January 2016 - MPlayer-dev-eng - lists.mplayerhq.hu

Relicensing MPlayer or parts of it to LGPL
by wm4 28 Sep '16

28 Sep '16

I'm maintainer of mpv, a MPlayer fork. One of the changes in mpv was turning the slave mode interface into a library (libmpv), which replaces the stdin/stdout parsing slave mode MPlayer uses. One problem with this is that it forces every client to be GPL. libmpv basically exhibits the slave mode interface as API, but since the application code actually links to it, and the player runs in the same address space, the GPL technically applies (in my interpretation). To avoid this, I'm thinking of relicensing mpv to LGPL 2.1 (with the "or later"). While many things have changed in mpv compared to MPlayer (and many parts have been rewritten or removed), much of its core is still based on MPlayer, and I need agreement from the original authors of the code. I'm thinking it would be beneficial for MPlayer as well to relicense parts to LGPL (or all of it, if enough contributors can be reached). Alternatively, I'd just ask for permission to relicense the affected source files in mpv. Would there be any welcoming or opposing to such a move from MPlayer developers?

8 20

[PATCH] Added support for arm64
by jpinto 24 May '16

24 May '16

Signed-off-by: Joao Pinto <jpinto(a)synopsys.com> Tested-by: Joao Pinto <jpinto(a)synopsys.com> --- configure | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/configure b/configure index 722b8a4..d17e61e 100755 --- a/configure +++ b/configure @@ -2493,6 +2493,11 @@ EOF arch='generic' ;; + aarch64) + arch='arm64' + iproc='arm64' + ;; + *) echo "The architecture of your CPU ($host_arch) is not supported by this configure script" echo "It seems nobody has ported MPlayer to your OS or CPU type yet." -- 1.8.1.5

3 3

Re: [MPlayer-dev-eng] [MPlayer-users] Build regression on FreeBSD
by Alexander Strasser 15 Feb '16

15 Feb '16

Hi Thomas, AFAICT there are multiple problems at work, that make a quick solution rather involved or hackish. That's why I have Cc'ed MPlayer development mailing list. On 2015-12-19 20:21 +0100, Thomas Zander wrote: > porting an up-to-date snapshot to FreeBSD I noticed two build > regressions and want to kindly ask for resolving those upstream: > > 1) Makefile (at root dir) > codec-cfg.c now includes <vdpau/vdpau.h> > On FreeBSD this file ends up in LOCALBASE (by default > /usr/local/include) which is not included in the default compiler > search path, so in its build target > codec-cfg$(EXESUF) .... the compile command > $(HOST_CC) $(HOSTCFLAGS) -o $@ $< > is not sufficient. The include path for localbase needs to be added. I guess r37454 introduced this problem, by including vdpau.h in our img_format.h header. It cannot be reverted because that would probably break the build with future FFmpeg versions. The problem is if we want to use external stuff, which cannot be forward declared, in img_format.h we need to add additional flags to HOSTCFLAGS as well. Furthermore it is interesing that vdpau gets detected at all on your system. I suspect this is because of manual addition to CFLAGS or because of modification by previous configure tests. I might be missing a few things, but all in all I do not see why we need the structure vdpau_frame_data in codec-cfg.c source. So maybe the structure should be defined in a different header? All in all I see no really easy way to set HOSTCFLAGS correctly. Additionally I am not sure if the work is justified, because if I am not mistaken above it is not needed in codec-cfg. [...] Alexander

4 12

Coverity
by Ingo Brückl 09 Feb '16

09 Feb '16

Could someone please execute Coverity scans on a more regular basis (like every two or three months)? I found the defects discovered by this tools very interesting and helpful. There is a batch in repository's TOOLS. Ingo

