Paul Fertser [Sun, 22 Sep 2013 18:09:26 +0000 (22:09 +0400)]
board: add TM4C123G LaunchPad config
Almost the same as the LM4F120 LaunchPad but the chip is using the new
naming scheme (Tiva C series), supports USB OTG and has some new PWM
hardware blocks.
Change-Id: I6a7b2df76768766471f366cddaf64df3cc63f5f2 Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/1647 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
arm_adi_v5: Rewrite MEM-AP transfer implementation
Create a single pair of relatively simple functions to handle all variants
of MEM-AP transfers. This replaces the many separate functions that
handled different access sizes and packed or non-packed transfers, which
were all implemented rather differently.
With this single implementation, performance should be more consistent,
regardless of transfer type.
Change-Id: I89960e437fc6ba68a389c074fab8eac91abcf844 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1658 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
cortex_m: mem_ap access functions take number of bytes, not words
Accessing one byte of memory using a 16-bit access is not well defined.
The current implementation is forgiving and rounds up, but it should not
be relied upon.
Also, I suspect this code might fail if the byte order differs between
target and host, but I have no way of verifying it so I left it as it is.
Change-Id: I8d6a511151a194ed419f141703201f0632d84fc8 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1657 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
arm_adi_v5: Fix packed transfers crossing TAR auto-increment block
The word count returned from max_tar_block_size() was compared with the
count of half-word/bytes in the u16 and u8 packed access functions,
causing an infinite loop if the access actually crossed the boundary.
Change max_tar_block_size() to return a byte count, and scale at the call
site.
Change-Id: I2fe9b5941eb485f3d8219cfdd29fb71e02006de4 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1649 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
arm_adi_v5: Fix packed and unaligned memory writes
For packed and/or unaligned accesses, the write functions reordered the
source buffer in place. Causing in the best case a segfault, in the worst
case silent data corruption.
Rewrite the data fetching to directly match the byte lane mapping
according to IHI0031C, without destroying the buffer.
Also slightly clean up variable usage and harmonize all the write
functions.
Change-Id: I9a01cfc5578653f9ec02043ff6b61a7a20f90d67 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1646 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Franck Jullien [Thu, 8 Aug 2013 21:45:47 +0000 (23:45 +0200)]
Add new target type: OpenRISC
Add support for OpenRISC target. This implementation
supports the adv_debug_sys debug unit core. The mohor
dbg_if is not supported. Support for mohor TAP core
and Altera Virtual JTAG core are also provided.
Philip Craig [Mon, 17 Jun 2013 03:56:33 +0000 (13:56 +1000)]
jlink: improve USB read during jlink_tap_execute
Previously, when doing EMU_CMD_HW_JTAG3 commands we would do two reads,
one to read the data, and one to read the result. However, we can just
do a single larger read instead.
The motivation for this change is a weird problem. If I run the Segger
software before running OpenOCD, then the first read always fails:
Error: usb_bulk_read failed (requested=1, result=0)
Error: jlink_tap_execute, wrong result -107 (expected 1)
Sniffing the USB traffic shows that the J-Link is returning an
overflow error, meaning it is expecting to return the full result
in a single read.
Change-Id: I75e020d3b3732c9a74ee3d31838fdf17a7fac24c Signed-off-by: Philip Craig <phil@blackmoth.com.au>
Reviewed-on: http://openocd.zylin.com/1447 Tested-by: jenkins Reviewed-by: Xiaofan <xiaofanc@gmail.com> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
bus_blaster: do not fail on missing get_latency_timer
Patch submitted from Trac #64
Seems certain clones do not correctly emulate the ftdi chip, if
the call to read the latency timer fails then do not treat this as a
failure.
Spencer Oliver [Thu, 12 Sep 2013 11:31:49 +0000 (12:31 +0100)]
gdb: add default description reg types
Add support for the default gdb register description type "int" and "float".
When this is given to gdb it will use the bitsize to determine the reg size.
Change-Id: Iaeed594d1feece54044128eae1baff9858bdcae0 Signed-off-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-on: http://openocd.zylin.com/1622 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Half-word and byte writes using mwh/mwb has been completely broken since bf71e34cbf, because it dispatched all writes, regardless of access size,
to target_write_buffer(), which uses as large accesses as possible.
