April 2008 - MPlayer-dev-eng - lists.mplayerhq.hu

[PATCH] Unbreak building
by Luca Barbato 19 Apr '08

19 Apr '08

The recent ffmpeg changes make use of ifdef HAVE_$foo, mplayer configure does define HAVE_$foo in any case. lu -- Luca Barbato Gentoo Council Member Gentoo/linux Gentoo/PPC http://dev.gentoo.org/~lu_zero Index: configure =================================================================== --- configure (revision 26459) +++ configure (working copy) @@ -7818,17 +7818,17 @@ $_target_arch $_target_arch_x86 TARGET_CPU=$iproc -HAVE_MMX = $_mmx -HAVE_MMX2 = $_mmxext -HAVE_3DNOW = $_3dnow -HAVE_3DNOWEX = $_3dnowext -HAVE_SSE = $_sse -HAVE_CMOV = $_cmov -HAVE_ALTIVEC = $_altivec -HAVE_ARMV5TE = $_armv5te -HAVE_ARMV6 = $_armv6 -HAVE_IWMMXT = $_iwmmxt -HAVE_VIS = $_vis +$(test $_mmx != yes && echo "#") HAVE_MMX = $_mmx +$(test $_mmxext != yes && echo "#") HAVE_MMX2 = $_mmxext +$(test $_3dnow != yes && echo "#") HAVE_3DNOW = $_3dnow +$(test $_3dnowext != yes && echo "#") HAVE_3DNOWEX = $_3dnowext +$(test $_sse != yes && echo "#") HAVE_SSE = $_sse +$(test $_cmov != yes && echo "#") HAVE_CMOV = $_cmov +$(test $_altivec != yes && echo "#") HAVE_ALTIVEC = $_altivec +$(test $_armv5te != yes && echo "#") HAVE_ARMV5TE = $_armv5te +$(test $_armv6 != yes && echo "#") HAVE_ARMV6 = $_armv6 +$(test $_iwmmxt != yes && echo "#") HAVE_IWMMXT = $_iwmmxt +$(test $_vis != yes && echo "#") HAVE_VIS = $_vis # --- GUI stuff --- GUI_GTK = $_gui_gtk

1 1

[RFC] auto adjustment of playback speed
by Nico Sabbi 19 Apr '08

19 Apr '08

This patch implements (quite naively) an auto-adjustment mechanism of the playback speed based on the proportion between the real delta_time elapsed from the beginning of the playback and the delta_reference_clock found in the multiplex. At the moment the only mux that carries this information is MPEG-TS and only if demuxed by demux_ts.c. The patch is just in an initial state, but I'd like to know if this code is (or resembles) what Uoti and Rich had in mind when they objected to my commit in ds_fill_buffer() a couple of months ago. The patch should always be used with -af scaletempo, or the audio stream will sound distorted. Possible parameters are: -autospeed to enable the auto-adjustment -autospeed-period (in seconds) indicates the smallest period between two consecutive adjustments -autospeed-threshold is the minimum difference between the two delta_times after which the adjustment will be considered necessary Unfortunately even with this patch applies the video stream sometimes still stutters ... Ideas and comments welcome

2 3

[PATCH] [BUG] make install is broken for mplayer with non-english manual
by Andrew Savchenko 18 Apr '08

18 Apr '08

Hello, r26355 brokes make install for mplayer manual if it is not in DOCS/en/, because of renaming $$i shell variable to $$lang only in the loop declaration. r26383 is an only partial fix for that: it does not replace *all* $$i to $$lang inside a loop (take a look at svn diff Makefile -r 26382:26383, I don't post it here). So make install failes on non-english manuals. The following patch fixes this:

3 2

[PATCH] configure: unset some CPU-specific variables not relevant to arch
by Jan Knutar 16 Apr '08

16 Apr '08

The Problem: Parameters such as HAVE_MMX appear in config.mak with value = auto, a recent-ish change in libavcodec's Makefile makes compilation of mmx and others depend on ifdef HAVE_MMX. When that variable is left defined as "auto", the build system tries to build MMX and SSE code on sparc64, and others. Other suspicious things can be found with 'grep auto config.mak' after configure. The patch: Unset _mmx & co if not on x86, unset altivec if not on ppc|powerpc, unset some arm parameters if not on arm. Fixes the compile on sparc64/netbsd and sparc64/freebsd, and keeps mmx/sse on x86_64 and x86_32. Is it the right place to unset, does it break crosscompilation? I don't know..

2 2

[PATCH] -noconfig option
by Andrew Savchenko 15 Apr '08

15 Apr '08

Hello, Sometimes it is usefull no ignore system and user-specific config files and use only command line options. Especially this is usefull on debugging or figuring out what is wrong with playback. Of course, it possible to override all config file options on the command line, but with large and rather complicated configs (e.g. as mine one) it is a pain. Usually I just temporary rename my mplayer.conf or ~/.mplayer/config to somewhat other name, but this is incovenient too. So I propose a patch which adds appropriate -noconfig command line option to both mplayer and mencoder. It will disable config files parsing at all. The first patch adds functionality described above. Note that option must be checked before configs parsing in the same way as -really-quiet. The second patch is just cosmetics required after first patch applying. Anyone can figure out that there is some code duplication in mplayer.c and mencoder.c. I dislike code duplication, so the third patch is intended to fix this by moving common code (related to the first patch) from these files to mpcommon.[ch]. This patch requires the first patch to be applied. And finally the fourth patch adds new option description to the manual page.

