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.
 
 
 
 
 
 

317 lines
12 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. - Isolate all TCL command support:
  18. - Pure C CLI implementations using --disable-builtin-tcl.
  19. - Allow developers to build new dongles using OpenOCD's JTAG core.
  20. - At first, provide only low-level JTAG support; target layer and
  21. above rely heavily on scripting event mechanisms.
  22. - Allow full TCL support? add --with-tcl=/path/to/installed/tcl
  23. - Move TCL support out of foo.[ch] and into foo_tcl.[ch] (other ideas?)
  24. - See src/jtag/core.c and src/jtag/tcl.c for an example.
  25. - allow some of these TCL command modules to be dynamically loadable?
  26. @section thelistjtag JTAG
  27. This section list issues that need to be resolved in the JTAG layer.
  28. @subsection thelistjtagcore JTAG Core
  29. The following tasks have been suggested for cleaning up the JTAG layer:
  30. - use tap_set_state everywhere to allow logging TAP state transitions
  31. - rename other tap_states to use standard JTAG names (suggested by ML)
  32. - Encapsulate cmd_queue_cur_state and related varaible handling.
  33. - add slick 32 bit versions of jtag_add_xxx_scan() that avoids
  34. buf_set_u32() calls and other evidence of poor impedance match between
  35. API and calling code. New API should cut down # of lines in calling
  36. code by 100's and make things clearer. Also potentially be supported
  37. directly in minidriver API for better embedded host performance.
  38. The following tasks have been suggested for adding new core JTAG support:
  39. - autodetect devices present on the scan chain
  40. - implement 'discover_taps' command
  41. - SPI/UART emulation:
  42. - (ab)use bit-banging JTAG interfaces to emulate SPI/UART
  43. - allow SPI to program flash, MCUs, etc.
  44. @subsection thelistjtaginterfaces JTAG Interfaces
  45. There are some known bugs to fix in JTAG adapter drivers:
  46. - For JTAG_STATEMOVE to TAP_RESET, all drivers must ignore the current
  47. recorded state. The tap_get_state() call won't necessarily return
  48. the correct value, especially at server startup. Fix is easy: in
  49. that case, always issue five clocks with TMS high.
  50. - amt_jtagaccel.c
  51. - arm-jtag-ew.c
  52. - bitbang.c
  53. - bitq.c
  54. - gw16012.c
  55. - jlink.c
  56. - usbprog.c
  57. - vsllink.c
  58. - rlink/rlink.c
  59. The following tasks have been suggeted for improving OpenOCD's JTAG
  60. interface support:
  61. - rework USB communication to be more robust. Two possible options are:
  62. -# use libusb-1.0.1 with libusb-compat-0.1.1 (non-blocking I/O wrapper)
  63. -# rewrite implementation to use non-blocking I/O
  64. - J-Link driver:
  65. - fix to work with long scan chains, such as R.Doss's svf test.
  66. - FT2232 (libftdi):
  67. - make performance comparable to alternatives (on Win32, D2XX is faster)
  68. - make usability comparable to alternatives
  69. The following tasks have been suggested for adding new JTAG interfaces:
  70. - TCP driver: allow client/server for remote JTAG interface control.
  71. @section thelistswd Serial Wire Debug
  72. - implement Serial Wire Debug interface
  73. @section thelistbs Boundary Scan Support
  74. - add STAPL support?
  75. - add BSDL support?
  76. A few possible options for the above:
  77. -# Fake a TCL equivalent?
  78. -# Integrate an existing library?
  79. -# Write a new C implementation a la Jim?
  80. Once the above are completed:
  81. - add support for programming flash using boundary scan techniques
  82. - add integration with a modified gerber view program:
  83. - provide means to view the PCB and select pins and traces
  84. - allow use-cases such as the following:
  85. - @b Stimulus
  86. -# Double-click on a pin (or trace) with the mouse.
  87. - @b Effects
  88. -# The trace starts blinking, and
  89. -# OpenOCD toggles the pin(s) 0/1.
