gcc-2.95.2

gcc-2.95.2

[Agung Yogaswara]

I have gcc-2.95.2 installed with glibc-2.2 updated for Red Hat 7.0 plus
the patch. But I got this error when compiling a program:

In file included from
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream.h:31,
                 from
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream:6,
                 from ../include/JTC/Types.h:58,
                 from Cond.cpp:11:
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h: In
method `struct streampos streambuf::pubseekoff(long long int,
ios::seek_dir, int = 3)':
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h:362: conversion
from `__off64_t' to non-scalar type `streampos' requested
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h: In
method `struct streampos streambuf::pubseekpos(_G_fpos64_t, int = 3)':
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h:364: `struct
streampos' used where a `long long int' was expected

In file included from Cond.cpp:28:
/usr/include/sys/time.h: At top level:
/usr/include/sys/time.h:69: two or more data types in declaration of
`__tz'


Any ideas?

Thanks

Re: gcc-2.95.2

[Alexandre Oliva]

On Feb 22, 2001, Agung Yogaswara <azung@xerxes.intern.trabas.com> wrote:

> I have gcc-2.95.2 installed with glibc-2.2 updated for Red Hat 7.0 plus
> the patch.

Which patch?  The one in the platform-specific installation
instructions at gcc.gnu.org?

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist    *Please* write to mailing lists, not to me

Re: gcc-2.95.2

[Jerry Miller]

The only version of <sys/time.h> where I could find '__tz'
was on my Web host.  (It's also the only Linux system to
which I have access.)  The lines where I found it are:

extern int __gettimeofday __P ((struct timeval * __tp,
   struct timezone * __tz));
extern int gettimeofday __P ((struct timeval * __tp,
   struct timezone * __tz));
extern int __settimeofday __P ((__const struct timeval *__tv,
   __const struct timezone *__tz));
extern int settimeofday __P ((__const struct timeval *__tv,
   __const struct timezone *__tz));

If yours looks anything like this, the only explanation I can
see is that '__const' is for some reason defined as something
other than 'const'.  (A declaration of "int float x;" gives the
identical error message.)

Just on a hunch, try checking Cond.cpp and ../include/JTC/Types.h
for any #undef directives.  Sorry I don't have any more enlightening
ideas.

----- Original Message -----
From: Agung Yogaswara <azung@xerxes.intern.trabas.com>
To: <gcc-help@gcc.gnu.org>
Sent: Thursday, February 22, 2001 1:31 AM
Subject: gcc-2.95.2


> I have gcc-2.95.2 installed with glibc-2.2 updated for Red Hat 7.0 plus
> the patch. But I got this error when compiling a program:
>
> In file included from
>
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream
.h:31,
>                  from
>
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream
:6,
>                  from ../include/JTC/Types.h:58,
>                  from Cond.cpp:11:
>
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambu
f.h: In
> method `struct streampos streambuf::pubseekoff(long long int,
> ios::seek_dir, int = 3)':
>
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambu
f.h:362: conversion
> from `__off64_t' to non-scalar type `streampos' requested
>
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambu
f.h: In
> method `struct streampos streambuf::pubseekpos(_G_fpos64_t, int = 3)':
>
/usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambu
f.h:364: `struct
> streampos' used where a `long long int' was expected
>
> In file included from Cond.cpp:28:
> /usr/include/sys/time.h: At top level:
> /usr/include/sys/time.h:69: two or more data types in declaration of
> `__tz'
>
>
> Any ideas?
>
> Thanks
>
>

Re: gcc-2.95.2

[sherry]

I have seen something like .... in my case, I was missing the _D_GNU_SOURCE .
For more info check /usr/include/features.h, to get the right flavor..
hope it helps

Agung Yogaswara wrote:

