The logic in `init_registers()` was quite convoluted.
Initialization of each `struct reg` field is separated into function
`gdb_regno_<field_name>()`.
IMHO, this makes it much easier to reason about the code.
Change-Id: Id7faa1464ce026cc5025585d0a6a95a01fb39cee
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
Add access to fpr and cp1 registers.
GDB can now check the FPRs with `info reg f` and change them.
Checkpatch-ignore: MACRO_ARG_REUSE
Change-Id: I63896ab6f6737054d8108db105a13a58e1446fbc
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7866
Tested-by: jenkins
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Extend the existing code to support Monitor mode in AArch32.
Change-Id: Ia43df98d1497baac48aea67b92d81344c24f0635
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8169
Tested-by: jenkins
ADIv6 brought more complicated rules for DP reg 0 banking.
Neither the original implementation [1] nor the later
modification [2] respected that the DP reg 0 is banked
for read only, not for write. Enforcing of an useless
SELECT write before a write to ABORT register may trigger
FAULT (CTRL/STAT bits ORUNDETECT and STICKYORUN are set)
or WAIT (DP is stalled by an outstanding previous operation)
and therefore make ABORT register virtually unusable
on some adapters (bitbang, CMSIS-DAP).
There are DP ABORT specific functions swd_queue_ap_abort()
and swd_clear_sticky_errors() which worked around the problem
using the lowest level swd->write_reg(). Using a specific
write procedure for a single DP register was error prone
(there are other DP_ABORT writes using swd_queue_dp_write_inner())
and also the Tcl command 'xx.dap dpreg 0 value' suffered
from unwanted SELECT write.
Other smaller discords in DP banking probably do not
influence normal DP operation however they may complicate
debugging in corner cases.
Adhere strictly to the DP banking rules for both ADI versions.
Fixes: [1] commit 72fb88613f ("adiv6: add low level swd transport")
Fixes: [2] commit ee3fb5a0ea ("target/arm_adi_v5: fix DP SELECT logic")
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Change-Id: I3328748c1c3e0661c5ecd6eb070ac519b190ace2
Reviewed-on: https://review.openocd.org/c/openocd/+/8154
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Allow direct pointer to struct adiv5_private_config
for targets with adiv5_private_config inside of a bigger
private config container. Use it instead of the private_config
pointer toggling hack in aarch64.c
Allow optional use of -dap parameter and use it instead
of the static variable hack in xtensa_chip.c
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Change-Id: I7260c79332940adfa49d57b45cae39325cdaf432
Reviewed-on: https://review.openocd.org/c/openocd/+/8138
Tested-by: jenkins
Reviewed-by: Ian Thompson <ianst@cadence.com>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Add access to dsp registers and a command for dsp related operations.
Checkpatch-ignore: MACRO_ARG_REUSE
Change-Id: I30aec0b9e4984896965edb1663f74216ad41101e
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7867
Tested-by: jenkins
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
After commit d9b2607ca0 ("gdb_server: support sparse register
maps"), GDB crashes while requesting the value of 'cpsr' because
the fake register is tagged as not existing.
Change the logic and set all register as existing, while still
limiting the list for the initial GDB request at connect.
Change-Id: I1c4e274c06147683db2a59a8920ae5ccd863e15c
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8163
Tested-by: jenkins
For the sake of https://review.openocd.org/c/openocd/+/7957
Instead of "coreid", 'target smp' command call order used as
an index
Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com>
Change-Id: Iab86b81868d37c0bf8663707ee11367c41f6b96d
Reviewed-on: https://review.openocd.org/c/openocd/+/8162
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
find members of a union, nested in struct.
Allows file to be compiled with GCC 4.0
Signed-off-by: Sevan Janiyan <venture37@geeklan.co.uk>
Change-Id: Ied68668d3b5f811573a20e11e83aceff268963eb
Reviewed-on: https://review.openocd.org/c/openocd/+/8120
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Allows build with legacy toolchains which do not support
C23 nor GCC extension for binary literals.
Change-Id: I742d3a8a86bf16f81421d11c59d3cb155ee17aed
Signed-off-by: Sevan Janiyan <venture37@geeklan.co.uk>
Reviewed-on: https://review.openocd.org/c/openocd/+/8123
Tested-by: jenkins
Reviewed-by: Jörg Wunsch <openocd@uriah.heep.sax.de>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Add more debug messages connected with triggers.
Update names for internal flags to make them more clarified.
Change-Id: I5642346ce4a1e9bf79b22cdbf36bd757a7beffa8
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Fix obsoleted references to Cortex-M3 from the time
when M3 was the only supported Cortex.
Fix typo.
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Change-Id: I6f93265f1b9328fec063fecd819210deb28aaf2c
Reviewed-on: https://review.openocd.org/c/openocd/+/8099
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
In 2008 the commit 1829361253 ("define resetting
the target into the halted or running state as an atomic operation.")
introduced the target_halt() call to the end of cortex_m3_assert_reset(),
Checkpatch-ignore: GIT_COMMIT_ID
A year later the commit ed36a8d15d
("... Updated halt handling for cortex_m3")
prevented cortex_m3_halt() take any action in case of TARGET_RESET state.
This narrowed the target_halt() called from cortex_m3_assert_reset()
to setting target->halt_issued and storing a time stamp.
