-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Please ignore this message if you haven't already :) I found I had
screwed-up ce libraries as you noted while fighting with a dsplink
issue.

Thanks,
Caglar

Yusuf Caglar AKYUZ wrote:
> Ring, Chris wrote:
>> 0x80008018 is RMS_EINVPROT (EINVPROT == Error, invalid protocol), found in 
>> ti/sdo/ce/rms.h.  This error occurs if the stubs (ARM-side) and skeletons 
>> (DSP-side) don't "speak the same protocol".  That is, the marshalling 
>> protocol of arguments in the stubs doesn't match the unmarshalling protocol 
>> of the arguments in the skeletons.
> 
>> In short, the version of Codec Engine built into the DSP-side doesn't match 
>> the version built into the ARM-side.
> 
> 
> This explains the weird behavior here. However, I can not find the
> source. I double checked everything, and still looking for what I'm
> doing wrong.
> 
> One issue I should mention is I'm creating a CE shared library and
> using it in my applications. I stole this from
> video_copy/dualcpu_seperateconfig_dll. In that example it mentions
> using dlopen and friends. Instead of this, I'm linking *so with ld.
> I hope I'm not being too smart at this.
> 
>> During startup, when CE_DEBUG=2 is set, the versions of components in the 
>> system should be displayed, so you should be able to compare which versions 
>> were used on each side.
> 
> 
> I'm lost at version numbers. I ran ce sanity check application(out
> of the box without re-building) as:
> 
> CE_DEBUG=2 ./app.out in.dat out.dat | grep dsplink
> @0,414,540us: [+4 T:0x4001cf40] OG -     package dsplink.gpp
> (/db/atree/library/trees/ce-j10x/src/codec_engine_2_21/cetools/packages/dsplink/gpp/)
> [5,0,0,1224878287000]
> @0,415,347us: [+4 T:0x4001cf40] OG -     package
> ti.sdo.ce.ipc.dsplink
> (/db/atree/library/trees/ce-j10x/src/codec_engine_2_21/packages/ti/sdo/ce/ipc/dsplink/)
> [2,0,1,1226392278611]
> [DSP] @0,026,140tk: [+4 T:0x8fa4794c] OG -     package
> ti.sdo.ce.ipc.dsplink.dsp
> (/db/atree/library/trees/ce-j10x/src/codec_engine_2_21/packages/ti/sdo/ce/ipc/dsplink/dsp/)
> [2,0,1,1226392278316]
> [DSP] @0,026,532tk: [+4 T:0x8fa4794c] OG -     package dsplink.dsp
> (/db/atree/library/trees/ce-j10x/src/codec_engine_2_21/cetools/packages/dsplink/dsp/)
> [1,4,0,1224878295000]
> 
> Then I ran my own application as:
> 
> LD_LIBRARY_PATH=. CE_DEBUG=3 ./begtester -a test_mpeg4 | grep dsplink
> @0,744,829us: [+4 T:0x4001d310 S:0xbe952b9c] OG -     package
> dsplink.gpp
> (/home/caglar/myfs/work/common/ti/codec_engine_2_21/cetools/packages/dsplink/gpp/)
> [5,0,0,1224878287000]
> @0,752,839us: [+4 T:0x4001d310 S:0xbe952b9c] OG -     package
> ti.sdo.ce.ipc.dsplink
> (/home/caglar/myfs/work/common/ti/codec_engine_2_21/packages/ti/sdo/ce/ipc/dsplink/)
> [2,0,1,1226392278611]
> [DSP] @0,058,051tk: [+4 T:0x8b80006c S:0x8b80ffe4] OG -     package
> ti.sdo.ce.ipc.dsplink.dsp
> (/home/caglar/myfs/work/common/ti/codec_engine_2_21/packages/ti/sdo/ce/ipc/dsplink/dsp/)
> [2,0,1,1226392278316]
> [DSP] @0,058,376tk: [+4 T:0x8b80006c S:0x8b80ffe4] OG -     package
> dsplink.dsp
> (/home/caglar/myfs/work/common/ti/codec_engine_2_21/cetools/packages/dsplink/dsp/)
> [1,4,0,1224878295000]
> 
> I don't see what's wrong here. Anyway, I keep digging. You're
> explanations are pointing us* to the right direction as always.
> Thanks for your time.
> 
> Caglar
> 
> (*) People who are directed to dead ends like grepping the source of
> dsplink for 80008018 error!!!
> 
>> Chris 
> 

_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org

iEYEARECAAYFAkm1cU4ACgkQ/nL+S5dojegP+wCfYlg2tdLD4Azxzlfe4uCHyCZU
ShcAoLC0VXGm3sPeBkJ44ycOHqCYTWOd
=1JNm
-----END PGP SIGNATURE-----

_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to