mirrors/cpython
0
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2024-11-25 10:54:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

bugfix for issue 8052 fixes on *BSD platforms.

Browse Source
This commit is contained in:
Gregory P. Smith 2012-01-21 15:20:08 -08:00
commit 61f0da8c02
1 changed files with 13 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
Modules/_posixsubprocess.c
View File

@ -15,12 +15,17 @@
#include <dirent.h>
#endif
#if defined(sun) && !defined(HAVE_DIRFD)
#if defined(sun)
/* readdir64 is used to work around Solaris 9 bug 6395699. */
# define readdir readdir64
# define dirent dirent64
# if !defined(HAVE_DIRFD)
/* Some versions of Solaris lack dirfd(). */
# define DIRFD(dirp) ((dirp)->dd_fd)
# define HAVE_DIRFD
#else
# define DIRFD(dirp) (dirfd(dirp))
# define DIRFD(dirp) ((dirp)->dd_fd)
# define HAVE_DIRFD
# else
# define DIRFD(dirp) (dirfd(dirp))
# endif
#endif
#define LINUX_SOLARIS_FD_DIR "/proc/self/fd"
@ -206,7 +211,7 @@ static void _close_open_fd_range_safe(int start_fd, int end_fd,
* exclusive. Do not close any in the sorted py_fds_to_keep list.
*
* This function violates the strict use of async signal safe functions. :(
* It calls opendir(), readdir64() and closedir(). Of these, the one most
* It calls opendir(), readdir() and closedir(). Of these, the one most
* likely to ever cause a problem is opendir() as it performs an internal
* malloc(). Practically this should not be a problem. The Java VM makes the
* same calls between fork and exec in its own UNIXProcess_md.c implementation.
@ -243,15 +248,14 @@ static void _close_open_fd_range_maybe_unsafe(int start_fd, int end_fd,
/* No way to get a list of open fds. */
_close_fds_by_brute_force(start_fd, end_fd, py_fds_to_keep);
} else {
struct dirent64 *dir_entry;
struct dirent *dir_entry;
#ifdef HAVE_DIRFD
int fd_used_by_opendir = DIRFD(proc_fd_dir);
#else
int fd_used_by_opendir = start_fd - 1;
#endif
errno = 0;
/* readdir64 is used to work around Solaris 9 bug 6395699. */
while ((dir_entry = readdir64(proc_fd_dir))) {
while ((dir_entry = readdir(proc_fd_dir))) {
int fd;
if ((fd = _pos_int_from_ascii(dir_entry->d_name)) < 0)
continue; /* Not a number. */