  90. @section thelisttargets Target Support
  91. - general layer cleanup: @par
  92. https://lists.berlios.de/pipermail/openocd-development/2009-May/006590.html
  93. - regression: xscale does not place debug_handler.bin into the right spot. workaround:
  94. use -s option on command line to place xscale/debug_handler.bin in search path @par
  95. https://lists.berlios.de/pipermail/openocd-development/2009-July/009338.html
  96. - bug: either USBprog is broken with new tms sequence or there is a general
  97. problem with XScale and the new tms sequence. Workaround: use "tms_sequence long"
  98. @par
  99. https://lists.berlios.de/pipermail/openocd-development/2009-July/009426.html
  100. - regression: "reset halt" between 729(works) and 788(fails): @par
  101. https://lists.berlios.de/pipermail/openocd-development/2009-July/009206.html
  102. - ARM7/9:
  103. - add reset option to allow programming embedded ice while srst is asserted.
  104. Some CPUs will gate the JTAG clock when srst is asserted and in this case,
  105. it is necessary to program embedded ice and then assert srst afterwards.
  106. - ARM923EJS:
  107. - reset run/halt/step is not robust; needs testing to map out problems.
  108. - ARM11 improvements (MB?)
  109. - fix single stepping (reported by ØH). Need to automatically
  110. use hardware stepping if available.
  111. - hunt down and add timeouts to all infinite loops, e.g. arm11_run_instr_no_data would
  112. lock up in infinite loop if e.g. an "mdh" command tries to read memory from invalid memory location.
  113. Try mdh 0x40000000 on i.MX31 PDK
  114. - mdb can return garbage data if read byte operation fails for
  115. a memory region(16 & 32 byte access modes may be supported). Is this
  116. a bug in the .MX31 PDK init script? Try on i.MX31 PDK:
  117. mdw 0xb80005f0 0x8, mdh 0xb80005f0 0x10, mdb 0xb80005f0 0x20. mdb returns
  118. garabage.
  119. - implement missing functionality (grep FNC_INFO_NOTIMPLEMENTED ...)
  120. - thumb support is missing: ISTR ARMv6 requires Thumb.
  121. ARM1156 has Thumb2; ARM1136 doesn't.
  122. - Cortex A8 support (ML)
  123. - add target implementation (ML)
  124. - Generic ARM run_algorithm() interface
  125. - tagged struct wrapping ARM instructions and metadata
  126. - not revision-specific (current: ARMv4+ARMv5 -or- ARMv6 -or- ARMv7)
  127. - usable with at least arm_nandwrite() and generic CFI drivers
  128. - MC1322x support (JW/DE?)
  129. - integrate and test support from JW (and DE?)
  130. - get working with a known good interface (i.e. not today's jlink)
  131. - AT91SAM92xx:
  132. - improvements for unknown-board-atmel-at91sam9260.cfg (RD)
  133. - STR9x: (ZW)
  134. - improvements to str912.cfg to be more general purpose
  135. - AVR: (SQ)
  136. - independently verify implementation
  137. - incrementally improve working prototype in trunk. (SQ)
  138. - work out how to debug this target
  139. - AVR debugging protocol.
  140. - FPGA:
  141. - Altera Nios Soft-CPU support
  142. - Coldfire (suggested by NC)
  143. - can we draw from the BDM project? @par
  144. http://bdm.sourceforge.net/
  145. or the OSBDM package @par
  146. http://forums.freescale.com/freescale/board/message?board.id=OSBDM08&thread.id=422
  147. @section thelistsvf SVF/XSVF
  148. - develop SVF unit tests
  149. - develop XSVF unit tests
  150. @section thelistflash Flash Support
  151. - finish documentation for the following flash drivers:
  152. - avr
  153. - ecosflash
  154. - pic32mx
  155. - ocl
  156. - str9xpec
  157. @subsection thelistflashcfi CFI
  158. - finish implementing bus width/chip width handling (suggested by NC)
  159. - factor vendor-specific code into separate source files
  160. - add new callback interface for vendor-specific code
  161. - investigate/implement "thin wrapper" to use eCos CFI drivers (ØH)
  162. @section thelistdebug Debugger Support
  163. - breakpoints can get lost in some circumstances: @par
  164. https://lists.berlios.de/pipermail/openocd-development/2009-June/008853.html
  165. - integrate Keil AGDI interface to OpenOCD? (submitted by Dario Vecchio)
  166. @section thelisttesting Testing Suite
  167. This section includes several related groups of ideas:
  168. - @ref thelistunittests
  169. - @ref thelistsmoketests
  170. - @ref thelisttestreports
  171. - @ref thelisttestgenerichw
  172. @subsection thelistunittests Unit Tests
  173. - add testing skeleton to provide frameworks for adding tests
  174. - implement server unit tests
  175. - implement JTAG core unit tests
  176. - implement JTAG interface unit tests
  177. - implement flash unit tests
  178. - implement target unit tests
  179. @subsection thelistsmoketests Smoke Test Tools
  180. -# extend 'make check' with a smoketest app
  181. - checks for OOCD_TEST_CONFIG, etc. in environment (or config file)
  182. - if properly set, runs the smoke test with specified parameters
  183. - openocd -f ${OOCD_TEST_CONFIG}
  184. - implies a modular test suite (see below)
  185. - should be able to run some minimal tests with dummy interface:
  186. - compare results of baseline sanity checks with expected results
  187. -# builds a more complete test suite:
  188. - existing testing/examples/ look like a great start
  189. - all targets should be tested fully and for all capabilities
  190. - we do NOT want a "lowest common denominator" test suite
  191. - ... but can we start with one to get going?
