Re: Patch to cache TLS in %fp7 to avoid _lwp_getprivate syscalls

[Jake Hamby <jake.hamby%gmail.com@localhost>]

Thanks for the link. I'll investigate the two approaches you discussed
and try to implement one of them. I was pleased with the hack that I
came up with, since it modified so little code and works great on my
system, but I completely agree that it's not the correct solution for
the platform.

Michael, thanks for the context about the TAS instruction. It's good
to know that it's not a problem for 68020+.

-Jake


On Tue, Jan 17, 2012 at 2:14 PM, Joerg Sonnenberger
<joerg%britannica.bec.de@localhost> wrote:
> On Tue, Jan 17, 2012 at 02:03:32PM -0800, Jake Hamby wrote:
>> I recently started hacking on NetBSD/amiga and have been working on a
>> patch that I'd like to contribute to -current. I still need to wrap
>> the code with an appropriate #ifdef and do some more testing, but I'd
>> like to get some feedback and find out if there's an open PR for this
>> issue, or if I should create a new PR for the patch. I've attached the
>> current version with comments below. It's only beneficial for m68k
>> systems with a hardware FPU.
>
> May I interest you in looking at a different approach for this?
> It would be far preferable to not depend on the %fp7 register to be
> unused and instead implement either of the two approaches in
>
> http://mail-index.netbsd.org/port-m68k/2011/11/19/msg000303.html
>
> Other than that, disabling the spinning in the mutexes as platform
> option sounds valid.
>
> Joerg