> I have gcc-2.95.2 installed with glibc-2.2 updated for Red Hat 7.0 plus
> the patch. But I got this error when compiling a program:
>
> In file included from
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream.h:31,
>                  from
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream:6,
>                  from ../include/JTC/Types.h:58,
>                  from Cond.cpp:11:
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h: In
> method `struct streampos streambuf::pubseekoff(long long int,
> ios::seek_dir, int = 3)':
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h:362: conversion
> from `__off64_t' to non-scalar type `streampos' requested
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h: In
> method `struct streampos streambuf::pubseekpos(_G_fpos64_t, int = 3)':
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h:364: `struct
> streampos' used where a `long long int' was expected
>
> In file included from Cond.cpp:28:
> /usr/include/sys/time.h: At top level:
> /usr/include/sys/time.h:69: two or more data types in declaration of
> `__tz'
>
> Any ideas?
>
> Thanks

Re: gcc-2.95.2

[sherry]

I have seen something like .... in my case, I was missing the _D_GNU_SOURCE .
For more info check /usr/include/features.h, to get the right flavor..
hope it helps


Agung Yogaswara wrote:

> I have gcc-2.95.2 installed with glibc-2.2 updated for Red Hat 7.0 plus
> the patch. But I got this error when compiling a program:
>
> In file included from
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream.h:31,
>                  from
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/iostream:6,
>                  from ../include/JTC/Types.h:58,
>                  from Cond.cpp:11:
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h: In
> method `struct streampos streambuf::pubseekoff(long long int,
> ios::seek_dir, int = 3)':
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h:362: conversion
> from `__off64_t' to non-scalar type `streampos' requested
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h: In
> method `struct streampos streambuf::pubseekpos(_G_fpos64_t, int = 3)':
> /usr/lib/gcc-lib/i586-pc-linux-gnu/2.95.2/../../../../include/g++-3/streambuf.h:364: `struct
> streampos' used where a `long long int' was expected
>
> In file included from Cond.cpp:28:
> /usr/include/sys/time.h: At top level:
> /usr/include/sys/time.h:69: two or more data types in declaration of
> `__tz'
>
> Any ideas?
>
> Thanks

Re: gcc-2.95.2

[Agung Yogaswara]

On 22 Feb 2001, Alexandre Oliva wrote:

> 
> Which patch?  The one in the platform-specific installation
> instructions at gcc.gnu.org?

the patch applied: glibc-2.2.patch as attachment

Re: gcc-2.95.2

[Agung Yogaswara]

On Thu, 22 Feb 2001, Jerry Miller wrote:

> The only version of <sys/time.h> where I could find '__tz'
> was on my Web host.  (It's also the only Linux system to
> which I have access.)  The lines where I found it are:
> 
> extern int __gettimeofday __P ((struct timeval * __tp,
>    struct timezone * __tz));
> extern int gettimeofday __P ((struct timeval * __tp,
>    struct timezone * __tz));
> extern int __settimeofday __P ((__const struct timeval *__tv,
>    __const struct timezone *__tz));
> extern int settimeofday __P ((__const struct timeval *__tv,
>    __const struct timezone *__tz));
> 
> If yours looks anything like this, the only explanation I can
> see is that '__const' is for some reason defined as something
> other than 'const'.  (A declaration of "int float x;" gives the
> identical error message.)
> 

thanks.
 here is my version of '__tz" in /usr/include/sys/time.h (the
only one):

struct timezone
  {
    int tz_minuteswest;     /* Minutes west of GMT.  */
    int tz_dsttime;     /* Nonzero if DST is ever in effect.  */
  };

#if defined __USE_GNU || defined __USE_BSD
typedef struct timezone *__timezone_ptr_t;
#else
typedef void *__timezone_ptr_t;
#endif

extern int gettimeofday (struct timeval *__restrict __tv,
             __timezone_ptr_t __restrict __tz) __THROW;

extern int settimeofday (__const struct timeval *__tv,
             __const struct timezone *__tz) __THROW;


thats pretty much different, isn't it?
am i miss something?

btw, i'm using kernel 2.3.4. 

