aarch64: use symbolic constant for register count

Aarch64 has 34 registers, but use ARMV8_LAST_REG instead of
raw integer constant.

Change-Id: I86481899ade74f27fc90eff9f367d444c03e535e
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: remove armv7-a virt-to-phys code

Page table layout in aarch64 is very different from armv7-a layout.
Remove the incorrect handling, to be replaced correct armv8 code in a
later patch

Change-Id: I64c728a72a24f9f4177726ccc07a02a8ca0d56ce
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: clear breakpoint value register on removal

Not only null control but also value of the breakpoint when it is
removed.

Change-Id: Id99c7e3644729c64e563f1fa8b0577f350be6a98
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: correct breakpoint register offset

armv8 breakpoint register spacing is 16, not 4 as in armv7-a

Change-Id: I0d49d06878a0c9dab35cde478064e5366f01a8e0
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix cache identification

Use correct instructions to access CLIDR, CSSELR and CCSIDR.

Change-Id: I319b96c03a44fdb59fcb18a00f816f6af0261f0a
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix reading of translation table registers

Correctly access and parse aarch64 ttbcr.

Change-Id: I1b1652791a6b5200f58033925286292d838e8410
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix entry into debug state

- armv8 EDSCR has no ITR_EN bit, ITR is always enabled. Writes to this
  bit are ignored but we should not do them anyway
- use dpmv8 function to report the reason for debug entry
- WFAR is a 64bit register

Change-Id: I07b81ecf105ceb7c3ae2f764bb408eb973c1d1de
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: use symbolic opcodes instead of hex values

Use opcode definitions from armv8_opcodes.h where appropriate

Change-Id: Iead33fb8e62eb2dd2419ef8932f7d46c087f51a8
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix accesses to SCTLR_ELn register

The system control register has several instances, depending on the
exception level. Make sure to access always access the correct one.

Change-Id: I9e867f4dbd9625762042f20ed905064ea4e3270f
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix error recovery in aarch64_dpm_prepare

Flush DTRRX with a dummy read if it's full, clear sticky errors
by writing CSE bit to EDRCR register.

Change-Id: Ia42ae9d3859ba6cbe892d48584e21acdd4e25c84
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: formalize use of CTI in halt and resume

Use configured CTI base address instead of hardcoded value, if
available.
Use symbolic constants instead of raw hex offsets.
Trim halt and resume code to what is actually necessary.

Change-Id: I4997c2bcca7cebf5ad78859a6a12abe8639594ed
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix context and hybrid hardware breakpoints

Fix 64bit address setting
Fix register spacing (16 instead of 4)
Set HMC bit for all but linked context match breakpoints,
where the bit is ignored anyway

Change-Id: I48428f39154a6fe5fadc075ca918d1500a0bb241
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: deconflict debug register names

CPUDBG_ -> CPUV8_DBG_ for armv8 debug registers.

Change-Id: I3d24cc209309fa9bbeb5c3e6c88a572383c9360e
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Implement MA mode for bulk memory reads and writes

- 64bit addresses are supported
- Aarch32 state is supported

Change-Id: I8c37fa166954d09195d08c6963b8017194e350f5
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: make DCC read/write functions operate on struct armv8_common

Change the signature of aarch64_(read|write)_dcc[_64] to take a
"struct armv8_common *" as the context to operate on. No functional
change.

Change-Id: Ie501113f65ea22aff2eee173ec717f6908a63494
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: introduce dpm extension for ARMv8

Add or move ARMv8 related dpm function to their own source module

Change-Id: Id93d50be0b8635bd40ddb2a74fe8746ff840d736
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

target: add -ctibase config option in addition to -dbgbase

Some vendors don't fully populate the ROM table, e.g. BCM2357 (used in
Raspberry Pi 3) doesn't list CTI, however it is mandatory for halting
an ARMv8 core and therefore it's always present (and required),
regardless of the ROM table listing it or not.

Change-Id: Ia18a4f1b5b931ccd19805b188ebf737c837c6b54
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: fix reading of MPIDR

read MPIDR register through correct MSR instruction.

Change-Id: I7e2d00c2871191c4168b177a7a809443b0db4c82
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: add symbolic definitions for armv8 opcodes

To replace hexadecimal constants with descriptive names and increase
flexibility.

Change-Id: I6f7b6f045866ed8b9360f54b640ecdb307eebc51
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Correct target state for hardware step

When using hardware step for doing stepping, the existing DSCR records
the event as external debug request.  This will generate a SIGINT event
to GDB and causes it to stop the stepping process.
For aarch64, read DESR to check if the event is a hardware step and set
state to DBG_REASON_SINGLESTEP.
With this patch, GDB can now do source level stepping.

Change-Id: I1d06f819578c74b3ac17376c67f882adddea1f52
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Enable resuming with address

Enable resuming to an address.