2 5

Building mplayer with ASAN support
by Gustavo Grieco 07 Feb '16

07 Feb '16

Hello, I'm trying to build the last cve snapshot (export 2015-12-18) with address sanitizer in order to test in deep the last version of mplayer. I used: $ ./configure --extra-cflags="-fsanitize=address" but unfortunately, this is not working: $ make cc -MMD -MP -Wundef -Wall -Wno-switch -Wno-parentheses -Wpointer-arith -Wredundant-decls -Werror=format-security -Wstrict-prototypes -Wmissing-prototypes -Wdisabled-optimization -Wno-pointer-sign -Wdeclaration-after-statement -std=gnu99 -Werror-implicit-function-declaration -D_POSIX_C_SOURCE=200112 -D_XOPEN_SOURCE=600 -D_ISOC99_SOURCE -I. -Iffmpeg -O4 -march=native -mtune=native -pipe -ffast-math -fomit-frame-pointer -fno-tree-vectorize -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE64_SOURCE -fsanitize=address -fpie -DPIC -D_REENTRANT -I/usr/include/p11-kit-1 -D_REENTRANT -I/usr/include/freetype2 -DZLIB_CONST -c -o command.o command.c In file included from command.c:27:0: config.h:124:23: warning: missing terminating " character [enabled by default] #define EXTERN_PREFIX "__asan_init_v1 U ^ config.h:125:1: warning: missing terminating " character [enabled by default] " ^ config.h:125:1: error: missing terminating " character In file included from stream/stream.h:22:0, from command.c:30: ./config.h:124:23: warning: missing terminating " character [enabled by default] #define EXTERN_PREFIX "__asan_init_v1 U ^z ./config.h:125:1: warning: missing terminating " character [enabled by default] " ^ In file included from ./mp_msg.h:143:0, from stream/stream.h:24, from command.c:30: ./config.h:124:23: warning: missing terminating " character [enabled by default] #define EXTERN_PREFIX "__asan_init_v1 U ^ ./config.h:125:1: warning: missing terminating " character [enabled by default] " ^ command.c: In function 'mp_property_sub': command.c:1453:22: warning: variable 'reset_spu' set but not used [-Wunused-but-set-variable] int source = -1, reset_spu = 0; ^ make: *** [command.o] Error 1 If i disable the custom CFLAGS, mplayer will compile correctly. Is ASAN supported somehow? Thanks!

2 1

MPlayer 1.2.1 release
by Roberto Togni 27 Jan '16

27 Jan '16

Hi I plan to release 1.2.1 in a few days. 1.2.1 will be made from the 1.2 branch, and will include FFmpeg 2.8.4 In addition to what is already backported, I'd like to include also these changes, even if they are not really bug fixes - r37545 compile with external FFmpeg without requiring x86/asm.h This should make life easier for distributions - r37549 creation of docs without requiring FFmpeg Also useful for distributions, already included in Debian - r37546 silence warning from FFmpeg about network initialization I'm undecided about this, but should be harmless Opinions? If you have anything else that you think is relevant for inclusion in 1.2.1 please commit it, or replay this message. Ciao, Roberto

5 11

SPDIF and ticket #2277
by Roberto Togni 27 Jan '16

27 Jan '16

Hi, I'm looking into ticket #2277. The file was already fixed for normal playback, but it still fails with -ao null. In this case the ad_spdif decoder is invoked, and crashes int lavf. Tha attached patch fixes it, but I have no way to test this. By looking at the change, I don't see any relation between my fix and the fuzzed file; I suspect that spdif was just broken by some changes in ffmpeg and nobody noticed. Can anybody with spdif check if playback is broken with and without my patch? Bonus points if you also check the 1.2 branch. Ciao, Roberto

3 5

MPlayer and raw video
by Mats Peterson 26 Jan '16

26 Jan '16

MPlayer naively assumes that 'rawvideo' raw video from libavcodec (whether it's contained in AVI, QuickTime or something else) contains 24-bit RGB triplets. It *can*, for pixel formats like rgb24 and bgr24, but it can also be palettized or monochrome. Anything that's not compressed/encoded. The following files with raw video will play without video in MPlayer: QuickTime 1 bpp: https://drive.google.com/open?id=0B3_pEBoLs0faU05mTVJ3b3VZcmc QuickTime 2 bpp: https://drive.google.com/open?id=0B3_pEBoLs0faZUc4VEVwalhGSE0 QuickTime 4 bpp: https://drive.google.com/open?id=0B3_pEBoLs0facHJkWnhvYXdtbms QuickTime 8 bpp: https://drive.google.com/open?id=0B3_pEBoLs0fabUdRcTVxRl9yVWM AVI 1 bpp: https://drive.google.com/open?id=0B3_pEBoLs0faelN5RXpKLWtfTnM AVI 2 bpp: https://drive.google.com/open?id=0B3_pEBoLs0fabnQyNm5VV2NXNGM AVI 4 bpp: https://drive.google.com/open?id=0B3_pEBoLs0faTnppb3NjdFQzMmc AVI 8 bpp: https://drive.google.com/open?id=0B3_pEBoLs0faekxWUTdkVGNTTlk Mats -- Mats Peterson http://matsp888.no-ip.org/~mats/