Introducing ocd_process_reset(_inner) made the setting of halt_issued
and halt_issued_time useless. The Tcl function waits for halt
of all targets if applicable.
cortex_m_halt() and also target_halt() does not work as expected
if the cached target state is TARGET_RESET (although the core could
be out of reset and ready to be halted, just have not been polled).
Explicit Tcl arp_poll must be issued in many scenarios.
Remove the useless hack.
Also remove the explicit error return from cortex_m_halt_one()
in case of RESET_SRST_PULLS_TRST and asserted srst. If the communication
with the target is gated by any reset, cortex_m_write_debug_halt_mask()
fails. Propagate the error return of this call instead.
Change-Id: I0da05b87f43c3d0facb78e54d8f00c1728fe7c46
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/8098
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
In a corner case when debug_ap is not available,
cortex_m_assert_reset() asserts reset to restore
communication with the target.
Prevent to do so on targets with defer_examine,
as such targets need some special handling to enable them
after reset anyway.
The change makes possible to handle a multicore Cortex-M SoC with
an auxiliary Cortex-M core(s) switched of by default
even with 'reset_config srst_gates_jtag'
Change-Id: I8cec7a816423e588d5e2e4f7904c81c776eddc42
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/8097
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Function riscv_program_exec() contains code for restoring
of register values after progbuf execution. This code is
not used anymore by current OpenOCD, and hence removed.
Related discussion can be found under:
https://github.com/riscv/riscv-openocd/issues/982
Change-Id: I4c79bec081522b6fc0d16367cef51ed19a131962
Signed-off-by: Jan Matyas <jan.matyas@codasip.com>
Fixes:
- Data types of address & data parameters in riscv_batch_add_*()
and riscv*_fill_dm*() changed to uint64_t and uint32_t.
- Corrected the comparison in riscv_batch_full().
- Corrected assertions in riscv_batch_get_dmi_read_op()
and riscv_batch_get_dmi_read_data().
Cleanup:
- Simplified calloc() fail handling in riscv_batch_alloc().
- Added explicit NULL assignments in riscv_batch_alloc()
for clarity and readability. Don't rely on calloc().
- Removed suffix `_u64` from riscv_*_fill_dm*() since it
does not have any meaning.
- Renamed *dmi_write_u64_bits() to *get_dmi_scan_length()
which better describes its purpose.
Change-Id: Id70e5968528d64b2ee5476f1c00e08459a1e291d
Signed-off-by: Jan Matyas <jan.matyas@codasip.com>
Cleanup, non-functional changes:
- Replaced one call of low-level function dm_op()
by high-level dm_read().
- Made sure that truncation of values passed to dm_*
is explicit. (Added explicit casts.)
Change-Id: I1d1b2f29a822b6841373f3313de2b1e96f514116
Signed-off-by: Jan Matyas <jan.matyas@codasip.com>
If targets are in a halt group, and a target in the group reaches a
breakpoint, the target's state was able to remain `TARGET_RUNNING`.
Addresses issue #1010
Change-Id: I734bc6da71d289c4d05b417c8bf67a7d1a56574f
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
clang build triggers an error for an uninitialized value of the
variable 'instr'.
This is a false positive, as the macro
#define MIPS32_CONFIG3_ISA_MASK (3 << MIPS32_CONFIG3_ISA_SHIFT)
guarantees the switch/case already covers all the possible values
with cases 0, 1, 2 and 3.
Silent clang by adding a useless default case to the switch.
While there, fix the indentation of the switch/case accordingly to
OpenOCD coding style.
Change-Id: I0ae316754ce7d091dd8366bf314b8e6ee780e313
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Fixes: 7de4b1202d ("target/mips32: add cpu info detection")
Reviewed-on: https://review.openocd.org/c/openocd/+/8065
Tested-by: jenkins
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Commit e370e06b72 ("target: Deprecate 'array2mem' and
'mem2array''") has already replaced the deprecated root versions
of commands mem2array and array2mem with TCL proc's that use
'read_memory' and 'write_memory'. It has left the deprecated code
of the target's version of the commands because the effort to code
the TCL replacement was not considered valuable.
To drop the last jim_handler commands, I consider much easier and
less error-prone to code them in TCL instead of converting the
deprecated code to COMMAND_HANDLER.
Drop the code in target.c and extend the TCL proc's.
While there, add the TCL procs to _telnet_autocomplete_skip.
Change-Id: I97d2370d8af479434ddf5af68541f90913982bc0
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8052
Tested-by: jenkins
When requested by the core code (handle_breakpoints = true),
arc_resume() should be able to advance over a potential breakpoint set
at the resume address instead of getting stuck in one place. This is
achieved by removing the breakpoint, executing one instruction,
resetting the breakpoint, then proceeding forward as normal.
With this patch applied, openocd is now able to resume from a
breakpoint halt when debugging ARCv2 targets via telnet.
This has previously been committed to the Zephyr project's openocd repo
(see https://github.com/zephyrproject-rtos/openocd/pull/31).
Change-Id: I17dba0dcea311d394b303c587bc2dfaa99d67859
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
Signed-off-by: Artemiy Volkov <artemiy@synopsys.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7817
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
Presently, we rely on gdb to restore break/watchpoints upon resuming
execution in arc_resume(). To match this behavior in absence of gdb
(more specifically, when handle_breakpoints is true), this patch
explicitly re-enables all breakpoints and watchpoints in arc_resume().
