213541 - Assumes Darwin = PPC

Status: RESOLVED FIXED

(SeaMonkey :: Build Config, defect)

Description

[Ed Waldmire]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.4b) Gecko/20030516 Mozilla Firebird/0.6
Build Identifier: 

When trying to build mozilla on Darwin/x86 the build stops because there is code
that associates Darwin with a powerpc processor.

Reproducible: Always

Steps to Reproduce:

Comment 1

[cls]

More information is required. Which code is assuming ppc?  Note that building
against the mac or cocoa toolkits isn't supported for non-ppc builds.  Afaik,
you can only build against the x11 toolkits.

Comment 2

[cls]

Marking invalid for lack of additional requested information.

Comment 3

[Ed Waldmire]

from the file:

../mozilla/xpcom/reflect/xptcall/src/md/unix/Makefile

**********************************************
ifeq ($(OS_ARCH),Darwin)
ASFLAGS         += -x assembler-with-cpp   # assumes $(AS) == $(CC)
ifdef HAVE_GCC3_ABI
ASFLAGS         += -DHAVE_GCC3_ABI
endif
CPPSRCS         := xptcinvoke_ppc_rhapsody.cpp xptcstubs_ppc_rhapsody.cpp
ASFILES         := xptcinvoke_asm_ppc_rhapsody.s xptcstubs_asm_ppc_darwin.s
endif
**********************************************

Comment 4

[cls]

Attachment: v1.0

Make darwin/x86 build using the generic unixish_x86 sources and put a cpu test
around the ppc case.

Comment 5

[cls]

Comment on attachment 129769 [details] [diff] [review]
v1.0

xptcall makefile changes to properly detect darwin/x86. Low risk.

Comment 6

[Asa Dotzler [:asa]]

Comment on attachment 129769 [details] [diff] [review]
v1.0

a=asa (on behalf of drivers) for checkin to 1.5beta

Comment 7

[cls]

The patch has been checked in.  Shall we wait to see what else turns up?

Comment 8

[Ed Waldmire]

The patch causes the following build error: 
 
c++ -o xptcall.o -c -DOSTYPE=\"Darwin6.6\" -DOSARCH=\"Darwin\" 
-DEXPORT_XPTC_API -D_IMPL_NS_COM -D_IMPL_NS_BASE   
-I../../../../dist/include/xpcom -I../../../../dist/include 
-I/Users/waldmire/mozilla/dist/include/nspr     -I/usr/X11R6/include   -fPIC  
-I/usr/X11R6/include -fno-rtti -fno-exceptions -Wall -Wconversion -Wpointer-arith 
-Wcast-align -Woverloaded-virtual -Wsynth -Wno-ctor-dtor-privacy -Wno-long-long 
-fpascal-strings -no-cpp-precomp -fno-common -fshort-wchar -pipe  -DDEBUG 
-D_DEBUG -DDEBUG_waldmire -DTRACING -g  -I/usr/X11R6/include 
-DMOZILLA_CLIENT -include ../../../../mozilla-config.h -Wp,-MD,.deps/xptcall.pp 
xptcall.cpp 
rm -f libxptcall.a 
ar cr libxptcall.a xptcall.o 
ranlib libxptcall.a 
../../../../config/nsinstall -L /Users/waldmire/mozilla/xpcom/reflect/xptcall/src -m 644 
libxptcall.a ../../../../dist/lib 
xptcinvoke_unixish_x86.cpp 
c++ -o xptcinvoke_unixish_x86.o -c -DOSTYPE=\"Darwin6.6\" 
-DOSARCH=\"Darwin\" -DEXPORT_XPTC_API   -I../../../../../../dist/include/xpcom 
-I../../../../../../dist/include -I/Users/waldmire/mozilla/dist/include/nspr     -I./../.. 
-I/usr/X11R6/include   -fPIC  -I/usr/X11R6/include -fno-rtti -fno-exceptions -Wall 
-Wconversion -Wpointer-arith -Wcast-align -Woverloaded-virtual -Wsynth 
-Wno-ctor-dtor-privacy -Wno-long-long -fpascal-strings -no-cpp-precomp 
-fno-common -fshort-wchar -pipe  -DDEBUG -D_DEBUG -DDEBUG_waldmire 
-DTRACING -g  -I/usr/X11R6/include -DMOZILLA_CLIENT -include 
../../../../../../mozilla-config.h -Wp,-MD,.deps/xptcinvoke_unixish_x86.pp 
xptcinvoke_unixish_x86.cpp 
In file included from xptcinvoke_unixish_x86.cpp:41: 
xptc_platforms_unixish_x86.h:147:2: #error "need a platform define if using unixish 
x86 code" 
xptc_platforms_unixish_x86.h:153:2: #error "need to define 'this' adjust scheme" 
make[7]: *** [xptcinvoke_unixish_x86.o] Error 1 
make[6]: *** [libs] Error 2 
make[5]: *** [libs] Error 2 
make[4]: *** [libs] Error 2 
make[3]: *** [libs] Error 2 
make[2]: *** [libs] Error 2 
make[1]: *** [tier_2] Error 2 
make: *** [default] Error 2 
 

Comment 9

[cls]

Attachment: v2.0

Try this patch.  You may have to change __APPLE__ to the appropriate define as
I'm not sure what are standard defines for gcc on Darwin/x86.

Comment 10

[Ed Waldmire]

the define should be as follows:

#elif defined (__APPLE__) && (__MACH__)

Comment 11

[Ed Waldmire]

I testing the patches, and the build gets past this point with both patches (the
second patch needs changed to reflect comment #10.  I haven't yet determined the
problem with the next build error, but it is in a different directory.

Comment 12

[Ed Waldmire]

If we get the second patch committed to CVS (with the change noted in comment #10) then we 
could probably mark this bug as fixed.  I haven't gotten throught the entire build yet, but so far I 
haven't found any other code that is assuming Darwin means PPC.

Comment 13

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

Mass reassign of Build/Config bugs to Leaf.

Comment 14

[Simon Paquet [:sipaq]]

Comment on attachment 130010 [details] [diff] [review]
v2.0

Leaf can you please review this together with the change proposed in comment 10
and check this in if applicable? Thanks.

Comment 15

[Boris 'pi' Piwinger]

There was an approval from drivers, but the bug is marked UNCONFIRMED? I don't
get that. Is it correct to assume that there is still a problem in 1.5 or later?

pi

Comment 16

[Simon Paquet [:sipaq]]

Attachment: v2.1

Adjusting cls's patch to he stuff said in comment 10.

Comment 17

[Simon Paquet [:sipaq]]

Comment on attachment 134979 [details] [diff] [review]
v2.1

Brian, can you please review and check this in if you like this patch?

Comment 18

[Simon Paquet [:sipaq]]

I think we can confirm this.

Comment 19

[Simon Paquet [:sipaq]]

Comment on attachment 130010 [details] [diff] [review]
v2.0

Clearing review-request of an obsolete patch

Comment 20

[Ed Waldmire]

I tested Mozilla 1.5 with patch v2.1 with the newest version of Darwin (7.0) and it builds with no 
additional changes needed.  Mozilla crashes immediately, but the viewer runs.  I think it would be 
safe to commit patch v2.1.

Comment 21

[Simon Paquet [:sipaq]]

Boris Zbarsky checked this in at 10.17 AM pacific time.