Re: gcc-2.95.2

[Jerry Miller]

In one sense, it is, but both versions contain the ostensibly
offending declaration:

"__const struct timezone *__tz"

----- Original Message ----- 
From: Agung Yogaswara <azung@xerxes.intern.trabas.com>
To: Jerry Miller <gmiller@cs.sunysb.edu>
Cc: <gcc-help@gcc.gnu.org>
Sent: Thursday, February 22, 2001 7:31 AM
Subject: Re: gcc-2.95.2


> On Thu, 22 Feb 2001, Jerry Miller wrote:
> 
> > The only version of <sys/time.h> where I could find '__tz'
> > was on my Web host.  (It's also the only Linux system to
> > which I have access.)  The lines where I found it are:
> > 
> > extern int __gettimeofday __P ((struct timeval * __tp,
> >    struct timezone * __tz));
> > extern int gettimeofday __P ((struct timeval * __tp,
> >    struct timezone * __tz));
> > extern int __settimeofday __P ((__const struct timeval *__tv,
> >    __const struct timezone *__tz));
> > extern int settimeofday __P ((__const struct timeval *__tv,
> >    __const struct timezone *__tz));
> > 
> > If yours looks anything like this, the only explanation I can
> > see is that '__const' is for some reason defined as something
> > other than 'const'.  (A declaration of "int float x;" gives the
> > identical error message.)
> > 
> 
> thanks.
>  here is my version of '__tz" in /usr/include/sys/time.h (the
> only one):
> 
> struct timezone
>   {
>     int tz_minuteswest;     /* Minutes west of GMT.  */
>     int tz_dsttime;     /* Nonzero if DST is ever in effect.  */
>   };
> 
> #if defined __USE_GNU || defined __USE_BSD
> typedef struct timezone *__timezone_ptr_t;
> #else
> typedef void *__timezone_ptr_t;
> #endif
> 
> extern int gettimeofday (struct timeval *__restrict __tv,
>              __timezone_ptr_t __restrict __tz) __THROW;
> 
> extern int settimeofday (__const struct timeval *__tv,
>              __const struct timezone *__tz) __THROW;
> 
> 
> thats pretty much different, isn't it?
> am i miss something?
> 
> btw, i'm using kernel 2.3.4. 
> 
> 

Re: gcc-2.95.2

[Agung Yogaswara]

On Thu, 22 Feb 2001, Agung Yogaswara wrote:

> On Thu, 22 Feb 2001, Jerry Miller wrote:
> 
> > The only version of <sys/time.h> where I could find '__tz'
> > was on my Web host.  (It's also the only Linux system to
> > which I have access.)  The lines where I found it are:
> > 
> > extern int __gettimeofday __P ((struct timeval * __tp,
> >    struct timezone * __tz));
> > extern int gettimeofday __P ((struct timeval * __tp,
> >    struct timezone * __tz));
> > extern int __settimeofday __P ((__const struct timeval *__tv,
> >    __const struct timezone *__tz));
> > extern int settimeofday __P ((__const struct timeval *__tv,
> >    __const struct timezone *__tz));
> > 
> > If yours looks anything like this, the only explanation I can
> > see is that '__const' is for some reason defined as something
> > other than 'const'.  (A declaration of "int float x;" gives the
> > identical error message.)
> > 
> 
 thanks.
  here is my version of '__tz" in /usr/include/sys/time.h (the
 only one):
 
 struct timezone
   {
     int tz_minuteswest;     /* Minutes west of GMT.  */
     int tz_dsttime;     /* Nonzero if DST is ever in effect.  */
   };
 
 #if defined __USE_GNU || defined __USE_BSD
 typedef struct timezone *__timezone_ptr_t;
 #else
 typedef void *__timezone_ptr_t;
 #endif
 
 extern int gettimeofday (struct timeval *__restrict __tv,
              __timezone_ptr_t __restrict __tz) __THROW;
 
 extern int settimeofday (__const struct timeval *__tv,
              __const struct timezone *__tz) __THROW;
 
 
 thats pretty much different, isn't it?
 am i miss something?

 btw, i'm using kernel 2.3.4. 
 
 

Re: gcc-2.95.2

[Agung Yogaswara]

On Thu, 22 Feb 2001, Agung Yogaswara wrote:

> On 22 Feb 2001, Alexandre Oliva wrote:
> 
> > 
> > Which patch?  The one in the platform-specific installation
> > instructions at gcc.gnu.org?
> 

the patch applied: glibc-2.2.patch as attachment
 

gcc-2.95.2

[sachin lamsoge]

hi 
i am getting the following error while installing
gcc-2.95.2 on redhat 6.2 kernel 2.2.14-5.0
kindly help

../../../../../libstdc++/src/string-inst.cc:64:  
instantiated from here
../../../../../libstdc++/bits/basic_string.h:675:
warning: can't inline call to `class
std::basic_string<char,st
d::char_traits<char>,std::allocator<char> > &
std::basic_string<char,std::char_traits<char>,std::allocator<char>
 >::replace<std::__normal_iterator<char
*,std::basic_string<char,std::char_traits<char>,std::allocator<char>
> >
 >(std::__normal_iterator<char
*,std::basic_string<char,std::char_traits<char>,std::allocator<char>
> >, std::__
normal_iterator<char
*,std::basic_string<char,std::char_traits<char>,std::allocator<char>
> >, std::__normal_ite
sic_string<char,std::char_traits<char>,std::allocator<char>
> &, unsigned int, unsigned int)':
../../../../../libstdc++/src/string-inst.cc:64:  
instantiated from here
../../../../../libstdc++/bits/basic_string.h:541:
warning: can't inline call to `class
std::basic_string<char,std::char_traits<char>,std::allocator<char>
> &
std::basic_string<char,std::char_traits<char>,std::allocator<char>
>::assign<std::__normal_iterator<char
*,std::basic_string<char,std::char_traits<char>,std::allocator<char>
> >
>(std::__normal_iterator<char
*,std::basic_string<char,std::char_traits<char>,std::allocator<char>
> >, std::__normal_iterator<char
*,std::basic_string<char,std::char_traits<char>,std::allocator<char>
> >)'
../../../../../libstdc++/bits/basic_string.h:523:
warning: called from here
make[4]: *** [stringMAIN.lo] Error 1
make[4]: Leaving directory
`/root/gcc-2.95.2/build/object/i686-pc-linux-gnu/libstdc++/src'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory
`/root/gcc-2.95.2/build/object/i686-pc-linux-gnu/libstdc++'
make[2]: *** [all-recursive-am] Error 2
make[2]: Leaving directory
`/root/gcc-2.95.2/build/object/i686-pc-linux-gnu/libstdc++'
make[1]: *** [all-target-libstdc++] Error 2
make[1]: Leaving directory
`/root/gcc-2.95.2/build/object'
make: *** [bootstrap] Error 2
[root@gateway object]#

__________________________________________________
Do You Yahoo!?
Make a great connection at Yahoo! Personals.
http://personals.yahoo.com

RE: gcc-2.95.2

[David Korn]

>-----Original Message-----
>From: Jumblat, Ghassan [ mailto:Ghassan.Jumblat@schwabcanada.com ]
>Sent: 16 March 2001 15:27


>Tough at the configuration level when running ./configure I 
>keep getting
>error messages such as:
>Configuring for a sparc-sun-solaris2.8 host.
>Created "Makefile" in /usr/objdir using "mh-frag"
>/tmp/gcc-2.95.2/configure: /tmp/gcc-2.95.2/gcc: cannot execute
>*** The command '/tmp/gcc-2.95.2/gcc -o conftest -O2   
>conftest.c' failed.
>*** You must set the environment variable CC to a working compiler.
>
>I think my problem is I do not have a native compiler on my system and
>therefore

  You are absolutely 100% right; you do need some kind of compiler to get
started with.  If you don't have the Sun development tools to install,
your best bet is to download a precompiled version of Gcc from 

http://www.sunfreeware.com/

        DaveK
-- 
 All your base are belong to us!


