flash/nor/lpc2000: properly free working area used in get_lpc2000_part_id()

The IAP working area needs to be freed here, just like in all the
other driver functions since an automatic local variable is used to
store a pointer to it.

This was reported by quitte (Jonas Meyer) on IRC as a strange totally
unrelated segfault after doing certain operations (leading to target
reset) from GDB. He has provided me with remote access to the specific
machine and configuration that exposed the issue, and after some
debugging it became apparent that a auto local variable (holding the
gdb connection pointer) gets overwritten somehow. Placing an
appropriate breakpoint just before the event and using a watchpoint
made the cause apparent: reset lead to freeing of all working areas,
and there was one holding a pointer to a variable that was auto local
in get_lpc2000_part_id().

Change-Id: I7e634d890135ca0f3b4b311e09e8385a03982bd6
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2695
Tested-by: jenkins

jtag/startup.tcl: fix regression with autoselecting JTAG

This regression was introduced with d90b86d8. "transport select" doesn't
throw an error anymore and autoselects the first available transport on
its own.

Reported by moyix on IRC.

Change-Id: I3f303c0372e915931cca4b28af450694abc1a63e
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2693
Tested-by: jenkins

ChibiOS: fix crash on auto detection

The detection framework assumes rtos->symbols is dynamically allocated,
an assumption that the ChibiOS variant breaks by providing a raw statically
allocated symbol list.

Change-Id: I379bcc2af99006912608ddd3f646ff7085606f47
Signed-off-by: Richard Braun <rbraun@sceen.net>
Reviewed-on: http://openocd.zylin.com/2597
Tested-by: jenkins
Reviewed-by: Stian Skjelstad <stian@nixia.no>
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Tcl exception codes cleanup, shutdown command amendments

This patch might influence openocd Tcl commands behaviour in subtle
ways, please give it a nice testing.

The idea is that if an OpenOCD Tcl command returns an error, an
exception is raised, and then the return code is propogated all the
way up (or to the "catch" if present). This allows to detect
"shutdown" which is not actually an error but has to raise an
exception to stop execution of the commands that follow it in the
script.

openocd_thread special-cases shutdown because it should then terminate
OpenOCD with a success error code, unless shutdown was called with an
optional "error" argument which means terminate with a non-zero exit
code.

Change-Id: I7b6fa8a2e24c947dc45d8def0008b4b007c478b3
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2600
Tested-by: jenkins
Reviewed-by: Juha Niskanen <juha.niskanen@haltian.com>
Reviewed-by: Jens Bauer <jens@gpio.dk>
Reviewed-by: Oleksij Rempel <linux@rempel-privat.de>

helper: shutdown command should return with 0 exit status

Commit a35712a85c42 caused a regression where command

openocd -c "echo a1; shutdown; echo a2"

always returned non-zero exit status to operating system,
even when commands before shutdown all succeeded. This patch
attempt to fix this.

Change-Id: I3f478c2c51d100af810ea0171d2fd4c8fcc657f3
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>
Reviewed-on: http://openocd.zylin.com/2589
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jtag/tcl: fix incorrect memcpy in jim_newtap_expected_id

Found by clang static checker.

On the very first call of jim_newtap_expected_id() pTap->expected_ids
and expected_len are null, and there's nothing to copy. This patch
changes this cryptic code to use realloc() instead.

Change-Id: Ic0b5140d08257a906f15b55a2ae64db7bc06d5f1
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2562
Reviewed-by: Stian Skjelstad <stian@nixia.no>
Tested-by: jenkins

turtelizer2-revC: Tested with real hardware.

Also some hints around deprecated tcl/interface/turtelizer2.cfg added.

Change-Id: Ifa57b49febffaeddd5d8ff0a48833d3544927b10
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2584
Tested-by: jenkins
Reviewed-by: Harald Kipp
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

lpc2000: Removed cmd51_can_xxxxb variables.

Current flash driver can fail to write across the sector boundary.
This patch fixes "thisrun_bytes" set to "cmd51_dst_boundary" value instead of "cmd51_can_xxxxb"

Referred to SevenW's post and fix,thanks.
http://sourceforge.net/p/openocd/mailman/openocd-devel/thread/fa32e7d6654df22470dc5f4a3dbc984a%40familiebenschop.nl/#msg33594152

Tested on below listed chips.
LPC810,LPC811,LPC812,LPC824,LPC1115,LPC1343,LPC1347,LPC1227,
LPC1769,LPC1788,LPC54102,LPC4088 and LPC2388.

Change-Id: If1c6a1daa58ca27c405bd959a784e599a7a8f4d4
Signed-off-by: Nemui Trinomius <nemuisan_kawausogasuki@live.jp>
Reviewed-on: http://openocd.zylin.com/2607
Tested-by: jenkins
Reviewed-by: Jens Bauer <jens@gpio.dk>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl/board: Add Infineon XMC 2Go config

Tested with "J-Link Lite-XMC4200 Rev.1 compiled Jan 10 2014 20:31:33".

Change-Id: Iefa9185372341d889db2b5f1f93bce126450b535
Cc: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2564
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

Added Atmel SAM4SA16C chip (chip ID 0x28a70ce0).

