hm, is it me, or src.illumos.org is returning 504 Gateway Time-out

Timing out for me too

ok....

btw, regarding kadmind crash on omnios - it happens because of garbage value in krb5_keytab structure (data field should point to krb5_ktfile_data but has bad pointer), butI do not know yet why we have it there and why mech_krb5.so.1 built from illumos-gate does not have this problem.

Bizarre.

[illumos-gate] 16519 panic when removing SADB entry with tcpkey delete -- Andy Fiddaman <illumos⊙fn>

[illumos-gate] 16502 panic in zfs:abd_iter_map -- Bill Sommerfeld <sommerfeld⊙ho>

lordy opengrok is a pain in the arse

indeed it is

I'm sorting it out, but it'll be a little bit haha

I think it probably needs a bigger VM now for whatever reason

them Java beans don't come cheap

well, we ask it to look at a lot

though I can't see what, because it's broke, haha

I just know freebsd used to be in there, and that's a lot of history

on my local instance, linux seems to be a lot worse than freebsd

it would be, because of their process, as I understand it

they have many many revisions

freebsd only have 300,000

(that's also why I never asked jclulow to add gcc to src.illumos.org)

richlowe: yes - the linux process has an extreme bias towards tiny patches and long patch series and lots of branches and merges.

saw stuff like "add new driver for device X" start by introducing a driver skeleton and then patching it a bit at a time into a working driver.

Is that Erik N's classic line "debug into existence"?

IMHO no.

because the driver is "done" (or at least functionally complete) before the patches are mailed in.

That is what I see happening in FreeBSD's bhyve too of late, but some of the tiny interim patches there don't even compile.

i mean, i can see breaking it up into smaller functional pieces to aid in reviewing, but also a bunch of trivial patches can also be annoying too...

Yes, there has to be a balance somewhere

interim states that don't compile or cause breakage make bisection harder.

amen

this is rule #1

I never realized I depended so much in opengrok, I'll return to cscope in the meantime

[illumos-gate] 16491 netcat should support setting the outgoing and minimum TTL -- Robert Mustacchi <rm⊙fo>

[illumos-gate] 16508 snoop incorrectly prints the IPv6 Traffic Class -- Robert Mustacchi <rm⊙fo>

[illumos-gate] 16352 strftime_l() should be more careful with pointer arithmetic -- Bill Sommerfeld <sommerfeld⊙ho>