Bugs item #428045, was updated on 2001-05-28 13:14
You can respond by visiting:
http://sourceforge.net/tracker/?func=detail&atid=110493&aid=428045&group_id=10493
Category: build
Group: None
Status: Open
Resolution: None
Priority: 5
Submitted By: Brian D. Carlstrom (bdc)
Assigned to: Brian D. Carlstrom (bdc)
Summary: building on Irix 6.5
Initial Comment:
From: Dave Love <fx@gnu.org>
To: scsh-bugs@zurich.ai.mit.edu
Subject: building on Irix 6.5
Date: 18 Oct 1999 11:47:56 +0100
I had to make this change to build scsh-0.5.2 and
check on Irix
6.5.3m. Presumably this breaks it on other Irix
versions, but I'm
afraid I don't know the issues.
cd /user2/fx/scsh-0.5.2/scsh/machine/
diff -c /user2/fx/scsh-
0.5.2/scsh/machine/sigset.h\~ /user2/fx/scsh-
0.5.2/scsh/\
machine/sigset.h
*** /user2/fx/scsh-0.5.2/scsh/machine/sigset.h~ Sat
Oct 14 04:34:21 1995
--- /user2/fx/scsh-0.5.2/scsh/machine/sigset.h Fri
Oct 15 18:38:42 1999
***************
*** 2,10 ****
** These macros are OS-dependent, and must be
defined per-OS.
*/
! #define make_sigset(maskp, hi, lo) ((maskp)->sigbits
[0]=((hi)<<24)|(lo))
/* Not a procedure: */
#define split_sigset(mask, hip, lop) \
! ((*(hip)=((mask).sigbits[0]>>24)&0xff), \
! (*(lop)=((mask).sigbits[0]&0xffffff)))
--- 2,10 ----
** These macros are OS-dependent, and must be
defined per-OS.
*/
! #define make_sigset(maskp, hi, lo) ((maskp)-
>__sigbits[0]=((hi)<<24)|(lo))
/* Not a procedure: */
#define split_sigset(mask, hip, lop) \
! ((*(hip)=((mask).__sigbits[0]>>24)&0xff), \
! (*(lop)=((mask).__sigbits[0]&0xffffff)))
Diff finished at Mon Oct 18 11:43:39
----------------------------------------------------------------------
>Comment By: Brian D. Carlstrom (bdc)
Date: 2001-05-28 19:55
Message:
Logged In: YES
user_id=27364
Return-Path: <kimball@sgrail.com>
Message-ID: <3884BA20.24A05E5F@sgrail.com>
X-Mailer: Mozilla 4.7C-SGI [en] (X11; U; IRIX64 6.5 IP30)
X-Accept-Language: en
MIME-Version: 1.0
Content-Type: multipart/mixed;
boundary="------------524950EE7ACB72AA943A01AB"
From: Kimball Thurston <kimball@sgrail.com>
To: shivers@ai.mit.edu, bdc@ai.mit.edu
Subject: fyi: scsh 0.5.2
Date: Tue, 18 Jan 2000 11:08:16 -0800
--text follows this line--
This is a multi-part message in MIME format.
--------------524950EE7ACB72AA943A01AB
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Compiling scsh under Irix 6.5 with SGI MIPSpro compiler 7.3
something or
another results in an error - the irix/sigset.h I had to
hack to be the
enclosed file (note the change from sigbits[0] to __sigbits
[0])... I'm
not sure if there was a better way to do this...
later,
--
Kimball
mailto:kimball@sgrail.com
---------------
There is a difference between knowing the path and walking
the path.
- Morpheus
--------------524950EE7ACB72AA943A01AB
Content-Type: application/octet-stream;
name="sigset.h"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
filename="sigset.h"
LyogQ29udmVydCBiZXR3ZWVuIGEgbG8yNC9oaSBpbnRlZ2VyLXBhaXIgYml0
c2V0IGFuZCBh
IHNpZ3NldF90IHZhbHVlLgoqKiBUaGVzZSBtYWNyb3MgYXJlIE9TLWRlcGVu
ZGVudCwgYW5k
IG11c3QgYmUgZGVmaW5lZCBwZXItT1MuCiovCgojZGVmaW5lIG1ha2Vfc2ln
c2V0KG1hc2tw
LCBoaSwgbG8pICgobWFza3ApLT5fX3NpZ2JpdHNbMF09KChoaSk8PDI0KXwo
bG8pKQoKLyog
Tm90IGEgcHJvY2VkdXJlOiAqLwojZGVmaW5lIHNwbGl0X3NpZ3NldChtYXNr
LCBoaXAsIGxv
cCkgXAoJKCgqKGhpcCk9KChtYXNrKS5fX3NpZ2JpdHNbMF0+PjI0KSYweGZm
KSwgXAoJICgq
KGxvcCk9KChtYXNrKS5fX3NpZ2JpdHNbMF0mMHhmZmZmZmYpKSkK
--------------524950EE7ACB72AA943A01AB--
----------------------------------------------------------------------
Comment By: Brian D. Carlstrom (bdc)
Date: 2001-05-28 13:18
Message:
Logged In: YES
user_id=27364
From: Paul Emsley <paule@chem.gla.ac.uk>
To: scsh-news@zurich.ai.mit.edu
Subject: Re: sigset_t / IRIX 6.5
Date: 02 Aug 2000 14:15:25 +0100
>>>>> "R-TH" == Rolf-Thomas Happe
<rthappe@balrog.mathematik.uni-freiburg.de> w\
rites:
R-TH> Paul Emsley:
>> (& (,*plot-viewer* ,plot-file-name))
>>
>> However, this does not fork off into the
background. Instead,
>> it stays in the foreground and main-line computation
is
>> supended until the viewer is terminated.
