incubator-nuttx/libc
Gregory Nutt b901f08d90 Trivial changes from review of last PR. 2018-02-26 11:48:57 -06:00
..
aio This change renames all internal, private NuttX signal-related functions to use the prefix nxsig_ so that they cannot be confused with application interfaces that begin, primarily, with sig_ 2017-10-05 13:25:25 -06:00
audio Merged in hardlulz/modem-3.0-nuttx/fix-sem-EINTR (pull request #603) 2018-02-20 18:24:53 +00:00
bin Changes for ZDS-II ez80 build 2013-11-28 17:47:03 -06:00
dirent Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
dllfcn libc/Kconfig: Use menus vs. comments to better organize the C library menus 2017-02-10 10:04:56 -06:00
fixedmath SDCC Fixes: Change some prototypes and some assumptions about the size of unsigned int to get to a clean SDCC compile. 2017-10-08 08:50:18 -06:00
hex2bin Make sure that labeling is used consistently in all function headers (part 3). 2018-02-01 12:17:03 -06:00
inttypes Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
kbin Changes for ZDS-II ez80 build 2013-11-28 17:47:03 -06:00
libgen Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
locale Fix lots of occurrences of 'the the', 'the there', 'the these', 'the then', 'the they. 2017-05-11 13:35:56 -06:00
machine Merged in ccondurache/nuttx/fix_elf_arm_checkarch (pull request #422) 2017-06-28 16:19:51 +00:00
math libc/match: Use of exp() vs expf() in logf() caused function to be slow. 2017-10-19 09:14:34 -06:00
misc Merged in hardlulz/modem-3.0-nuttx/fix-sem-EINTR (pull request #603) 2018-02-20 18:24:53 +00:00
modlib Merged in hardlulz/modem-3.0-nuttx/fix-sem-EINTR (pull request #603) 2018-02-20 18:24:53 +00:00
net Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
netdb Trivial changes from review of last PR. 2018-02-26 11:48:57 -06:00
pthread Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
queue Make sure that labeling is used consistently in all function headers (part 2). 2018-02-01 12:03:55 -06:00
sched Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
semaphore Fix various issues noted by Coverity 2018-02-06 09:13:16 -06:00
signal Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
spawn Squashed commit of the following: 2018-01-30 11:08:18 -06:00
stdio Make sure that labeling is used consistently in all function headers (part 2). 2018-02-01 12:03:55 -06:00
stdlib Merged in hardlulz/modem-3.0-nuttx/fix-sem-EINTR (pull request #603) 2018-02-20 18:24:53 +00:00
string Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
symtab Modules: Add mod_setsymtab to set global symtol table once. Now we can remove symbol table parameters from the insmod call. This will make implementing an NSH insmod command much easier 2015-12-13 08:10:01 -06:00
syslog 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
termios Make sure that labeling is used consistently in all function headers (part 2). 2018-02-01 12:03:55 -06:00
time Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
tls libc/Kconfig: Remove source'ing file twice. Use cleaner dependencies in some menus. 2017-02-10 10:28:34 -06:00
ubin Changes for ZDS-II ez80 build 2013-11-28 17:47:03 -06:00
uio libc: Add support for readv() and write(). Also includes some cosmetic changes to some unrelated files. 2017-11-11 11:44:14 -06:00
unistd Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
userfs Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
wchar Make sure that labeling is used consistently in all function headers (part 3). 2018-02-01 12:17:03 -06:00
wctype Remove dangling whitespace at the end of some lines. 2017-06-28 10:39:54 -06:00
wqueue Merged in hardlulz/modem-3.0-nuttx/fix-sem-EINTR (pull request #603) 2018-02-20 18:24:53 +00:00
zoneinfo Squashed commit of the following: 2017-10-30 18:07:42 -06:00
.gitignore Changes for ZDS-II ez80 build 2013-11-28 17:47:03 -06:00
Kconfig Squashed commit of the following: 2017-10-30 18:07:42 -06:00
Makefile Build system: Fix CONFIG_BUILD_KERNEL logic directories that have ubin and kbin subdirectories. Conditional logic was fine for CONFIG_BUILD_FLAT and CONFIG_BUILD_PROTECTED but generated useless dependencies if CONFIG_BUILD_KERNEL. 2017-11-15 07:54:09 -06:00
README.txt Documentation: Update NSH manual to include truncate command. 2018-01-04 12:51:01 -06:00
libc.csv pthreads: Move pthread_barrier_init, pthread_barrier_destroy, and pthread_barrier_wait from sched/pthreads to libc/pthreads. This just coordinate other OS interface calls but are not a fundamental OS interfaces and, hence, do not belong within the OS. 2017-06-14 07:59:54 -06:00
libc.h tools/mkconfig.c: Add logic to keep all of the buffering options in sync. 2017-02-09 16:17:54 -06:00
math.csv libc: Add support for readv() and write(). Also includes some cosmetic changes to some unrelated files. 2017-11-11 11:44:14 -06:00

