Add a dependency on MULTITHREADING for the
STACK_SENTINEL feature, so it may not get
enabled in single-thread Zephyr builds.
Signed-off-by: Ioannis Glaropoulos <Ioannis.Glaropoulos@nordicsemi.no>
Increase the default thread analyzer stack size. On ARM systems the
stack usage is higher with CONFIG_FPU enabled.
The default of 512 is not enough in this case and lead to stack
overflow.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Remove this intrusive tracing feature in favor of the new object tracing
using the main tracing feature in zephyr. See #33603 for the new tracing
coverage for all objects.
This will allow for support in more tools and less reliance on GDB for
tracing objects.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Remove the config BOOT_TIME_MEASUREMENT and corresponding #ifdef'd code
throughout (kernel/init.c, idle.c, core/common.S , reset.S, ... ) which
hold the extern hooks for z_timestamp_main and z_timestamp_idle in the
removed boot_time test suite.
Signed-off-by: Jennifer Williams <jennifer.m.williams@intel.com>
Spin lock validation is touching threads. Allow only when
multithreading is enabled.
Signed-off-by: Krzysztof Chruscinski <krzysztof.chruscinski@nordicsemi.no>
The ARM64 port is currently using SP_EL0 for everything: kernel threads,
user threads and exceptions. In addition when taking an exception the
exception code is still using the thread SP without relying on any
interrupt stack.
If from one hand this makes the context switch really quick because the
thread context is already on the thread stack so we have only to save
one register (SP) for the whole context, on the other hand the major
limitation introduced by this choice is that if for some reason the
thread SP is corrupted or pointing to some unaccessible location (for
example in case of stack overflow), the exception code is unable to
recover or even deal with it.
The usual way of dealing with this kind of problems is to use a
dedicated interrupt stack on SP_EL1 when servicing the exceptions. The
real drawback of this is that, in case of context switch, all the
context must be copied from the shared interrupt stack into a
thread-specific stack or structure, so it is really slow.
We use here an hybrid approach, sacrificing a bit of stack space for a
quicker context switch. While nothing really changes for kernel threads,
for user threads we now use the privileged stack (already present to
service syscalls) as interrupt stack.
When an exception arrives the code now switches to use SP_EL1 that for
user threads is always pointing inside the privileged portion of the
stack of the current running thread. This achieves two things: (1)
isolate exceptions and syscall code to use a stack that is isolated,
privileged and not accessible to user threads and (2) the thread SP is
not touched at all during exceptions, so it can be invalid or corrupted
without any direct consequence.
Signed-off-by: Carlo Caione <ccaione@baylibre.com>
There's no need to duplicate the linker section for each architecture.
Instead, move the section declaration to common-rom.ld.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
The exported structures that were originally introduced for OpenOCD have
since then been reused for other debugger plugins, including PyOCD and
Segger J-Link.
Rename the Kconfig option and the implementation from openocd to debug
thread info, so that it reflects the fact that this is no longer
specifically tied to OpenOCD.
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
If CONFIG_LOG_IMMEDIATE=n, go to panic mode with LOG_PANIC() so full
coredump is logged.
Add missing log_strdup() call.
Signed-off-by: Pete Skeggs <peter.skeggs@nordicsemi.no>
the implementation of spinlock validation uses two LSB bits in the
bottom of a pointer union to store a CPU index, which only has space
for 4 CPUS. the MP_NUM_CPUS should be <= 4.
Signed-off-by: Watson Zeng <zhiwei@synopsys.com>
This removes the z_ prefix those (functions, enums, etc.) that
are being used outside the coredump subsys. This aligns better
with the naming convention.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds some shell commands to the logging backend. Since
this is a simple backend, only get/clear errors are
implemented.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
This adds two new APIs to the coredump subsystem to perform
query and command. These can be used to query coredump subsys
for information, and to perform commands such as finding
out if there is a stored coredump.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
The coredump frontend (mostly) consists of wrappers of backend
functions so there is really no need to track errors at
the frontend level. Let the backends deal with their own errors
and this simplifies the code a bit.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Originally there was a null backend but it was removed before
code was merged. However, some leftover code still refers to
this null backend. So remove the leftovers.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Add thread runtime statistics to the thread analyser.
With CONFIG_THREAD_RUNTIME_STATS enabled:
Booting from ROM..*** Booting Zephyr OS build zephyr-v2.4.0-2330-g77be0e93e65b ***
thread_a: Hello World from cpu 0 on qemu_x86!
