public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
* bootstrapping with `cc -n32' on IRIX 6.3 fails compare
@ 1999-03-21 10:01 Alexandre Oliva
  1999-03-31 23:46 ` Alexandre Oliva
  0 siblings, 1 reply; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-21 10:01 UTC (permalink / raw)
  To: egcs

Can anyone else confirm that it's not just my cc for IRIX 6.3 that's
broken?  I've never been able to bootstrap egcs with `cc -n32' (the
-n32 is needed for libiberty); it always fails `make compare' for all
object files (I haven't really checked if some are missing, but the
list of failures is huge).  Then, if a proceed with `bootstrap2', it
moves the stage3 compiler into stage2, builds a new stage3 and then
compare succeeds.  Does this sound familiar to anyone?  Is this
expected?  Do we care about fixing it, or should we just document it
and put the blame on SGI? :-)

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists

^ permalink raw reply	[flat|nested] 20+ messages in thread

* bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-21 10:01 bootstrapping with `cc -n32' on IRIX 6.3 fails compare Alexandre Oliva
@ 1999-03-31 23:46 ` Alexandre Oliva
  0 siblings, 0 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-31 23:46 UTC (permalink / raw)
  To: egcs

Can anyone else confirm that it's not just my cc for IRIX 6.3 that's
broken?  I've never been able to bootstrap egcs with `cc -n32' (the
-n32 is needed for libiberty); it always fails `make compare' for all
object files (I haven't really checked if some are missing, but the
list of failures is huge).  Then, if a proceed with `bootstrap2', it
moves the stage3 compiler into stage2, builds a new stage3 and then
compare succeeds.  Does this sound familiar to anyone?  Is this
expected?  Do we care about fixing it, or should we just document it
and put the blame on SGI? :-)

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists


^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  0:04 ` Alexandre Oliva
       [not found]   ` <tprvhfruwva.fsf@cs.ucsb.edu>
@ 1999-03-31 23:46   ` Alexandre Oliva
  1 sibling, 0 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-31 23:46 UTC (permalink / raw)
  To: Kaveh R. Ghazi; +Cc: egcs

On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:

>> From: Alexandre Oliva <oliva@dcc.unicamp.br> 

>> I've never been able to bootstrap egcs with `cc -n32' (the -n32 is
>> needed for libiberty); it always fails `make compare'

> 	I've had success with cc -n32 when bootstrapping on Irix 6.2.

> I don't know if the version of the compiler is tied to the OS or not. 
> What version of the compiler are you using?

Dunno, how can I find it out?  I tried various command-line flags and
nothing helped.  I even looked at the manual! :-D

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists


^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  6:16 N8TM
@ 1999-03-31 23:46 ` N8TM
  0 siblings, 0 replies; 20+ messages in thread
From: N8TM @ 1999-03-31 23:46 UTC (permalink / raw)
  To: ghazi, jerry, oliva; +Cc: egcs

In a message dated 3/24/99 5:13:43 AM Pacific Standard Time,
ghazi@caip.rutgers.edu writes:

> It looks like we have 6.2 and 7.1, but the dates on the 7.1 entry are
>  _older_ than the 6.2 one. (?)  Wierd.
7.1 is the worst of the versions 6.2, 7.0, and 7.1.  My installation
supposedly upgraded to 7.2.1 a week ago but I don't see any fixes for
documented bugs of 7.2.

