Hi Artem, On Tue, Feb 4, 2014 at 10:36 PM, Artem Rakhov <arakhov@xxxxxxxxxx> wrote: > Hi Luiz, > > We are currently working on the Bluetooth SIG qualification, and we are > failing AVDTP test case "TP/SIG/SMG/BI-28-C". I've found your test in the > BlueZ source (unit/test-avdtp.c), which seems to contain implementation of > this test case. Could you please give us some clue on how to use it to do > the actual testing and to produce a log? Interesting, apparently the test spec suggests resetting the signal identifier to 0x00: Pass verdict The lower tester receives the AVDTP signaling message with the fields: Transaction_label = Transaction LowerTester Other 10 bits = All set to '0'. While the spec suggests to keep the original signal identifier that is being rejected see Table 8.42: General Reject response format. We are following the spec text in this respect, thus our unit test checks for the original signal identifier skipping the RFA bits which is what our AVDTP implementation does when responding when receiving such command, either way it is easy to fix but we probably need to file an errata so the spec and test spec are more consistent regarding general reject response. For the purpose of qualification I think we need the full trace up to HCI so the unit test are not enough, but there are plans to do that via emulator. -- Luiz Augusto von Dentz -- To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
