235207 - Build failure in mozilla/xpcom/string/src/nsStringObsolete.cpp

Status: RESOLVED FIXED

(Core :: XPCOM, defect)

Description

[Roland Mainz]

CVS pull from "Sun Feb 22 00:51:52 PST 2004" bulid with Sun Workshop 8 on
Solaris 2.8/SPARC, the build fails like this:
-- snip --
nsStringObsolete.cpp
Building deps for ../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp
/opt/SUNWspro/bin/CC -o nsStringObsolete.o -c -DOSTYPE=\"SunOS5\"
-DOSARCH=\"SunOS\" -D_IMPL_NS_COM  -I../../../dist/include/xpcom
-I../../../dist/include/string -I../../../dist/include
-I/shared/bigtmp/mozilla/nightlybuilds/north/objdir_gtk1/dist/include/nspr    
-I/usr/openwin/include   -KPIC  -I/usr/openwin/include -xbuiltin=%all -mt 
-DNDEBUG -DTRIMMED -xO2  -I/usr/openwin/include -DSOLARIS=1
-DNSCAP_DISABLE_DEBUG_PTR_TYPES=1 -DD_INO=d_ino -DSTDC_HEADERS=1
-DHAVE_ST_BLKSIZE=1 -DHAVE_SIGINFO_T=1 -DHAVE_INT16_T=1 -DHAVE_INT32_T=1
-DHAVE_INT64_T=1 -DHAVE_UINT=1 -DHAVE_UINT_T=1 -DHAVE_UINT16_T=1
-DHAVE_DIRENT_H=1 -DHAVE_SYS_BYTEORDER_H=1 -DHAVE_MEMORY_H=1 -DHAVE_UNISTD_H=1
-DHAVE_NL_TYPES_H=1 -DHAVE_MALLOC_H=1 -DHAVE_X11_XKBLIB_H=1
-DHAVE_SYS_STATVFS_H=1 -DHAVE_SYS_STATFS_H=1 -DHAVE_LIBM=1 -DHAVE_LIBDL=1
-DHAVE_LIBSOCKET=1 -DFUNCPROTO=15 -DHAVE_XSHM=1 -D_REENTRANT=1 -DHAVE_RANDOM=1
-DHAVE_STRERROR=1 -DHAVE_LCHOWN=1 -DHAVE_FCHMOD=1 -DHAVE_SNPRINTF=1
-DHAVE_MEMMOVE=1 -DHAVE_RINT=1 -DHAVE_NL_LANGINFO=1 -DHAVE_FLOCKFILE=1
-DHAVE_LOCALTIME_R=1 -DHAVE_STRTOK_R=1 -DVA_COPY=va_copy -DHAVE_VA_COPY=1
-DHAVE_I18N_LC_MESSAGES=1 -DMOZ_DEFAULT_TOOLKIT=\"gtk\" -DMOZ_WIDGET_GTK=1
-DMOZ_ENABLE_XREMOTE=1 -DMOZ_X11=1 -DMOZ_ENABLE_COREXFONTS=1
-DMOZ_EXTRA_X11CONVERTERS=1 -DOJI=1 -DIBMBIDI=1 -DMOZ_VIEW_SOURCE=1
-DACCESSIBILITY=1 -DMOZ_XPINSTALL=1 -DMOZ_JSLOADER=1 -DMOZ_MATHML=1 -DMOZ_SVG=1
-DMOZ_SVG_RENDERER_LIBART=1 -DMOZ_LOGGING=1 -DMOZ_USER_DIR=\".mozilla\"
-DMOZ_XUL=1 -DMOZ_PROFILESHARING=1 -DMOZ_PROFILELOCKING=1 -DSUNCTL=1
-DMOZ_DLL_SUFFIX=\".so\" -DXP_UNIX=1 -DUNIX_ASYNC_DNS=1 -DJS_THREADSAFE=1
-DNS_PRINT_PREVIEW=1 -DNS_PRINTING=1 -DMOZILLA_VERSION=\"1.7b\"
-DMOZILLA_LOCALE_VERSION=\"1.7b\" -DMOZILLA_REGION_VERSION=\"1.7b\" 
-D_MOZILLA_CONFIG_H_ -DMOZILLA_CLIENT
../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp
"../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp", line 568: Warning
(Anachronism): "static" is not allowed and is being ignored.
"../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp", line 684: Warning
(Anachronism): "static" is not allowed and is being ignored.
"../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp", line 703: Warning
(Anachronism): "static" is not allowed and is being ignored.
"../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp", line 724: Warning
(Anachronism): "static" is not allowed and is being ignored.
"../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp", line 751: Warning
(Anachronism): "static" is not allowed and is being ignored.
"../../../../mozilla/xpcom/string/src/nsStringObsolete.cpp", line 642: Error:
The name Compare2To1(const unsigned short*, const char*, unsigned, int) is
unusable in static nsBufferRoutines<unsigned short>::compare(const unsigned
short*, const char*, unsigned, int).
1 Error(s) and 5 Warning(s) detected.
gmake[4]: *** [nsStringObsolete.o] Error 1
gmake[4]: Leaving directory
`/shared/bigtmp/mozilla/nightlybuilds/north/objdir_gtk1/xpcom/string/src'
gmake[3]: *** [libs] Error 2
gmake[3]: Leaving directory
`/shared/bigtmp/mozilla/nightlybuilds/north/objdir_gtk1/xpcom/string'
gmake[2]: *** [libs] Error 2
gmake[2]: Leaving directory
`/shared/bigtmp/mozilla/nightlybuilds/north/objdir_gtk1/xpcom'
gmake[1]: *** [tier_2] Error 2
gmake[1]: Leaving directory
`/shared/bigtmp/mozilla/nightlybuilds/north/objdir_gtk1'
gmake: *** [default] Error 2
-- snip --

