mesa: 5039ad6bc5cec5c49f01380aa270248aae77811e (master) $ ./autogen.sh --with-dri-drivers= --with-gallium-drivers=swrast [...] $ make [...] CC dri_screen.lo In file included from dri_screen.c:33:0: ../../../../../src/mesa/drivers/dri/common/xmlpool.h:96:29: fatal error: xmlpool/options.h: No such file or directory b587a7595e7aae0809a8e7392e89c39df0c955b1 is the first bad commit commit b587a7595e7aae0809a8e7392e89c39df0c955b1 Author: Carl Worth <cworth@cworth.org> Date: Mon Dec 3 15:43:19 2012 -0800 driconf: Add translation-generation to build system, don't track generated files Previously, the xmlpool directory had a lone Makefile to assist poeple in manually invoking a deep make in order to update the translations in options.h. We can observe that this wasn't happening in fact, (new translations had been added to de.po without being generated into options.h, and new options had been manually added directly to options.h rather than to t_options.h). Prevent both of these problems from occurring in the future by automatically generating options.h as part of the standard build of mesa. For this, the generated options.h is now removed from version control, (along with Makefile in favor of Makefile.am). [chadv: Port the Autotools changes to Android.] Signed-off-by: Chad Versace <chad.versace@linux.intel.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> :100644 100644 1b2083df5e930948293e4ebc276ad6f8f168c049 72df83c0cd640d21df506a969e991627d08725b7 M configure.ac :040000 040000 ba43e3327c3a7bd75de5497eaa6ee065aaaaba88 c56772871d8761f3416a193a19cae9f263403e84 M src bisect run success
Either src/mesa/drivers/dri/common/xmlpool/options.h was removed by mistake or a reference to it was not removed as supposed.
Identified problem: src/mesa/drivers/dri/common/xmlpool.h is not built before being needed. Manually building it lets finish the build process.
In src/mesa/drivers/dri/Makefile.am, if HAVE_COMMON_DRI defines if it should go further in the "common" folder. It seems HAVE_COMMON_DRI is not defined, which explains why it's not building what's in that folder.
This could be related to an autogen.sh option: I'm using "--with-dri-drivers= " which sets any dri drivers off. However, while this is OK to do and it makes sense to disable HAVE_COMMON_DRI, it seems dri state_tracker needs xmlpoo.h. In other words, HAVE_COMMON_DRI should also be set as TRUE if the dri state_tracker is to be built.
Created attachment 72887 [details] [review] Fixes missing xmlpool.h Please review and commit if OK. Fixes missing xmlpool.h when using dri state tracker.
Build works with patch, but I can't load the driver afterwards: libGL: OpenDriver: trying /usr/lib/xorg/modules/dri/tls/r600_dri.so libGL: OpenDriver: trying /usr/lib/xorg/modules/dri/r600_dri.so libGL error: dlopen /usr/lib/xorg/modules/dri/r600_dri.so failed (/usr/lib/xorg/modules/dri/r600_dri.so: undefined symbol: _ZTVN10__cxxabiv120__si_class_type_infoE) libGL error: unable to load driver: r600_dri.so libGL error: driver pointer missing libGL error: failed to load driver: r600 I don't know if this is related to the patch.
(In reply to comment #6) > Build works with patch, but I can't load the driver afterwards: > > libGL: OpenDriver: trying /usr/lib/xorg/modules/dri/tls/r600_dri.so > libGL: OpenDriver: trying /usr/lib/xorg/modules/dri/r600_dri.so > libGL error: dlopen /usr/lib/xorg/modules/dri/r600_dri.so failed > (/usr/lib/xorg/modules/dri/r600_dri.so: undefined symbol: > _ZTVN10__cxxabiv120__si_class_type_infoE) > libGL error: unable to load driver: r600_dri.so > libGL error: driver pointer missing > libGL error: failed to load driver: r600 > > I don't know if this is related to the patch. Pretty sure it is not related since it fixes only the fact xmlpool.h was previously always present before being modified by culprit commit.
(In reply to comment #7) > (In reply to comment #6) > > Build works with patch, but I can't load the driver afterwards: > > > > libGL: OpenDriver: trying /usr/lib/xorg/modules/dri/tls/r600_dri.so > > libGL: OpenDriver: trying /usr/lib/xorg/modules/dri/r600_dri.so > > libGL error: dlopen /usr/lib/xorg/modules/dri/r600_dri.so failed > > (/usr/lib/xorg/modules/dri/r600_dri.so: undefined symbol: > > _ZTVN10__cxxabiv120__si_class_type_infoE) > > libGL error: unable to load driver: r600_dri.so > > libGL error: driver pointer missing > > libGL error: failed to load driver: r600 > > > > I don't know if this is related to the patch. > > Pretty sure it is not related since it fixes only the fact xmlpool.h was > previously always present before being modified by culprit commit. After searching a bit, you are hitting bug 59282 which is unrelated to the current bug/bad commit and patch. Just in case, make a clean build of mesa. Also, you could update the ddx driver, I think I saw something that could be related being pushed yesterday (not sure though). That's what I did and it works OK over here.
Thanks! Fixed by 67ef755908375661c2cc1e7a26922682b981dc97
THank you, Alexandre Demers. Quick and effective fix.
re-broken again by 998d975e38db7478781130b42a4fb68d6b181055
(In reply to comment #11) > re-broken again by 998d975e38db7478781130b42a4fb68d6b181055 Hi Mark, I'd like to help with this bug, but "re-broken again" doesn't give me enough information to even start looking. For example, for me the build is working fine. So I would need answers to the same questions as if you were submitting a new What failure mode are you hitting? What error message are you getting? What steps are needed to replicate the failure, (command-line options to configure, etc.). Thanks. -Carl
I get the same error building mesa as of today: CC dri2.lo In file included from dri_screen.c:33:0: ../../../../../src/mesa/drivers/dri/common/xmlpool.h:96:29: fatal error: xmlpool/options.h: No such file or directory compilation terminated. configure options: ./autogen.sh --prefix=/usr \ --sysconfdir=/etc \ --with-dri-driverdir=/usr/lib/xorg/modules/dri \ --with-gallium-drivers=r600,swrast \ --with-dri-drivers="" \ --with-egl-platforms=x11,drm \ --enable-gallium-llvm \ --enable-egl \ --enable-gallium-egl \ --enable-shared-glapi \ --enable-gbm \ --enable-glx-tls \ --enable-dri \ --enable-glx \ --enable-osmesa \ --enable-gles1 \ --enable-gles2 \ --enable-texture-float \ --enable-xa \ --enable-vdpau \ --enable-gallium-gbm \ --enable-gallium-g3dvl \ --enable-r600-llvm-compiler
If I would have to make a wild guess I would blame this commit: http://cgit.freedesktop.org/mesa/mesa/commit/?id=998d975e38db7478781130b42a4fb68d6b181055 Since it changes the exact line the first patch for this bug changed.
Thanks for the configure command. With that I was able to replicate the bug. For this latest reopening of this bug, there's already a new, minimal bug report showing the bisect down to the buggy commit. So I'll mark this as a duplicate and we can track the problem there. -Carl *** This bug has been marked as a duplicate of bug 61821 ***
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.