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  5:10 Kaveh R. Ghazi
@ 1999-03-31 23:46 ` Kaveh R. Ghazi
  0 siblings, 0 replies; 20+ messages in thread
From: Kaveh R. Ghazi @ 1999-03-31 23:46 UTC (permalink / raw)
  To: jerry, oliva; +Cc: egcs

 > From: Alexandre Oliva <oliva@dcc.unicamp.br>
 >  
 > On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:
 >  
 > > On Mar 24, 1999, you wrote:
 > >> On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:
 >  
 > > [snip]
 >  
 > >> > I don't know if the version of the compiler is tied to the OS or not. 
 > >> > What version of the compiler are you using?
 > >> 
 > >> Dunno, how can I find it out?  I tried various command-line flags and
 > >> nothing helped.  I even looked at the manual! :-D
 >  
 > > Use the "versions" command.
 >  
 > Thanks.  I get:
 >  
 > oliva@duerer% versions | grep c_dev
 > I  c_dev                11/12/97  C, 7.0
 >  
 > Pretty old, huh? :-(


	Hmm, The full listing I get is:

I  c_dev                07/29/1998  C, 6.2
I  c_dev.src            07/29/1998  C Optional Sources
I  c_dev.src.acpp       07/29/1998  C acpp Source
I  c_dev.src.customalloc  07/29/1998  C customalloc Source
I  c_dev.sw             07/29/1998  C Software
I  c_dev.sw.speclib     07/29/1998  C NonShared Libraries for Benchmarks
I  c_dev.sw32           07/29/1998  C Software
I  c_dev.sw32.speclib   07/29/1998  C NonShared Libraries for Benchmarks
I  c_dev                06/02/1997  C, 7.1 on irix 6.2
I  c_dev.books          06/02/1997  C IRIS InSight Books
I  c_dev.books.CLanguageRef  06/02/1997  C Language Reference Manual
I  c_dev.hdr            06/02/1997  C Headers
I  c_dev.hdr.lib        06/02/1997  C Library Headers
I  c_dev.man            06/02/1997  C Manual Pages
I  c_dev.man.c          06/02/1997  C Compiler Man Pages
I  c_dev.man.copt       06/02/1997  C Copt Man Pages
I  c_dev.man.relnotes   06/02/1997  C Release Notes
I  c_dev.man.util       06/02/1997  C Utility Man Pages
I  c_dev.sw             06/02/1997  C Software
I  c_dev.sw.c           06/02/1997  C Compiler
I  c_dev.sw.copt        06/02/1997  C Scalar Optimizer
I  c_dev.sw.lib         06/02/1997  C Libraries
I  c_dev.sw.util        06/02/1997  C Utilities
I  c_dev.sw.v6_c        06/02/1997  C Compiler Version 6
I  c_dev.sw.v6_copt     06/02/1997  C Scalar Optimizer Version 6
I  c_dev.sw64           06/02/1997  C Software
I  c_dev.sw64.lib       06/02/1997  C Libraries 64bit

It looks like we have 6.2 and 7.1, but the dates on the 7.1 entry are
_older_ than the 6.2 one. (?)  Wierd.

--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24 12:41   ` Jim Wilson
@ 1999-03-31 23:46     ` Jim Wilson
  0 siblings, 0 replies; 20+ messages in thread
From: Jim Wilson @ 1999-03-31 23:46 UTC (permalink / raw)
  To: oliva; +Cc: egcs

In article <oriubr70o5.fsf.cygnus.egcs@dcc.unicamp.br> you write:
>oliva@duerer% versions | grep c_dev
>I  c_dev                11/12/97  C, 7.0

All versions of SGI C before v7.1beta have bugs that cause gcc to be
miscompiled when using the -n32 option.

With SGI C v7.0, there was only one bug that I found.  It miscompiles a little
bit of code in one of the optimization passes, with the result that the
optimization does not happen.  You still get correct code though.  Hence
you can successfully bootstrap if you go a fourth stage.
stage1: compiled with one optimization disabled
stage2: compiled OK, but only partially optimized
stage3: compiled OK, and completely optimized
stage4: compiled OK, and completely optimized

I don't recall what the v7.0 bug was.  I reported it to SGI about 3 years ago,
and it was fixed by v7.1beta about 3 years ago.

As far as I know, all versions from v7.1beta and later are OK.

