In CCing Jörg.

Richard Henderson schrieb:
On 08/02/2011 12:52 AM, Georg-Johann Lay wrote:

There are still unrecognizables:

gcc.c-torture/execute/complex-7.c:56:1: error: unrecognizable insn:
(insn 17 14 18 3 (set (mem:SF (post_dec:HI (reg/f:HI 32 __SP_L__)) [0 S4 A8])
       (reg:SF 43 [ f5.0+4 ]))
    (nil))

I was pretty sure I ran the compile tests.  I've tried several
times to come up with an environment that would properly run
the simulator, without success.  AVR support seems to be in
too many different places, none of which properly communicate
with each other.

Just ask :-)

For questions/answers you may want to read/post to the looow traffic avr-gcc-list:
   http://lists.gnu.org/archive/html/avr-gcc-list/

== binutils ==

configure plain vanilla: --target=avr --prefix=<same-as-avr-gcc>
build and install

== gcc ==

You've done that already. I'm using something around
configure --target=avr --enable-languages=c,c++ --disable-nls --prefix=<same-as-binutils> --with-dwarf2

== avr-libc ==

Needed to run because it provides startup code and C libs.
A bit tricky, easiest to use is current CVS head.

With the latest version 1.7.1 from
http://download.savannah.gnu.org/releases/avr-libc/
you will need the patches
http://svn.sv.gnu.org/viewvc?view=rev&root=avr-libc&revision=2239
http://svn.sv.gnu.org/viewvc?view=rev&root=avr-libc&revision=2241

configure with --host=avr --prefix=<same-as-avr-gcc> CC=avr-gcc
If you give CC= (e.g. if your avr-gcc is noz in PATH) note that
the name must contain "avr" i.e. build using CC=xgcc or so does
not work. In-tree build is not supported.

Building is currently blocked by PR49864 so you have a dead-lock
and may want to downgrade gcc or avr BE to r177070 to build avr-libc.

Install it.

If I overlooked something Joerg will correct me.

== avrtest ==

There is a text
http://lists.gnu.org/archive/html/avr-gcc-list/2011-06/msg00015.html
and a README:
http://winavr.cvs.sourceforge.net/viewvc/winavr/avrtest/README?view=markup

In the case there are questions: Ask.

That said, this fixes that test case, committed as obvious.

Didn't try it yet. Is that capable of fixing the runtime FAILs?

Johann


r~

Reply via email to