Change-Id: I45e8e807a36c39940b910b3edb40698c7d8dabd6
Signed-off-by: Patrick Noffke <patrick@noffke.me>
Reviewed-on: http://openocd.zylin.com/2625
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/board: Add Infineon XMC1100 Boot Kit config

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

Change-Id: Icc03172cefe38f2217bf44a73f94f8a6fb93dfba
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2472
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/target: Add Infineon XMC1000 family target config

Basic target config, verified against XMC1100 AA/AB, XMC1200 AA/AB,
XMC1300 AA/AB manuals.

The default adapter_khz was tested with the XMC1100 Boot Kit.

Change-Id: Iff6ed52d875ccb83c0d4ff0d555b90e0f28b860c
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2471
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/board: Add Spansion SK-FM4-176L-S6E2CC config

Works best after update to firmware v2.3.

Change-Id: Id2d3a0ae28bba014ee5338df9280fe39773c3398
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2570
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/target: Add Spansion FM4 config

Assembled by trial-and-error for an S6E2CC.

Change-Id: I317c12d24c9075ce3de286455fa3ee45731c5c81
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2569
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

startup: Fixed measure_clk to return kHz instead of MHz.

The original code had iterated 10,000,000 times and taken the elapsed
time divided by 10,000, to yield kHz which is mathematically correct
only if we were measuring time in seconds, but we are measuring time in
 milliseconds, so the correct divisor is actually 10,000,000.  Previous
code would report 0.500 for actual measured speed of 500 kHz.

Change-Id: Iba4c4961fe3973e7ccfa6dfa11d606a966ceb50c
Signed-off-by: Ed Beroset <beroset@ieee.org>
Reviewed-on: http://openocd.zylin.com/2573
Tested-by: jenkins
Reviewed-by: Jens Bauer <jens@gpio.dk>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

em357: Corrected EM357 support including errata details

Original submitted code had only been tested with em358, but testing with
actual em357 revealed errors that this patch corrects.

Change-Id: I70cf31210de8ed84e3755a56e76261ad200322bb
Signed-off-by: Ed Beroset <beroset@ieee.org>
Reviewed-on: http://openocd.zylin.com/2581
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

Cortex-A: Don't flush the data/unified cache if MMU is off

When the SCTLR has C set but M unset (i.e. Caching on, but MMU off) the cache
if effectively off. So only flush the cache if MMU is on, otherwise stale
entries might be committed to memory.

Change-Id: Iaff8b6f25b7a41ba838b91d45684c98f99fc0b27
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-on: http://openocd.zylin.com/2429
Tested-by: jenkins
Reviewed-by: Christopher Head <chead@zaber.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Vladimir Svoboda <ze.vlad@gmail.com>

RTOS: Add logging to FreeRTOS and general RTOS

Change-Id: I43d14f3b59daae7f90c344abdf71eaa8f74ef7ef
Signed-off-by: Evan Hunter <ehunter@broadcom.com>
Reviewed-on: http://openocd.zylin.com/2391
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/interface/parport: default to sane value on non-windows systems

When using ppdev driver 0 is the most appropriate default value as it
corresponds to /dev/parport0. Raw port address is suitable only for
direct access (I think that's parport-giveio on windows).

Reported by danitool on IRC.

Change-Id: I983c22251de6601b433ad31aaf660fb664cee7e9
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2572
Reviewed-by: Andreas Färber <afaerber@suse.de>
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

jtag/core: honour SRST timings in SWD mode

Since SWD is currently abusing the JTAG command queue for reset and
sleep handling (and all other operations are performed in a different
way), sleep needs to be forced explicitly to ensure correct timings.

Change-Id: I5b0da6cbb7d0560154e4077b261aa6828cefc892
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2591
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/board: Add Infineon XMC4500 Relax Kit config

Tested with "J-Link Lite-XMC4000 Rev.1 compiled Dec  7 2012 19:23:07"
on XMC4500 Relax Lite Kit V1.

Change-Id: Ib680a444fa4cadbf640afba15d607c0e6bd4ab2c
Signed-off-by: Jeff Ciesielski <jeffciesielski@gmail.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2567
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/board: Add Infineon XMC4200 Application Kit config

Change-Id: I3a946ba3745da651d470cd574222ba5c7147ac9d
Signed-off-by: Jeff Ciesielski <jeffciesielski@gmail.com>
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2566
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/board: Add Infineon XMC4500 Application Kit configs

Tested CPU_45A-V2 (General Purpose) with ARM-USB-TINY-H (via SWD) and
with "J-Link Lite-Cortex-M V8 compiled Jul 17 2014 11:40:12" firmware.

Tested CPU_45B-V1-006 (SDRAM) with ARM-USB-TINY-H (via SWD) and
with "J-Link Lite-XMC4200 Rev.1 compiled Jul 18 2014 17:28:26" firmware.

Change-Id: I3451117606551671ec77722f6adf7d7a6a4bd576
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2481
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/target: Add Infineon XMC4000 family config

Base config, verified against XMC4500, XMC4400 and XMC4100/XMC4200 manuals.

