mirror of
https://github.com/AsahiLinux/u-boot
synced 2024-11-14 17:07:38 +00:00
62584db191
This implementation uses opendir()/readdir() to access the directory information and then puts it in a linked list for the caller's use. Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Tom Rini <trini@ti.com>
364 lines
7.9 KiB
C
364 lines
7.9 KiB
C
/*
|
|
* Copyright (c) 2011 The Chromium OS Authors.
|
|
* See file CREDITS for list of people who contributed to this
|
|
* project.
|
|
*
|
|
* This program is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU General Public License as
|
|
* published by the Free Software Foundation; either version 2 of
|
|
* the License, or (at your option) any later version.
|
|
*
|
|
* This program is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License
|
|
* along with this program; if not, write to the Free Software
|
|
* Foundation, Inc., 59 Temple Place, Suite 330, Boston,
|
|
* MA 02111-1307 USA
|
|
*/
|
|
|
|
#include <dirent.h>
|
|
#include <errno.h>
|
|
#include <fcntl.h>
|
|
#include <getopt.h>
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <termios.h>
|
|
#include <time.h>
|
|
#include <unistd.h>
|
|
#include <sys/mman.h>
|
|
#include <sys/stat.h>
|
|
#include <sys/time.h>
|
|
#include <sys/types.h>
|
|
#include <linux/types.h>
|
|
|
|
#include <asm/getopt.h>
|
|
#include <asm/sections.h>
|
|
#include <asm/state.h>
|
|
#include <os.h>
|
|
|
|
/* Operating System Interface */
|
|
|
|
ssize_t os_read(int fd, void *buf, size_t count)
|
|
{
|
|
return read(fd, buf, count);
|
|
}
|
|
|
|
ssize_t os_read_no_block(int fd, void *buf, size_t count)
|
|
{
|
|
const int flags = fcntl(fd, F_GETFL, 0);
|
|
|
|
fcntl(fd, F_SETFL, flags | O_NONBLOCK);
|
|
return os_read(fd, buf, count);
|
|
}
|
|
|
|
ssize_t os_write(int fd, const void *buf, size_t count)
|
|
{
|
|
return write(fd, buf, count);
|
|
}
|
|
|
|
off_t os_lseek(int fd, off_t offset, int whence)
|
|
{
|
|
if (whence == OS_SEEK_SET)
|
|
whence = SEEK_SET;
|
|
else if (whence == OS_SEEK_CUR)
|
|
whence = SEEK_CUR;
|
|
else if (whence == OS_SEEK_END)
|
|
whence = SEEK_END;
|
|
else
|
|
os_exit(1);
|
|
return lseek(fd, offset, whence);
|
|
}
|
|
|
|
int os_open(const char *pathname, int os_flags)
|
|
{
|
|
int flags;
|
|
|
|
switch (os_flags & OS_O_MASK) {
|
|
case OS_O_RDONLY:
|
|
default:
|
|
flags = O_RDONLY;
|
|
break;
|
|
|
|
case OS_O_WRONLY:
|
|
flags = O_WRONLY;
|
|
break;
|
|
|
|
case OS_O_RDWR:
|
|
flags = O_RDWR;
|
|
break;
|
|
}
|
|
|
|
if (os_flags & OS_O_CREAT)
|
|
flags |= O_CREAT;
|
|
|
|
return open(pathname, flags, 0777);
|
|
}
|
|
|
|
int os_close(int fd)
|
|
{
|
|
return close(fd);
|
|
}
|
|
|
|
void os_exit(int exit_code)
|
|
{
|
|
exit(exit_code);
|
|
}
|
|
|
|
/* Restore tty state when we exit */
|
|
static struct termios orig_term;
|
|
|
|
static void os_fd_restore(void)
|
|
{
|
|
tcsetattr(0, TCSANOW, &orig_term);
|
|
}
|
|
|
|
/* Put tty into raw mode so <tab> and <ctrl+c> work */
|
|
void os_tty_raw(int fd)
|
|
{
|
|
