Fix test bench scripts

[xiulipan]

Add status check and fix type size check.
re-enable the test bench check in Travis

fix #2752

[xiulipan]

@singalsu It seems we got vol host test bench failed

error: loading controls
error: parsing topology
debug: ipc and scheduler initialized
debug: topology parsing start
debug: type: 8, size: 0x70 count: 1 index: 0
debug: type: 5, size: 0x680 count: 6 index: 1
debug: number of DAPM widgets 6
debug: loading widget PCM0P id 11
debug: registered file comp driver
debug: file_new()
debug: loading widget PGA1.0 id 4
debug: registered comp driver for volume
debug: opening shared lib libsof_volume.so
debug: loading widget BUF1.0 id 16
debug: loading widget  id 4
debug: loading widget  id 0
info: Widget type not supported 0
error: loading widget
debug: topology parsing end

[singalsu]

Strange, I'm able to run volume component tests this way:

cd tools/test/audio
octave
process_test('volume', 16, 16, 48e3);

The verdict is fail due to volume ramp no getting scheduled and resulting wrong gain vs. expected but the data flow is OK. Though I have not rebased today. Is your testbench binary up-to-date?

[lgirdwood]

Code looks fine to me, I assume you are saying here that script is now correctly detecting a testbench failure.

[singalsu]

@xiulipan @lgirdwood I had the testbench up-to-date but used old test topologies, maybe two weeks old. After I re-ran today "scripts/build-tools.sh -t" the testbench volume topology parsing failed. I'll start to debug what has happened. Thanks for reporting this!

[lgirdwood]

SOFCI TEST

[singalsu]

The volume volume topology parsing fix is in #2824. I'd assume the PR is in quite good shape now after some review rounds.

[lgirdwood]

SOFCI TEST

[singalsu]

Why are we waiting to merge #2824 and this?

[lgirdwood]

@singalsu can you check the CI output, I'll run it again.

[lgirdwood]

SOFCI TEST

[singalsu]

@lgirdwood The fails above look unrelated. I tested it also locally. With git master from today and this patch applied this run passes now:

$ scripts/host-testbench.sh 
==========================================================
test volume with ./volume_run.sh 16 16 48000 zeros_in.raw volume_out.raw
volume test passed!
volume_out size check passed!
==========================================================
test src with ./src_run.sh 32 32 44100 48000 zeros_in.raw src_out.raw
src test passed!
src_out size check passed!
==========================================================
test eqiir with ./eqiir_run.sh 16 16 48000 zeros_in.raw eqiir_out.raw
eqiir test passed!
eqiir_out size check passed!

[xiulipan]

@lgirdwood @singalsu I re trigger the Travis test check if the testbench can work now.

[lgirdwood]

@xiulipan looks like the CI version of xtensa-buildall.sh is missing some machines from the build. It's an unrelated failure. Jenkins is usual issues.

[xiulipan]

@lgirdwood What is missed here? I do not see any issue in sof-ci/jenkins

[lgirdwood]

@xiulipan it could be a stale result. I've seen this before.