Change-Id: I10907bdf307bc6d11dc5454bf5391758de49dc30
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2480
Tested-by: jenkins
Reviewed-by: Jeff Ciesielski <jeffciesielski@gmail.com>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl: icepick: add icepick_d_set_coreid

this is just to avoid open coding that in
icepick_d_tapenable. Cleanup only, no functional
changes.

Change-Id: Iabd20291b7bdd95957afa1c74f52171789201227
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2624
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target: icepick: call icepick_c_setup

it also works on icepick_d.

Change-Id: I50c0c81286aae673c94ea77e47454ff48eab1668
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2623
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target: icepick: switch over to icepick_c_router

Reusing what's already there to ease maintainability.

Change-Id: I2030581669c644e2d9d9f9968075ab6344445d04
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2622
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: target: am437x: pass coreid

This commit is only for the sake of completeness as
default coreid is zero. In any case, coreids 1-4 are
used for the different PRU cores inside the SoC.

Change-Id: I775f2f444b1a908ffaf7bdbc43bcc966f19668c4
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2621
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target: am437x: use more descriptive names

Use more descriptive names for JRC and DAPs
so they more closely match documentation.

For example there's no Cortex A9 DAP, that's
the DebugSS DAP where Cortex A9 target sits. In
that same DAP we have have ETM, STM and both
dual-PRU subsystems.

Change-Id: I0e66ebb6299763f96606fae3e4c62e5785c804f2
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2620
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target: am437x: fix DebugSS DAP ID

Main DAP (where Cortex A9 sits) ID is actually
0x46b6902f. Fix it.

Change-Id: Ifa3335186bcf60d264d4ecea477bfe2f5ca10ead
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2619
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: board: add AM437x IDK support

Add support for Texas Instruments AM437x
Industrial Development Kit support.

Change-Id: I33ed71c7392c3805a86cf2c8adce83c0e8aa323d
Tested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2617
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: am437x: add reset-init event handler

this event handler will configure and lock PLLs
and configure DDR so platform is placed in usable
state.

Change-Id: Idd02f4c9789181d69578f8606ac3576ea1dd8a0b
Tested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2616
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: am437x: disable watchdog on reset-end

sometimes, watchdog might be left running and
it could expire in the middle of a debug session,
to prevent that, just make sure to disable watchdog
on reset-end if current state is 'halted'.

Change-Id: Ib4f2a2321cba17cd8c56ca3ae63114a563a6de90
Tested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2615
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: am437x: define registers to be used later

a later commit will implement a proper reset-init
handler to lock pll and configure ddr as we should.

Change-Id: I432cf28a5a944bfa83c20aed7298dbd29df30e38
Tested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2614
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: board: ti_am43xx_evm: follow xds100v2 requirements

xds100v2 asks us to call these three commands to
guarantee proper behavior, so do it.

Change-Id: Iecf9c148ce7c2082ef915b46eeb511ceea395cc3
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2613
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: board: ti_am437x_evm: select jtag transport

default already is jtag, but this silences a
warning during startup.

Change-Id: I94478327bbb259649500ef74a5b5c10d51e2a517
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2612
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: target: am437x: use entire SRAM area

According to AM4379 TRM [1], table 2-1 L3 Memory Map,
we have a total of 256KiB and there's no reason not to
use it all.

[1] http://www.ti.com/lit/ug/spruhl7b/spruhl7b.pdf

Change-Id: I117f2afe721bc4e3f0df304d3542e1a91aa69d12
Signed-off-by: Felipe Balbi <balbi@ti.com>
Reviewed-on: http://openocd.zylin.com/2611
Tested-by: jenkins
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl: target: am437x.cfg: pass correct dbgbase

Since commit ec9ccaa28849 (arm_adi_v5: make dap_lookup_cs_component()
traverse subtables and handle multicore) AM437x devices can't be used
with OpenOCD anymore. The reason is that dbgbase used to be set to zero
before that commit and that just happens to work with AM437x devices.

A more robust solution is to pass correct dbgbase when creating the
target, which this commit does.

Signed-off-by: Felipe Balbi <balbi@ti.com>
Change-Id: Iaf2617804324de8094b25137943e08b84f14c75f
Reviewed-on: http://openocd.zylin.com/2602
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>

target/target.c: fixed rp check bug in asynchronous flash write algorithm.

Bug in read pointer check within flash write algorithm made incorrect check
if block size is more than 4 bytes (bug was detected with 16 bytes block size).

Change-Id: I5b8e7ebca619a0a85ae6e9e496ff792248134d81
Signed-off-by: DmitryShpak <disona@yandex.ru>
Reviewed-on: http://openocd.zylin.com/2657
Tested-by: jenkins
Reviewed-by: Jens Bauer <jens@gpio.dk>
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

jtag/drivers/ftdi.c: removed memory leak

swd_cmd_queue buffer memory leak removed

Change-Id: Iafcdf034d32a37d577b58b6256c8fd9b064ce228
Signed-off-by: Mateusz Manowiecki <segmentation@fault.pl>
Reviewed-on: http://openocd.zylin.com/2563
Tested-by: jenkins
Reviewed-by: Andreas Färber <afaerber@suse.de>
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

doc: update RTOS section, add 'mqx' info

Change-Id: I20b93d49d275e6daaf5e39c49b4508c70c6fc56f
Signed-off-by: Marian Cingel <cingel.marian@gmail.com>
Reviewed-on: http://openocd.zylin.com/2592
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

Add configuration for EFM32 Gecko boards with Segger J-Link.

