3 Stuff that needs to be done
6 doesn't understand (), lots of susv3 stuff.
9 The command shell situation is a big mess. We have three or four different
10 shells that don't really share any code, and the "standalone shell" doesn't
11 work all that well (especially not in a chroot environment), due to apps not
12 being reentrant. Unifying the various shells and figuring out a configurable
13 way of adding the minimal set of bash features a given script uses is a big
14 job, but it be a big improvement.
16 Note: Rob Landley (rob@landley.net) is working on this one, but very slowly...
19 We should have a diff -u command. We have patch, we should have diff
20 (we only need to support unified diffs though).
23 Would be nice. The basic susv3 options, plus fuser -k.
26 should have simple fuzz factor support to apply patches at an offset which
27 shouldn't take up too much space.
30 It would be nice to have a man command. Not one that handles troff or
31 anything, just one that can handle preformatted ascii man pages, possibly
32 compressed. This could probably be a script in the extras directory that
33 calls cat/zcat/bzcat | less
36 Compression-side support.
42 Look at the full Single Unix Specification version 3 (available online at
43 "http://www.opengroup.org/onlinepubs/009695399/nfindex.html") and
44 figure out which of our apps are compliant, and what we're missing that
45 we might actually care about.
47 Even better would be some kind of automated compliance test harness that
48 exercises each command line option and the various corner cases.
51 Lots of archivers have the same general infrastructure. The directory
52 traversal code should be factored out, and the guts of each archiver could
53 be some setup code and a series of callbacks for "add this file",
54 "add this directory", "add this symlink" and so on.
56 This could clean up tar and zip, and make it cheaper to add cpio and ar
57 write support, and possibly even cheaply add things like mkisofs someday,
58 if it becomes relevant.
61 Several existing applets (sort, vi, less...) read
62 a whole file into memory and act on it. There might be an opportunity
63 for shared code in there that could be moved into libbb...
65 Individual compilation of applets.
66 It would be nice if busybox had the option to compile to individual applets,
67 for people who want an alternate implementation less bloated than the gnu
68 utils (or simply with less political baggage), but without it being one big
71 Turning libbb into a real dll is another possibility, especially if libbb
72 could export some of the other library interfaces we've already more or less
73 got the code for (like zlib).
75 buildroot - Make a "dogfood" option
76 Busybox 1.1 will be capable of replacing most gnu packages for real world use,
77 such as developing software or in a live CD. It needs wider testing.
79 Busybox should now be able to replace bzip2, coreutils, e2fsprogs, file,
80 findutils, gawk, grep, inetutils, less, modutils, net-tools, patch, procps,
81 sed, shadow, sysklogd, sysvinit, tar, util-linux, and vim. The resulting
82 system should be self-hosting (I.E. able to rebuild itself from source code).
83 This means it would need (at least) binutils, gcc, and make, or equivalents.
85 It would be a good "eating our own dogfood" test if buildroot had the option
86 of using a "make allyesconfig" busybox instead of the all of the above
87 packages. Anything that's wrong with the resulting system, we can fix. (It
88 would be nice to be able to upgrade busybox to be able to replace bash and
89 diffutils as well, but we're not there yet.)
91 One example of an existing system that does this already is Firmware Linux:
92 http://www.landley.net/code/firmware
95 We have a CONFIG_BUFFER mechanism that lets us select whether to do memory
96 allocation on the stack or the heap. Unfortunately, we're not using it much.
97 We need to audit our memory allocations and turn a lot of malloc/free calls
98 into RESERVE_CONFIG_BUFFER/RELEASE_CONFIG_BUFFER.
100 And while we're at it, many of the CONFIG_FEATURE_CLEAN_UP #ifdefs will be
101 optimized out by the compiler in the stack allocation case (since there's no
102 free for an alloca()), and this means that various cleanup loops that just
103 call free might also be optimized out by the compiler if written right, so
104 we can yank those #ifdefs too, and generally clean up the code.
106 Switch CONFIG_SYMBOLS to ENABLE_SYMBOLS
108 In busybox 1.0 and earlier, configuration was done by CONFIG_SYMBOLS
109 that were either defined or undefined to indicate whether the symbol was
110 selected in the .config file. They were used with #ifdefs, ala:
118 In 1.1, we have new ENABLE_SYMBOLS which are always defined (as 0 or 1),
119 meaning you can still use them for preprocessor tests by replacing
120 "#ifdef CONFIG_SYMBOL" with "#if ENABLE_SYMBOL". But more importantly, we
121 can use them as a true or false test in normal C code:
123 if (ENABLE_SYMBOL && other_test) {
127 (Optimizing away if() statements that resolve to a constant value
128 is known as "dead code elimination", an optimization so old and simple that
129 Turbo Pascal for DOS did it twenty years ago. Even modern mini-compilers
130 like the Tiny C Compiler (tcc) and the Small Device C Compiler (SDCC)
131 perform dead code elimination.)
133 Right now, busybox.h is #including both "config.h" (defining the
134 CONFIG_SYMBOLS) and "bb_config.h" (defining the ENABLE_SYMBOLS). At some
135 point in the future, it would be nice to wean ourselves off of the
136 CONFIG versions. (Among other things, some defective build environments
137 leak the Linux kernel's CONFIG_SYMBOLS into the system's standard #include
138 files. We've experienced collisions before.)
141 This is more an unresolved issue than a to-do item. More thought is needed.
143 Normally we rely on exit() to free memory, close files, and unmap segments
144 for us. This makes most calls to free(), close(), and unmap() optional in
145 busybox applets that don't intend to run for very long, and optional stuff
146 can be omitted to save size.
148 The idea was raised that we could simulate fork/exit with setjmp/longjmp
149 for _really_ brainless embedded systems, or speed up the standalone shell
150 by not forking. Doing so would require a reliable FEATURE_CLEAN_UP.
151 Unfortunately, this isn't as easy as it sounds.
153 The problem is, lots of things exit(), sometimes unexpectedly (xmalloc())
154 and sometimes reliably (bb_perror_msg_and_die() or show_usage()). This
155 jumps out of the normal flow control and bypasses any cleanup code we
156 put at the end of our applets.
158 It's possible to add hooks to libbb functions like xmalloc() and bb_xopen()
159 to add their entries to a linked list, which could be traversed and
160 freed/closed automatically. (This would need to be able to free just the
161 entries after a checkpoint to be usable for a forkless standalone shell.
162 You don't want to free the shell's own resources.)
164 Right now, FEATURE_CLEAN_UP is more or less a debugging aid, to make things
165 like valgrind happy. It's also documentation of _what_ we're trusting
166 exit() to clean up for us. But new infrastructure to auto-free stuff would
167 render the existing FEATURE_CLEAN_UP code redundant.
169 For right now, exit() handles it just fine.