public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris
@ 2004-04-20 16:33 papadopo at shfj dot cea dot fr
  2004-04-20 16:44 ` [Bug bootstrap/15029] " pinskia at gcc dot gnu dot org
                   ` (22 more replies)
  0 siblings, 23 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-20 16:33 UTC (permalink / raw)
  To: gcc-bugs

gcc 3.4.0 fails to build on Solaris 8:

$ cd /tmp
$ gtar xzf gcc-3.4.0.tar.gz
$ mkdir GCC
$ cd GCC
$ ../gcc-3.4.0/configure --prefix=/usr/local/gcc-3.4.0 \
	--with-as=/usr/local/binutils/bin/as \
	--with-ld=/usr/local/binutils/bin/ld
[...]
$ gmake bootstrap
[...]
/tmp/gcc-3.4.0/libffi/configure: test: argument expected
gmake[1]: *** [configure-target-libffi] Error 1
gmake[1]: Leaving directory `/tmp/GCC'
gmake: *** [bootstrap] Error 2
$ 

The C compiler has already been built, so this can't be too bad. I'm not sure
how to get more verbose output to debug this though.

-- 
           Summary: gcc 3.4.0 doesn't build on Solaris
           Product: gcc
           Version: 3.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: bootstrap
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: papadopo at shfj dot cea dot fr
                CC: gcc-bugs at gcc dot gnu dot org
  GCC host triplet: sparc-sun-solaris2.8


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
@ 2004-04-20 16:44 ` pinskia at gcc dot gnu dot org
  2004-04-20 16:53 ` papadopo at shfj dot cea dot fr
                   ` (21 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-04-20 16:44 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From pinskia at gcc dot gnu dot org  2004-04-20 16:33 -------
Did you read: <http://gcc.gnu.org/install/specific.html#*-*-solaris2*>?

The Solaris 2 /bin/sh will often fail to configure libstdc++-v3, boehm-gc or libjava. We therefore 
recommend to use the following sequence of commands to bootstrap and install GCC:
        % CONFIG_SHELL=/bin/ksh
        % export CONFIG_SHELL
and then proceed as described in the build instructions, where we strongly recommend using GNU 
make and specifying an absolute path to invoke srcdir/configure.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
  2004-04-20 16:44 ` [Bug bootstrap/15029] " pinskia at gcc dot gnu dot org
@ 2004-04-20 16:53 ` papadopo at shfj dot cea dot fr
  2004-04-20 17:04 ` papadopo at shfj dot cea dot fr
                   ` (20 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-20 16:53 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From papadopo at shfj dot cea dot fr  2004-04-20 16:40 -------
> The C compiler has already been built, so this can't be too bad. I'm not sure
> how to get more verbose output to debug this though.

It seems to be enough to change line 2500 of configure from:

	if test $ac_cv_header_sys_mman_h != yes \
	 || test $ac_cv_func_mmap != yes; then

to:

	if test "x$ac_cv_header_sys_mman_h" != xyes \
	 || test "x$ac_cv_func_mmap" != xyes; then

Unfortunately I'm not sure which source file needs to be changed. This specific
test seems to be generated without configure.in or Makefile.am specifying
anything relevant, so this would rather be an autoconf issue.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
  2004-04-20 16:44 ` [Bug bootstrap/15029] " pinskia at gcc dot gnu dot org
  2004-04-20 16:53 ` papadopo at shfj dot cea dot fr
@ 2004-04-20 17:04 ` papadopo at shfj dot cea dot fr
  2004-04-20 17:11 ` pinskia at gcc dot gnu dot org
                   ` (19 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-20 17:04 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From papadopo at shfj dot cea dot fr  2004-04-20 16:46 -------
Mmmh... The comment in
http://gcc.gnu.org/install/specific.html#*-*-solaris2*
seems to have been added recently. I don't remember having seen it in the past.