static int setup = 0;
|
|
struct termios term;
|
|
|
|
if (setup)
|
|
return;
|
|
setup = 1;
|
|
|
|
/* If not a tty, don't complain */
|
|
if (tcgetattr(fd, &orig_term))
|
|
return;
|
|
|
|
term = orig_term;
|
|
term.c_iflag = IGNBRK | IGNPAR;
|
|
term.c_oflag = OPOST | ONLCR;
|
|
term.c_cflag = CS8 | CREAD | CLOCAL;
|
|
term.c_lflag = 0;
|
|
if (tcsetattr(fd, TCSANOW, &term))
|
|
return;
|
|
|
|
atexit(os_fd_restore);
|
|
}
|
|
|
|
void *os_malloc(size_t length)
|
|
{
|
|
return mmap(NULL, length, PROT_READ | PROT_WRITE,
|
|
MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
|
|
}
|
|
|
|
void os_usleep(unsigned long usec)
|
|
{
|
|
usleep(usec);
|
|
}
|
|
|
|
u64 os_get_nsec(void)
|
|
{
|
|
#if defined(CLOCK_MONOTONIC) && defined(_POSIX_MONOTONIC_CLOCK)
|
|
struct timespec tp;
|
|
if (EINVAL == clock_gettime(CLOCK_MONOTONIC, &tp)) {
|
|
struct timeval tv;
|
|
|
|
gettimeofday(&tv, NULL);
|
|
tp.tv_sec = tv.tv_sec;
|
|
tp.tv_nsec = tv.tv_usec * 1000;
|
|
}
|
|
return tp.tv_sec * 1000000000ULL + tp.tv_nsec;
|
|
#else
|
|
struct timeval tv;
|
|
gettimeofday(&tv, NULL);
|
|
return tv.tv_sec * 1000000000ULL + tv.tv_usec * 1000;
|
|
#endif
|
|
}
|
|
|
|
static char *short_opts;
|
|
static struct option *long_opts;
|
|
|
|
int os_parse_args(struct sandbox_state *state, int argc, char *argv[])
|
|
{
|
|
struct sb_cmdline_option **sb_opt = __u_boot_sandbox_option_start;
|
|
size_t num_options = __u_boot_sandbox_option_count();
|
|
size_t i;
|
|
|
|
int hidden_short_opt;
|
|
size_t si;
|
|
|
|
int c;
|
|
|
|
if (short_opts || long_opts)
|
|
return 1;
|
|
|
|
state->argc = argc;
|
|
state->argv = argv;
|
|
|
|
/* dynamically construct the arguments to the system getopt_long */
|
|
short_opts = os_malloc(sizeof(*short_opts) * num_options * 2 + 1);
|
|
long_opts = os_malloc(sizeof(*long_opts) * num_options);
|
|
if (!short_opts || !long_opts)
|
|
return 1;
|
|
|
|
/*
|
|
* getopt_long requires "val" to be unique (since that is what the
|
|
* func returns), so generate unique values automatically for flags
|
|
* that don't have a short option. pick 0x100 as that is above the
|
|
* single byte range (where ASCII/ISO-XXXX-X charsets live).
|
|
*/
|
|
hidden_short_opt = 0x100;
|
|
si = 0;
|
|
for (i = 0; i < num_options; ++i) {
|
|
long_opts[i].name = sb_opt[i]->flag;
|
|
long_opts[i].has_arg = sb_opt[i]->has_arg ?
|
|
required_argument : no_argument;
|
|
long_opts[i].flag = NULL;
|
|
|
|
if (sb_opt[i]->flag_short) {
|
|
short_opts[si++] = long_opts[i].val = sb_opt[i]->flag_short;
|
|
if (long_opts[i].has_arg == required_argument)
|
|
short_opts[si++] = ':';
|
|
} else
|
|
long_opts[i].val = sb_opt[i]->flag_short = hidden_short_opt++;
|
|
}
|
|
short_opts[si] = '\0';
|
|
|
|
/* we need to handle output ourselves since u-boot provides printf */
|
|
opterr = 0;
|
|
|
|
/*
|
|
* walk all of the options the user gave us on the command line,
|
|
* figure out what u-boot option structure they belong to (via
|
|
* the unique short val key), and call the appropriate callback.