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Why is using Compare2To2 OK but using Compare2To1 an error?

Comment 2

[Roland Mainz]

David Baron wrote:
> Why is using Compare2To2 OK but using Compare2To1 an error

Compare2To2(), Compare2To1() and Compare1To1() return an error, the compiler
just thinks one of these errors is enougth to report... :-/

Comment 3

[Roland Mainz]

Attachment: Patch for 2004-02-22-trunk to fix the problem

The patch makes the Compare[12]To[12]() functions |inline| to get rid of the
compiler internal confusion about scope and kills the anachronism problems,
too.

Comment 4

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

What's the minimum patch needed to fix the bustage?  All of these changes are
likely to cause changes we don't want.

Comment 5

[Roland Mainz]

David Baron wrote:
> What's the minimum patch needed to fix the bustage?

Erm... the patch is the minimum patch. Unfortunately we have also bustage
elsehwere:
-- snip --
nsLocalFileCommon.cpp
Building deps for ../../../mozilla/xpcom/io/nsLocalFileCommon.cpp
/opt/SUNWspro/bin/CC -o nsLocalFileCommon.o -c -DOSTYPE=\"SunOS5\"
-DOSARCH=\"SunOS\" -D_IMPL_NS_COM -I.. -I../../dist/include/string
-I../../dist/include/xpcom -I../../dist/include
-I/shared/bigtmp/mozilla/nightlybuilds/north/objdir_gtk1/dist/include/nspr    
-I/usr/openwin/include   -KPIC  -I/usr/openwin/include -xbuiltin=%all -mt 
-DNDEBUG -DTRIMMED -xO2  -I/usr/openwin/include -DSOLARIS=1
-DNSCAP_DISABLE_DEBUG_PTR_TYPES=1 -DD_INO=d_ino -DSTDC_HEADERS=1
-DHAVE_ST_BLKSIZE=1 -DHAVE_SIGINFO_T=1 -DHAVE_INT16_T=1 -DHAVE_INT32_T=1
-DHAVE_INT64_T=1 -DHAVE_UINT=1 -DHAVE_UINT_T=1 -DHAVE_UINT16_T=1
-DHAVE_DIRENT_H=1 -DHAVE_SYS_BYTEORDER_H=1 -DHAVE_MEMORY_H=1 -DHAVE_UNISTD_H=1
-DHAVE_NL_TYPES_H=1 -DHAVE_MALLOC_H=1 -DHAVE_X11_XKBLIB_H=1
-DHAVE_SYS_STATVFS_H=1 -DHAVE_SYS_STATFS_H=1 -DHAVE_LIBM=1 -DHAVE_LIBDL=1
-DHAVE_LIBSOCKET=1 -DFUNCPROTO=15 -DHAVE_XSHM=1 -D_REENTRANT=1 -DHAVE_RANDOM=1
-DHAVE_STRERROR=1 -DHAVE_LCHOWN=1 -DHAVE_FCHMOD=1 -DHAVE_SNPRINTF=1
-DHAVE_MEMMOVE=1 -DHAVE_RINT=1 -DHAVE_NL_LANGINFO=1 -DHAVE_FLOCKFILE=1
-DHAVE_LOCALTIME_R=1 -DHAVE_STRTOK_R=1 -DVA_COPY=va_copy -DHAVE_VA_COPY=1
-DHAVE_I18N_LC_MESSAGES=1 -DMOZ_DEFAULT_TOOLKIT=\"gtk\" -DMOZ_WIDGET_GTK=1
-DMOZ_ENABLE_XREMOTE=1 -DMOZ_X11=1 -DMOZ_ENABLE_COREXFONTS=1
-DMOZ_EXTRA_X11CONVERTERS=1 -DOJI=1 -DIBMBIDI=1 -DMOZ_VIEW_SOURCE=1
-DACCESSIBILITY=1 -DMOZ_XPINSTALL=1 -DMOZ_JSLOADER=1 -DMOZ_MATHML=1 -DMOZ_SVG=1