This has previously been committed to the Zephyr project's openocd repo
(see https://github.com/zephyrproject-rtos/openocd/pull/31).
Change-Id: I59e9c91270ef0b5fd19cfc570663dc67a6022dbd
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
Signed-off-by: Artemiy Volkov <artemiy@synopsys.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7816
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
For Armv8.1-M based cores, detect if the core implements the optional
M-profile vector extension (MVE), using MVFR1 register.
While at there rework armv7m->fp_feature detection based on MVFR0
and MVFR1 registers.
Change-Id: I92d5b1759aea9f7561d285f46acdec51d6efb7b4
Signed-off-by: Tarek BOCHKATI <tarek.bouchkati@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/6950
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
The target esirisc does not free the allocated memory resources,
causing memory leaks at OpenOCD exit.
Add esirisc_free_reg_cache() and esirisc_deinit_target() and use
them to free all the allocated resources.
Change-Id: I17b8ebff54906fa25a37f2d96c01d010a98cffbd
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8094
Tested-by: jenkins
Reviewed-by: Steven Stallion <sstallion@gmail.com>
The function in question does not need to change target state. It is a
target-type-dependant function, however, IMHO, it is safe to assume that
any target type would not need to change type-independant state of a
target to figure out the arch.
Change-Id: I607cb3aee6529cd5a97bc1200a0226cf6ef43caf
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8093
Tested-by: jenkins
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
There are quite a lot of "getters" in target interface.
They do not change target structure, nevertheless the structure is
passed to these functions via a plain pointer.
The intention is to clarify the purpouse of these functions by passing
the `target` structure as a pointer to constant data.
Change-Id: Ida4a798da94938753b86a293a308d93b091d1bf3
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8092
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
This reverts commits 2e920a212f and
8dbb1250f5.
The reason is, after `openocd_is_shutdown_pending()` becomes true,
arbitrary command may be executed:
* In `target_destroy()` and the corresponding
`target->type->deinit_target()`.
* In user-specifyed `pre_shutdown_commands` list.
Change-Id: Icd00d1d954cd45e255880a6f76c3a74c098d6a17
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
The reason for the change is a conflict: `dcsr[5]` is `dcsr.v` in
current spec, but it is `dcsr.debugint` in 0.11. This causes `priv`
register to be read incorrectly.
Change-Id: If2d8fdcd8536afa4c7149c453101b00ce0df1ce0
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
There was an error in `dm_read`/`dm_write`: DMI address was checked
against DM registers disregarding DM base address.
To solve the issue `dmi_address()` function was introduced.
Change-Id: Ia3be619417b5f5b53db5dfe302db05170d6787c9
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
The motivalion for the change:
* `riscv_hart_count()` is used only once to print the value into the log
during exmination.
* The returned value is a bit confusing: it's not the total number of
targets on the TAP. It is the number of targets accessable through the
same DM. So the name of the function is misleading.
* This value is already reported on `-d3` level.
So the function seems redundant and can be safely removed.
Change-Id: Iac9021af59ba8dba2cfb6b9dd15eebc98fe42a08
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
The function is used for commands:
- jtag configure
- jtag cget
While there, add the missing .usage field.
Change-Id: I97ddc4898259ddb7fd2d057a997f33a6f4b0e2a8
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8058
Tested-by: jenkins
clang build triggers an error for an uninitialized value of the
variable 'instr'.
This is a false positive, as the macro
#define MIPS32_CONFIG3_ISA_MASK (3 << MIPS32_CONFIG3_ISA_SHIFT)
guarantees the switch/case already covers all the possible values
with cases 0, 1, 2 and 3.
Silent clang by adding a useless default case to the switch.
While there, fix the indentation of the switch/case accordingly to
OpenOCD coding style.
Change-Id: I0ae316754ce7d091dd8366bf314b8e6ee780e313
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Fixes: 7de4b1202d ("target/mips32: add cpu info detection")
Reviewed-on: https://review.openocd.org/c/openocd/+/8065
Tested-by: jenkins
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Commit e370e06b72 ("target: Deprecate 'array2mem' and
'mem2array''") has already replaced the deprecated root versions
of commands mem2array and array2mem with TCL proc's that use
'read_memory' and 'write_memory'. It has left the deprecated code
of the target's version of the commands because the effort to code
the TCL replacement was not considered valuable.
To drop the last jim_handler commands, I consider much easier and
less error-prone to code them in TCL instead of converting the
deprecated code to COMMAND_HANDLER.
Drop the code in target.c and extend the TCL proc's.
While there, add the TCL procs to _telnet_autocomplete_skip.
Change-Id: I97d2370d8af479434ddf5af68541f90913982bc0
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8052
Tested-by: jenkins
When requested by the core code (handle_breakpoints = true),
arc_resume() should be able to advance over a potential breakpoint set
at the resume address instead of getting stuck in one place. This is
achieved by removing the breakpoint, executing one instruction,
resetting the breakpoint, then proceeding forward as normal.
With this patch applied, openocd is now able to resume from a
breakpoint halt when debugging ARCv2 targets via telnet.
This has previously been committed to the Zephyr project's openocd repo
(see https://github.com/zephyrproject-rtos/openocd/pull/31).
