1 // This file is part of the Doxygen Developer Manual
2 /** @page tasks Pending and Open Tasks
4 This page lists pending and open tasks being considered or worked upon
5 by the OpenOCD community.
7 @section thelist The List
9 Most items are open for the taking, but please post to the mailing list
10 before spending much time working on anything lists here. The community
11 may have evolved an idea since it was added here.
13 Feel free to send patches to add or clarify items on this list, too.
15 @section thelisttcl TCL
17 This section provides possible things to improve with OpenOCD's TCL support.
19 - Fix problem with incorrect line numbers reported for a syntax
20 error in a reset init event.
22 - organize the TCL configurations:
23 - provide more directory structure for boards/targets?
24 - factor configurations into layers (encapsulation and re-use)
26 - Fix handling of variables between multiple command line "-c" and "-f"
27 parameters. Currently variables assigned through one such parameter
28 command/script are unset before the next one is invoked.
30 - Isolate all TCL command support:
31 - Pure C CLI implementations using --disable-builtin-tcl.
32 - Allow developers to build new dongles using OpenOCD's JTAG core.
33 - At first, provide only low-level JTAG support; target layer and
34 above rely heavily on scripting event mechanisms.
35 - Allow full TCL support? add --with-tcl=/path/to/installed/tcl
36 - Move TCL support out of foo.[ch] and into foo_tcl.[ch] (other ideas?)
37 - See src/jtag/core.c and src/jtag/tcl.c for an example.
38 - allow some of these TCL command modules to be dynamically loadable?
40 @section thelistjtag JTAG
42 This section list issues that need to be resolved in the JTAG layer.
44 @subsection thelistjtagcore JTAG Core
46 The following tasks have been suggested for cleaning up the JTAG layer:
48 - use tap_set_state everywhere to allow logging TAP state transitions
49 - Encapsulate cmd_queue_cur_state and related variable handling.
50 - add slick 32 bit versions of jtag_add_xxx_scan() that avoids
51 buf_set_u32() calls and other evidence of poor impedance match between
52 API and calling code. New API should cut down # of lines in calling
53 code by 100's and make things clearer. Also potentially be supported
54 directly in minidriver API for better embedded host performance.
56 The following tasks have been suggested for adding new core JTAG support:
58 - Improve autodetection of TAPs by supporting tcl escape procedures that
59 can configure discovered TAPs based on IDCODE value ... they could:
60 - Remove guessing for irlen
61 - Allow non-default irmask/ircapture values
63 - (ab)use bit-banging JTAG interfaces to emulate SPI/UART
64 - allow SPI to program flash, MCUs, etc.
66 @subsection thelistjtaginterfaces JTAG Interfaces
68 There are some known bugs to fix in JTAG adapter drivers:
70 - For JTAG_STATEMOVE to TAP_RESET, all drivers must ignore the current
71 recorded state. The tap_get_state() call won't necessarily return
72 the correct value, especially at server startup. Fix is easy: in
73 that case, always issue five clocks with TMS high.
83 - bug: USBprog is broken with new tms sequence; it needs 7-clock cycles.
84 Fix promised from Peter Denison openwrt at marshadder.org
85 Workaround: use "tms_sequence long" @par
86 https://lists.berlios.de/pipermail/openocd-development/2009-July/009426.html
88 The following tasks have been suggested for improving OpenOCD's JTAG
91 - rework USB communication to be more robust. Two possible options are:
92 -# use libusb-1.0.1 with libusb-compat-0.1.1 (non-blocking I/O wrapper)
93 -# rewrite implementation to use non-blocking I/O
95 - fix to work with long scan chains, such as R.Doss's svf test.
96 - Autodetect USB based adapters; this should be easy on Linux. If there's
97 more than one, list the options; otherwise, just select that one.
99 The following tasks have been suggested for adding new JTAG interfaces:
101 - TCP driver: allow client/server for remote JTAG interface control.
102 This requires a client and a server. The server is built into the
103 normal OpenOCD and takes commands from the client and executes
104 them on the interface returning the result of TCP/IP. The client
105 is an OpenOCD which is built with a TCP/IP minidriver. The use
106 of a minidriver is required to capture all the jtag_add_xxx()
107 fn's at a high enough level and repackage these cmd's as
108 TCP/IP packets handled by the server.
110 @section thelistswd Serial Wire Debug
112 - implement Serial Wire Debug interface
114 @section thelistbs Boundary Scan Support
119 A few possible options for the above:
120 -# Fake a TCL equivalent?
121 -# Integrate an existing library?
122 -# Write a new C implementation a la Jim?
124 Once the above are completed:
125 - add support for programming flash using boundary scan techniques
126 - add integration with a modified gerber view program:
127 - provide means to view the PCB and select pins and traces
128 - allow use-cases such as the following:
130 -# Double-click on a pin (or trace) with the mouse.
132 -# The trace starts blinking, and
133 -# OpenOCD toggles the pin(s) 0/1.