-DMOZ_SVG_RENDERER_LIBART=1 -DMOZ_LOGGING=1 -DMOZ_USER_DIR=\".mozilla\"
-DMOZ_XUL=1 -DMOZ_PROFILESHARING=1 -DMOZ_PROFILELOCKING=1 -DSUNCTL=1
-DMOZ_DLL_SUFFIX=\".so\" -DXP_UNIX=1 -DUNIX_ASYNC_DNS=1 -DJS_THREADSAFE=1
-DNS_PRINT_PREVIEW=1 -DNS_PRINTING=1 -DMOZILLA_VERSION=\"1.7b\"
-DMOZILLA_LOCALE_VERSION=\"1.7b\" -DMOZILLA_REGION_VERSION=\"1.7b\" 
-D_MOZILLA_CONFIG_H_ -DMOZILLA_CLIENT
../../../mozilla/xpcom/io/nsLocalFileCommon.cpp
"../../../mozilla/xpcom/io/nsLocalFileCommon.cpp", line 110: Error:
nsACString::nsACString(const nsACString&) is not accessible from
nsLocalFile::CreateUnique(unsigned, unsigned).
1 Error(s) detected.
gmake[3]: *** [nsLocalFileCommon.o] Error 1
-- snip --
I wish there would still be a Tinderbox with Sun Workshop around...

Comment 6

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

The anachronism messages were warnings, not errors.

Comment 7

[Roland Mainz]

dbaron:
Thank you for not caring about the fact that the code doesn't compile without
that.

Comment 8

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

If the compiler issues a warning that it's ignoring something, then one wouldn't
expect removing the thing that it's *ignoring* to fix an error.

Did you even try it with only the other changes?

Comment 9

[Roland Mainz]

David Baron wrote:
> If the compiler issues a warning that it's ignoring something, then one 
> wouldn't expect removing the thing that it's *ignoring* to fix an error.
>
> Did you even try it with only the other changes?

Yes (CC: Sun C++ 5.5 Patch 113817-03 2003/10/14, SunOS meridian 5.8
Generic_108528-27 sun4u sparc SUNW,Ultra-1, 256MB). Without that the |inline|
change is a NO-OP (don't ask me why, I don't have the 5.5 compiler sources). 
Just guess why I added the change. I didn't do that for fun or just to annoy
you. *Grumble*

Comment 10

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I'd find this patch acceptable if all the changes were appropriately ifdef-ed
for the compiler in question.

Comment 11

[Roland Mainz]

Attachment: patch

Comment 12

[Christian :Biesinger (don't email me, ping me on IRC)]

reopening, marking this bug invalid like that is just silly

Comment 13

[timeless]

SunOS DEATHSTARII 5.9 Generic_112233-11 sun4u sparc SUNW,Ultra-80
CC: Forte Developer 7 C++ 5.4 Patch 111715-05 2003/02/09
is also affected (i pulled, started a build, and then committed the fix, i'll
update the build when cvs-mirror picks it up)

Comment 14

[timeless]

checked in
bug 235273 covers the next build failure