Change-Id: I17dba0dcea311d394b303c587bc2dfaa99d67859
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
Signed-off-by: Artemiy Volkov <artemiy@synopsys.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7817
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
Presently, we rely on gdb to restore break/watchpoints upon resuming
execution in arc_resume(). To match this behavior in absence of gdb
(more specifically, when handle_breakpoints is true), this patch
explicitly re-enables all breakpoints and watchpoints in arc_resume().
This has previously been committed to the Zephyr project's openocd repo
(see https://github.com/zephyrproject-rtos/openocd/pull/31).
Change-Id: I59e9c91270ef0b5fd19cfc570663dc67a6022dbd
Signed-off-by: Evgeniy Didin <didin@synopsys.com>
Signed-off-by: Stephanos Ioannidis <root@stephanos.io>
Signed-off-by: Artemiy Volkov <artemiy@synopsys.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7816
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
For Armv8.1-M based cores, detect if the core implements the optional
M-profile vector extension (MVE), using MVFR1 register.
While at there rework armv7m->fp_feature detection based on MVFR0
and MVFR1 registers.
Change-Id: I92d5b1759aea9f7561d285f46acdec51d6efb7b4
Signed-off-by: Tarek BOCHKATI <tarek.bouchkati@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/6950
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
In loops that typically take longer time to complete, check if there is
a pending shutdown request. If so, terminate the loop.
This allows to respond to a signal requesting a shutdown during some
loops which do not return control to main OpenOCD loop.
Change-Id: Iace0b58eddde1237832d0f9333a7c7b930565674
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
Based on Daniel Goehring's [1] and Peter Collingbourne's [2] work.
Probe for support of 8, 16 bit and if the large data extension is available
also probe for 64, 128 and 256 bit operations.
Probe for the ability of packing 8 and 16 bit data
(formerly probed in mem_ap_init()). The probe is integrated to
mem_ap_read/write() routines and takes place just before the first memory
access of the specific size.
Add 64, 128 and 256 bit MEM-AP read/writes.
Introduce specific error codes for unsupported transfer size
and for unsupported packing.
Change-Id: I180c4ef17d2fc3189e8e2f14bafd22d857f29608
Link: 7191: target/adiv5: add MEM-AP 64-bit access support | https://review.openocd.org/c/openocd/+/7191
Link: 7436: arm_adi_v5: Support reads wider than 32 bits | https://review.openocd.org/c/openocd/+/7436
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7576
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
Prevent packed writes with Nuvoton NPCX quirks because the workaround
uses all byte lanes for one byte or halfword and thus precludes packing.
Eliminate quirk code for size 4 as it is equivalent to the common code.
Make the quirk code for sizes 2 and 1 easier readable.
Change-Id: I72324e56a49b4712bd3769e03dce01427d9fcd73
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7575
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Introduce ti_be_lane_xor for byte lane correction
and use common code for both quirk and regular conversion.
The same lane correction takes place in both mem_ap_read/write()
- it was obfuscated in original code with different bitwise and arithmetic
operations.
Change-Id: I6a30672b908770323d30813a714e06ab8695fe26
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7574
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Update mips32 instructions, add barrier and sync related insts.
Add SYNC and barrier instruction blocks for memory access safety.
These instructions are not supported on Lexra and/or MIPSr1 CPUs,
detections were added and they will be executed conditionally.
Rework mips32_pracc_read/write_regs function.
Checkpatch-ignore: MACRO_ARG_REUSE
Change-Id: Ib14112f37ff1f060b1633df73d671a6b09bb2178
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7865
Tested-by: jenkins
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Add detection for mips cpu types by using prid.
Add cpuinfo command for inspecting more verbose info.
Add MIPS Architecture specs in openocd docs.
Change-Id: I28573b7c51783628db986bad0e226dcc399b4fa6
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7912
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Some devices with an internal multi-tap JTAG router require a vendor
specific bypass instruction to bypass the master TAP when addressing
slave taps internal to the same device. On these devices the standard
bypass instruction bypasses the whole device.
Change-Id: I4506f0e67c9e4dfe39b7fa18c63d67900313e594
Signed-off-by: Henrik Nordström <henrik.nordstrom@addiva.se>
Reviewed-on: https://review.openocd.org/c/openocd/+/8041
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
swd_connect_multidrop() sent DORMANT_TO_SWD and called
swd_multidrop_select_inner(). DORMANT_TO_SWD sequence ends
with a LINE_RESET sequence.
swd_multidrop_select_inner() sent LINE_RESET sequence again.
It was useless in this case.
swd_connect_multidrop() emited JTAG_TO_DORMANT and DORMANT_TO_SWD
sequences before connecting each DAP in SWD multidrop bus.
It is sufficient to emit JTAG_TO_DORMANT and DORMANT_TO_SWD
just once and emit the shorter LINE_RESET instead for subsequent DAPs.
Introduce a global variable swd_multidrop_in_swd_state
and use it to control what sequence is emitted.
In case of reconnect after an error, always use the full switch
JTAG_TO_DORMANT and DORMANT_TO_SWD.
Change-Id: Iba21620f6a9680793208bf398960ed0eb59df3b1
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7218
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
SWD multidrop requires some initialization once before connecting
all daps. Provide an optional pre-connect dap operation.
Change-Id: I778215c512c56423a425dda80ab19a739f22f285
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7542
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
The original code supported ADIv5 only, just one SELECT register
with some reserved bits - the pseudo value DP_SELECT_INVALID was
just fine to indicate the DP SELECT register is in an unknown state.
