x264 Release 745 problem!

[audyovydeo]

from 749 on the versioning is correct

Hello ?
from 751 on the versioning is NOT correct: see my post.

I doublechecked file sizes only to check whether the 751 executable somehow got distributed across the 75x download branches. That's obviously not the case. So something is amiss with the versioning script re:749.

cheers
audyovydeo

[bob0r]

audyovydeo:
Did you compile the .exe files yourself or are you refering to the x264.nl builds?
The x264.nl .exe files indeed have problems with updating the version.

The 757 modified build i just posted does show the correct version, so somewhere along the line when using git pull and re-compile something goes wrong.

Reported to pengvado.

However for 757 ill start with a fresh dir but keep a copy of the messed version dir.

[Sharktooth]

where did you got those builds?

[audyovydeo]

bob0r, Sharktooth

I *only* & always use builds from x264.nl.

cheers
a/v

[gruntster]

The shared lib doesn't build with r757 using MinGW:

gcc -shared -o libx264.so.58 common/mc.o common/predict.o common/pixel.o common/macroblock.o common/frame.o common/dct.o common/cpu.o common/cabac.o common/common.o common/mdate.o common/set.o common/quant.o encoder/analyse.o encoder/me.o encoder/ratecontrol.o encoder/set.o encoder/macroblock.o encoder/cabac.o encoder/cavlc.o encoder/encoder.o encoder/eval.o common/x86/mc-c.o common/x86/predict-c.o common/x86/dct-a.o common/x86/deblock-a.o common/x86/mc-a.o common/x86/mc-a2.o common/x86/pixel-a.o common/x86/predict-a.o common/x86/quant-a.o common/x86/sad-a.o common/x86/cpu-32.o common/x86/dct-32.o common/x86/pixel-32.o -Wl,-soname,libx264.so.58 -lvfw32 -s
Cannot export x264_pixel_ads_mvs: symbol not found
collect2: ld returned 1 exit status
make: *** [libx264.so.58] Error 1

[bob0r]

Also reported to pengvado.


x264.nl now has a proper version .exe online for 757, let's see how 758+ will go :)

[LoRd_MuldeR]

x264.nl now has a proper version .exe online for 757, let's see how 758+ will go :)

Could you also provide a .dll file, so we can test it with Avidemux?
Also a VAQ version would be very much appreciated.

:thanks:

[bob0r]

x264.nl script now will rerun configure each time a new update is detected, thanks for reporting the version issue.

a shared lib (.dll) file will be compiled when compiling is fixed.

[audyovydeo]

x264.nl script now will rerun configure each time a new update is detected, thanks for reporting the version issue.

a shared lib (.dll) file will be compiled when compiling is fixed.

Thanks.

Also happy to report that r757 is around 2% faster than r745 on my test clip, using High@L3.1 settings.

cheers
a/v

[LoRd_MuldeR]

a shared lib (.dll) file will be compiled when compiling is fixed.

Thanks. Looking forward... :)

[bob0r]

Could you also provide a .dll file, so we can test it with Avidemux?
Also a VAQ version would be very much appreciated.

:thanks:


http://files.x264.nl/libx264.0.58.763.modified.dll

compiled make fprofiled, with all patches.

[LoRd_MuldeR]

http://files.x264.nl/libx264.0.58.763.modified.dll

compiled make fprofiled, with all patches.

Thanks a lot! Gonna test it out :)

[gruntster]

I'm still getting the crash in r763.


Program received signal SIGSEGV, Segmentation fault.
#0 0x024823c8 in predict_8x8_ddl_sse2 ()
from c:\dev\avidemux_2.4_build\libx264.dll
#1 0x0243d5e6 in x264_slicetype_mb_cost (h=0x6bf4210, a=0x22b528,
frames=0x22b4e0, p0=0, p1=0, b=0, dist_scale_factor=128)
at encoder/slicetype.c:226


Dump of assembler code from 0x24823a8 to 0x24823e8:
0x024823a8 <predict_8x8_ddr_mmxext+216>: or %cl,(%edi)
0x024823aa <predict_8x8_ddr_mmxext+218>: jmp 0x248236e <predict_8x8_ddr_mmxext+158>
0x024823ac <predict_8x8_ddr_mmxext+220>: movq %mm0,0x20(%eax)
0x024823b0 <predict_8x8_ddr_mmxext+224>: psrlq $0x8,%mm0
0x024823b4 <predict_8x8_ddr_mmxext+228>: psllq $0x38,%mm1
0x024823b8 <predict_8x8_ddr_mmxext+232>: por %mm1,%mm0
0x024823bb <predict_8x8_ddr_mmxext+235>: movq %mm0,(%eax)
0x024823be <predict_8x8_ddr_mmxext+238>: ret
0x024823bf <predict_8x8_ddr_mmxext+239>: nop
0x024823c0 <predict_8x8_ddl_sse2+0>: mov 0x4(%esp),%eax
0x024823c4 <predict_8x8_ddl_sse2+4>: mov 0x8(%esp),%ecx
0x024823c8 <predict_8x8_ddl_sse2+8>: movdqa 0x10(%ecx),%xmm3
0x024823cd <predict_8x8_ddl_sse2+13>: movdqu 0x11(%ecx),%xmm2
0x024823d2 <predict_8x8_ddl_sse2+18>: movdqa %xmm3,%xmm1
0x024823d6 <predict_8x8_ddl_sse2+22>: pslldq $0x1,%xmm1
0x024823db <predict_8x8_ddl_sse2+27>: movdqa %xmm1,%xmm4
0x024823df <predict_8x8_ddl_sse2+31>: pavgb %xmm2,%xmm1
0x024823e3 <predict_8x8_ddl_sse2+35>: pxor %xmm4,%xmm2
0x024823e7 <predict_8x8_ddl_sse2+39>: movdqa %xmm3,%xmm0