From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:470:142:3::10]:59894 helo=eggs1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jQGBd-0004DW-6k for guix-patches@gnu.org; Sun, 19 Apr 2020 16:05:06 -0400 Received: from Debian-exim by eggs1p.gnu.org with spam-scanned (Exim 4.90_1) (envelope-from ) id 1jQGBa-0005Fq-Ts for guix-patches@gnu.org; Sun, 19 Apr 2020 16:05:04 -0400 Received: from debbugs.gnu.org ([209.51.188.43]:34513) by eggs1p.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1jQGBa-0005FT-H7 for guix-patches@gnu.org; Sun, 19 Apr 2020 16:05:02 -0400 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1jQGBa-0007Vf-Ai for guix-patches@gnu.org; Sun, 19 Apr 2020 16:05:02 -0400 Subject: [bug#40677] [PATCH 10/30] gnu: Add ffmpeg-jami. Resent-Message-ID: Date: Sun, 19 Apr 2020 22:03:57 +0200 From: Jan Message-ID: <20200419220357.1181934c@kompiuter> In-Reply-To: <87pnc5hsaw.fsf@gmail.com> References: <20200417003433.5608-1-tona_kosmicznego_smiecia@interia.pl> <20200417003433.5608-10-tona_kosmicznego_smiecia@interia.pl> <87pnc5hsaw.fsf@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: Mathieu Othacehe Cc: 40677@debbugs.gnu.org On Sat, 18 Apr 2020 14:49:59 +0200 Mathieu Othacehe wrote: > Do we need this particular commit? If so the version should be: > > --8<---------------cut here---------------start------------->8--- > (git-version "4.2.1" revision commit) > --8<---------------cut here---------------end--------------->8--- > > you will find some examples in packages such as guile-webutils, or > boost-sync. Because of the strange bugs I get in Jami, I tried to be as close as possible to the official Jami package. They use this particular commit and if I remember correctly, applying the patches failed without it. The patches add: -Hardware acceleration to use GPU instead CPU to encode/decode video whenever possible -All codecs available (but nor really important, it will be ok with h264 + vp8) -REMB feedback (./ffmpeg/rtp_ext_abs_send_time.patch) to automatically adapt the bitrate and avoid dirty video. > This long list is a bit scary. Could inherit from the one of ffmpeg > and add/remove only what's necessary? The default values should also > be omitted. I tried asking Jami devs how important are these flags, but never got answer. I can ask again, but for now I must check one more thing related to the strange bug I get. > > + ;; Comes from > > + ;; > > "ring-project/daemon/contrib/src/pjproject/rules.mak". > > + ;; WARNING: These amount for huge changes in > > pjproject. > > That's only three patches here? You can maybe remove this comment. > > Mathieu Yes, right. Jan Wielkiewicz