Change-Id: I29c7d3b56f6cbf8b3cd02c93733fc96f45000af3
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Add instruction stepping support using hardware step

Use AARCH64's hardware step event to do stepping.

Change-Id: I2d029ceeadd381913d0c3355c8787b11dacff7f7
Signed-off-by: pierre Kuo <vichy.kuo@gmail.com>
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Enable halting debug mode on breakpoint set

Ensure that we allow halting debug mode after setting breakpoint

Change-Id: I6f0d7a4a4775a93c133fb1ec31dfe3324d9f7395
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Add hardware breakpoint support

Enable the use of hardware breakpoint on AARCH64.

Change-Id: I59caaa6d92ac60278af8938625b1790a1787372f
Signed-off-by: pierre Kuo <vichy.kuo@gmail.com>
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

aarch64: Add ARMv8 AARCH64 support files

Add new AARCH64 target and ARMv8 support files.
This is an instantiation from the cortex_a files but modified to support
64bit ARMv8. Not all features are complete, notably breakpts and single
stepping are not yet implemented.
Currently it lets you halt of the processors, resume, dump cpu
registers,
read/write memory and getting a stack trace with gdb.

> halt
invalid mode value encountered 5
target state: halted
unrecognized psr mode: 0x5
target halted in ARM state due to debug-request, current mode:
UNRECOGNIZED
cpsr: 0x600001c5 pc: 0x00093528
MMU: disabled, D-Cache: disabled, I-Cache: disabled

> targets
    TargetName         Type       Endian TapName            State
--  ------------------ ---------- ------ ------------------ ------------
 0* cpu0               aarch64    little cpu.dap            halted

> reg
===== arm v8 registers
(0) r0 (/64): 0x00000000FFFFFFED (dirty)
(1) r1 (/64): 0x00000000F76E4000
(2) r2 (/64): 0x0000000000000000
(3) r3 (/64): 0x0000000000010000
(4) r4 (/64): 0xFFFFFFC06E2939E1
(5) r5 (/64): 0x0000000000000018
(6) r6 (/64): 0x003A699CFB3C8480
(7) r7 (/64): 0x0000000053555555
(8) r8 (/64): 0x00FFFFFFFFFFFFFF
(9) r9 (/64): 0x000000001FFEF992
(10) r10 (/64): 0x0000000000000001
(11) r11 (/64): 0x0000000000000000
(12) r12 (/64): 0x00000000000000F0
(13) r13 (/64): 0x00000000EFDFEAC8
(14) r14 (/64): 0x00000000F6DDA659
(15) r15 (/64): 0x0000000000000000
(16) r16 (/64): 0xFFFFFFC0000F9094
(17) r17 (/64): 0x0000000000000000
(18) r18 (/64): 0x0000000000000000
(19) r19 (/64): 0xFFFFFFC00087C000
(20) r20 (/64): 0x0000000000000002
(21) r21 (/64): 0xFFFFFFC000867C28
(22) r22 (/64): 0xFFFFFFC000916A52
(23) r23 (/64): 0xFFFFFFC00116D8B0
(24) r24 (/64): 0xFFFFFFC000774A0C
(25) r25 (/64): 0x000000008007B000
(26) r26 (/64): 0x000000008007D000
(27) r27 (/64): 0xFFFFFFC000080450
(28) r28 (/64): 0x0000004080000000
(29) r29 (/64): 0xFFFFFFC00087FF20
(30) r30 (/64): 0xFFFFFFC000085114
(31) sp (/64): 0xFFFFFFC00087FF20
(32) pc (/64): 0xFFFFFFC000093528
(33) xPSR (/64): 0x00000000600001C5

And from gdb

(gdb) bt
 #0  cpu_do_idle () at /mnt/host/source/src/third_party/kernel/3.14/arch/arm64/mm/proc.S:87
 #1  0xffffffc000085114 in arch_cpu_idle () at /mnt/host/source/src/third_party/kernel/3.14/arch/arm64/kernel/process.c:107
 #2  0x0000000000000000 in ?? ()

Change-Id: Iccb1d15c7d8ace7b9e811dac3c9757ced4d0f618
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

arm_dpm: Add 64bit register handling.

Add various function to read/write ARMv8 registers.

Change-Id: I16f2829bdd0e87b050a51e414ff675d5c21bcbae
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

arm_dpm: Add new state ARM_STATE_AARCH64

Add new enum ARM_STATE_AARCH64 to the list of possible states.

Change-Id: I3cb2df70f8d5803a63d8374bf3eb75de988e24f8
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

server: Allow 64 address to be send over GBD server

Accept 64 bit addresses from GDB read memory packet.
Also allow breakpoint/stepping addresses to take 64bit values.

