A good place to check for historical runs is the weekly merge PRs, I'll take a look.

hadfl's build times have been 36:02 (2nd of June) 35:40 (29th) and 35:35 (8th of July)

my builds have been ~1:05 consistently in early june, but ~1:12 now

I just went back to the old BE, and the build time got better.

but I can't see any evidence of _why_

Can you break it down by phase from mail_msg?

not in a way that helps

the "build time" time stamp only got 2 minutes longer

I'm not user the tools build would hold up to the rest.

s/user/sure/

for my last two builds from before/after I went back to the june 9th BE: full time new BE: 1:11:23, old be: 1:03:16. "elapsed build time" new BE: 57:03.1 old BE: 54:24.2

normally I'd ignore it, but I'm both desparate to confirm it's not my fault in my project work, and worried because last time it felt off but I didn't check, was the gcc thing you fixed

We've definitely had build time regressions in the past, which is how I know to go and look at the ~weekly PRs

so I wanted to at least report, and see if anyone else could reproduce/had noticed/something.

It doesn't look like we've had any on those omnios upstreaming builds

thanks

andyf: does the non-hotload version of loading the microcode persist across reboots, or is this something where to remain safe you must stay on up-to-date versions of illumos (or Linux or whatever)? I just want to be certain before advising anyone else.

wrong channel. sigh

gonna move my question to #illumos for the sake of posterity.

There will be an OmniOS update in a day or two to get this loaded during boot, but it has to get through review and testing.