X-Git-Url: https://git.librecmc.org/?a=blobdiff_plain;f=TODO;h=dd9ae11f3f9e8b6176dea3979e83e6290cef4dc4;hb=94c3331d47ae08167e5122a3023ffc3340274698;hp=909f8c51c313f492cbff5b2578e7d824e2224bb3;hpb=5afc864422e8c572a13b3e48df47fd0e56cfbb74;p=oweals%2Fbusybox.git diff --git a/TODO b/TODO index 909f8c51c..dd9ae11f3 100644 --- a/TODO +++ b/TODO @@ -1,462 +1,182 @@ -TODO list for busybox in no particular order. Just because something -is listed here doesn't mean that it is going to be added to busybox, -or that doing so is even a good idea. It just means that I _might_ get -around to it some time. If you have any good ideas, please let me know. - -* login/sulogin/passwd/getty/etc are part of tinylogin, and so are not - needed or wanted in busybox (or else I'd have to link to libcrypt). - -* Networking apps are probably going to be split out some time soon into a - separate package (named perhaps tiny-netkit?). This currently includes - hostid, hostname, mnc, and ping. - - - -Erik - ------------ - -* Make insmod actually work -* dnsdomainname -* traceroute/netstat -* rdate -* hwclock -* stty -* cut -* expr -* wget (or whatever I call it) -* tftp -* ftp - - ------------------------ - -Add a compile option to turn off all the compiled in help -and usage information. Some folks don't need or want it... - ------------------------ - -Running the following: - - rm -f busybox && make LDFLAGS+=-nostdlib 2>&1 | \ - sed -ne 's/.*undefined reference to `\(.*\)..*/\1/gp' | sort | uniq - -reveals the list of all external (i.e. libc) things that BusyBox depends on. -It would be a very nice thing to reduce this list to an absolute minimum, and -then create a microLibc to provide these functions. There is no good reason -for GNU libc to be so big. I'm sure it can be a lot better. - -(BTW, this is more informative if BB_FEATURE_NFSMOUNT is turned off...) - -Most wanted list: - - [andersen@slag busybox]$ grep -l getgroups *.[ch] - test.c - -Policy violation. getgroups uses libc nss, which is unlikely -to be present in an embedded system. - - [andersen@slag busybox]$ grep -l getopt *.[ch] - dmesg.c - gunzip.c - hostname.c - mkfs_minix.c - printf.c - sfdisk.c - - This includes the symbols: - getopt_long - optarg - opterr - optind - -To be replaced with a non-getopt parser. - - [andersen@slag busybox]$ grep -l glob *.[ch] - gunzip.c - gzip.c - sh.c - tar.c - telnet.c - -Can check_wildcard_match() from utility.c do this job? - - ------------------------ - -Compile with debugging on, run 'nm --size-sort ./busybox' -and then start with the biggest things and make them smaller... - ------------------------ - -busybox.defs.h is too big and hard to follow. - -Perhaps I need to add a better build system (like the Linux kernel?) - ------------------------ - -Feature request: - -/bin/busybox --install -s which makes all links to commands that it - can support (an optionnal -s should be used for symbolic links instead - of hard links). - ------------------------ - - -> Have you ever thought of doig network logging in busybox syslogd ? It -> would quite make sense on embedded systems... :) - -So far I had not considered it. Basically, you wish to have -messages from the embedded box logged to a remote network -syslog box, right? I can see that this would be useful. -I'll add this to the TODO list, - - ------------------------ - - - I think that the add_inode &c in utility.c needs to also stow the - st_dev field, and that du.c should NOT call `reset_inode_list' - because there can be hard links from inside one argv/ to inside - another argv/. du.c probably ought to have an -x switch like GNU du - does also... - - ------------------------------------------------------------------- - - -Date: Fri, 25 Feb 2000 08:23:12 +0000 (GMT) -From: Riley Williams -X-Sender: rhw@moo.cus.org.uk -To: almesber@lrc.di.epfl.ch -Cc: "Albert D. Cahalan" , - Linux Kernel -Subject: Re: What /proc should contain [was: /proc/driver/microcode] -In-Reply-To: <20000224165245.A29790@lrc.di.epfl.ch> -Message-ID: -Sender: owner-linux-kernel@vger.rutgers.edu -Precedence: bulk - -Hi there. - - >> Nope, most /proc access is does via programs written in C. - - > That varies a lot from file to file. E.g. I haven't seen any - > programs that are specificly designed to read /proc/atm/* yet, - > and I know of only one (fuser) that reads /proc/mounts, - > extracting only partial information (just to pick two examples - > that I'm quite familiar with). - -As a point of reference, here's a slightly tweaked analysis of the -programs on the system I'm reading your mail on. Basically, I ran the -following script... - - Q> #!