April 2009 - MPlayer-dev-eng - lists.mplayerhq.hu

[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

[PATCH] Network synchronized playback using UDP
by Jason Holt 08 Sep '10

08 Sep '10

I hacked in a simple mechanism to synchronize playback between multiple instances of mplayer over a network. The master sends a UDP datagram to announce its position in the video each time it plays a frame. The slaves check to see if they're within epsilon of the master. If they are, they just advance by a frame. If not, they seek to where the master is. This works out nicely because I can dynamically tweak frame rate, seek and pause on the master and the slaves all follow along. The patch is not compliant with your coding standards. Before I do that work, I wanted to make sure it's a patch you're interested in, and see if you have any particular guidance on things I should move or change.

10 78

[PATCH]Set default mc to 1
by Carl Eugen Hoyos 08 Jun '10

08 Jun '10

Hi! Attached patch makes all PAFF samples playable (with a small, for some samples even acceptable delay), while I do not know of any sample that does not desync immediately with latest svn. Please comment, Carl Eugen

3 3

rtp support in mplayer
by chas williams - CONTRACTOR 30 Apr '10

30 Apr '10

while trying to play from rtp/mpeg-es streams i found i needed the following patch: --- stream/rtp.c.000 2008-05-13 17:37:23.000000000 -0400 +++ stream/rtp.c 2008-05-13 18:02:02.000000000 -0400 @@ -246,6 +246,9 @@ headerSize = 12 + 4*rh->b.cc; /* in bytes */ + if (rh->b.pt == 32) /* 32 = MPV from RFC 1890 */ + headerSize += 4; /* skip MPEG video-specific header RFC 2250 */ + *lengthData = lengthPacket - headerSize; *data = (char*) buf + headerSize; to skip over the 4-byte mpeg header present in the file. however, after looking at mplayer a bit harder, i see that there are two rtp decoders/demuxers. the one supplied in libavformat (apparently derived from ffmpeg) seems like it has more smarts. which one should i be using?

4 13

[PATCH] Seeking in multirate RealMedia files.
by Zuxy Meng 21 Feb '10

21 Feb '10

Hi, The attached implements seeking in multirate RM files, with some code clean up. Now both normal (interleaved) and multirate RM files maintain current_apacket and current_vpacket and use these counters in seeking. Tested with several normal and multirate files and seemed to work fine. -- Zuxy Beauty is truth, While truth is beauty. PGP KeyID: E8555ED6

4 7

[patch] play and seek files inside multivolume uncompressed rar files
by Emiel Neggers 28 Dec '09

28 Dec '09

Hello mplayer devs, Attached to this email is a patch which i wrote about a year ago that lets you play files in uncompressed, multi-volume rar files, as long as there is only one video file inside the rars (which is often the case). It probably (surely) isn't ready for inclusion in mplayer as-is, has no error reporting, and doesn't work on rar files with compression. However, I've been using it a lot for about a year now without any problems. Do with it what you wish :) I also attached the bash file that i use to play these rar's which shows how to use this patch. Regards, Emiel p.s. this is a 2nd email, because the previous one is being held in moderation as i was not yet subscribed. the kind people on irc told me it was better to send it again so there.