**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.

This footnote also confirms that this email message has been swept by
MIMEsweeper for the presence of computer viruses.

www.mimesweeper.com
**********************************************************************

gcc-2.95.2

[Jumblat, Ghassan]

Hi..
I downloaded the file gcc-2.95.2.tar.gz
gunzip the file and untar.

I am trying to install gcc compiler on my system.
I am following the instructions from the documents on the following site:
http://gcc.gnu.org/install/index.html

Tough at the configuration level when running ./configure I keep getting
error messages such as:
Configuring for a sparc-sun-solaris2.8 host.
Created "Makefile" in /usr/objdir using "mh-frag"
/tmp/gcc-2.95.2/configure: /tmp/gcc-2.95.2/gcc: cannot execute
*** The command '/tmp/gcc-2.95.2/gcc -o conftest -O2   conftest.c' failed.
*** You must set the environment variable CC to a working compiler.

I think my problem is I do not have a native compiler on my system and
therefore
I get the error messages.

How do I install gcc on my system from scratch  ??

Thank you

regards,
ghassan
System Engineer
w: 416-682-6428
c: 416-985-8300

Warning:  All e-mail sent to or from this address will be received by the
Charles Schwab corporate e-mail system and is subject to archival and review
by someone other than the recipient.

Re: gcc-2.95.2

[Inferno]

I don't quite understand what you are saying.  I have a working gcc-2.95.1 compiler
installed.  I am trying to compile and install gcc-2.95.2 overtop of it.  I went
through the following steps to try to compile:
cd /root/gcc-2.95.2/
mkdir obj
cd obj
CONFIG_SHELL=bash bash ../configure --prefix=/Cygnus/Cygwin-b20
--exec-prefix=/Cygnus/Cygwin-b20/H-i586-cygwin32 -v i586-cygwin32
make

It then goes through about an hour worth of compiling and hits where i pasted on.
I have also just tried configure with no options with also no luck.  Ideas?

Thanks,
Inferno

Tim Prince wrote:

> > trying to install gcc-2.95.2 on my cygwin b20.1 system
>
> You should have said so in the subject; you'll be lucky if you get relevant
> responses.
> >_muldi3
> >./libgcc2.c:41: stdlib.h: No such file or directory
> >./libgcc2.c:42: unistd.h: No such file or directory
>
> You don't have it configured with --prefix and --target pointing to a directory
> which contains the include files.  You could patch over this by putting them in
> /usr/include, but that should be a last resort, as you may have other problems,
> particularly if you're on W9X.  Supposing that your current installation is
> like --prefix=//c/cygnus/cygwin-b20/H-i586-cygwin32 --host=i586-cygwin32 and
> you haven't installed the recommended snapshots, your configure would include
> the above and --target=i586-cygwin32, unless you want to copy the necessary
> directories from where there are now to where you say you're intending to
> install.
>
> I have a (mostly done) treatise on this at
> http://members.aol.com/n8tm/cygwingccg77.htm .
> Tim Prince
> tprince@computer.org

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Tim Prince]

> trying to install gcc-2.95.2 on my cygwin b20.1 system

You should have said so in the subject; you'll be lucky if you get relevant
responses.
>_muldi3
>./libgcc2.c:41: stdlib.h: No such file or directory
>./libgcc2.c:42: unistd.h: No such file or directory

You don't have it configured with --prefix and --target pointing to a directory
which contains the include files.  You could patch over this by putting them in
/usr/include, but that should be a last resort, as you may have other problems,
particularly if you're on W9X.  Supposing that your current installation is
like --prefix=//c/cygnus/cygwin-b20/H-i586-cygwin32 --host=i586-cygwin32 and
you haven't installed the recommended snapshots, your configure would include
the above and --target=i586-cygwin32, unless you want to copy the necessary
directories from where there are now to where you say you're intending to
install.

