To use experimental features one have to pass -E to the daemon.
Before I forget, you probaly want to add a fix for https://github.com/bluez/bluez/issues/102 which is https://github.com/BluezTestBot/bluez/commit/28ddec8d6b829e002fa268c07b71e4c564ba9e16
I cannot test the openrc script... but if you have a patch for its init.d file... you are welcome :)
Created attachment 692976 [details, diff] Fix both issues
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=e7d2a2805d8c54cfdeab02e3e60935629f471ac7 commit e7d2a2805d8c54cfdeab02e3e60935629f471ac7 Author: Pacho Ramos <pacho@gentoo.org> AuthorDate: 2021-03-22 21:16:26 +0000 Commit: Pacho Ramos <pacho@gentoo.org> CommitDate: 2021-03-22 21:16:26 +0000 net-wireless/bluez: Fix reconnect issues and allow to pass options... Upstream fix for reconnect issues and allow to pass options to init.d script. Also drop obsolete btattach hacks as Fedora did. Closes: https://bugs.gentoo.org/777000 Thanks-to: Joakim Tjernlund Package-Manager: Portage-3.0.17, Repoman-3.0.2 Signed-off-by: Pacho Ramos <pacho@gentoo.org> net-wireless/bluez/bluez-5.56-r1.ebuild | 295 +++++++++++++++++++++ net-wireless/bluez/files/bluetooth-conf.d | 1 + net-wireless/bluez/files/bluetooth-init.d-r5 | 13 + .../bluez/files/bluez-5.56-avdtp-disconnects.patch | 41 +++ 4 files changed, 350 insertions(+)
Thanks, I did note an old bug though: -e "s:cupsdir = \$(libdir)/cups:cupsdir = $(cups-config --serverbin):" \ Makefile.{in,tools} || die should be Makefile.{am,... a -9999 ebuild would be great too: if [[ ${PV} = 9999 ]]; then inherit git-r3 SRC_URI="" EGIT_BRANCH="master" EGIT_REPO_URI="https://git.kernel.org/pub/scm/bluetooth/bluez.git" fi
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=55b826b94264ce7e979c20f416504541d3cf1aaf commit 55b826b94264ce7e979c20f416504541d3cf1aaf Author: Pacho Ramos <pacho@gentoo.org> AuthorDate: 2021-03-23 21:59:45 +0000 Commit: Pacho Ramos <pacho@gentoo.org> CommitDate: 2021-03-23 21:59:45 +0000 net-wireless/bluez: Fix sed for cupsdir Bug: https://bugs.gentoo.org/777000 Thanks-to: Joakim Tjernlund Package-Manager: Portage-3.0.17, Repoman-3.0.2 Signed-off-by: Pacho Ramos <pacho@gentoo.org> net-wireless/bluez/bluez-5.56-r1.ebuild | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Regarding the live ebuild... I never liked then much and, as I wouldn't use it... it is likely to get broken without noticing. If there are important fixes... it's better to try to apply them to normal releases or, in the worst case, provide a snapshot I think :/
(In reply to Pacho Ramos from comment #7) > Regarding the live ebuild... I never liked then much and, as I wouldn't use > it... it is likely to get broken without noticing. If there are important > fixes... it's better to try to apply them to normal releases or, in the > worst case, provide a snapshot I think :/ Well, I like them. Makes it easy to track something you are interested in or need. You would notice breakage each time you bump them. Anyhow, I think I tricked you a bit before. eautoreconf should be before sed as autoreconf may touch Makefile.in: eautoreconf if use cups; then sed -i \ -e "s:cupsdir = \$(libdir)/cups:cupsdir = $(cups-config --serverbin):" \ Makefile.{in,tools} || die fi drop the am part, just Makefile.{in,tools}
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=af81c2377b8238d5c51509c1cf743b97f37d352f commit af81c2377b8238d5c51509c1cf743b97f37d352f Author: Pacho Ramos <pacho@gentoo.org> AuthorDate: 2021-03-24 07:57:47 +0000 Commit: Pacho Ramos <pacho@gentoo.org> CommitDate: 2021-03-24 07:57:58 +0000 net-wireless/bluez: Not require eautoreconf because of cupsdir Bug: https://bugs.gentoo.org/777000 Package-Manager: Portage-3.0.17, Repoman-3.0.2 Signed-off-by: Pacho Ramos <pacho@gentoo.org> net-wireless/bluez/bluez-5.56-r1.ebuild | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
