On ARM64 the build will fail with coverage and GCOV enabled:
CONFIG_COVERAGE=y
CONFIG_COVERAGE_GCOV=y
CONFIG_FORCE_COVERAGE=y
west build -b rpi_5 -p always samples/hello_world -- \
-DCONFIG_COVERAGE=y -DCONFIG_FORCE_COVERAGE=y
Failure:
"linker_zephyr_pre0.cmd:181: undefined symbol `__gcov_bss_end' referenced
in expression"
Fix build of ARM64 platforms with coverage and GCOV enabled by adding for
ARM64 the same gcov linker sections as it is done for ARM.
Signed-off-by: Grygorii Strashko <grygorii_strashko@epam.com>
number of counters have changed in gcc 14 by commit
https://github.com/gcc-mirror/gcc/commit/08a52331803
Specifically, a new counter was added in gcc/gcov-counter.def:
```
/* Conditions. The counter is interpreted as a bit-set. */
DEF_GCOV_COUNTER(GCOV_COUNTER_CONDS, "conditions", _ior)
```
which in turn updates the value of GCOV_COUNTERS that is defined in
gcov-io.h like this:
```
enum {
GCOV_COUNTERS
};
```
Signed-off-by: Roman Studenikin <srv@meta.com>
Don't attempt to lock the scheduler is trying to dump coverage
information from an ISR. The scheduler won't run while the ISR is in
progress.
Signed-off-by: Jordan Yates <jordan@embeint.com>
Add an option to reset gcov counters before running tests.
This ensures that only code lines triggered by test itself are counted in
the coverage report and all the board initialization code and pre-test
bootstrap is not counted. This is useful when, for example, you are
testing code that is also executed during bootup
Test Plan:
west build -p always -b qemu_x86_64 tests/ztest/base/ -- \
-DCONFIG_COVERAGE=y -DCONFIG_COVERAGE_GCOV=y -DCONFIG_COVERAGE_DUMP=y \
-DCONFIG_ZTEST_COVERAGE_RESET_BEFORE_TESTS=y
ninja -Cbuild run | tee log.log
Signed-off-by: Roman Studenikin <srv@meta.com>
Ability to reset gcov counters allows better isolation of tests
coverage. Specifically it allows to:
1. Not include counters for any code that was executed prior running the
test, which includes all the code executed during the board boot.
2. Run multiple tests without firmware reload by resetting counters
between each.
Signed-off-by: Roman Studenikin <srv@meta.com>
This allows external applications wishing to report coverage data over
different interfaces the ability to do so by exposing the gcov list.
Signed-off-by: Joshua Lilly <jgl@meta.com>
When there is no buffer for gcov hex data,
dump them directly to console.
This saves RAM which can be used to hold more gcov data.
Reduce number of gcov hex data send by console -
do not send space between every two digits.
Signed-off-by: Piotr Kosycarz <piotr.kosycarz@nordicsemi.no>
While porting the coverage.c file from RIOT to Zephyr, which employs
different compiler flags, I noticed several -Wcast-align GCC warnings on
arm. I think, as is, the current implementation may perform unaligned
memory accesses which may not be supported on certain platforms. To
workaround that, I have rewritten the code for RIOT using
bytewise-writes with `memcpy`.
Signed-off-by: Sören Tempel <tempel@uni-bremen.de>
The heap size for serial dump of gcov data is currently
defined by MALLOC_MAX_HEAP_SIZE, which cannot be
adjusted by kconfig. This commit adds a new kconfig
option, CONFIG_COVERAGE_GCOV_HEAP_SIZE, which retains
the behavior or MALLOC_MAX_HEAP_SIZE by default.
Signed-off-by: Jason Wright <jwright@synchron.com>
The GCC/GCOV version over 12 has slight format change of the gcno
and gcda. Make some adaption in the gcov dump function to fix the
code coverage report.
Mainly two places change:
1. Added the checksum in the struct gcov_info. This fix the crash
in qemu_x86, and mps2_an385 when run with --coverage.
2. Adjust the GCOV_TAG_FUNCTION_LENGTH accroding to gcov-io.h. It's
length unit is caculated by bytes now.
Fixes#50255.
Fixes#50257.
Signed-off-by: Enjia Mai <enjia.mai@intel.com>
As of today <zephyr/zephyr.h> is 100% equivalent to <zephyr/kernel.h>.
This patch proposes to then include <zephyr/kernel.h> instead of
<zephyr/zephyr.h> since it is more clear that you are including the
Kernel APIs and (probably) nothing else. <zephyr/zephyr.h> sounds like a
catch-all header that may be confusing. Most applications need to
include a bunch of other things to compile, e.g. driver headers or
subsystem headers like BT, logging, etc.