I have a (mostly done) treatise on this at
http://members.aol.com/n8tm/cygwingccg77.htm .  
Tim Prince
tprince@computer.org

Re: gcc-2.95.2

[Mumit Khan]

Just pick up my 2.95.2 binary distribution from
http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ .

Regards,
Mumit

gcc-2.95.2

[Inferno]

I was just trying to install gcc-2.95.2 on my cygwin b20.1 system and
got the following errors on compile:
  /root/gcc-2.95.2/gcc/xgcc -B/root/gcc-2.95.2/gcc/
-B/H-i586-cygwin32/i686-pc-cygwin32/bin/ -I/H-i5
86-cygwin32/i686-pc-cygwin32/include -O2 -I./../winsup/include
-DIN_GCC     -g -O2 -I./include   -g
1  -DIN_LIBGCC2 -D__GCC_FLOAT_NOT_NEEDED   -I. -I. -I./config
-I./../include -c -DL${name} \
       ./libgcc2.c -o ${name}.o; \
  if [ $? -eq 0 ] ; then true; else exit 1; fi; \
  ar rc tmplibgcc2.a ${name}.o; \
  rm -f ${name}.o; \
done
_muldi3
./libgcc2.c:41: stdlib.h: No such file or directory
./libgcc2.c:42: unistd.h: No such file or directory
make[1]: *** [libgcc2.a] Error 1
make: *** [all-gcc] Error 2
#


I was using the configure --prefix=/H-i586-cygwin32    program
configuration.  I let it do defaults for the rest.  What went wrong?

Thanks,
Inferno

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Inferno]

how did you manage to compile the source?  Just curious for reference
later on.

Thanks,
Inferno

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Tim Prince]

>when i install prebuilt binaries, is it less
>efficient than binaries built on my own system

I suppose this question could be interpreted various ways, and I'm not fully
competent to answer, as I haven't tried both ways.  Maybe I'll get Mumit's
version and compare output files.  I expect the results would be the same
except for possible alignment differences; I've put some emphasis on getting
the .p2align scheme working when building natively, even though it is not fully
effective because of cygwin not providing 128-bit stack alignment.  You would
see this by comparing .s files, for example; the alignment directives may
differ.  Also, you probably wouldn't be using exactly the same binutils, if
Mumit is recommending you install his binaries for both binutils and the
compilers.
Tim Prince
tprince@computer.org

Re: gcc-2.95.2

[Inferno]

Out of curiousity, when i install prebuilt binaries, is it less
efficient than binaries built on my own system or is it the exact same?

Thanks,
Inferno

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Tim Prince]

>when i install prebuilt binaries, is it less
>efficient than binaries built on my own system

I suppose this question could be interpreted various ways, and I'm not fully
competent to answer, as I haven't tried both ways.  Maybe I'll get Mumit's
version and compare output files.  I expect the results would be the same
except for possible alignment differences; I've put some emphasis on getting
the .p2align scheme working when building natively, even though it is not fully
effective because of cygwin not providing 128-bit stack alignment.  You would
see this by comparing .s files, for example; the alignment directives may
differ.  Also, you probably wouldn't be using exactly the same binutils, if
Mumit is recommending you install his binaries for both binutils and the
compilers.
Tim Prince
tprince@computer.org

Re: gcc-2.95.2

[Inferno]

Out of curiousity, when i install prebuilt binaries, is it less
efficient than binaries built on my own system or is it the exact same?

Thanks,
Inferno

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Inferno]

how did you manage to compile the source?  Just curious for reference
later on.

Thanks,
Inferno

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Mumit Khan]

Just pick up my 2.95.2 binary distribution from
http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ .

Regards,
Mumit

Re: gcc-2.95.2

[Inferno]

I don't quite understand what you are saying.  I have a working gcc-2.95.1 compiler
installed.  I am trying to compile and install gcc-2.95.2 overtop of it.  I went
through the following steps to try to compile:
cd /root/gcc-2.95.2/
mkdir obj
cd obj
CONFIG_SHELL=bash bash ../configure --prefix=/Cygnus/Cygwin-b20
--exec-prefix=/Cygnus/Cygwin-b20/H-i586-cygwin32 -v i586-cygwin32
make

It then goes through about an hour worth of compiling and hits where i pasted on.
I have also just tried configure with no options with also no luck.  Ideas?

Thanks,
Inferno

Tim Prince wrote:

> > trying to install gcc-2.95.2 on my cygwin b20.1 system
>
> You should have said so in the subject; you'll be lucky if you get relevant
> responses.
> >_muldi3
> >./libgcc2.c:41: stdlib.h: No such file or directory
> >./libgcc2.c:42: unistd.h: No such file or directory
>
> You don't have it configured with --prefix and --target pointing to a directory
> which contains the include files.  You could patch over this by putting them in
> /usr/include, but that should be a last resort, as you may have other problems,
> particularly if you're on W9X.  Supposing that your current installation is
> like --prefix=//c/cygnus/cygwin-b20/H-i586-cygwin32 --host=i586-cygwin32 and
> you haven't installed the recommended snapshots, your configure would include
> the above and --target=i586-cygwin32, unless you want to copy the necessary
> directories from where there are now to where you say you're intending to
> install.
>
> I have a (mostly done) treatise on this at
> http://members.aol.com/n8tm/cygwingccg77.htm .
> Tim Prince
> tprince@computer.org

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno

Re: gcc-2.95.2

[Tim Prince]

> trying to install gcc-2.95.2 on my cygwin b20.1 system

You should have said so in the subject; you'll be lucky if you get relevant
responses.
>_muldi3
>./libgcc2.c:41: stdlib.h: No such file or directory
>./libgcc2.c:42: unistd.h: No such file or directory

You don't have it configured with --prefix and --target pointing to a directory
which contains the include files.  You could patch over this by putting them in
/usr/include, but that should be a last resort, as you may have other problems,
particularly if you're on W9X.  Supposing that your current installation is
like --prefix=//c/cygnus/cygwin-b20/H-i586-cygwin32 --host=i586-cygwin32 and
you haven't installed the recommended snapshots, your configure would include
the above and --target=i586-cygwin32, unless you want to copy the necessary
directories from where there are now to where you say you're intending to
install.

I have a (mostly done) treatise on this at
http://members.aol.com/n8tm/cygwingccg77.htm .  
Tim Prince
tprince@computer.org

gcc-2.95.2

[Inferno]

I was just trying to install gcc-2.95.2 on my cygwin b20.1 system and
got the following errors on compile:
  /root/gcc-2.95.2/gcc/xgcc -B/root/gcc-2.95.2/gcc/
-B/H-i586-cygwin32/i686-pc-cygwin32/bin/ -I/H-i5
86-cygwin32/i686-pc-cygwin32/include -O2 -I./../winsup/include
-DIN_GCC     -g -O2 -I./include   -g
1  -DIN_LIBGCC2 -D__GCC_FLOAT_NOT_NEEDED   -I. -I. -I./config
-I./../include -c -DL${name} \
       ./libgcc2.c -o ${name}.o; \
  if [ $? -eq 0 ] ; then true; else exit 1; fi; \
  ar rc tmplibgcc2.a ${name}.o; \
  rm -f ${name}.o; \
done
_muldi3
./libgcc2.c:41: stdlib.h: No such file or directory
./libgcc2.c:42: unistd.h: No such file or directory
make[1]: *** [libgcc2.a] Error 1
make: *** [all-gcc] Error 2
#


I was using the configure --prefix=/H-i586-cygwin32    program
configuration.  I let it do defaults for the rest.  What went wrong?

Thanks,
Inferno

--
"Life is the whim of several billion cells to be you for a while."
                                            --Inferno