BOOTSTRAP FAILURE: segementation fault in genattrtab under hpux 10.20

[John David Anglin]

> On Sun, Nov 12, 2000 at 02:49:29PM -0500, John David Anglin wrote:
> > > > As can be seen, the return pointer rp has been used as a scratch register.
> > > > Its value at the first return is 0.
> > 
> > I added a `(use (reg:SI 2))' to the return insn to work around this problem.
> 
> Hmm, it appears some general cleanup is needed in this area.
> Try the following.

Tried it.  The result was the same as with my hack (see below).

> > Another problem with regrename?
> 
> Dunno.. does it work with -fno-rename-registers?

Still have this problem.  Will try to dig deeper tomorrow.

Dave
-- 
J. David Anglin                                  dave.anglin@nrc.ca
National Research Council of Canada              (613) 990-0752 (FAX: 952-6605)

2000-11-11  John David Anglin  <dave@hiauly1.hia.nrc.ca>

	* pa.md (return, return_internal): Modify patterns to prevent regrename
	mucking with the return pointer.

--- pa.md.orig	Fri Nov 10 16:00:56 2000
+++ pa.md	Sat Nov 11 16:02:37 2000
@@ -5667,7 +5667,9 @@
 ;; Unconditional and other jump instructions.
 
 (define_insn "return"
-  [(return)]
+  [(return)
+   (use (reg:SI 2))
+   (const_int 0)]
   "hppa_can_use_return_insn_p ()"
   "*
 {
@@ -5682,7 +5684,8 @@
 ;; epilogues so as not to confuse jump and reorg.
 (define_insn "return_internal"
   [(return)
-   (use (reg:SI 2))]
+   (use (reg:SI 2))
+   (const_int 1)]
   ""
   "*
 {