1 1

How to report a couple of crashes with malformed wav files
by Gustavo Grieco 25 Jan '16

25 Jan '16

Hello, I'm the main developer and maintainer of QuickFuzz, a free and open-source experimental grammar fuzzer. I recently made a quick test of Mplayer version shipped with Ubuntu 14.04 and found a few interesting crashes trying to play malformed wav files. These crashes are de-duplicated by Honggfuzz, so they should be more or less independent (although, some are definitely related). The list is here: SIGBUS.PC.5704be.STACK.5c9a551.CODE.128.ADDR.(nil).INSTR.callq__0xfffffffffff3c6e2.fuzz SIGFPE.PC.432d0e.STACK.18b3c0fcd4.CODE.1.ADDR.0x432d0e.INSTR.idivl__0x560ca8(%rip)________#_0x0000000000560cae.fuzz SIGFPE.PC.4bf2c3.STACK.1bca543b66.CODE.1.ADDR.0x4bf2c3.INSTR.divl___0x80(%rsi).fuzz SIGFPE.PC.7ffff3ceed83.STACK.d7d8808dd.CODE.1.ADDR.0x7ffff3ceed83.INSTR.idiv___%r8d.fuzz SIGSEGV.PC.4bd833.STACK.18b2dd10ac.CODE.1.ADDR.0xa.INSTR.movzwl_0xa(%rax),%ecx.fuzz SIGSEGV.PC.4be46b.STACK.1aba63653d.CODE.1.ADDR.(nil).INSTR.movzbl_(%rcx,%rdx,1),%ecx.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7ffffaaf0e08.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7ffffefcecd8.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff48eff8.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff54e258.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff60c128.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff60d498.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff60d4a8.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff60daa8.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff60db58.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5704be.STACK.5c9a551.CODE.1.ADDR.0x7fffff6429a8.INSTR.callq__0xfffffffffff3c6e2.fuzz SIGSEGV.PC.5947b5.STACK.1b0df30f87.CODE.1.ADDR.(nil).INSTR.movzbl_(%rcx,%rdx,1),%ecx.fuzz SIGSEGV.PC.7fffefe7314e.STACK.18f3cc3594.CODE.1.ADDR.(nil).INSTR.movdqu_%xmm8,(%rdi,%rcx,1).fuzz I think some of them can be security issues (that's why they are not linked in this email). I want to handle such crashes to some trusted developers of Mplayer (i found no email address or contact form for security issues) and avoid spamming the bug tracker. Which is the best way to collaborate with the Mplayer project? Regards, Gus.

3 7

[RFC] Remove internal adpcm audio codecs
by Roberto Togni 24 Jan '16

24 Jan '16

Hi I propose to remove the internal adpcm audio codecs from MPlayer, and keep only the ffmpeg version The files involved are ad_dk3adpcm.c DK3 adpcm ad_imaadpcm.c MS ima adpcm, QT ima adpcm, DK4 adpcm ad_msadpcm.c MS adpcm FFmpeg decoder is already preferred for all of them, so nothing should change for normal usage. The internal codecs are less robust that the FFmpeg implementation; so when the initialization of the ffmpeg codec fails on broken files (eg. fuzzed samples), the internal codecs is used. So the internal codecs are now used only on broken files, and they crash. Surely it's possible to fix them, and some trivial fixes were done by Carl and me in the last weeks, but I see no good reason to keep spending resources on them. So if nobody is against it, I'll remove them shortly. Ciao, Roberto

4 4