Change-Id: I9bf7b44affe24839cf30897c55ad17fdd29edf14
Signed-off-by: David Ung <david.ung.42@gmail.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

target: Add 64-bit target address support

Define a target_addr_t type to support 32-bit and 64-bit addresses at
the same time. Also define matching TARGET_PRI*ADDR format macros as
well as a convenient TARGET_ADDR_FMT.

In targets that are 32-bit (avr32, nds32, arm7/9/11, fm4, xmc1000)
be least invasive by leaving the formatting unchanged apart from the
type;
for generic code adopt TARGET_ADDR_FMT as unified address format.

Don't silently change gdb formatting here, leave that to later.

Add COMMAND_PARSE_ADDRESS() macro to abstract the address type.
Implement it using its own parse_target_addr() function, in the hopes
of catching pointer type mismatches better.

Add '--disable-target64' configure option to revert to previous 32-bit
target address behavior.

Change-Id: I2e91d205862ceb14f94b3e72a7e99ee0373a85d5
Signed-off-by: Dongxue Zhang <elta.era@gmail.com>
Signed-off-by: David Ung <david.ung.42@gmail.com>
[AF: Default to enabling (Paul Fertser), rename macros, simplify]
Signed-off-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>

Restore normal development cycle

Signed-off-by: Paul Fertser <fercerpav@gmail.com>

The openocd-0.10.0 release

Signed-off-by: Paul Fertser <fercerpav@gmail.com>

NEWS: last pre-release changes

Change-Id: I93203717f9096880298c10efebf05d59f888f34b
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3954
Tested-by: jenkins

arm_dpm: avoid duplicating the register cache

This bug was already attempted to fix in an earlier patch but
merging the "defer-examine" feature caused a regression, which this patch
tries to fix again.

Change-Id: Ie1ad1516f0d7f130d44e003d6c29dcc1a02a82ca
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/3951
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Restore -dev suffix

Signed-off-by: Paul Fertser <fercerpav@gmail.com>

The openocd-0.10.0-rc2 release candidate

Signed-off-by: Paul Fertser <fercerpav@gmail.com>

NEWS: mention SWO capture facility of the new J-Link driver

Change-Id: I4fc150584f3f0c5df5c39ae512125b42893331f0
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3948
Tested-by: jenkins

jtag: core: report TRST and SRST as asserted only when they really are

When SRST is never enabled, the global jtag_srst variable is left at its
initial state, that is, -1, and it does _not_ mean SRST is currently
asserted. Same about TRST.

Fixes "reset halt" in cases when srst_pulls_trst but srst usage is not
enabled.

Change-Id: I8d2e9120479de4cfbf5561033926c9ef945eecc9
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3943
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>

flash/nor: fix doc/help and range test for flash protect

Commit 77a1c01ccbb1150ffe749a7373cf6c4dc15ecad0 introduced infrastructure
for utilizing protection blocks of different size than erase sector.
Parts of doc/help kept reading 'sector' instead of 'protection block'.
flash_driver_protect() parameter range testing did not switched
to bank->num_prot_blocks.
This change fixes it.

Change-Id: Iec301761190a1a1bcc4cb005a519b9e5e4fede51
Reported-by: Mark Odell <mark@odell.ws>
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/3917
Tested-by: jenkins
Reviewed-by: Mark Odell <mrfirmware@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl at91samdXX.cfg: partially revert change 3722

Commit 25d7ba19c9e70cf5b912f660cf6aaa93d9ca120f introduced a problem
with 'reset halt' due to setting srst_pulls_trst:

Error: cortex_m.c:595 cortex_m_halt(): can't request a halt while
  in reset if nSRST pulls nTRST

Sorry, I don't know why I overlooked it when I tested #3722.

Change-Id: I41e9473dd91a86d93cf3e78b1fbbdfe1dd188d83
Reported-by: Ladislav Laska <laska@kam.mff.cuni.cz>
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/3942
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

adi_v5_jtag: make sure SSTICKYERR is cleared after a POR

Don't terminate the transaction end-check early if debug power-loss
was detected, without clearing SSTICKYERR.

Change-Id: I83b6a4a20523eea42e48a15297f972a730aa21a8
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/3947
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

doc: Improve ftdi driver section

Fix typo and a sentence that sounds strange since the ft2232 driver
was removed.

Add documentation for the SWD signal usage.

Remove the text mentioning a default VID:PID which doesn't exist.

Change-Id: I27eba571f2f7c46bdb6ae623ab285595018e99d9
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3935
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

helper,rtos,server: Restore missing warning flags

These libraries override the used CFLAGS without adding the
defaults. This didn't have any effect until change
http://openocd.zylin.com/3870 (ef4c139). Restore by adding
AM_CLAGS to the per-target CFLAGS.