/bin/bash - Q> function use() { - Q> for Z in $* ; do - Q> strings $Z | grep /proc | sed "s=^=$Z =" - Q> done - Q> } - Q> use /{,s}bin/* /usr/{,s}bin/* | sort -u | tee proc-usage - -...and then went through it removing comments and print format -strings. Replace /proc with /dev and you'd soon have an equivalent -list for that - although I can report that such is MUCH larger... - -There are three programs therein that refer to /proc/mounts ... - - /bin/mount - /bin/umount - /usr/bin/eject - -...and, as you stated, none that refer to /proc/atm on this system. -However, as this is a RedHat Linux 5.0 based system, that's not -necessarily an up to date reference thereto... - -Here's the list anyway... - -/bin/kill /proc/%d/cmdline -/bin/kill /proc/%d/stat - -/bin/mount /proc/devices -/bin/mount /proc/filesystems -/bin/mount /proc/mounts - -/bin/netstat /proc/net -/bin/netstat /proc/net/appletalk -/bin/netstat /proc/net/ax25 -/bin/netstat /proc/net/ax25_route -/bin/netstat /proc/net/dev -/bin/netstat /proc/net/ip_masquerade -/bin/netstat /proc/net/ipx -/bin/netstat /proc/net/ipx_route -/bin/netstat /proc/net/netstat -/bin/netstat /proc/net/nr -/bin/netstat /proc/net/nr_neigh -/bin/netstat /proc/net/nr_nodes -/bin/netstat /proc/net/raw -/bin/netstat /proc/net/route -/bin/netstat /proc/net/rt_cache -/bin/netstat /proc/net/snmp -/bin/netstat /proc/net/tcp -/bin/netstat /proc/net/udp -/bin/netstat /proc/net/unix - -/bin/umount /proc/devices -/bin/umount /proc/mounts - -/sbin/arp /proc/net/appletalk -/sbin/arp /proc/net/arp -/sbin/arp /proc/net/ax25 -/sbin/arp /proc/net/ipx -/sbin/arp /proc/net/nr -/sbin/arp /proc/net/unix - -/sbin/cardctl /proc/devices - -/sbin/cardmgr /proc/devices - -/sbin/fdisk /proc/ide/%s/media -/sbin/fdisk /proc/scsi/scsi - -/sbin/getty /proc/version - -/sbin/ifconfig /proc/net -/sbin/ifconfig /proc/net/appletalk -/sbin/ifconfig /proc/net/ax25 -/sbin/ifconfig /proc/net/dev -/sbin/ifconfig /proc/net/ipx -/sbin/ifconfig /proc/net/nr -/sbin/ifconfig /proc/net/unix - -/sbin/ifup /proc/sys/kernel/modprobe - -/sbin/ipchains /proc/net/ip_fwchains -/sbin/ipchains /proc/net/ip_fwnames -/sbin/ipchains /proc/net/ip_masquerade -/sbin/ipchains /proc/sys/net/ipv4/ip_forward - -/sbin/ipmaddr /proc/net/dev_mcast -/sbin/ipmaddr /proc/net/igmp -/sbin/ipmaddr /proc/net/igmp6 - -/sbin/iptunnel /proc/net/dev - -/sbin/killall5 /proc/%s/cmdline -/sbin/killall5 /proc/%s/exe -/sbin/killall5 /proc/%s/stat -/sbin/killall5 /proc/version - -/sbin/klogd /proc/kmsg - -/sbin/lsmod /proc/modules - -/sbin/modprobe /proc/modules - -/sbin/pidof /proc/%s/cmdline -/sbin/pidof /proc/%s/exe -/sbin/pidof /proc/%s/stat -/sbin/pidof /proc/version - -/sbin/probe /proc/pci - -/sbin/rarp /proc/net/ax25 -/sbin/rarp /proc/net/nr -/sbin/rarp /proc/net/rarp - -/sbin/rmmod /proc/modules - -/sbin/rmmod.static /proc/modules - -/sbin/route /proc/net/appletalk -/sbin/route /proc/net/ax25 -/sbin/route /proc/net/ax25_route -/sbin/route /proc/net/ipx -/sbin/route /proc/net/ipx_route -/sbin/route /proc/net/nr -/sbin/route /proc/net/nr_neigh -/sbin/route /proc/net/nr_nodes -/sbin/route /proc/net/route -/sbin/route /proc/net/rt_cache -/sbin/route /proc/net/unix - -/sbin/scsi_info /proc/scsi -/sbin/scsi_info /proc/scsi/%s -/sbin/scsi_info /proc/scsi/scsi - -/sbin/slattach /proc/net/ax25 -/sbin/slattach /proc/net/nr - -/sbin/swapoff /proc/swaps - -/sbin/swapon /proc/swaps - -/sbin/uugetty /proc/version - -/usr/bin/dig /proc/ -/usr/bin/dig /proc/interrupts -/usr/bin/dig /proc/meminfo -/usr/bin/dig /proc/rtc -/usr/bin/dig /proc/self/status -/usr/bin/dig /proc/stat - -/usr/bin/dnsquery /proc/ -/usr/bin/dnsquery /proc/interrupts -/usr/bin/dnsquery /proc/meminfo -/usr/bin/dnsquery /proc/rtc -/usr/bin/dnsquery /proc/self/status -/usr/bin/dnsquery /proc/stat - -/usr/bin/eject /proc/mounts - -/usr/bin/emacs /proc/loadavg - -/usr/bin/fetchmail /proc/net/dev - -/usr/bin/free /proc/meminfo - -/usr/bin/gmake /proc/loadavg - -/usr/bin/gpm-root /proc/loadavg -/usr/bin/gpm-root /proc/meminfo - -/usr/bin/host /proc/ -/usr/bin/host /proc/interrupts -/usr/bin/host /proc/meminfo -/usr/bin/host /proc/rtc -/usr/bin/host /proc/self/status -/usr/bin/host /proc/stat - -/usr/bin/hoststat /proc/loadavg - -/usr/bin/hwdiag /proc/cpuinfo -/usr/bin/hwdiag /proc/pci -/usr/bin/hwdiag /proc/scsi/scsi -/usr/bin/hwdiag /proc/version - -/usr/bin/lsdev /proc/dma -/usr/bin/lsdev /proc/interrupts -/usr/bin/lsdev /proc/ioports - -/usr/bin/mailq /proc/loadavg - -/usr/bin/make /proc/loadavg - -/usr/bin/mcookie /proc/loadavg -/usr/bin/mcookie /proc/stat - -/usr/bin/newaliases /proc/loadavg - -/usr/bin/nslookup /proc/ -/usr/bin/nslookup /proc/interrupts -/usr/bin/nslookup /proc/meminfo -/usr/bin/nslookup /proc/rtc -/usr/bin/nslookup /proc/self/status -/usr/bin/nslookup /proc/stat - -/usr/bin/nsupdate /proc/ -/usr/bin/nsupdate /proc/interrupts -/usr/bin/nsupdate /proc/meminfo -/usr/bin/nsupdate /proc/rtc -/usr/bin/nsupdate /proc/self/status -/usr/bin/nsupdate /proc/stat - -/usr/bin/pgp /proc/version -/usr/bin/pgpe /proc/version -/usr/bin/pgpk /proc/version -/usr/bin/pgps /proc/version -/usr/bin/pgpv /proc/version - -/usr/bin/procinfo /proc/cmdline -/usr/bin/procinfo /proc/devices -/usr/bin/procinfo /proc/dma -/usr/bin/procinfo /proc/filesystems -/usr/bin/procinfo /proc/interrupts -/usr/bin/procinfo /proc/loadavg -/usr/bin/procinfo /proc/meminfo -/usr/bin/procinfo /proc/modules -/usr/bin/procinfo /proc/stat -/usr/bin/procinfo /proc/uptime -/usr/bin/procinfo /proc/version - -/usr/bin/purgestat /proc/loadavg - -/usr/bin/screen /proc/loadavg - -/usr/bin/strace /proc/%d/stat - -/usr/bin/top /proc/cpuinfo -/usr/bin/top /proc/meminfo -/usr/bin/top /proc/stat - -/usr/bin/vmstat /proc/%s/stat -/usr/bin/vmstat /proc/meminfo -/usr/bin/vmstat /proc/stat - -/usr/sbin/atd /proc/loadavg - -/usr/sbin/dnskeygen /proc/ -/usr/sbin/dnskeygen /proc/interrupts -/usr/sbin/dnskeygen /proc/meminfo -/usr/sbin/dnskeygen /proc/rtc -/usr/sbin/dnskeygen /proc/self/status -/usr/sbin/dnskeygen /proc/stat - -/usr/sbin/fuser /proc/%d/stat -/usr/sbin/fuser /proc/net/%s -/usr/sbin/fuser /proc/net/unix - -/usr/sbin/in.identd /proc/net/tcp - -/usr/sbin/irpd /proc/ -/usr/sbin/irpd /proc/interrupts -/usr/sbin/irpd /proc/meminfo -/usr/sbin/irpd /proc/rtc -/usr/sbin/irpd /proc/self/status -/usr/sbin/irpd /proc/stat - -/usr/sbin/named /proc/ -/usr/sbin/named /proc/interrupts -/usr/sbin/named /proc/meminfo -/usr/sbin/named /proc/rtc -/usr/sbin/named /proc/self/status -/usr/sbin/named /proc/stat - -/usr/sbin/named-xfer /proc/ -/usr/sbin/named-xfer /proc/interrupts -/usr/sbin/named-xfer /proc/meminfo -/usr/sbin/named-xfer /proc/rtc -/usr/sbin/named-xfer /proc/self/status -/usr/sbin/named-xfer /proc/stat - -/usr/sbin/readprofile /proc/profile - -/usr/sbin/rwhod /proc/loadavg -/usr/sbin/rwhod /proc/uptime - -/usr/sbin/sendmail /proc/loadavg - -/usr/sbin/setconsole /proc/openprom/options -/usr/sbin/setconsole /proc/openprom/options/${console}-mode -/usr/sbin/setconsole /proc/openprom/options/input-device -/usr/sbin/setconsole /proc/openprom/options/output-device - -Best wishes from Riley. - - * Copyright (C) 1999, Memory Alpha Systems. - * All rights and wrongs reserved. - -+----------------------------------------------------------------------+ -| There is something frustrating about the quality and speed of Linux | -| development, ie., the quality is too high and the speed is too high, | -| in other words, I can implement this XXXX feature, but I bet someone | -| else has already done so and is just about to release their patch. | -+----------------------------------------------------------------------+ - * http://www.memalpha.cx/Linux/Kernel/ - - -- -To unsubscribe from this list: send the line "unsubscribe linux-kernel" in -the body of a message to majordomo@vger.rutgers.edu -Please