Re: bcm283x_dwcotg0: usbpf: Out of memory

On 4/18/26 03:26, void wrote:
> On Fri, Apr 17, 2026 at 10:45:27AM -0700, Mark Millard wrote:
>> I'll note that the warnings about potential mis-tuning are based on the
>> total swap space enabled, not based on separate swap-space sizes that
>> contribute.
> 
> With 3* 1.5G partitions, the warning went away.
> Tested with swapoff -a then swapon -a which will normally
> provoke the warning if there's an issue.

(For reading reference: As noted in a later reply that you sent, the
warning was actually still being produced.)

> 
>> Normally WITH_META_MODE use automatically sets WITHOUT_CLEAN (NO_CLEAN)
>> when appropriate (or acts as if it had). 
> 
> Ah, I didn't know this.
> 
>> Explicitly setting
>> WITHOUT_CLEAN (NO_CLEAN) might set it when such is inappropriate. (A
>> possibility might be when newvers.sh has changed.)
> 
> Huh. Well, as it's going from 15-stable to current, I'll stop
> the build and take those parameters out of src.conf and start it
> again.
> 
>> FYI: To my knowledge, Bob P. does not use ccache for his build
>> activities. (Nor do I.) That might make comparison/contrast of results
>> messy.
> 
> How long does a buildworld in those circumstances take?

Bob builds on multiple types of systems, as do I. My note was general,
not RPi2B v1.1 specific. I've not done buildworld buildkernel on an
RPi2B in a very long time.

Bob does buildworld buildkernel on RPi2B v1.1's on a regular basis. But
he likely would have to answer with a fairly wide range of time frames,
because of incremental-build variability and full builds --and use of
WITH_META_MODE.


-- 
===
Mark Millard
marklmi at yahoo.com