You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

360 lines
15 KiB

  1. // This file is part of the Doyxgen Developer Manual
  2. /** @page tasks Pending and Open Tasks
  3. This page lists pending and open tasks being considered or worked upon
  4. by the OpenOCD community.
  5. @section thelist The List
  6. Most items are open for the taking, but please post to the mailing list
  7. before spending much time working on anything lists here. The community
  8. may have evolved an idea since it was added here.
  9. Feel free to send patches to add or clarify items on this list, too.
  10. @section thelisttcl TCL
  11. This section provides possible things to improve with OpenOCD's TCL support.
  12. - Fix problem with incorrect line numbers reported for a syntax
  13. error in a reset init event.
  14. - organize the TCL configurations:
  15. - provide more directory structure for boards/targets?
  16. - factor configurations into layers (encapsulation and re-use)
  17. - Fix handling of variables between multiple command line "-c" and "-f"
  18. parameters. Currently variables assigned through one such parameter
  19. command/script are unset before the next one is invoked.
  20. - Isolate all TCL command support:
  21. - Pure C CLI implementations using --disable-builtin-tcl.
  22. - Allow developers to build new dongles using OpenOCD's JTAG core.
  23. - At first, provide only low-level JTAG support; target layer and
  24. above rely heavily on scripting event mechanisms.
  25. - Allow full TCL support? add --with-tcl=/path/to/installed/tcl
  26. - Move TCL support out of foo.[ch] and into foo_tcl.[ch] (other ideas?)
  27. - See src/jtag/core.c and src/jtag/tcl.c for an example.
  28. - allow some of these TCL command modules to be dynamically loadable?
  29. @section thelistjtag JTAG
  30. This section list issues that need to be resolved in the JTAG layer.
  31. @subsection thelistjtagcore JTAG Core
  32. The following tasks have been suggested for cleaning up the JTAG layer:
  33. - use tap_set_state everywhere to allow logging TAP state transitions
  34. - Encapsulate cmd_queue_cur_state and related varaible handling.
  35. - add slick 32 bit versions of jtag_add_xxx_scan() that avoids
  36. buf_set_u32() calls and other evidence of poor impedance match between
  37. API and calling code. New API should cut down # of lines in calling
  38. code by 100's and make things clearer. Also potentially be supported
  39. directly in minidriver API for better embedded host performance.
  40. The following tasks have been suggested for adding new core JTAG support:
  41. - Improve autodetection of TAPs by supporting tcl escape procedures that
  42. can configure discovered TAPs based on IDCODE value ... they could:
  43. - Remove guessing for irlen
  44. - Allow non-default irmask/ircapture values
  45. - SPI/UART emulation:
  46. - (ab)use bit-banging JTAG interfaces to emulate SPI/UART
  47. - allow SPI to program flash, MCUs, etc.
  48. @subsection thelistjtaginterfaces JTAG Interfaces
  49. There are some known bugs to fix in JTAG adapter drivers:
  50. - For JTAG_STATEMOVE to TAP_RESET, all drivers must ignore the current
  51. recorded state. The tap_get_state() call won't necessarily return
  52. the correct value, especially at server startup. Fix is easy: in
  53. that case, always issue five clocks with TMS high.
  54. - amt_jtagaccel.c
  55. - arm-jtag-ew.c
  56. - bitbang.c
  57. - bitq.c
  58. - gw16012.c
  59. - jlink.c
  60. - usbprog.c
  61. - vsllink.c
  62. - rlink/rlink.c
  63. - bug: USBprog is broken with new tms sequence; it needs 7-clock cycles.
  64. Fix promised from Peter Denison openwrt at marshadder.org
  65. Workaround: use "tms_sequence long" @par
  66. https://lists.berlios.de/pipermail/openocd-development/2009-July/009426.html
  67. The following tasks have been suggeted for improving OpenOCD's JTAG
  68. interface support:
  69. - rework USB communication to be more robust. Two possible options are:
  70. -# use libusb-1.0.1 with libusb-compat-0.1.1 (non-blocking I/O wrapper)
  71. -# rewrite implementation to use non-blocking I/O
  72. - J-Link driver:
  73. - fix to work with long scan chains, such as R.Doss's svf test.
