15:27:03] *** spectral` <spectral`!~julian@os.k/spectral> has quit IRC (Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net)
[15:27:21] *** baojg <baojg!~baojg@162.243.44.213> has quit IRC (Remote host closed the connection)
[15:30:45] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has joined #illumos
[15:34:08] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has quit IRC (Remote host closed the connection)
[15:34:48] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[15:49:38] *** jzu_ <jzu_!~jussi@pupu.jus.si> has quit IRC (Ping timeout: 245 seconds)
[15:54:55] *** igork1 <igork1!~igork@91.204.56.74> has joined #illumos
[15:56:42] *** igork <igork!~igork@91.204.56.74> has quit IRC (Ping timeout: 245 seconds)
[15:57:33] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[15:58:10] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[16:00:40] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has quit IRC (Ping timeout: 258 seconds)
[16:04:44] *** psydroid <psydroid!psydroidma@gateway/shell/matrix.org/x-jgrfmewepegnoadp> has quit IRC (Changing host)
[16:04:44] *** psydroid <psydroid!psydroidma@unaffiliated/psydroid> has joined #illumos
[16:04:44] *** psydroid <psydroid!psydroidma@unaffiliated/psydroid> has quit IRC (Changing host)
[16:04:44] *** psydroid <psydroid!psydroidma@gateway/shell/matrix.org/x-jgrfmewepegnoadp> has joined #illumos
[16:16:07] *** neirac <neirac!~cneir@190.162.109.190> has joined #illumos
[16:18:47] *** spectral` <spectral`!~julian@spectral.powered.by.lunarbnc.net> has joined #illumos
[16:18:47] *** spectral` <spectral`!~julian@spectral.powered.by.lunarbnc.net> has quit IRC (Changing host)
[16:18:47] *** spectral` <spectral`!~julian@os.k/spectral> has joined #illumos
[16:22:32] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has joined #illumos
[16:22:38] *** igork1 is now known as igork
[16:22:44] *** jzu_ <jzu_!~jussi@pupu.jus.si> has joined #illumos
[16:24:26] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Ping timeout: 258 seconds)
[16:55:40] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has quit IRC (Ping timeout: 268 seconds)
[17:01:43] *** merzo <merzo!~merzo@185.39.197.205> has quit IRC (Ping timeout: 245 seconds)
[17:04:12] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[17:04:37] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has quit IRC (Ping timeout: 245 seconds)
[17:04:58] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has joined #illumos
[17:05:39] *** baojg <baojg!~baojg@162.243.44.213> has joined #illumos
[17:15:02] <ptribble> I must stop reading man pages
[17:20:23] <andyf> :)
[17:21:12] <rmustacc> Don't worry, we'll keep writing new ones for you to read.
[17:24:50] <andyf> https://paste.ec/paste/DVjtgXjB#Hwlx7LKPppcAhC1pjOedDwvVNpVqP0Pp+9GLMRysy66
[17:25:04] <andyf> New one...
[17:25:58] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC ()
[17:30:45] <andyf> reproducible with `rem_drv vioif` so presumably something related to the recent improvements there
[17:50:10] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has quit IRC (Ping timeout: 276 seconds)
[17:50:27] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Quit: Leaving.)
[17:58:43] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[17:59:47] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has joined #illumos
[18:02:49] *** merzo <merzo!~merzo@185.39.197.205> has joined #illumos
[18:10:16] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has joined #illumos
[18:13:57] <rmustacc> LeftWing: ^
[18:15:07] <andyf> and another crash on attach failure in mac_fini_ops() which does not like .cb_str being NULL
[18:15:27] <andyf> rmustacc, thanks, I'll ping him directly.
[18:20:14] <rmustacc> I'll help take a look a bit later today.
[18:27:18] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has quit IRC (Ping timeout: 272 seconds)
[18:28:08] *** idodeclare <idodeclare!~textual@cpe-76-185-177-63.satx.res.rr.com> has joined #illumos
[18:29:33] *** tru_tru <tru_tru!~tru@sillage.bis.pasteur.fr> has quit IRC (Quit: leaving)
[18:30:05] *** tru_tru <tru_tru!~tru@sillage.bis.pasteur.fr> has joined #illumos
[18:30:29] *** tru_tru <tru_tru!~tru@sillage.bis.pasteur.fr> has quit IRC (Read error: error:1408F10B:SSL routines:ssl3_get_record:wrong version number)
[18:32:00] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Quit: man_u)
[18:33:06] *** tru_tru <tru_tru!~tru@sillage.bis.pasteur.fr> has joined #illumos
[19:04:48] *** sarvet <sarvet!~sarvet@i59F508FA.versanet.de> has joined #illumos
[19:10:22] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has quit IRC (Ping timeout: 272 seconds)
[19:27:45] *** jimklimov <jimklimov!~jimklimov@ip-86-49-243-242.net.upcbroadband.cz> has joined #illumos
[19:31:12] *** merzo <merzo!~merzo@185.39.197.205> has quit IRC (Ping timeout: 244 seconds)
[19:35:59] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Quit: Leaving)
[19:37:13] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Remote host closed the connection)
[19:37:40] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[19:40:04] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[19:41:16] *** wacki <wacki!~wacki@i577B8BAD.versanet.de> has joined #illumos
[19:51:01] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Remote host closed the connection)
[19:51:28] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[19:57:00] <richlowe> can't decide whether perlmacos(1) or perlhurd(1) is least useful
[19:57:06] <richlowe> thanks andyf for setting me off on this adventure
[19:58:37] <andyf> hmm?