Partially revert the commit by changing back to the correct write method.
Change-Id: I60731fc576bf4a076a7da02bee7879e121c21d17 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1628 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Spencer Oliver [Wed, 18 Sep 2013 21:41:54 +0000 (22:41 +0100)]
adiv5: check packed transfers are supported
Currently we try and use MEM-AP packed transfers as much as possible for
8/16bit transfers. However not all targets support packed transfers, so
check before using.
Change-Id: I66256007f25ccd0c583f23db5acf6d1aa8b5e57d Signed-off-by: Spencer Oliver <spen@spen-soft.co.uk>
Reviewed-on: http://openocd.zylin.com/1602 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Paul Fertser [Wed, 18 Sep 2013 09:48:21 +0000 (13:48 +0400)]
README.Windows: pkgconfig-lite should go into MinGW, not MSYS
As autoconf is part of MinGW, it uses its root to find m4 files, pkg-config
should be installed accordingly.
Change-Id: I551797e61a3d044f751768bd7d57d8e26e7202cc Reported-by: Tomáš Voda <Voda.Tomas@seznam.cz> Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/1630 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Stephan Linz [Fri, 6 Sep 2013 13:03:54 +0000 (15:03 +0200)]
configure: expand library search for libftd2xx
Error was:
checking for library containing FT_GetLibraryVersion... no
configure: error: You appear to be missing the FTD2xx driver library.
Reason were two cases:
Case one:
The FTD2xx library uses clock_gettime() and in normal cases
the linker will find the function in the libc library. Most
(older) systems require the program be linked with the librt
library to use these (see man clock_gettime: Link with -lrt).
Case two:
You can rebuild the FTD2xx library (relink from the object
files) against the primary (system wide installed) libusb-1.0.
So you avoid mixed code (GPL with proprietary code) in the
FTD2xx library. In this case, the FTD2xx library provides no
private (linked in) libusb-1.0 functionality and require the
program be linked with the libftd2xx __AND__ the libusb-1.0
to resolve all the libusb-1.0 symbols.
Change-Id: Iaf9a35ab4257e37b98dccd47667378ad2a64b7ed Signed-off-by: Stephan Linz <linz@li-pro.net>
Reviewed-on: http://openocd.zylin.com/1614 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com> Reviewed-by: Xiaofan <xiaofanc@gmail.com>
Force the libusb-1.0 include dir to be treated as a system include path
which suppresses the cast-align warnings otherwise failing the build on
some platforms.
Change-Id: I738cfd793052b8772731d2feeef968ee893bd4bd Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1534 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Paul Fertser [Wed, 11 Sep 2013 21:25:19 +0000 (23:25 +0200)]
configure: auto-enable USB adapters based on libusb availability
This unifies the way the USB adapters are handled, and enables them
automatically whenever possible (until explicitly disabled). If an
adapter is explicitly enabled but can't be built, abort the
configure.
Also add infrastructure for generic handling of adapter drivers in
configure and print a summary of the configuration results after
finishing.
The m4 quoting is as conservative as I could get it, and seems
appropriate.
Change-Id: I1655691e5ea0d8eb9e3f67830b96992ffe33640a Signed-off-by: Paul Fertser <fercerpav@gmail.com> Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1475 Reviewed-by: Xiaofan <xiaofanc@gmail.com> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Tested-by: Spencer Oliver <spen@spen-soft.co.uk>
Paul Fertser [Wed, 11 Sep 2013 21:07:59 +0000 (23:07 +0200)]
libusb: require pkg-config support
An alternative approach to show how much cleaner the pure pkg-config
way is.
This changes the discovery procedures for libusb-1.0 and libusb-0.1,
making them depend on pkg-config being properly installed and
configured, including the necessary build host configuration for the
cross-builds (see
http://www.flameeyes.eu/autotools-mythbuster/pkgconfig/cross-compiling.html)
It should make it possible to compile OpenOCD without changes and
extra effort on GNU/Linux, FreeBSD users would need to supply a .pc
file for their libusb implementation or add LIBUSB1_LIBS and
LIBUSB1_CFLAGS to the configure environment.
