blob: f18e0ec2df5e960f64d5b1416ab280bd2ed624c5 [file] [log] [blame]
Glenn L McGrath90d2bff2004-05-01 00:49:49 +00001Busybox TODO
2
Rob Landleydf4cdaf2006-05-19 20:47:55 +00003Stuff that needs to be done. This is organized by who plans to get around to
4doing it eventually, but that doesn't mean they "own" the item. If you want to
5do one of these bounce an email off the person it's listed under to see if they
6have any suggestions how they plan to go about it, and to minimize conflicts
7between your work and theirs. But otherwise, all of these are fair game.
Glenn L McGrath90d2bff2004-05-01 00:49:49 +00008
Rob Landleydf4cdaf2006-05-19 20:47:55 +00009Rob Landley <rob@landley.net>:
Rob Landley4b146e62006-05-30 18:31:37 +000010 Add BB_NOMMU to platform.h and migrate __uClinux__ tests to that.
11 #if defined __UCLIBC__ && !defined __ARCH_USE_MMU__
Rob Landley0582ee02006-05-29 05:06:06 +000012 Add a libbb/platform.c
13 Implement fdprintf() for platforms that haven't got one.
14 Implement bb_realpath() that can handle NULL on non-glibc.
15 Cleanup bb_asprintf()
16
Rob Landleydf4cdaf2006-05-19 20:47:55 +000017 Migrate calloc() and bb_calloc() occurrences to bb_xzalloc().
18 Remove obsolete _() wrapper crud for internationalization we don't do.
19 Figure out where we need utf8 support, and add it.
20
21 sh
22 The command shell situation is a big mess. We have three or four different
23 shells that don't really share any code, and the "standalone shell" doesn't
24 work all that well (especially not in a chroot environment), due to apps not
25 being reentrant. I'm writing a new shell (bbsh) to unify the various
26 shells and configurably add the minimal set of bash features people
27 actually use. The hardest part is it has to configure down as small as
28 lash while providing lash's features. The rest is easy in comparison.
29 bzip2
30 Compression-side support.
31 init
Bernhard Reutner-Fischer0da069d2006-05-29 12:54:16 +000032 General cleanup (should use ENABLE_FEATURE_INIT_SYSLOG and ENABLE_FEATURE_INIT_DEBUG).
Ned Ludd5df6d9f2006-06-30 21:48:38 +000033 depmod
34 busybox lacks a way to update module deps when running from firmware without the
35 use of the depmod.pl (perl is to bloated for most embedded setups) and or orig
Denis Vlasenko9213a9e2006-09-17 16:28:10 +000036 modutils. The orig depmod is rather pointless to have to add to a firmware image
Ned Ludd5df6d9f2006-06-30 21:48:38 +000037 in when we already have a insmod/rmmod and friends.
Rob Landleydf4cdaf2006-05-19 20:47:55 +000038 Unify base64 handling.
39 There's base64 encoding and decoding going on in:
40 networking/wget.c:base64enc()
41 coreutils/uudecode.c:read_base64()
42 coreutils/uuencode.c:tbl_base64[]
43 networking/httpd.c:decodeBase64()
44 And probably elsewhere. That needs to be unified into libbb functions.
45 Do a SUSv3 audit
46 Look at the full Single Unix Specification version 3 (available online at
47 "http://www.opengroup.org/onlinepubs/009695399/nfindex.html") and
48 figure out which of our apps are compliant, and what we're missing that
49 we might actually care about.
50
51 Even better would be some kind of automated compliance test harness that
52 exercises each command line option and the various corner cases.
53 Internationalization
54 How much internationalization should we do?
55
56 The low hanging fruit is UTF-8 character set support. We should do this.
57 (Vodz pointed out the shell's cmdedit as needing work here. What else?)
58
59 We also have lots of hardwired english text messages. Consolidating this
60 into some kind of message table not only makes translation easier, but
61 also allows us to consolidate redundant (or close) strings.