Anyway, I have been building gcc on Solaris since gcc 2.7.2.x and I don't
remember having problems with /bin/sh. Maybe this specific issue could be fixed?
That's the only problem I have encoutered so far. The build process has now
entered the java tree and things seem OK so far.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (2 preceding siblings ...)
  2004-04-20 17:04 ` papadopo at shfj dot cea dot fr
@ 2004-04-20 17:11 ` pinskia at gcc dot gnu dot org
  2004-04-20 17:22 ` papadopo at shfj dot cea dot fr
                   ` (18 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-04-20 17:11 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From pinskia at gcc dot gnu dot org  2004-04-20 16:55 -------
This comment has been there for a while now (well most of it has been there almost a year now but the 
rest about not using solaris's sh has been there since at least two years).  Can you try building with how 
the instructions are because they are known to work.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (3 preceding siblings ...)
  2004-04-20 17:11 ` pinskia at gcc dot gnu dot org
@ 2004-04-20 17:22 ` papadopo at shfj dot cea dot fr
  2004-04-20 17:38 ` ebotcazou at gcc dot gnu dot org
                   ` (17 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-20 17:22 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From papadopo at shfj dot cea dot fr  2004-04-20 17:04 -------
I'd rather have the problem fixed in gcc.

I have encountered only this single problem. The build process is now building
the Java libraries, so I'm almost there. It would be a pity not to fix the problem.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (4 preceding siblings ...)
  2004-04-20 17:22 ` papadopo at shfj dot cea dot fr
@ 2004-04-20 17:38 ` ebotcazou at gcc dot gnu dot org
  2004-04-20 17:54 ` papadopo at shfj dot cea dot fr
                   ` (16 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-20 17:38 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-04-20 17:11 -------
The problem is that /bin/sh is buggy on Solaris.  We don't control /bin/sh, ergo
we recommend to use a shell that is known to be less buggy.  If you don't want
to follow the recommendations, fine, but then don't complain.

FYI I built GCC 3.4.0 on Solaris 2.5.1, 2.6, 2.7, 2.8, 2.9 yesterday without a
hitch.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ebotcazou at gcc dot gnu dot
                   |                            |org
             Status|WAITING                     |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (5 preceding siblings ...)
  2004-04-20 17:38 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-20 17:54 ` papadopo at shfj dot cea dot fr
  2004-04-20 19:45 ` pinskia at gcc dot gnu dot org
                   ` (15 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-20 17:54 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From papadopo at shfj dot cea dot fr  2004-04-20 17:20 -------
> The problem is that /bin/sh is buggy on Solaris.

Any large piece of software is buggy - and so are GNU bash or ksh. It's just
that in this case it bites us.

> We don't control /bin/sh, ergo we recommend to use a shell that is known to
> be less buggy.

I can understand that, but if you push that too far, you'll end up recommending
a kernel and a C library that is known to work :-) I'd have no problem with
that, except I have to support Solaris for a few more years. It's just that I
feel it's worth applying simple fixes for the sake of portability.

> If you don't want to follow the recommendations, fine, but then don't
> complain.
> 
> FYI I built GCC 3.4.0 on Solaris 2.5.1, 2.6, 2.7, 2.8, 2.9 yesterday without a
> hitch.

I don't complain and I am able to build gcc 3.4.0. I'm just trying to help
improving portability. Of course you'll find bugs on Solaris. But if they're not
very severe, I feel it's worth working around them.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|INVALID                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (6 preceding siblings ...)
  2004-04-20 17:54 ` papadopo at shfj dot cea dot fr
@ 2004-04-20 19:45 ` pinskia at gcc dot gnu dot org
  2004-04-20 21:27 ` papadopo at shfj dot cea dot fr
                   ` (14 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-04-20 19:45 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From pinskia at gcc dot gnu dot org  2004-04-20 17:22 -------
Just wait until you install, there will be more problems.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (7 preceding siblings ...)
  2004-04-20 19:45 ` pinskia at gcc dot gnu dot org
@ 2004-04-20 21:27 ` papadopo at shfj dot cea dot fr
  2004-04-20 21:42 ` ebotcazou at gcc dot gnu dot org
                   ` (13 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-20 21:27 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From papadopo at shfj dot cea dot fr  2004-04-20 17:26 -------
I've already heard that about Sun's as/ld in the past. It'll never work and so
on. Today gcc supports Sun's as/ld. Of course I guess Sun have worked on fixing
as and ld, but this is different, it shouldn't be too difficult to get gcc to
work with /bin/sh.