read the FAQ at http://www.tux.org/lkml/ - +Busybox TODO + +Stuff that needs to be done + +tr - missing SuS3 features in busybox 1.0pre10 + +tr doesnt support [:blank:], [:digit:] or other predefined classes, [=equiv=] +support is also missing. +---- +find + doesn't understand () or -exec, and these are actually used out in the real + world. The "make uninstall" of lots of things (including busybox itself) + breaks because of this, and sometimes even "make install" (like udev). +---- +sh + The command shell situation is a big mess. We have three or four different + shells that don't really share any code, and the "standalone shell" doesn't + work all that well (especially not in a chroot environment), due to apps not + being reentrant. Unifying the various shells and figuring out a configurable + way of adding the minimal set of bash features a given script uses is a big + job, but it be a big improvement. + + Note: Rob Landley (rob@landley.net) is working on this one, but very slowly... +--- +gzip + Can't handle compressing multiple files at once. (I don't mean making a + multiple file archive, I mean compressing more than one file at a time.) + Some global variables aren't re-initialized between runs. +--- +gunzip + same problem as gzip. "gunzip one.gz two.gz three.gz" doesn't work for + two.gz and three.gz due to global variables not getting reset. +--- +diff + We should have a diff -u command. We have patch, we should have diff + (we only need to support unified diffs though). +--- +fuser + Would be nice. The basic susv3 options, plus fuser -k. +--- +patch + should have simple fuzz factor support to apply patches at an offset which + shouldn't take up too much space. +--- +man + It would be nice to have a man command. Not one that handles troff or + anything, just one that can handle preformatted ascii man pages, possibly + compressed. This could probably be a script in the extras directory that + calls cat/zcatbzcat | more +--- +bzip2 + Compression-side support. + + +Architectural issues: + +Do a SUSv3 audit + Look at the full Single Unix Specification version 3 (available online at + "http://www.opengroup.org/onlinepubs/009695399/nfindex.html") and + figure out which of our apps are compliant, and what we're missing that + we might actually care about. + + Even better would be some kind of automated compliance test harness that + exercises each command line option and the various corner cases. +-- +Unify archivers + Lots of archivers have the same general infrastructure. The directory + traversal code should be factored out, and the guts of each archiver could + be some setup code and a series of callbacks for "add this file", + "add this directory", "add this symlink" and so on. + + This could clean up tar and zip, and make it cheaper to add cpio and ar + write support, and possibly even cheaply add things like mkisofs someday, + if it becomes relevant. +--- +Text buffer support. + Several existing applets and potential additions (sort, vi, less...) read + a whole file into memory and act on it. There might be an opportunity + for shared code in there that could be moved into libbb... +--- +Individual compilation of applets. + It would be nice if busybox had the option to compile to individual applets, + for people who want an alternate implementation less bloated than the gnu + utils (or simply with less political baggage), but without it being one big + executable. + + Turning libbb into a real dll is another possibility, especially if libbb + could export some of the other library interfaces we've already more or less + got the code for (like zlib). +--- +buildroot - Make a "dogfood" option + Busybox is now capable of replacing most gnu packages for real world use, + such as developing software or in a live CD. A system built from busybox + (1.00 with updated sort.c), uclibc 0.9.27, gcc, binutils, make, and a few + other development tools (http://www.landley.net/code/firmware has an example + system using autoconf, automake, bison, flex, libtools, m4, zlib, + and groff: dunno what subset of that is actually necessary) is capable of + rebuilding itself, from