Hi Jake,

OK. I never thought about K2.

Should having another line for K2 similar to ti33x/ti44x and point to the same commit ID - "d0160ca810be30bf2b2e7681f5047933402efb52" fix the problem?
Most of the changes that TI is doing on top of the default GST plugins are related to hardware acceleration. And we don't have the IVAHD on the AM3 and AM4, they are pointing to the default a commit ID that corresponds to the default GST upstream package.

Could you please try the change for K2 and submit a patch if that works? I don't have board or environment to fix that really quick.

Regards
Karthik

On 01-Aug-16 4:22 PM, Stiffler, Jacob wrote:

Karthik, Denys,

 

We are seeing a failure with building the gstreamer1.0-plugins-bad for k2 since upgrading to 1.6.

 

In file included from ../../../../git/gst-libs/gst/drm/gstdrmallocator.c:32:0:

../../../../git/gst-libs/gst/drm/gstdrmallocator.h:39:21: fatal error: xf86drm.h: No such file or directory

compilation terminated.

Makefile:866: recipe for target 'libgstdrm_1.0_la-gstdrmallocator.lo' failed

make[4]: *** [libgstdrm_1.0_la-gstdrmallocator.lo] Error 1

make[4]: Leaving directory '/home/gtbldadm/processor-sdk-linux-krogoth-build/build-CORTEX_1/arago-tmp-external-linaro-toolchain/work/k2hk_evm-linux-gnueabi/gstreamer1.0-plugins-bad/1.6.3-r20/build/gst-libs/gst/drm'

Makefile:1050: recipe for target 'drm' failed

 

 

This appears to be related to http://git.ti.com/glsdk/gstreamer1-0-plugins-bad/commit/04ee4d353888a210c5ee5792ff402eb8c3d3e6e4. Is there a config option to disable the drm allocator?

 

I am also curious as to why we are using different versions of these plugins for ti33x and ti43x. Should k2 also use that commit? I am referring to the lines in gstreamer1.0-plugins-bad_git.bbappend:

 

  38 SRCREV = "f0694b5fcec997036ef7df4d382b28e833c8ebfd"

  39 SRCREV_ti43x = "d0160ca810be30bf2b2e7681f5047933402efb52"

  40 SRCREV_ti33x = "d0160ca810be30bf2b2e7681f5047933402efb52"

 

 

Thank you,

Jake