Disabled the CONFIG_COVERAGE for benchmarks and other tests.
This is needed because it interferes with normal behavior of the
test case.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
These tests need to use stack size as a function of
CONFIG_TEST_EXTRA_STACKSIZE. These test will fail when
CONFIG_COVERAGE is enabled.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
This patch adds a x86_64 architecture and qemu_x86_64 board to Zephyr.
Only the basic architecture support needed to run 64 bit code is
added; no drivers are added, though a low-level console exists and is
wired to printk().
The support is built on top of a "X86 underkernel" layer, which can be
built in isolation as a unit test on a Linux host.
Limitations:
+ Right now the SDK lacks an x86_64 toolchain. The build will fall
back to a host toolchain if it finds no cross compiler defined,
which is tested to work on gcc 8.2.1 right now.
+ No x87/SSE/AVX usage is allowed. This is a stronger limitation than
other architectures where the instructions work from one thread even
if the context switch code doesn't support it. We are passing
-no-sse to prevent gcc from automatically generating SSE
instructions for non-floating-point purposes, which has the side
effect of changing the ABI. Future work to handle the FPU registers
will need to be combined with an "application" ABI distinct from the
kernel one (or just to require USERSPACE).
+ Paging is enabled (it has to be in long mode), but is a 1:1 mapping
of all memory. No MMU/USERSPACE support yet.
+ We are building with -mno-red-zone for stack size reasons, but this
is a valuable optimization. Enabling it requires automatic stack
switching, which requires a TSS, which means it has to happen after
MMU support.
+ The OS runs in 64 bit mode, but for compatibility reasons is
compiled to the 32 bit "X32" ABI. So while the full 64 bit
registers and instruction set are available, C pointers are 32 bits
long and Zephyr is constrained to run in the bottom 4G of memory.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
It's worth using custom timing information on a few systems to save
cycles or gain precision. But make the use of k_cycle_get_32() a
proper default instead of hardcoding all the platforms and failing to
build on new ones. On Xtensa and RISC-V (and now x86_64) the cycle
informatoin from that call is a very fast wrapper around the native
counters anyway -- all you would save would be the function call
overhead.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
With the new implementation we do not need a NULL terminated list
of kobjects. Therefore the list will only contain valid entries
of kobjects.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Move to latest cmake version with many bug fixes and enhancements.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
The constant NSEC_PER_USEC is already defined in sys_clock.h, there is
no need to define it here.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
These tests was written to try to eliminate timer interrupts, but the
mechanism chosen (setting TICKS_PER_SECOND to 1) is dangerously
susceptible to overflow conditions on systems with fast cycle counters
and high timeout durations.
Actually the tests pass just fine if you use a conventional tick rate
and use a tickless-capable driver (which eliminates interrupts too,
which is the whole point), but there's no easy way in kconfig to do an
"if" to select that condition for capable systems only. Just disable
tickless to keep the same behavior.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
When using an IDE (e.g. Eclipse, Qt Creator), the project name gets
displayed. This greatly simplifies the navigation between projects when
having many of them open at the same time. Naming every project "NONE"
defeats this functionality.
This patch tries to use sensible project names while not duplicating
too much of what is already represented in the path. This is done by
using the name of the directory the relevant CMakeLists.txt file is
stored in. To ensure unique project names in the samples (and again, in
the tests folder) folder, small manual adjustments have been done.
Signed-off-by: Reto Schneider <code@reto-schneider.ch>
I was pretty careful, but these snuck in. Most of them are due to
overbroad string replacements in comments. The pull request is very
large, and I'm too lazy to find exactly where to back-merge all of
these.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Now that the API has been fixed up, replace the existing timeout queue
with a much smaller version. The basic algorithm is unchanged:
timeouts are stored in a sorted dlist with each node nolding a delta
time from the previous node in the list; the announce call just walks
this list pulling off the heads as needed. Advantages:
* Properly spinlocked and SMP-aware. The earlier timer implementation
relied on only CPU 0 doing timeout work, and on an irq_lock() being
taken before entry (something that was violated in a few spots).
Now any CPU can wake up for an event (or all of them) and everything
works correctly.
* The *_thread_timeout() API is now expressible as a clean wrapping
(just one liners) around the lower-level interface based on function
pointer callbacks. As a result the timeout objects no longer need
to store backpointers to the thread and wait_q and have shrunk by
33%.
* MUCH smaller, to the tune of hundreds of lines of code removed.
* Future proof, in that all operations on the queue are now fronted by
just two entry points (_add_timeout() and z_clock_announce()) which
can easily be augmented with fancier data structures.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The system tick count is a 64 bit quantity that gets updated from
interrupt context, meaning that it's dangerously non-atomic and has to
be locked. The core kernel clock code did this right.
But the value was also exposed to the rest of the universe as a global
variable, and virtually nothing else was doing this correctly. Even
in the timer ISRs themselves, the interrupts may be themselves
preempted (most of our architectures support nested interrupts) by
code that wants to set timeouts and inspect system uptime.
Define a z_tick_{get,set}() API, eliminate the old variable, and make
sure everyone uses the right mechanism.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
This just got turned into a function from a "variable" API, but
post-the-most-recent-patch it turns out to be degenerate anyway.
