Re: Build time measurements

To: Andrew Doran <ad%netbsd.org@localhost>
Subject: Re: Build time measurements
From: Andreas Gustafsson <gson%gson.org@localhost>
Date: Wed, 25 Mar 2020 15:40:25 +0200

Andrew,

You wrote:
> Thank you for doing this, and for bisecting the performance losses over
> time (I fixed the vnode regression you found BTW).

Thank you for the fix and the other performance improvements!

> There are two options enabled in -current that spoil performance on multi
> processor machines: DIAGNOSTIC and acpicpu.  I'm guessing that you had both
> enabled during your test runs.

Yes, my tests so far have all been using unmodified GENERIC kernels.

> We ship releases without DIAGNOSTIC, and acpicpu really needs to be
> fixed.

I will rerun the 24-core tests with these disabled for comparison.
-- 
Andreas Gustafsson, gson%gson.org@localhost

References:
- Increases in build system time
  - From: Andreas Gustafsson
- Build time measurements
  - From: Andreas Gustafsson
- Re: Build time measurements
  - From: Andrew Doran

Prev by Date: Re: PATCH: Relax fdatasync checks to IEEE Std 1003.1-2008
Next by Date: Re: PATCH: Relax fdatasync checks to IEEE Std 1003.1-2008
Previous by Thread: Re: Build time measurements
Next by Thread: Re: Build time measurements
Indexes:

Home | Main Index | Thread Index | Old Index