00:23:00 <rmustacc> it's also worth noting that the IP interface has an MTU that starts from the underlying datalink, but is truly separate. That said, the question of whether it binds or copies is dependent on several things, such as the number of mblks, etc.
00:24:01 <rmustacc> The choice to bind / copy is done well before that point though. The ring tx path calls mlxcx_buf_bind_or_copy() which iterates over the mblk chain.
00:24:50 <rmustacc> I would look to see some of the stats and DTrace probes that are there to help figure out what's going on.
00:25:18 <rmustacc> The only copying that takes place in the aforementioned function is for the headers when they're inlined into the descriptor. Otherwise the descriptor chain is all set up in advance of that.
13:52:55 <danmcd> yuripv: Sorry for that delay.  Running a safety nightly now before I push both.
14:22:33 <jbk> gwr: while looking at something unrelated, I came across this which might be relevant to your question: https://src.illumos.org/source/xref/illumos-gate/usr/src/stand/lib/tcp/tcp.c?r=b531f6d1#632
14:23:16 <jbk> (it does make me wonder what the impact of that on things like LRO might be)
14:32:15 <yuripv> danmcd: no worries, thank you
14:32:17 <gitomat> [illumos-gate] 13313 tools build uses cw from /opt/onbld even if -t flag is specified -- Yuri Pankov <ypankov⊙tc>
14:32:17 <gitomat> [illumos-gate] 15012 nightly should fail if tools build failed -- Yuri Pankov <ypankov⊙tc>
14:32:23 <danmcd> Good timing... 
14:32:28 <yuripv> :)
14:33:01 <danmcd> How often does that allocb fail, jbk?
14:33:18 <danmcd> Oof, also it uses "0" as PRI, which is unreadable AF.
14:34:22 <jbk> err
14:34:26 <jbk> too early still
14:39:46 <danmcd> ENOCAFFEINE perhaps.  I still stand by using 0 as priority is Bad.
14:40:41 <jbk> yeah, but that's a separate tcp stack :) which i realized later
14:41:33 <jbk> which uses a lot of the same function names as the 'real' tcp stack
14:42:16 <jbk> and at the very real danger of ENOCAFFEINE twice...
14:43:13 <jbk> i don't suppose anyone still has a setup where bhyve<->non-VM zone still hangs regularly... I was never able to reproduce it locally
14:44:03 <jbk> but while hunting down something unrelated (mostly just curious)... I found something that'd be nice to at least eliminate
14:51:50 <jbk> (it still annoys me i was never able to chase that down)
14:52:19 <toppk> i'm having lots of pkg issues on hipster, e.g. cannot install git, cannot update.  is that expected?
15:04:15 <jbk> what sort of errors are you getting?
15:13:26 <yuripv> can't gerrit auto-close the reviews once it's integrated? "abandon" feels wrong somehow
15:19:01 <toppk> install git gives "This version is excluded by installed incorporation consolidation/userland/userland-incorporation⊙0"
15:19:45 <toppk> update gives "No version for 'incorporate' dependency on system/library/processor⊙0 can be found"
15:22:11 <toppk> trying to run "pkg update -v pkg://openindiana.org/consolidation/userland/userland-incorporation⊙0:20221009T143209Z" encounters lots of  No version matching 'require' dependency errors
15:29:40 <yuripv> looks like "noincorp" was forgotten when obsoletting system/library/processor?
15:31:58 <Smithx10> Has anyone created a prometheus node_exporter that supports illumos?
15:33:45 <jperkin> well, node_exporter runs on illumos and we have a package for it..
15:33:54 <jperkin> or was there something specific you needed?
15:39:35 <Smithx10> nah,  just curious that it ran 
15:40:24 <jperkin> yeh it seems to work fine and at least gives basic OS metrics
15:47:25 <toppk> yuripv: seems that way, I will create a PR and see what happens.   Any advice for the "pkg install git" issue?  "This version is excluded by installed incorporation consolidation/userland/userland-incorporation⊙0"
15:48:01 <toppk> Am I supposed to do a change-facet or just remove the incorporation?  (im new to all this)
15:52:00 <Kernspin> toppk: After installing/upgraqding with IPS pkg manager, you most probably need to reboot into the boot environment.