Interestingly, automake seems to clear the CFLAGS for the target
even if the override variable is only mentioned within a non-active
conditional branch, such as the IS_MINGW for the affected libraries.

Change-Id: I805206865e59e3fa33a7ea3c0d3472e51219351c
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3927
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jtag: drivers: build usb_blaster only if enabled

Fix build failure when libusb dev files are not available and
so usb_blaster is disabled.

Change-Id: I4cda7df689cdb4b62b733cbbab813241cf561e29
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3920
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

libjaylink: Update to 0.1.0 release

Change-Id: I5a905ccda346c09187f7f8e8229ae71d3353128f
Signed-off-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-on: http://openocd.zylin.com/3946
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Restore -dev suffix

Signed-off-by: Paul Fertser <fercerpav@gmail.com>

The openocd-0.10.0-rc1 release candidate

Signed-off-by: Paul Fertser <fercerpav@gmail.com>

NEWS: prepare for new release

Change-Id: I141ff17f6dcab7e164a19c91b975ddd455d26dbc
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3854
Tested-by: jenkins

doc: fix underfull and overfull boxes

This is needed to generate PDF user manual without typesetting
artifacts.

Change-Id: Ibcbd804dac2b9415459327f53f6fad0dc38fa5c6
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3919
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

doc: add a cross-reference to GDB threading commands

Change-Id: If4766ba0053fc94cd47495e442fcf288241af218
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3557
Tested-by: jenkins

target: armv7a: remove unused level_num field from armv7a_cachesize

This was never used and produces warnings on some systems.

Change-Id: I48d2c5b79890bb2d70c5fae95278b8eb62743398
Reported-by: Tommy Murphy <tm1234@users.sf.net>
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3891
Tested-by: jenkins

tcl: target: imx5{1,3}: remove bogus tapenable command

This seems to be a leftover from borrowing from omap3* configs. Since
SJC is not enabling or disabling the DAP tap, and the tap is always
available, the extra tapenable command causes warnings on startup
(can't enable what's already enabled).

Change-Id: I7514436d565aa5b91876dbdab547956f36dcab77
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3892
Tested-by: jenkins

target: cortex_a: fix segfault when SPSR is not properly handled

OpenOCD doesn't (yet) know how to handle HYP mode properly so spsr
register is not getting initialised when OpenOCD connects to a target
stopped in this mode.

Reported on IRC by thinkfat and nearffxx.

Change-Id: I4bda9ba0c582c8e9cacefe708cc4a3d947151f84
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3906
Tested-by: jenkins
Reviewed-by: Matthias Welwarsky <matthias@welwarsky.de>
Reviewed-by: Chengyu Zheng <chengyu.zheng@polimi.it>

README: fix a typo

Change-Id: I1ad1ba001f8fad0fffa9b6bd5b21dd58be4f6181
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3902
Tested-by: jenkins
Reviewed-by: Chengyu Zheng <chengyu.zheng@polimi.it>

Search for scripts relative to the executable on all(?) platforms

Add a helper to hide the platform-dependent method to get a
canonical, absolute, /-separated path to the executable.

Use this and the relative path from BINDIR to PKGDATADIR to
construct a search path that finds the scripts even if the
installation dir is moved, as long as the structure below $prefix
is maintained.

This method should fully support all the tricks you can to with
autotools to customize the installed layout such as overriding the
default directories at configure-time and overriding the configured
directories at build-time.

The exe path detection methods are combined from
http://openocd.zylin.com/3388 by Rick Foos and
http://openocd.zylin.com/3537 by Steven Stallion, as well as tips
found all over internet.

Change-Id: Ifc9cc9dd0bf52fbd67b1b0f2383318cda0c422c4
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Signed-off-by: Steven Stallion <sstallion@gmail.com>
Reviewed-on: http://openocd.zylin.com/3889
Tested-by: jenkins
Reviewed-by: Rick Foos <rfoos@solengtech.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jimtcl: update to 0.77, the current version, enable only specific modules

This is needed to fix build on aarch64 to get updated config.guess.

Because some newer JimTcl modules that get enabled are failing to
build on some of the platforms OpenOCD supports, enable only a
fixed set that shouldn't cause any issues.

We also disable running Jim Tcl tests on distcheck because they never
worked (before 85358e479e5bbf295a5aaf743f3c29a5e1a4fd1c Jim commit)
but didn't break Jenkins build; with newer version they're getting run
and fail due to limited permissions for filesystem and networking
operations.

Change-Id: I0b6c6a00bf2cf0902bbb01c9c0224fe93c74ac94
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3700
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

contrib: udev: modify rules for users physically in front of machine (TAG+="uaccess")

Modern desktop systems with systemd recommend this way to give users
access to devices. We change permissions to sane value along the way.