Jim

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  0:32     ` Alexandre Oliva
       [not found]       ` <tprsoavuvw0.fsf@cs.ucsb.edu>
@ 1999-03-31 23:46       ` Alexandre Oliva
  1 sibling, 0 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-31 23:46 UTC (permalink / raw)
  To: Jerry James; +Cc: Kaveh R. Ghazi, egcs

On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:

> On Mar 24, 1999, you wrote:
>> On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:

> [snip]

>> > I don't know if the version of the compiler is tied to the OS or not. 
>> > What version of the compiler are you using?
>> 
>> Dunno, how can I find it out?  I tried various command-line flags and
>> nothing helped.  I even looked at the manual! :-D

> Use the "versions" command.

Thanks.  I get:

oliva@duerer% versions | grep c_dev
I  c_dev                11/12/97  C, 7.0

Pretty old, huh? :-(

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists


^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  5:13 Kaveh R. Ghazi
  1999-03-24  5:20 ` Alexandre Oliva
@ 1999-03-31 23:46 ` Kaveh R. Ghazi
  1 sibling, 0 replies; 20+ messages in thread
From: Kaveh R. Ghazi @ 1999-03-31 23:46 UTC (permalink / raw)
  To: jerry, oliva; +Cc: egcs

 > From: Alexandre Oliva <oliva@dcc.unicamp.br>
 >  
 > On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:
 >  
 > > On Mar 24, 1999, you wrote:
 > >> On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:
 >  
 > > [snip]
 >  
 > >> > I don't know if the version of the compiler is tied to the OS or not. 
 > >> > What version of the compiler are you using?
 > >> 
 > >> Dunno, how can I find it out?  I tried various command-line flags and
 > >> nothing helped.  I even looked at the manual! :-D
 >  
 > > Use the "versions" command.
 >  
 > Thanks.  I get:
 >  
 > oliva@duerer% versions | grep c_dev
 > I  c_dev                11/12/97  C, 7.0
 >  
 > Pretty old, huh? :-(

	PS: I think I recall Jim W. specifically mentioning that some
versions of 7.0 had bugs which caused cc -n32 to fail bootstrapping gcc. 
This came up in our discussion about the o32/n32 libiberty problem but
you'd have to ask him or search the archives back through last November
to get details. 

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  0:45         ` Alexandre Oliva
@ 1999-03-31 23:46           ` Alexandre Oliva
  0 siblings, 0 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-31 23:46 UTC (permalink / raw)
  To: Jerry James; +Cc: egcs

On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:

> Alexandre Oliva <oliva@dcc.unicamp.br>, on 24 Mar 1999 at 05:27:06 -0300
> (how's that for overkill?) you wrote:
>> Thanks.  I get:
>> 
>> oliva@duerer% versions | grep c_dev
>> I  c_dev                11/12/97  C, 7.0
>> 
>> Pretty old, huh? :-(

> Wow, you have one of the original version 7 compilers!

I'm a tough guy, ain't I? :-D

> That's a pretty buggy compiler you've got there, I'm afraid.

Then it's amazing that egcs can bootstrap with it, in spite of needing
4 stages!

> You'd be better off with the version 6.2 compiler.

No way.  We don't have it.  In fact, I believe we've got this old
compiler because people forgot to buy it and someone found us
miserable enough to give us a license for it :-)

Moreover, no one really uses cc here; the only use I have for it is to 
test egcs :-)

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists


^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  5:20 ` Alexandre Oliva
@ 1999-03-31 23:46   ` Alexandre Oliva
  0 siblings, 0 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-31 23:46 UTC (permalink / raw)
  To: Kaveh R. Ghazi; +Cc: jerry, egcs

On Mar 24, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:

> 	PS: I think I recall Jim W. specifically mentioning that some
> versions of 7.0 had bugs which caused cc -n32 to fail bootstrapping gcc. 

Now that you mention it, I think I recall that too.  I'm not sure he
was talking about version 7.0, but I'm pretty sure he mentioned the
buggyness of -n32 on some IRIX compilers.  Well, I'll keep trying :-)

It's good to know that, even though cc -n32 is so broken, we can still
get a working compiler out of it.

