S390[x] configuration update

S390[x] configuration update

[Andreas Schwab]

This updates the S390/S390x configuration to follow the latest wisdom.

2002-04-28  Andreas Schwab  <schwab@suse.de>

	* config/s390/s390.mh (NATDEPFILES): Remove slib.o and
	core-regset.o, add linux-proc.o and gcore.o.

--- gdb/config/s390/s390.mh	2002/04/27 18:51:58	1.1
+++ gdb/config/s390/s390.mh	2002/04/27 18:55:00
@@ -4,9 +4,8 @@ XM_FILE= xm-linux.h
 XM_CLIBS=
 
 NAT_FILE= nm-linux.h
-NATDEPFILES= infptrace.o solib.o inftarg.o fork-child.o corelow.o s390-nat.o core-aout.o core-regset.o
-# post 5.0 natdepfiles.
-NATDEPFILES+= thread-db.o lin-lwp.o proc-service.o 
+NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o s390-nat.o \
+	core-aout.o linux-proc.o gcore.o thread-db.o lin-lwp.o proc-service.o
 LOADLIBES = -ldl -rdynamic
 
 

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: S390[x] configuration update

[Daniel Jacobowitz]

On Sun, Apr 28, 2002 at 12:41:03AM +0200, Andreas Schwab wrote:
> This updates the S390/S390x configuration to follow the latest wisdom.
> 
> 2002-04-28  Andreas Schwab  <schwab@suse.de>
> 
> 	* config/s390/s390.mh (NATDEPFILES): Remove slib.o and
> 	core-regset.o, add linux-proc.o and gcore.o.

Er.... what?  Why remove solib.o and core-regset.o?

Also, I bet you didn't build test this.  If you don't add linux-proc.o
then it won't link properly.  I forgot to submit the patch :(

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer

Re: S390[x] configuration update

[Andreas Schwab]

Daniel Jacobowitz <drow@mvista.com> writes:

|> On Sun, Apr 28, 2002 at 12:41:03AM +0200, Andreas Schwab wrote:
|> > This updates the S390/S390x configuration to follow the latest wisdom.
|> > 
|> > 2002-04-28  Andreas Schwab  <schwab@suse.de>
|> > 
|> > 	* config/s390/s390.mh (NATDEPFILES): Remove slib.o and
|> > 	core-regset.o, add linux-proc.o and gcore.o.
|> 
|> Er.... what?  Why remove solib.o and core-regset.o?

solib.o is in TDEPFILES, the same as all other linux targets.  But
removing core-regset.o was wrong.  Misunderstanding from my side, sorry.

|> Also, I bet you didn't build test this.

Of course, I did.  But I forgot to test corefile functionality.

Btw, config/i386/linux.mh NATDEPFILES lists linux-proc.o twice.

2002-04-28  Andreas Schwab  <schwab@suse.de>

	* config/s390/s390.mh (NATDEPFILES): Remove slib.o, add
	linux-proc.o and gcore.o.

--- gdb/config/s390/s390.mh	2002/04/27 18:51:58	1.1
+++ gdb/config/s390/s390.mh	2002/04/27 23:42:01
@@ -4,9 +4,9 @@ XM_FILE= xm-linux.h
 XM_CLIBS=
 
 NAT_FILE= nm-linux.h
-NATDEPFILES= infptrace.o solib.o inftarg.o fork-child.o corelow.o s390-nat.o core-aout.o core-regset.o
-# post 5.0 natdepfiles.
-NATDEPFILES+= thread-db.o lin-lwp.o proc-service.o 
+NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o s390-nat.o \
+	core-aout.o core-regset.o linux-proc.o gcore.o thread-db.o lin-lwp.o \
+	proc-service.o
 LOADLIBES = -ldl -rdynamic
 
 

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: S390[x] configuration update

[Daniel Jacobowitz]

On Sun, Apr 28, 2002 at 02:13:34AM +0200, Andreas Schwab wrote:
> Daniel Jacobowitz <drow@mvista.com> writes:
> 
> |> On Sun, Apr 28, 2002 at 12:41:03AM +0200, Andreas Schwab wrote:
> |> > This updates the S390/S390x configuration to follow the latest wisdom.
> |> > 
> |> > 2002-04-28  Andreas Schwab  <schwab@suse.de>
> |> > 
> |> > 	* config/s390/s390.mh (NATDEPFILES): Remove slib.o and
> |> > 	core-regset.o, add linux-proc.o and gcore.o.
> |> 
> |> Er.... what?  Why remove solib.o and core-regset.o?
> 
> solib.o is in TDEPFILES, the same as all other linux targets.  But
> removing core-regset.o was wrong.  Misunderstanding from my side, sorry.

