[illumos-gate] 16569 want a more foolproof method to build native tools -- Richard Lowe <richlowe⊙rn>

nice:)

hi folks

nice to see so many peeps in the channel

I've not touch sun* tech in some time

used to love solaris 11

happy to see folks using illumos and it's derivatives

Looking at gdb, from the OI build. It has a few problems.

(vers 14.2)

Just ran across a strange one: Some sigacthandler is recursing until it runs out of stack.

[illumos-gate] 16633 nvme firmware updates cannot enforce update granularity -- Andy Fiddaman <illumos⊙fn>

The manpage for readdir_r says you have to call pathconf(directory, _PC_NAME_MAX) to find the size of the d_name[] array in it. Is there are documented maximum size that pathconf(directory, _PC_NAME_MAX) can return?

AustinWise: I suspect it will in practice always be less than PATH_MAX!

gwr: that's exactly the bug I filed against omnios-extra

gwr: omniosorg/omnios-extra #1476#issuecomment-2209509237

gwr: of particular note, in case it gets lost in my spew there, is when I try to debug it stops breaking

don't want you to miss that part and run into the same frustration by surprise

it also seemed to need me to be attached to a process in a different tty (for hadfl), but I'm not sure it needed that for me...

gwr: my next hope was to dtrace for delivery of sigint, and see the kernel stack

but then everything else started needing my attention

did any of you do a build today, and notice any packaging noise?

something odd happened to my most recent changeset, my workspace, or both.

did not see

==== package build errors (non-DEBUG) ==== this one is clean

I did an RTI build earlier, didn't see anything odd

hopefully git did something confused then. thank you.