aed8a602b8
stdio: Add support for remove()
2016-08-25 08:40:34 -06:00
6a58f046e6
CONFIG_LIBC_DIFFTIME is no longer used, but was still in Kconfig file. Noted by Sebastien Lorquet
2016-08-25 08:09:23 -06:00
51596dc457
Updata difftime. Add a version of difftime for the case where the platform does not support type double
2016-08-25 08:00:50 -06:00
a626ba5b70
libc/time: This commit adds the difftime() function. Since the function returns a double, I have isolated it in a CONFIG_LIBC_DIFFTIME option (It also depends on the toolchain-dependent CONFIG_HAVE_DOUBLE so is not available on tiny platforms).
2016-08-24 15:43:57 -06:00
67851849da
Upate ChangeLog
2016-08-22 10:31:52 -06:00
60b70f7dbb
Make xorshift128 re-entrant so that we do do have to suffer the overhad of serialization controls.
2016-08-21 07:47:53 -06:00
757023a909
XorShift128: Add some protection to handle the case where this logic could be called from the interrupt level.
2016-08-20 17:27:41 -06:00
6f833be9d5
Separate XorShift128 PRNG from /dev/urandom and make it generally available.
2016-08-20 11:36:02 -06:00
6e5010e0d0
floor(), floorf(), and floorl(): Fix logic error. Was not correctly handling negative integral value.
2016-08-11 18:21:29 -06:00
155055d564
strtod(): Was not returning endptr on error conditions.
2016-08-11 18:20:25 -06:00
0e35bad987
Update some comments
2016-08-11 10:12:04 -06:00
986c568d34
Correct file header comments
2016-08-07 10:04:02 -06:00
4daa553328
lib_asinl.c edited online with Bitbucket. Add missing # on endif. Noted by David Alessio.
2016-07-31 16:47:25 +00:00
c145159c6b
This commit fixes the following libc/math issues:
...
1) asin[f l]() use Newton’s method to converge on a solution. But Newton’s method converges very slowly (> 500,000 iterations) for values of x close to 1.0; and, in the case of asinl(), sometimes fails to converge (loops forever). The attached patch uses an trig identity for values of x > sqrt(2). The resultant functions converge in no more than 5 iterations, 6 for asinl().
2) The NuttX erf[f l]() functions are based on Chebyshev fitting to a good guess. The problem there’s a bug in the implementation that causes the functions to blow up with x near -3.0. This patch fixes that problem. It should be noted that this method returns the error function erf(x) with fractional error less than 1.2E-07 and that’s fine for the float version erff(), but the same method is used for double and long double version which will yield only slightly better precision. This patch doesn't address the issue of lower precision for erf() and erfl().
3) a faster version of copysignf() for floats is included.
2016-07-30 15:43:56 -06:00
d36d9d61be
Trivial spacing change
2016-07-28 09:34:56 -06:00
2b3bc90ba5
Correct a comment
2016-07-27 09:09:40 -06:00
d4f3954b35
lib_dumpbuffer: Now prints a large on-stack buffer first to avoid problems when the syslog output is prefixed with time.
2016-07-27 08:40:46 -06:00
fd42f898f6
libc: Add mkfifo() and pipe() to libc CSV file
2016-07-22 11:02:51 -06:00
7370d3171a
Move include/nuttx/lib.h to include/nuttx/lib/lib.h
2016-07-21 14:05:44 -06:00
779d3f98ed
Move include/nuttx/math32.h to include/nuttx/lib/math32.h
2016-07-21 14:00:15 -06:00
23a876ab9c
Move include/nuttx/math.h to include/nuttx/lib/math.h
2016-07-21 13:58:09 -06:00
72a45c82b7
Move include/nuttx/regex.h to include/nuttx/lib/regex.h
2016-07-21 13:42:18 -06:00
ee9c66186c
ramdisk.h moved from include/fs/nuttx/ to include/nuttx/drivers.