Oh, right.  I forgot that it had moved.

> |> Also, I bet you didn't build test this.
> 
> Of course, I did.  But I forgot to test corefile functionality.
> 
> Btw, config/i386/linux.mh NATDEPFILES lists linux-proc.o twice.

I wonder why I thought that... I was assuming you hadn't added
linux-proc.o, which you obviously did.  Sorry.  The updated patch looks
good to me; you can probably just commit it, since it fixes a build
failure and is quite straightforward.


> 2002-04-28  Andreas Schwab  <schwab@suse.de>
> 
> 	* config/s390/s390.mh (NATDEPFILES): Remove slib.o, add
> 	linux-proc.o and gcore.o.

"solib.o"

> 
> --- gdb/config/s390/s390.mh	2002/04/27 18:51:58	1.1
> +++ gdb/config/s390/s390.mh	2002/04/27 23:42:01
> @@ -4,9 +4,9 @@ XM_FILE= xm-linux.h
>  XM_CLIBS=
>  
>  NAT_FILE= nm-linux.h
> -NATDEPFILES= infptrace.o solib.o inftarg.o fork-child.o corelow.o s390-nat.o core-aout.o core-regset.o
> -# post 5.0 natdepfiles.
> -NATDEPFILES+= thread-db.o lin-lwp.o proc-service.o 
> +NATDEPFILES= infptrace.o inftarg.o fork-child.o corelow.o s390-nat.o \
> +	core-aout.o core-regset.o linux-proc.o gcore.o thread-db.o lin-lwp.o \
> +	proc-service.o
>  LOADLIBES = -ldl -rdynamic
>  
>  
> 
> Andreas.
> 
> -- 
> Andreas Schwab, SuSE Labs, schwab@suse.de
> SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
> Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
> "And now for something completely different."
> 

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer

Re: S390[x] configuration update

[Andreas Schwab]

Daniel Jacobowitz <drow@mvista.com> writes:

|> The updated patch looks
|> good to me; you can probably just commit it, since it fixes a build
|> failure and is quite straightforward.

Done.  What about the 5.2 branch?  Is it still possible to get this into
the release?

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: S390[x] configuration update

[Daniel Jacobowitz]

On Sun, Apr 28, 2002 at 02:31:33AM +0200, Andreas Schwab wrote:
> Daniel Jacobowitz <drow@mvista.com> writes:
> 
> |> The updated patch looks
> |> good to me; you can probably just commit it, since it fixes a build
> |> failure and is quite straightforward.
> 
> Done.  What about the 5.2 branch?  Is it still possible to get this into
> the release?

Don't think so, since Andrew seems to have already rolled the release
tarball.

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer

Re: S390[x] configuration update

[Andrew Cagney]

> On Sun, Apr 28, 2002 at 02:31:33AM +0200, Andreas Schwab wrote:
> 
>> Daniel Jacobowitz <drow@mvista.com> writes:
>> 
>> |> The updated patch looks
>> |> good to me; you can probably just commit it, since it fixes a build
>> |> failure and is quite straightforward.
>> 
>> Done.  What about the 5.2 branch?  Is it still possible to get this into
>> the release?
> 
> 
> Don't think so, since Andrew seems to have already rolled the release
> tarball.

Yes, definitly closed.

I also suspect there is no benefit in pulling it into the 5.2 branch 
once unfrozen.  The existing stuff ``works'' - as they say, if it ain't 
broke don't fix it :-)

enjoy,
Andrew

Re: S390[x] configuration update

[Daniel Jacobowitz]

On Mon, Apr 29, 2002 at 12:18:40PM -0400, Andrew Cagney wrote:
> >On Sun, Apr 28, 2002 at 02:31:33AM +0200, Andreas Schwab wrote:
> >
> >>Daniel Jacobowitz <drow@mvista.com> writes:
> >>
> >>|> The updated patch looks
> >>|> good to me; you can probably just commit it, since it fixes a build
> >>|> failure and is quite straightforward.
> >>
> >>Done.  What about the 5.2 branch?  Is it still possible to get this into
> >>the release?
> >
> >
> >Don't think so, since Andrew seems to have already rolled the release
> >tarball.
> 
> Yes, definitly closed.
> 
> I also suspect there is no benefit in pulling it into the 5.2 branch 
> once unfrozen.  The existing stuff ``works'' - as they say, if it ain't 
> broke don't fix it :-)

By ``works'' you mean doesn't build? :)  See above; 5.2 won't build on
S/390.

