05:52:17 <tsoome> rmustacc sure. 
05:54:54 <tsoome> rmustacc but I have identified the problem is that we have already closed the errlog as part of the fmd shutdown, we were attempting to close fltlog, but got error while doing it, and since errlog is already closed, reporting that error did end up with dereferencing NULL pointer and crash
06:14:13 <tsoome> rmustacc so the basic question is, should we just check this pointer to be not NULL, or should fmd.running flag be reset earlier or something like that...
13:07:16 <ChadGPT> hello. for science(tm) i'm trying to find out real-world 'nice' handling on various systems. here is a test program which compiles on linux and the bsds, i expect it will also compile with minor tweaks on illumos
13:07:24 <ChadGPT> https://dpaste.com/8LJ6FUNLJ
13:07:34 <ChadGPT> and here is a test script for it: https://dpaste.com/C9GB22FJ3
13:07:51 <ChadGPT> looking for someone(tm) with a *bare metal* illumos install who would be willing to run this
13:08:13 <ChadGPT> for the test to work it will need to be bound to one cpu, i don't know how to do it on illumos
13:08:24 <ChadGPT> on linux it would taskset
13:08:57 <ChadGPT> looks like pbind(8) will do it
13:09:21 <ChadGPT> pbind -e 3 sh script.sh
13:09:22 <ChadGPT> or so
13:09:40 <ChadGPT> cc -O0 -o cpuburner-prio cpuburner-prio.c
13:16:09 <otis> let me try it on illumos
13:17:09 <ChadGPT> make sure the box is mostly idle, or at least nothing messes with the target cpu
13:17:13 <ChadGPT> also thank you :)
13:20:37 <otis>  68678: prio  19 ops      14796
13:20:37 <otis>  68679: prio   0 ops      14821
13:21:04 <ChadGPT> can you verify on top or similar that they stay on the same cpu for the duration of the test?
13:23:09 <otis> oh no, it's being migrated. let me exec using pbind
13:24:21 <otis> ok, pbind -e 3 helped (3 is the most idle core)
13:24:38 <otis>  69320: prio   0 ops       7419
13:24:39 <otis>  69318: prio  19 ops       7430
13:25:06 <ChadGPT> ok this adds up, but is also broken :)
13:25:22 <ChadGPT> what's your uname
13:25:39 <otis> joyent_20230309T001400Z
13:25:49 <otis> (smartos)
13:25:58 <ChadGPT> thank you for testing
17:33:42 <sommerfeld> ChadGPT: the answer can vary dramatically depending on the choice of scheduling class.  
17:34:13 <sommerfeld> TS: 196184: prio  19 ops       9968
17:34:13 <sommerfeld> 196185: prio   0 ops      11399
17:34:32 <sommerfeld> IA: 196199: prio   0 ops      21377
17:34:32 <sommerfeld> 196198: prio  19 ops         41
17:34:55 <sommerfeld> FSS: 196210: prio   0 ops      10656
17:34:55 <sommerfeld> 196209: prio  19 ops      10693
17:35:48 <sommerfeld> RT: + nice -n 20 ./memory_burner 1048576
17:35:48 <sommerfeld> nice: invalid operation; scheduling class does not support nice
17:35:48 <sommerfeld> + nice -n 0 ./memory_burner 1048576
17:35:48 <sommerfeld> nice: invalid operation; scheduling class does not support nice
17:35:51 <sommerfeld> :-)
17:42:05 <sommerfeld> (not sure why the impact is so high for IA vs TS as the documentation makes them seem near-identical)
17:49:00 <alanc> IA should be just TS plus the ability for the X server to give boosts to processes with focus
17:52:43 <sommerfeld> it appears that when you put a process into IA it starts boosted.
18:53:33 <richlowe> tsoome: I noticed due to a mistake that libscf has at least one smatch warning that seems worth checking, and I guess clang might notice too.  You might want to remove the SMOFF and take a glance
18:56:36 <richlowe> I'm starting to suspect that we use calloc() with the arguments the wrong way around more often than not
18:59:14 <richlowe> tsoome: see msg
19:51:06 <jbk> i've noticed that
21:28:11 <jbk> of course a big chunk of the uses of calloc are basically as 'zalloc()' for userland
21:35:20 <alanc> I don't think any of the calloc implementations in the gate actually care - it's not like they try to align or pad the individual unit size, just multiply it out