Re: cvs-src summary for August 2-9

Ruslan Ermilov <ru_at_freebsd.org> wrote:
> On Tue, Aug 10, 2004 at 10:28:39PM -0500, Mark Johnston wrote:
> > Cryptography in releases and legal concerns
> > -------------------------------------------
> >
> > Nate Lawson (njl) moved the crypto distribution into base, making all
> > releases cryptography-enabled.  He noted, "The -DNOCRYPT build option
> > still exists for anyone who really wants to build non-cryptographic
> > binaries [ . . . ]."
>
> It was Colin Percival who did the change.

Argh!  My apologies to Colin - I still can't figure out how I managed to do 
that.

> > mbuf exhaustion panic fixed
> > ---------------------------
> > Brian Feldman (green) changed the UMA (uniform memory access) code,
>
> UMA stands for the Universal Memory Allocator.

Ah, thanks.  I had only heard of UMA as part of the term "NUMA", and I was 
extrapolating from that, but actually looking up what NUMA is shows "uniform 
memory access" to be clearly not something you'd write special code for.  For 
anyone who shared my confusion, "non-uniform memory access" is when some 
memory is slower than other memory, like in a clustered system.  "Uniform 
memory access", where all access is the same speed, is obviously the normal 
case.

For those who just use the Web page, I'm going to correct these errors in the 
published summary.  I'm not trying to pretend I don't screw up, but I think 
the technical accuracy of what's on the Web is more important than preserving 
a record there of everything that was said.  If this surprises or upsets you, 
please feel free to use the -current Web archives to continue to refer to the 
original version of the summary, including errors.

Mark