Would you at least accept patches?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (8 preceding siblings ...)
  2004-04-20 21:27 ` papadopo at shfj dot cea dot fr
@ 2004-04-20 21:42 ` ebotcazou at gcc dot gnu dot org
  2004-04-26  8:45 ` dhighley at highley-recommended dot com
                   ` (12 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-20 21:42 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-04-20 17:54 -------
> I've already heard that about Sun's as/ld in the past. It'll never work and so
> on. Today gcc supports Sun's as/ld. Of course I guess Sun have worked on
> fixing as and ld, but this is different, it shouldn't be too difficult to get 
> gcc to work with /bin/sh.

There are 2 big problems:
- Sun has pledged not to fix /bin/sh for the sake of compatibility,
- GCC is mainly developed on systems with relatively robust shells, so we would
need to spend a fair amount of time permanently adding workarounds left and right.

> Would you at least accept patches?

Try to post one to gcc-patches@gcc.gnu.org and see what happens.  I suspect it
would not make everybody happy :-)

As one of the SPARC maintainers mainly concerned by Solaris, I do not consider
these problems as ours.  I'd suggest to bug Sun about them instead.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (9 preceding siblings ...)
  2004-04-20 21:42 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-26  8:45 ` dhighley at highley-recommended dot com
  2004-04-26  8:46 ` ebotcazou at gcc dot gnu dot org
                   ` (11 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: dhighley at highley-recommended dot com @ 2004-04-26  8:45 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From dhighley at highley-recommended dot com  2004-04-26 05:08 -------
I ran into the same issue. The problem here is not a bug in Solaris's shell. It
is  a faulty configure script. Anyone who has written shell scripts knows that
you need to protect against a variable not being defined in a test by putting
quotation marks around the variable. So the discussion about the Solaris shell
is pure garbage in this context.

After fixing this issue and completing the build the install will fail. The
Makefile INSTALL macro was not tree position leveled so I
ended up creating these symlinks under gcc-3.4.0-obj:
./gcc-3.4.0
./sparc-sun-solaris2.9/libstdc++-v3/gcc-3.4.0
./sparc-sun-solaris2.9/sparcv9/libstdc++-v3/gcc-3.4.0
./sparc-sun-solaris2.9/sparcv9/libffi/gcc-3.4.0
./sparc-sun-solaris2.9/sparcv9/libjava/gcc-3.4.0
./sparc-sun-solaris2.9/sparcv9/gcc-3.4.0
./sparc-sun-solaris2.9/libffi/gcc-3.4.0
./sparc-sun-solaris2.9/libjava/gcc-3.4.0
./sparc-sun-solaris2.9/gcc-3.4.0

Ended up repeating gmake install to discover each symlink needed. So I have
built tested and installed on a Solaris 9 platform. I did poke around in the
make file templates, but did not discover where to fix this issue. 

Just a closing comment. I have built and used the GCC compiler for years with
little problems until we reached the 3.x versions. I have experienced install
issues with every release in the 3.x series.
 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (10 preceding siblings ...)
  2004-04-26  8:45 ` dhighley at highley-recommended dot com
@ 2004-04-26  8:46 ` ebotcazou at gcc dot gnu dot org
  2004-04-26 10:01 ` papadopo at shfj dot cea dot fr
                   ` (10 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-26  8:46 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-04-26 05:20 -------
> I ran into the same issue. The problem here is not a bug in Solaris's shell.
> It is  a faulty configure script. Anyone who has written shell scripts knows
> that you need to protect against a variable not being defined in a test by
> putting quotation marks around the variable. So the discussion about the
> Solaris shell is pure garbage in this context.