5 17

[PATCH] [RESEND] -aspect-tolerenace option
by Kurt Garloff 14 Apr '08

14 Apr '08

Hi, this patch introduces the aspect tolerance option. When decodding mpeg material, there is a prescaling step involved that takes care of expanding the frames to the right aspect ratio. Scaling is a process that does however reduce the quality, so ideally we would avoid it. The option actually is meant to avoid if if the prescaling done for the aspect ratio is actually only making a tiny correction. This can happen due to roundoff errors, e.g. 1366x768 vs 1360x768 or 1360x766. Default is 0.01, which means that we'll avoid prescaling for up to 0.01 error in the aspect ratio (0.56% for 16:9). Aspect inaccuracies of that size are typically more acceptable than the loss of quality by prescaling (at least for high quality material). Behavior can be controlled via the -aspect-tolerance option. Best, -- Kurt Garloff <kurt(a)garloff.de> [Koeln, DE] Physics:Plasma modeling <garloff(a)plasimo.phys.tue.nl> [TU Eindhoven, NL] Linux: VP PM OPS <garloff(a)suse.de> [Novell Inc]

4 7

Need a guide to implement a codec library using hardware video codec
by Lee Kwangwoo 14 Apr '08

14 Apr '08

Hello, I'm porting the mplayer with a hardware video codec in a SoC. It is capable of decoding H.264 and MPEG4. It can also encode frame with such formats. In the libmpcodecs/vd.c you explicitly note a comment - "do not add any new decoders here" for mpcodes_vd_drivers[]. So if I use the libavcodec/h264.c, should I replace that file with my own decoder file or add feature to support a hardware decoder with a define macro? I think that I need to replace the functions, for example - decode_nal_units(), because the hardware decoder requires one frame starting with nal header and output a decoded YUV420 frame. Any comments and guide will be very helpful. Thank you, Kwangwoo Lee

2 1

fw: HTML OSD
by Dion Rasmussen 13 Apr '08

13 Apr '08

That script I posted earlier, if you're going to use it for anything serious lol (it's only an example); the gmplayer and firefox system-calls arguments should be in list format instead to disallow subscripts or whatever... But an HTML OSD would be quite useful, don't you think?

1 0

[PATCH] Use the normal option parser for command line preparsing
by Alban Bedel 13 Apr '08

13 Apr '08

Hi all, with regard to the -noconfig patch I fixed the command line preparsing to allow using options normally. I put the preparser code with the config file parser together bcs this code is also used by both mplayer and mencoder. After applying the patch command line like "foo.avi -sub -really-quiet" will behave correctly (ie. -really-quiet won't be interpreted as an option). Testing welcome, I plan to apply it tonight unless there is some problem. Albeu

2 7

HTML OSD
by Dion Rasmussen 13 Apr '08

13 Apr '08

Hello again, Now my intelligent library for movies as well as documentation and photos is quite ready and as an example I've written a HTML OSD mechanism in perl utilizing firefox which is following: -snip- #!/usr/bin/perl -w # csmplayer # CutSys extension for mplayer # (c) dion.rasmussen(a)gmail.com use IO::Handle; # fork a player control process to write my $info_pipe_pid = open IPR, '-|'; IPR->autoflush(); die "can't open stdout" unless defined $info_pipe_pid; if(!$info_pipe_pid) { system("killall gmplayer"); system("gmplayer -user Mandrake -passwd mandrake\(a)gangsterfreak.com-skin Terminator3 -vo xv -prefer-ipv4 -quiet -playlist $ARGV[0]"); exit; } open IPW, '>test-pipe' or die "can't open test-pipe $!..."; IPW->autoflush(); print IPW "New session:\n"; my $cut_id; my $ssh_pid; while(<IPR>){ print; if( /^Playing/ ){ # append to the player_log ;) print IPW; # kill the screensaver hold if ($ssh_pid) { kill 15, $ssh_pid; wait; undef $ssh_pid; } # open firefox/osd for CutSys info if( /(http:\/\/[^\/]+)\/cut\/(\d+)/ && (!defined $cut_id ||$2 != $cut_id) && `pidof firefox-bin`) { $cut_id = $2; system "firefox $1/uf/$cut_id"; } } if (/^VIDEO/) { die "couldn't fork" unless defined ($ssh_pid = fork); if(!$ssh_pid) { close IPW; close IPR; # deactivate the screensaver system 'gnome-screensaver-command -d'; # hold the screensaver exec 'gnome-screensaver-command -i'; die "couldn't exec"; } } } print IPW "Lost the pipe...\n"; close IPW; wait; close IPR; if ($ssh_pid) { kill 15, $ssh_pid; wait; } print "done"; -snip- The killall is a problem when that patch I posted earlier ain't applied, but it's showing a quite delicious representation of the current album or series domain which is currently playing. Apropos GUI talk earlier I thought of using libgtkhtml with some perl-hooks or whatever for retrieving track-info and showing it in a HTML-rendered part of the GUI instead of firefox. Just a thought! Regards, Dion

1 0