November 2011 - MPlayer-dev-eng - lists.mplayerhq.hu

Homepage skin download issue
by Ingo Brückl 09 Sep '15

09 Sep '15

Yesterday, I updated hwswskin to 1.3, but it still isn't downloadable. Is the generation of the skin download tarballs working? Ingo

3 8

Homepage DOCS needs update
by Ingo Brückl 24 Apr '14

24 Apr '14

If it's already on the to-do list ignore, but the documentation on the homepage doesn't get updated. Ingo

3 12

[PATCH] xvmc: handle shm xvimage allocation error
by Marcin Slusarz 06 Jan '14

06 Jan '14

- handle xvimage/shm allocation failure and fallback to non-shm xv - assert xvimage->data_size!=0 because 0 size will lead to memory corruptions (I hit it because of bug in Nouveau's early xvmc implementation) Index: libvo/vo_xvmc.c =================================================================== --- libvo/vo_xvmc.c (wersja 34098) +++ libvo/vo_xvmc.c (kopia robocza) @@ -181,6 +181,15 @@ return cur_render->state || (cur_render->mpi && cur_render->mpi->usage_count); } +static void allocate_xvimage_non_shm(int xvimage_width, int xvimage_height, int xv_format) +{ + xvimage = (XvImage *) XvCreateImage(mDisplay, xv_port, xv_format, NULL, + xvimage_width, xvimage_height); + assert(xvimage->data_size); + xvimage->data = malloc(xvimage->data_size); + XSync(mDisplay,False); +} + static void allocate_xvimage(int xvimage_width,int xvimage_height,int xv_format) { /* @@ -188,7 +197,8 @@ * mit-shm this will bomb... trzing to fix ::atmos */ #ifdef HAVE_SHM - if ( mLocalDisplay && XShmQueryExtension( mDisplay ) ) Shmem_Flag = 1; + if ( mLocalDisplay && XShmQueryExtension( mDisplay ) ) + Shmem_Flag = 0; else { Shmem_Flag = 0; @@ -198,25 +208,41 @@ { xvimage = (XvImage *) XvShmCreateImage(mDisplay, xv_port, xv_format, NULL, xvimage_width, xvimage_height, &Shminfo); + if (!xvimage) + goto noshmimage; + assert(xvimage->data_size); Shminfo.shmid = shmget(IPC_PRIVATE, xvimage->data_size, IPC_CREAT | 0777); + if (Shminfo.shmid == -1) + goto shmgetfail; + Shminfo.shmaddr = (char *) shmat(Shminfo.shmid, 0, 0); + if ((long)Shminfo.shmaddr == -1) + goto shmatfail; + Shminfo.readOnly = False; xvimage->data = Shminfo.shmaddr; - XShmAttach(mDisplay, &Shminfo); + if (!XShmAttach(mDisplay, &Shminfo)) + goto shmattachfail; + XSync(mDisplay, False); shmctl(Shminfo.shmid, IPC_RMID, 0); } else #endif - { - xvimage = (XvImage *) XvCreateImage(mDisplay, xv_port, xv_format, NULL, xvimage_width, xvimage_height); - xvimage->data = malloc(xvimage->data_size); - XSync(mDisplay,False); - } -// memset(xvimage->data,128,xvimage->data_size); + allocate_xvimage_non_shm(xvimage_width, xvimage_height, xv_format); return; + +shmattachfail: + shmdt(Shminfo.shmaddr); +shmatfail: + shmctl(Shminfo.shmid, IPC_RMID, 0); +shmgetfail: + XFree(xvimage); +noshmimage: + Shmem_Flag = 0; + allocate_xvimage_non_shm(xvimage_width, xvimage_height, xv_format); } static void deallocate_xvimage(void)