Added ADIv6 support required DP SELECT and SELECT1 registers
without reserved bits. Therefore DP_SELECT_INVALID value became
reachable as a (fortunately not really used) ADIv6 AP ADDR.
JTAG DPBANKSEL setting support introduced with ADIv6 does not
honor DP_SELECT_INVALID correctly: required select value
gets compared to DP_SELECT_INVALID value and the most common zero
bank does not trigger DP SELECT write.
DP banked registers need just to set DP SELECT. ADIv6 AP register
addressing scheme may use both DP SELECT and SELECT1. This further
complicates using a single invalid value.
Moreover the difference how the SWD line reset influences
DPBANKSEL field between ADIv5 and ADIv6 deserves better handling
than setting select cache to zero and then to DP_SELECT_INVALID
in a very specific code positions.
Introduce bool flags indicating the validity of each SELECT
register and one SWD specific for DPBANKSEL field.
Use the latter to prevent selecting DP BANK before taking
the connection out of reset by reading DPIDR.
Treat DP SELECT and SELECT1 individually in ADIv6 64-bit mode.
Update comments to reflect the difference between ADIv5 and ADIv6
in SWD line reset.
Change-Id: Ibbb0b06cb592be072571218b666566a13d8dff0e
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7541
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
In loops that typically take longer time to complete, check if there is
a pending shutdown request. If so, terminate the loop.
This allows to respond to a signal requesting a shutdown during some
loops which do not return control to main OpenOCD loop.
Change-Id: Iace0b58eddde1237832d0f9333a7c7b930565674
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8032
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
While at it, fix some coding style issues.
Change-Id: I8196045f46ce043ed0d28cb95470132b3a7de1bb
Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8039
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
The field backup_working_area is always used as a boolean value.
Use bool type for backup_working_area.
Change-Id: I55c68d717dbbe9e5caf60fd1db368527c6d1b995
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8036
Tested-by: jenkins
Reviewed-by: zapb <dev@zapb.de>
I don't think there are any real bugs here, but at least this gives us a
clean slate moving forward.
Change-Id: I29c6c398c28dfe580f9a2deb3bdbcfc491a2ceb6
Signed-off-by: Tim Newsome <tim@sifive.com>
These messages helps to clarify current status of examination process
Change-Id: I5d93903c4680deed2c1bf707d8f7ef0b48ffdc9a
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/8013
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
This functionality can be useful for;
1-ESP flashing code to load flasher stub on target and
write/read/erase flash.
2-ESP GCOV command uses some of these functions to run
onboard routines to dump coverage info.
This is high level api for the Espressif xtensa and riscv targets
Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com>
Change-Id: I5e618b960bb6566ee618d4ba261f51af97a7cb0e
Reviewed-on: https://review.openocd.org/c/openocd/+/7759
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
The goal of this commit is to provide more robust error handling in
`write_memory_progbuf()`. This is achieved by rewriting it in a fashion
similar to `read_memory_progbuf()`.
The motivation is: some instability in `load_image` was encountered. No
stable reproduction could be obtained, so the root cause was not
determined. Therefore, it was decided to clean-up the code, that may be
implicated in such failures.
Examples of unhanded errors in the code prior to this commit:
* Most of `dmi_write()` return values are discarded.
* If `dm_read()` on `abstractcs` failed (line 4546), `abstractauto` was
not cleared.
Furthermore, the structure of the code was quite complicated, which made
it hard to analyze and reason whether or not all possible failures are
handled properly.
Change-Id: I8a100b686e594855fbf34acf5ccf0e1550f18869
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
While at it, fix the 'wp' command documentation.
Change-Id: I70f3110e8ce286051f8f810260f1857b2285e634
Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8022
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
By definition in `target/target.h`, `coreid` is not a unique identifier
of a target -- it can be the same for targets on different TAPs.
Change-Id: Ifce78da55fffe28dd8b6b06ecae7d8c4e305c0a2
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7997
Tested-by: jenkins
Reviewed-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
When user requested a change on cp0 status register,
it may contain changes on EXL/ERL bits, and changes on
these bits could lead to differnt behaviours on writing
to other cp0 registers.
Change-Id: Ic83039988c29c06ee134226b52de943c46d19da2
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7914
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
OpenOCD style guide(`doc/manual/style.txt`) prohibits use of VLA:
> - use malloc() to create dynamic arrays. Do @b not use @c alloca
> or variable length arrays on the stack. non-MMU hosts(uClinux) and
> pthreads require modest and predictable stack usage.
Change-Id: I12e4a5087fd056d69866137237af6deca27f5d33
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
`LOG_TARGET_DEBUG()` reports file, line and function name at the call
site. This information is not helpfull if it always points to the same
location inside `log_debug_reg()`.
Change-Id: Ib73be0344fb5c80c9ac8e5fdee1084d405522eb7
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
If we can't remove bp/wp, we will stuck in endless loop
Change-Id: I44c0a164db1d15c0a0637d33c75087a49cf5c0f4
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7940
Tested-by: jenkins
Reviewed-by: Anatoly P <kupokupokupopo@gmail.com>
Reviewed-by: Tim Newsome <tim@sifive.com>
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
There was a big conflict with this one incoming change. This version of
the code passes all the tests (which don't test for the changed behavior
at all AFAIK), and at least passes errors back everywhere.