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists


^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-23  9:35 Kaveh R. Ghazi
  1999-03-24  0:04 ` Alexandre Oliva
       [not found] ` <oriubr70o5.fsf.cygnus.egcs@dcc.unicamp.br>
@ 1999-03-31 23:46 ` Kaveh R. Ghazi
  2 siblings, 0 replies; 20+ messages in thread
From: Kaveh R. Ghazi @ 1999-03-31 23:46 UTC (permalink / raw)
  To: oliva; +Cc: egcs

 > From: Alexandre Oliva <oliva@dcc.unicamp.br> 
 > 
 > Can anyone else confirm that it's not just my cc for IRIX 6.3 that's
 > broken?  I've never been able to bootstrap egcs with `cc -n32' (the
 > -n32 is needed for libiberty); it always fails `make compare' for all
 > object files (I haven't really checked if some are missing, but the
 > list of failures is huge).  Then, if a proceed with `bootstrap2', it
 > moves the stage3 compiler into stage2, builds a new stage3 and then
 > compare succeeds.  Does this sound familiar to anyone?  Is this
 > expected?  Do we care about fixing it, or should we just document it
 > and put the blame on SGI? :-)

	I've had success with cc -n32 when bootstrapping on Irix 6.2.

I don't know if the version of the compiler is tied to the OS or not. 
What version of the compiler are you using?

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
       [not found] ` <oriubr70o5.fsf.cygnus.egcs@dcc.unicamp.br>
@ 1999-03-24 12:41   ` Jim Wilson
  1999-03-31 23:46     ` Jim Wilson
  0 siblings, 1 reply; 20+ messages in thread
From: Jim Wilson @ 1999-03-24 12:41 UTC (permalink / raw)
  To: oliva; +Cc: egcs

In article <oriubr70o5.fsf.cygnus.egcs@dcc.unicamp.br> you write:
>oliva@duerer% versions | grep c_dev
>I  c_dev                11/12/97  C, 7.0

All versions of SGI C before v7.1beta have bugs that cause gcc to be
miscompiled when using the -n32 option.

With SGI C v7.0, there was only one bug that I found.  It miscompiles a little
bit of code in one of the optimization passes, with the result that the
optimization does not happen.  You still get correct code though.  Hence
you can successfully bootstrap if you go a fourth stage.
stage1: compiled with one optimization disabled
stage2: compiled OK, but only partially optimized
stage3: compiled OK, and completely optimized
stage4: compiled OK, and completely optimized

I don't recall what the v7.0 bug was.  I reported it to SGI about 3 years ago,
and it was fixed by v7.1beta about 3 years ago.

As far as I know, all versions from v7.1beta and later are OK.

Jim

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
@ 1999-03-24  6:16 N8TM
  1999-03-31 23:46 ` N8TM
  0 siblings, 1 reply; 20+ messages in thread
From: N8TM @ 1999-03-24  6:16 UTC (permalink / raw)
  To: ghazi, jerry, oliva; +Cc: egcs

In a message dated 3/24/99 5:13:43 AM Pacific Standard Time,
ghazi@caip.rutgers.edu writes:

> It looks like we have 6.2 and 7.1, but the dates on the 7.1 entry are
>  _older_ than the 6.2 one. (?)  Wierd.
7.1 is the worst of the versions 6.2, 7.0, and 7.1.  My installation
supposedly upgraded to 7.2.1 a week ago but I don't see any fixes for
documented bugs of 7.2.

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-24  5:13 Kaveh R. Ghazi
@ 1999-03-24  5:20 ` Alexandre Oliva
  1999-03-31 23:46   ` Alexandre Oliva
  1999-03-31 23:46 ` Kaveh R. Ghazi
  1 sibling, 1 reply; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-24  5:20 UTC (permalink / raw)
  To: Kaveh R. Ghazi; +Cc: jerry, egcs

On Mar 24, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:

> 	PS: I think I recall Jim W. specifically mentioning that some
> versions of 7.0 had bugs which caused cc -n32 to fail bootstrapping gcc. 

