aboutsummaryrefslogtreecommitdiff
path: root/toys
AgeCommit message (Collapse)Author
2014-06-11Cleanup pass on loginRob Landley
2014-06-11Enable fallocate in defconfig.Rob Landley
2014-06-11Cleanup pass on host.cRob Landley
2014-06-11Promote strings.Rob Landley
2014-06-11Cleanup strings.Rob Landley
2014-06-10pending/useradd: unbreak buildIsaac Dunham
When useradd started using xfork(), the conditional in else if (pid > 0) became unnecessary, since else means pid is nonzero and xfork makes it non-negative. However, the "if" was not deleted.
2014-06-09Promote rfkill.Rob Landley
2014-06-09Cleanup pass on rfkill.Rob Landley
2014-06-09When locale is enabled, sprintf("%.123s", str) is counting characters, not ↵Rob Landley
bytes, so we can't globally enable locale without opening stack/heap smashing vulnerabilities. Make commands individually request setlocale() using TOYFLAGS instead.
2014-06-08Warning fix from Rich Felker.Rob Landley
2014-06-08Add host by Rich Felker.Rob Landley
2014-06-07Make md5sum/sha1sum -b flag be "brief" output (just the hash).Rob Landley
(It's not like systems implementing -b binary and -t text still matter.)
2014-06-03mount: start on option parsing, implement loopback and bind mount autodetection.Rob Landley
2014-06-02Help text should have a blank line after usage: lines, and a couple other ↵Rob Landley
whitespace tweaks.
2014-06-01cpio: archive more filesIsaac Dunham
While writing tests for cpio, I found that cpio tries to open empty files if they're regular files, and fails to archive them if unreadable. This can be easily avoided, and is not the usual behavior.
2014-05-31iconv: some fixesFelix Janda
- fix problem with sequences at buffer boundaries - add (ignored) -c and -s options - don't try to continue with a file when read() fails
2014-05-31Promote partprobe.Rob Landley
2014-05-31Cleanup partprobe.Rob Landley
2014-05-31Introduce xfork() and make commands use it, and make some WEXITSTATUS() use ↵Rob Landley
WIFEXITED() and WTERMSIG()+127.
2014-05-29killall5 - kill all the processes not in its session.Ashwini Sharma
2014-05-29makedevs - making devices/nodes in a range. Supports reading the tabled ↵Ashwini Sharma
entry from file.
2014-05-29strings - print the strings in the file.Ashwini Sharma
2014-05-29First stab at mount, very incomplete.Rob Landley
2014-05-29Make "losetup /dev/loop0 filename" work.Rob Landley
Sigh. Implement the complex cases and you screw up the simple cases you already tested...
2014-05-29Switch mtab_list to doubly linked so we can traverse in either order. ↵Rob Landley
Convert umount and df. Add dlist_terminate() to break lists for traversal in either direction.
2014-05-26Brush the dust off toysh.Rob Landley
Simplify the config micromanagement tangle to just a single "interactive" option. Fix an unused variable and wrong variable type.
2014-05-25I have attached a patch adding a program that allows re-reading the ↵Bertold Van den Bergh
partition table. This is often used on embedded systems booting from SD/USB devices that need to resize partitions on first boot.
2014-05-25atolx_range() is already added into lib/lib.c, in vconfig.c atolx_range() ↵Ashwini Sharma
can be used in place of locally defined strtorange() function.
2014-05-25logname and whoami are the same as id -un, so merge them.Isaac Dunham
Since the starting letters are greater than those for 'id' or 'groups', we cn just check if the first letter is greater than 'i'.
2014-05-24Bugfix from Isaac Dunham (new pci database has # comments in it) and some ↵Rob Landley
minor cleanups I had in my tree already.
2014-05-24Pending commands should default nRob Landley
2014-05-24Single-user login.Ashwini Sharma
2014-05-24Remove debug detritus I didn't mean to to check in, and treat an "this ↵Rob Landley
variable can never actually be used uninitialized but gcc's warning generator can't tell and fails spamwards" warning.
2014-05-23rfkill - enable/disable the radio devicesAshwini Sharma
e.g. Wireless adapter, Bluetooth devices...
2014-05-23inotifyd - watch for filesystem events.Ashwini Sharma
2014-05-23Ashwini Sharma pointed out I screwed up last.c.Rob Landley
Renamed the function, missed a user...
2014-05-22date: add -ds, document +FORMAT escapes.Rob Landley
2014-05-21Convert bootchartd to generic_signal().Rob Landley
2014-05-21Make telnet use generic_signal(), minor in-passing cleanups.Rob Landley
2014-05-21Make telnetd use generic_signal(), inline kill_session(), close race window ↵Rob Landley
where a SIGCHLD could get lost.
2014-05-21First pass init cleanup: use sigatexit() to set multiple signal handlers, ↵Rob Landley
rename set_sane_term() to reset_term() and have it take the fd it works on as an argument, some whitespace and help text tweaks.
2014-05-21Make fsck.c use common list free function, collate cleanup code and move inline.Rob Landley
2014-05-21Make last use common llist free function, minor cleanups.Rob Landley
2014-05-21dhcp client had a segfault, when DHCP message contained 'pad' option.Ashwini Sharma
The parsing logic kept checking for other options beyond __pad__ option, without checking if it was __end__ option after that or not.
2014-05-20lspci text output: add -i, allow -n when disabledIsaac Dunham
-i is how standard lspci handles changing the PCI ID database. lspci -n should be a no-op when text is disabled.
2014-05-19Cleanup pass on bootchartd.Rob Landley
Might have broken something, don't actually have a test case for bootchartd yet.
2014-05-18Quick cleanup pass on ps.Rob Landley
2014-05-18Cleanup pass on lspciRob Landley
2014-05-15bootchartd by Bilal Qureshi.Rob Landley
2014-05-15Here's a quick cleanup of md5sum. Executive summary: smaller and faster.Daniel Verkamp
On my machine, for a 2.2 GB file of random bytes, the timings with warm cache are: toybox before: 11.4 seconds toybox after: 8.3 seconds GNU md5sum: 3.9 seconds openssl dgst -md5: 3.5 seconds This is clearly better than before (3x openssl), but still slow (2x openssl). I suspect there is more low-hanging fruit to be had by eliminating the memcpy in hash_update (maybe not too much - hash_update accounts for about 4% of total runtime versus 92% for md5_transform according to perf - but this would also help sha1sum). make bloatcheck on x86_64 gcc 4.8.2 -Os: name old new delta ----------------------------------------------------------------------- md5rot 0 64 64 md5_transform 365 223 -142 ----------------------------------------------------------------------- -78 total Rationale for the changes: Move definition of 'rol' up so it can be used in md5_transform. This is purely cosmetic; it expands to exactly the same code. Put rotation counts in a lookup table instead of calculating them on the fly. This is mostly a wash size-wise, +5 bytes total, but worthwhile for readability and speed. Instead of accessing the state array using a rotating index (the variable formerly known as 'a'), access the state with constant offsets and rotate the contents of the array instead. This is the big win - it eliminates all the crazy memory addressing math inside the loop.