With SOF secondary cores are booted later at run-time instead
of the traditional simultaneous booting of all the cores.
Adjust arch_start_cpu() to make that possible.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
smp.c only has to be built if CONFIG_SMP is enabled. Remove
preprocessor checks from the file itself and update cmake rules
instead.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
To disable the IDC interrupt on the interrupt controller a bit
must be set in the MSD register instead of clearing the bit in
the MCD register, which has no effect.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Currently cavs_idc_smp_init() is called from a system
initialisation sequence, which only runs on the main CPU.
However, it must also run on secondary CPUs if those are
powered on later instead of simultaneously with the main
CPU.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Currently P4WQ supports queues with sets of user-provided
worked threads of arbitrary numbers. These threads are started
immediately upon initialisation.
This patch adds support for 3 more thread implementation options:
1. queue per thread. It adds a K_P4WQ_ARRAY_DEFINE() macro which
initialises an array of queues and threads of the same number.
These threads are then uniquely assigned to respective queues.
2. delayed start. With this option threads aren't started
immediately upon queue initialisation. Instead a new function
k_p4wq_enable_static_thread() has to be called to enable those
threads individually.
3. queue per CPU. With this option the user can assign CPU masks
to threads when calling k_p4wq_enable_static_thread().
Otherwise the cpu_mask parameter to that function is ignored.
Currently enabling this option implies option 2 above. Also so
far to enable queues per CPU the user has to use
K_P4WQ_ARRAY_DEFINE(), which means this option also implies 1
above, but both these restrictions can be relaxed in the
future if required.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Work items in P4WQ currently belong to the user before submission
and after exit from the handler, therefore, unless the handler
re-submits the item, accessing it in p4wq_loop() in such cases
is racy. To fix this we re-define work item ownership. Now the
item belongs to the P4WQ core until the user calls
k_p4wq_wait(). If the work item has its .sync flag set, the
function will sleep until the handler completes processing the
work item or until the timeout expires. If .sync isn't set and
the handler hasn't processed the item yet, the function returns
-EBUSY.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
When SMP is disabled, the SMP initialisation level is
undefined, therefore a different level must be used.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
When secondary cores are booted, they use the dummy thread and
the IRQ stack until they switch over to a real thread. Therefore
dummy threads shouldn't be skipped when cohering outgoing thread
stack, only threads with zero stack size should be skipped.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Usually Zephyr boots all secondary CPUs as a part of system
boot. Some applications however need an ability to boot on
the main CPU only and enable secondary CPUs selectively at
run-time. Add a Kconfig option to support this behaviour.
When booting CPUs on demand applications also need helpers
to initialise a dummy thread and begin threaded execution
on those CPUs, add two such helpers.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
irq_enable() should be called with the composite IRQ code as its
argument, not just the Xtensa proper part of it.
Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Support this when debugging also.
Tweak the style for brevity also while we're here by propertizing the
supports_nogui method, etc.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This makes it easier for runners to check if the binary has thread
info support turned on, allowing automatic configuration of the
underlying tool to support threads, if possible.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Set options that are definitely true or false to True or False in the
options dict. Add a 'getboolean' method that also allows a fallback to
False in case the option is not mentioned in .config due to unmet
dependencies. This allows calling code to just ask about the option
they are interested in, even if the .config file doesn't mention the
option at all.
Propagate this to users within the runners package and 'west sign',
taking advantage of the new build_conf property.
Rename the 'bcfg' internal variable in sign.py to 'build_conf' to
match other source files that use BuildConfiguration instances, to
make it easier to grep for users.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Instead of mocking out the BuildConfiguration class, just create its
input file and let the real class do the work.
This in turn exposes a bug in the way the board name is being pulled
out of the BuildConfiguration, which we fix to keep the tests passing.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This is really verbose, and I doubt anyone cares unless there is a
problem. Keep it around when run as 'west -v flash' to allow for
debugging, though.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Defer loading .config until we really need it, when we are flashing a
binary. Pre-emptively loading it is wasted effort if we're flashing a
.hex, which has been the default behavior when possible since
dcaabb860f ("west: runners: jlink: prefer .hex over .bin").
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Put it all in one log message rather than splitting it up.
This makes it look cleaner now that each log message is prefixed with
'runners.link:'.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
JLink versions like 'V7.0a' do not conform to PEP 440 version
conventions; the 'a' part is used by PEP 440 compliant versions for
alphas. It gets parsed to a legacy type by the packaging library,
which always is treated as a lower value when compared with a
conforming version string.
To fix, get the version from the shared library distributed with the
JLink tools. This has the side benefit of making the code work on
Windows. That's merely a nice to have for -nogui 1 detection for now,
but will be essential in the next commit.
Reported-by: Jake Mercer <jake.mercer@civica.co.uk>
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This will be used to deal with the Segger shared library in a portable
way in the jlink runner.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Make it easier to get a BuildConfiguration from runner code.
Stash the result so it only has to be computed once.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
This option has existed since the beginning of the runners package,
which greatly predates the way DT is used in zephyr right now. It
never really worked the way I wanted it to but it's too much work to
fix it now. Try to improve the help a bit at least while I'm looking
at it again.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Copy a fix from test_nrfjprog.py to the other runner test suites. The
current code will enter an infinite recursion if you hit the path
where os.path.isfile is called, since it's been patched to
os_path_isfile_patch in the calling context. The fix is to cache the
'real' version in the parent scope and call it directly as a fallback.
Signed-off-by: Martí Bolívar <marti.bolivar@nordicsemi.no>
Fix buffer leak in L2CAP and SMP when attempting to send over a
a connection that has been disconnected.
Both L2CAP and SMP now have ownership of the buffer in case of failure
to send it and need to unref it in case of send failure.
Change the return type of bt_l2cap_send since the user of this function
has to handle releasing of the buffer when this fails.
Regression from: a77f809faa47ff9922692d2b4eab33062b8a2b4d
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Remove the comment about bt_l2cap_send_cb not being able to fail when
called from RX thread. It will fail in the case where the connection
has been disconnected.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Improved the documentation for the ISO data paths, as well
as making the bt_iso_setup_data_path function a bit more
clear.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
Since we are only setting up a single data path for
broadcasters (either receive or transmit), we should
only remove that same path.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
The BIGinfo reports were almost identical to the
periodic advertising reports when printet. Updated to
mark the start of the line with BIGinfo.
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
When setting the iso data path for a BIS, it shall only be set for
one direction (depending on whether we are a broadcaster or
a receiver).
Signed-off-by: Emil Gydesen <emil.gydesen@nordicsemi.no>
For non-specified archs, including those out-of-tree, the possibility to
use a specific implementation has been reintroduced.
CONFIG_ATOMIC_OPERATIONS_ARCH must be selected to utilize this.
Signed-off-by: Tommie Skriver <tosk@demant.com>
This adds FPU sharing support with a lazy context switching algorithm.
Every thread is allowed to use FPU/SIMD registers. In fact, the compiler
may insert FPU reg accesses in anycontext to optimize even non-FP code
unless the -mgeneral-regs-only compiler flag is used, but Zephyr
currently doesn't support such a build.
It is therefore possible to do FP access in IRS as well with this patch
although IRQs are then disabled to prevent nested IRQs in such cases.
Because the thread object grows in size, some tests have to be adjusted.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
Add the exception depth count to tpidrro_el0 and make it available
through the arch_exception_depth() accessor.
The IN_EL0 flag is now updated unconditionally even if userspace is
not configured. Doing otherwise made the code rather hairy and
I doubt the overhead is measurable.
Signed-off-by: Nicolas Pitre <npitre@baylibre.com>