2016-07-20 14:02:18 -06:00
4b4dbc79a2
Move driver related prototypes out of include/nuttx/fs/fs.h and into new include/drivers/drivers.h
2016-07-20 13:15:37 -06:00
e73e82a923
Move prototypes of mkfifo2() from sys/stat.h and pipe2() from unistd.h. They are non-standard and should not be in such a public place. They are not in include/nuttx/fs/fs.h (which isn't a great place either).
2016-07-20 11:57:18 -06:00
b28fa8a609
Move pipe() and mkpipe() to nuttx/libc. Change syscalls to pipe2() and mkfifo2()
2016-07-19 14:15:26 -06:00
5a35abb488
Improve some comments
2016-07-17 08:31:02 -06:00
07e20479ad
/dev/random: Add configuration option to use the congruential PRNG.
2016-07-17 07:56:25 -06:00
ffd3a31649
Rename lib_rand.c to lib_srand.c
2016-07-17 07:11:09 -06:00
95555a0199
PTY: Add ioctls to support locking and unlocking of the slave. LIBC: Add implementation of unlockpt()
2016-07-15 09:39:33 -06:00
6139860c0b
PTY: Added PTY IOCTL command definitions. Add support for TIOCGPTN. LIB: Implement ptsname() and ptsname_r()
2016-07-15 08:33:47 -06:00
912ad2d345
Fix bug in logf() algorithm that caused erroneous INFINITY results.
2016-07-14 20:15:37 -06:00
96edfdb4c7
libm: Fix/improve range checks in lib_expi() and lib_expif().
2016-07-12 09:59:10 -06:00
3bc504b685
Per comment from David Alession: M_E128 will never be used since it is greater than FLT_MAX.
2016-07-12 09:59:08 -06:00
dfe8596e53
Cosmetic
2016-07-11 14:15:34 -06:00
f9d4b50207
If there are no streams, let printf() fall back to use syslog() for output.
2016-07-11 13:52:21 -06:00
29eae10232
lib_expi/f(): Exponental table should be 'const'. Dimension was wrong for float version.
2016-07-11 13:53:31 -06:00
6a081e118d
Add guess at missing lib_libexpif.c file
2016-07-11 13:43:43 -06:00
261358f1f5
This change should significantly improve the performance of single precision floating point math library functions. The vast majority of changes have to do with preventing the compiler from needlessly promoting floats to doubles, performing the calculation with doubles, only to demote the result to float. These changes only affect the math lib functions that return float.
2016-07-11 07:02:50 -06:00
830af912a1
Add sig_raise()
2016-07-04 09:32:36 -06:00
bde90de12f
Fix libc/zoneinfo build
2016-06-30 16:20:43 -06:00
aa96a1d2ec
Move the TZ/Olson zoneinfo data set from apps/system/zoneinfo to libc/zoneinfo
2016-06-30 15:40:37 -06:00
67ea112767
Update ChangeLog
2016-06-27 13:08:40 -06:00
2a19e16eb0
libc/hex2bin: Move the portable library portion of apps/system/hex2bin to nuttx/libc/hex2bin where it can be shared with the OS internals.
2016-06-27 11:22:38 -06:00
7f16548f57
Replaces last three commits. Does the same thing, but does it in a way that does not change the usage model.
2016-06-21 05:26:08 -06:00
b82743612f
Even the 'unconditional' _alert() macro should be disabled if all debug features are disabled.
2016-06-20 22:38:54 -06:00
505ca542e8
Remove some last traces of lowvsyslog that were missed; Add a SYSLOG emergency channel for handling assertion output more cleanly
2016-06-20 16:11:50 -06:00
2a751068e6
Without lowsyslog() *llerr() is not useful. Eliminate and replace with *err().
2016-06-20 12:44:38 -06:00
43eb04bb8f
Without lowsyslog() *llinfo() is not useful. Eliminate and replace with *info().
2016-06-20 11:59:15 -06:00
d40a473f72
Without lowsyslog() *llwarn() is not useful. Eliminate and replace with *warn().
2016-06-20 09:37:08 -06:00
Sebastien Lorquet
Gregory Nutt
David S. Alessio
Pierre-noel Bouteville
David Alessio