6 16

Another sample from norway
by Håkon Alstadheim 27 Aug '09

27 Aug '09

This is just to motivate anybody having a spare 5 minutes into working receiving the norwegian dvb-t broadcasts ... (should also interest people from NZ) :-) I've uploaded another sample to incoming, filename hakon.alstadheim.priv.no.nrk3.ts. Attaching the standard output and error when playing below. As you can see I have patched my mplayer to force 25 fps framerate. Seems that patch no longer works as intended, but the straight (non-patched) mplayer gives similarily discouraging results when viewing. Happy hacking everybody :-) . If you want more debugging output, just tell me what and I'll try to comply. -------------------- MPlayer dev-SVN-r26171-4.1.2 (C) 2000-2008 MPlayer Team CPU: Intel(R) Pentium(R) 4 CPU 2.00GHz (Family: 15, Model: 2, Stepping: 4) CPUflags: MMX: 1 MMX2: 1 3DNow: 0 3DNow2: 0 SSE: 1 SSE2: 1 Compiled for x86 CPU with extensions: MMX MMX2 SSE SSE2 Playing hakon.alstadheim.priv.no.nrk3.ts. TS file format detected. VIDEO H264(pid=521) AUDIO AAC(pid=676) NO SUBS (yet)! PROGRAM N. 103 FPS seems to be: 25.000000 ========================================================================== Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family Selected video codec: [ffh264] vfm: ffmpeg (FFmpeg H.264) ========================================================================== ========================================================================== Opening audio decoder: [faad] AAC (MPEG2/4 Advanced Audio Coding) versionA not supported FAAD: compressed input bitrate missing, assuming 128kbit/s! AUDIO: 44100 Hz, 2 ch, s16le, 128.0 kbit/9.07% (ratio: 16000->176400) Selected audio codec: [faad] afm: faad (FAAD AAC (MPEG-2/MPEG-4 Audio) decoder) ========================================================================== AO: [oss] 44100Hz 2ch s16le (2 bytes per sample) Starting playback... FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: Failed to decode frame: Invalid number of channels [h264 @ 0x88d3070]left block unavailable for requested intra mode at 0 4 [h264 @ 0x88d3070]error while decoding MB 0 4, bytestream (24666) FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Quantised value out of range, trying to resync! FAAD: error: Quantised value out of range, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: Failed to decode frame: Gain control not yet implemented VDec: vo config request - 720 x 576 (preferred colorspace: Planar YV12) VDec: using Planar YV12 as output csp (no 0) Movie-Aspect is 1.82:1 - prescaling to correct movie aspect. VO: [xv] 720x576 => 1048x576 Planar YV12 [ASPECT] Warning: No suitable new res found! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Unexpected fill element with SBR data, trying to resync! FAAD: Failed to decode frame: Unexpected fill element with SBR data [h264 @ 0x88d3070]B picture before any references, skipping [h264 @ 0x88d3070]decode_slice_header error [h264 @ 0x88d3070]no frame! Error while decoding frame! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Scalefactor out of range, trying to resync! FAAD: error: Pulse coding not allowed in short blocks, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: Failed to decode frame: Gain control not yet implemented [h264 @ 0x88d3070]B picture before any references, skipping [h264 @ 0x88d3070]decode_slice_header error [h264 @ 0x88d3070]no frame! Error while decoding frame! FAAD: error: Invalid number of channels, trying to resync! ??,?% 0 0 FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: Failed to decode frame: Maximum number of scalefactor bands exceeded [h264 @ 0x88d3070]B picture before any references, skipping [h264 @ 0x88d3070]decode_slice_header error [h264 @ 0x88d3070]no frame! Error while decoding frame! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Gain control not yet implemented, trying to resync! FAAD: error: Channel coupling not yet implemented, trying to resync! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: error: Invalid number of channels, trying to resync! FAAD: Failed to decode frame: Invalid number of channels [h264 @ 0x88d3070]B picture before any references, skipping [h264 @ 0x88d3070]decode_slice_header error [h264 @ 0x88d3070]no frame! Error while decoding frame! FAAD: error: Maximum number of scalefactor bands exceeded, trying to resync! FAAD: error: Quantised value out of range, trying to resync! MPlayer interrupted by signal 11 in module: decode_audio - MPlayer crashed by bad usage of CPU/FPU/RAM. Recompile MPlayer with --enable-debug and make a 'gdb' backtrace and disassembly. Details in DOCS/HTML/en/bugreports_what.html#bugreports_crash. - MPlayer crashed. This shouldn't happen. It can be a bug in the MPlayer code _or_ in your drivers _or_ in your gcc version. If you think it's MPlayer's fault, please read DOCS/HTML/en/bugreports.html and follow the instructions there. We can't and won't help unless you provide this information when reporting a possible bug.

3 5

[PATCH] cmd max size from 256 to 1024
by Mathieu SCHROETER 26 Jun '09

26 Jun '09

Hi, I suggest to increase the cmd max size from 256 to at least 1024. With the slave command `loadfile <path>`, 256 chars is easily reached. Regards, -- Mathieu SCHROETER

3 2

[PATCH]: af_hrtf SIMD optimization (3DNOW!/SSE/SSE3/plain C)
by Zhou Zongyi 09 Jun '09