Change-Id: I826e378dd1e0d101a549a573b2c63212a7e00b64 Signed-off-by: Paul Fertser <fercerpav@gmail.com> Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1467 Reviewed-by: Xiaofan <xiaofanc@gmail.com> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Tested-by: Spencer Oliver <spen@spen-soft.co.uk>
Paul Fertser [Wed, 11 Sep 2013 21:05:16 +0000 (23:05 +0200)]
README* refactoring
This is an attempt to bring the README files in line with the current
status of the OpenOCD development.
- remove some obsolete information and duplicated instructions
- reword some statements
- restructure in a way more appealing to a regular user
- add a supported hardware list to allow a potential user to determine
if his/her usecase is covered by a freetext keyword search through
the document
- Add OSX notes (courtesy GrizzlyAdams and inca)
Since most ftdi-based adapters are now covered by the ftdi driver, I
think it's ok to remove some of the libftdi/ftd2xx instructions, the
few users who still need them should refer to upstream docs instead.
I'm not sure if README.Windows should come with the DOS line endings,
but i'd expect many windows users to use their silly notepad to view
it, and notepad ignores LF apparently. (Decided to use LF anyway.)
I understand discussing and reviewing such a massive README change is
a somewhat demanding task but I feel it's a necessity to move forward
maintaining proper documentation.
Change-Id: Idfde3014c72dd5c32ad292ee1ab205322e51a138 Signed-off-by: Paul Fertser <fercerpav@gmail.com> Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1503 Tested-by: jenkins Reviewed-by: Xiaofan <xiaofanc@gmail.com>
Hsiangkai Wang [Wed, 14 Aug 2013 06:46:58 +0000 (14:46 +0800)]
target: enhance target profiling
1. gprof uses 2-bytes as minimum bucket size.
2. As user wants to use gprof --sum to summarize multiple
profiling data files, the range MUST be the same.
Add new arguments to specify profiling range.
Change-Id: Ie7e6afa6a4d82250e2d194a0eed2b428c1479ea1 Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1572 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Hsiangkai Wang [Fri, 30 Aug 2013 05:45:42 +0000 (13:45 +0800)]
target: increase the maximum number of buckets
I do not know what is the reasonable number of buckets.
If there are enough buckets, the result will be accurate.
I propose increase the maximum number of buckets to 128K.
If the size of program text section is less than 256KB, every
two bytes will be occupied by one buckets.
(The minimum size of one buckets is 2 bytes in gprof implementation.)
Change-Id: If9147743cefdc36f40f21e6dc73b9b28f28c9e1e Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1608 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Anton Kolesov [Tue, 27 Aug 2013 09:38:50 +0000 (13:38 +0400)]
gdb_server: Do not set gdb_con->sync to true for new connections
In GDB connected to OpenOCD there is a command "monitor gdb_sync" which makes
next stepi command to be ignored while GDB still will get an updated target
state. This command sets gdb_connection->sync field to true to notify that stepi
should be ignored. This field is set to true for all new connection and is set
to false after first "continue" command. However if first resume command is
stepi/nexti then it will be ignored and result will confuse GDB client, it will
report that target received signal SIGINT. This patch sets this field to false
for new connections, thus stepi/nexti will work properly when it is a first
resume command.
Change-Id: I7c9ebd69c3dc35f3e316041aa99f4e9d3425c0b6 Signed-off-by: Anton Kolesov <akolesov@synopsys.com>
Reviewed-on: http://openocd.zylin.com/1587 Tested-by: jenkins Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Rewrite the target_*_buffer_default to generate as large accesses as
possible while maintaining natural alignment.
These versions are easy to extend to generate 8-byte accesses to support
64-bit targets, although it requires some conformity from all target
implementations (i.e. they need to refuse unsupported access sizes with
some defined error code, so we can try again with a smaller one).