See e.g. https://lists.debian.org/debian-devel-announce/2016/11/msg00008.html

We also change the filename to put it in order with 71-seat.rules, 60
is default for dh_installudev so pick that as a reasonable value.

Change-Id: I15f6fb1b6be853ac097d0ca91955609f9e5eb9cf
Signed-off-by: Jiri Kastner <cz172638@gmail.com>
Reviewed-on: http://openocd.zylin.com/2804
Tested-by: jenkins
Reviewed-by: Robert Jordens <jordens@gmail.com>
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

jtag: drivers: ulink: fix uninitialised variable warning

This prevents clean build when --enable-verbose-jtag-io is used.

Change-Id: I5c9e6968cfa425b1f6f92f59156b6ae38cb9af18
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3907
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

board: add configuration for freescale twr-vf65gs10 system module

This patch contains configuration for the twr-vf65gs10 system module
of the Freescale (NXP) tower system. It provides support for both the
CMSIS-DAP available via USB and the Cortex+ETM connector
on J5. The configuration also contains DDR and clock init code
hooked into the reset-init event handler.

Change-Id: I68303e0038e137dcadc57525e662428769fb69f2
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/3564
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

stm32l0.cfg: Add examine-end event like on other STM32 targets.

Enable debug in standby/stop/sleep.
Stop watchdogs during halt.

Change-Id: I8383a191cd897118bd88bf78528d05943f3a368e
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/3882
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Fail on enabling sysfsgpio on non-linux OS

Change-Id: I5036d8eca6a6e6ed47e3b9ff78bc3a3bf34491fb
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3888
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Deprecate ZY1000, ioutils and oocd_trace

These have not been maintained for a very long time so I say it's
time we remove them.

Change-Id: Ic091978d734be09347e271736df0f7f7f9095243
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3885
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: De-duplicate code in AC_CHECK_HEADERS

Combine all checks that have the same required includes.

Change-Id: I9b4567766e037922c2ffb737f32e9c12af43bf0e
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3884
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Convert to non-recursive make

Change-Id: I11f8bc8553957e2ff083c09e72e16881e4d3bb6f
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3865
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Consolidate platform defines

Change-Id: Iebc775baadeeb5e963fb48c0a9e8d87d26d9035a
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3863
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Don't set user variables

CFLAGS is a user variable and shouldn't be set in either
configure.ac or Makefile.am because that makes it impossible to
override compiler flags at configure or make time.

AC_SUBST the default warning flags into AM_CFLAGS instead, which
is emitted before CFLAGS on the compiler command line.

Do the same thing with a mingw-specific flag as well.

Change-Id: I6ac3a3c8b7e0abf6f0e60ea239b549d68ea6d370
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3870
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Add USE_* conditionals for hidapi and libjaylink

For consistency with other libs.

Change-Id: Id37329a00db5ab668f4c2c9ce53d2c41976649b1
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3869
Tested-by: jenkins
Reviewed-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Use same case for libjaylink as for other libs

Change-Id: I60d91a0543d9b8c580254cdc5f04b2e60209b98b
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3868
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target: LS1_SAP: fix "declaration of 'read' shadows a global" warning

Change-Id: I7070193819cc134f9fe1427c20a11160ec415ccb
Reported-by: Jens Bauer <jens-lists@gpio.dk>
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3877
Tested-by: jenkins

target: Replace malloc+manual zeroing with calloc

Change-Id: I3c782c34b59cb36b8ca1d36e9c804c67bae5cb45
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3667
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Replace shell constructs with M4sh macros

AS_IF and AS_CASE should be used instead of if and case to properly
handle macros in conditional branches.

Also guard all tests against empty variables and embedded spaces which
makes it possible to remove some variable defaults.

Change-Id: Ib675177690d406a80ee98dd5d105296a7cdc7b8a
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3669
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jlink: Output libjaylink version

Output the package version of libjaylink to ease debugging.

Change-Id: I3b9da6d046d140ba850056c98e67bed22c885ee0
Signed-off-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-on: http://openocd.zylin.com/3887
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

libjaylink: Update to latest master branch

The changes since the last update are mainly related to the build
system, preparations and cleanups for the upcoming release.

If there are no bugs reported, this will be the last update before the
libjaylink 0.1.0 release.

Change-Id: I4cec9bb61159f6153690aaf39c1d12ba0baacf9b
Signed-off-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-on: http://openocd.zylin.com/3886
Tested-by: jenkins
Reviewed-by: Xiaofan <xiaofanc@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jlink: Add EMUCOM support

EMUCOM is a way to communicate with a J-Link device via so called
channels. A channel can either be read or written in a single
operation.

Beside the reserved channels for SEGGER, there are channels available to
implement vendor and/or device specific functionalities. For example,
EMUCOM is used on many starter and development kits from Silicon Labs to
access power measurements and various other information and settings.