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer

Re: S390[x] configuration update

[Andrew Cagney]

>> Yes, definitly closed.
>> 
>> I also suspect there is no benefit in pulling it into the 5.2 branch 
>> once unfrozen.  The existing stuff ``works'' - as they say, if it ain't 
>> broke don't fix it :-)
> 
> 
> By ``works'' you mean doesn't build? :)  See above; 5.2 won't build on
> S/390.

My understanding of Andreas's original post is that it updated things to 
current wisdom rather than fixed a pre-existing build problem.

http://sources.redhat.com/ml/gdb-patches/2002-04/msg01104.html

enjoy,
Andrew

Re: S390[x] configuration update

[Daniel Jacobowitz]

On Mon, Apr 29, 2002 at 02:03:00PM -0400, Andrew Cagney wrote:
> 
> >>Yes, definitly closed.
> >>
> >>I also suspect there is no benefit in pulling it into the 5.2 branch 
> >>once unfrozen.  The existing stuff ``works'' - as they say, if it ain't 
> >>broke don't fix it :-)
> >
> >
> >By ``works'' you mean doesn't build? :)  See above; 5.2 won't build on
> >S/390.
> 
> My understanding of Andreas's original post is that it updated things to 
> current wisdom rather than fixed a pre-existing build problem.
> 
> http://sources.redhat.com/ml/gdb-patches/2002-04/msg01104.html

Different interpretation of his phrasing, perhaps.  But no GNU/Linux
target that uses config/nm-linux.h can link without including
linux-proc.o; I assume it was missed because the .mh file is named
s390.mh instead of s390/linux.mh.  S/390 doesn't build in 5.2 without
this patch.

-- 
Daniel Jacobowitz                           Carnegie Mellon University
MontaVista Software                         Debian GNU/Linux Developer

Re: S390[x] configuration update

[Andrew Cagney]

>> http://sources.redhat.com/ml/gdb-patches/2002-04/msg01104.html
> 
> 
> Different interpretation of his phrasing, perhaps.

Just had another look, and still can't see a ``5.2 does not build'' 
anywhere :-)

>  But no GNU/Linux
> target that uses config/nm-linux.h can link without including
> linux-proc.o; I assume it was missed because the .mh file is named
> s390.mh instead of s390/linux.mh.  S/390 doesn't build in 5.2 without
> this patch.

M'kay, adding it to the 5.2 release notes.

s390, FreeBSD 4.4 and other stuff suggest a 5.2.1

enjoy,
Andrew

Re: S390[x] configuration update

[Andreas Schwab]

Andrew Cagney <ac131313@cygnus.com> writes:

|> >> Yes, definitly closed.
|> >> I also suspect there is no benefit in pulling it into the 5.2 branch
|> >> once unfrozen.  The existing stuff ``works'' - as they say, if it ain't
|> >> broke don't fix it :-)
|> > By ``works'' you mean doesn't build? :)  See above; 5.2 won't build on
|> > S/390.
|> 
|> My understanding of Andreas's original post is that it updated things to
|> current wisdom rather than fixed a pre-existing build problem.

Sloppy phrasing, sorry.  The current wisdom includes that linux-proc.o has
to be linked in.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: S390[x] configuration update

[Andrew Cagney]

> Andrew Cagney <ac131313@cygnus.com> writes:
> 
> |> >> Yes, definitly closed.
> |> >> I also suspect there is no benefit in pulling it into the 5.2 branch
> |> >> once unfrozen.  The existing stuff ``works'' - as they say, if it ain't
> |> >> broke don't fix it :-)
> |> > By ``works'' you mean doesn't build? :)  See above; 5.2 won't build on
> |> > S/390.
> |> 
> |> My understanding of Andreas's original post is that it updated things to
> |> current wisdom rather than fixed a pre-existing build problem.
> 
> Sloppy phrasing, sorry.  The current wisdom includes that linux-proc.o has
> to be linked in.

Ah, ok, as the branch is unfrozen, it can go in (per daniel's suggestion).

Andrew