135 @section thelisttargets Target Support
137 - Many common ARM cores could be autodetected using IDCODE
138 - general layer cleanup: @par
139 https://lists.berlios.de/pipermail/openocd-development/2009-May/006590.html
140 - regression: "reset halt" between 729(works) and 788(fails): @par
141 https://lists.berlios.de/pipermail/openocd-development/2009-July/009206.html
143 - add flush-value operation, call them all on resume/reset
144 - mcr/mrc target->type support
145 - missing from ARM920t, ARM966e, XScale.
146 It's possible that the current syntax is unable to support read-modify-write
147 operations(see arm966e).
148 - mcr/mrc - retire cp15 commands when there the mrc/mrc commands have been
149 tested from: arm926ejs, arm720t, cortex_a8
151 - clean up "arm9tdmi vector_catch". Available for some arm7 cores? @par
152 https://lists.berlios.de/pipermail/openocd-development/2009-October/011488.html
153 https://lists.berlios.de/pipermail/openocd-development/2009-October/011506.html
154 - add reset option to allow programming embedded ice while srst is asserted.
155 Some CPUs will gate the JTAG clock when srst is asserted and in this case,
156 it is necessary to program embedded ice and then assert srst afterwards.
158 - reset run/halt/step is not robust; needs testing to map out problems.
159 - ARM11 improvements (MB?)
160 - add support for asserting srst to reset the core.
161 - Single stepping works, but should automatically
162 use hardware stepping if available.
163 - mdb can return garbage data if read byte operation fails for
164 a memory region(16 & 32 byte access modes may be supported). Is this
165 a bug in the .MX31 PDK init script? Try on i.MX31 PDK:
166 mdw 0xb80005f0 0x8, mdh 0xb80005f0 0x10, mdb 0xb80005f0 0x20. mdb returns
168 - implement missing functionality (grep FNC_INFO_NOTIMPLEMENTED ...)
169 - Thumb2 single stepping: ARM1156T2 needs simulator support
170 - Cortex-A8 support (ML)
171 - add target implementation (ML)
173 - when stepping, only write dirtied registers (be faster)
174 - when connecting to halted core, fetch registers (startup is quirky)
175 - Generic ARM run_algorithm() interface
176 - tagged struct wrapping ARM instructions and metadata
177 - not revision-specific (current: ARMv4+ARMv5 -or- ARMv6 -or- ARMv7)
178 - usable with at least arm_nandwrite() and generic CFI drivers
180 - don't show FIFOFULL registers if they're not supported
181 - use comparators to get more breakpoints and watchpoints
182 - add "etm drivers" command
183 - trace driver init() via examine() paths only, not setup()/reset
184 - MC1322x support (JW/DE?)
185 - integrate and test support from JW (and DE?)
186 - get working with a known good interface (i.e. not today's jlink)
188 - improvements for unknown-board-atmel-at91sam9260.cfg (RD)
190 - improvements to str912.cfg to be more general purpose
192 - independently verify implementation
193 - incrementally improve working prototype in trunk. (SQ)
194 - work out how to debug this target
195 - AVR debugging protocol.
197 - Altera Nios Soft-CPU support
198 - Coldfire (suggested by NC)
199 - can we draw from the BDM project? @par
200 http://bdm.sourceforge.net/
202 or the OSBDM package @par
203 http://forums.freescale.com/freescale/board/message?board.id=OSBDM08&thread.id=422
205 @section thelistsvf SVF/XSVF
207 - develop SVF unit tests
208 - develop XSVF unit tests
210 @section thelistflash Flash Support
212 - finish documentation for the following flash drivers:
218 - Don't expect writing all-ones to be a safe way to write without
219 changing bit values. Minimally it loses on flash modules with
220 internal ECC, where it may change the ECC.
221 - NOR flash_write_unlock() does that between sectors
222 - there may be other cases too
224 - Make sure all commands accept either a bank name or a bank number,
225 and be sure both identifiers show up in "flash banks" and "nand list".
226 Right now the user-friendly names are pretty much hidden...
228 @subsection thelistflashcfi CFI
230 - finish implementing bus width/chip width handling (suggested by NC)
231 - factor vendor-specific code into separate source files
232 - add new callback interface for vendor-specific code
233 - investigate/implement "thin wrapper" to use eCos CFI drivers (ØH)
235 @section thelistdebug Debugger Support
237 - add support for masks in watchpoints? @par
238 https://lists.berlios.de/pipermail/openocd-development/2009-October/011507.html
239 - breakpoints can get lost in some circumstances: @par
240 https://lists.berlios.de/pipermail/openocd-development/2009-June/008853.html
241 - add support for masks in watchpoints. The trick is that GDB does not
242 support a breakpoint mask in the remote protocol. One way to work around
243 this is to add a separate command "watchpoint_mask add/rem <addr> <mask>", that
244 is run to register a list of masks that the gdb_server knows to use with
245 a particular watchpoint address.