4 6

GUI and vo_x11 functions
by Ingo Brückl 02 Mar '13

02 Mar '13

I'm about to commit the vo_x11_ewmh_fullscreen() patch which adds a window parameter to the function, but I'm not quite sure whether this really is ok, although I got permission. The question is whether vo_x11_ewmh_fullscreen() and all the other vo_x11 functions should only operate on the global vo variables (like vo_window in that case) or whether they should be reusable functions, offering services for whoever wants to call them (the GUI in my case). The reason why I raise this issue is that the GUI does call three other vo_x11 functions (with relative success), namely vo_x11_decoration() vo_x11_sizehint() vo_x11_setlayer() While vo_x11_decoration() seems to be problem-free (it has some useful parameters, but uses global variables as well), vo_x11_sizehint() mostly uses global variables (which causes some X errors with the GUI then if called at a time when these variables aren't yet set). vo_x11_setlayer() is pointlessly called by the GUI, because it doesn't do anything in case of WinID >= 0. So, what is the right way to continue? Setting all necessary global vo variables in the GUI - even prior to the vo driver (vo_window is known to become WinID later anyway, for example), trying to modestly patch these four functions so that they can be reused or copy the code and letting the GUI use its own copies? All these four functions are window specific and window handling should be GUI's task although commonly used code is convenient. Pros and cons should be obvious. Ingo

3 16

[PATCH] fix OSD decimal output for dvdnav
by Roger Pack 22 Apr '12

22 Apr '12

Hello. Apologies for the flood. Anyway, currently with the OSD decimal output, it gets the "second" portion from demuxer_get_current_time (the NAV packets) and gets the "decimal portion" of the output from sh_video->pts. Since these don't match with DVD's (the NAV packets only occur every 0.5s, and are usually about 0.1s "off" the MPEG timestamps), it can lead to odd OSD output, for example: $ mplayer dvdnav://1 -osd-fractions 1 -osdlevel 2 OSD output: ... 00:00:04.04 00:00:04.08 00:00:04.12 00:00:04.16 00:00:04.20 00:00:04.24 00:00:04.29 00:00:04.33 00:00:05.37 # note the shift here... 00:00:05.41 00:00:05.45 00:00:05.50 00:00:05.55 00:00:05.60 00:00:05.65 00:00:05.70 00:00:05.75 00:00:05.80 00:00:05.84 00:00:05.90 00:00:05.95 00:00:05.99 00:00:05.04 # shift again, because the two are offset. ... [repeat] Also if there has been a timestamp reset ever, it leads to even odder output numbers, like 01:20:05.-8 displayed in the OSD. This patch makes the OSD "monotonically increase" and doesn't seem to hurt other stream types. For dvdnav, it now outputs the equivalent of: ... 00:00:00.40 00:00:00.40 00:00:00.40 00:00:00.40 00:00:00.80 00:00:00.80 00:00:00.80 00:00:00.80 00:00:00.80 00:00:00.80 00:00:00.80 00:00:01.20 00:00:01.20 ... which, though less precise, seems more accurate (and works past reset's et al). My hope is to (should these patches eventually be applied) submit another diff that makes these timestamps even more accurate, but this would is an improvement, at least from my perspective. I also am considering another similar patch for EDL timestamps checking, so that EDL's can work with dvdnav (esp. past any MPEG PTs resets). Feedback welcome. Thanks. -roger-

2 8

playing AC3 through HDMI abnormal sound issue
by AnnieLiu＠viatech.com.cn 12 Feb '12

