forks/toybox/lib, branch 0.0.2 Toybox with POSIX patches from E5ten https://git.carbslinux.org/forks/toybox/atom?h=0.0.2 2007-01-25T21:11:23+00:00 Update args.c to implement numeric arguments. 2007-01-25T21:11:23+00:00 Rob Landley rob@landley.net 2007-01-25T21:11:23+00:00 urn:sha1:028a54424e08cb8bf88bd2014ffd7d485f15ba51 The fdlength() ioctl apparently doesn't work on files (and the lseek trick 2007-01-25T21:10:37+00:00 Rob Landley rob@landley.net 2007-01-25T21:10:37+00:00 urn:sha1:b3a3382abf49573ba1f79d22417e9fc50125d28d doesn't work on some devices, and we can't always tell _when_ it failed), so go to the binary search for now. More random progress on mke2fs. Nothing to see yet. 2007-01-23T18:20:38+00:00 Rob Landley rob@landley.net 2007-01-23T18:20:38+00:00 urn:sha1:e2580dbebbd43f668913b3d2ae6c0636161636ed Add "echo". Has -n and -e (but not \0123 yet). 2007-01-20T23:04:20+00:00 Rob Landley rob@landley.net 2007-01-20T23:04:20+00:00 urn:sha1:24d1d45ee901e74b3388e7053543eb16aea09ada Merge a memset with an existing loop, tweak comments. 2007-01-20T17:30:19+00:00 Rob Landley rob@landley.net 2007-01-20T17:30:19+00:00 urn:sha1:125a2c74f2ef9438b09930b208dd88bf7538bac8 Merge two similar loops and expand a comment. 2007-01-19T21:31:11+00:00 Rob Landley rob@landley.net 2007-01-19T21:31:11+00:00 urn:sha1:94515a2803d8db4b5a0c1cc51554f79a98803891 Another suggestion from Manuel: Grab 2 bits instead of 1 inside a loop. Saves 2007-01-19T21:01:54+00:00 Rob Landley rob@landley.net 2007-01-19T21:01:54+00:00 urn:sha1:59b301786890b909d64ddc61e279394184fd57d7 4 bytes and reduces running time by one half of one percent. In bunzip replace setjmp/longjmp handling with error_exit(), replace string 2007-01-19T03:00:12+00:00 Rob Landley rob@landley.net 2007-01-19T03:00:12+00:00 urn:sha1:6000f13e8ac8c17c1f01fcafc8066bb7b419f224 based handling of a 6-byte header with with two 24-bit integer reads. Use xmalloc() and xzalloc(). Add fdprintf(). Remove reread() and rewrite() which handle -EINTR, which 2007-01-19T02:54:08+00:00 Rob Landley rob@landley.net 2007-01-19T02:54:08+00:00 urn:sha1:901637760b4206e968e73dd5ff7430c107c27b57 shouldn't be a problem if we register signal handlers with sigaction(SA_RESTART) Straighten out count and len (I generally consistently use "count" for the current progress and "len" for the total, but this time I got them backwards for some reason and don't want to confuse myself in future.) More optimizations originally suggested by Manuel Nova: Use a sentinel value 2007-01-18T23:16:11+00:00 Rob Landley rob@landley.net 2007-01-18T23:16:11+00:00 urn:sha1:813840c9d199bf9e26d23291cf23e22a68e8ad82 for limit[] to move a test out of a loop. Unroll a single-bit get_bits() to avoid a function call in the common case on a hot path. And one more application of the old "two tests in one via typecasting and/or math" trick.