Home | History | Annotate | Line # | Download | only in ddb
TODO revision 1.1 
      1 $NetBSD: TODO,v 1.1 2000/06/06 05:41:54 jhawk Exp $
      2 
      3 In rough order.
      4 
      5 Bugs to be fixed:
      6 
      7 1.	command history sometimes gets slightly confused
      8 
      9 2.	uvm_fault()s can loop infinitely (kern/10016)
     10 
     11 3.	examine /m (hex + character dump) doesn't work.
     12 
     13 4.	"dot", "next", and "last" are not set properly and are usually
     14 	the same.
     15 
     16 5.	The vax port supporst decimal pids in "kill", etc. This should be
     17 	MI or not exist at all.
     18 
     19 6.	double-quote delimitted strings don't work in "print", nor do
     20 	multiple arguments (like the manpage says)
     21 
     22 7.	Fix "ditto" character in manpage vs. code.
     23 
     24 
     25 Features to be implemented:
     26 
     27 1.	Have "ps" print pids with a 0t prefix to reduce radix
     28 	confusion.
     29 
     30 2.	"set" should display the old value as "write" does.
     31 
     32 3.	Support 64-bit longs in "write". /L or /q? both? long versus quad?
     33 
     34 4.	Note only some radixes are supported.
     35 
     36 5.	Allow a mechanism for DDB to not fail under X (on the i386).
     37 	Probably just skip DDB in such cases. 
     38 
     39 6.	Enable DDB in GENERIC on all architectures not starved for
     40 	space.
     41 
     42 7.	Numbers starting with [a-f] should work, but symbols
     43 	of the same name should have priority.
     44 
     45 8.	Implement the unimplemented C operators, especially bitwise-OR (|).
     46 
     47 9.	"cond" -- conditionally execute a specified command on a breakpoint.
     48 	from Mach.
     49 
     50 10.	macros ("macro", "dmacro", "show macro"). from Mach.
     51 
     52 11.	!! command repitition (mostly unnecessary, just for consistency).
     53 	from Mach.
     54 
     55 12.	"continue" should honor _count as a number of breakpoints to ignore,
     56 	just as ",5:c" ignores 5 breakpoints in adb.
     57 
     58 13.	There should be a "step over" command (like :e in adb) that sets
     59 	a breakpoint at the next instruction if the current instruction
     60 	is a call (or jump?).
     61 
     62 14.	Specification of up to 4 ASCII chars as a 32-bit number
     63 	in an expression, like ' in adb:
     64 	     'cccc'          ASCII value of up to 4 characters.
     65 	I'm not sure what character to use for this, if the single-quote
     66 	is used for ditto, and the double-quote for strings in "print".
     67 
     68 15.	Add crossreference of gdb/ddb commands to manpage. i.e.
     69 	ddb "next" ~= gdb "finish".
     70 
     71 16.	Correct "ditto" to be a single-quote? Or something else?
     72 
     73 17.	option? To not lose system time when in ddb [reset clock on exit?]
     74 
     75 18.	Consider an XSERVER_DDB option [is it worth it?]
     76 
     77 19.	Eliminate vestigal DDB-specific kprintf() format specifiers
     78 
     79 20.	Figure out what's up with "ed_style"
     80 
     81 21.	Some sort of "immediate" step-in, step-over keybindings, like [
     82 	and ] in kadb (but not in adb):
     83 	     [           Like :e in adb(1), but requires only  one  keys-
     84 			 troke and no RETURN character.
     85 	
     86 	     ]           Like :s in adb(1), but requires only  one  keys-
     87 			 troke and no RETURN character.
     88 	Does this make people queasy?
     89 
     90 22.	"search" should display useful output (by default? with /v?)
     91 	indicating success/failure rather than simply setting "dot".
     92 	Perhaps verbose by default unless called from a macro (ugh?).
     93 
     94 23.	kern/9544, some symbolic representation of boot flags
     95 	suitable for documentation (esp. for RB_NOSYNC).
     96 
     97 24.	Soren requests in-band symbol table storage for ddb.
     98 	"dbsym"? Needs investigation/thought.
     99 
    100 25.	Support multiple address spaces (e.g. io space).
    101 	Suggested by eeh in <Pine.NEB.4.21.0005250944210.25299-100000@ehorvath>
    102 
    103 26.	Expand help to provide usage/synopsis per-command.
    104 
    105 27.	Ensure MD commands have consistent names and factor out
    106 	common code.
    107