>>
>> I don't know why this happens on the SGI executable,
but I
>> suspect that it is sigset_t related.
R-TH> Is it just in particular situations when &
doesn't work for
R-TH> you? I cannot reproduce the misbehaviour you
describe. In
R-TH> other words, my SGI executable performs (&
(xcalc)) in the
R-TH> expected way.
Hmmm indeed...
(& (xcalc)) performs in the expected way for me too.
I shall try to pin down exactly where it doesn't
work (might
be a bit tricky).
Cheers,
Paul.
--
Paul.Emsley@chem.gla.ac.uk
http://www.chem.gla.ac.uk/~paule
----------------------------------------------------------------------
Comment By: Brian D. Carlstrom (bdc)
Date: 2001-05-28 13:17
Message:
Logged In: YES
user_id=27364
From: Rolf-Thomas Happe <rthappe@balrog.mathematik.uni-
freiburg.de>
To: scsh-news@zurich.ai.mit.edu
Subject: Re: sigset_t / IRIX 6.5
Date: 02 Aug 2000 13:38:32 +0200
Paul Emsley:
> (& (,*plot-viewer* ,plot-file-name))
>
> However, this does not fork off into the
background. Instead,
> it stays in the foreground and main-line
computation is supended
> until the viewer is terminated.
>
> I don't know why this happens on the SGI
executable, but I
> suspect that it is sigset_t related.
Is it just in particular situations when & doesn't work for
you?
I cannot reproduce the misbehaviour you describe. In other
words,
my SGI executable performs (& (xcalc)) in the expected way.
rthappe
----------------------------------------------------------------------
Comment By: Brian D. Carlstrom (bdc)
Date: 2001-05-28 13:17
Message:
Logged In: YES
user_id=27364
From: Paul Emsley <paule@chem.gla.ac.uk>
To: scsh-bugs@zurich.ai.mit.edu
Subject: Re: sigset_t / IRIX 6.5
Date: 02 Aug 2000 12:14:38 +0100
[snipped SGI sigset_t issues]
Yes indeed. Dave Love and others have mentioned
this.
I however have problems with job control using an
executable
built with such a change.
More specifically, I have a line:
(& (,*plot-viewer* ,plot-file-name))
However, this does not fork off into the
background. Instead,
it stays in the foreground and main-line computation
is supended
until the viewer is terminated.
I don't know why this happens on the SGI
executable, but I
suspect that it is sigset_t related.
Thoughts?
Paul.
--
Paul.Emsley@chem.gla.ac.uk
http://www.chem.gla.ac.uk/~paule
----------------------------------------------------------------------
Comment By: Brian D. Carlstrom (bdc)
Date: 2001-05-28 13:16
Message:
Logged In: YES
user_id=27364
From: Rolf-Thomas Happe <rthappe@balrog.mathematik.uni-
freiburg.de>
To: scsh-bugs@zurich.ai.mit.edu
Subject: sigset_t / IRIX 6.5
Date: Tue, 1 Aug 2000 19:50:22 +0200 (CEST)
I had to change sigbits into __sigbits in order to build
scsh 0.5.2
on an IRIX 6.5 IP32 platform. This regards
scsh-0.5.2/scsh/irix/sigset.h :
5c5
< #define make_sigset(maskp, hi, lo) ((maskp)->sigbits[0]=
((hi)<<24)|(lo))
---
> #define make_sigset(maskp, hi, lo) ((maskp)->__sigbits[0]=
((hi)<<24)|(lo))
9,10c9,10
< ((*(hip)=((mask).sigbits[0]>>24)&0xff), \
< (*(lop)=((mask).sigbits[0]&0xffffff)))
---
> ((*(hip)=((mask).__sigbits[0]>>24)&0xff), \
> (*(lop)=((mask).__sigbits[0]&0xffffff)))
rthappe
----------------------------------------------------------------------
Comment By: Brian D. Carlstrom (bdc)
Date: 2001-05-28 13:14
Message:
Logged In: YES
user_id=27364
From: Dave Love <d.love@dl.ac.uk>
To: "Brian D. Carlstrom" <bdc@ai.mit.edu>
Subject: Re: building on Irix 6.5
Date: 18 Oct 1999 19:54:12 +0100
You wrote:
> Dave Love writes:
>> I had to make this change to build scsh-0.5.2 and check
on Irix
>> 6.5.3m. Presumably this breaks it on other Irix
versions, but I'm
>> afraid I don't know the issues.
> Is there something we can ifdef for? IRIX6 or something?
or better a
> version IRIXVERSION > 60? For example on HP, I can get
gcc to tell me
> what symbols it defines like below.
If you need the OS version, you'd get it via autoconf --
cpp doesn't
help -- but I've no idea what might work in exactly what
version of
Irix, sorry. I think you need an Irix guru. I'm a bit
surprised if
no-one else has run into this.
----------------------------------------------------------------------
You can respond by visiting:
http://sourceforge.net/tracker/?func=detail&atid=110493&aid=428045&group_id=10493
|