| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
signal handling.
|
|
TOYBOX_DEBUG enabled.
One again gcc manages to be incompetent about producing warnings. For example,
in bzcat we error_exit() with a message that's an error code fed into an
array of string constants, and apparently dereferencing an array of string
literals doesn't give you a string literal according to -Wformat-security.
Not breaking the code to humor the compiler here, the compiler is wrong.
|
|
i'll be AFK for a week, so here's the patch i've been using this
evening to find other format string mistakes. BSD uses __printflike
and takes two arguments instead of hard-coding (1,2), but i figured
that as long as you don't need the generality you'd prefer not to have
it. and it's easy enough to retrofit if we ever do have a formatting
function that takes other arguments.
|
|
|
|
Inline open_wall_clock_rtc() into rtc_open(), factor out xtzset(),
inline set_sysclock_from_hwclock(), set_hwclock_from_sysclock(),
and set_sysclock_timezone().
/etc/adjtime is short enough we don't need to bother with a break.
The final "else" case in main() should always trigger because >0 in optstr
says "at most zero arguments", so the || at the end is always true, so
take the test off.
|
|
|
|
faccessat(AT_SYMLINK_NOFOLLOW) is not supported.
|
|
|
|
a static inline in portability.h, and prototype dirname() while we're at it.
|
|
I considered a #else on the big #ifdef __GLIBC__ above, but that
seemed harder to follow.
|
|
|
|
|
|
|
|
the SELinux commands (such as chcon) and the SELinux-specific options
to regular commands (such as ls -Z).
This lets us #include <selinux/selinux.h> in portability.h.
I've also fixed chcon to insist on being given the a context argument.
This patch also adds -Z to id and fixes id's regular output (-G should
be separated by spaces, non-G output should be separated by commas,
and you don't want a double comma where the egid is omitted from the
list of groups).
|
|
|
|
add estrtol() (which clears errno first), and xstrtol() (which error_exit()s on overflow).
|
|
|
|
macros for a disabled command (needed when multiple commands share infrastructure with a common set of flags).
This means the flag space is no longer packed, but leaves gaps where the zeroes
go. (Actual flag bit positions are the same for all configs.) Since the
option parsing needs to know where the holes are, the OPTSTR values are
now generated as part of flags.h with ascii 1 values for the disabled values.
(So generated/oldflags.h went away.)
This also means that the option string argument for OLDTOY() went away, it now
uses the same arguments as the NEWTOY() it references.
|
|
random other symbol name (because gnu) and this screws up nontrivial macro expansions of NEWTOY(basename), so work around it in portability.h.
|
|
|
|
|
|
The tizen guys wanted this. Yeah, I know there's base64 code in
uuencode/uudecode, but that this has -i, input lines aren't of fixed length,
encode/decode are in same file, there's no prefix/suffix code, it always
writes to stdout... Eliminating the code duplication wouldn't be worth
the if/else I'd have to add, so I just did a new one.
Factored out the base64 table init into lib.c though: that was worth sharing.
|
|
The semantics of strncat() and strncpy() are non-obvious, so let's not use 'em.
Both zero all their remaining buffer space, and with strncat() the size is
the space left at the _end_ of the string (not the size of the buffer) so
it's way too easy to stomp memory you don't own. As long as we have to measure
stuff ourselves to get it right, just use memcpy().
|
|
xfork() to portability.h and #ifdef based on __uClinux__ (which seems to be the nommu compiler define).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The CFG_* symbols are always defined so if() can use them as compile-time
constants, so don't if defined() them.
Doing USE_BLAH() around variable definitions opens up the same potential for
config-dependent build breaks as #ifdefs do, just make the whole command
depend on the symbol for now, factor out the utmpx infrastructure later.
The PTY probe was always failing because it used NULL without #including
the header that defines it. Substitute 0 instead.
|
|
Android is missing all of these; we need to probe for some so we have
a config symbol to depend on.
sethostname() is easily replaced.
We got termios.h via pty.h; now it's not included in configure-step tools,
so we need termios.h to generate globals.
|
|
|
|
attempts to read from stdout instead of stdin for "-" or no arguments.
|
|
|
|
|
|
|
|
|
|
The Linux man page says I can use AT_SYMLINK_NOFOLLOW. It works in glibc,
uclibc, and klibc, but musl returns -EINVAL any time you pass in that flag
and the maintainer says that's not a bug and insists the man page and those
other libraries all change to match musl's behavior.
Toybox uses it to avoid scheduling unnecessary metadata writes for things we're
about to delete (have to chmod unreadable directories so we can descend into
them to delete their contents, the chmod happens before we descend so the
disk I/O has plenty of time to be scheduled) because the extra writes wear out
SSD faster. It's just an optimization and I don't really care if it works
_well_ (the fchmodat call _also_ takes AT_SYMLINK_NOFOLLOW so that's covered),
but musl's behavior uniquely makes the check always error and thus breaks normal
"rm -r".
Yes this workaround is checking #ifdef __MUSL__ which the library does not
supply (because its code is perfect and will thus never need to be worked
around). You can CFLAGS=-D__MUSL__ if you don't echo "#define __MUSL__" >>
include/features.h when installing the library.
|
|
|
|
another command's help.
|
|
(line separator), and "/" (filename separator).
Restricting usernames to the legacy posix character allowed set (for filenames,
so the $HOME directory is creatable on VFAT and similar) means you can't have
UTF-8 usernames. Linux allows any character but / and NUL in filenames.
Since root is creating these entries, we assume root knows what it's doing.
|
|
Sharma.
dirtree->name is an array, not a pointer, so can't be zero. Remove the test.
We dereference plen without checking it for null but calling dirtree_path(0, 0)
is pilot error: only the _first_ call can have plen = 0. Add a comment.
|
|
I've been locally patching uClibc to not violate posix-2008 (you don't need
to define a GNU macro to get a posix function), but uClibc is obsolete and
moribund (development peaked in 2006, last bugfix release was over 2 years
ago), and the largest remaining user (buildroot) doesn't bother to apply such
a patch. Since even buildroot is slowly migrating to musl-libc, just do the
portability tweak for what the last release of the old thing actually did.
|
|
|
|
we want to redirect both, one, or neither of stdin/stdout.
|