Change-Id: I00ddcbb1d2fd33f9f8b99cb448cc93505a2421fc Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1221 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
The only remaining user is arm7_9 so remove it from the target API and add
it to struct arm7_9_common to support all its variants with minimal
changes. Many of the variants are likely not correct in the cache/mmu
handling when the bulk write is triggered. This patch does nothing to
change that, except for arm946e, where it was easier to do what might be
the right thing.
Change-Id: Ie73ac07507ff0936fefdb90760046cc8810ed182 Signed-off-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Reviewed-on: http://openocd.zylin.com/1220 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Cosmin Gorgovan [Fri, 9 Aug 2013 16:52:46 +0000 (17:52 +0100)]
mini51: support for Nuvoton NuMicro Mini51 series flash memory
Adds a flash driver for Nuvoton MINI51, MINI52 and MINI54 microcontrollers.
At the moment, it only supports the erase and write operations.
These microcontrollers have a 4 / 8 / 16 KB APROM for application code and a
2 KB LDROM for bootloaders. When the MCU has booted off the APROM, the LDROM
isn't mapped in memory but can be programmed, and the other way around.
This means that the ARM core is typically rebooted for programming. After a
successful write or erase operation, it is rebooted again, using the initial
boot source.
This driver only supports programming the APROM.
This driver is a pure JTAG implementation, it doesn't use any SRAM.
I've tested it on a MINI54ZAN microcontroller using an ST-LINK/V2. With the
microcontroller running at the default clock frequency of 22.1184 MHz, speed
seems to be around 1.1 KB/s.
This patch adds a driver for the jtag_vpi server [1]. This server is
now part of the ORPSoC version 3 (OpenRISC Reference Platform SoC).
The jtag_vpi server provides an interface between OpenOCD and a simulated
core.
Hsiangkai Wang [Wed, 28 Aug 2013 02:59:29 +0000 (10:59 +0800)]
target: Remove error messages as no .get_gdb_fileio_info
If target does not support semi-hosting function, it has
no need to provide .get_gdb_fileio_info callback. OpenOCD
will use default function target_get_gdb_fileio_info_default.
The default function just return ERROR_FAIL and gdb_server
will treat every halted condition as normal halted and
return "Txx" to gdb.
Change-Id: I9ddb2be3a1145eae2ef5b712bdea89eb2e0fbc20 Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1586 Tested-by: jenkins Reviewed-by: Nemui Trinomius <nemuisan_kawausogasuki@live.jp> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Andrey Smirnov [Tue, 20 Aug 2013 19:26:19 +0000 (12:26 -0700)]
stlink: Add workaround for intermittent FW info retrieval failure
It appears that on some host USB configurations(2012 MacBook Air)
multiple restarts of openocd tool cause the FW on STLINKv2 dongle to
go into a weird state in which it will no longer respond to
STLINK_GET_VERSION command. This patch adds code that, if said request
fails for the first time, attempts to reset the device and retry to
initialize it and obtain FW information one more time.
Change-Id: I7227fc972adb49d52ae700ad48ab9f66b2aaa72c Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Reviewed-on: http://openocd.zylin.com/1561 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
This commit adds two tcl configuration files, one for the Altera
Cyclone V SoC series, and one for the SoCkit development board.
The board configuration is able to halt and resume the cpu cores,
and dump register contents etc. It has not been fully tested, however.
Unfortunately the Medium+ density and 0x436 devices have their F_SIZE register
at a different location: 0x1FF800CC instead of 0x1FF8004C. Fix this for
the 0x427 Medium+ devices and also the 0x436 devices. Furthermore, for
0x436 devices the flash size is reported as a 0 or 1 code rather than
the size in Kb. Please see RM0038 r8 or newer for an explanation, as
noted in the comments.
Paul Fertser [Wed, 17 Jul 2013 05:47:43 +0000 (09:47 +0400)]
target: clear running_alg flag after reset
After the target was reset we can be sure it's not running any
algorithm.
This fixes the following failure scenario:
On my STM32F103 board after I start the firmware and then stop and try
to "load" in gdb (before doing mon reset halt), I get
Error: timeout waiting for algorithm, a target reset is recommended
However, target reset doesn't help as the flag is still there ("Error:
Target is already running an algorithm"), so I have no choice but to
restart the OpenOCD process.
