public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* all-gcc should depend on maybe-all-{flex,texinfo}
@ 2004-07-08 21:11 Alexandre Oliva
  2004-07-09  9:12 ` Paolo Bonzini
  0 siblings, 1 reply; 6+ messages in thread
From: Alexandre Oliva @ 2004-07-08 21:11 UTC (permalink / raw)
  To: gcc-patches

[-- Attachment #1: Type: text/plain, Size: 721 bytes --]

Since gcc configure is going to set things up to use flex from the
build tree if it finds flex sources in the source tree, we need this
dependency.  In general, if you have a unified tree, you won't run
into a problem because gcc depends on ld, that depends on flex.  But
if you happen to disable ld, or if it's not supported on your
platform, you end up without the dependency, and lose.

This patch fixes it, while at the same time fixing the incompatibility
between the dependencies of all-gcc and all-bootstrap.  Hmm...  Maybe
we could generate the all-gcc deps in the same way we generate the
all-bootstrap ones?  Does it seem to make sense to anyone like it does
to me?

I'm checking this in gcc mainline and src.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gcc-needs-flex.patch --]
[-- Type: text/x-patch, Size: 3831 bytes --]

Index: ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>

	* Makefile.def (host_modules): Set bootstrap=true for flex.
	* Makefile.tpl (all-gcc): Depend on texinfo and flex.
	* Makefile.in: Rebuilt.

Index: Makefile.def
===================================================================
RCS file: /cvs/gcc/gcc/Makefile.def,v
retrieving revision 1.28
diff -u -p -r1.28 Makefile.def
--- Makefile.def 2 Jul 2004 07:55:35 -0000 1.28
+++ Makefile.def 8 Jul 2004 19:28:57 -0000
@@ -50,7 +50,7 @@ host_modules= { module= fastjar; no_chec
 host_modules= { module= fileutils; };
 host_modules= { module= findutils; };
 host_modules= { module= find; };
-host_modules= { module= flex; no_check_cross= true; };
+host_modules= { module= flex; no_check_cross= true; bootstrap=true; };
 host_modules= { module= gas; bootstrap=true; };
 host_modules= { module= gawk; };
 host_modules= { module= gettext; };
Index: Makefile.in
===================================================================
RCS file: /cvs/gcc/gcc/Makefile.in,v
retrieving revision 1.228
diff -u -p -r1.228 Makefile.in
--- Makefile.in 2 Jul 2004 07:55:35 -0000 1.228
+++ Makefile.in 8 Jul 2004 19:29:03 -0000
@@ -29195,13 +29195,13 @@ all-build-m4: maybe-all-build-libiberty 
 # GCC needs to identify certain tools.
 # GCC also needs the information exported by the intl configure script.
 configure-gcc: maybe-configure-intl maybe-configure-binutils maybe-configure-gas maybe-configure-ld maybe-configure-bison maybe-configure-flex
-all-gcc: maybe-all-libiberty maybe-all-intl maybe-all-bison maybe-all-byacc maybe-all-binutils maybe-all-gas maybe-all-ld maybe-all-zlib maybe-all-libbanshee maybe-all-libcpp
+all-gcc: maybe-all-libiberty maybe-all-intl maybe-all-texinfo maybe-all-bison maybe-all-byacc maybe-all-flex maybe-all-binutils maybe-all-gas maybe-all-ld maybe-all-zlib maybe-all-libbanshee maybe-all-libcpp
 configure-libcpp: maybe-configure-libiberty maybe-configure-intl
 all-libcpp: maybe-all-libiberty maybe-all-intl
 # This is a slightly kludgy method of getting dependencies on 
 # all-build-libiberty correct; it would be better to build it every time.
 all-gcc: maybe-all-build-libiberty
-all-bootstrap: maybe-all-binutils maybe-all-bison maybe-all-byacc maybe-all-gas maybe-all-intl maybe-all-ld maybe-all-libcpp maybe-all-libbanshee maybe-all-libiberty maybe-all-texinfo maybe-all-zlib 
+all-bootstrap: maybe-all-binutils maybe-all-bison maybe-all-byacc maybe-all-flex maybe-all-gas maybe-all-intl maybe-all-ld maybe-all-libcpp maybe-all-libbanshee maybe-all-libiberty maybe-all-texinfo maybe-all-zlib 
 
 # Host modules specific to gdb.
 # GDB needs to know that the simulator is being built.
Index: Makefile.tpl
===================================================================
RCS file: /cvs/gcc/gcc/Makefile.tpl,v
retrieving revision 1.106
diff -u -p -r1.106 Makefile.tpl
--- Makefile.tpl 2 Jul 2004 07:55:36 -0000 1.106
+++ Makefile.tpl 8 Jul 2004 19:29:04 -0000
@@ -1675,7 +1675,7 @@ all-build-m4: maybe-all-build-libiberty 
 # GCC needs to identify certain tools.
 # GCC also needs the information exported by the intl configure script.
 configure-gcc: maybe-configure-intl maybe-configure-binutils maybe-configure-gas maybe-configure-ld maybe-configure-bison maybe-configure-flex
-all-gcc: maybe-all-libiberty maybe-all-intl maybe-all-bison maybe-all-byacc maybe-all-binutils maybe-all-gas maybe-all-ld maybe-all-zlib maybe-all-libbanshee maybe-all-libcpp
+all-gcc: maybe-all-libiberty maybe-all-intl maybe-all-texinfo maybe-all-bison maybe-all-byacc maybe-all-flex maybe-all-binutils maybe-all-gas maybe-all-ld maybe-all-zlib maybe-all-libbanshee maybe-all-libcpp
 configure-libcpp: maybe-configure-libiberty maybe-configure-intl
 all-libcpp: maybe-all-libiberty maybe-all-intl
 # This is a slightly kludgy method of getting dependencies on 