Tested with Tiny Gecko.

Change-Id: Iddb87170b6aaaf7f15d3c4517c8661dd6394be8f
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2585
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

flash/nor: mrvlqspi: fix printf formatting issues

Change-Id: I74cfce7bb8dbc13fbc3005b5a96213417f93a9f2
Signed-off-by: Mahavir Jain <mjain@marvell.com>
Reviewed-on: http://openocd.zylin.com/2577
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Stian Skjelstad <stian@nixia.no>

transport: make 'transport select' auto-select the first available transport if not set

This should allow most of the existing configurations for older
versions to remain compatible without forcing the user to change his
or her config to explicitly select transport.

Also in some circumstances can remove the need to chain a "-c transport
select X" when building custom configs on the command line, which seems
like a common new user pitfall.

Change-Id: Ic87a38c0b9b88e88fb6d106385efce2f39381d3d
Suggested-by: Petteri Aimonen <jpa@git.mail.kapsi.fi>
Signed-off-by: Angus Gratton <gus@projectgus.com>
Reviewed-on: http://openocd.zylin.com/2551
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

target/adi_v5_swd, cortex_m: properly handle more cases requiring reconnect

This brings SWD reconnection procedure in line with the ARM
documentation and changes cortex_m reset procedure to make use of it.

The motivation behind this patch is to make SAM4L "reset" and "reset
halt" properly without SRST. The complication here is that EDBG issues
an additional read of DP_RDBUFF automatically right after writing
SYSRESETREQ, that leads to a FAULT which needs to be dealt with
properly. With this patch the very first ahbap_debugport_init DAP
access will make SWD layer properly reinitialise the link before
continuing.

Runtime tested with mbed CMIS-DAP + KL25 only.

Change-Id: Ic506f9db30931dfa60860036b83f73b897975909
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2596
Tested-by: jenkins
Reviewed-by: Andrey Yurovsky <yurovsky@gmail.com>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

drivers/cmsis-dap: port to common SWD framework

Valgrind-tested.

Comparison of flashing performance on an FRDM-KL25Z board running mbed
CMSIS-DAP variant, 5MHz clock, old driver:

wrote 28096 bytes from file demo.elf in 26.833590s (1.023 KiB/s)
verified 27264 bytes in 1.754972s (15.171 KiB/s)

this implementation:

wrote 28096 bytes from file demo.elf in 3.691939s (7.432 KiB/s)
verified 27264 bytes in 0.598987s (44.450 KiB/s)

Also tested "Keil ULINK-ME CMSIS-DAP" with an STM32F100 target, 5MHz
clock, results reading from flash, old driver:

dumped 131072 bytes in 98.445305s (1.300 KiB/s)

this implementation:

dumped 131072 bytes in 8.242686s (15.529 KiB/s)

Change-Id: Ic64d3124b1d6cd9dd1016445bb627c71e189ae95
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2356
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/board/twr-k60f120m: remove useless flash bank 0 definition

Since the very first flash bank is already defined in target/kx.cfg,
there's no sense in repeating it.

Reported and tested by Richard Braun.

Change-Id: I417b7072b5e6675ddbf824446e7581b8b7da8f4b
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2595
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

flash/nor/kinetis: do not attempt mass-erase in place of a bank erase

Many kinetis parts come in multi-bank configuration, so this
optimisation here can't be performed safely.

Investigated and fixed by Richard Braun.

Change-Id: I2b56614b47951595c403a1a8edd3afe11b85679b
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2594
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

armv7m_trace, stlink: provide APIs to capture trace with an adapter

Change-Id: I9d193dd5af382912e4fe838bd4f612cffd11b295
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2540
Tested-by: jenkins

armv7m: add generic trace support (TPIU, ITM, etc.)

This provides support for various trace-related subsystems in a
generic and expandable way.

Change-Id: I3a27fa7b8cfb111753088bb8c3d760dd12d1395f
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2538
Tested-by: jenkins

stm32l1.cfg: Add missing dash to fix f7394049d3 commit.

Change-Id: Ifeb2d4fc2b43813edbc6fe2cf08bfd4c55cd1e86
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2590
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Stian Skjelstad <stian@nixia.no>

tcl: Add default hooks for STM32F3x

Keep clocks running in low power modes. Stop watchdogs from interfering
with the debug session. Set up PLL and increase clock at reset init.

Change-Id: I984d2018f7d47a1042f1e12894563154fa7b566c
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/2196
Tested-by: jenkins
Reviewed-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

flash/nor/stellaris: allow to recover a locked device that can't be examined

Change-Id: I28536184053e2d1ba906620e728f7fad6ba39f0a
Reported-by: Ed Beroset <beroset@mindspring.com>
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2552
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-by: Ed Beroset <beroset@ieee.org>

jlink: Add variant "J-Link Lite-XMC4000"

Avoids "J-Link hw type unknown 0x10" on the Infineon Relax Lite Kit.