Now that you mention it, I think I recall that too.  I'm not sure he
was talking about version 7.0, but I'm pretty sure he mentioned the
buggyness of -n32 on some IRIX compilers.  Well, I'll keep trying :-)

It's good to know that, even though cc -n32 is so broken, we can still
get a working compiler out of it.

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
@ 1999-03-24  5:13 Kaveh R. Ghazi
  1999-03-24  5:20 ` Alexandre Oliva
  1999-03-31 23:46 ` Kaveh R. Ghazi
  0 siblings, 2 replies; 20+ messages in thread
From: Kaveh R. Ghazi @ 1999-03-24  5:13 UTC (permalink / raw)
  To: jerry, oliva; +Cc: egcs

 > From: Alexandre Oliva <oliva@dcc.unicamp.br>
 >  
 > On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:
 >  
 > > On Mar 24, 1999, you wrote:
 > >> On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:
 >  
 > > [snip]
 >  
 > >> > I don't know if the version of the compiler is tied to the OS or not. 
 > >> > What version of the compiler are you using?
 > >> 
 > >> Dunno, how can I find it out?  I tried various command-line flags and
 > >> nothing helped.  I even looked at the manual! :-D
 >  
 > > Use the "versions" command.
 >  
 > Thanks.  I get:
 >  
 > oliva@duerer% versions | grep c_dev
 > I  c_dev                11/12/97  C, 7.0
 >  
 > Pretty old, huh? :-(

	PS: I think I recall Jim W. specifically mentioning that some
versions of 7.0 had bugs which caused cc -n32 to fail bootstrapping gcc. 
This came up in our discussion about the o32/n32 libiberty problem but
you'd have to ask him or search the archives back through last November
to get details. 

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
@ 1999-03-24  5:10 Kaveh R. Ghazi
  1999-03-31 23:46 ` Kaveh R. Ghazi
  0 siblings, 1 reply; 20+ messages in thread
From: Kaveh R. Ghazi @ 1999-03-24  5:10 UTC (permalink / raw)
  To: jerry, oliva; +Cc: egcs

 > From: Alexandre Oliva <oliva@dcc.unicamp.br>
 >  
 > On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:
 >  
 > > On Mar 24, 1999, you wrote:
 > >> On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:
 >  
 > > [snip]
 >  
 > >> > I don't know if the version of the compiler is tied to the OS or not. 
 > >> > What version of the compiler are you using?
 > >> 
 > >> Dunno, how can I find it out?  I tried various command-line flags and
 > >> nothing helped.  I even looked at the manual! :-D
 >  
 > > Use the "versions" command.
 >  
 > Thanks.  I get:
 >  
 > oliva@duerer% versions | grep c_dev
 > I  c_dev                11/12/97  C, 7.0
 >  
 > Pretty old, huh? :-(


	Hmm, The full listing I get is:

I  c_dev                07/29/1998  C, 6.2
I  c_dev.src            07/29/1998  C Optional Sources
I  c_dev.src.acpp       07/29/1998  C acpp Source
I  c_dev.src.customalloc  07/29/1998  C customalloc Source
I  c_dev.sw             07/29/1998  C Software
I  c_dev.sw.speclib     07/29/1998  C NonShared Libraries for Benchmarks
I  c_dev.sw32           07/29/1998  C Software
I  c_dev.sw32.speclib   07/29/1998  C NonShared Libraries for Benchmarks
I  c_dev                06/02/1997  C, 7.1 on irix 6.2
I  c_dev.books          06/02/1997  C IRIS InSight Books
I  c_dev.books.CLanguageRef  06/02/1997  C Language Reference Manual
I  c_dev.hdr            06/02/1997  C Headers
I  c_dev.hdr.lib        06/02/1997  C Library Headers
I  c_dev.man            06/02/1997  C Manual Pages
I  c_dev.man.c          06/02/1997  C Compiler Man Pages
I  c_dev.man.copt       06/02/1997  C Copt Man Pages
I  c_dev.man.relnotes   06/02/1997  C Release Notes
I  c_dev.man.util       06/02/1997  C Utility Man Pages
I  c_dev.sw             06/02/1997  C Software
I  c_dev.sw.c           06/02/1997  C Compiler
I  c_dev.sw.copt        06/02/1997  C Scalar Optimizer
I  c_dev.sw.lib         06/02/1997  C Libraries
I  c_dev.sw.util        06/02/1997  C Utilities
I  c_dev.sw.v6_c        06/02/1997  C Compiler Version 6
I  c_dev.sw.v6_copt     06/02/1997  C Scalar Optimizer Version 6
I  c_dev.sw64           06/02/1997  C Software
I  c_dev.sw64.lib       06/02/1997  C Libraries 64bit