[19:58:51] <LeftWing> whatis?
[20:01:20] <richlowe> andyf: I went looking for where there were important manual pages that weren't covered by your approach to finding them.
[20:02:07] <richlowe> though describing various of the perl manual pages as important is a stretch. :)
[20:03:42] <richlowe> I'd forgot about them shipping what amounts to a changelog in the manual, too.
[20:27:37] <gitomat> [illumos-gate] 3386 ld(1) should document -z symbol-cap -- Robert Mustacchi <rm at joyent dot com>
[20:29:06] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC ()
[20:29:16] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[20:29:54] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Client Quit)
[20:30:05] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[20:30:41] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Client Quit)
[20:30:50] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[20:31:29] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Client Quit)
[20:31:39] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[20:32:16] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Client Quit)
[20:32:27] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[20:33:04] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Client Quit)
[20:33:13] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has joined #illumos
[20:33:51] *** _alhazred <_alhazred!~Alex@mobile-access-bcee0a-247.dhcp.inet.fi> has quit IRC (Client Quit)
[20:49:48] <jlevon> richlowe: not around much this evening so can't look further, but you might be interested / depressed by https://smartos.org/bugview/OS-7973
[20:51:15] <richlowe> Is that literally a thing we document should already have been checked?!
[20:52:14] <rmustacc> richlowe: Probably, but I made the mistake of going one gcc7 vesrion to the next with -il3. So I wasn't comparing against 4.
[20:52:37] <rmustacc> So that's how I missed it with the retpoline change there.
[20:52:54] <rmustacc> (Which I hope makes sense)
[20:57:00] *** wiedi <wiedi!~wiedi@p5B119C32.dip0.t-ipconnect.de> has joined #illumos
[20:57:54] <richlowe> jlevon: Is it knowing about the prologue(?), or is it hoisting above the prologue because it "knows" it doesn't need it?
[20:58:53] *** wacki <wacki!~wacki@i577B8BAD.versanet.de> has quit IRC (Quit: Lingo: www.lingoirc.com)
[21:00:57] <jlevon> richlowe: I don't know what's happening
[21:01:02] <jlevon> and yeah it should be in our list
[21:23:50] *** wiedi <wiedi!~wiedi@p5B119C32.dip0.t-ipconnect.de> has quit IRC (Quit: ^C)
[21:42:34] *** neirac <neirac!~cneir@190.162.109.190> has quit IRC (Read error: Connection reset by peer)
[21:44:38] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[22:04:09] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has joined #illumos
[22:10:20] *** danmcd <danmcd!~danmcd@static-71-174-113-16.bstnma.fios.verizon.net> has quit IRC (Ping timeout: 244 seconds)
[22:11:30] *** phyre__ <phyre__!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 272 seconds)
[22:20:01] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[22:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[22:20:20] *** gwr <gwr!~gwr@151.203.114.175> has joined #illumos
[22:49:12] <richlowe> rmustacc: don't suppose you're around?
[22:49:41] <rmustacc> richlowe: I am.
[22:49:51] <rmustacc> What can I do for you?
[22:55:54] *** gwr <gwr!~gwr@151.203.114.175> has quit IRC (Quit: This computer has gone to sleep)
[23:07:27] <richlowe> could someone link the joyent bug again?
[23:07:31] <ryaeng> jimklimov: errmsg.c appears to check the input but allows it to create the user even though it doesn't pass the check. Is this correct or should this be more stringent, i.e. return with error?
[23:07:36] <ryaeng> http://src.illumos.org/source/xref/illumos-gate/usr/src/cmd/oamuser/lib/errmsg.c
[23:07:56] <yuripv> re: man pages locations, we could have a /etc/man.d/ where packages would install the .conf files containing man pages locations
[23:08:03] <ryaeng> Also, who has the final say on issues like these?
[23:08:13] <richlowe> the peanut gallery, usually. :)
[23:09:06] *** gwr <gwr!~gwr@151.203.114.175> has joined #illumos
[23:12:05] <jbk> richlowe: OS-7973 ?
[23:12:16] <richlowe> ... I... wouldn't know? that's the problem?
[23:12:47] <jbk> https://smartos.org/bugview/OS-7973 <-- that the bug you meant
[23:13:07] <jbk> ?
[23:14:03] <richlowe> yes, thanks.
[23:19:24] <richlowe> well so far this is just making me angrier and angrier without bound
[23:21:34] <jlevon> uh oh
[23:23:47] <richlowe> waiting for rmustacc to confirm I'm only as crazy as usual.