Change-Id: I6094109c043b34aed4a40ceabe71f30ff896bf1d
Signed-off-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-on: http://openocd.zylin.com/3794
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

flash/nor/efm32: Support flash size smaller than 32k

The current implementation fails on devices with less than 32k of
flash (such as several devices in the Zero Gecko family) because
the 'assert' assumes (incorrectly) that the number of flash banks
will always be >= 32.

This change ensures that at least one word of lock bits is always read
in order to support devices with less than 32k of flash.

Signed-off-by: Kevlar Harness <software@klystron.com>
Change-Id: I59febe2cb690c893a5057a5f72918e146cf2afe4
Reviewed-on: http://openocd.zylin.com/3806
Tested-by: jenkins
Reviewed-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

xscale: Move debug handler to contrib/loaders

Avoid special rules to generate array at compile time by shipping
the generated file. Convert to Makefile build like the other
loaders.

Change-Id: I5a05edddcfaff3d395086cd3aa33120f8a7aa9dc
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3864
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

cortex_a: fix reset for SWD transport

Change b0698501b0e789091e9f9f1427b32af304888769 fixed
reset for i.MX6 and TI Sitara SoCs but broke reset for
cortex-a targets that use SWD. This patch is a work-
around that forces asserting SRST when SWD is used.

Change-Id: I7e39f2a469b9b4b2b74ad48ba49f2eeb58528921
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/3641
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Remove build date from banner for releases

In support for reproducible builds, see
https://wiki.debian.org/ReproducibleBuilds

Fixes Debian bug #834316.

Change-Id: Id81ec72a87bf6dd99abfd2a0ae074658111bc9a3
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3866
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Remove unused BUILD_BITQ and BUILD_BITBANG define

Change-Id: Ibfb5e414dd77d2ce4b835007722eb18da3820f3a
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3668
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Remove conflicting CMSIS_DAP conditional

It's already defined by PROCESS_ADAPTERS.

Change-Id: I3bbcb0f77c645e9c5a98c2d5069c178358d1516b
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3239
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

configure: Detect libftdi using pkgconfig

Auto-enable libftdi adapter drivers if found, just like libusb and
hidapi based adapters.

Change-Id: I71a10c90c8b999c4cae9b4f8fb5e97971d03083b
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3238
Tested-by: jenkins
Reviewed-by: Xiaofan <xiaofanc@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

cortex_m: allow setting debug ap during create

This patch adds a Cortex-M private configuration option
that allows setting the acess point during target
creation. This circumvents situations in hybrid systems
when the correct access point can not be automatically
detected.

Change-Id: If313a5250e6e66509bb9080f3498feab7781dced
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/3639
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jtag/drivers/openjtag: Add support for Cypress CY7C65215

The Cypress CY7C65215 Dual Channel USB-Serial Bridge Controller [1]
understands the OpenJTAG protocol over a proprietary USB interface.

This patch adds support for the CY7C65215 to the openjtag interface
driver.  A new configuration option, `openjtag_variant`, allows to
select the transport to use.

Libusb (1.x or 0.1) is now a hard dependency of the openjtag driver.
This should not be a big issue as libftdi also depends on it.

[1] http://www.cypress.com/?rID=82870

Change-Id: I55ffb3fd9e006eb311e405d9fb836bb119644bfd
Signed-off-by: Vianney le Clément de Saint-Marcq <vianney.leclement@essensium.com>
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Reviewed-on: http://openocd.zylin.com/2805
Tested-by: jenkins
Reviewed-by: Jiri Kastner <cz172638@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

helper: Code cleanup for hexify()

Simplify hexify() and do not longer use 0 as special case for the
parameter 'count' to determine the string length of the binary input.
Instead, use strlen() outside of the function if needed.
Additionally, fix the return value and return the length of the
converted string. The old function always returned 2 * count.

Also, use more appropriate data types for the function parameters and
add a small documentation.

Change-Id: I133a8ab786b8f7c1296afcaf9c0a0b43881e5112
Signed-off-by: Marc Schink <openocd-dev@marcschink.de>
Reviewed-on: http://openocd.zylin.com/3793
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>

jtag: drivers: bcm2835gpio: set 4ma drive, slow slew rate

Both the drive strength and slew rate are currently set to very high
values.  This causes the waveforms to overshoot and be less reliable
when operating at high speed.

This patch lowers the slew rate and sets the drive strength to 4ma,
improving the waveform output.

Change-Id: I761d35cd64bc54a9e94043904ef00a003b056af0
Signed-off-by: Sean Cross <sean@xobs.io>
Reviewed-on: http://openocd.zylin.com/3707
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Philipp Guehring <pg@futureware.at>

gdb_server: support qXfer:threads:read packet

This patch adds support for the qXfer:threads:read packet. In addition
to providing a more efficient method of updating thread state, recent
versions of GDB (7.11.1 and up) can also report remote thread names.
While thread names are not enabled in this patch due to its limited
applicability at the moment, it can be enabled at a later date with
little effort.