Change-Id: I3091623ead2e84b67ac20d9866307ccbb3f26f66
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2568
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

jlink: Add variant "J-Link Lite-XMC4200"

Avoids "J-Link hw type unknown 0x11" on various Infineon boards.

Change-Id: If20b9e21110d2acc02be57f5faf28c5e6a39e2c9
Signed-off-by: Andreas Färber <afaerber@suse.de>
Reviewed-on: http://openocd.zylin.com/2565
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

server, target, cortex_m: add deinit_target to the API to free resources

This should facilitate dynamic target creation and removal.

Currently it helps with getting 0 bytes lost report from Valgrind on
exit (after talking to a nucleo board). However, 1,223,886 bytes in
5,268 blocks are still reachable which means the app holds pointers to
that data on exit. The majority comes from the jtag command queue,
there're also many blocks from TCL command registration.

Change-Id: I7523234bb90fffd26f7d29cdd7648ddd221d46ab
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2544
Tested-by: jenkins
Reviewed-by: Stian Skjelstad <stian@nixia.no>

target/target: call event handlers around examine when polling resumes

The target might be using Tcl examine-start and examine-end handlers,
they need to be called when the target gets reexamined after polling
succeeds again.

Change-Id: I371380c6f3c427ec7a0206d73426f6589f18a9bd
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2536
Tested-by: jenkins
Reviewed-by: Stian Skjelstad <stian@nixia.no>

hla/hla_interface: call HLA layout API close() on quit

This bug was exposed by Valgrind.

Change-Id: I2e2bc036b49ca3ff22f78f765ee4537763350096
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2543
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

target/cortex_m: do not leak memory on reexamination

This bug was exposed by Valgrind.

Change-Id: If50878664d928c0a44e309ca1452089c1ac71466
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2542
Tested-by: jenkins
Reviewed-by: Stian Skjelstad <stian@nixia.no>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

target: fix timer callbacks processing

Warning, behaviour change: before this patch if a timer callback
returned an error, the other handlers in the list were not called.

This patch fixes two different issues with the way timer callbacks are
called:

1. The function is not designed to be reentrant but a nested call is
possible via: target_handle timer event -> poll -> target events
before/after reexaminantion -> script_command_run ->
target_call_timer_callbacks_now . This patch makes function a no-op
when called recursively;

2. The current code can deal with the case when calling a handler
leads to its removal but not when it leads to removal of the next
callback in the list. This patch defers actual removal to consolidate
it with the calling loop.

These bugs were exposed by Valgrind.

Change-Id: Ia628a744634f5d2911eb329747e826cb9772e789
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2541
Tested-by: jenkins
Reviewed-by: Stian Skjelstad <stian@nixia.no>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

arm11: initialise DPM and register cache before reading DSCR for the first time

When target was already halted during the initial examination,
arm11_check_init() was trying to read, store and interpret DSCR
contents before the DPM structure is initialised. This caused
a segfault like described on
http://sourceforge.net/apps/trac/openocd/ticket/65 .

This is a totally untested attempt to fix this issue.

Change-Id: I2fff115679a3f0023e7a88c749ccb5f045d6cf01
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2043
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

stlink: avoid null pointer dereference in stlink_usb_close()

Otherwise it happens if stlink can not be opened on start.

Change-Id: I7088f10e61508dae230eccfe576a51498c92f5b8
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2550
Tested-by: jenkins
Reviewed-by: Andreas Färber <afaerber@suse.de>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-by: Stian Skjelstad <stian@nixia.no>

openrisc: add profiling function

Change-Id: Ifee89b289069590e6086a4713b165989578e29ec
Signed-off-by: Franck Jullien <franck.jullien@gmail.com>
Reviewed-on: http://openocd.zylin.com/2494
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

armv7m: do not access FPU registers when not present

This is runtime and valgrind tested with l0, l1 and f3 hla boards.

Change-Id: I49b0b042253d5f3bf216997f0203583db319fe23
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2516
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

armv7m: add FPU registers support

This patch adds the fpv4-sp-d16 registers to the armv7m register set.

The work is inspired by Mathias K but takes a different approach:
instead of having both double and single presicion registers in the
cache this patch works only with the doubles and counts on GDB to
split the data in halves whenever needed.

Tested with HLA only (on an STM32F334 disco board).

Currently this patch makes all ARMv7-M targets report an FPU-enabled
target description to GDB. It shouldn't harm if the user is not trying
to access non-existing FPU. However, the plan is to make this depend
on actual FPU presence later.

Change-Id: Ifcc72c80ef745230c42e4dc3995f792753fc4e7a
Signed-off-by: Mathias K <kesmtp@freenet.de>
[fercerpav@gmail.com: rework to fit target description framework]
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/514
Tested-by: jenkins
Reviewed-by: Peter Stuge <peter@stuge.se>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

jimtcl: update to 0.76 release version

It makes sense to use the latest released version to ensure
compatibility.