It looks like we have 6.2 and 7.1, but the dates on the 7.1 entry are
_older_ than the 6.2 one. (?)  Wierd.

--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
       [not found]       ` <tprsoavuvw0.fsf@cs.ucsb.edu>
@ 1999-03-24  0:45         ` Alexandre Oliva
  1999-03-31 23:46           ` Alexandre Oliva
  0 siblings, 1 reply; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-24  0:45 UTC (permalink / raw)
  To: Jerry James; +Cc: egcs

On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:

> Alexandre Oliva <oliva@dcc.unicamp.br>, on 24 Mar 1999 at 05:27:06 -0300
> (how's that for overkill?) you wrote:
>> Thanks.  I get:
>> 
>> oliva@duerer% versions | grep c_dev
>> I  c_dev                11/12/97  C, 7.0
>> 
>> Pretty old, huh? :-(

> Wow, you have one of the original version 7 compilers!

I'm a tough guy, ain't I? :-D

> That's a pretty buggy compiler you've got there, I'm afraid.

Then it's amazing that egcs can bootstrap with it, in spite of needing
4 stages!

> You'd be better off with the version 6.2 compiler.

No way.  We don't have it.  In fact, I believe we've got this old
compiler because people forgot to buy it and someone found us
miserable enough to give us a license for it :-)

Moreover, no one really uses cc here; the only use I have for it is to 
test egcs :-)

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
       [not found]   ` <tprvhfruwva.fsf@cs.ucsb.edu>
@ 1999-03-24  0:32     ` Alexandre Oliva
       [not found]       ` <tprsoavuvw0.fsf@cs.ucsb.edu>
  1999-03-31 23:46       ` Alexandre Oliva
  0 siblings, 2 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-24  0:32 UTC (permalink / raw)
  To: Jerry James; +Cc: Kaveh R. Ghazi, egcs

On Mar 24, 1999, jerry@cs.ucsb.edu (Jerry James) wrote:

> On Mar 24, 1999, you wrote:
>> On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:

> [snip]

>> > I don't know if the version of the compiler is tied to the OS or not. 
>> > What version of the compiler are you using?
>> 
>> Dunno, how can I find it out?  I tried various command-line flags and
>> nothing helped.  I even looked at the manual! :-D

> Use the "versions" command.

Thanks.  I get:

oliva@duerer% versions | grep c_dev
I  c_dev                11/12/97  C, 7.0

Pretty old, huh? :-(

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
  1999-03-23  9:35 Kaveh R. Ghazi
@ 1999-03-24  0:04 ` Alexandre Oliva
       [not found]   ` <tprvhfruwva.fsf@cs.ucsb.edu>
  1999-03-31 23:46   ` Alexandre Oliva
       [not found] ` <oriubr70o5.fsf.cygnus.egcs@dcc.unicamp.br>
  1999-03-31 23:46 ` Kaveh R. Ghazi
  2 siblings, 2 replies; 20+ messages in thread
From: Alexandre Oliva @ 1999-03-24  0:04 UTC (permalink / raw)
  To: Kaveh R. Ghazi; +Cc: egcs

On Mar 23, 1999, "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu> wrote:

>> From: Alexandre Oliva <oliva@dcc.unicamp.br> 

>> I've never been able to bootstrap egcs with `cc -n32' (the -n32 is
>> needed for libiberty); it always fails `make compare'

> 	I've had success with cc -n32 when bootstrapping on Irix 6.2.