62
63 We probably don't want to be bloated with locale support. (Not unless we
64 can cleanly export it from our underlying C library without having to
65 concern ourselves with it directly. Perhaps a few specific things like a
66 config option for "date" are low hanging fruit here?)
67
68 What level should things happen at? How much do we care about
69 internationalizing the text console when X11 and xterms are so much better
70 at it? (There's some infrastructure here we don't implement: The
71 "unicode_start" and "unicode_stop" shell scripts need "vt-is-UTF8" and a
72 --unicode option to loadkeys. That implies a real loadkeys/dumpkeys
73 implementation to replace loadkmap/dumpkmap. Plus messing with console font
74 loading. Is it worth it, or do we just say "use X"?)
75
76 Individual compilation of applets.
77 It would be nice if busybox had the option to compile to individual applets,
78 for people who want an alternate implementation less bloated than the gnu
79 utils (or simply with less political baggage), but without it being one big
80 executable.
81
82 Turning libbb into a real dll is another possibility, especially if libbb
83 could export some of the other library interfaces we've already more or less
84 got the code for (like zlib).
85 buildroot - Make a "dogfood" option
86 Busybox 1.1 will be capable of replacing most gnu packages for real world
87 use, such as developing software or in a live CD. It needs wider testing.
88
89 Busybox should now be able to replace bzip2, coreutils, e2fsprogs, file,
90 findutils, gawk, grep, inetutils, less, modutils, net-tools, patch, procps,
91 sed, shadow, sysklogd, sysvinit, tar, util-linux, and vim. The resulting
92 system should be self-hosting (I.E. able to rebuild itself from source
93 code). This means it would need (at least) binutils, gcc, and make, or
94 equivalents.
95
96 It would be a good "eating our own dogfood" test if buildroot had the option
97 of using a "make allyesconfig" busybox instead of the all of the above
98 packages. Anything that's wrong with the resulting system, we can fix. (It
99 would be nice to be able to upgrade busybox to be able to replace bash and
100 diffutils as well, but we're not there yet.)
101
102 One example of an existing system that does this already is Firmware Linux:
103 http://www.landley.net/code/firmware
104 initramfs
105 Busybox should have a sample initramfs build script. This depends on
106 bbsh, mdev, and switch_root.
Rob Landley9c0e4f02006-06-30 16:38:09 +0000107 mkdep
108 Write a mkdep that doesn't segfault if there's a directory it doesn't
109 have permission to read, isn't based on manually editing the output of
110 lexx and yacc, doesn't make such a mess under include/config, etc.
111 Group globals into unions of structures.
112 Go through and turn all the global and static variables into structures,
113 and have all those structures be in a big union shared between processes,
114 so busybox uses less bss. (This is a big win on nommu machines.) See
115 sed.c and mdev.c for examples.
116 Go through bugs.busybox.net and close out all of that somehow.
117 This one's open to everybody, but I'll wind up doing it...
Rob Landleydf4cdaf2006-05-19 20:47:55 +0000118
119
Bernhard Reutner-Fischer05592ac2006-09-21 22:58:38 +0000120Bernhard Fischer <busybox@busybox.net> suggests to look at these:
Rob Landleydf4cdaf2006-05-19 20:47:55 +0000121 Makefile stuff:
Bernhard Reutner-Fischer05592ac2006-09-21 22:58:38 +0000122 make -j is broken. klibc make infrastructure (vda?)
Rob Landley0582ee02006-05-29 05:06:06 +0000123 New debug options:
124 -Wlarger-than-127
Bernhard Reutner-Fischer05592ac2006-09-21 22:58:38 +0000125 Cleanup any big users
126 -Wunused-parameter
127 Facilitate applet PROTOTYPES to provide means for having applets that
128 do a) not take any arguments b) need only one of argc or argv c) need
129 both argc and argv. All of these three options should go for the most
130 feature complete denominator.