We do *not* support /bin/sh on Solaris.  So arguing about that is a pure waste
of time in this context.

> After fixing this issue and completing the build the install will fail. The
> Makefile INSTALL macro was not tree position leveled so I
> ended up creating these symlinks under gcc-3.4.0-obj:
 
You did not follow the build instructions for Solaris.  Do not bother reporting
problems in that case.

> Just a closing comment. I have built and used the GCC compiler for years with
> little problems until we reached the 3.x versions. I have experienced install
> issues with every release in the 3.x series.

Right.  That's why we took care to document a method that works on any Solaris
version.  You chose not to follow it, fine, but then do not complain.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (11 preceding siblings ...)
  2004-04-26  8:46 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-26 10:01 ` papadopo at shfj dot cea dot fr
  2004-04-26 11:34 ` ebotcazou at gcc dot gnu dot org
                   ` (9 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: papadopo at shfj dot cea dot fr @ 2004-04-26 10:01 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From papadopo at shfj dot cea dot fr  2004-04-26 08:01 -------
Subject: Re:  gcc 3.4.0 doesn't build on Solaris

Hi,

> We do *not* support /bin/sh on Solaris.  So arguing about that is a pure waste
> of time in this context.

Mmmmh... /bin/sh the default shell and this this is getting ridiculous.

There are some bugs in /bin/sh indeed, but they aren't that serious and 
they're easily worked around. I see a lot of fuss over minor issues here.

The attitude "Sun is garbage and we don't support this and that" seems 
to have been very common among gcc developers for years. I can find no 
rational reason for it. I agree the Solaris platform has bugs, but the 
same can be told of all platforms. That's why there are configure 
scripts after all. If you look at any configure script, you'll see it's 
plenty of workarounds to handle bugs and differences between platforms.

I can understand there may not be enough manpower to handle all these 
issues. However the reasons invoked here seem irrational to me (the "Sun 
is garbage" kind). A more constructive answer would be: "We are aware of 
the problems, but have no time to handle them, feel free to send a patch."

Anyway, you're free to do as you wish about that, whatever the reasons.


> Right.  That's why we took care to document a method that works on any Solaris
> version.  You chose not to follow it, fine, but then do not complain.

No one's complaining. People are trying to be constructive. You aren't.

Regards,
--
Dimitri Papadopoulos


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (12 preceding siblings ...)
  2004-04-26 10:01 ` papadopo at shfj dot cea dot fr
@ 2004-04-26 11:34 ` ebotcazou at gcc dot gnu dot org
  2004-04-27 10:25 ` ebotcazou at gcc dot gnu dot org
                   ` (8 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-26 11:34 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-04-26 11:11 -------
> Mmmmh... /bin/sh the default shell and this this is getting ridiculous.

I don't see why.  /bin/sh has bugs that /bin/ksh doesn't have.  GCC is a complex
piece of software with a dozen of configure scripts that very often break with
/bin/sh.  Ergo we recommend to use /bin/ksh.  This is pragmatism.

> There are some bugs in /bin/sh indeed, but they aren't that serious and 
> they're easily worked around. I see a lot of fuss over minor issues here.

We have more interesting things to do than working around bugs in shells.

> The attitude "Sun is garbage and we don't support this and that" seems 
> to have been very common among gcc developers for years. I can find no 
> rational reason for it

Neither me.  SPARC/Solaris is one of the best Unix platforms out there.

> I can understand there may not be enough manpower to handle all these 
> issues. However the reasons invoked here seem irrational to me (the "Sun 
> is garbage" kind). A more constructive answer would be: "We are aware of 
> the problems, but have no time to handle them, feel free to send a patch."

