Enhance the linker script so that all statically-defined kernel
objects of a given type are co-located.
Note 1: This capability is needed so that static kernel objects
can be added to the associated object tracing list. (Some kernel
object types are already co-located to permit the kernel to do
other initialzation on statically-defined objects.)
Note 2: A follow-up commit is needed to ensure statically-
defined kernel objects are placed in the sections referenced
in the linker script.
Note 3: The legacy script info remains for now to allow
applications using the microkernel or nanokernel to continue
to build.
Change-Id: I40d831f9e183fb121c950e30fa8298b6d529375b
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Microkernel and nanokernel versions of the test now use distinct
sets of source files. This eliminates the dependency on the
MICROKERNEL and NANOKERNEL configuration options, which the
unified kernel doesn't set correctly in all cases.
Change-Id: I59e1d4a73265a433f526d030d2fd04d4b8de1973
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Some HW needs longet timeouts for this TC to complete succesfully.
Change-Id: If6cd4ad93b6e15293f1f8d8e606e159271f0e4d1
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
If application prompt handler returns NULL curently selected module
prompt should also be checked before of printing default shell prompt.
Change-Id: Iba9c191041951c31036766b1cc8070b040655445
Signed-off-by: Szymon Janc <szymon.janc@codecoup.pl>
ztest execute syncronization depends on sem take and give
between TC. Current issues are
*) test execution go to next one before current TC finish
*) one TC failure will block whole testing.
Fix this issue by move k_sem_take from init_testing to
run_test and add k_sem_give in run_test_fail.
Meanwhile, set init sem value range from 0 to 1
Issue: ZEP-1164
Change-Id: I7b2d38501d0965455a71863a4729ee81472a63ec
Signed-off-by: jing wang <jing.j.wang@intel.com>
As we transition to the unified kernel, we need to check on there
being enough memory in said case (nano kernel always had it based on
higher granularity filtering with 'kernel = nano').
Change-Id: Ie21aea2773feb9337d8d34dcf45858e4df879469
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
When running in real HW, testcases take more time to execute than the
default 60s. 200s seems to work for all HW types as of this commit
(FRDM k64f, Arduino Due, Arduino 101 / x86, EMSK 9d).
Change-Id: I663a67161bd7b4c349a483efb8e9cbf14c09421c
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Several legacy API macros had the format:
#define some_legacy_api(<parameters>) other_api(<same parameters>)
This prevents passing 'legacy_api()' as a function pointer. Remove the
parenthesis where possible and instead use the format:
#define some_legacy_api other_api
This allows passing 'legacy_api' as a function pointer by having the
preprocessor resolve it to 'new_api' directly.
Change-Id: Ib680a15a22097727b1dade4b7b4785267c004751
Signed-off-by: Benjamin Walsh <walsh.benj@gmail.com>
Ability to use Zephyr shell by multiple modules simultaneously, each
module for its own usage.
Old shell implementation enabled the user to call only one module
commands, not all of the modules simultaneously.
Change-Id: I0ef8fa2fd190b7490c44fe91d1016363258302c9
Signed-off-by: Yael Avramovich <yael.avramovich@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
In the unified kernel legacy layer, both these functions are
implemented as #defines which can't be put in a struct.
Eventually this driver should be updated to use new APIs, but
this will get it to at least build which is the priority.
Change-Id: Ic5dfb5b85ff3bd0d1646f542340642549a2b2e05
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Not completely ported to V2 APIs yet, but enough that we pull in
all the kernel.h APIs for the microkernel test, bypassing the legacy
layer.
We pull in a lot more than the legacy test did, so results are not
directly comparable.
Change-Id: I35e74bfd1aa15ee59951f68fb7dd8d811619c3f8
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The commit:
commit f25ac092b9
"Makefile: Restructure for multilibs"
breaks external HAL library linking, by overriding, in the
the second Makefile pass, the LIB_INCLUDE_DIR variable
set in the first Makefile pass from ext/hal Makefiles.
This breaks use of the CONFIG_CC3200SDK_LIBRARY=y option in
the cc3200 SDK, and also likely the QMSI ext/hal/qmsi/Makefile,
when CONFIG_QMSI_LIBRARY=y is set.
