public inbox for gcc-prs@sourceware.org
help / color / mirror / Atom feed
* Re: optimization/1762, please reopen
@ 2001-06-25 11:16 Richard Henderson
0 siblings, 0 replies; 3+ messages in thread
From: Richard Henderson @ 2001-06-25 11:16 UTC (permalink / raw)
To: rth; +Cc: gcc-prs
The following reply was made to PR optimization/1762; it has been noted by GNATS.
From: Richard Henderson <rth@redhat.com>
To: Reichelt <reichelt@igpm.rwth-aachen.de>
Cc: rth@gcc.gnu.org, gcc-gnats@gcc.gnu.org, gcc-prs@gcc.gnu.org,
gcc-bugs@gcc.gnu.org
Subject: Re: optimization/1762, please reopen
Date: Mon, 25 Jun 2001 11:08:34 -0700
On Mon, Jun 25, 2001 at 06:57:58PM +0200, Reichelt wrote:
> I'm afraid the bug reappeared with the 3.0 release.
> So, please reopen the PR.
The test case still passes for me. Is the test not in fact
representative of your code?
r~
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: optimization/1762, please reopen
@ 2001-06-26 4:36 Reichelt
0 siblings, 0 replies; 3+ messages in thread
From: Reichelt @ 2001-06-26 4:36 UTC (permalink / raw)
To: rth; +Cc: gcc-prs
The following reply was made to PR optimization/1762; it has been noted by GNATS.
From: Reichelt <reichelt@igpm.rwth-aachen.de>
To: rth@redhat.com
Cc: rth@gcc.gnu.org, gcc-gnats@gcc.gnu.org, gcc-bugs@gcc.gnu.org
Subject: Re: optimization/1762, please reopen
Date: Tue, 26 Jun 2001 13:43:18 +0200
---2038259397-2078917053-993555807=:708930
Content-Type: TEXT/plain; charset=us-ascii
Hello again.
On 25 Jun, Richard Henderson wrote:
> On Mon, Jun 25, 2001 at 08:39:11PM +0200, Reichelt wrote:
>> I still get the wrong results with the code snippet from the PR.
>> But it's target specific (maybe I should have mentioned that more
>> explicitly): On an x86-linux machine everything works fine
>> (return value "0"), but on the SGI I get "1" instead.
>>
>> Did you check it on an SGI?
>
> Yes.
>
> $ ./cc1 -quiet -v -O z.c
> GNU CPP version 3.0.1 20010622 (prerelease) (cpplib) [AL 1.1, MM 40] SGI running IRIX 6.x
> ignoring nonexistent directory "/usr/local/include"
> ignoring nonexistent directory "..."
> ignoring nonexistent directory "..."
> #include "..." search starts here:
> #include <...> search starts here:
> /usr/include
> End of search list.
> $ gcc z.s
> $ ./a.out
> $ echo $?
> 0
> $ uname -a
> IRIX ada 6.3 12161207 IP32
>
> So. What are you doing differently?
I took me a while to find out, but apparently it's a matter of mips3
or mips4 (with mips4 generating the wrong code):
$ gcc -mips4 -mabi=n32 -O NaNBug1762.c; ./a.out; echo $?
1
$ gcc -mips3 -mabi=n32 -O NaNBug1762.c; ./a.out; echo $?
ld32: WARNING 158: Expecting MIPS3 objects: /usr/LOCAL/lib/gcc-lib/mips-sgi-irix6.5/3.0/crtbegin.o is MIPS4.
0
I'm using IRIX 6.5 and the default configuration is mips4.
You seem to be using IRIX 6.3. Maybe your default configuration is
mips3? The assembler files show the following differences:
$ diff -c MIPS3.s MIPS4.s
*** MIPS3.s Tue Jun 26 13:31:38 2001
--- MIPS4.s Tue Jun 26 13:31:56 2001
***************
*** 26,36 ****
daddu $gp,$1,$25
.set at
dmtc1 $0,$f0
! c.lt.d $f12,$f13
#nop
! bc1f .L3
! l.d $f0,.LC0
! .L3:
ld $28,16($sp)
#nop
.set noreorder
--- 26,35 ----
daddu $gp,$1,$25
.set at
dmtc1 $0,$f0
! c.le.d $f13,$f12
! l.d $f1,.LC0
#nop
! movf.d $f0,$f1,$fcc0
ld $28,16($sp)
#nop
.set noreorder
Greetings,
Volker
---2038259397-2078917053-993555807=:708930
Content-Type: APPLICATION/octet-stream; name="MIPS3.s.gz"
Content-Transfer-Encoding: BASE64
Content-Disposition: attachment; filename="MIPS3.s.gz"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---2038259397-2078917053-993555807=:708930
Content-Type: APPLICATION/octet-stream; name="MIPS4.s.gz"
Content-Transfer-Encoding: BASE64
Content-Disposition: attachment; filename="MIPS4.s.gz"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---2038259397-2078917053-993555807=:708930--
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: optimization/1762, please reopen
@ 2001-06-25 9:56 Reichelt
0 siblings, 0 replies; 3+ messages in thread
From: Reichelt @ 2001-06-25 9:56 UTC (permalink / raw)
To: rth; +Cc: gcc-prs
The following reply was made to PR optimization/1762; it has been noted by GNATS.
From: Reichelt <reichelt@igpm.rwth-aachen.de>
To: rth@gcc.gnu.org, gcc-gnats@gcc.gnu.org, gcc-prs@gcc.gnu.org,
gcc-bugs@gcc.gnu.org
Cc:
Subject: Re: optimization/1762, please reopen
Date: Mon, 25 Jun 2001 18:57:58 +0200
Hello,
I'm afraid the bug reappeared with the 3.0 release.
So, please reopen the PR.
Greetings,
Volker Reichelt
http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=1762&database=gcc
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2001-06-26 4:36 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-06-25 11:16 optimization/1762, please reopen Richard Henderson
-- strict thread matches above, loose matches on Subject: below --
2001-06-26 4:36 Reichelt
2001-06-25 9:56 Reichelt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).