Re: gcc-2.95.3, but not gcc-2.95.2, requires gas on hppa2.0-hp-hpux10.20

[Bernd Schmidt <bernds at redhat dot com>]

On Tue, 23 Jan 2001, Jeffrey A Law wrote:

>
>
>   In message <Pine.LNX.4.30.0101231637410.1128-100000@host140.cambridge.redhat.
> com>you write:
>   > So... do you recommend that this should be treated as a showstopper bug,
>   > or is gcc not expected to work using the HP assembler?  I'm trying to
>   > figure out why it now fails, but if there are random bugs in the HP
>   > assembler there's probably not much of a point in doing so.
> The failures aren't random -- they're pretty predictable.
>
> The one that crops up most often is there's no way to generate suitable
> code for an indirect call to a static function where one of the parameters
> passed in registers is a floating point value.
>
> What kinds of problems are you seeing with gcc-2.95.3 on hpux?  [ I really
> haven't followed the discussion. ]

It turns out it isn't an assembler problem, but a compiler problem.  I was
suprised to notice that gcc emits different code depending on whether it uses
gas or not; the problem only occurs if not using gas.

The problem is that we have an apparently unused exception handler which is
optimized out during flow, but a reference to its label survives in a
constant pool.  It's a bit odd that the label got deleted, since the load
from the constant pool has a REG_LABEL note and you'd think flow would count
that as a use.


Bernd