Change-Id: I0957f927d825fa6fb6a1594ca17bcca46ee6c1a6
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2522
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

target/arm_disassembler: add exception related disassembly

Add ERET/HVC/SMC disassebly decoding flow, below is testing result

> mdw 0x5c 4
0x0000005c: e160006e e1400072 e1600073 ee110f10
> arm disassemble 0x5c 4
0x0000005c 0xe160006e ERET
0x00000060 0xe1400072 HVC 0x0002
0x00000064 0xe1600073 SMC 0x0003
0x00000068 0xee110f10 MRC p15, 0x00, r0, c1, c0, 0x00
>

Change-Id: I1beccff885b5b37747edd0b2e9fb2297ce466a00
Signed-off-by: pierre Kuo <vichy.kuo@gmail.com>
Reviewed-on: http://openocd.zylin.com/2548
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

target/cortex_a: remove dead assignment

Found by clang static checker.

Change-Id: I77b0dc18188328fdb28d07b9e5c52e06182d9e2b
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2561
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

flash/nor/stellaris: remove dead assignment

Found by clang static checker.

Change-Id: Ifa58ba383092341c7343916e5cc8ec3c72ab2f60
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2560
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

flash/nor/sim3x: remove dead assignment

Found by clang static code checker.

Change-Id: Ic1370f8d7a48f08da6514afec5aacde38af7dfb6
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2559
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

contrib/itmdump: fix UB in show_swit, and few compile warnings

Change-Id: I1c5c99f190f7b4d405dc6fa06533e7ff37a652ec
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2533
Tested-by: jenkins
Reviewed-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

contrib: itmdump: fix incorrect format printf

Change-Id: I29100e4b284e031613586a66daa74987d86ac9e1
Signed-off-by: Karl Palsson <karlp@tweak.net.au>
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2449
Tested-by: jenkins
Reviewed-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-by: Stian Skjelstad <stian@nixia.no>
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

server: tcl_notifications command

Implements async target notifications to the tcl server

Change-Id: I4d83e9fa209e95426c440030597f99e9f0c3b260
Signed-off-by: Austin Morton <austinpmorton@gmail.com>
Reviewed-on: http://openocd.zylin.com/2336
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

rtos: Freescale MQX rtos support

ARMv7E-M (CortexM4) architecture

- fix position offset of r2,r3 registers on exception stack
- switch 'calloc' arguments
- remove prototypes of internal function and typedefs
- add NULL check for alloc functions
- remove last line of license "Franklin Street, Fifth Floor"
  because of 'checkpatch' validation
- environment: jlink + twrk60n512

Change-Id: I70840ded15b17dd945ca190ce31e2775078da2d9
Signed-off-by: Marian Cingel <cingel.marian@gmail.com>
Reviewed-on: http://openocd.zylin.com/2353
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>

tcl/interface/ftdi/xdsv2: fix and clarify EMU* signals

The signal names are changed for consistency with TI's docs and
sources.

Change-Id: Ic5c5314daa20f6f610be8a848399f951d47aa137
Signed-off-by: Matthijs van Duin <matthijs@rinnic-vaude.nl>
Reviewed-on: http://openocd.zylin.com/2231
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

stm32l1.cfg: Add ID Code of Cat.2 devices.

Change-Id: I4eb5020858c1896e294d633213f3df3fa45b6250
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2517
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Tested-by: jenkins

stm32f3x.cfg: Remove duplicate item.

Change-Id: I812c36688add73fab2e74fc112c733c5d3c201a6
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2554
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

lpc2000: Add LPC407x/8x flash size auto detection

This patch adds auto flash size detection for LPC407x/8x series.

Tested on below listed chips.
LPC4088
LPC1788(regression test)

Change-Id: I82f62678a04eac9b84658bd6d1cfdf45be64c931
Signed-off-by: Nemui Trinomius <nemuisan_kawausogasuki@live.jp>
Reviewed-on: http://openocd.zylin.com/2555
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Jens Bauer <jens@gpio.dk>

atmega: add support for the at90usb128 flash

Add support for the at90usb128 flash (tested on the RZUSBstick)

Change-Id: Ic042d7c403b20a5cc533da00c30ae6e2139bbd10
Signed-off-by: Olivier Esver <olg.esver@gmail.com>
Reviewed-on: http://openocd.zylin.com/2557
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Added system signal handling to Linux version

(with http://www.cons.org/cracauer/sigint.html in mind)

Change-Id: I15f559bc1122a408c3fb9338ba55c16fab3187e1
Signed-off-by: Mateusz Manowiecki <segmentation@fault.pl>
Reviewed-on: http://openocd.zylin.com/2443
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

jtag/drivers/buspirate: add JTAG_STABLECLOCKS cmd

Solution found on the internet

Change-Id: Ied6f7d9b28131a7ac83b203e4c64d4e9ffec0595
Signed-off-by: Mateusz Manowiecki <segmentation@fault.pl>
Reviewed-on: http://openocd.zylin.com/2496
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

target & board: AT91SAM7A2 and Olimex SAM7-LA2

Initial Support for AT91SAM7A2 on Olimex SAM7-LA2 board.
The board seems not to be able to reset into halted mode, as srst is
connected to NRESET of the cpu (configured srst_pulls_trst).
JTAG RCLK is connected to CLK.
Tested with interface/ftdi/olimex-arm-usb-ocd-h.cfg.