Nothing irrational, only pragmatism.  One year ago the bug database was riddled
with configuration/installation problems on Solaris.  We fixed our part but we
can't fix /bin/sh.  So we decided not to support /bin/sh anymore.

> No one's complaining. People are trying to be constructive. You aren't.

Admittedly.  But there is no point in being constructive here, the decision not
to support /bin/sh on Solaris is final.

Thanks,


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (13 preceding siblings ...)
  2004-04-26 11:34 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-27 10:25 ` ebotcazou at gcc dot gnu dot org
  2004-04-27 13:29 ` lhecking+gccbug at nmrc dot ucc dot ie
                   ` (7 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-27 10:25 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-04-27 09:21 -------
*** Bug 15163 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |lhecking+gccbug at nmrc dot
                   |                            |ucc dot ie


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (14 preceding siblings ...)
  2004-04-27 10:25 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-27 13:29 ` lhecking+gccbug at nmrc dot ucc dot ie
  2004-04-27 17:04 ` ebotcazou at gcc dot gnu dot org
                   ` (6 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: lhecking+gccbug at nmrc dot ucc dot ie @ 2004-04-27 13:29 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From lhecking+gccbug at nmrc dot ucc dot ie  2004-04-27 13:07 -------
The purpose of a configure script is to take care of platform specific details.
I fully concur with comments #11 and #13. Please fix your scripts.

Also, can you please address the iconv issue mentioned in bug 15163.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (15 preceding siblings ...)
  2004-04-27 13:29 ` lhecking+gccbug at nmrc dot ucc dot ie
@ 2004-04-27 17:04 ` ebotcazou at gcc dot gnu dot org
  2004-05-26 14:00 ` bangerth at dealii dot org
                   ` (5 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-27 17:04 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-04-27 16:53 -------
> I fully concur with comments #11 and #13. Please fix your scripts.

Please read the Solaris-specific build instructions.

> Also, can you please address the iconv issue mentioned in bug 15163.

Can you please file another PR for the problem?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (16 preceding siblings ...)
  2004-04-27 17:04 ` ebotcazou at gcc dot gnu dot org
@ 2004-05-26 14:00 ` bangerth at dealii dot org
  2004-05-26 14:00 ` bangerth at dealii dot org
                   ` (4 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: bangerth at dealii dot org @ 2004-05-26 14:00 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From bangerth at dealii dot org  2004-05-25 18:26 -------
...as a duplicate. 

*** This bug has been marked as a duplicate of 15093 ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |DUPLICATE


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (18 preceding siblings ...)
  2004-05-26 14:00 ` bangerth at dealii dot org
@ 2004-05-26 14:00 ` fcusack at fcusack dot com
  2004-05-26 14:04 ` pinskia at gcc dot gnu dot org
                   ` (2 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: fcusack at fcusack dot com @ 2004-05-26 14:00 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From fcusack at fcusack dot com  2004-05-25 18:20 -------
Fixed in bug 15093.  I assume I don't have privilege to mark this as a duplicate;
even if I did I'd rather leave it to you folks.  May I comment on the suboptimal
way this bug was handled?  So what if Solaris /bin/sh isn't supported.  A very
short investigation of the problem shows a clear error in the configure script.
This does not appear to be a bug in /bin/sh, although I don't see how it works
on other platforms.  Note that in bug 15093, the CONFIG_SHELL workaround is
shown as clearly wrong (simply masks the problem).  To the best of my reading.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (17 preceding siblings ...)
  2004-05-26 14:00 ` bangerth at dealii dot org
@ 2004-05-26 14:00 ` bangerth at dealii dot org
  2004-05-26 14:00 ` fcusack at fcusack dot com
                   ` (3 subsequent siblings)
  22 siblings, 0 replies; 24+ messages in thread
From: bangerth at dealii dot org @ 2004-05-26 14:00 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From bangerth at dealii dot org  2004-05-25 18:25 -------
Reopen to mark... 