Thread analyze:
thread_b : STACK: unused 740 usage 284 / 1024 (27 %); CPU: 0 %
thread_analyzer : STACK: unused 8 usage 504 / 512 (98 %); CPU: 0 %
thread_a : STACK: unused 648 usage 376 / 1024 (36 %); CPU: 98 %
idle 00 : STACK: unused 204 usage 116 / 320 (36 %); CPU: 0 %
thread_b: Hello World from cpu 0 on qemu_x86!
thread_a: Hello World from cpu 0 on qemu_x86!
thread_b: Hello World from cpu 0 on qemu_x86!
thread_a: Hello World from cpu 0 on qemu_x86!
thread_b: Hello World from cpu 0 on qemu_x86!
thread_a: Hello World from cpu 0 on qemu_x86!
thread_b: Hello World from cpu 0 on qemu_x86!
thread_a: Hello World from cpu 0 on qemu_x86!
Thread analyze:
thread_b : STACK: unused 648 usage 376 / 1024 (36 %); CPU: 7 %
thread_analyzer : STACK: unused 8 usage 504 / 512 (98 %); CPU: 0 %
thread_a : STACK: unused 648 usage 376 / 1024 (36 %); CPU: 9 %
idle 00 : STACK: unused 204 usage 116 / 320 (36 %); CPU: 82 %
thread_b: Hello World from cpu 0 on qemu_x86!
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This adds the pieces needed for openocd.c to compile when ARCH=SPARC.
In particular, it allows the tracing.osawareness.openocd sample to
build and run.
Signed-off-by: Martin Åberg <martin.aberg@gaisler.com>
GDB on host may ask the guest to read the a pointer contents even if
it is pointing to NULL. Send an error code when this occurs.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
It implements gdb remote protocol to talk with a host gdb during the
debug session. The implementation is divided in three layers:
1 - The top layer that is responsible for the gdb remote protocol.
2 - An architecture specific layer responsible to write/read registers,
set breakpoints, handle exceptions, ...
3 - A transport layer to be used to communicate with the host
The communication with GDB in the host is synchronous and the systems
stops execution waiting for instructions and return its execution after
a "continue" or "step" command. The protocol has an exception that is
when the host sends a packet to cause an interruption, usually triggered
by a Ctrl-C. This implementation ignores this instruction though.
This initial work supports only X86 using uart as backend.
Signed-off-by: Flavio Ceolin <flavio.ceolin@intel.com>
This adds a very primitive coredump mechanism under subsys/debug
where during fatal error, register and memory content can be
dumped to coredump backend. One such backend utilizing log
module for output is included. Once the coredump log is converted
to a binary file, it can be used with the ELF output file as
inputs to an overly simplified implementation of a GDB server.
This GDB server can be attached via the target remote command of
GDB and will be serving register and memory content. This allows
using GDB to examine stack and memory where the fatal error
occurred.
Signed-off-by: Daniel Leung <daniel.leung@intel.com>
Sort entries alphabetically and cleanup top level menu for each
subsystem. Move stats subsystem Kconfig from debug into its own Kconfig.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
This one was appearing in the top level Kconfig menu when viewed in
menuconfig. Move it to where it belongs.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
The boot time measurement can also run with the HPET timer so there's
no reason to restrict it to APIC.
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
In case of SMP (i.e. multiple execution units processing
the same list of tasks) we cannot use the same data structures for
getting data about active tasks as with just one processor (UP).
So until explicit support of SMP is added make sure we don't allow
to select both OPENOCD & SMP simultaneously.
Moreover starting from
commit a203d21962 ("kernel: remove legacy fields in _kernel")
this will lead to build-time error if MP_NUM_CPUS > 1.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
This commit renames the Kconfig `FP_SHARING` symbol to `FPU_SHARING`,
since this symbol specifically refers to the hardware FPU sharing
support by means of FPU context preservation, and the "FP" prefix is
not fully descriptive of that; leaving room for ambiguity.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
This commit renames the Kconfig `FLOAT` symbol to `FPU`, since this
symbol only indicates that the hardware Floating Point Unit (FPU) is
used and does not imply and/or indicate the general availability of
toolchain-level floating point support (i.e. this symbol is not
selected when building for an FPU-less platform that supports floating
point operations through the toolchain-provided software floating point
library).
Moreover, given that the symbol that indicates the availability of FPU
is named `CPU_HAS_FPU`, it only makes sense to use "FPU" in the name of
the symbol that enables the FPU.
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
I think people might be reading differences into 'if' and 'depends on'
that aren't there, like maybe 'if' being needed to "hide" a symbol,
while 'depends on' just adds a dependency.
There are no differences between 'if' and 'depends on'. 'if' is just a
shorthand for 'depends on'. They work the same when it comes to creating
implicit menus too.