  74. - FT2232 (libftdi):
  75. - make performance comparable to alternatives (on Win32, D2XX is faster)
  76. - make usability comparable to alternatives
  77. - Autodetect USB based adapters; this should be easy on Linux. If there's
  78. more than one, list the options; otherwise, just select that one.
  79. The following tasks have been suggested for adding new JTAG interfaces:
  80. - TCP driver: allow client/server for remote JTAG interface control.
  81. This requires a client and a server. The server is built into the
  82. normal OpenOCD and takes commands from the client and executes
  83. them on the interface returning the result of TCP/IP. The client
  84. is an OpenOCD which is built with a TCP/IP minidriver. The use
  85. of a minidriver is required to capture all the jtag_add_xxx()
  86. fn's at a high enough level and repackage these cmd's as
  87. TCP/IP packets handled by the server.
  88. @section thelistswd Serial Wire Debug
  89. - implement Serial Wire Debug interface
  90. @section thelistbs Boundary Scan Support
  91. - add STAPL support?
  92. - add BSDL support?
  93. A few possible options for the above:
  94. -# Fake a TCL equivalent?
  95. -# Integrate an existing library?
  96. -# Write a new C implementation a la Jim?
  97. Once the above are completed:
  98. - add support for programming flash using boundary scan techniques
  99. - add integration with a modified gerber view program:
  100. - provide means to view the PCB and select pins and traces
  101. - allow use-cases such as the following:
  102. - @b Stimulus
  103. -# Double-click on a pin (or trace) with the mouse.
  104. - @b Effects
  105. -# The trace starts blinking, and
  106. -# OpenOCD toggles the pin(s) 0/1.
  107. @section thelisttargets Target Support
  108. - Many common ARM cores could be autodetected using IDCODE
  109. - general layer cleanup: @par
  110. https://lists.berlios.de/pipermail/openocd-development/2009-May/006590.html
  111. - regression: "reset halt" between 729(works) and 788(fails): @par
  112. https://lists.berlios.de/pipermail/openocd-development/2009-July/009206.html
  113. - mcr/mrc target->type support
  114. - missing from ARM920t, ARM966e, XScale.
  115. It's possible that the current syntax is unable to support read-modify-write
  116. operations(see arm966e).
  117. - mcr/mrc - retire cp15 commands when there the mrc/mrc commands have been
  118. tested from: arm926ejs, arm720t, cortex_a8
  119. - ARM7/9:
  120. - clean up "arm9tdmi vector_catch". Available for some arm7 cores? @par
  121. https://lists.berlios.de/pipermail/openocd-development/2009-October/011488.html
  122. https://lists.berlios.de/pipermail/openocd-development/2009-October/011506.html
  123. - add reset option to allow programming embedded ice while srst is asserted.
  124. Some CPUs will gate the JTAG clock when srst is asserted and in this case,
  125. it is necessary to program embedded ice and then assert srst afterwards.
  126. - ARM926EJS:
  127. - reset run/halt/step is not robust; needs testing to map out problems.
  128. - ARM11 improvements (MB?)
  129. - add support for asserting srst to reset the core.
  130. - Single stepping works, but should automatically
  131. use hardware stepping if available.
  132. - mdb can return garbage data if read byte operation fails for
  133. a memory region(16 & 32 byte access modes may be supported). Is this
  134. a bug in the .MX31 PDK init script? Try on i.MX31 PDK:
  135. mdw 0xb80005f0 0x8, mdh 0xb80005f0 0x10, mdb 0xb80005f0 0x20. mdb returns
  136. garabage.
  137. - implement missing functionality (grep FNC_INFO_NOTIMPLEMENTED ...)