Conflicts:
src/target/breakpoints.c
Change-Id: I72f75a3e08deda7e624e8bb82e1a9ea07a7a9276
Instead of returning an 'error string', throw an error. This makes it
much easier to handle errors in Tcl scripts or in tools that use Tcl RPC.
Change-Id: I75c48750cfad7430fa5e6bc88fe04ebd59d34cea
Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/8006
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
The mips opcode macro for the SRL opcode was using the wrong constant
value:
SRA -- Shift right arithmetic
Encoding: 0000 00-- ---t tttt dddd dhhh hh00 0011
SRL -- Shift right logical
Encoding: 0000 00-- ---t tttt dddd dhhh hh00 0010
This corrects the opcode constant for SRL and adds the SRA opcode for
completeness.
There is only one user of MIPS32_OP_SRL in src/flash/nor/cfi.c:
Since the mask constant (0x00000080 for the DQ7 mask) shifted in this
case would never have the sign bit set, it worked fine even though it
was accidentally using the SRA opcode instead of SRL.
Change-Id: I0a80746e2075c7df1ce35b9db00d9d0b997a3feb
Signed-off-by: Tobias Diedrich <ranma+openocd@tdiedrich.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/3613
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
We were previously inadvertently clearing the top 32 bits of SCTLR_EL1
during read_memory/write_memory as a result of using 32-bit operations
to access the register and because the fields used to temporarily
store the register were 32-bit. Fix it.
Change-Id: I657d7f949e1f7ab6bf90609e3f91cae09cade31a
Signed-off-by: Peter Collingbourne <pcc@google.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7939
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Associativity=0 means that only one CMO is needed for all ways.
It could also mean that the cache is 1-way, but this is less likely.
Currently Associativity=0 causes us to hang in the while loop in
decode_cache_reg. Fix it by skipping the loop in this case. We can let
way_shift be set to the arbitrary value of 0 because in the case where
Associativity=0 we only ever shift 0 by it before ORing it into the
CMO operand.
Change-Id: I7c1de68d33f6b3ed627cbb1e2401d43185e4c1e3
Signed-off-by: Peter Collingbourne <pcc@google.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7916
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
This patch adds target logging to breakpoint
handling code. This makes it easier to
debug multicore/multithread systems.
Change-Id: I6bea8079a457070a8f63d0ce381a4ece6f5a190a
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7922
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
If we can't remove bp/wp, we will stuck in endless loop
Change-Id: I44c0a164db1d15c0a0637d33c75087a49cf5c0f4
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7940
Tested-by: jenkins
Reviewed-by: Anatoly P <kupokupokupopo@gmail.com>
Reviewed-by: Tim Newsome <tim@sifive.com>
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Break- and watchpoints are not removed when a target is destroyed
which introduces a memory leak.
Change-Id: I6143d48f7efd765b7752a12fdc337da3496d896f
Signed-off-by: Marc Schink <dev@zapb.de>
Reviewed-on: https://review.openocd.org/c/openocd/+/7956
Reviewed-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
In the unusual (and even incorrect) case of running the command
target create ...
before defining an adapter and the associated transport, the
command causes a segmentation fault. E.g.:
openocd -c 'target create cpu cortex-m -endian little'
Check that get_current_transport() returns a valid pointer before
referencing it.
Change-Id: I9796a7e92196ef3df5c7152b27c34102045dc9e7
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7962
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Tested-by: jenkins
Now it returns `resource not available`
Change-Id: Ifbbd468bdf62023850690eb96fe8a16f4114e915
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
The existing code didn't seem to work right at all. I have spike
modifications that exercise these new cases. I'll merge those once this
has merged.
Change-Id: I89bd336f34f1b208a76f25b6b41fe3877800765b
Signed-off-by: Tim Newsome <tim@sifive.com>
Compilers are good at optimizing, and with functions it's abundantly
clear what all the types involved are. This change means we don't have
to be super careful about the type of values because of what the macro
might do to them that might cause overflow.
The only place where the return type matters is in printf-style
functions, and I made get_value32() for those cases where a change was
needed.
This should set the stage for simply copying the latest debug_defines.h
from the debug spec build again.
Change-Id: I5fb19d0cfc1e20137832a7b344b05db215ce00e1
Signed-off-by: Tim Newsome <tim@sifive.com>
The reasoning for the change:
* `__func__` is part of C99, `__PRETTY_FUNCTION__` is GNU extension.
* `__PRETTY_FUNCTION__` is defined to be the same as `__func__` for C
sources by GCC documentation but differ for C++ sources (full
signature instead of just a name).
* Currently Clang does support `__PRETTY_FUNCTION__`, though it uses
GCC's C++ variant across C and C++.
Therefore using `__PRETTY_FUNCTION__` creates confusion and does not
provide any valueble information in the logs.
Change-Id: Ie0db6d73f602784b6752a30911dcef3dd7ee4594
Sometimes, the value from of some DMI scans has no meaning (e.g. when
`op` is read). Such values should not be decoded. To make the dumps more
consistent, `<no decoding available>` is printed when there is no
decoding for a register.
Change-Id: I415f06a5a80f2fc8fb8ab3f79132bdf0602c8ad6
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
This allows programmatically determining the debug reason.
Change-Id: I0c3e85cebb6dc28fc0fc212beca84a484ac654a5
Signed-off-by: Peter Collingbourne <pcc@google.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7952
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
Fixes below Sparse tool warning.
warning: symbol 'breakpoint_watchpoint_remove_all' was not declared.