[23:23:50] <richlowe> then I'll share the love.
[23:24:06] *** ptribble <ptribble!~ptribble@cpc92716-cmbg20-2-0-cust138.5-4.cable.virginm.net> has quit IRC (Quit: Leaving)
[23:32:13] <richlowe> jlevon: I believe the problem is shrink wrapping.
[23:32:27] <richlowe> jlevon: -fno-shrink-wrap should bring things back into line
[23:32:35] <jlevon> one mo
[23:33:14] <richlowe> jlevon: I'm unclear on whether the _problem_ is the interaction of shrink-wrap and save args again, or whether it's purely that when shrink wrapped some things can get hoisted over the prologue
[23:35:02] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Read error: Connection reset by peer)
[23:35:10] <jlevon> man I was so close with -fno-shrink-wrap-separate
[23:35:12] <richlowe> with shrink wrap and no save-args, we still hoist bits, but differently.
[23:35:27] <richlowe> in that we end up with pushq %rbp, testl %edx,%edx
[23:35:33] <richlowe> which would still scupper things, I think?
[23:35:38] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[23:35:43] <jlevon> no that's fine now iirc
[23:36:13] <richlowe> we were unclear on whether fbt needed the push and mov, or just the push
[23:36:29] <jlevon> looks like on 64-bit it's only the push
[23:36:42] <rmustacc> Yeah, that was my read of it even though the comment says otherwise.
[23:36:51] <jlevon> I swear it used to be that
[23:36:55] <richlowe> then the code I asked rzezeski to put in the bug needs the trailing 3 bytes removed from the check.
[23:36:56] *** sarvet <sarvet!~sarvet@i59F508FA.versanet.de> has quit IRC (Quit: sarvet)
[23:37:17] <richlowe> and you should be able to disable shrink-wrap at least as a workaround for right now
[23:37:18] <jlevon> I'll start a build with -fno-shrink-wrap and see where we go
[23:37:24] <jlevon> thanks as always
[23:37:51] <richlowe> and I'll try again to find what's actually going on in here
[23:38:18] *** andy_js <andy_js!~andy@94.12.192.123> has quit IRC (Quit: andy_js)
[23:38:21] <richlowe> that might be tomorrow though, when I'm feeling less like breaking stuff. :)
[23:38:27] <rzezeski> 10-4
[23:38:45] <rmustacc> Is there a good explanation of the option or the reason it exists?
[23:39:05] <jlevon> Emit function prologues only before parts of the function that need it, rather than at the top of the function. This flag is enabled by default at -O and higher.
[23:39:12] <jlevon> It's pretty dead on hah
[23:39:20] <richlowe> I mean, in general I'd just like to not do that.
[23:39:21] <richlowe> but..
[23:39:25] <jlevon> somehow I missed that instead of the -separate version.
[23:39:35] <richlowe> somehow we still don't convince gcc we need it
[23:39:46] <rmustacc> Right, I read that. Just was looking for the why I guess.
[23:39:58] <richlowe> but the wording does suggest whether save-args tells it the right thing or not, sometimes it's going to hurt us, right?
[23:40:13] <richlowe> we don't want to rely on save-args needing a prologue, and happening first, to mean we get a prologue for DTrace surely?
[23:40:17] <richlowe> we just want to... have one
[23:40:36] *** gwr <gwr!~gwr@151.203.114.175> has quit IRC (Quit: This computer has gone to sleep)
[23:41:21] <richlowe> rmustacc: my assumption would be that you can have code that only in one branch needs a prologue, and with shrink-wrap the code's arranged so we only go adjusting the stack, etc, in that arm.
[23:41:57] <richlowe> hm, and in rzezeski's disassembly we _do_ get the prologue before save-args
[23:42:04] <richlowe> we just... other stuff hoisted over it.
[23:42:11] <rmustacc> I see, OK.
[23:42:40] <jlevon> in that particular case it knows that if we're not going to drop into the switch, why bother with a frame
[23:42:42] <richlowe> I think it's hoisting a fast path completely over both the prologue and save-args?
[23:42:45] <jlevon> right
[23:43:16] <richlowe> jlevon: I think that's us agreeing, at least, yes.
[23:46:31] *** user888 <user888!~user888@pool-74-105-193-56.nwrknj.fios.verizon.net> has quit IRC (Ping timeout: 246 seconds)
[23:49:36] <LeftWing> oof
[23:49:44] <LeftWing> c o m p i l e r s
[23:50:12] <rzezeski> They know what's best for you Josh
[23:55:13] *** danmcd <danmcd!~danmcd@static-71-174-113-16.bstnma.fios.verizon.net> has joined #illumos
[23:57:52] <rzezeski> richlowe: attached your program to the ticket, thank you
[23:58:52] *** aszeszo <aszeszo!~aszeszo@unaffiliated/aszeszo> has joined #illumos
top

   August 29, 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 | 31 | >