As a part of revamping how threads are presented to GDB, extra info
strings for each of the supported RTOSes were updated to match
conventions present in the GDB source code. For more information, see
remote_threads_extra_info() in remote.c. This results in a much smoother
experience when interacting with GDB.

It is also worth mentioning that use of qXfer:threads:read works around
a number of regressions in older versions of GDB regarding remote thread
display. Trust me, it's great.

Change-Id: I97dd6a93c342ceb9b9d0023b6359db0e5604c6e6
Signed-off-by: Steven Stallion <stallion@squareup.com>
Reviewed-on: http://openocd.zylin.com/3559
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

semihosting: support fileio operation

This patch adds support for bridging semihosting to GDB's File-I/O
remote protocol extension. For the most part operations match up 1:1,
however some require a working area to complete successfully, namely
operations that devolve to read, stat, and gettimeofday.

A new command was added to enable support for fileio named `arm
semihosting_fileio`, which ensures that the default behavior remains
intact for those that prefer it.

Finally, redundant logging was removed from the target_arch_state
function; this permits ARM targets to quiesce log output when polling
for a fileio reply. This prevents filling the logs with halt/resume
messages when using semihosting fileio.

Change-Id: Ifbb864fc2373336a501cc0332675b887b552e1ee
Signed-off-by: Steven Stallion <stallion@squareup.com>
Reviewed-on: http://openocd.zylin.com/3566
Tested-by: jenkins
Reviewed-by: Steven Stallion <sstallion@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

at91sam4: Add missing SAM4S family CHIPIDs and remove FWS=6.

Add missing CHIPID values for all SAM4S parts listed in revision K of
Atmel-11100-32-bit Cortex-M4-Microcontroller-SAM4S_Datasheet.pdf. I have
also removed the FWS=6 workaround, as this appears to be a copy-paste error
from the SAM3X family.

Change-Id: I1ce1d82911f39d6fcb8f04034f5c9c9bf2818466
Signed-off-by: Owen Kirby <oskirby@gmail.com>
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/3837
Tested-by: jenkins

at91sam4: Add flash description and chipid for SAM4Cxx variants.

Chip ID and flash layout taken from
Atmel-11102F-ATARM-SAM4C32-SAM4C16-SAM4C8-SAM4C4-Datasheet_27-Mar-15
and tested on a SAM4C32-EK (rev A).

Change-Id: I68aae5b60994c0b5964ea9031d40bc76ba025675
Signed-off-by: Owen Kirby <oskirby@gmail.com>
Reviewed-on: http://openocd.zylin.com/3527
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>

jtag: drivers: cmsis-dap: fix operation with SAMD10 Xplained board

Change-Id: I9164edeb83c9a9106e74798e98b9517177e83358
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/3856
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>

helper: Update jep106 database to JEP106AT

Change-Id: I2dac416189d16938597c073fd35ad654bca7484c
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/3871
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Cortex-R: Force usage of physical memory read/write since there is no MMU

Change-Id: I69d6e6301ab66744258fe650a76c7241bebe00a8
Signed-off-by: Evan Hunter <ehunter@broadcom.com>
Reviewed-on: http://openocd.zylin.com/3203
Tested-by: jenkins
Reviewed-by: Girts Folkmanis <opensource@girts.me>
Reviewed-by: Matthias Welwarsky <matthias@welwarsky.de>
Reviewed-by: Jiri Kastner <cz172638@gmail.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

rtos: add support for uC/OS-III

This patch introduces RTOS support for uC/OS-III. Currently, only
FPU-less ARM Cortex-M targets are supported. Due to the configurability
of the RTOS, an OpenOCD-specific file must be linked along with the
project to determine the correct offsets within the OS_TCB structure.

In addition to the above, a crash was fixed in rtos_get_gdb_reg_list
such that RTOS support could be used between resets without restarting
OpenOCD and support for the Hg packet was cleaned up.

Change-Id: Ide004a689e6b886185df665c00fb644629eb31d1
Signed-off-by: Steven Stallion <stallion@squareup.com>
Reviewed-on: http://openocd.zylin.com/3556
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target: Add verify_image_checksum command

This avoids the secondary binary search if the checksum is different

Change-Id: I986ba7687cea76f30e37a6bca58aabde18198263
Signed-off-by: Evan Hunter <ehunter@broadcom.com>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/2869
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl/board: Add Infineon XMC4300 Relax EtherCAT Kit config

Tested with "J-Link Lite-XMC4200 Rev.1 compiled Jul 18 2014 17:28:26".