|
|
*/
|
|
while ((c = getopt_long(argc, argv, short_opts, long_opts, NULL)) != -1) {
|
|
for (i = 0; i < num_options; ++i) {
|
|
if (sb_opt[i]->flag_short == c) {
|
|
if (sb_opt[i]->callback(state, optarg)) {
|
|
state->parse_err = sb_opt[i]->flag;
|
|
return 0;
|
|
}
|
|
break;
|
|
}
|
|
}
|
|
if (i == num_options) {
|
|
/*
|
|
* store the faulting flag for later display. we have to
|
|
* store the flag itself as the getopt parsing itself is
|
|
* tricky: need to handle the following flags (assume all
|
|
* of the below are unknown):
|
|
* -a optopt='a' optind=<next>
|
|
* -abbbb optopt='a' optind=<this>
|
|
* -aaaaa optopt='a' optind=<this>
|
|
* --a optopt=0 optind=<this>
|
|
* as you can see, it is impossible to determine the exact
|
|
* faulting flag without doing the parsing ourselves, so
|
|
* we just report the specific flag that failed.
|
|
*/
|
|
if (optopt) {
|
|
static char parse_err[3] = { '-', 0, '\0', };
|
|
parse_err[1] = optopt;
|
|
state->parse_err = parse_err;
|
|
} else
|
|
state->parse_err = argv[optind - 1];
|
|
break;
|
|
}
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
void os_dirent_free(struct os_dirent_node *node)
|
|
{
|
|
struct os_dirent_node *next;
|
|
|
|
while (node) {
|
|
next = node->next;
|
|
free(node);
|
|
node = next;
|
|
}
|
|
}
|
|
|
|
int os_dirent_ls(const char *dirname, struct os_dirent_node **headp)
|
|
{
|
|
struct dirent entry, *result;
|
|
struct os_dirent_node *head, *node, *next;
|
|
struct stat buf;
|
|
DIR *dir;
|
|
int ret;
|
|
char *fname;
|
|
int len;
|
|
|
|
*headp = NULL;
|
|
dir = opendir(dirname);
|
|
if (!dir)
|
|
return -1;
|
|
|
|
/* Create a buffer for the maximum filename length */
|
|
len = sizeof(entry.d_name) + strlen(dirname) + 2;
|
|
fname = malloc(len);
|
|
if (!fname) {
|
|
ret = -ENOMEM;
|
|
goto done;
|
|
}
|
|
|
|
for (node = head = NULL;; node = next) {
|
|
ret = readdir_r(dir, &entry, &result);
|
|
if (ret || !result)
|
|
break;
|
|
next = malloc(sizeof(*node) + strlen(entry.d_name) + 1);
|
|
if (!next) {
|
|
os_dirent_free(head);
|
|
ret = -ENOMEM;
|
|
goto done;
|
|
}
|
|
strcpy(next->name, entry.d_name);
|
|
switch (entry.d_type) {
|
|
case DT_REG:
|
|
next->type = OS_FILET_REG;
|
|
break;
|
|
case DT_DIR:
|
|
next->type = OS_FILET_DIR;
|
|
break;
|
|
case DT_LNK:
|
|
next->type = OS_FILET_LNK;
|
|
break;
|
|
}
|
|
next->size = 0;
|
|
snprintf(fname, len, "%s/%s", dirname, next->name);
|
|
if (!stat(fname, &buf))
|
|
next->size = buf.st_size;
|
|
if (node)
|
|
node->next = next;
|
|
if (!head)
|
|
head = node;
|
|
}
|
|
*headp = head;
|
|
|
|
done:
|
|
closedir(dir);
|
|
return ret;
|
|
}
|
|
|
|
const char *os_dirent_typename[OS_FILET_COUNT] = {
|
|
" ",
|
|
"SYM",
|
|
"DIR",
|
|
"???",
|
|
};
|
|
|
|
const char *os_dirent_get_typename(enum os_dirent_t type)
|
|
{
|
|
if (type >= 0 && type < OS_FILET_COUNT)
|
|
return os_dirent_typename[type];
|
|
|
|
return os_dirent_typename[OS_FILET_UNKNOWN];
|
|
}
|
|
|
|
ssize_t os_get_filesize(const char *fname)
|
|
{
|
|
struct stat buf;
|
|
int ret;
|
|
|
|
ret = stat(fname, &buf);
|
|
if (ret)
|
|
return ret;
|
|
return buf.st_size;
|
|
}
|