  138. - Thumb2 single stepping: ARM1156T2 needs simulator support
  139. - Cortex A8 support (ML)
  140. - add target implementation (ML)
  141. - Generic ARM run_algorithm() interface
  142. - tagged struct wrapping ARM instructions and metadata
  143. - not revision-specific (current: ARMv4+ARMv5 -or- ARMv6 -or- ARMv7)
  144. - usable with at least arm_nandwrite() and generic CFI drivers
  145. - MC1322x support (JW/DE?)
  146. - integrate and test support from JW (and DE?)
  147. - get working with a known good interface (i.e. not today's jlink)
  148. - AT91SAM92xx:
  149. - improvements for unknown-board-atmel-at91sam9260.cfg (RD)
  150. - STR9x: (ZW)
  151. - improvements to str912.cfg to be more general purpose
  152. - AVR: (SQ)
  153. - independently verify implementation
  154. - incrementally improve working prototype in trunk. (SQ)
  155. - work out how to debug this target
  156. - AVR debugging protocol.
  157. - FPGA:
  158. - Altera Nios Soft-CPU support
  159. - Coldfire (suggested by NC)
  160. - can we draw from the BDM project? @par
  161. http://bdm.sourceforge.net/
  162. or the OSBDM package @par
  163. http://forums.freescale.com/freescale/board/message?board.id=OSBDM08&thread.id=422
  164. @section thelistsvf SVF/XSVF
  165. - develop SVF unit tests
  166. - develop XSVF unit tests
  167. @section thelistflash Flash Support
  168. - finish documentation for the following flash drivers:
  169. - avr
  170. - ecosflash
  171. - pic32mx
  172. - ocl
  173. - str9xpec
  174. @subsection thelistflashcfi CFI
  175. - finish implementing bus width/chip width handling (suggested by NC)
  176. - factor vendor-specific code into separate source files
  177. - add new callback interface for vendor-specific code
  178. - investigate/implement "thin wrapper" to use eCos CFI drivers (ØH)
  179. @section thelistdebug Debugger Support
  180. - add support for masks in watchpoints? @par
  181. https://lists.berlios.de/pipermail/openocd-development/2009-October/011507.html
  182. - breakpoints can get lost in some circumstances: @par
  183. https://lists.berlios.de/pipermail/openocd-development/2009-June/008853.html
  184. - add support for masks in watchpoints. The trick is that GDB does not
  185. support a breakpoint mask in the remote protocol. One way to work around
  186. this is to add a separate command "watchpoint_mask add/rem <addr> <mask>", that
  187. is run to register a list of masks that the gdb_server knows to use with
  188. a particular watchpoint address.
  189. - integrate Keil AGDI interface to OpenOCD? (submitted by Dario Vecchio)
  190. @section thelisttesting Testing Suite
  191. This section includes several related groups of ideas:
  192. - @ref thelistunittests
  193. - @ref thelistsmoketests
  194. - @ref thelisttestreports
  195. - @ref thelisttestgenerichw
  196. @subsection thelistunittests Unit Tests
  197. - add testing skeleton to provide frameworks for adding tests
  198. - implement server unit tests
  199. - implement JTAG core unit tests
  200. - implement JTAG interface unit tests
  201. - implement flash unit tests
  202. - implement target unit tests
  203. @subsection thelistsmoketests Smoke Test Tools
  204. -# extend 'make check' with a smoketest app
  205. - checks for OOCD_TEST_CONFIG, etc. in environment (or config file)
  206. - if properly set, runs the smoke test with specified parameters
  207. - openocd -f ${OOCD_TEST_CONFIG}
  208. - implies a modular test suite (see below)
  209. - should be able to run some minimal tests with dummy interface:
  210. - compare results of baseline sanity checks with expected results
  211. -# builds a more complete test suite:
  212. - existing testing/examples/ look like a great start
  213. - all targets should be tested fully and for all capabilities
  214. - we do NOT want a "lowest common denominator" test suite
  215. - ... but can we start with one to get going?
