forks/toybox/scripts, branch 0.7.2Toybox with POSIX patches from E5ten
https://git.carbslinux.org/forks/toybox/atom?h=0.7.22016-10-20T20:26:38+00:00Always regenerate config.h for now, so make single and make don't clash.2016-10-20T20:26:38+00:00Rob Landleyrob@landley.net2016-10-20T20:26:38+00:00urn:sha1:3214c87ccf7f4a3426fe0d30296af1af85fc864e
The problem is that .config and .singleconfig produce the same header file,
so checking that the header is newer than the .config file won't help if
it was last generated from the other file.
Test infrastructure: collate make "test_single" and "make tests" into common2016-10-18T21:52:17+00:00Rob Landleyrob@landley.net2016-10-18T21:52:17+00:00urn:sha1:ee14fc396dff50a263b62670f5484efcd6316aeb
function, and add $C variable with an absolute path to the command being tested
(you need to call things like printf by path to avoid shell builtins, might as
well be consistent).
Add install_airlock target for hermetic build environment setup, and rebuild2016-10-03T03:41:55+00:00Rob Landleyrob@landley.net2016-10-03T03:41:55+00:00urn:sha1:6c3188cf0e551cd07e28398262cffb8b5b82a114
headers during install to make other install targets robust against single
command builds between "make" and "make install" of multiplexer version.
Kernels between 2006 and 2010 (such as the one in Centos 6.6) provide unshare()2016-09-28T19:22:34+00:00Rob Landleyrob@landley.net2016-09-28T19:22:34+00:00urn:sha1:c99cbd12ca56cadbd5d3dc6dfc77508701d94130
but not nsenter(). So probe for both.
make test_cmp caused the make invocation to exit with an error because the2016-09-05T00:13:31+00:00Rob Landleyrob@landley.net2016-09-05T00:13:31+00:00urn:sha1:8361fbd8ea9c6f29602982d31f67351f42f736b0
last test exited with nonzero status and the script preserved it. In the
current context, that's counterproductive.
The recent option parsing changes broke the multiplexer, because the "toybox"2016-07-17T07:14:35+00:00Rob Landleyrob@landley.net2016-07-17T07:14:35+00:00urn:sha1:48ea55969907286d1a125fcd655be9d8307a3449
command was trying to parse their options and failing. (Commands
that aren't just disabled but have no option string even in the allyesconfig
case need an OPTSTR of 0.)
This is probably only a partial fix, there's some design work to think through
here.
Add optional openssl accelerated versions of hash functions, loosely based on2016-07-15T09:45:08+00:00Rob Landleyrob@landley.net2016-07-15T09:45:08+00:00urn:sha1:adef5dcb1857f524e2a24dd0223f31db7cd7a8b9
a patch from Elliott Hughes, who said:
[PATCH] Add support for libcrypto for MD5/SHA.
Orders of magnitude faster (for architectures where OpenSSL/BoringSSL
has optimized assembler).
Also adds sha224sum, sha256sum, sha384sum, and sha512sum for folks
building with libcrypto.
The fallback portable C implementations could easily be refactored
to be API-compatible, but I don't know whether they'd stay here or
move to lib/ so I've left that part alone for now.
Fix generated/make.sh. (Quotes in TOYBOX_VERSION need to wind up in2016-07-06T00:36:00+00:00Rob Landleyrob@landley.net2016-07-06T00:36:00+00:00urn:sha1:d620f77f21638d2e9142ebd523a781e831757251
resulting macro.)
Fix option parsing infrastructure bug.2016-07-01T01:41:07+00:00Rob Landleyrob@landley.net2016-07-01T01:41:07+00:00urn:sha1:ec0b482e9fd27cb994b30e0f1b8a4d39a85735cc
In main.c: used adjusted string with placeholers for removed options (so flag
values don't move based on config, allowing FORCE_FLAGS to work).
In scripts/mkflags.c: test was wrong, a bare longopt never matches a single
letter command, but it was treated as always matching when comparing
allyesconfig vs current config to determine which options were disabled, so
there was a corner case that got the flag values wrong.
Add NOSPACE=1 to allow tests to pass with diff -b2016-06-30T15:39:23+00:00Rob Landleyrob@landley.net2016-06-30T15:39:23+00:00urn:sha1:fcbf55ad565861e22e1cb297e6954b9d82f5094b