target/6869: G++ 3.1 generates incorrect assembler instructions for HPPA 2.0

[law@redhat.com]

In message <200205310617.g4V6Hof3004691@hiauly1.hia.nrc.ca>, "John David Anglin
" writes:
 > > -       ble  R%$$dyncall(%sr4,%r2)       <- for PA1.1
 > > +       be,l  R%$$dyncall(%sr4,%r2)      <- for PA2.0
 > 
 > I have applied the following patch to the mainline.  It has been tested
 > with a bootstrap under hppa2.0w-hp-hpux11.11 with "-mpa-risc-2-0".
 > 
 > It should fix your problem.  Would you let me know if it does and
 > I will close the PR.
 > 
 > Dave
 > -- 
 > J. David Anglin                                  dave.anglin@nrc.ca
 > National Research Council of Canada              (613) 990-0752 (FAX: 952-66
 > 05)
 > 
 > 2002-05-31  John David Anglin  <dave@hiauly1.hia.nrc.ca>
 > 
 > 	* pa.c (output_millicode_call): Correct "be,l" insn for TARGET_PA_20.
 > 	(output_call): Likewise.
Ah yes.  Now I recall.  We fixed the assembler to require the explicit 
%sr0,%r31
per the PA2.0 manuals, but apparently forgot to update GCC at the same time.
Sigh.  It's really annoying when the syntax for an instruction changes like
this :(

Thanks for fixing this.

jeff