I'm not sure yet what exactly prevents load from working after my
firmware is initialised, most probably some interrupt is firing and my
handler produces a fault due to garbled RAM.
Change-Id: Idd977f2780a64d84800e3abd412cffc1ab6801b0 Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/1512 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Paul Fertser [Mon, 29 Jul 2013 13:22:07 +0000 (17:22 +0400)]
mdr32fx: support for Milandr's MDR32Fx internal flash memory
This adds example config and flash driver for russian Cortex-M3
microcontroller model.
Run-time tested on MDR32F9Q2I evaluation board; the flash driver
should be compatible with MDR32F2x (Cortex-M0) too but I lack hardware
to test.
There're no status bits at all, the datasheets specifies some delays
for flash operations instead. All being in <100us range, they're hard
to violate with JTAG, I hope. There're also no flash identification
registers so the flash size and type has to be hardcoded into the
config.
The flashing is considerably complicated because the flash is split
into pages, and each page consists of 4 interleaved non-consecutive
"sectors" (on MDR32F9 only, MDR32F2 is single-sectored), so the
fastest way is to latch the page and sector address and then write
only the part that should go into the current page and current sector.
Performance testing results with adapter_khz 1000 and the chip running
on its default HSI 8MHz oscillator:
When working area is specified, a target helper algorithm is used:
wrote 131072 bytes from file testfile.bin in 3.698427s (34.609 KiB/s)
This can theoretically be sped up by ~1.4 times if the helper
algorithm is fed some kind of "loader instructions stream" to allow
sector-by-sector writing.
Pure JTAG implementation (when target memory area is not available)
flashes all the 128k memory in 49.5s.
Flashing "info" memory region is also implemented, but due to the
overlapping memory addresses (resulting in incorrect memory map
calculations for GDB) it can't be used at the same time, so OpenOCD
needs to be started this way: -c "set IMEMORY true" -f
target/mdr32f9q2i.cfg
It also can't be read/verified because it's not memory-mapped anywhere
ever, and OpenOCD NOR framework doesn't really allow to provide a
custom handler that would be used when verifying.
Change-Id: I80c0632da686d49856fdbf9e05d908846dd44316 Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Reviewed-on: http://openocd.zylin.com/1532 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Enable reading the SWO trace output via STLinkv2 dongles that support
it.
This adds an optional initialization parameter "trace" with which the user
specifies a destination file where SWO trace output is appended as it comes in
as well as the trace module's source clock rate.
STLink will be configured for a 2MHz SWO data rate (STLink's highest
supported rate) if the source clock is > 2MHz, otherwise the source
clock is used as the data rate directly.
If "trace" is specified with a usable file path, the stlink_usb driver will
attempt to configure and read SWO trace data as follows:
- on _run(), the target's TPI and TMI are configured and the STLinkv2 is told
to enable tracing. Only generic ARM TPI and TMI registers are
configured, any MCU-specific settings (ex: pin routing) are the
responsibility of the target firmware. The configuration applied is
based on the STLinkv2's capabilities (UART emulation).
- on _v2_get_status(), the trace data (if any) is fetched from the
STLink after the target status is checked and the target is found to
be running.
- on _halt(), the STLink is told to disable tracing.
When fetching trace data, the entire trace frame is written to the output file
and that data is flushed. An external tool may be used to parse the
trace data into a more human-readable format.
Tested on ARM Cortex M4F and M3 MCUs (STM32F407 and STM32L152).
Change-Id: Ic3983d46c82ba77010c23b0e18ce7b275d917f12 Signed-off-by: Andrey Yurovsky <yurovsky@gmail.com>
Reviewed-on: http://openocd.zylin.com/1524 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
Hsiangkai Wang [Thu, 11 Jul 2013 02:44:59 +0000 (10:44 +0800)]
gdb_server: check target before executing event callback
As debugging multi-targets, every target has its own gdb connection.
If there are two connections, gdb_target_callback_event_handler will
be registered twice. Everytime event occurs, the registered callback
will be executed twice. If both targets are running, as user issues
ctrl-c in one gdb client, both connections will send "stop reply" to
GDB clients even TARGET_EVENT_GDB_HALT is caused by one of them.