Change-Id: I31482734af7621593a244ffac9dd223408470609
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/3876
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

xmc4xxx: Add support for XMC4300

XMC4300 can reuse the existing XMC4700/XMC4800 sectors support.
Add support for XMC4300 AA to the info command.

Change-Id: Id929a51d20c73bd869a4457ffedc48ad5fa3f2df
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/3875
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

dap_dp_init: remove loop

current loop sounds to me like 'we don't know what we do, let's do it ten times, maybe we will have luck'.
should be enough to 'ping' debug port using reading CRTL_STAT.

tested on cortex-a8, snapdragon, jetson k1, cortex-r5, cortex-r4

Change-Id: Ibc62ac1eca06c141f4fccd5de7b11350ca1f35fd
Signed-off-by: Jiri Kastner <cz172638@gmail.com>
Tested-by: Jiri Kastner <cz172638@gmail.com>
Reviewed-on: http://openocd.zylin.com/3193
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Liviu Dudau <liviu@dudau.co.uk>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

cortex_a: remove partnum magic from arp_examine and dbginit

Depending on the Debug implementation the "OS Lock" feature might be
implemented or not. It is not actually depending on the part number of the
implemented ARM core but on the DBGOSLSR.OSLM bits. This patch removes
querying the part number and implements proper parsing of OSLM. Result is
a more generic approach that will work out-of-box on more devices.

Change-Id: I79e052869c2f9af1d7fdedef42faddb7292e7332
Signed-off-by: Matthias Welwarsky <matthias@welwarsky.de>
Reviewed-on: http://openocd.zylin.com/3213
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Add -defer-examine option to target create command

The '-defer-examine' option to target create allows declaring targets
that are present on the chain, but not fully functional.  They will
be skipped by the initial arp_examine as well as arp_examine after
reset.

Manual examine using 'arp_examine' is needed to examine them, with the
idea that some kind of actions is neeed to bring them to a state where
examine will succeed (if at all possible).

In order to allow value less options to target command, I had to relax
the goi.argc check in jim_target_configure().

Change-Id: I9bf4e8d27eb6476dd9353d15f48965a8cfd5c122
Signed-off-by: Esben Haabendal <esben@haabendal.dk>
Signed-off-by: Matthias Welwarsky <matthias.welwarsky@sysgo.com>
Reviewed-on: http://openocd.zylin.com/3076
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>

tcl: Support for Analog Devices ADSP-SC58x

target tcl to enable debugging of ARM Cortex-A5 on ADSP-SC58x

Change-Id: I378f8b94b7d6d6b9d0567985abc0e36aea7c8dea
Signed-off-by: Peter Lawrence <majbthrd@gmail.com>
Reviewed-on: http://openocd.zylin.com/3125
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-by: Matthias Welwarsky <matthias@welwarsky.de>

flash/nor: at91samd protection bits write fix

Flash protection set on a device with MANW=1 was lost after reset.
Since #2903 the driver honored MANW bit and issued Write Page command just
for main flash write. This change adds similar technique to
samd_modify_user_row().

Minor code improvements:
samd_check_error() returns error code corresponding to error type
instead of bool.

samd_check_error() does not clear STATUS register if no error bit is set.

Eliminated double error check in call sequence samd_issue_nvmctrl_command()
folowed by samd_check_error().

Missing error code ERROR_FLASH_PROTECTED added to src/flash/common.h.

Change-Id: Icf59ab8803305d0cb3170c8a5089b8f9828b99f8
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/3550
Tested-by: jenkins
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

1024 SHA256:YKx8b7u5ZWdcbp7/4AeXNaqElP49m6QrwfXaqQGJAOk gerrit-code-review@openocd.zylin.com (DSA)
384 SHA256:jHIbSQa4REvwCFG4cq5LBlBLxmxSqelQPem/EXIrxjk gerrit-code-review@openocd.org (ECDSA)
521 SHA256:UAOPYkU9Fjtcao0Ul/Rrlnj/OsQvt+pgdYSZ4jOYdgs gerrit-code-review@openocd.org (ECDSA)
256 SHA256:A13M5QlnozFOvTllybRZH6vm7iSt0XLxbA48yfc2yfY gerrit-code-review@openocd.org (ECDSA)
256 SHA256:spYMBqEYoAOtK7yZBrcwE8ZpYt6b68Cfh9yEVetvbXg gerrit-code-review@openocd.org (ED25519)
+--[ED25519 256]--+
|=..              |
|+o..   .         |
|*.o   . .        |
|+B . . .         |
|Bo. = o S        |
|Oo.+ + =         |
|oB=.* = . o      |
| =+=.+   + E     |
|. .=o   . o      |
+----[SHA256]-----+
2048 SHA256:0Onrb7/PHjpo6iVZ7xQX2riKN83FJ3KGU0TvI0TaFG4 gerrit-code-review@openocd.zylin.com (RSA)