192313 - the setup for the dec compilers needs to have -std1 removed and -pthreads added

Status: RESOLVED WONTFIX

(SeaMonkey :: Build Config, defect)

Description

[David Warren]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0

gtk fails with -std1 due to the inline directive and pthreads fails without -pthread

Reproducible: Always

Steps to Reproduce:
1.setenv CC cc setenv CXX cxx
2.configure --with-pthreads
3.gmake

Actual Results:  
configure fails until changes are made in all configure scripts, after which it
configures and compiles

Expected Results:  
configure and build

Comment 1

[hacker formerly known as [e-mail]]

Can you elaborate on the -std1 issue?  And which version of OSF/1 and the
compilers are you using?

Comment 2

[David Warren]

Her is the requested info:

We are running V5.1A + patches:
OSF1 sunny.atmos.washington.edu V5.1 1885 alpha
I am using the DEC C and CXX compilers. 
 -std1
      Selects the strict ANSI language mode. Strictly enforces the ANSI C
      standard and all its prohibitions (such as those that apply to the han-
      dling of void types, the definition of lvalues in expressions, the mix-
      ing of integrals and pointers, and the modification of rvalues).

      This option does not restrict the Tru64 UNIX name space (for example,
      the names defined in system header files). To restrict that name space
      so that only ANSI C reserved names are visible from the ANSI C header
      files, use the _ANSI_C_SOURCE macro. See standards(5) for more details.

      This option sets the macro __STDC__ to 1. Note that this option also
      affects linker-defined symbols. See ld(1) for more information.

      This option, combined with -O3 or higher, turns on ANSI aliasing rules
      (-ansi_alias option).


This breaks gtk, as the gtk headers use the inline keyword, which is not ansi.
Without -std1, the compiler takes all ansi constructs and is more forgiving of
things that are extentions.We are running V5.1A + patches:
OSF1 sunny.atmos.washington.edu V5.1 1885 alpha
I am using the DEC C and CXX compilers. 
 -std1
      Selects the strict ANSI language mode. Strictly enforces the ANSI C
      standard and all its prohibitions (such as those that apply to the han-
      dling of void types, the definition of lvalues in expressions, the mix-
      ing of integrals and pointers, and the modification of rvalues).

      This option does not restrict the Tru64 UNIX name space (for example,
      the names defined in system header files). To restrict that name space
      so that only ANSI C reserved names are visible from the ANSI C header
      files, use the _ANSI_C_SOURCE macro. See standards(5) for more details.

      This option sets the macro __STDC__ to 1. Note that this option also
      affects linker-defined symbols. See ld(1) for more information.

      This option, combined with -O3 or higher, turns on ANSI aliasing rules
      (-ansi_alias option).


This breaks gtk, as the gtk headers use the inline keyword, which is not ansi.
Without -std1, the compiler takes all ansi constructs and is more forgiving of
things that are extentions.We are running V5.1A + patches:
OSF1 sunny.atmos.washington.edu V5.1 1885 alpha
I am using the DEC C and CXX compilers. 
 -std1
      Selects the strict ANSI language mode. Strictly enforces the ANSI C
      standard and all its prohibitions (such as those that apply to the han-
      dling of void types, the definition of lvalues in expressions, the mix-
      ing of integrals and pointers, and the modification of rvalues).

      This option does not restrict the Tru64 UNIX name space (for example,
      the names defined in system header files). To restrict that name space
      so that only ANSI C reserved names are visible from the ANSI C header
      files, use the _ANSI_C_SOURCE macro. See standards(5) for more details.

      This option sets the macro __STDC__ to 1. Note that this option also
      affects linker-defined symbols. See ld(1) for more information.

      This option, combined with -O3 or higher, turns on ANSI aliasing rules
      (-ansi_alias option).


This breaks gtk, as the gtk headers use the inline keyword, which is not ansi.
Without -std1, the compiler takes all ansi constructs and is more forgiving of
things that are extentions.

Comment 3

[hacker formerly known as [e-mail]]

Hrm. Ok, the -std issue sounds like the typical 'compiling glib/gtk with
different flags used to compile mozilla' problem.  Based upon experience with
Solaris, if you use a strict ansi mode to compile glib/gtk, then they won't have
the inline keyword in their headers either. 

Wan-Teh, any thoughts? It looks like you added that flag way back in revision
1.104 of configure.in.  I'm guessing that those flags were just carried over
from NSPR?

Comment 4

[Wan-Teh Chang]

You should be able to compile gtk without using the inline
key word.  Do you *want* to or *need* to use the inline
key word?

If removing the -std1 compiler flag is the most expedient
solution, that's fine by me.

Comment 5

[hacker formerly known as [e-mail]]

Attachment: Remove -std1 uses

Comment 6

[hacker formerly known as [e-mail]]

Attachment: Just mozilla changes

Comment 7

[Wan-Teh Chang]

Comment on attachment 117262 [details] [diff] [review]
Just mozilla changes

r=wtc.

NSPR and LDAP C SDK should not be changed because they
don't include gtk header files.

Comment 8

[hacker formerly known as [e-mail]]

btw, the -std1 patch was checked in a few days ago.  I still need to work on the
-pthreads issue.

Comment 9

[hacker formerly known as [e-mail]]

So apparently removing the -std1 cause some sort of ripple effect in the build?
bug 201490 & maybe bug 201574.  I'm thinking that we might want to put it back.

Comment 10

[Jim Brown]

bug 201490 is definitely caused by the lack of -std1.
I don't think bug 201574 is caused by the absense of -std1.

Comment 11

[hacker formerly known as [e-mail]]

Mass reassign to new default build assignee

Comment 12

[(default assignee for unassigned bugs. does not receive bugmail)]

Mass reassign of Build/Config bugs to Leaf.

Comment 13

[Chase Phillips]

Mass reassign of open bugs for [e-mail] to [e-mail].

Comment 14

[J. Paul Reed [:preed]]

Resolving WONTFIX; this bug is pretty old.

If someone wants to revive the patch for 1.7.14, feel free to reopen and please
assign to yourself.