The way symbols get "hidden" is through their dependencies not being
satisfied ('if'/'depends on' get copied up as a dependency on the
prompt).
Since 'if' and 'depends on' are the same, an 'if' with just a single
symbol in it can be replaced with a 'depends on'. IMO, it's best to
avoid 'if' there as a style choice too, because it confuses people into
thinking there's deep Kconfig magic going on that requires 'if'.
Going for 'depends on' can also remove some nested 'if's, which
generates nicer symbol information and docs, because nested 'if's really
are so simple/dumb that they just add the dependencies from both 'if's
to all symbols within.
Replace a bunch of single-symbol 'if's with 'depends on' to despam the
Kconfig files a bit and make it clearer how things work. Also do some
other minor related dependency refactoring.
The replacement isn't complete. Will fix up the rest later. Splitting it
a bit to make it more manageable.
(Everything above is true for choices, menus, and comments as well.)
Detected by tweaking the Kconfiglib parsing code. It's impossible to
detect after parsing, because 'if' turns into 'depends on'.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
Move rtt configuration options to drivers/debug and split the
systemview configuration.
drivers/debug will service for this class of drivers that are enabled in
debug mode only and provide a hardware interface to the system.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Tracing subsystem is growing and although related to debugging, it does
deserve to belong into its own subsystem.
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
Add Kconfig option TRACING_TEST which can be used to
customize tracing packet format (currently supporting
string format and data format) for testing purpose.
Signed-off-by: Wentong Wu <wentong.wu@intel.com>
Add TRACING_ISR Kconfig to help high latency backend working well.
Currently the ISR tracing hook function is put at the begining and
ending of ISR wrapper, when there is ISR needed in the tracing path
(especially tracing backend), it will cause tracing buffer easily
be exhausted if async tracing method enabled. Also it will increase
system latency if all the ISRs are traced. So add TRACING_ISR to
enable/disable ISR tracing here. Later a filter out mechanism based
on irq number will be added.
Signed-off-by: Wentong Wu <wentong.wu@intel.com>
First, this commit adds user interface in tracing_format.h which
can trace both string format and data format packet.
Second, it adds method both for asynchronous and synchronous way.
For asynchronous method, tracing packet will be buffered in tracing
buffer first, tracing thread will output the stream data with the
help of tracing backend when tracing thread get scheduled.
Third, it adds UART and USB tracing backend for asynchronous
tracing method, and adds POSIX tracing backend for synchronous
tracing way.
Also it can receive command from host to dynamically enable and
disable tracing to have host capture tracing data conveniently.
Signed-off-by: Wentong Wu <wentong.wu@intel.com>
Function cpu_stats_get_ns from CPU stats module returns
invalid time, because of unnecessary conversion into u32_t.
Higher bits of time value are removed during this conversion.
All time values are u64_t, so there is no need to convert anything.
Signed-off-by: Dominik Kilian <Dominik.Kilian@nordicsemi.no>
HAS_SEGGER_RTT is assigned by various Nordic boards, but assignments
have no effect on promptless symbols. This symbol is enabled through
being select'ed by SOC_SERIES_NRF52X.
Holyiot, nRF52833-PCA10100, nRF52840-PCA10056, and nRF52-PCA10040 only
assign HAS_SEGGER_RTT without assigning USE_SEGGER_RTT. They probably
meant to enable USE_SEGGER_RTT, so do that instead.
Also add a help text to HAS_SEGGER_RTT and a warning re. HAS_SEGGER_RTT
vs. USE_SEGGER_RTT to the documentation.
Flagged by https://github.com/zephyrproject-rtos/zephyr/pull/20742.
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
RTT_CONSOLE depends on CONSOLE, but SEGGER_SYSTEMVIEW and
SHELL_BACKEND_RTT select RTT_CONSOLE without also selecting CONSOLE.
This leads to a build failure in drivers/console/rtt_console.c (compiled
if RTT_CONSOLE is enabled) unless CONSOLE is enabled some other way.
Symbols like CONFIG_RTT_RETRY_DELAY_MS won't be defined, because they
depend on CONSOLE.
Came up in https://github.com/zephyrproject-rtos/zephyr/pull/21860. This
fix was verified there.
(Would be nice to get rid of some 'select's in the console subsystem at
some point. This is a quick fix.)
Signed-off-by: Ulf Magnusson <Ulf.Magnusson@nordicsemi.no>
Add option to disable the assertion message, this makes all __ASSERT
behave as __ASSERT_NO_MSG instead.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
Add verbose option which would control if the assertion mechanism prints
any information at all. With this disabled they application will have to
use the stack-frame to locate the assertion location.
Signed-off-by: Joakim Andersson <joakim.andersson@nordicsemi.no>
