Bug 428282 – net-wireless/bluez-4.99 expected specifier-qualifier-list before ‘typeof’ in bluetooth.h

• Product: Gentoo Linux
• Status: RESOLVED TEST-REQUEST

[Description W. Trevor King 2012-07-27 09:54:59 UTC]

Created attachment 319370 [details]
Log of failing build.

While emerging app-mobilephone/obexpushd-9999 from the my wtk overlay I ran into:

  cd /var/tmp/portage/app-mobilephone/obexpushd-9999/work/obexpushd-9999_build/src && /usr/bin/x86_64-pc-linux-gnu-gcc  -D_GNU_SOURCE -DUSE_LIBGCRYPT -DHAVE_SDPLIB -DUSB_GADGET_SUPPORT -DENABLE_TCPWRAP -DUSE_XATTR -DOPENOBEX_TCPOBEX=1 -DDEFINITIONS  -DNDEBUG -O2 -march=native -pipe  -std=c99 -I/var/tmp/portage/app-mobilephone/obexpushd-9999/work/obexpushd-9999/src -I/var/tmp/portage/app-mobilephone/obexpushd-9999/work/obexpushd-9999_build/src    -o CMakeFiles/obexpushd.dir/obexpushd.c.o -c /var/tmp/portage/app-mobilephone/obexpushd-9999/work/obexpushd-9999/src/obexpushd.c
  In file included from /var/tmp/portage/app-mobilephone/obexpushd-9999/work/obexpushd-9999/src/obexpushd.c:18:0:
  /usr/include/bluetooth/bluetooth.h: In function ‘bt_get_le64’:
  /usr/include/bluetooth/bluetooth.h:159:9: error: expected specifier-qualifier-list before ‘typeof’
  /usr/include/bluetooth/bluetooth.h:159:9: error: called object ‘typeof(__p)’ is not a function
  /usr/include/bluetooth/bluetooth.h:159:9: error: ‘struct <anonymous>’ has no member named ‘__v’

This seems to be the same as Debian's 661879 [1], which they fixed by patching bluez [2].


[1]: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661879
[2]: http://patch-tracker.debian.org/patch/series/view/bluez/4.101-1/09_fix_ftbfs_with_c99.patch

[Comment 1 W. Trevor King 2012-07-27 09:55:25 UTC]

Created attachment 319372 [details]
emerge --info

[Comment 2 W. Trevor King 2012-07-27 09:56:16 UTC]

Created attachment 319374 [details, diff]
Patch from Debian (http://patch-tracker.debian.org/patch/series/view/bluez/4.101-1/09_fix_ftbfs_with_c99.patch)

[Comment 3 W. Trevor King 2012-07-27 17:56:04 UTC]

I can confirm that the patch does fix the issue for me.

The patch made it into the upstream mailing list back in March, but doesn't seem to be included in their repository yet.  I've bumped them, and will report back on what I hear.

[Comment 4 Pacho Ramos 2012-09-05 17:56:06 UTC]

It could be a gcc bug per:
https://bugzilla.redhat.com/show_bug.cgi?id=786966#c5

[Comment 5 Tiziano Müller (RETIRED) 2012-10-28 18:52:00 UTC]

reassigning to pacho (thanks!) since I stepped down as maintainer for bluez

[Comment 6 Pacho Ramos 2012-11-18 14:08:25 UTC]

(In reply to comment #4)
> It could be a gcc bug per:
> https://bugzilla.redhat.com/show_bug.cgi?id=786966#c5

But no idea about what gcc version includes a fix for that

[Comment 7 W. Trevor King 2012-11-18 16:20:38 UTC]

(In reply to comment #3)
> I've bumped them, and will report back on what I hear.

Thanks Pacho for pushing this forward!  Upstream threads (so others don't need to dig through Gmane):

http://thread.gmane.org/gmane.linux.bluez.kernel/20372  (2012-01-16)
http://thread.gmane.org/gmane.linux.bluez.kernel/22306  (2012-03-05)
http://thread.gmane.org/gmane.linux.bluez.kernel/27931  (2012-07-29)

[Comment 8 Pacho Ramos 2013-02-12 19:32:06 UTC]

Still valid with gcc-4.6.3?

[Comment 9 W. Trevor King 2013-02-12 21:43:00 UTC]

(In reply to comment #8)
> Still valid with gcc-4.6.3?

Yeah, I got the same error using gcc-4.6.3 and bluez-4.101-r5.

[Comment 10 Pacho Ramos 2013-10-03 06:12:33 UTC]

And with gcc-4.7.3-r1?

[Comment 11 W. Trevor King 2013-10-11 23:00:44 UTC]

(In reply to Pacho Ramos from comment #10)
> And with gcc-4.7.3-r1?

I'll let you know once something in the 4.7 series stabilizes ;).

[Comment 12 Pacho Ramos 2013-10-12 06:07:44 UTC]

It's in progress :)
https://bugs.gentoo.org/show_bug.cgi?id=467274