The commit fix above problem as debugging multi-targets.
Change-Id: I1e12d4846927d7dcf1e3bb9aeb1affabc80df813 Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1501 Tested-by: jenkins Reviewed-by: Sergey Borshch <sb-sf@users.sourceforge.net> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Hsiangkai Wang [Wed, 6 Mar 2013 01:39:52 +0000 (09:39 +0800)]
gdb_server: add target_debug_reason for program exit detection
Currently, there is no way to notify gdb that program has exited.
Add new target_debug_reason called DBG_REASON_EXIT to notify gdb
the condition has occured. If the debug reason is DBG_REASON_EXIT,
gdb_server will send 'W' packet to tell gdb the process has exited.
Change-Id: I7a371da292716a3e6ac4cc2c31b009a651fe047a Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1242 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Hsiangkai Wang [Wed, 2 Jan 2013 04:02:00 +0000 (12:02 +0800)]
gdb_server: support File-I/O Remote Protocol Extension
The File I/O remote protocol extension allows the target to use the
host's file system and console I/O to perform various system calls.
To use the function, targets need to prepare two callback functions:
* get_gdb_finish_info: to get file I/O parameters from target
* gdb_fileio_end: pass file I/O response to target
As target is halted, gdb_server will try to get file-I/O information
from target through target_get_gdb_fileio_info(). If the callback function
returns ERROR_OK, gdb_server will initiate a file-I/O request to gdb.
After gdb finishes system call, gdb will pass response of the system call
to target through target_gdb_fileio_end() and continue to run(continue or step).
To implement the function, I add a new data structure in struct target,
called struct gdb_fileio_info, to record file I/O name and parameters.
Details refer to GDB manual "File-I/O Remote Protocol Extension"
Change-Id: I7f4d45e7c9e967b6d898dc79ba01d86bc46315d3 Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1102 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Hsiangkai Wang [Wed, 26 Dec 2012 11:11:03 +0000 (19:11 +0800)]
gdb server: new feature, add stop reason in stop reply packet for gdb
In GDB remote serial protocol, the stop reply packet could contain more
detail stop reason. The currently defined stop reasons are listed below.
* watch
* rwatch
* awatch
* library
* replaylog
This commit adds stop reason, watch/rwatch/awatch, in stop reply packet for
just hit watchpoint. As manual indicates, at most one stop reason should be present.
The function needs target to implement new hook, hit_watchpoint. The hook will fill
the hit watchpoint in second parameter. The information will assist gdb to locate
the watchpoint. If no such information, gdb needs to scan all watchpoints by itself.
Refer to GDB Manual, D.3 Stop Reply Packets
Change-Id: I1f70a1a9cc772e88e641b6171f1a009629a43bd1 Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1092 Tested-by: jenkins Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk>
Hsiangkai Wang [Tue, 7 May 2013 13:43:35 +0000 (21:43 +0800)]
gdb_server: support gdb target description
* Add a parameter in .get_gdb_reg_list() to return different
register lists as generating target description.
* Modify STRUCT REG to let gdb generate target description
according to register information.
The modified structure of register is
struct reg {
const char *name;
uint32_t number; /* for regnum="num" */
struct reg_feature *feature; /* for register group feature name */
bool caller_save; /* for save-restore="yes|no" */
void *value;
bool dirty;
bool valid;
bool exist;
uint32_t size;
struct reg_data_type *reg_data_type; /* for type="type" */
const char *group; /* for group="general|float|vector" */
void *arch_info;
const struct reg_arch_type *type;
};
Change-Id: I2096b67adf94518ba0b8b23d8c6a9f64ad7932b8 Signed-off-by: Hsiangkai Wang <hsiangkai@gmail.com>
Reviewed-on: http://openocd.zylin.com/1382 Tested-by: jenkins Reviewed-by: Franck Jullien <franck.jullien@gmail.com> Reviewed-by: Spencer Oliver <spen@spen-soft.co.uk> Reviewed-by: Andreas Fritiofson <andreas.fritiofson@gmail.com>
projects / openocd / log