This will add basic support for sending bodies of data that exceed the
size of a single UDP packet.
Block-wise transfers are defined in the recently adopted RFC 7959[1].
The RFC defines four options, two negotiating the block transfer, and
two for informing the size of the transfer. Depending whether the packet
is a request or a response, each option is defined as
"control" (informative) or descriptive (describes what's in the
payload).
Change-Id: Ic71275558c4afed0298d20e8712f76d53904f89f
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
The value for CONFIG_SYS_CLOCK_HW_CYCLES_PER_SEC needs to be set
in the SOC, since each SOC will have a different frequency
on the EM Starterkit. The EM7D, for example, has a 30Mhz CPU clock.
It was already being defined in the SOC, but the board setting
would override the SOC.
Change-Id: I4daf3b94f15bad99c0f3c8674a15ad225aa3d274
Signed-off-by: Chuck Jordan <cjordan@synopsys.com>
This refactor is in preparation for making driver API structures
const.
The console driver provides a mechanism to install an input and an
output hook function. These are primarily used by the onboard
gdb-server. The output hook is entirely implemented within the
console driver.
The input hook is partially implemented in the top of the uart driver
and within the console driver. The hook function itself is installed
in the uart API structure, but is invoked only by the console driver.
Installing the hook function directly into the uart API structure
prevents the API structure being const. There are two approaches to
fixing this:
1) Implement setting of the input hook in the same way as
uart_irq_callback_set().
2) Move the input hook entirely to the console driver.
We implement the latter. This approach has two benefits, first it
removes the need for every uart driver to implement the behaviour and
second, the current placement of the callback function in the uart API
seems odd given that the callback is only invoked by the console
driver, never by a uart driver.
Change-Id: I258b312d3055df1c2bdeb896bd4f4f39c40838f7
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
The variable is defined as SYS_LOG_GROVE_LEVEL in Kconfig..
Change-Id: I0f77336df2293694ece71f4f5fccb5283f6dc58c
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Microkernel and nanokernel tests now use customized source code
to eliminate use of MICROKERNEL and NANOKERNEL config options.
Change-Id: Ic3617df34487911af1607ab46f469c5e1212d3f7
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Microkernel and nanokernel tests now use customized source code
to eliminate use of MICROKERNEL and NANOKERNEL config options.
Change-Id: I7f9aff4729a7a257c4a0a0f5939ba0f5525af460
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Enhances the microkernel memory pool test application to
include tests for dynamic memory allocation and freeing
from the heap memory pool using kernel APIs that behave
like malloc() and free().
This enhancement works under both the microkernel and
unified kernel.
Change-Id: Ibc485877ea9d60307edb8f93c54a0b94ebacb017
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Fleshes out the prototype heap memory pool support
to make it fully operational. Noteworthy changes are
listed below:
Tweaks arguments to k_malloc() and k_free() to be more like
malloc() and free(). Similarly, modifies k_free() to take
no action when passed a NULL pointer.
Now stores the complete block descriptor at the start
of any block allocated from the heap memory pool. This
increases memory overhead by 4 bytes per block, but
streamlines the allocation and freeing algorithms. It also
ensures that the routines will work if the block descriptor
internals are changed in the future.
Now allows the heap memory pool to be defined using the
HEAP_MEM_POOL_SIZE configuration option. This will be the
official configuration approach in the unified kernel.
Also allows the heap memory pool to be defined using the
(undocumented) HEAP_SIZE entry in the MDEF. This is provided
for legacy reasons only.
Co-locates memory pool initialization code to keep the line
that causes memory pool initialization to be done during booting
right next to the routine that does the initialization.
Change-Id: Ifea9d88142fb434d4bea38bb1fcc4856a3853d8d
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Reworks k_work_q_start() so that it accepts its 3 configuration
settings directly, rather than forcing the caller to pass in a
configuration data structure.
Change-Id: Ic0bd1b94f1a1c8e0f8a84b3bd3677d59d0708734
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Aligns the APIs for defining a thread at compile time and for
spawning a thread at run time.
Change-Id: Ic5df450cbe4d0eb562fb4a608f1ac5a8a7cb4b96
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
The "__noinit" was accidentally lost during initial prototyping
of the unified kernel. This just restores it ...
Change-Id: Id13e0e9a323c1bcd49c28a5d8da73943b0177890
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
Drop the _m3 from the test name since this can run on M0, M3, M4, etc.
Change-Id: Ia12ece62fc7b42e28f37e191c90c0dead48d40d0
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
This will make it possible for us to optimize the list of tests we run, for
example, we could exclude footprint tests from the main run because those are
run as part of the footprint checks later in the CI job.
Use like this: sanitycheck -e footprint
Change-Id: I4e7a3aa6fac2ba1c9c99b356f08459da97fda777
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The name read16() collides with other definitions within Zephyr. We
don't actually use this function here, so remove the definition. Fix
various other unnecessary name exports while we are here by adding
'static'.
Change-Id: I7eee8c527a62fea4e6e1bdae8a4874d8ce66596c
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
The read16() name collides with other definitions within zephyr. Make
it static and fix a few other unnecessary name exports while we are
here.
Change-Id: Id72d44a6b84b8d72c7ba2d0f34b29dbc1d02f2b6
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
The build_all tests contain an ever growing list of device drivers to
build. Ideally we minimize the number of images we build, but we
already observe that some of the tests, notably sensors, is too big
for some of our supported boards.
Rather disable an ever growing list of boards as the build_all tests
get bigger it would be better to split the tests into smaller chunks
that can reasonably be expected to run on any supported board.
We split the sensor test set into two, the division is arbitrary,
based on the name of the driver. This allows us to remove the filter
on the quark_d2000.
The current split into two groups is arbirary, in the future it is
inevitable that as the list of supported drivers grows, we will need
to further subdivied the larger tests.
Change-Id: If7ee00b3c8e1749c4c827f83d7cbc2feb70e56ad
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
Introduce an architecture sorting of boards. This is to allow for
easier maintenance going forward as the number of boards grows. It
will be easier for any scripts to know the board/arch mapping without
having to maintain an explicit list of what boards are associated with
which arch. We can also do things like have architecture maintainers
cover reviews and branches for arch/${ARCH} and boards/${ARCH} going
forward.
Change-Id: I02e0a30292b31fad58fb5dfab2682ad1c5a7d5a7
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
This will signify that we are in the development phase. Having the 1.5.0 on
master between two major releases might be confusing.
Jira: ZEP-1100
Change-Id: Ifbea81fbb8afe544af1fb30c79cb67dadb6d731d
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
