OpenOCD memory leak in `hwthread_update_threads()` Update `os_free()` from `rtos/rtos.c:96` to fix memory leaks Change-Id: Id7b4c775291b735a0b4423151c2930afce6bf0bd Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com> Reviewed-on: https://review.openocd.org/c/openocd/+/7934 Reviewed-by: Tim Newsome <tim@sifive.com> Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
rtos/zephyr: arm: fetch arm exc return offset Since zephyrproject-rtos/zephyr@c3eeae8, Zephyr OS exposes offset of mode_exc_return in the arch struct for ARM. Accounting for this allows for consistency and enables logic with further offsets that may be added after this. Signed-off-by: Bruno Mendes <bd_mendes@outlook.com> Change-Id: Id53ebd80c5d98a7d94eb6b00ad638ce51e719822 Reviewed-on: https://review.openocd.org/c/openocd/+/7851 Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Tested-by: jenkins
rtos: uCOS-III: split struct ucos_iii_params The static analyser 'sparse' complains about using sizeof() on a struct that has variable size: src/rtos/uCOS-III.c:267:32: warning: using sizeof on a flexible structure src/rtos/uCOS-III.c:269:41: warning: using sizeof on a flexible structure src/rtos/uCOS-III.c:275:66: warning: using sizeof on a flexible structure The struct ucos_iii_params contains either constants values for different target type and variable fields. The last field is an variable size array, always allocated to UCOS_III_MAX_THREADS items. It's not practical to fix this size because we would get too huge initialization in data segment. Split away from struct ucos_iii_params all the variable fields and put them in struct ucos_iii_private. Add in the new struct a pointer to the selected element of ucos_iii_params_list[] and fix the size of array threads[] to its maximum value; this would be allocated at run-time, avoiding impacts to data segment. Change-Id: I569011a257783d35a8795adbda06e942b4157f2a Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/7678 Tested-by: jenkins
rtos: move prototype of rtos_thread_packet() in rtos.h The function rtos_thread_packet() is used across rtos and declared locally as extern. Move the prototype of the function in common include rtos.h Change-Id: I50d311b583148a2de628de0997ef1afc9103a70e Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/7677 Tested-by: jenkins
rtos: move in rtos.h the rtos_type's declaration The static analyser 'sparse' complains, while compiling a rtos' file, that the struct rtos_type is declared in the file as non static, but it is not exposed through an include file. The message is: warning: symbol 'XXX' was not declared. Should it be static? Move the list of rtos_type's declaration in rtos.h Change-Id: Ia96dff077407a6653b11920519c1724e4c1167a3 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/7660 Tested-by: jenkins
rtos: with pointers, use NULL instead of 0 Don't compare pointers with 0, use NULL when needed. Don't assign pointer to 0, use NULL. Detected through 'sparse' tool. Change-Id: Ifa81ba961c0d490cc74880b4a46b620e6358f779 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/7598 Tested-by: jenkins
rtos/linux.c: Fix Linux user space border check Linux kernel and user space border is 0xc0000000 not 0xc000000 Signed-off-by: panciyan <panciyan@eswincomputing.com> Change-Id: I6b487cce62ac31737deca97d5f5f7bbc081280f4 Reviewed-on: https://review.openocd.org/c/openocd/+/7570 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos/FreeRTOS: some optimization of freertos_update_threads() 1. update the rtos->thread_count in time, to make sure the allocated thread_name_str and extra_info_str could be freed by rtos_free_threadlist(). Otherwise the abnormal return may cause a memory leak. 2. remove a redundant assignment to threadid. Signed-off-by: Chao Du <duchao@eswincomputing.com> Change-Id: Ifabc59d501c925b3d6aec8b04b2856d2c31cc4e2 Reviewed-on: https://review.openocd.org/c/openocd/+/7549 Tested-by: jenkins Reviewed-by: Tomas Vanek <vanekt@fbl.cz> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos/nuttx: add Espressif target support and refactoring Almost written from the beginning in a modern OpenOCD way. - Endiannes support - Proper variable types - Align with the other rtos implementations Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com> Change-Id: I0868a22da2ed2ab664c82b17c171dc59ede78d10 Reviewed-on: https://review.openocd.org/c/openocd/+/7444 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos: remove config.h includes from stackings headers And add its own header to the rtos_xxx_stackings.c Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com> Change-Id: I084130fde7ee8645129a7cf60bb7bf59448e2f39 Reviewed-on: https://review.openocd.org/c/openocd/+/7441 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos: Fix constness of struct rtos_type Change-Id: Iaa89f2ff4036c23f944ffb4f37fe0c7afaf5069b Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/6680 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos: Support rt-kernel Works for the proprietary rt-kernel from rt-labs. See: https://rt-labs.com/product/rt-kernel/ Change-Id: Id2c2e292c15fb17eab25e3d07db05014daa2a2b0 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/6668 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos/nuttx: add stacking info for Espressif Xtensa chips ESP32, ESP32-S2 and ESP32-S3 stack register offsets added Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com> Change-Id: Ic6215c1d0152853fd08f82cbd3c138c7d62dbc46 Reviewed-on: https://review.openocd.org/c/openocd/+/7443 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos: add custom stack read function This is optional field for the targets which has to implement their custom stack read function. Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com> Change-Id: Icbc9ed66a052fc2cc0ef67e3ec4d85ab0c2c1b94 Reviewed-on: https://review.openocd.org/c/openocd/+/7442 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos/nuttx: add riscv stacking info Tested with Espressif ESP32-C3 MCU Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com> Change-Id: Ia71ace4909f2dc93ddc07a2ec5524cf374f1937c Reviewed-on: https://review.openocd.org/c/openocd/+/7251 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
rtos/nuttx: move stacking info to the new nuttx stackings file Other OSes have separate files to keep stack register offsets. Adding them for NuttX as well will provide a clearer way to expand. Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com> Change-Id: I1428fefefa199a95431f2073c0279dd7028ad8da Reviewed-on: https://review.openocd.org/c/openocd/+/7250 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
src/rtos/eCos : Support per-build configuration specific structure layouts Extended feature. This is a large patch, but is self-contained to the eCos support and does not affect any other openocd functionality. It does NOT affect existing eCos RTOS plugin users where their applications do not provide the extended symbolic helper information. If the helper symbols are not available the rtos support code will behave as before. This "dynamic" functionality is *required* because eCos does NOT have a fixed/hardwired, known, layout for the thread descriptor structure. The per-application build eCos configuration can affect the shape of the thread descriptor structure (field presence, and hence offsets of subsequent fields) such that constant values cannot be used to consistently interpret all possible eCos application configurations. For historical reasons, there is not yet a consistent namespace for the helper symbols across eCos HALs hence the support is currently limited to specific architectures (Cortex-M and ARM/Cortex-A). No new Clang analyser warnings are raised by this changeset. Change-Id: Ib3a36877326eeb56595cbca55e21b9e59a59c98a Signed-off-by: James G. Smith <jsmith@rallysmith.co.uk> Reviewed-on: https://review.openocd.org/c/openocd/+/6275 Reviewed-by: Alex Schuilenburg <alex.schuilenburg@gmail.com> Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com> Tested-by: jenkins
nds32: drop it, together with aice adapter driver The target nds32 and its companion adapter aice have not received any real improvement since 2013. It has been hard to keep them aligned during the evolution of OpenOCD code, with no way for maintainers to really check if they are still working. No real documentation is present for them in OpenOCD. The nds32 code triggers ~50 errors/warnings with scan-build. The arch nds32 has been dropped from Linux kernel v5.18-rc1. For all the reasons above, this code has been deprecated with commit 2e5df83de7f2 ("nds32: deprecate it, together with aice adapter driver") and tagged to be dropped before v0.13.0. Let it r.i.p. in OpenOCD git history. While there, drop from checkpatch list the camelcase symbols that where only used in this code. Change-Id: Ide52a217f2228e9da2f1cc5036c48f3536f26952 Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com> Reviewed-on: https://review.openocd.org/c/openocd/+/7382 Tested-by: jenkins
hwthread: Restore current_threadid in hwthread_update_threads When OpenOCD receives a step-execution command from GDB and the target is configured as rtos=hwthread, OpenOCD reconstructs the thread-info. However, OpenOCD does not restore the thread id which is currently selected by GDB. Due to this issue, OpenOCD sends the information of wrong thread to GDB after the step execution. This commit fixes the above issue by adding a code to save/restore the thread id selected by GDB. Signed-off-by: Koudai Iwahori <koudai@google.com> Change-Id: I761a1141c04d48f1290e4f09baa7c7024f86f36a Reviewed-on: https://review.openocd.org/c/openocd/+/7358 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
hwthread: Add register validity check in get_thread_reg_list When OpenOCD receives 'g' packet (read general registers) from GDB and target is configured as rtos=hwthread, hwthread_get_thread_reg_list is called. However, it does not check if the register valid or not. Due to this issue, OpenOCD returns invalid register values to GDB. This commit adds a validity check to hwthread_get_thread_reg_list. If the register is not valid, it tries to read the register from the target. Signed-off-by: Koudai Iwahori <koudai@google.com> Change-Id: Iad6424b62124271ec411b1dfc044b57dfc460280 Reviewed-on: https://review.openocd.org/c/openocd/+/7357 Tested-by: jenkins Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>