coro: Asm coro_transfer for arm7

Marc Lehmann schmorp at
Sat Dec 5 21:50:14 CET 2015

On Wed, Dec 02, 2015 at 02:25:52PM +0300, Nick Zavaritsky <mejedi at> wrote:
> Sorry for nagging, but I’ve got no response on this.
> Could you please consider my patch?

Sorry for the delay, the good patches take a bit longer :)

And yes, arm and mips are somewhat high on my todo list, so thank you for
doing this work.

I have a few questions on the specifics of your patch, though:

1. in the header, you test for __arm__, but in the implementation,
you only support __ARM_ARCH=7, this will obviously fail. Is this an
oversight? I would assume testing for __ARM_ARCH everywhere is the right

2. The streamlined CORO_STARTUP code is something I'd like to avoid unless
it's superuseful (I'd like to avoid the complications because I don't
think they are worth the speed gained, unless it would be substantial) -
do you have any benchmark or some other data that caused you to optimise
this case, or did you just do it while you were at it?

You don't have to send an update to your patch, the above questions are
just so I have a better idea of what to do.


                The choice of a       Deliantra, the free code+content MORPG
      -----==-     _GNU_    
      ----==-- _       generation
      ---==---(_)__  __ ____  __      Marc Lehmann
      --==---/ / _ \/ // /\ \/ /      schmorp at
      -=====/_/_//_/\_,_/ /_/\_\

More information about the libev mailing list