The idea of a catch-all header in Zephyr is probably not feasible
anyway. Reason is that Zephyr is not a library, like it could be for
example `libpython`. Zephyr provides many utilities nowadays: a kernel,
drivers, subsystems, etc and things will likely grow. A catch-all header
would be massive, difficult to keep up-to-date. It is also likely that
an application will only build a small subset. Note that subsystem-level
headers may use a catch-all approach to make things easier, though.
NOTE: This patch is **NOT** removing the header, just removing its usage
in-tree. I'd advocate for its deprecation (add a #warning on it), but I
understand many people will have concerns.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
This adds the bits to pin the GCOV sections to the pinned
sections so they can be accessed during boot when demand
paging is enabled. Or else accessing them would result in
page faults.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds an ALIGN statement to the rodata linker snippet
for coverage. Without this, sometimes the section is not
aligned, but __init_array_start indicates an aligned
address, resulting in incorrect function pointers.
So align it.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
In order to bring consistency in-tree, migrate all subsystems code to
the new prefix <zephyr/...>. Note that the conversion has been scripted,
refer to zephyrproject-rtos#45388 for more details.
Signed-off-by: Gerard Marull-Paretas <gerard.marull@nordicsemi.no>
The overall code coverage report of mps2_an385 was blocked by the
tests/net/lib/coap, the error message shows "No Mem available to
continue dump". So we enlarge the gcov heap size to prevent this
situation, try to make the report can be generated at least.
Signed-off-by: Enjia Mai <enjia.mai@intel.com>
We need to do a few things differently if we are to support
a virtual memory map, i.e. CONFIG_MMU where CONFIG_KERNEL_VM_BASE
is not the same as CONFIG_SRAM_BASE_ADDRESS.
- All sections must be specified with a VMA and LMA, where
VMA is the virtual address and LMA is the physical memory
location.
- All sections must be specified with ALIGN_WITH_INPUT to
keep VMAs and LMAs synchronized
To do this, the existing linker macros need some adjustment:
- GROUP_LINK_IN undefined when CONFIG_KERNEL_VM_BASE is not
the same as CONFIG_SRAM_BASE_ADDRESS.
- New macro GROUP_ROM_LINK_IN for text/rodata sections
- New macro GROUP_NOLOAD_LINK_IN for bss/noinit sections
- Implicit ALIGN_WITH_INPUT for all sections
GROUP_FOLLOWS_AT is unused anywhere in the kernel for years
now and has been removed.
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
* add toolchain abstraction for coverage
* add select HAS_COVERAGE_SUPPORT to kconfig
* port gcov linker code to CKake for arc
* give user permission to gcov bss section
* expand the size of iccm and dccm to 1M
Signed-off-by: Jingru Wang <jingru@synopsys.com>
During coverage reports generation in C++ code gcov_coverage_dump()
function would get stuck in endless loop. Fix by checking list head
pointer with current list pointer.
Signed-off-by: Marko Poljanić <mpoljanic@gmail.com>
Use the core k_heap API pervasively within our tree instead of the
z_mem_pool wrapper that provided compatibility with the older mempool
implementation.
Almost all of this is straightforward swapping of one alloc/free call
for another. In a few cases where code was holding onto an old-style
"mem_block" a local compatibility struct with a single field has been
swapped in to keep the invasiveness of the changes down.
Note that not all the relevant changes in this patch have in-tree test
coverage, though I validated that it all builds.
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Mark all k_mem_pool APIs deprecated for future code. Remaining
internal usage now uses equivalent "z_mem_pool" symbols instead.
Fixes#24358
Signed-off-by: Andy Ross <andrew.j.ross@intel.com>
Updated the GCOV_COUNTERS value, without which the coverage data
was corrupted when gcc 10.2 was used.
Signed-off-by: Kumar Gala <kumar.gala@linaro.org>
* add toolchain abstraction for coverage
* add select HAS_COVERAGE_SUPPORT to kconfig
* port gcov linker code to CKake for arc
Signed-off-by: Jingru Wang <jingru@synopsys.com>
The .init_array vector is, in fact, read-only data, so put it there.
Previously it was placed at the end of the ROM, but was unknown to
the x86 memory protection code with XIP enabled (because it was not
part of the text, rodata, or kernel RAM). Until recently, the XIP
implementation artificially bloated _image_rodata_size to cover the
entire ROM, so the (mis)placement of .init_array went unnoticed.
Signed-off-by: Charles E. Youse <charles.youse@intel.com>
Update the files which contain no license information with the
'Apache-2.0' SPDX license identifier. Many source files in the tree are
missing licensing information, which makes it harder for compliance
tools to determine the correct license.
By default all files without license information are under the default
license of Zephyr, which is Apache version 2.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This is to fix build errors complaining about undefined reference
to __gcov_exit(). There is no special processing required here
so leave the function empty.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This part of tests was forgotten when we move to subsys/
Fixes#13729
Signed-off-by: Wentong Wu <wentong.wu@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