scratch, under itself. + + It would be a good "eating our own dogfood" test if buildroot had the option + of using busybox instead of bzip2, coreutils, file, findutils, gawk, grep, + inetutils, modutils, net-tools, procps, sed, shadow, sysklogd, sysvinit, tar, + util-linux, and vim. Anything that's wrong with the resulting system, we + can fix. (It would be nice to be able to upgrade busybox to be able to + replace bash, diffutils, gzip, less, and patch as well.) +--- +Memory Allocation + We have a CONFIG_BUFFER mechanism that lets us select whether to do memory + allocation on the stack or the heap. Unfortunately, we're not using it much. + We need to audit our memory allocations and turn a lot of malloc/free calls + into RESERVE_CONFIG_BUFFER/RELEASE_CONFIG_BUFFER. + + And while we're at it, many of the CONFIG_FEATURE_CLEAN_UP #ifdefs will be + optimized out by the compiler in the stack allocation case (since there's no + free for an alloca()), and this means that various cleanup loops that just + call free might also be optimized out by the compiler if written right, so + we can yank those #ifdefs too, and generally clean up the code. +--- +Switch CONFIG_SYMBOLS to ENABLE_SYMBOLS + + In busybox 1.0 and earlier, configuration was done by CONFIG_SYMBOLS + that were either defined or undefined to indicate whether the symbol was + selected in the .config file. They were used with #ifdefs, ala: + + #ifdef CONFIG_SYMBOL + if (other_test) { + do_code(); + } + #endif + + In 1.1, we have new ENABLE_SYMBOLS which are always defined (as 0 or 1), + meaning you can still use them for preprocessor tests by replacing + "#ifdef CONFIG_SYMBOL" with "#if ENABLE_SYMBOL". But more importantly, we + can use them as a true or false test in normal C code: + + if (ENABLE_SYMBOL && other_test) { + do_code(); + } + + (Optimizing away if() statements that resolve to a constant value + is known as "dead code elimination", an optimization so old and simple that + Turbo Pascal for DOS did it twenty years ago. Even modern mini-compilers + like the Tiny C Compiler (tcc) and the Small Device C Compiler (SDCC) + perform dead code elimination.) + + Right now, busybox.h is #including both "config.h" (defining the + CONFIG_SYMBOLS) and "bb_config.h" (defining the ENABLE_SYMBOLS). At some + point in the future, it would be nice to wean ourselves off of the + CONFIG versions. (Among other things, some defective build environments + leak the Linux kernel's CONFIG_SYMBOLS into the system's standard #include + files. We've experienced collisions before.) +--- +FEATURE_CLEAN_UP + This is more an unresolved issue than a to-do item. More thought is needed. + + Normally we rely on exit() to free memory, close files, and unmap segments + for us. This makes most calls to free(), close(), and unmap() optional in + busybox applets that don't intend to run for very long, and optional stuff + can be omitted to save size. + + The idea was raised that we could simulate fork/exit with setjmp/longjmp + for _really_ brainless embedded systems, or speed up the standalone shell + by not forking. Doing so would require a reliable FEATURE_CLEAN_UP. + Unfortunately, this isn't as easy as it sounds. + + The problem is, lots of things exit(), sometimes unexpectedly (xmalloc()) + and sometimes reliably (bb_perror_msg_and_die() or show_usage()). This + jumps out of the normal flow control and bypasses any cleanup code we + put at the end of our applets. + + It's possible to add hooks to libbb functions like xmalloc() and bb_xopen() + to add their entries to a linked list, which could be traversed and + freed/closed automatically. (This would need to be able to free just the + entries after a checkpoint to be usable for a forkless standalone shell. + You don't want to free the shell's own resources.) + + Right now, FEATURE_CLEAN_UP is more or less a debugging aid, to make things + like valgrind happy. It's also documentation of _what_ we're trusting + exit() to clean up for us. But new infrastructure to auto-free stuff would + render the existing FEATURE_CLEAN_UP code redundant. + + For right now, exit() handles it just fine.