Rob Landley0582ee02006-05-29 05:06:06 +0000131 Collate BUFSIZ IOBUF_SIZE MY_BUF_SIZE PIPE_PROGRESS_SIZE BUFSIZE PIPESIZE
Bernhard Reutner-Fischer05592ac2006-09-21 22:58:38 +0000132 make bb_common_bufsiz1 configurable, size wise.
133 make pipesize configurable, size wise.
134 Use bb_common_bufsiz1 throughout applets!
Rob Landleydf4cdaf2006-05-19 20:47:55 +0000135
136As yet unclaimed:
137
Mike Frysingerb38673f2006-02-02 01:41:53 +0000138----
Rob Landleyf4bb2122005-01-24 06:56:24 +0000139find
Rob Landleyc58fd152005-10-25 20:22:50 +0000140 doesn't understand (), lots of susv3 stuff.
Rob Landleyf4bb2122005-01-24 06:56:24 +0000141----
Rob Landleyf4bb2122005-01-24 06:56:24 +0000142diff
Rob Landleydf4cdaf2006-05-19 20:47:55 +0000143 Make sure we handle empty files properly:
Rob Landley8bcc6e92006-01-09 00:54:46 +0000144 From the patch man page:
145
Mike Frysinger0b7dfb52006-07-27 03:42:30 +0000146 you can remove a file by sending out a context diff that compares
147 the file to be deleted with an empty file dated the Epoch. The
148 file will be removed unless patch is conforming to POSIX and the
149 -E or --remove-empty-files option is not given.
Rob Landleyf4bb2122005-01-24 06:56:24 +0000150---
151patch
Rob Landleyc9c959c2005-10-27 00:57:50 +0000152 Should have simple fuzz factor support to apply patches at an offset which
Rob Landley078bacf2005-09-01 03:02:23 +0000153 shouldn't take up too much space.
Rob Landleyc9c959c2005-10-27 00:57:50 +0000154
155 And while we're at it, a new patch filename quoting format is apparently
156 coming soon: http://marc.theaimsgroup.com/?l=git&m=112927316408690&w=2
Rob Landleyf4bb2122005-01-24 06:56:24 +0000157---
158man
159 It would be nice to have a man command. Not one that handles troff or
160 anything, just one that can handle preformatted ascii man pages, possibly
161 compressed. This could probably be a script in the extras directory that
Rob Landleyc58fd152005-10-25 20:22:50 +0000162 calls cat/zcat/bzcat | less
Rob Landley8bcc6e92006-01-09 00:54:46 +0000163
164 (How doclifter might work into this is anybody's guess.)
Rob Landleyf4bb2122005-01-24 06:56:24 +0000165---
Rob Landley8bcc6e92006-01-09 00:54:46 +0000166ar
167 Write support?
168---
Bernhard Reutner-Fischerfc477f42006-04-26 19:40:15 +0000169crond
170 turn FEATURE_DEBUG_OPT into ENABLE_FEATURE_CROND_DEBUG_OPT
Rob Landleyf4bb2122005-01-24 06:56:24 +0000171
172Architectural issues:
173
Rob Landley7b7c99c2005-11-04 20:45:54 +0000174bb_close() with fsync()
175 We should have a bb_close() in place of normal close, with a CONFIG_ option
176 to not just check the return value of close() for an error, but fsync().
177 Close can't reliably report anything useful because if write() accepted the
Rob Landley8bcc6e92006-01-09 00:54:46 +0000178 data then it either went out to the network or it's in cache or a pipe
179 buffer. Either way, there's no guarantee it'll make it to its final
180 destination before close() gets called, so there's no guarantee that any
181 error will be reported.
182
Rob Landley7b7c99c2005-11-04 20:45:54 +0000183 You need to call fsync() if you care about errors that occur after write(),
184 but that can have a big performance impact. So make it a config option.