Change-Id: I2bdd67e3683e45f1119c5850bad294aa107891d8
Signed-off-by: Arne Wichmann <arne.wichmann@gmail.com>
Reviewed-on: http://openocd.zylin.com/2318
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

psoc4 flash driver: cleaned printf PRI... formats

Failed build on Mac OS X 10.10.2 was reported in OpenOCD-devel.
Cleaning types and printf formats. uint32_t prefered for flash/sector sizes.
2 minor changes in comments.
Removed redundant bracket.

Change-Id: Ia06b77af59c2c0ffd10869a4b263a760ca8b0a7a
Signed-off-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-on: http://openocd.zylin.com/2558
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Xiaofan <xiaofanc@gmail.com>

nrf51: Update known devices table.

Added new entries to the nrf51_known_devices_table array. New entries
are documented in the "nRF51 Series Compatability Matrix V1.0" found on
the Nordic Semi web site. Reordered entries to match the order found in
the document.

Also added an entry for an undocumented hwid discovered while flashing
the PCA10031 and PCA10028 dev boards.

Change-Id: Icca7da103d437dc28e651f27ab937fe953b9aac9
Signed-off-by: Theodore A. Roth <troth@openavr.org>
Reviewed-on: http://openocd.zylin.com/2514
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

flash/nor/spi: add GigaDevice SPI flash

Signed-off-by: dmitry pervushin <dpervushin@gmail.com>
Change-Id: I5a239dc67754ef4be1d9ec36186f434b09aa1e25
Reviewed-on: http://openocd.zylin.com/2530
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

cfg: add board and target configs for TI SimpleLink Wi-Fi CC3200 LaunchPad

Change-Id: I4396ee737c1dad380aa23894bbd1faf75f26d072
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2465
Tested-by: jenkins

Provide genuine F3 nucleo config and source it  for STM32F334 Discovery board.

The F334 disco board has a stlink V.2-1 as F3 nucleo boards. Normal F3 disco
boards use stlink v2 and can't ne used.

Change-Id: I77ebef93b184592f25ff18bb2da776d636f60ff0
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2434
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

tcl/target|board: add configs for Alphascale asm9260t

This adds configs for Alphascale asm9260t ARM based SoC and
Evaluation Kit based on this chip.

Change-Id: Id8d3a1ef204e3ae84540c2693e3d62650ba82f73
Signed-off-by: Oleksij Rempel <linux@rempel-privat.de>
Reviewed-on: http://openocd.zylin.com/2515
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Remove long-deprecated "target count" and "target number" commands.

Given that the manual states that these two subcommands are
deprecated and were scheduled to be removed back in 2010,
remove them and the corresponding documentation from the
manual.

Change-Id: Iaac633349d7fcb8b7f964109c7d26dd0cc5fc233
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Reviewed-on: http://openocd.zylin.com/1860
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

cfg: Fix Kinetis kwikstik/kx config

The flash definition belongs in the target cfg. Add some working area
and suitable reset_config.

Make kx.cfg more similar to klx.cfg.

Disable rclk as it is dead slow and a fixed 1MHz clock seems to work.

Change-Id: I8328f179c3a33be64403da93616abb48651bdfe6
Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/2227
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Add SWD protocol support to sysfsgpio

Based on the initial work on bcm2835gpio.c by Paul Fertser with many
additions. Modifications to the GPIO handling was minimal in this
patch. A more big modification is required before cleanup the
interface between bitbang and sysfsgpio.

Change-Id: I54bf2a2aa2ca059368b0e0e105dff6084b73d624
Signed-off-by: Jean-Christian de Rivaz <jc@eclis.ch>
Reviewed-on: http://openocd.zylin.com/2438
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

Add SWD protocol support to bitbang

This is based on the initial work by Paul Fertser with addition of the
switch sequences and new ACK handling.  In case of WAIT response, the
sticky bits are cleared and the last operation is repeated. The ACK
handling is based on the interpretation of the 8 February 2006 ARM
Debug Interface v5 Architecture Specification

Change-Id: Id50855b1ffff310177ccf9883dc9eb0d1b4458c8
Signed-off-by: Jean-Christian de Rivaz <jc@eclis.ch>
Reviewed-on: http://openocd.zylin.com/2437
Tested-by: jenkins
Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

gdb_server: ignore stray + in ACK mode

I couldn't make OpenOCD to work with GDB. I was always getting this in GDB:
(gdb) target remote localhost:3333
Remote debugging using localhost:3333
Ignoring packet error, continuing...
Ignoring packet error, continuing...
Ignoring packet error, continuing...
Ignoring packet error, continuing...
Malformed response to offset query, timeout
(gdb)
While debugging gdb remote protocol, I have seen that gdb responds with:
w ++$?#3f
And those two '+' seems to confuse the OpenOCD parser, if it sees another
'+' sign it emits the DEBUG output and sets the noack_mode to 2. The
problem is that we weren't even IN noack mode, this was set to 0 and then
it explicitly sets it to 2 and thus turning the noack mode on.

