re: Kernel memory allocation oddities in NetBSD-10.99.12

[matthew green <mrg%eterna23.net@localhost>]

> 	The allocation category which fails the most is the pcgnormal, which is a fixed buffer of
> 256 bytes per allocation.  
>
> 	One question I have is, why would pcgnormal fail so often, while, at the same time, 
>
> execargs    262144 2436031   0  2436030 30522 30509    13    16     0    16   12
>
> which allocates far larger block sizes far more often, experience no allocation failures?  
>
> If execargs can always find 200,000 bytes to allocate, how is it that pcgnormal or xnfrx, which
> allocate 256 bytes and 4096 bytes respectively, cannot?

it's probably due to a feature of pools.

there are only ever a small number of active execargs in use,
but after the system has been up for a while, most CPUs will
have at least one of them cached.  this means that normally it
will always find something to return for exec's usage.  these
other pools are far more active, usually, in both their number
of active allocations and rate of allocation.


can you explain why you think the lock up is related to this?
i've seen one lock up of recent -current one so far.

thanks.


.mrg.