[Bug 1740] Change PowerPC multi-libs for e200 and e500 cores

[bugzilla-daemon at rtems.org]

https://www.rtems.org/bugzilla/show_bug.cgi?id=1740

--- Comment #4 from Ralf Corsepius <ralf.corsepius at rtems.org> 2011-02-16 07:14:13 CST ---
(In reply to comment #0)
> Created an attachment (id=1109)
 --> (https://www.rtems.org/bugzilla/attachment.cgi?id=1109) [details]

> One question to a part of it in t-rtems:

> +# Map e500, 8548 to 8540
> +MULTILIB_MATCHES       += mcpu?8540=mcpu?e500

> Why do we need this?
Seemingly no ... I believe to recall it once was used in RTEMS and once was a
valid gcc option ... this can be removed.

(In reply to comment #1)
> Are the names "embfp32" and "embfp64" in use elsewhere?
I will not allow the RTEMS toolchains being infected by trademarks and brands.

(In reply to comment #2)
> Ok, we can also remove the MULTILIB_DIRNAMES completely, but I guess this will
> break some Makefiles.
It should not, because MULTILIB_DIRNAMES are supposed to be transparent
=> Any Makefile which explicitly uses them, has to be considered 
  "broken by definition"

The purpose of MULTILIB_DIRNAMES is to convert "cryptic" gcc options into
something better readable, so gprs1 or gprs2 would be possible candidates (the
names gcc uses internally) or even gprs_float or gprs_double.


Final questions (I haven't tried this patch, yet):
* How many multilibs is this patch adding?
* Do all current RTEMS-4.10 and rtems-4.11 BSPs build with these changes
applied (w/ and w/o multilibs)?

-- 
Configure bugmail: https://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.