[-- Attachment #3: Type: text/plain, Size: 188 bytes --]


-- 
Alexandre Oliva             http://www.ic.unicamp.br/~oliva/
Red Hat Compiler Engineer   aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist  oliva@{lsd.ic.unicamp.br, gnu.org}

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: all-gcc should depend on maybe-all-{flex,texinfo}
  2004-07-08 21:11 all-gcc should depend on maybe-all-{flex,texinfo} Alexandre Oliva
@ 2004-07-09  9:12 ` Paolo Bonzini
  2004-07-09 13:27   ` Wouter van Heyst
  2004-07-09 14:40   ` Andreas Schwab
  0 siblings, 2 replies; 6+ messages in thread
From: Paolo Bonzini @ 2004-07-09  9:12 UTC (permalink / raw)
  To: gcc-patches

> This patch fixes it, while at the same time fixing the incompatibility
> between the dependencies of all-gcc and all-bootstrap.  Hmm...  Maybe
> we could generate the all-gcc deps in the same way we generate the
> all-bootstrap ones?  Does it seem to make sense to anyone like it does
> to me?

Why cannot all-gcc depend on all-bootstrap?

Paolo

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: all-gcc should depend on maybe-all-{flex,texinfo}
  2004-07-09  9:12 ` Paolo Bonzini
@ 2004-07-09 13:27   ` Wouter van Heyst
  2004-07-09 14:40   ` Andreas Schwab
  1 sibling, 0 replies; 6+ messages in thread
From: Wouter van Heyst @ 2004-07-09 13:27 UTC (permalink / raw)
  To: gcc-patches

On Fri, Jul 09, 2004 at 09:18:41AM +0200, Paolo Bonzini wrote:
> >This patch fixes it, while at the same time fixing the incompatibility
> >between the dependencies of all-gcc and all-bootstrap.  Hmm...  Maybe
> >we could generate the all-gcc deps in the same way we generate the
> >all-bootstrap ones?  Does it seem to make sense to anyone like it does
> >to me?
> 
> Why cannot all-gcc depend on all-bootstrap?

I'm not actually familiar with how the Makefiles are set up, but
all-bootstrap suggests it can't be used with crossbuilds, where all-gcc
is. Of course, getting lost looking at Makefile.in, I could be wrong.

Wouter van Heyst

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: all-gcc should depend on maybe-all-{flex,texinfo}
  2004-07-09  9:12 ` Paolo Bonzini
  2004-07-09 13:27   ` Wouter van Heyst
@ 2004-07-09 14:40   ` Andreas Schwab
  2004-07-09 14:46     ` Paolo Bonzini
  1 sibling, 1 reply; 6+ messages in thread
From: Andreas Schwab @ 2004-07-09 14:40 UTC (permalink / raw)
  To: Paolo Bonzini; +Cc: gcc-patches

Paolo Bonzini <bonzini@gnu.org> writes:

>> This patch fixes it, while at the same time fixing the incompatibility
>> between the dependencies of all-gcc and all-bootstrap.  Hmm...  Maybe
>> we could generate the all-gcc deps in the same way we generate the
>> all-bootstrap ones?  Does it seem to make sense to anyone like it does
>> to me?
>
> Why cannot all-gcc depend on all-bootstrap?

There should be a way to build the compiler without going through the
whole bootstrap process.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux AG, Maxfeldstraße 5, 90409 Nürnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: all-gcc should depend on maybe-all-{flex,texinfo}
  2004-07-09 14:40   ` Andreas Schwab
@ 2004-07-09 14:46     ` Paolo Bonzini
  2004-07-09 15:09       ` Alexandre Oliva
  0 siblings, 1 reply; 6+ messages in thread
From: Paolo Bonzini @ 2004-07-09 14:46 UTC (permalink / raw)
  To: Andreas Schwab; +Cc: Paolo Bonzini, gcc-patches

> There should be a way to build the compiler without going through the
> whole bootstrap process.

all-bootstrap was ill-named.  It should be actually *all-prebootstrap*, 
that's why I proposed that all-gcc depended on it.

Paolo

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: all-gcc should depend on maybe-all-{flex,texinfo}
  2004-07-09 14:46     ` Paolo Bonzini
@ 2004-07-09 15:09       ` Alexandre Oliva
  0 siblings, 0 replies; 6+ messages in thread
From: Alexandre Oliva @ 2004-07-09 15:09 UTC (permalink / raw)
  To: Paolo Bonzini; +Cc: Andreas Schwab, Paolo Bonzini, gcc-patches

On Jul  9, 2004, Paolo Bonzini <paolo.bonzini@polimi.it> wrote:

>> There should be a way to build the compiler without going through the
>> whole bootstrap process.

> all-bootstrap was ill-named.  It should be actually
> *all-prebootstrap*, that's why I proposed that all-gcc depended on it.

all-gcc-deps sounds good to me.

-- 
Alexandre Oliva             http://www.ic.unicamp.br/~oliva/
Red Hat Compiler Engineer   aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist  oliva@{lsd.ic.unicamp.br, gnu.org}

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2004-07-09 14:21 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-07-08 21:11 all-gcc should depend on maybe-all-{flex,texinfo} Alexandre Oliva
2004-07-09  9:12 ` Paolo Bonzini
2004-07-09 13:27   ` Wouter van Heyst
2004-07-09 14:40   ` Andreas Schwab
2004-07-09 14:46     ` Paolo Bonzini
2004-07-09 15:09       ` Alexandre Oliva

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).