15:52:14 <Kernspin> toppk: (new boot environment)
16:00:48 <toppk> kernspin: I should have mentioned, the first update updated 27 packages, and seemed to work fine, but new BE cannot do another update.  neither the original nor the new BE could install git
16:03:27 <Kernspin> toppk: Ok!
16:14:25 <pmooney> tsoome: I'm seeing a slew of these clang tickets with 'unix' in the title when they really apply to a different module
16:15:22 <pmooney> which makes it seem like they apply to the core 'unix' portion of the kernel, rather than separately loaded modules
16:16:03 <tsoome> pmooney oh, vmm is built separate modure, my bad, I'll fix it.
16:16:11 <pmooney> thanks
16:16:26 <tsoome> pmooney I was running make in i86pc ;)
16:16:44 <pmooney> its built under i86pc because it reaches into some machine-arch stuff
16:16:54 <pmooney> but in the end it emits a separate module
16:17:37 <tsoome> yep, there are some...
16:18:29 <pmooney> probably applies to 15066 as well (although being grouped under pcie)
19:54:11 <tsoome> yep
20:01:00 <KungFuJesus> errmm, what's this one mean: https://pastebin.com/MefHdS5F
20:01:48 <richlowe> system/library/processor is obsolete, and it seems confused
20:02:17 <richlowe> perhaps something still depends on it somewhere?
20:02:27 <richlowe> or could do, in whatever pkg is trying to achieve for you
20:02:40 <KungFuJesus> just doing "update"
20:02:51 <KungFuJesus> I did update earlier, and it seemed...suspiciously small
20:03:00 <KungFuJesus> so after the reboot I tried again
20:03:20 <KungFuJesus> image-update is just an alias to update, no?  Nothing different?
20:03:25 <richlowe> yeah, nothing different.
20:03:48 <richlowe> pkg search for things that need that package might help? if that's what's making it not just remove it
20:04:03 <KungFuJesus> how do I find dependent packages?
20:04:08 <richlowe> pkg(8) has examples of searching based on dependencis, I never remember it
20:04:19 <richlowe> :depend:... something
20:04:24 <richlowe> if you search the manual for :depend: it'll show up
20:04:27 <richlowe> or depend:?
20:09:12 <KungFuJesus> returns empty: 
20:09:24 <richlowe> then I'm out of immediate ideas :(
20:10:00 <KungFuJesus> https://pastebin.com/Az72eacA
20:17:20 <KungFuJesus> richlowe: so, doing -vvv does give more info: https://pastebin.com/70BpwQ6z
20:21:31 <richlowe> I have no idea what could cause you to have SUNW* packages still, so that looks weird.
20:23:37 <KungFuJesus> is it safe for me to remove system/library/processor?
20:23:51 <KungFuJesus> it seems the apache package on there is needed for system-repository
20:23:55 <richlowe> I don't think it will work, or help
20:24:09 <richlowe> the SUNW* one is?
20:24:34 <KungFuJesus> no, web/server/apache24, which comes up under pkg list
20:24:35 <richlowe> this seems like something you'd want to talk to #oi-dev about
20:24:57 <richlowe> yeah, it's the stuff about SUNWapch22  and whatnot I don't get
20:28:56 <KungFuJesus> pretty quiet in there :-/
20:29:28 <richlowe> only other place I know of is the mailing list unfortunately
20:33:41 <KungFuJesus> hmm, well, mailed the ML
20:35:57 <KungFuJesus> https://openindiana.org/pipermail/oi-dev/2022-October/033281.html
20:36:01 <KungFuJesus> hmm, seemingly not alone
21:02:58 <gitomat> [illumos-gate] 15040 remove support for unused sun studio inline asm -- Richard Lowe <richlowe⊙rn>