09 Jun '09

Hi all, This patch implements SIMD optimized FIR filter used by hrtf. They can also be used by other audio filters. Since SSE3 is used, you should apply SSE3 detection patch before applying this one: http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/2009-January/059824.html Index: libaf/af_hrtf.c =================================================================== --- libaf/af_hrtf.c (revision 28934) +++ libaf/af_hrtf.c (working copy) @@ -27,6 +27,7 @@ #include <math.h> +#include "cpudetect.h" #include "af.h" #include "dsp.h" @@ -77,19 +78,40 @@ * sk: convolution kernel * offset: offset on the ring buffer, can be */ -static float conv(const int nx, const int nk, const float *sx, const float *sk, - const int offset) -{ - /* k = reminder of offset / nx */ - int k = offset >= 0 ? offset % nx : nx + (offset % nx); - if(nk + k <= nx) - return af_filter_fir(nk, sx + k, sk); - else - return af_filter_fir(nk + k - nx, sx, sk + nx - k) + - af_filter_fir(nx - k, sx + k, sk); +#define conv_routine(x)\ +static float x(const int nx, const int nk, const float *sx, const float *sk, const int offset)\ +{\ + /* k = reminder of offset / nx */\ + int k = offset >= 0 ? offset % nx : nx + (offset % nx);\ + if(nk + k <= nx)\ + return af_filter_fir(nk, sx + k, sk);\ + else\ + return af_filter_fir(nk + k - nx, sx, sk + nx - k) +\ + af_filter_fir(nx - k, sx + k, sk);\ } + +#if HAVE_MMX +conv_routine(conv_c) + +#define af_filter_fir af_filter_fir_sse +conv_routine(conv_sse) +#undef af_filter_fir + +#define af_filter_fir af_filter_fir_sse3 +conv_routine(conv_sse3) +#undef af_filter_fir + +#define af_filter_fir af_filter_fir_3dnow +conv_routine(conv_3dnow) +#undef af_filter_fir + +static float (*conv)(const int, const int, const float*, const float*, const int) = NULL; +#else +conv_routine(conv) +#endif + /* Detect when the impulse response starts (significantly) */ static int pulse_detect(const float *sx) { @@ -107,14 +129,8 @@ /* Fuzzy matrix coefficient transfer function to "lock" the matrix on a effectively passive mode if the gain is approximately 1 */ -static inline float passive_lock(float x) -{ - const float x1 = x - 1; - const float ax1s = fabs(x - 1) * (1.0 / MATAGCLOCK); +#define passive_lock(x) ((float)((x-1)-(x-1)/(1+(x-1)*(x-1)*(1/MATAGCLOCK/MATAGCLOCK))+1)) - return x1 - x1 / (1 + ax1s * ax1s) + 1; -} - /* Unified active matrix decoder for 2 channel matrix encoded surround sources */ static inline void matrix_decode(short *in, const int k, const int il, @@ -668,6 +684,15 @@ for(i = 0; i < s->basslen; i++) s->ba_ir[i] *= BASSGAIN; +#if HAVE_MMX + if (!conv){ + conv = gCpuCaps.hasSSE3 ? conv_sse3 + :(gCpuCaps.hasSSE ? conv_sse + :(gCpuCaps.has3DNow? conv_3dnow + : conv_c)); + af_msg(AF_MSG_INFO,"[hrtf] Using %s optimized converter\n",gCpuCaps.hasSSE3?"SSE3":(gCpuCaps.hasSSE?"SSE":(gCpuCaps.has3DNow? "3DNow!":"non"))); + } +#endif return AF_OK; } Index: libaf/af_hrtf.h =================================================================== --- libaf/af_hrtf.h (revision 28934) +++ libaf/af_hrtf.h (working copy) @@ -106,7 +106,7 @@ /* Center front (-5 degree) - not 0 degree in order to create a clear front image from a finite distance */ -static const float cf_filt[128] = { +static const float cf_filt[128] __attribute__((aligned(16))) = { -0.00008638082319075036, 0.0003198059946385229, -0.0005010631339162132, 0.0011424741331126876, -0.001584220794688753, 0.001742715363246275, @@ -173,7 +173,7 @@ -0.005469203016468759, -0.004355784273189485 }; /* Front, same side (30 degree) */ -static const float af_filt[128] = { +static const float af_filt[128] __attribute__((aligned(16))) = { -0.004140580614755493, 0.005790934614385445, 0.003318916682081112, 0.014257145544366063, 0.007328442487127339, -0.06550381777876194, @@ -240,7 +240,7 @@ -0.0005083025643192044, 0.00035096963769606926 }; /* Front, opposite (-30 degree) */ -static const float of_filt[128] = { +static const float of_filt[128] __attribute__((aligned(16))) = { -0.000013472538374193126, -0.00008048061877079751, 0.000043927265781258155, -0.000017931700794858892, -0.000034774602476112886, -0.00009576223008735474, @@ -307,7 +307,7 @@ -0.0013726264946164232, -0.0020075119315034313 }; /* Rear, same side (135 degree) */ -static const float ar_filt[128] = { +static const float ar_filt[128] __attribute__((aligned(16))) = { 0.004573315040810066, 0.0013592578059426913, 0.01553271930902704, -0.0002356117224941317, -0.05746098219774702, 0.03430688963370445, @@ -374,7 +374,7 @@ -0.0026884314856593368, -0.004084181815125924 }; /* Rear, opposite (-135 degree) */ -static const float or_filt[128] = { +static const float or_filt[128] __attribute__((aligned(16))) = { 0.0001220944028243897, -0.000021785381808441314, 5.823057988603169e-6, -0.00001217768176447613, -0.00006123604397345513, 5.574117262531134e-6, @@ -441,7 +441,7 @@ -0.001852908777923165, -0.002540339553144362 }; /* Center rear (180 degree) */ -static const float cr_filt[128] = { +static const float cr_filt[128] __attribute__((aligned(16))) = { -0.00005989110716536726, -0.00022790291829128702, 0.0002659166098971966, -0.0003774772716776257, 0.0004540309551867803, -0.000420238187386368, Index: libaf/filter.c =================================================================== --- libaf/filter.c (revision 28934) +++ libaf/filter.c (working copy) @@ -22,6 +22,8 @@ #include <string.h> #include <math.h> +#include "config.h" +#include "cpudetect.h" #include "dsp.h" /****************************************************************************** @@ -39,13 +41,141 @@ { register FLOAT_TYPE y; // Output y = 0.0; - do{ - n--; - y+=w[n]*x[n]; - }while(n != 0); + while((int)(n-=4)>=0){ + y+=w[n]*x[n]+(w+1)[n]*(x+1)[n]+(w+2)[n]*(x+2)[n]+(w+3)[n]*(x+3)[n]; + } + switch(n) + { + case -1:y+=w[2]*x[2]; + case -2:y+=w[1]*x[1]; + case -3:y+=w[0]*x[0]; + } return y; } +#if HAVE_MMX +/* SIMD implementation of FIR filter on x86 */ +inline FLOAT_TYPE af_filter_fir_sse(register unsigned int n, const FLOAT_TYPE* w, + const FLOAT_TYPE* x) +{ + FLOAT_TYPE y = 0.0; // Output + __asm__( + "subl $8, %%"REG_c" \n\t" + "jl 2f \n\t" + "xorps %%xmm4, %%xmm4 \n\t" + ".align (1<<4) \n\t" + "1: \n\t" + "movups (%%"REG_a",%%"REG_c",4), %%xmm0 \n\t" + "movups (%%"REG_d",%%"REG_c",4), %%xmm1 \n\t" + "movups 16(%%"REG_a",%%"REG_c",4), %%xmm2 \n\t" + "movups 16(%%"REG_d",%%"REG_c",4), %%xmm3 \n\t" + "mulps %%xmm1, %%xmm0 \n\t" + "mulps %%xmm3, %%xmm2 \n\t" + "addps %%xmm2, %%xmm4 \n\t" + "addps %%xmm0, %%xmm4 \n\t" + "subl $8, %%"REG_c" \n\t" + "jge 1b \n\t" + "movhlps %%xmm4, %%xmm0 \n\t" + "addps %%xmm0, %%xmm4 \n\t" + "movaps %%xmm4, %%xmm0 \n\t" + "shufps $0x01, %%xmm4, %%xmm4 \n\t" + "addps %%xmm0, %%xmm4 \n\t" + "movss %%xmm4, %4 \n\t" + "2: \n\t" + :"=c"(n) + :"a"(w),"d"(x),"c"(n),"m"(y) + :"memory" + ); + switch(n) + { + case -1:y+=w[6]*x[6]; + case -2:y+=w[5]*x[5]; + case -3:y+=w[4]*x[4]; + case -4:y+=w[3]*x[3]; + case -5:y+=w[2]*x[2]; + case -6:y+=w[1]*x[1]; + case -7:y+=w[0]*x[0]; + } + return y; +} + +inline FLOAT_TYPE af_filter_fir_sse3(register unsigned int n, const FLOAT_TYPE* w, + const FLOAT_TYPE* x) +{ + register FLOAT_TYPE y = 0.0; // Output + __asm__( + "subl $8, %%"REG_c" \n\t" + "jl 2f \n\t" + "xorps %%xmm4, %%xmm4 \n\t" + ".align (1<<4) \n\t" + "1: \n\t" + "lddqu (%%"REG_a",%%"REG_c",4), %%xmm0 \n\t" + "lddqu (%%"REG_d",%%"REG_c",4), %%xmm1 \n\t" + "lddqu 16(%%"REG_a",%%"REG_c",4), %%xmm2 \n\t" + "lddqu 16(%%"REG_d",%%"REG_c",4), %%xmm3 \n\t" + "mulps %%xmm1, %%xmm0 \n\t" + "mulps %%xmm3, %%xmm2 \n\t" + "addps %%xmm2, %%xmm4 \n\t" + "addps %%xmm0, %%xmm4 \n\t" + "subl $8, %%"REG_c" \n\t" + "jge 1b \n\t" + "haddps %%xmm4, %%xmm4 \n\t" + "haddps %%xmm4, %%xmm4 \n\t" + "movd %%xmm4, %0 \n\t" + "2: \n\t" + :"=r"(y),"=c"(n) + :"a"(w),"d"(x),"c"(n) + ); + switch(n) + { + case -1:y+=w[6]*x[6]; + case -2:y+=w[5]*x[5]; + case -3:y+=w[4]*x[4]; + case -4:y+=w[3]*x[3]; + case -5:y+=w[2]*x[2]; + case -6:y+=w[1]*x[1]; + case -7:y+=w[0]*x[0]; + } + return y; +} + +inline FLOAT_TYPE af_filter_fir_3dnow(register unsigned int n, const FLOAT_TYPE* w, + const FLOAT_TYPE* x) +{ + register FLOAT_TYPE y = 0.0; // Output + __asm__( + "subl $4, %%"REG_c" \n\t" + "jl 2f \n\t" + "pxor %%mm4, %%mm4 \n\t" + ".align (1<<4) \n\t" + "1: \n\t" + "movq (%%"REG_a",%%"REG_c",4), %%mm0 \n\t" + "movq (%%"REG_d",%%"REG_c",4), %%mm1 \n\t" + "movq 8(%%"REG_a",%%"REG_c",4), %%mm2 \n\t" + "movq 8(%%"REG_d",%%"REG_c",4), %%mm3 \n\t" + "pfmul %%mm1, %%mm0 \n\t" + "pfmul %%mm3, %%mm2 \n\t" + "pfadd %%mm2, %%mm4 \n\t" + "pfadd %%mm0, %%mm4 \n\t" + "subl $4, %%"REG_c" \n\t" + "jge 1b \n\t" + "pfacc %%mm4, %%mm4 \n\t" + "movd %%mm4, %0 \n\t" + "femms \n\t" + "2: \n\t" + :"=r"(y),"=c"(n) + :"a"(w),"d"(x),"c"(n) + ); + switch(n) + { + case -1:y+=w[2]*x[2]; + case -2:y+=w[1]*x[1]; + case -3:y+=w[0]*x[0]; + } + return y; +} +#endif + /* C implementation of parallel FIR filter y(k)=w(k) * x(k) (where * denotes convolution) n number of filter taps, where mod(n,4)==0 Index: libaf/filter.h =================================================================== --- libaf/filter.h (revision 28934) +++ libaf/filter.h (working copy) @@ -56,6 +56,12 @@ // Exported functions FLOAT_TYPE af_filter_fir(unsigned int n, const FLOAT_TYPE* w, const FLOAT_TYPE* x); +#if HAVE_MMX +FLOAT_TYPE af_filter_fir_sse(unsigned int n, const FLOAT_TYPE* w, const FLOAT_TYPE* x); +FLOAT_TYPE af_filter_fir_sse3(unsigned int n, const FLOAT_TYPE* w, const FLOAT_TYPE* x); +FLOAT_TYPE af_filter_fir_3dnow(unsigned int n, const FLOAT_TYPE* w, const FLOAT_TYPE* x); +#endif + FLOAT_TYPE* af_filter_pfir(unsigned int n, unsigned int k, unsigned int xi, const FLOAT_TYPE** w, const FLOAT_TYPE** x, FLOAT_TYPE* y, Regards, Zhou Zongyi

2 1

Re: [MPlayer-dev-eng] [PATCH] vo_macosx: process events with shared_buffer (and patch roundup)
by Michael Guntsche 02 Jun '09

02 Jun '09

Hi, looking at the patch Adrian posted I wondered if we even NEED to forward the events to Nsapp anyway, since normally in the case of running with a shared_buffer the GUI application takes care of parsing all the events anyway. Attached is a far more easier patch that works for me. It is pretty similar to Adrian's the only difference is that it just discards any events that it gets. I tested this with mplayerosx.extented and did not see any problems. Kind regards, Michael

5 22