Should it be static?
Signed-off-by: Erhan Kurubas <erhan.kurubas@espressif.com>
Change-Id: I4af1d5aa54abcb45f746b877513ba0b5fccbeb47
Reviewed-on: https://review.openocd.org/c/openocd/+/7955
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: Marek Vrbka <marek.vrbka@codasip.com>
Enhance `mips32_read_config_regs` to better detect MIPS32 ISA,
DSP, and FPU features, allowing user to get more detailed
target information.
Most of these information will be used in MIPS m/iAptiv support.
Change-Id: I23571a626ec64fa019acac91bdbfcb434373bfc1
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7911
Tested-by: jenkins
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
This patch forces a read of register from the target
in the get_reg command in case the register cache
does not hold a valid value at that moment.
Note that the command "reg" already handles it
correctly, no fix is needed there.
Change-Id: I75fad25188e94ee4e06162ab6d600ea24dbf590a
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7958
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: zapb <dev@zapb.de>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
By definition in `target/target.h`, `coreid` is not a unique identifier
of a target -- it can be the same for targets on different TAPs.
Change-Id: Ifce78da55fffe28dd8b6b06ecae7d8c4e305c0a2
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
This patch replaces ~(typeof(watchpoint->mask))0 with
WATCHPOINT_IGNORE_DATA_VALUE_MASK. This improves
readability and moves the RISCV target in line with
other targets.
Change-Id: I15ac4d4ee76098b304d9b22f720911ba4329c190
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Depending on configuration, the existing implementation of watchpoints is
rather inefficient for certain scenarios. Consider HW that:
1. triggers 0-3 can be used as instruction breakpoints
2. triggers 4-7 can be used as data breakpoints (watchpoints)
3. NAPOT triggers are not supported.
Now, consider that we have a pending watchpoint. And we perform a "step"
operation. According to the current implementation:
* OpenOCD will disable watchpoints
* Perform a single-step
* Will try to restore the original watchpoints. It will need 12 attempts
to find a suitable trigger: (8 attempts to try NAPOT, and another 4 to try
GE+LE).
This patch introduces a dedicated cache for requests to triggers. It
significantly speeds things up, since we cache failed attempts and no
additional interactions with HW is necessary.
Change-Id: Ic272895eaa763a7ae84d14f7633790afd015ca9d
Signed-off-by: Anastasiya Chernikova <anastasiya.chernikova@syntacore.com>
Otherwise it won't compile for me. Not sure why that doesn't affect the
automated builds.
Change-Id: Ic66c743e1698c4c0772e5601723cb5c711b4fa5c
Signed-off-by: Tim Newsome <tim@sifive.com>
Since DPC is WARL (same rules as MEPC according to
the specification), it is possible that
writes to it won't result in the exact value present.
Therefore, writes to it shouldn't be cached, same as
with other WARL registers.
Change-Id: I818c0cef9727b999b7d84b19f9f42cd706c99d69
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Added the ability to enter dimensionless registers
Change-Id: I1b781959ce4690ec65304142bd9a7c6f540b3e86
Signed-off-by: Anastasiya Chernikova <anastasiya.chernikova@syntacore.com>
Now:
1) If mmu is disabled, virt2phys succeeded and returns physical address
2) If mmu is enbaled, but translation fails, read/write_memory fails
Change-Id: I312309c660239014b3278cb77cadc5618de8e4de
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Add result check for mips32_pracc_read_regs in mips32_save_context.
Change-Id: Ie796d2b05a9feb11e246c2d0771b52cad4fb70db
Signed-off-by: Walter Ji <walter.ji@oss.cipunited.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7932
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Previously, when listing the watchpoints, OpenOCD printed
numbers 0, 1 and 2 representing READ, WRITE and ACCESS type
watchpoints.
This patch changes it to 'r', 'w' and 'a'. This increases the
clarity as what type the watchpoint actually is.
Change-Id: I9eac72dfd0bb2a9596a5b0c080a3f584556ed599
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7909
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
This patch adds the "all" option to the rwp command.
It removes all watchpoints, much like rbp all removes
all breakpoints.
Change-Id: Id58dd103085e558f17afa4a287888cf085566ca9
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7907
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
If wp/bp is missing at address rwp/rbp won't return zero code (on smp).
Now it fixed.
Fixes: 022e438292 ("target: Change policy of removing watchpoints/breakpoints.")
Change-Id: I3a3c245f7088fc23227b286d2191fc7f3edba702
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7910
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Add support for the ARM MCRR/MRRC instructions which require the use of
two registers to transfer a 64-bit co-processor registers. We are going
to use this in a subsequent patch in order to properly dump 64-bit page
table descriptors that exist on ARMv7A with VMSA extensions.
We make use of r0 and r1 to transfer 64-bit quantities to/from DCC.
Change-Id: Ic4975026c1ae4f2853795575ac7701d541248736
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Michael Chalfant <michael.chalfant@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/5228
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Currently, instruction cache is being invalidated in
arc_{un,}set_breakpoint() regardless of whether the breakpoint's type is
HW or SW. For SW breakpoints, this has no net effect as the caches are
flushed as a by-product of overwriting instructions in main memory and
is thus merely unnecessary; but for HW breakpoints this invalidation is
not preceded by a flush and might lead to loss of data. This patch
removes the invalidate() call altogether to correct this undesired
behavior for HW breakpoints.
