On Jun 9, 1:37 pm, Jason <ja...@njkfrudils.plus.com> wrote:
> Hi
>
> Now that 2.4 is almost ready to go , here's some ideas for mpir-2.5
>
> 1) Release on 1st September
>
> 2) upgrade internals if needed ie yasm,autotools,gnu config
>
> 3) New x64 assembler code
>
> 4) remove explicit support for old cpus sh,m68k,alpha?


There are few cpu's which I think MPIR has not been tested on

mpn/sh  see

http://en.wikipedia.org/wiki/SuperH

This architecture has been expanded to 64bit :) , but the only asm
code MPIR has is sh2 which is used in automotive apps :(

mpn/m68k see

http://en.wikipedia.org/wiki/Motorola_68000_family

The last incarnation of this family was approx equal to a pentium ,
and considering the comments from the readme
"The code herein is old and poorly maintained.  If somebody really
cared, it could be optimized substantially. "

mpn/thumb see

http://en.wikipedia.org/wiki/Thumb-2#Thumb-2

This was a variant of the ARM chips with a compressed instruction
stream , and we only have add/sun.asm anyway and how optimized is the
old code for the current microarchitecture of ARM chips?

For MPIR-2.5 I'm going to remove all explicit support for the above
arches , generic C will of course work for them and at least it's been
tested.

mpn/alpha see

http://en.wikipedia.org/wiki/Alpha_CPU

GCC-farm's alphaev56 has been down for ages , and thats the only one I
know of , and 56 is OLD , another dead server chip (replace by the
itanium....) , there is considerable work been done for it though , it
does seem a waste. The advantage of turning to generic C for the
alpha , is that quite  lot of simplification of the autotools
configure can be done :) ie a lot of special case code for something
we never use and that gets in the way whenever we try to simplify
things.

For MPIR-2.5 I'm pretty sure this should be removed as well , it's a
little more contentious than the other three and if you object speak
up.

All the other specific arches we do test on , although I haven't
checked any sub-arches.

Jason

-- 
You received this message because you are subscribed to the Google Groups 
"mpir-devel" group.
To post to this group, send email to mpir-devel@googlegroups.com.
To unsubscribe from this group, send email to 
mpir-devel+unsubscr...@googlegroups.com.
For more options, visit this group at 
http://groups.google.com/group/mpir-devel?hl=en.

Reply via email to