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|INVALID                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (19 preceding siblings ...)
  2004-05-26 14:00 ` fcusack at fcusack dot com
@ 2004-05-26 14:04 ` pinskia at gcc dot gnu dot org
  2004-05-27 15:17 ` fcusack at fcusack dot com
  2004-05-27 15:45 ` pinskia at gcc dot gnu dot org
  22 siblings, 0 replies; 24+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-05-26 14:04 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From pinskia at gcc dot gnu dot org  2004-05-25 18:27 -------
Note you would run into other problems even after fixing the libffi problem, solaris's sh has problems 
figuring out the right patch for install.sh.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (20 preceding siblings ...)
  2004-05-26 14:04 ` pinskia at gcc dot gnu dot org
@ 2004-05-27 15:17 ` fcusack at fcusack dot com
  2004-05-27 15:45 ` pinskia at gcc dot gnu dot org
  22 siblings, 0 replies; 24+ messages in thread
From: fcusack at fcusack dot com @ 2004-05-27 15:17 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From fcusack at fcusack dot com  2004-05-26 22:38 -------
Well, that's half right.  See bug 15668.  (I've no idea how you could
come up with a diagnosis that /bin/sh is at fault.)

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

* [Bug bootstrap/15029] gcc 3.4.0 doesn't build on Solaris
  2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
                   ` (21 preceding siblings ...)
  2004-05-27 15:17 ` fcusack at fcusack dot com
@ 2004-05-27 15:45 ` pinskia at gcc dot gnu dot org
  22 siblings, 0 replies; 24+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-05-27 15:45 UTC (permalink / raw)
  To: gcc-bugs


------- Additional Comments From pinskia at gcc dot gnu dot org  2004-05-26 23:01 -------
Because it works with anyother real POSIX shell, that is why.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15029


^ permalink raw reply	[flat|nested] 24+ messages in thread

end of thread, other threads:[~2004-05-26 23:01 UTC | newest]

Thread overview: 24+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-04-20 16:33 [Bug bootstrap/15029] New: gcc 3.4.0 doesn't build on Solaris papadopo at shfj dot cea dot fr
2004-04-20 16:44 ` [Bug bootstrap/15029] " pinskia at gcc dot gnu dot org
2004-04-20 16:53 ` papadopo at shfj dot cea dot fr
2004-04-20 17:04 ` papadopo at shfj dot cea dot fr
2004-04-20 17:11 ` pinskia at gcc dot gnu dot org
2004-04-20 17:22 ` papadopo at shfj dot cea dot fr
2004-04-20 17:38 ` ebotcazou at gcc dot gnu dot org
2004-04-20 17:54 ` papadopo at shfj dot cea dot fr
2004-04-20 19:45 ` pinskia at gcc dot gnu dot org
2004-04-20 21:27 ` papadopo at shfj dot cea dot fr
2004-04-20 21:42 ` ebotcazou at gcc dot gnu dot org
2004-04-26  8:45 ` dhighley at highley-recommended dot com
2004-04-26  8:46 ` ebotcazou at gcc dot gnu dot org
2004-04-26 10:01 ` papadopo at shfj dot cea dot fr
2004-04-26 11:34 ` ebotcazou at gcc dot gnu dot org
2004-04-27 10:25 ` ebotcazou at gcc dot gnu dot org
2004-04-27 13:29 ` lhecking+gccbug at nmrc dot ucc dot ie
2004-04-27 17:04 ` ebotcazou at gcc dot gnu dot org
2004-05-26 14:00 ` bangerth at dealii dot org
2004-05-26 14:00 ` bangerth at dealii dot org
2004-05-26 14:00 ` fcusack at fcusack dot com
2004-05-26 14:04 ` pinskia at gcc dot gnu dot org
2004-05-27 15:17 ` fcusack at fcusack dot com
2004-05-27 15:45 ` pinskia at gcc dot gnu dot org

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).