README.txt

lib
===

This directory contains numerous, small functions typically associated with
what you would expect to find in a standard C library.  The sub-directories
in this directory contain standard interface that can be executed by user-
mode programs.

Normally, NuttX is built with no protection and all threads running in kerne-
mode.  In that model, there is no real architectural distinction between
what is a kernel-mode program and what is a user-mode program; the system is
more like on multi-threaded program that all runs in kernel-mode.

But if the CONFIG_BUILD_PROTECTED option is selected, NuttX will be built
into distinct user-mode and kernel-mode sections.  In that case, most of the
code in the nuttx/ directory will run in kernel-mode with with exceptions
of (1) the user-mode "proxies" found in syscall/proxies, and (2) the
standard C library functions found in this directory.  In this build model,
it is critical to separate the user-mode OS interfaces in this way.

If CONFIG_BUILD_KERNEL is selected, then only a NuttX kernel will be built
with no applications.

Sub-Directories
===============

The files in the libc/ directory are organized (mostly) according which file
in the include/ directory provides the prototype for library functions.  So
we have:

  audio     - This part of the audio system: nuttx/audio/audio.h
  dllfcn    - dllfcn.h
  hex2bin   - hex2bin.h
  libgen    - libgen.h
  locale    - locale.h
  fixedmath - fixedmath.h
  inttypes  - inttypes.h
  machine   - Various architecture-specific implementations.
  math      - math.h
  modlib    - Part of module and shared library logic: nuttx/lib/modlib.h
  net       - Various network-related header files: netinet/ether.h, arpa/inet.h
  pthread   - pthread.h
  queue     - queue.h
  sched     - sched.h
  semaphore - semaphore.h
  stdio     - stdio.h
  stdlib    - stdlib.h
  string    - string.h (and legacy strings.h)
  time      - time.h
  uio       - sys/uio.h
  unistd    - unistd.h
  wchar     - wchar.h
  wctype    - wctype.h

Most of these are "standard" header files; some are not: hex2bin.h and
fixemath.h are non-standard.

There is also a misc/ subdirectory that contains various internal functions
and interfaces from header files that are too few to warrant their own sub-
directory:

  misc      - Nonstandard "glue" logic, debug.h, crc32.h, dirent.h

Library Database
================

Information about functions available in the NuttX C library information is
maintained in a database.  That "database" is implemented as a simple comma-
separated-value file, libc.csv.  Most spreadsheets programs will accept this
format and can be used to maintain the library database.

This library database will (eventually) be used to generate symbol library
symbol table information that can be exported to external applications.

The format of the CSV file for each line is:

  Field 1: Function name
  Field 2: The header file that contains the function prototype
  Field 3: Condition for compilation
  Field 4: The type of function return value.
  Field 5 - N+5: The type of each of the N formal parameters of the function

Each type field has a format as follows:

  type name:
        For all simpler types
  formal type | actual type:
        For array types where the form of the formal (eg. int parm[2])
        differs from the type of actual passed parameter (eg. int*).  This
        is necessary because you cannot do simple casts to array types.
  formal type | union member actual type | union member fieldname:
        A similar situation exists for unions.  For example, the formal
        parameter type union sigval -- You cannot cast a uintptr_t to
        a union sigval, but you can cast to the type of one of the union
        member types when passing the actual parameter.  Similarly, we
        cannot cast a union sigval to a uinptr_t either.  Rather, we need
        to cast a specific union member fieldname to uintptr_t.

NOTE: The tool mksymtab can be used to generate a symbol table from this CSV
file.  See nuttx/tools/README.txt for further details about the use of mksymtab.