  192. - probably requires one test configuration file per board/target
  193. - modularization can occur here, just like with targets/boards/chips
  194. - coverage can increase over time, building up bundles of tests
  195. -# add new 'smoketest' Makefile target:
  196. - calls 'make check' (and the smoketest app)
  197. - gather inputs and output into a report file
  198. @subsection thelisttestreports Test Feedback Tools
  199. These ideas were first introduced here: @par
  200. https://lists.berlios.de/pipermail/openocd-development/2009-May/006358.html
  201. - provide report submission scripts for e-mail and web forms
  202. - add new Makefile targets to post the report:
  203. - 'checkreportsend' -- send to list via e-mail (via sendmail)
  204. - 'checkreportpost' -- send web form (via curl or other script)
  205. @subsection thelisttestgenerichw Generic Hardware Tester
  206. - implement VHDL to use for FPGA-based JTAG TAP testing device
  207. - develop test suite that utilizes this testing device
  208. @section thelistautotools Autotools Build System
  209. - make entire configure process require less user consideration:
  210. - automatically detect the features that are available, unless
  211. options were specifically provided to configure
  212. - provide a report of the drivers that will be build at the end of
  213. running configure, so the users can verify which driverswill be
  214. built during 'make' (and their options) .
  215. - eliminate sources of confusion in @c bootstrap script:
  216. -# Make @c bootstrap call 'configure --enable-maintainer-mode \<opts\>'?
  217. -# Add @c buildstrap script to assist with boostrap and configure steps.
  218. - automatically build tool-chains required for cross-compiling
  219. - produce mingw32, arm-elf, others using in-tree scripts
  220. - build all required target code from sources
  221. - make JTAG and USB debug output a run-time configuration option
  222. @section thelistarchitecture Architectural Tasks
  223. The following architectural tasks need to be accomplished and should be
  224. fairly easy to complete:
  225. - clean-up code to match style guides
  226. - factor code to eliminate duplicated functionality
  227. - rewrite code that uses casts to access 16-bit and larger types
  228. from unaligned memory addresses
  229. - libopenocd support: @par
  230. https://lists.berlios.de/pipermail/openocd-development/2009-May/006405.html
  231. - review and clean up interface/target/flash APIs
  232. The following strategic tasks will require ambition, knowledge, and time
  233. to complete:
  234. - overhaul use of types to improve 32/64-bit portability
  235. - types for both host and target word sizes?
  236. - can we use GDB's CORE_TYPE support?
  237. - Allow N:M:P mapping of servers, targets, and interfaces
  238. - loadable module support for interface/target/flash drivers and commands
  239. - support both static and dynamic modules.
  240. - should probably use libltdl for dynamic library handing.
  241. @section thelistadmin Documentation Tasks
  242. - Develop milestone and release guidelines, processes, and scripts.
  243. - Develop "style" guidelines (and scripts) for maintainers:
  244. - reviewing patches
  245. - committing to Subversion
  246. - Review The Guide for OpenOCD Users for documentation errors or omissions
  247. - Update The Manual for OpenOCD Developerrs:
  248. - Add documentation describing the architecture of each module
  249. - Provide more Technical Primers to bootstrap contributor knowledge
  250. */
  251. /** @file
  252. This file contains the @ref thelist page.
  253. */