246 - integrate Keil AGDI interface to OpenOCD? (submitted by Dario Vecchio)
248 @section thelisttesting Testing Suite
250 This section includes several related groups of ideas:
251 - @ref thelistunittests
252 - @ref thelistsmoketests
253 - @ref thelisttestreports
254 - @ref thelisttestgenerichw
256 @subsection thelistunittests Unit Tests
258 - add testing skeleton to provide frameworks for adding tests
259 - implement server unit tests
260 - implement JTAG core unit tests
261 - implement JTAG interface unit tests
262 - implement flash unit tests
263 - implement target unit tests
265 @subsection thelistsmoketests Smoke Test Tools
267 -# extend 'make check' with a smoketest app
268 - checks for OOCD_TEST_CONFIG, etc. in environment (or config file)
269 - if properly set, runs the smoke test with specified parameters
270 - openocd -f ${OOCD_TEST_CONFIG}
271 - implies a modular test suite (see below)
272 - should be able to run some minimal tests with dummy interface:
273 - compare results of baseline sanity checks with expected results
275 -# builds a more complete test suite:
276 - existing testing/examples/ look like a great start
277 - all targets should be tested fully and for all capabilities
278 - we do NOT want a "lowest common denominator" test suite
279 - ... but can we start with one to get going?
280 - probably requires one test configuration file per board/target
281 - modularization can occur here, just like with targets/boards/chips
282 - coverage can increase over time, building up bundles of tests
284 -# add new 'smoketest' Makefile target:
285 - calls 'make check' (and the smoketest app)
286 - gather inputs and output into a report file
288 @subsection thelisttestreports Test Feedback Tools
290 These ideas were first introduced here: @par
291 https://lists.berlios.de/pipermail/openocd-development/2009-May/006358.html
293 - provide report submission scripts for e-mail and web forms
294 - add new Makefile targets to post the report:
295 - 'checkreportsend' -- send to list via e-mail (via sendmail)
296 - 'checkreportpost' -- send web form (via curl or other script)
298 @subsection thelisttestgenerichw Generic Hardware Tester
300 - implement VHDL to use for FPGA-based JTAG TAP testing device
301 - develop test suite that utilizes this testing device
303 @section thelistautotools Autotools Build System
305 - make entire configure process require less user consideration:
306 - automatically detect the features that are available, unless
307 options were specifically provided to configure
308 - provide a report of the drivers that will be build at the end of
309 running configure, so the users can verify which drivers will be
310 built during 'make' (and their options) .
311 - eliminate sources of confusion in @c bootstrap script:
312 -# Make @c bootstrap call 'configure --enable-maintainer-mode \<opts\>'?
313 -# Add @c buildstrap script to assist with bootstrap and configure steps.
314 - automatically build tool-chains required for cross-compiling
315 - produce mingw32, arm-elf, others using in-tree scripts
316 - build all required target code from sources
317 - make JTAG and USB debug output a run-time configuration option
319 @section thelistarchitecture Architectural Tasks
321 The following architectural tasks need to be accomplished and should be
322 fairly easy to complete:
325 - use dynamic allocations for working memory. Scan & fix code
326 for excessive stack allocations. take linux/scripts/checkstack.pl and
327 see what the worst offenders are. Dynamic stack allocations are found
328 at the bottom of the list below. Example, on amd64:
330 $ objdump -d | checkstack.pl | head -10
331 0x004311e3 image_open [openocd]: 13464
332 0x00431301 image_open [openocd]: 13464
333 0x004237a4 target_array2mem [openocd]: 4376
334 0x0042382b target_array2mem [openocd]: 4376
335 0x00423e74 target_mem2array [openocd]: 4360
336 0x00423ef9 target_mem2array [openocd]: 4360
337 0x00404aed handle_svf_command [openocd]: 2248
338 0x00404b7e handle_svf_command [openocd]: 2248
339 0x00413581 handle_flash_fill_command [openocd]: 2200
340 0x004135fa handle_flash_fill_command [openocd]: 2200
341 - clean-up code to match style guides
342 - factor code to eliminate duplicated functionality
343 - rewrite code that uses casts to access 16-bit and larger types
344 from unaligned memory addresses
345 - libopenocd support: @par
346 https://lists.berlios.de/pipermail/openocd-development/2009-May/006405.html
347 - review and clean up interface/target/flash APIs
349 The following strategic tasks will require ambition, knowledge, and time
352 - overhaul use of types to improve 32/64-bit portability
353 - types for both host and target word sizes?
354 - can we use GDB's CORE_TYPE support?
355 - Allow N:M:P mapping of servers, targets, and interfaces
356 - loadable module support for interface/target/flash drivers and commands
357 - support both static and dynamic modules.
358 - should probably use libltdl for dynamic library handing.
360 @section thelistadmin Documentation Tasks
362 - Develop milestone and release guidelines, processes, and scripts.
363 - Develop "style" guidelines (and scripts) for maintainers:
366 - Review Users' Guide for documentation errors or omissions
367 - "capture" and "ocd_find" commands
368 - "ocd_" prefix on various stuff
369 - Update Developer's Manual (doxygen output)
370 - Add documentation describing the architecture of each module
371 - Provide more Technical Primers to bootstrap contributor knowledge
375 This file contains the @ref thelist page.