This test can't run properly because it already defines
an IPM console sender, and instantiating the dummy one
prevents any messages from being forwarded to the x86 side.
Issue: ZEP-708
Change-Id: Ib13c5df5db67f3d9fde960f8e5cda354c60efae1
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
A previous re-work of IRQ priorities was led astray by an incorrect
comment. Priority level 1 is not a non-maskable interrupt priority.
In addition, zero latency IRQs are not implemented on ARC.
Timer driver now doesn't specify IRQ_ZERO_LATENCY (as that wouldn't be
correct) and its IRQ priority is now tunable in Kconfig. The default is 0.
IPM driver on both ARC and x86 side were being configured with hard-coded
priority of 2, which wasn't valid for ARC and caused an assertion failure.
The priority level is now tunable with Kconfig and defaults to 1 for ARC.
Issue: ZEP-693
Change-Id: If76dbfee214be7630d787be0bce4549a1ecbcb5b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The QEMU bug in the SDK has been fixed.
This reverts commit eb63710960.
Change-Id: Ic405c9df0f7aebb2a9d20180c7b0c04889669610
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Per spec document, RXEN bit corresponds to bit 2
of the ECON1 register instead of bit 1.
Change-Id: I1bbfd048be248575a558679985ba9daadc43df0e
Signed-off-by: Juan Manuel Cruz Alcaraz <juan.m.cruz.alcaraz@intel.com>
The rx fiber call contiki IP stack functions to register
received frames. This interaction requires more stack memory
in the rx fiber stack to avoid memory corruption.
Change-Id: I106339e7dd1ca9282426adfb54590e34297755b3
Signed-off-by: Juan Manuel Cruz Alcaraz <juan.m.cruz.alcaraz@intel.com>
SPI bus is a shared resource between tx and rx processes.
The access to it must be synchronized to allow a rx process to
happen even when a tx process is taking place.
ECON1 register must be saved by the rx process and recovered
at the end to ensure that the tx process will continue operating
in the same register bank.
Change-Id: Ie9358bf02bef8ddb5bdf76c8847e998a627e5395
Signed-off-by: Juan Manuel Cruz Alcaraz <juan.m.cruz.alcaraz@intel.com>
Transmision status is polled. Remove rx interruption enabling and
clearance.
Change-Id: I2893f89533adef5f77c711a36b5c1bbe21c48a23
Signed-off-by: Juan Manuel Cruz Alcaraz <juan.m.cruz.alcaraz@intel.com>
Even values in ERXRDPT register may corrupt the receive buffer.
Set ERXRDPT value one address behind the next value.
If it is zero, set the value to the reception buffer end address.
Change-Id: I7150e5190c5bfb7bfafbff598eea1c26959643d0
Signed-off-by: Juan Manuel Cruz Alcaraz <juan.m.cruz.alcaraz@intel.com>
These console messages are for debugging and if sent too
rapidly, could be discarded. This was causing false positives on
test cases which rely on console output to determine success or
failure.
Issue: ZEP-704
Change-Id: I5a86c761311cde90f295cd2e65f2e70608f875fa
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The Receive Packet Pending Interrupt Flag (EIR.PKTIF) does not
reliably/accurately report the status of pending packets.
Reception fails intermittently. The EPKTCNT register is used instead.
Change-Id: I599c3b4df1b74d769dd7104e5c6e253d9cf63bd4
Signed-off-by: Juan Manuel Cruz Alcaraz <juan.m.cruz.alcaraz@intel.com>
The (currently) unused ipm_device field wasn't initialized.
Issue: ZEP-704
Change-Id: Icbeb50031c52d2513a1417f83a844e25b43429a7
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The implementation was using different typedefs than the prototypes
in the header.
Issue: ZEP-695
Change-Id: Ia369daafe1bce39830535d184bd4388b13c986b4
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Update the built-in QMSI build to add qm_flash.c into the default
build due to that the clk.c from QMSI 1.1.0 added a new dependency
on qm_flash file.
(https://gerrit.zephyrproject.org/r/#/c/4000/3/ext/hal/qmsi/drivers/clk.c)
Jira: ZEP-703
Change-Id: I9f64e308cec21477a9b96183dec9ecca37ead373
Signed-off-by: Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
At some point these were moved to include/drivers/
Change-Id: I5130a4d48b3bbff6ab03b6103bec0f53d88deb7b
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
0 should be fine; no need to make this an FIRQ or non-maskable
Change-Id: Ifdf89a72e4864a2c2bbd83752cd814e2cb9aa16e
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Replaced the term "platform" with "board" or "SoC" depending
on context as per, ZEP-534.