185---
Rob Landleyf4bb2122005-01-24 06:56:24 +0000186Unify archivers
187 Lots of archivers have the same general infrastructure. The directory
188 traversal code should be factored out, and the guts of each archiver could
189 be some setup code and a series of callbacks for "add this file",
190 "add this directory", "add this symlink" and so on.
191
192 This could clean up tar and zip, and make it cheaper to add cpio and ar
Rob Landley8bcc6e92006-01-09 00:54:46 +0000193 write support, and possibly even cheaply add things like mkisofs or
194 mksquashfs someday, if they become relevant.
Rob Landleyf4bb2122005-01-24 06:56:24 +0000195---
196Text buffer support.
Rob Landleyc58fd152005-10-25 20:22:50 +0000197 Several existing applets (sort, vi, less...) read
Rob Landleyf4bb2122005-01-24 06:56:24 +0000198 a whole file into memory and act on it. There might be an opportunity
199 for shared code in there that could be moved into libbb...
200---
Rob Landley958fa2a2005-06-11 22:10:42 +0000201Memory Allocation
202 We have a CONFIG_BUFFER mechanism that lets us select whether to do memory
203 allocation on the stack or the heap. Unfortunately, we're not using it much.
204 We need to audit our memory allocations and turn a lot of malloc/free calls
205 into RESERVE_CONFIG_BUFFER/RELEASE_CONFIG_BUFFER.
Rob Landley1fa4a942006-06-22 22:05:00 +0000206 For a start, see e.g. make EXTRA_CFLAGS=-Wlarger-than-64
Rob Landleya8821262005-09-16 14:58:55 +0000207
Rob Landley958fa2a2005-06-11 22:10:42 +0000208 And while we're at it, many of the CONFIG_FEATURE_CLEAN_UP #ifdefs will be
209 optimized out by the compiler in the stack allocation case (since there's no
210 free for an alloca()), and this means that various cleanup loops that just
211 call free might also be optimized out by the compiler if written right, so
212 we can yank those #ifdefs too, and generally clean up the code.
Rob Landleya8821262005-09-16 14:58:55 +0000213---
214Switch CONFIG_SYMBOLS to ENABLE_SYMBOLS
215
216 In busybox 1.0 and earlier, configuration was done by CONFIG_SYMBOLS
217 that were either defined or undefined to indicate whether the symbol was
218 selected in the .config file. They were used with #ifdefs, ala:
219
220 #ifdef CONFIG_SYMBOL
221 if (other_test) {
222 do_code();
223 }
224 #endif
225
226 In 1.1, we have new ENABLE_SYMBOLS which are always defined (as 0 or 1),
227 meaning you can still use them for preprocessor tests by replacing
228 "#ifdef CONFIG_SYMBOL" with "#if ENABLE_SYMBOL". But more importantly, we
229 can use them as a true or false test in normal C code:
230
231 if (ENABLE_SYMBOL && other_test) {
232 do_code();
233 }
234
235 (Optimizing away if() statements that resolve to a constant value
236 is known as "dead code elimination", an optimization so old and simple that
237 Turbo Pascal for DOS did it twenty years ago. Even modern mini-compilers
238 like the Tiny C Compiler (tcc) and the Small Device C Compiler (SDCC)
239 perform dead code elimination.)
240
241 Right now, busybox.h is #including both "config.h" (defining the
242 CONFIG_SYMBOLS) and "bb_config.h" (defining the ENABLE_SYMBOLS). At some
243 point in the future, it would be nice to wean ourselves off of the
244 CONFIG versions. (Among other things, some defective build environments
245 leak the Linux kernel's CONFIG_SYMBOLS into the system's standard #include
246 files. We've experienced collisions before.)
247---
248FEATURE_CLEAN_UP
249 This is more an unresolved issue than a to-do item. More thought is needed.
250
251 Normally we rely on exit() to free memory, close files, and unmap segments
252 for us. This makes most calls to free(), close(), and unmap() optional in
253 busybox applets that don't intend to run for very long, and optional stuff
254 can be omitted to save size.