Change-Id: If267c9226e57fa83121ded09cf69829f8f0b4b93
Signed-off-by: Matej Kupljen <matej.kupljen@gmail.com>
Reviewed-on: http://openocd.zylin.com/2545
Tested-by: jenkins
Reviewed-by: Tomas Vanek <vanekt@fbl.cz>
Reviewed-by: Paul Fertser <fercerpav@gmail.com>

 cortex_m: Add Cortex-M0 identification to ROM-table display.

Change-Id: Id7715a83ba9793844475629aaffd10a81ce586b6
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2549
Tested-by: jenkins
Reviewed-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-by: Stian Skjelstad <stian@nixia.no>

Cortex A: fix extra memory read and non-word sizes

Without this patch, to perform a memory read, OpenOCD first issues an
LDC instruction into DBGITR in Stall mode (thus executing the
instruction), then switches to Fast mode and reads from DBGDTRTX once
for each word to transfer.

At the very end of the transfer, the final Fast mode read of DBGDTRTX
has, as always, the side effect of re-issuing the LDC instruction. This
causes two problems:

(1) If the word immediately beyond the end of the requested region is
inaccessible, this spurious LDC will cause a fault. On a fast CPU, the
LDC will finish executing by the time the poll of DSCR takes place,
failing the entire memory read. On a slow CPU, the LDC might finish
executing later, leaving an unexpected and confusing sticky fault lying
around for the next operation to see.

(2) If the LDC succeeds, it will leave the loaded word in DBGDTRTX, thus
setting DBGDSCR.TXFULL=1. The cortex_a_read_apb_ab_memory routine
completes without consuming that last word, thus confusing the next
routine that tries to use DBGDTRTX (this may not have any visible effect
on some implementations, because writing to DBGDTRTXint when TXFULL=1 is
defined as Unpredictable, but I believe it caused a visible problem for
me).

With this patch, the bulk mem_ap_sel_read_buf_noincr is modified to omit
the last word of the block. The second-to-last read of DBGDTRTX by that
function will cause the issue of the LDC for the last word. After
switching back to Normal mode and waiting for that instruction to
finish, do a final read of DBGDTRTX to extract the last word into the
buffer, leaving TXFULL=0.

Without this patch, memory accesses are always expanded such that they
are aligned to the access size. With this patch, accesses are issued
exactly as ordered by the caller. The caller is expected to handle
fragments at the beginning and end of the transfer if the address is
unaligned and an unaligned access is not desired.

Without this patch, the DFAR and DFSR registers, which report the
location and status of data faults, are ignored while performing memory
accesses, which could cause problems debugging an OS page fault handler.
With this patch, DFAR and DFSR are preserved across memory accesses, and
DFSR is decoded in the event of a synchronous fault to provide the
caller with more information about the reason for failure.

Thanks to Boris Brezillon for the original patch whose ideas led to the
non-word access mechanism implemented here and to various code reviewers
for their comments.

Change-Id: I11ae7104fbe69a522efadefc705c9a217a7eef41
Signed-off-by: Christopher Head <chead@zaber.com>
Reviewed-on: http://openocd.zylin.com/2381
Tested-by: jenkins
Reviewed-by: Olivier Schonken <olivier.schonken@gmail.com>
Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>

doc: stellaris driver supports Tiva C too

Change-Id: I3b77bf0617c0bbba85cfd678adece57aa7d03e32
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2509
Tested-by: jenkins

flash/nor/stm32lx: add all the IDs and revisions from current RM

RM0038 Rev.12 lists these new parts and introduces the category naming
scheme.

RM0367 Rev.2 (STM32L0x3 RM) doesn't add any new codes.

Change-Id: Id95dd48dda64d5f108dac57d265d29a7db3a1bd1
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Signed-off-by: Uwe Bonnes <bon@elektron.ikp.physik.tu-darmstadt.de>
Reviewed-on: http://openocd.zylin.com/2495
Tested-by: jenkins

README, doc: add mrvlqspi flash driver information

Change-Id: I7a270cdaf3d9119aa75285a8d1e063c2fe2a31b7
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2510
Tested-by: jenkins

doc, README: mention all the variants supported by lpc2000 driver

Change-Id: I66f9a201426a68fc1314ab7f02b27e36dcab33ba
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2508
Tested-by: jenkins

flash/nor/at91samd: add small delay before checking nvm status

OpenOCD's SWD subsystem doesn't currently have a consistent WAIT
handling (i.e. it doesn't ever retry, just returns an error), so right
after a row write a small delay is needed as AHB access is stalled
during the flashing operation.

The issue was exposed with a samd20 using ftdi SWD transport.

Change-Id: I07d99d3a96845cc689c3904a41f4d41344f200aa
Signed-off-by: Andrej Kazmin <funnyfish@funnyfish.botik.ru>
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/2268
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)