Change-Id: I14c13d4eed429fe6e41e2221d6ff6afe97e942eb
Signed-off-by: Evan Couzens <evanx.couzens@intel.com>
Disable tx fifo empty interrupt at the completion of a receive
transfer. If the API caller does not send "STOP" at the end of
a receive transfer, unnecessary tx fifo empty interrupt will
come one by one endlessly.
Jira: ZEP-556
Change-Id: I54739459dfab81c28713f6a3c77789e7903bb7f9
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Incase of slip or ethenet use, if these is too much incoming data
current buffer count is not sufficient. Noticed with TCP client
and server example. So increasing it to avoid not enough Rx buffers
issue.
Change-Id: I6eef51c28c7c4751955aa4403122bf17d55ce12c
Signed-off-by: Ravi kumar Veeramally <ravikumar.veeramally@linux.intel.com>
We are also seeing qemu failures in the synchronous wakeup tests,
where the fiber should be resuming instantly but in practice sees a
one-tick delay due to the emulation environment not being
deterministic. Allow one tick of slop in those too.
Change-Id: Idab7c45ea0b10bd955b90a98d3884b5fe0571187
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Fix i2c failure while multiple slaves are connected to the
same i2c interface. Higher load on the i2c bus makes i2c
signal less nice. The requirement for signal ramp up and
ramp down time needs to be loosened.
Jira: ZEP-646
Change-Id: If698d7ba741788857694f166ecc602c2a94cb12e
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
The app has updated the text it prints (which we were looking for) so
this was causing a false negative.
Change-Id: I3223a0dec8e44b0f89caf73626ffff82d74ab24d
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
The per-platform blocks were just specifying defaults and didn't
need to exist.
Change-Id: I9f4c7db4cd16417d5a2c172c3f850f7758dcbd00
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The GPIO and UART drivers were failing to build due to some
missing soc.h defines.
Change-Id: I6811e3699449da0a61ccc8376a8e11b96ad7a4e5
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The eventhandler gets called with NULL buf, but it can't handle it. Add
a NULL check to prevent crashing.
Change-Id: Id16c2aa093a145e0442dfc61dd0b6e1b81b701f5
Signed-off-by: Jaakko Hannikainen <jaakko.hannikainen@intel.com>
README files are updated to include information about the
warnings displayed by the compiler.
Applications affected: DNS, NATS and MQTT.
Jira: ZEP-681
Change-Id: I482c3bbf37c5d9af3bb43b7ef5e58957bac2852a
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
We map legacy and SSP link keys to equivalent security levels so there
is no need for storing if key is legacy. This flag was never used
anyway.
Change-Id: I6867f43a1ed132f67a564de34436cf83b922a138
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
There is no need to ifdef type definitions and whole file is already
under (CONFIG_BLUETOOTH_SMP) || defined(CONFIG_BLUETOOTH_BREDR).
This improves file readability.
Change-Id: I9bcceb01023f75e016ad71873f87b3393a09fe26
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
Add a Kconfig option to select between the Hard and Soft Float ABIs. We
also default to the Hard Float ABI as this is what older SDK versions
supported.
JIRA: ZEP-555
Change-Id: I2180c98cd7556ab49f5ca9b46b31add2c11bd07b
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
The data sample registers are ordered as X, Z, Y (not X, Y, Z), so the
driver also needs to extract the sample components in this order.
Jira: ZEP-679
Change-Id: I317994bb545539b049e2ebf3d2af94e63c7d2511
Signed-off-by: Bogdan Davidoaia <bogdan.m.davidoaia@intel.com>
This was merged prematurely.
This reverts commit 758355779c.
Change-Id: I19d27df0b9ea9acf49382d27b5ef6ce49b915542
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
The parent commit for this was merged prematurely.
This reverts commit d38208b977.
Change-Id: I4f05f303eac7bdf176fb0522e30eabb839602b25
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Disable tx fifo empty interrupt at the completion of a receive
transfer. If the API caller does not send "STOP" at the end of
a receive transfer, unnecessary tx fifo empty interrupt will
come one by one endlessly.
Jira: ZEP-556
Change-Id: I311aba843e28685e8cca36733ac3f1e09b70d900
Signed-off-by: Baohong Liu <baohong.liu@intel.com>