NetBSD/sys/ddb/TODO

$NetBSD: TODO,v 1.17 2005/12/11 12:20:53 christos Exp $

In rough order.

Bugs to be fixed:

done	command history sometimes gets slightly confused

2.	uvm_fault()s can loop infinitely (kern/10016)

done	examine /m (hex + character dump) doesn't work.

4.	"dot", "next", and "last" are not set properly and are usually
	the same.

5.	The vax port supports decimal pids in "trace", etc. This should be
	MI or not exist at all.

6.	double-quote delimited strings don't work in "print", nor do
	multiple arguments (like the manpage says)

Features to be implemented:

dead	Have "ps" print pids with a 0t prefix to reduce radix
	confusion. This is very ugly...needs thought.

done	"set" should display the old value as "write" does.

3.	Support 64-bit longs in "write". /L or /q? both? long versus quad?

4.	Note only some radixes are supported.

5.	Allow a mechanism for DDB to not fail under X (on the i386).
	Probably just skip DDB in such cases.

6.	Enable DDB in GENERIC on all architectures not starved for
	space.

done	Numbers starting with [a-f] should work, but symbols
	of the same name should have priority.

8.	Implement the unimplemented C operators, especially bitwise-OR (|).

9.	"cond" -- conditionally execute a specified command on a breakpoint.
	from Mach.

10.	macros ("macro", "dmacro", "show macro"). from Mach.

11.	!! command repitition (mostly unnecessary, just for consistency).
	from Mach.

12.	"continue" should honor _count as a number of breakpoints to ignore,
	just as ",5:c" ignores 5 breakpoints in adb.

13.	There should be a "step over" command (like :e in adb) that sets
	a breakpoint at the next instruction if the current instruction
	is a call (or jump?).

14.	Specification of up to 4 ASCII chars as a 32-bit number
	in an expression, like ' in adb:
	     'cccc'          ASCII value of up to 4 characters.
	I'm not sure what character to use for this, if the single-quote
	is used for ditto, and the double-quote for strings in "print".

15.	Add crossreference of gdb/ddb commands to manpage. i.e.
	ddb "next" ~= gdb "finish".

16.	Correct "ditto" to be a single-quote? Or something else?

17.	option? To not lose system time when in ddb [reset clock on exit?]

18.	Consider an XSERVER_DDB option [is it worth it?]

19.	Eliminate vestigal DDB-specific kprintf() format specifiers

20.	Figure out what's up with "ed_style"

21.	Some sort of "immediate" step-in, step-over keybindings, like [
	and ] in kadb (but not in adb):
	     [           Like :e in adb(1), but requires only  one  keys-
			 troke and no RETURN character.

	     ]           Like :s in adb(1), but requires only  one  keys-
			 troke and no RETURN character.
	Does this make people queasy?

22.	"search" should display useful output (by default? with /v?)
	indicating success/failure rather than simply setting "dot".
	Perhaps verbose by default unless called from a macro (ugh?).

23.	kern/9544, some symbolic representation of boot flags
	suitable for documentation (esp. for RB_NOSYNC).

24.	Soren requests in-band symbol table storage for ddb.
	"dbsym"? Needs investigation/thought.

25.	Support multiple address spaces (e.g. io space).
	Suggested by eeh in <Pine.NEB.4.21.0005250944210.25299-100000@ehorvath>

26.	Expand help to provide usage/synopsis per-command.

27.	Ensure MD commands have consistent names and factor out
	common code.

28.	Some way to track symbol table failures, since the message buffer
	isn't available when we detect them.

29.	Make "workXX" variables work.

30.	Write ddb(9) to document how ports should implement the debugger,
	and the canonical solution to some niggling annoying issues.

31.	Add support for cpus where sizeof(register_t) is not necessarily
	the same as sizeof(void *). This is the case on SH5 using the
	ILP32 ABI. On this CPU db_expr_t is, necessarily, 64-bits.
	Unfortunately, in ILP32 mode, ddb will only display the low 32-bits
	of any expression, including registers...

done	Add support for duplicating all ddb output to the message buffer for
	those unlucky souls that don't have a serial console.