Closed Bug 842920 Opened 7 years ago Closed 7 years ago
Thunderbird Windows builds are broken due to regression in disable-web
+++ This bug was initially created as a clone of Bug #842918 +++ Windows building with --disable-webRTC breaks with the following errors: WMFUtils.obj : error LNK2001: unresolved external symbol _CLSID_CMP3DecMediaObject WMFUtils.obj : error LNK2001: unresolved external symbol _CLSID_CMSH264DecoderMFT xul.dll : fatal error LNK1120: 2 unresolved externals
Pushed to comm-central: https://hg.mozilla.org/comm-central/rev/3225c78eb19b Sorry, forgot to push the Suite/ patch to mail/. This should fix the Windows builds.
To be honest, I'm not keen on this solution, as it means the functionality is available on different platforms. I'd rather get the core issue fixed as to why --disable-webrtc isn't linking correctly. Although it looks like even with this patch it isn't linking correctly.
An additional patch in Bug 839031 Comment 22 might help: "We need link wmcodecdspuuid even if no MOZ_WEBRTC_IN_LIBXUL."
https://tbpl.mozilla.org/php/getParsedLog.php?id=19902340&tree=Thunderbird-Trunk#error0 vp8_impl.obj : error LNK2019: unresolved external symbol _vpx_codec_vp8_cx referenced in function "private: int __thiscall webrtc::VP8EncoderImpl::InitAndSetControlSettings(struct webrtc::VideoCodec const *)" (?InitAndSetControlSettings@VP8EncoderImpl@webrtc@@AAEHPBUVideoCodec@2@@Z) xul.dll : fatal error LNK1120: 1 unresolved externals So in m-c http://mxr.mozilla.org/mozilla-central/source/configure.in#5233 We have: 5233 if test -n "$MOZ_WEBRTC"; then 5234 AC_DEFINE(MOZ_WEBRTC) 5235 MOZ_MEDIA=1 5236 MOZ_RAW=1 5237 MOZ_VP8=1 5238 MOZ_VP8_ENCODER=1 5239 MOZ_VP8_ERROR_CONCEALMENT=1 And then in http://mxr.mozilla.org/mozilla-central/source/layout/media/symbols.def.in#42 We have: 42 #ifdef MOZ_VP8_ENCODER 43 vpx_codec_vp8_cx 44 #endif After this I'm lost.
(In reply to Mark Banner (:standard8) from comment #3) > To be honest, I'm not keen on this solution, as it means the functionality > is available on different platforms. I'd rather get the core issue fixed as > to why --disable-webrtc isn't linking correctly. > > Although it looks like even with this patch it isn't linking correctly. If you like, I can backout my patches (both on trunk and aurora..) Aurora (atm) is WinOpt green.. but anyway.. I was hoping it'd be ok. Just didn't realize that on top of this bug, there was something else that didn't like linking. But ignorance isn't an excuse. ;/
This breaks Windows build of browser for me too. Worth a separate bug?
(In reply to Honza Bambas (:mayhemer) from comment #7) > This breaks Windows build of browser for me too. Worth a separate bug? If you're talking about Philip's addendum, it'd be Bug 842935.
Please file a new bug on the webrtc disabling issues. Its a core issue and should be fixed anyway.
(In reply to Mark Banner (:standard8) from comment #9) > Please file a new bug on the webrtc disabling issues. Its a core issue and > should be fixed anyway. Should I backout my mail/ patch? it's more than likely going to cause TB to burn on Windows.. with probably the same results as https://bugzilla.mozilla.org/show_bug.cgi?id=842918#c6. I backed out the suite/ one and that's exactly what happened. Or should I just wait until the webrtc disabling issue is fixed before backing it out?
I pushed the backout to try: https://tbpl.mozilla.org/?tree=Thunderbird-Try&rev=dee9366f8b1c If someone does backout that patch, it needs a clobber.
Status: ASSIGNED → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
I backed this out earlier today: https://hg.mozilla.org/comm-central/rev/13af57b245bc I'm currently also investigating about what to do on aurora.
Also backed out from Aurora: https://hg.mozilla.org/releases/comm-aurora/rev/64b92b660066
You need to log in before you can comment on or make changes to this bug.