This commit just changes the '=' to "+=".
Jira: ZEP-1163
Change-Id: I1aaaaee57378dcd01ab6f2ed4799281b2337eea2
Signed-off-by: Gil Pitney <gil.pitney@linaro.org>
Nanokernel part of the latency_measure test is a subset of the
microkekernel tests. For unified kernel there is no reason to
have nanokernel test - for unified kernel we have to run whole
the set of tests.
Change-Id: Ief176fd9d25e7355e3d3697c3bbc1e953b16655c
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
This allows to delete character after cursor.
Change-Id: I9acc5e1edeb688ea4ec36e91aef69ecb5f5f6354
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
This allows to navigate to begining and end of the line easily.
Change-Id: I3a839a0499961d55574a310c65811dd2fe048703
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
The unified kernel doesn't use the KERNEL_ENTRY() mechanism,
it doesn't do kernel calls in a special context. So there's
no reason to test this on the unified kernel and we can omit it.
Change-Id: I2b8aa2495385a39130e9b87c0d0324f2f205a2c8
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Removes reference to nanokernel/microkernel.
Change-Id: Ibac9c85570d7add45c1dbabc524cb5176c5a189d
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Removes references to nanokernel/microkernel and related obsolete
text.
Change-Id: I2b52243db7c580e14d18d39effee7d86428855d8
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Adapt text to reflect unified kernel changes, and also clean up
a few stylistics inconsistencies.
Change-Id: Ife6d97cae20c7feb974f17f54a47da583f1c1907
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Allows event objects to pend signals in a cumulative way using
the semaphore in a non-binary way.
Jira: ZEP-928
Change-Id: I3ce8a075ef89309118596ec5781c15d4f3289d34
Signed-off-by: Peter Mitsis <peter.mitsis@windriver.com>
This adds support for setting the minimun required security level to
l2cap_register command:
l2cap-register <psm> [sec_level]
Note that entering the security level is optional and the default is
BT_SECURITY_LOW(1).
Change-Id: Ib6dea92919357f41fe2f4de3050b9a10a7f553d6
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
This implements the security check for incoming LE CoC and in case the
connection cannot attend the minimun security level set by the server
return a proper error:
> ACL Data RX: Handle 72 flags 0x02 dlen 18
LE L2CAP: LE Connection Request (0x14) ident 1 len 10
PSM: 128 (0x0080)
Source CID: 64
MTU: 672
MPS: 230
Credits: 10
< ACL Data TX: Handle 72 flags 0x00 dlen 18
LE L2CAP: LE Connection Response (0x15) ident 1 len 10
Destination CID: 0
MTU: 0
MPS: 0
Credits: 0
Result: Connection refused - insufficient authentication (0x0005)
Change-Id: Ia8b2873c5d1b154eb9a7c1a49c94578c4ec22f6d
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Even though this is level 0, NONE as a symbolic description is more in
line with the other level names.
Change-Id: Ic9d7c38b928d8744d47e759fae804bbbc08c89bf
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Fix minor style issues with excessive whitespace for alignment and
lack of 'static' for the tx_fiber_stack.
Change-Id: I0c761e94e6a6789a3772b08b0774cd1a5e73c2ef
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
Right now the hci-uart samples doesn't really work anywhere else than
nRF5x, so make that the default configuration. Also rename the old
prj.conf to generic.conf to give a generic starting point for creating
configurations for other targets.
Change-Id: I1c940339b457d0e77c8ddad4b2474fc7e5b12a51
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
So far bt_security_t has completely missed out on security level 0,
i.e. its actual values have been one lower than in the core
specification.
To properly introduce for the new level (which is only applicable for
BR/EDR) add proper tracking for each channel and server, and make the
channels inherit their required level from the respective server.
Change-Id: I9a2384d883017125c2c117880aa6e0ade30520e4
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
If length is already 0 there is no point in calling the handler again
as that already mark the end of the operation.
Change-Id: I212a6d8bdd9aa3d9886099bb7ed3f9d3831bcf1a
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
This uses net_buf_simple_{save/restore} so the same buffer can be reused
if the buffer needs to be resent, also since the responses don't need to
be saved a pool with 1 element is enough while it keeps the code safe
from deadlocking when both request and responses use the same pool.
Change-Id: Ibaa8e7ef39f4b466d5cd4d55874bd609f0a1d67c
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
In 3 of our UART driver implementations the ISR is exclusively
used and enabled for the RX path. The existing logic was
susceptible to a stall situation where a polled out transmission
would be interrupted by a reception and then the ISR code
would loop forever due to the TX event being signalled (although
the interrupt itself was disabled) causing the ISR to keep
looping for an RX interrupt.
Change-Id: Ic379e58b1c974aca3cee37d2d81f12c3726fb160
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
This header file is not intended for applications, rather than for
drivers, so move it to the appropriate place.
Change-Id: I7ff8158418e7e839b538c6cccbb32f3af43c12c6
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
There's no need for this app to use the HCI driver API directly.
Change-Id: Ie5acfbfb09fc3f86c3f29f22fbf7da7db1ff2127
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
If there is a request ongoing it may block responses to be generated
since they were using the same buffer pool, so this introduces a
dedicated pool for responses making the code able to act as both
server and client at same time.
Change-Id: I5fe3e19f9c5c0c2e0dfadedf77b7684f0960572c
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
These options were only needed for a MyNewt-based nRF51 firmware on
these boards (the MyNewt BLE stack is called Nimble, hence the
prj_nimble.conf sample config files). With a Zephyr-based nRF51
firmware these options are no-longer needed, so it's not appropriate
to have them default to enabled. Instead, if they are needed, require
the app-specific configuration to enable them.
Change-Id: Iefbee4d97590af4e11bcedea05fe61f32a147b83
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
These are using old kernel APIs and need to be converted.
Change-Id: I7457270fb5e93e290023756316e82a384712e82b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
As we transition to the unified kernel, we need to check on there
being enough memory in said case (nano kernel always had it based on
higher granularity filtering with 'kernel = nano').
Change-Id: Ic36c352f06be09407cd3aac74a1dfe73bb2fc61d
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
This code is part of the infrastructure, it does not test any features really.
We are wasting cycles on a stub that does basically nothing.
Change-Id: I130f320d1b365abdac118e5c44b80fd866e5f502
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
We have many tests that are being built as stand-alone binaries for no good
reason. Those can be put in one single test to speed up testing and CI
verification.
Currently we do support the following in test_common:
- byteorder
- printk
- intmath
- slist
- atomic
- ring_buf
In addition, the test now uses the ztest framework.
Change-Id: I656ac7f4acf48b7de4ec81c9f5dafc9dea3da911
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
We are building for unified kernel by default. This avoids running the test
twice.
Change-Id: I2951464f448b3019c236ce5943df55c052b427d9
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Tests that do not use any of the legacy kernel APIs can be moved to unified
kernel to avoid double testing.
Change-Id: Ic2353feb23ee20d9d93f5459432d3b3739df8e03
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Move all kernel tests using legacy APIs into tests/legacy to continue testing
old APIs and compatibility. We keep in tests/kernel those tests that do not use
any kernel APIs and generic in nature, those should not be affected by the
unified kernel API change.
In tests/kernel we will start adding tests that are unified kernel only. Later
and when deprecation period is over the legacy tests would be dropped.
Change-Id: Icc7d8c7e5f2af65af350b75da3117f72396925f4
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
As unified kernel starts main() application function, remove test
thread.
Remove nanokernel part of the test since their code base is
identical.
Change-Id: I2afb4c0da2d139b159c89186166afdb5d9813d61
Signed-off-by: Dmitriy Korovkin <dmitriy.korovkin@windriver.com>
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
Currently the net/buf test fails on quark_d2000_crb due to RAM
overflow:
region `RAM' overflowed by 2256 bytes
Change-Id: I854960a3c22c7318f5f5e9341ca7aaaf8743d30d
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
Make it run on Qemu only, this test is not intended for HW.
Change-Id: I842d7456f9238221ef5b661749136f4418d5ab16
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