With this patch applied, all supported HW breakpoint tests from the gdb
testsuite are now passing with the arc-openocd backend.
Change-Id: I3d252b97f01f1a1e2bf0eb8fb257bdab0c544bc2
Signed-off-by: Artemiy Volkov <artemiy@synopsys.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7767
Tested-by: jenkins
Reviewed-by: Evgeniy Didin <didin@synopsys.com>
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
The comment above armv8_dpm_read_current_registers() doesn't match
the implementation, as the function reads all the registers from
ARMV8_PC and above.
The registers currently read are not relevant to answer to the
usual GDB initial request through the 'g' packet. Plus the lack of
differentiation per core state (AArch32 vs AArch64) causes the
read of not existing registers in AArch32 triggering errors, as
tentatively fixed by https://review.openocd.org/5517/
Fix the code to read the registers initially required by GDB.
Modify the comment to report the register list in AArch32 and in
AArch64.
Keep the extra checks inside the read loop, even if they are
mostly irrelevant; this could prevent errors if someone needs to
extend the number of registers to read.
The current implementation of the register's description in
OpenOCD does not allow to discriminate among AArch32 and AArch64
registers. Add a TODO comment to highlight it.
Change-Id: Icd47d93c19a9e1694a7b51bbc5ca7e21a578df41
Signed-off-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7887
Tested-by: jenkins
Move `examine_attempted` flag to target struct to make it target specific.
`Info` messages for retry and `Error` messages for failure added.
Change-Id: Id2fbe7dc68d746c936c8412289d0d149fbd80d71
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
The secure microcontroller Infineon SLx2 uses a custom Cortex-M33.
The register CPUID reports value 0x490FDB00.
Reference link to the product:
Link: https://www.infineon.com/cms/en/about-infineon/press/market-news/2022/INFCSS202211-034.html
Change-Id: I8911712c55bd50e24ed53cf49958352f470027a5
Signed-off-by: Ahmed Boughanmi <boughanmi.external@infineon.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7879
Reviewed-by: Karl Palsson <karlp@tweak.au>
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
This patch adds error reporting when unknown image type is specified.
Previously, OpenOCD replied with an empty string.
Change-Id: I16220b1f5deb3b966a21731f0adf7911a78e8959
Signed-off-by: Marek Vrbka <marek.vrbka@codasip.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7883
Tested-by: jenkins
Reviewed-by: Jan Matyas <jan.matyas@codasip.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Add a new riscv specific commands to disable triggers
Change-Id: Ic1842085aa66851c740e0abcbfbe0adbe930920e
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
Extend riscv set_ebreak* commands.
Now it can be called without args to print current value.
riscv_ebreak* flags are moved to riscv_info struct.
Change-Id: Ib46e6b6dfc0117599c7f6715c7aaf113e63bd7dc
Signed-off-by: Kirill Radkin <kirill.radkin@syntacore.com>
and DP_SELECT_DPBANK.
Use the defined symbols instead of magic numbers.
Change-Id: I19c86b183e57e42b96f76eed180c0492cd67bee1
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7539
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
DP_SELECT_APSEL and DP_SELECT_APBANK is no more used in ADIv6.
Change-Id: I4176574d46c6dc8eb3fe3aef6daab6e33492c050
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: https://review.openocd.org/c/openocd/+/7538
Tested-by: jenkins
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
`riscv_debug_reg_to_s()` can be used to decode register value. If the
pointer to buffer is `NULL` it does not print anything, just returns the
length of the string.
The format is:
`<register_value> { <field_name>=<field_value_name or field_value>, ..., }`
e.g:
`0x400382 { version=2, ... ndmresetpending=false, }`
`0x321009 { regno=0x1009, ... cmdtype=0, }`
Change-Id: I63733d8d36385d89ca15de1a43139134bc488c4f
Signed-off-by: Evgeniy Naydanov <evgeniy.naydanov@syntacore.com>
For ARMv8, add AArch64 mdd and mwd support. AArch32 not supported.
Change-Id: I25490471e16943e5a67d7649595d77643aa9a095
Signed-off-by: Daniel Goehring <dgoehrin@os.amperecomputing.com>
Reviewed-on: https://review.openocd.org/c/openocd/+/7192
Reviewed-by: Antonio Borneo <borneo.antonio@gmail.com>
Tested-by: jenkins
When the target isn't halted, simply return an error. This used to be
purely internal code so an assert was appropriate. Now after some
refactoring and with unavailable harts you could get here when the hart
is unavailable. In that case the right thing is simply to return an
error message.
Change-Id: I49d26a11fe7565c645fd2480e89a2c35ea9b1688
Signed-off-by: Tim Newsome <tim@sifive.com>
Evgeniy Naydanov
Walter Ji
Antonio Borneo
Tomas Vanek
Erhan Kurubas
Evgeniy Naydanov
Sevan Janiyan
Jan Matyas
Kirill Radkin
Jan Matyas
ianst
Evgeniy Didin
Tarek BOCHKATI
Marc Schink
Henrik Nordström
Peter Collingbourne
Tim Newsome
Parshintsev Anatoly
Karl Palsson
Tobias Diedrich
Marek Vrbka
Anastasiya Chernikova
Florian Fainelli
Artemiy Volkov
liangzhen
Ahmed Boughanmi
Fedi Bouzazi
Daniel Goehring