lkml.org 
[lkml]   [2009]   [Feb]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectFASYNC + dup + close gives the wrong fd in SIGIO
From
Hi,

There seems to be a small deficiency in the O_ASYNC/FASYNC
implementation. When the flag is set for a file descriptor, it does
not "follow" the fd across dup()+close(), so the SIGIO handler will
get the ->si_fd for the fd at the time when O_ASYNC was set. For
example:

pipe([3, 4]) = 0
getpid() = 23982
fcntl(3, F_SETOWN, 23982) = 0
fcntl(3, F_SETSIG, 0x1d) = 0
fcntl(3, F_SETFL, O_RDONLY|O_ASYNC) = 0
dup(3) = 5
close(3) = 0
write(4, "\0"..., 1) = 1
--- SIGIO (I/O possible) @ 0 (0) ---
[...]
SIGIO for fd 3

It seems a bit obscure, I agree, but I thought I'd mention it. Maybe
the fcntl() man page could carry this extra information?

Here's my test case:

#define _GNU_SOURCE

#include <fcntl.h>
#include <signal.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>

static int done = 0;

static void handle_sigio(int signo, siginfo_t *info, void *unused)
{
printf("SIGIO for fd %d\n", info->si_fd);
done = 1;
}

int main(void)
{
struct sigaction act;
int pipe_fd[2];

memset(&act, 0, sizeof(act));
act.sa_sigaction = &handle_sigio;
act.sa_flags = SA_SIGINFO;
sigaction(SIGIO, &act, NULL);

pipe(pipe_fd);
fcntl(pipe_fd[0], F_SETOWN, getpid());
fcntl(pipe_fd[0], F_SETSIG, SIGIO);
fcntl(pipe_fd[0], F_SETFL, O_ASYNC);
dup(pipe_fd[0]);
close(pipe_fd[0]);

write(pipe_fd[1], "", 1);

while (!done)
sleep(1);

return 0;
}



Vegard

--
"The animistic metaphor of the bug that maliciously sneaked in while
the programmer was not looking is intellectually dishonest as it
disguises that the error is the programmer's own creation."
-- E. W. Dijkstra, EWD1036


\
 
 \ /
  Last update: 2009-02-03 23:49    [W:0.036 / U:0.872 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site