> I don't know if the version of the compiler is tied to the OS or not. 
> What version of the compiler are you using?

Dunno, how can I find it out?  I tried various command-line flags and
nothing helped.  I even looked at the manual! :-D

-- 
Alexandre Oliva http://www.dcc.unicamp.br/~oliva IC-Unicamp, Brasil
{oliva,Alexandre.Oliva}@dcc.unicamp.br  aoliva@{acm.org,computer.org}
oliva@{gnu.org,kaffe.org,egcs.cygnus.com,samba.org}
*** E-mail about software projects will be forwarded to mailing lists

^ permalink raw reply	[flat|nested] 20+ messages in thread

* Re: bootstrapping with `cc -n32' on IRIX 6.3 fails compare
@ 1999-03-23  9:35 Kaveh R. Ghazi
  1999-03-24  0:04 ` Alexandre Oliva
                   ` (2 more replies)
  0 siblings, 3 replies; 20+ messages in thread
From: Kaveh R. Ghazi @ 1999-03-23  9:35 UTC (permalink / raw)
  To: oliva; +Cc: egcs

 > From: Alexandre Oliva <oliva@dcc.unicamp.br> 
 > 
 > Can anyone else confirm that it's not just my cc for IRIX 6.3 that's
 > broken?  I've never been able to bootstrap egcs with `cc -n32' (the
 > -n32 is needed for libiberty); it always fails `make compare' for all
 > object files (I haven't really checked if some are missing, but the
 > list of failures is huge).  Then, if a proceed with `bootstrap2', it
 > moves the stage3 compiler into stage2, builds a new stage3 and then
 > compare succeeds.  Does this sound familiar to anyone?  Is this
 > expected?  Do we care about fixing it, or should we just document it
 > and put the blame on SGI? :-)

	I've had success with cc -n32 when bootstrapping on Irix 6.2.

I don't know if the version of the compiler is tied to the OS or not. 
What version of the compiler are you using?

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

^ permalink raw reply	[flat|nested] 20+ messages in thread

end of thread, other threads:[~1999-03-31 23:46 UTC | newest]

Thread overview: 20+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1999-03-21 10:01 bootstrapping with `cc -n32' on IRIX 6.3 fails compare Alexandre Oliva
1999-03-31 23:46 ` Alexandre Oliva
1999-03-23  9:35 Kaveh R. Ghazi
1999-03-24  0:04 ` Alexandre Oliva
     [not found]   ` <tprvhfruwva.fsf@cs.ucsb.edu>
1999-03-24  0:32     ` Alexandre Oliva
     [not found]       ` <tprsoavuvw0.fsf@cs.ucsb.edu>
1999-03-24  0:45         ` Alexandre Oliva
1999-03-31 23:46           ` Alexandre Oliva
1999-03-31 23:46       ` Alexandre Oliva
1999-03-31 23:46   ` Alexandre Oliva
     [not found] ` <oriubr70o5.fsf.cygnus.egcs@dcc.unicamp.br>
1999-03-24 12:41   ` Jim Wilson
1999-03-31 23:46     ` Jim Wilson
1999-03-31 23:46 ` Kaveh R. Ghazi
1999-03-24  5:10 Kaveh R. Ghazi
1999-03-31 23:46 ` Kaveh R. Ghazi
1999-03-24  5:13 Kaveh R. Ghazi
1999-03-24  5:20 ` Alexandre Oliva
1999-03-31 23:46   ` Alexandre Oliva
1999-03-31 23:46 ` Kaveh R. Ghazi
1999-03-24  6:16 N8TM
1999-03-31 23:46 ` N8TM

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).