Everyone everywhere should always have been using the kconfig variable
directly, and it was only a weirdness in the tickless API that made it
confusing. Fix.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The existing API defined sys_clock_{hw_cycles,ticks}_per_sec as simple
"variables" to be shared, except that they were only real storage in
certain modes (the HPET driver, basically) and everywhere else they
were a build constant.
Properly, these should be an API defined by the timer driver (who
controls those rates) and consumed by the clock subsystem. So give
them function syntax as a stepping stone to get there.
Note that this also removes the deprecated variable
_sys_clock_us_per_tick rather than give it the same treatment.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
The $srctree environment variable gives the path relative to which
'(o)source' statements work (the current directory is used if $srctree
is unset). It is set to $ZEPHYR_BASE in cmake/kconfig.cmake, so there's
no need to qualify the source of Kconfig.zephyr in sample Kconfig files
(or in external projects).
All 'source's in Zephyr assume that the Zephyr root directory is used as
the srctree as well, and would break otherwise.
Remove the $(ZEPHYR_BASE)s to make it clearer that all 'source'
statements work relative to the Zephyr root. There was some user
confusion on IRC.
Also explain how things work in the documentation.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
Add ifdef guard to the z_reset_timeslice() to fix compilation
errors when CONFIG_TIMESLICING is disabled.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
It is no longer necessary to set the KCONFIG_ROOT variable when the
KConfig file is in the application root directory.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
The total number of loops being executed were taking far too long
to finish executing the test case. This was causing the timer handler
to execute in the middle of the test case. Thereby causing a test case
failure.
Now the number of loops have been reduced and this will make sure the
test case completes.
This change has no effect on the time being calculated by the
benchmark.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
With the new Kconfig preprocessor (described in
https://github.com/torvalds/linux/blob/master/Documentation/kbuild/
kconfig-macro-language.txt), the syntax for expanding environment
variables is $(FOO) rather than $FOO.
$(FOO) is a general preprocessor variable expansion, which falls back to
environment variables if the variable isn't set (like in Make). It can
also be used in prompts, 'comment's, etc.
The old syntax will probably be supported forever in Kconfiglib for
backwards compatibility, but might as well make it consistent now that
people might start using the preprocessor more.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
In NRF52 we need to write to a capture register to retrieve the current
counter value. This was missing for userspace implementation.
Fixes: GH-9676
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Few SoCs whose clock speed is very less will cause multiple ticks
to occur for 1 sec of sleep. For such a platform we cant run
the loops for over 5000 because the tick handler will get executed.
Thus rendering the test useless for such platforms. By reducing
the number of iterations we get the required benchmark results.
Fixes: GH-7906
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
This patch provides support needed to get timing related
information from riscv32 based SOC.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
This patch provides support needed to get timing related
information from nios2 based SOC.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
This patch provides support needed to get timing related
information from xtensa based SOC.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Discard selected measurements that occur immediately before and
after the timer interrupt occurs. This causes fluctuations in the
time measured.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
This patch provides support needed to get timing related
information from ARC based SOC.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Added new KPIs for the userspace benchmarks.
1. Drop to user mode.
2. User thread Creation.
3. Syscall overhead.
4. Validation overhead k object init.
5. Validation overhead k object permission.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Prepend the text 'cmake_minimum_required(VERSION 3.8.2)' into the
application and test build scripts.
Modern versions of CMake will spam users with a deprecation warning
when the toplevel CMakeLists.txt does not specify a CMake
version. This is documented in bug #8355.
To resolve this we include a cmake_minimum_required() line into the
toplevel build scripts. Additionally, cmake_minimum_required is
invoked from within boilerplate.cmake. The highest version will be
enforced.
This patch allows us to afterwards change CMake policy CMP000 from OLD
to NEW which in turn finally rids us of the verbose warning.
The extra boilerplate is considered more acceptable than the verbosity
of the CMP0000 policy.
Signed-off-by: Sebastian Bøe <sebastian.boe@nordicsemi.no>
Disable CONFIG_I2C and CONFIG_SPI for minimum footprint test,
since we are doing "minimum" footprint.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The return value from the k_mem_slab_alloc was not read. Hence
adding code to make use of this return value.
Fixes: GH-6681
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Bool symbols implicitly default to 'n'.
A 'default n' can make sense e.g. in a Kconfig.defconfig file, if you
want to override a 'default y' on the base definition of the symbol. It
isn't used like that on any of these symbols though.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
All architecture defines OCTET_TO_SIZEOFUNIT and SIZEOFUNIT_TO_OCTET
as identity functions. But the only user is tests/benchmarks/app_kernel.
It's effectively a no-op. Remove them.
Signed-off-by: Yasushi SHOJI <y-shoji@ispace-inc.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
Disable userspace for benchmarks because it changes the KPI
numbers. It will be re-enabled once the required code is
in place.
Signed-off-by: Adithya Baglody <adithya.nagaraj.baglody@intel.com>
These are no longer needed since commit 4dc9e5b2de ("kconfig: Get rid
of 'option env' bounce symbols"). The C tools are likely to get rid of
them soon too.
The APPLICATION_BASE symbol (option env="PROJECT_BASE") triggered a
compatibility warning, because 'option env' symbols now need to have the
same name as the environment variables they reference to be compatible
with Kconfiglib. APPLICATION_BASE is unused, so just remove it.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
Peter A. Bigot