tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

comments

Browse Source
This commit is contained in:
Jean-Francois Dockes 2015-07-17 11:00:42 +02:00
parent 119773358b
commit c8e9795f9e
1 changed files with 12 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
src/utils/closefrom.cpp
View File

@ -25,8 +25,8 @@
* libraries. * libraries.
* *
* There are many reasons for closing file descriptors before * There are many reasons for closing file descriptors before
* an exec (security, pipe control, the possibility that a bug will trigger * an exec (security, pipe control, the possibility that a bug will
* an unwanted write, etc.) * trigger an unwanted write, etc.)
* *
* A process has currently no POSIX way to determine the set of open file * A process has currently no POSIX way to determine the set of open file
* descriptors or at least the highest value. Closing all files (except a few), * descriptors or at least the highest value. Closing all files (except a few),
@ -54,9 +54,10 @@
* - Solaris 10+ has closefrom, and can specify closefrom to posix_spawn() * - Solaris 10+ has closefrom, and can specify closefrom to posix_spawn()
* *
* Linux: * Linux:
* - Has nothing. The method we used (opening /dev/fd) was very * - Has nothing. The method we initially used (listing /dev/fd) could
* unsafe in multithread fork/exec context. We now use a close() * deadlock in multithread fork/exec context. We now use a close()
* loop. glibc maintainers think that closefrom() is a bad idea * loop but there is no completely reliable way to determine the high limit.
* glibc maintainers think that closefrom() is a bad idea
* *especially* because it is implemented on *BSD and Solaris. Go * *especially* because it is implemented on *BSD and Solaris. Go
* figure...: https://sourceware.org/bugzilla/show_bug.cgi?id=10353 * figure...: https://sourceware.org/bugzilla/show_bug.cgi?id=10353
* *
@ -206,6 +207,12 @@ int libclf_closefrom(int fd0)
} }
#endif #endif
// Note that this will not work if the limit was lowered after a
// higher fd was opened. But we don't call setrlimit() inside recoll
// code, so we should be ok. It seems that sysconf(_SC_OPEN_MAX)
// usually reports the soft limit, so it's redundant, but it could be
// useful in case getrlimit() is not implemented (unlikely as they're
// both POSIX.1-2001?
int libclf_maxfd(int) int libclf_maxfd(int)
{ {
struct rlimit lim; struct rlimit lim;