  216. - probably requires one test configuration file per board/target
  217. - modularization can occur here, just like with targets/boards/chips
  218. - coverage can increase over time, building up bundles of tests
  219. -# add new 'smoketest' Makefile target:
  220. - calls 'make check' (and the smoketest app)
  221. - gather inputs and output into a report file
  222. @subsection thelisttestreports Test Feedback Tools
  223. These ideas were first introduced here: @par
  224. https://lists.berlios.de/pipermail/openocd-development/2009-May/006358.html
  225. - provide report submission scripts for e-mail and web forms
  226. - add new Makefile targets to post the report:
  227. - 'checkreportsend' -- send to list via e-mail (via sendmail)
  228. - 'checkreportpost' -- send web form (via curl or other script)
  229. @subsection thelisttestgenerichw Generic Hardware Tester
  230. - implement VHDL to use for FPGA-based JTAG TAP testing device
  231. - develop test suite that utilizes this testing device
  232. @section thelistautotools Autotools Build System
  233. - make entire configure process require less user consideration:
  234. - automatically detect the features that are available, unless
  235. options were specifically provided to configure
  236. - provide a report of the drivers that will be build at the end of
  237. running configure, so the users can verify which driverswill be
  238. built during 'make' (and their options) .
  239. - eliminate sources of confusion in @c bootstrap script:
  240. -# Make @c bootstrap call 'configure --enable-maintainer-mode \<opts\>'?
  241. -# Add @c buildstrap script to assist with boostrap and configure steps.
  242. - automatically build tool-chains required for cross-compiling
  243. - produce mingw32, arm-elf, others using in-tree scripts
  244. - build all required target code from sources
  245. - make JTAG and USB debug output a run-time configuration option
  246. @section thelistarchitecture Architectural Tasks
  247. The following architectural tasks need to be accomplished and should be
  248. fairly easy to complete:
  249. - use dynamic allocations for working memory. Scan & fix code
  250. for excessive stack allocations. take linux/scripts/checkstack.pl and
  251. see what the worst offenders are. Dynamic stack allocations are found
  252. at the bottom of the list below. Example, on amd64:
  253. $ objdump -d | checkstack.pl | head -10
  254. 0x004311e3 image_open [openocd]: 13464
  255. 0x00431301 image_open [openocd]: 13464
  256. 0x004237a4 target_array2mem [openocd]: 4376
  257. 0x0042382b target_array2mem [openocd]: 4376
  258. 0x00423e74 target_mem2array [openocd]: 4360
  259. 0x00423ef9 target_mem2array [openocd]: 4360
  260. 0x00404aed handle_svf_command [openocd]: 2248
  261. 0x00404b7e handle_svf_command [openocd]: 2248
  262. 0x00413581 handle_flash_fill_command [openocd]: 2200
  263. 0x004135fa handle_flash_fill_command [openocd]: 2200
  264. - clean-up code to match style guides
  265. - factor code to eliminate duplicated functionality
  266. - rewrite code that uses casts to access 16-bit and larger types
  267. from unaligned memory addresses
  268. - libopenocd support: @par
  269. https://lists.berlios.de/pipermail/openocd-development/2009-May/006405.html
  270. - review and clean up interface/target/flash APIs
  271. The following strategic tasks will require ambition, knowledge, and time
  272. to complete:
  273. - overhaul use of types to improve 32/64-bit portability
  274. - types for both host and target word sizes?
  275. - can we use GDB's CORE_TYPE support?
  276. - Allow N:M:P mapping of servers, targets, and interfaces
  277. - loadable module support for interface/target/flash drivers and commands
  278. - support both static and dynamic modules.
  279. - should probably use libltdl for dynamic library handing.
  280. @section thelistadmin Documentation Tasks
  281. - Develop milestone and release guidelines, processes, and scripts.
  282. - Develop "style" guidelines (and scripts) for maintainers:
  283. - reviewing patches
  284. - committing to Subversion
  285. - Review The Guide for OpenOCD Users for documentation errors or omissions
  286. - Update The Manual for OpenOCD Developerrs:
  287. - Add documentation describing the architecture of each module
  288. - Provide more Technical Primers to bootstrap contributor knowledge
  289. */
  290. /** @file
  291. This file contains the @ref thelist page.
  292. */