12 Feb '12

Dear All I am Annie Liu from VIA Technology. I met an issue when play ac3 through HDMI by mplayer and need your help. I added the codec patch into alsa-driver to support our HDMI audio and I declared it as “HDA_PCM_TYPE_HDMI”. It works fine when play wav file. Now, I met an issue when playing audio with AC3 format through HDMI. If I use VLC (A52->S/PDIF), the sound is Okay, and the audio format displayed on amplifier is “D” (Dolby) as well. But I use mplayer with option “-ac hwac3”, the audio sounds abnormal, although the amplifier display “D” (Dolby). I tried mplayer with “-ac a52”, but I found it decode AC3 to PCM. On the contrary, if I try “-ac hwac3” on a “digital codec” in our SB which declared as “HDA_PCM_TYPE_SPDIF”, the sound is fine; but if I use VLC (A52->S/PDIF) no sound outputted. In VLC, I also found some a52 libraries: /usr/lib/vlc/plugins/codec/liba52_plugin.so /usr/lib/vlc/plugins/audio_filter/liba52tospdif_plugin.so I wonder how I can play ac3 by mplayer through HDMI in Dolby? I tried to declare my HDMI as “HDA_PCM_TYPE_SPDIF” in codec patch, the abnormal issue happened as well. (Playing PCM is Okay). I wonder if the audio data encoded by hwac3 is compatible with HDMI. I searched a lot of topics related to playing ac3 through HDMI, and found someone said I should set AES0 as 0x02. I did it by iecset, it works. But when I play ac3 by mplayer, it always reports “[AO-ALSA] alsa-lib: pcm.c:2208: (snd_pcm_open_noupdate) Unknown PCM hw:0,3,AES0=6”. I don’t know why it happens. Could anyone give me favor on it? Thanks a lot in advance! Thanks a lot. BR//Annie Annie Liu （刘晓丽） ----------------------------------------------------------------- Software Team, 5th Floor, VIA Technologies (China) Inc., Ltd. VIA Building, Tsinghua Science Park building No.7 No.1 Zhongguancun East Road, Haidian District, Beijing, 100084 Tel: 86-10-59852288 ext. 3865 E-mail: annieliu(a)viatech.com.cn <blocked::mailto:annieliu@viatech.com.cn> ----------------------------------------------------------------- 北京市海淀区中关村东路1号院7号楼威盛中国芯大厦 5层，软件部邮编： 100084 -----------------------------------------------------------------

2 1

[RFC] Playing growing file
by Vlad Seryakov 07 Feb '12

07 Feb '12

Oops, sorry, sent wrong file. This is the correct one with copy of the whole message: Hi, This is just some crazy idea i am thinking about while trying to figure out how to play MPEG TS recordings with mplayer. Really this is not full timeshift solution and never be, player is just a player but supporting growing file may be very useful. I decided to try with adding one function to stream_file.c which will handle pipe:// urls and just use different fill_buffer function which indefinitely reads from file but process events to avoid hangs. Just wondetng what other think about this solution. Thanks

3 2

[PATCH] remove mp3lib
by Reimar Döffinger 05 Feb '12

05 Feb '12

Hello, I am still interested in seeing performance numbers and statistics from ffmp3 vs. mp3lib when ffmp3 is slower, but I guess it's time to start some more specific moves to get rid of it.

12 59

GUI Icon
by Ingo Brückl 10 Jan '12

10 Jan '12

Although I like the MPlayer logo, I think that it isn't really suitable for a desktop icon. On Bugzilla (http://bugzilla.mplayerhq.hu/show_bug.cgi?id=710) there is an offer for a GUI icon set that is quite nice. There is licence information for the Tango Desktop Project Icon Library, but it is missing for the Tango Fridays which supposably is a problem. If there is no general objection in using these icons for the GUI, I could try to contact the creator, Andreas Nilsson, and ask for a full set of these icons and a licence information to include both to the MPlayer repository. Would that be OK? Ingo PS: Where can I find the original (svg?) of the MPlayer logo?

2 2

[PATCH] add support of compression algorithm 3 in mkv demuxer
by Alexander Beregalov 02 Jan '12

02 Jan '12

Specs from http://www.matroska.org/technical/specs/ : ContentCompAlgo The compression algorithm used. Algorithms that have been specified so far are: 0 - zlib, 1 - bzlib, 2 - lzo1x, 3 - Header Stripping. ContentCompSettings Settings that might be needed by the decompressor. For Header Stripping (ContentCompAlgo=3), the bytes that were removed from the beggining of each frames of the track. I have such file, it works for me. | + A track | + Track number: 2 | + Track UID: 2 | + Track type: audio | + Lacing flag: 1 | + Default flag: 1 | + Codec ID: A_AC3 | + Content encodings | + Content encoding | + Content compression | + Algorithm: 3 (header removal) | + Settings: length 2, data: 0x0b 0x77 See an attachment. Please comment.

7 21