tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: mutexes, locks and so on...

On 11/15/10 14:55, Johnny Billquist wrote:
On 11/14/10 20:16, David Holland wrote:
On Sat, Nov 13, 2010 at 01:45:40AM +0900, Izumi Tsutsui wrote:
> > Wow. I guess you can add me to the list of people leaving.
> There is no perfect world and we don't have enough resources.
> Any help to keep support for ancient machines are appreciate, but
> complaints like "we should support it" which prevents improvements
> of mainstream will just make NetBSD rotten.

What "prevents improvements of mainstream" are we talking about here?
We have someone who wants to provide tuned vax-specific locking
primitives. The absolute worst possible cost to the "mainstream" that
this incurs is a bit of extra cpp and config hackery.

(Can we all please get a grip?)

Yes, please.

And... Wow, just for my seaching for a spin mutex lock against myself at
boot time, I started counting the number of times the uvm_fpageqlock
spin mutex is taken. By the time the
"boot device: xxx" is printed out, that spin-mutex have been taken
423137 times.

That looked like a big number to me atleast. (And yes, it is that mutex
that is taken twice, which traps me right now. Still looking at why it
happens, but I know that the "normal" code does not in fact check for
this condition, and happily just accept that you take a spin-lock that
is already taken, when you're not multiprocessor.)

Aw. Crap. Forget that. Error in my mathematics, since I had two numbers, and one happened to be decimal, and the other hex...

The actual number of spin mutexes taken up to that point is 14524. Still a hefty number, but nowhere near what I first though I saw.


Home | Main Index | Thread Index | Old Index