Closed Bug 893045 Opened 11 years ago Closed 11 years ago

ERROR PROCESSING MOZBUILD FILE: HOST_CSRCS should not be used.

Categories

(Firefox Build System :: General, defect)

x86
Linux
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED WORKSFORME

People

(Reporter: ishikawa, Unassigned)

References

(Blocks 1 open bug)

Details

Hi,

After fetching fresh copy of comm-central (and refreshing it
by "python client.py checkout"), I got the following error during
cofiguration.

There is one caveat. I had to add
the following to my MOZCONFIG file.

#
# Debian GNU/Linux does not have .15 version of gstreamer.
# It is still at .10 (as of July 12, 2013)
ac_add_options --disable-gstreame

(See someone who had to add the same line to
MOZCONFIG. Bug 892473 )



It seems that we should not use
    HOST_CSRCS



  [...]
checking for GL/glx.h... (cached) yes
creating ./config.status
Reticulating splines...
Traceback (most recent call last):
  File "./config.status", line 971, in <module>
    config_status(**args)
  File "/COMM-CENTRAL/comm-central/mozilla/build/ConfigStatus.py", line 122, in config_status
    summary = backend.consume(definitions)
  File "/COMM-CENTRAL/comm-central/mozilla/python/mozbuild/mozbuild/backend/base.py", line 176, in consume
    for obj in objs:
  File "/COMM-CENTRAL/comm-central/mozilla/python/mozbuild/mozbuild/frontend/emitter.py", line 42, in emit
    for out in output:
  File "/COMM-CENTRAL/comm-central/mozilla/python/mozbuild/mozbuild/frontend/reader.py", line 601, in read_mozbuild
    raise bre
mozbuild.frontend.reader.BuildReaderError: ==============================
ERROR PROCESSING MOZBUILD FILE
==============================

The error occurred while processing the following file:

    /REF-COMM-CENTRAL/comm-central/mozilla/config/moz.build

The error was triggered on line 17 of this file:

    HOST_CSRCS += [

The underlying problem is an attempt to read a reserved UPPERCASE variable that does not exist.

The variable read causing the error is:

    HOST_CSRCS

Please change the file to not use this variable.

For reference, the set of valid variables is:

ASFILES, CONFIGURE_SUBST_FILES, CPP_SOURCES, CSRCS, DEFINES, DIRS, EXPORTS, EXTERNAL_MAKE_DIRS, MODULE, PARALLEL_DIRS, PARALLEL_EXTERNAL_MAKE_DIRS, PROGRAM, SIMPLE_PROGRAMS, TEST_DIRS, TEST_TOOL_DIRS, TIERS, TOOL_DIRS, XPCSHELL_TESTS_MANIFESTS, XPIDL_FLAGS, XPIDL_MODULE, XPIDL_SOURCES

------ config.log ------
configure:25524: /usr/bin/ccache gcc  -DDEBUG=1 -DDEBUG_4GB_CHECK -DUSEHELGRIND=1 -o conftest.o -c  -std=gnu99 -fgnu89-inline -fno-strict-aliasing -ffunction-sections -fdata-sections -pthread -pipe  conftest.c 1>&5
configure:25526: /usr/bin/ccache gcc  -DDEBUG=1 -DDEBUG_4GB_CHECK -DUSEHELGRIND=1 -o conftest -lpthread  -Wl,-z,noexecstack -Wl,--build-id -Wl,--icf=safe conftest.list -ldl  1>&5
configure:25552: checking what kind of ordering can be done with the linker
configure:25567: /usr/bin/ccache gcc  -DDEBUG=1 -DDEBUG_4GB_CHECK -DUSEHELGRIND=1 -o conftest  -std=gnu99 -fgnu89-inline -fno-strict-aliasing -ffunction-sections -fdata-sections -pthread -pipe  -lpthread  -Wl,-z,noexecstack -Wl,--build-id -Wl,--icf=safe -Wl,--section-ordering-file,conftest.order conftest.c -ldl  1>&5
configure:25713: checking for glib-2.0 >= 1.3.7 gobject-2.0
configure:25720: checking GLIB_CFLAGS
configure:25725: checking GLIB_LIBS
configure:25841: checking for freetype2 >= 6.1.0
configure:25848: checking FT2_CFLAGS
configure:25853: checking FT2_LIBS
configure:25890: checking for FT_Bitmap_Size.y_ppem
configure:25934: checking for FT_GlyphSlot_Embolden
configure:25934: checking for FT_Load_Sfnt_Table
configure:25934: checking for FT_Select_Size
configure:26000: checking for fontconfig/fcfreetype.h
configure:26103: checking for fontconfig
configure:26110: checking _FONTCONFIG_CFLAGS
configure:26115: checking _FONTCONFIG_LIBS
configure:26309: checking for stdint.h
configure:26309: checking for inttypes.h
configure:26309: checking for sys/int_types.h
configure:27666: checking for posix_fallocate
configure:27676: /usr/bin/ccache gcc  -DDEBUG=1 -DDEBUG_4GB_CHECK -DUSEHELGRIND=1 -o conftest -Wall -Wpointer-arith -Wdeclaration-after-statement -Werror=return-type -Wtype-limits -Wempty-body -Wsign-compare -Wno-unused -Wcast-align -std=gnu99 -fgnu89-inline -fno-strict-aliasing -ffunction-sections -fdata-sections -pthread -pipe  -lpthread  -Wl,-z,noexecstack -Wl,--build-id -Wl,--icf=safe conftest.c -ldl  1>&5
configure:27727: /usr/bin/ccache gcc  -DDEBUG=1 -DDEBUG_4GB_CHECK -DUSEHELGRIND=1 -c -Wall -Wpointer-arith -Wdeclaration-after-statement -Werror=return-type -Wtype-limits -Wempty-body -Wsign-compare -Wno-unused -Wcast-align -std=gnu99 -fgnu89-inline -fno-strict-aliasing -ffunction-sections -fdata-sections -pthread -pipe   conftest.c 1>&5
configure:27747: checking for GL/glx.h
configure: error: /REF-COMM-CENTRAL/comm-central/mozilla/configure failed for mozilla
*** Fix above errors and then restart with               "make -f client.mk build"
make: *** [configure] Error 1



(I understand that m-c under ./mozilla directory
of comm-central may not be in sync with the real M-C either
due to timing delay or intentionally(?), so I am filing this as thunderbird bug, but please feel free to re-classify this to core if it is where this bug entry should be. )
Blocks: nomakefiles
This sounds like you have a bad checkout of mozilla-central, since HOST_CSRCS has been ported to moz.build for quite some time, and comm-central has no uses of that variable. What changeset of m-c are you on, and do you have any local patches?
Product: Thunderbird → Core
HOST_CSRCS is defined in the moz.build sandbox, as jcranmer pointed out in comment #1. I suspect your source checkout is out of sync or a clobber is required.
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → WORKSFORME
(In reply to Gregory Szorc [:gps] from comment #2)
> HOST_CSRCS is defined in the moz.build sandbox, as jcranmer pointed out in
> comment #1. I suspect your source checkout is out of sync or a clobber is
> required.

I think a clobber was required and it seemed clobbering fixed the issue.
(But I have no idea why the checked out source didn't suggest this, but come to think of it, a freshly checked out source tree had no notion of the previous object directory, etc.)

Thank you!
Product: Core → Firefox Build System
You need to log in before you can comment on or make changes to this bug.