Message-ID:	<000b01be9054$2fc75dc0$4264a8c0@azagrath.org>
From:	"Leo van den Berg" <goezzi AT xs4all DOT nl>
To:	<pgcc AT delorie DOT com>
References:	<tpr3e1mexh0 DOT fsf AT cs DOT ucsb DOT edu>
Subject:	Re: [1.1.3] genattrtab exposes optimization bug
Date:	Tue, 27 Apr 1999 04:18:00 +0200
MIME-Version:	1.0
X-Priority:	3
X-MSMail-Priority:	Normal
X-Mailer:	Microsoft Outlook Express 5.00.2014.211
X-MimeOLE:	Produced By Microsoft MimeOLE V5.00.2014.211
Reply-To:	pgcc AT delorie DOT com

respect for pgcc 1.1.3 now, I managed to sucessfully compile my kernel,
isdn4linux and pgcc itself with these flags on a i586-pc-linux-gnu system:
(pgcc partly with rtti & execptions :)

-Wall -Wstrict-prototypes -O6 -fomit-frame-pointer -mieee-fp -malign-double 
-mstack-align-double -funroll-all-loops -march=pentium -fno-exceptions -fno-
rtti -malign-loops=2 -malign-jumps=2 -malign-functions=2

this will become my standard CFLAGS if no further problems occur, any hints
on further optimalisations?

----- Original Message -----
From: Jerry James <jerry AT cs DOT ucsb DOT edu>
To: <pgcc AT delorie DOT com>
Sent: Tuesday, April 27, 1999 2:08 AM
Subject: [1.1.3] genattrtab exposes optimization bug


>
> I wanted to see how well pgcc can optimize itself, just to get some idea
> of how trustworthy its optimizations are.  My first build, with -O6 used
> for everything, crashed when it tried to run genattrtab (with a
> segmentation fault).  My experiments indicate that the problem is with
> the compilation of genattrtab.c itself, although I haven't tracked it
> down to a specific point in the code.  If anyone is interested in
> tracking this down, here's what I know so far:
>
>   -O3: works
>   -O4: segmentation fault
>   -O4 -fno-risc: works
>   -O5 -fno-risc: segmentation fault
>   -O5 -fno-risc -fno-omit-frame-pointer: works
>   -O6 -fno-risc -fno-omit-frame-pointer: works
>
> This is on a Pentium II-based Solaris 7 system.
> --
> Jerry James
> Email: jerry AT cs DOT ucsb DOT edu
> WWW:   http://www.cs.ucsb.edu/~jerry/
>

- Raw text -