255
256 The idea was raised that we could simulate fork/exit with setjmp/longjmp
257 for _really_ brainless embedded systems, or speed up the standalone shell
258 by not forking. Doing so would require a reliable FEATURE_CLEAN_UP.
259 Unfortunately, this isn't as easy as it sounds.
260
261 The problem is, lots of things exit(), sometimes unexpectedly (xmalloc())
262 and sometimes reliably (bb_perror_msg_and_die() or show_usage()). This
263 jumps out of the normal flow control and bypasses any cleanup code we
264 put at the end of our applets.
265
Bernhard Reutner-Fischer8c1eda52006-08-28 23:39:36 +0000266 It's possible to add hooks to libbb functions like xmalloc() and xopen()
Rob Landleya8821262005-09-16 14:58:55 +0000267 to add their entries to a linked list, which could be traversed and
268 freed/closed automatically. (This would need to be able to free just the
269 entries after a checkpoint to be usable for a forkless standalone shell.
270 You don't want to free the shell's own resources.)
271
272 Right now, FEATURE_CLEAN_UP is more or less a debugging aid, to make things
273 like valgrind happy. It's also documentation of _what_ we're trusting
274 exit() to clean up for us. But new infrastructure to auto-free stuff would
275 render the existing FEATURE_CLEAN_UP code redundant.
276
277 For right now, exit() handles it just fine.
Rob Landley8bcc6e92006-01-09 00:54:46 +0000278
279
280
281Minor stuff:
282 watchdog.c could autodetect the timer duration via:
283 if(!ioctl (fd, WDIOC_GETTIMEOUT, &tmo)) timer_duration = 1 + (tmo / 2);
284 Unfortunately, that needs linux/watchdog.h and that contains unfiltered
285 kernel types on some distros, which breaks the build.
Bernhard Reutner-Fischer863073d2006-06-04 16:05:02 +0000286---
Bernhard Reutner-Fischer395010b2006-06-04 15:37:59 +0000287 use bb_error_msg where appropriate: See
288 egrep "(printf.*\([[:space:]]*(stderr|2)|[^_]write.*\([[:space:]]*(stderr|2))"
Bernhard Reutner-Fischer863073d2006-06-04 16:05:02 +0000289---
Bernhard Reutner-Fischer395010b2006-06-04 15:37:59 +0000290 use bb_perror_msg where appropriate: See
291 egrep "[^_]perror"
Bernhard Reutner-Fischer863073d2006-06-04 16:05:02 +0000292---
293 Remove superfluous fmt occurances: e.g.
294 fprintf(stderr, "%s: %s not found\n", "unalias", *argptr);
295 -> fprintf(stderr, "unalias: %s not found\n", *argptr);
296---
Bernhard Reutner-Fischer213cc002006-06-04 17:58:19 +0000297 possible code duplication ingroup() and is_a_group_member()
298---
Bernhard Reutner-Fischerbc291dc2006-06-05 16:00:12 +0000299 unify itoa: netstat.c, hush.c, lash.c, msh.c
300 Put one single, robust version into e.g. safe_strtol.c
301---
Bernhard Reutner-Fischere02706f2006-07-02 10:33:10 +0000302 Move __get_hz() to a better place and (re)use it in route.c, ash.c, msh.c
303---
Bernhard Reutner-Fischer395010b2006-06-04 15:37:59 +0000304
Bernhard Reutner-Fischer2677cf12006-01-13 08:46:39 +0000305
306Code cleanup:
307
308Replace deprecated functions.
309
310bzero() -> memset()
311---
312sigblock(), siggetmask(), sigsetmask(), sigmask() -> sigprocmask et al
313---
Bernhard Reutner-Fischer34fc71f2006-04-12 18:50:19 +0000314vdprintf() -> similar sized functionality
315---
Bernhard Reutner-Fischer2677cf12006-01-13 08:46:39 +0000316