Closed Bug 227353 Opened 21 years ago Closed 21 years ago

GCC/Win32 build process bails on browser/app/module.rc:147 FILEVERSION

Categories

(Firefox Build System :: General, defect)

x86
Windows XP
defect
Not set
blocker

Tracking

(Not tracked)

VERIFIED FIXED

People

(Reporter: fredbezies, Assigned: bryner)

Details

Attachments

(1 file, 3 obsolete files)

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6b) Gecko/20031202 Firebird/0.7+ (Pentium 4 gcc 3.3.1 optimized - MozJF)
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.6b) Gecko/20031202 Firebird/0.7+ (Pentium 4 gcc 3.3.1 optimized - MozJF)

Simple. Keep your sources up to date, and try to build firebird with gcc 3.3.1
from mingw32. It was up-to-date at midnight mozilla.org time.

I've done twice make -f client.mk checkout to be sure not to be in a mid-checkin.

Reproducible: Always

Steps to Reproduce:
1.see details


Actual Results:  
Compilation is killed when creating firebird.exe !

Error is saying :

...\windres.exe : [path to sources]/browser/app/module.rc:147: syntax error

And then, build process kill itself :[



Expected Results:  
Making firebird exe.

Should be related to this ?

12/02/2003 23:10	ben%bengoodger.com 	mozilla/ browser/ app/ module.ver ?!


Yesterday, I built firebird using this .mozconfig without problem :|


CC=gcc
CXX=g++
CPP=cpp
AS=as
LD=ld

export MOZ_PHOENIX=1
mk_add_options MOZ_PHOENIX=1

ac_add_options --disable-ldap
ac_add_options --disable-mailnews
ac_add_options
--enable-extensions=cookie,xml-rpc,xmlextras,p3p,pref,transformiix,universalchardet,typeaheadfind,webservices,inspector
ac_add_options --enable-crypto
ac_add_options --disable-composer
ac_add_options --disable-profilesharing
ac_add_options --disable-installer

mk_add_options MOZ_OBJDIR=@TOPSRCDIR@/../fbbuild

ac_add_options --disable-accessibility
ac_add_options --disable-debug
ac_add_options --disable-tests

ac_add_options --enable-strip

ac_add_options --enable-optimize="-Os -march=pentium4 -msse2 -mfpmath=sse
-ffast-math -ffloat-store -fforce-addr -ftracer"
Additional infos : I thought my source was guilty. So I download (via CVS) a
brand new one. Same sad end :[
Flags: blocking0.8?
It also breaks when I use --enable-optimize="-Os -march=i686"

:[
Guilty line is saying :

FILEVERSION 0.7
Confirming on a setup more or less identical to Frederic's. Ben's recent
checkins to update version strings and such in Windows look suspect, I'll see if
I can't narrow it down.
windres expects the version string to be a quartet. 
Attached patch Make FILEVERSION a quartet (obsolete) — Splinter Review
Patch sets WIN32_MODULE_FILEVERSION=0.7.0.0 instead of 0.7.
Getting patch. Making CVS update, applying patch and trying to a build. Result
in about 2 or 3 hours :]
Frederic,

I think my initial patch is incorrect. If it doesn't work (and I don't think it
will), try it changing to the following:

WIN32_MODULE_FILEVERSION=0,7,0,0

(note commas instead of dots)
Ok. I will make a copy or your patch with "," instead of "." and I will tell
you. I am doing a checkout for now.
Attached patch Make FILEVERSION a quartet (obsolete) — Splinter Review
New patch with correct way of defining the quartet as per my previous comment
(commas instead of dots).
Attachment #136768 - Attachment is obsolete: true
It works !

Thanks a lot !
Attachment #136779 - Flags: review?(bugs)
Comment on attachment 136779 [details] [diff] [review]
Make FILEVERSION a quartet

r=ben@mozilla.org
Attachment #136779 - Flags: review?(bugs) → review+
Fix checked in at 12:38pm today.
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → FIXED
Hmm ... although a patch has been checked in, its not the same patch that Ben
r+'ed over here. Someone with a GCC build environment on Windows should probably
test with a clean tree and see if it works, and reopen the bug if necessary.
Am reopening. The patch that Ben checked in sets
WIN32_MODULE_PRODUCTVERSION=0,7,0,0 and leaves WIN32_MODULE_FILEVERSION=0.7
intact (whereas my patch that Frederic tested does not define
WIN32_MODULE_PRODUCTVERSION but sets WIN32_MODULE_FILEVERSION=0,7,0,0). Based on
that fact that people are still reporting that the build is failing on
forums.mozillazine.org with an error in module.rc:147 I'm reopening until
someone with a GCC build environment on Win32 can confirm that it really is
fixed. Sorry for bugspam.
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Summary: Build process with Gcc-win32 is killed near the end :[ → GCC/Win32 build process bails on browser/app/module.rc:147 FILEVERSION
Patch added sucks and kills build process in modules.rc:147 :|

Bad patch, change patch. Ali Ebrahim's one was working well.

Confirming that bug is still alive, so reopening was justified :[
This patch should fix the breakage as compared to the current state of CVS.
Ben's checkin appears to be incorrect, and this should fix that by removing
WIN32_MODULE_PRODUCTVERSION and correcting WIN32_MODULE_FILEVERSION.

Frederic can you confirm that this patch works on a freshly pulled clean CVS
tree?
Attachment #136779 - Attachment is obsolete: true
Replacing to remove some cruft that I had at the start of the patchfile. Sorry
for bugspam, I just woke up. :)
Attachment #136865 - Attachment is obsolete: true
I am making a fresh CVS pull. And I will apply your patch.

I will tell you if building go to its end.
This patch works (attachment 136867 [details] [diff] [review]). Please, check it in :)
Comment on attachment 136867 [details] [diff] [review]
New patch to fix new breakage. Remove PRODUCTVERSION and correct FILEVERSION.

Requesting review for attachment #136867 [details] [diff] [review].
Attachment #136867 - Flags: review?(bugs)
Comment on attachment 136867 [details] [diff] [review]
New patch to fix new breakage. Remove PRODUCTVERSION and correct FILEVERSION.

r=ben@mozilla.org
Attachment #136867 - Flags: review?(bugs) → review+
checked in
Status: REOPENED → RESOLVED
Closed: 21 years ago21 years ago
Resolution: --- → FIXED
Fix verified working on MSVC++ .NET 2003. File version is now correctly showing
0.7.0.0 instead of 0.0.0.0. It would be good if someone could also confirm on
GCC/MinGW.
Flags: blocking0.8?
Confirming that it works perfectly. And properties are Ok :)
V
Status: RESOLVED → VERIFIED
Component: Build Config → General
Product: Firefox → Firefox Build System
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: