[patch] epoll_pwait ...

[Davide Libenzi <davidel@xmailserver.org>]

The attached patch implements the epoll_pwait system call, that extend the 
event wait mechanism with the same logic ppoll and pselect do. The 
definition of epoll_pwait is:

int epoll_pwait(int epfd, struct epoll_event *events, int maxevents,
                int timeout, const sigset_t *sigmask, size_t sigsetsize);

The difference between the vanilla epoll_wait and epoll_pwait is that the 
latter allows the caller to specify a signal mask to be set while waiting 
for events. Hence epoll_pwait will wait until either one monitored event, 
or an unmasked signal happen. If sigmask is NULL, the epoll_pwait system 
call will act exactly like epoll_wait. For the POSIX definition of 
pselect, information is available here:

http://www.opengroup.org/onlinepubs/009695399/functions/select.html

This patch goes over 2.6.15-mm4 and the epoll_pwait definition depends on 
the TIF_RESTORE_SIGMASK bits.

ChangeLog:

o Make it conditional to TIF_RESTORE_SIGMASK definition to avoid breaking
  archs that do not still support it

o Changed the name to epoll_pwait

o Changed the order of "sigmask" and "error" checks in the epoll_pwait
  exit path



Singed-off-by: Davide Libenzi <davidel@xmailserver.org>



- Davide



arch/i386/kernel/syscall_table.S |    1
fs/eventpoll.c                   |   55 ++++++++++++++++++++++++++++++++++++---
include/asm-i386/unistd.h        |    3 +-
include/linux/syscalls.h         |    3 ++
4 files changed, 58 insertions(+), 4 deletions(-)

diff -Nru linux-2.6.15/arch/i386/kernel/syscall_table.S linux-2.6.15.mod/arch/i386/kernel/syscall_table.S
--- linux-2.6.15/arch/i386/kernel/syscall_table.S	2006-01-18 11:05:35.000000000 -0800
+++ linux-2.6.15.mod/arch/i386/kernel/syscall_table.S	2006-01-18 11:06:42.000000000 -0800
@@ -310,3 +310,4 @@
 	.long sys_pselect6
 	.long sys_ppoll
 	.long sys_unshare		/* 310 */
+	.long sys_epoll_pwait
diff -Nru linux-2.6.15/fs/eventpoll.c linux-2.6.15.mod/fs/eventpoll.c
--- linux-2.6.15/fs/eventpoll.c	2006-01-18 11:05:37.000000000 -0800
+++ linux-2.6.15.mod/fs/eventpoll.c	2006-01-18 11:31:12.000000000 -0800
@@ -105,6 +105,8 @@
 /* Maximum msec timeout value storeable in a long int */
 #define EP_MAX_MSTIMEO min(1000ULL * MAX_SCHEDULE_TIMEOUT / HZ, (LONG_MAX - 999ULL) / HZ)
 
+#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
+
 
 struct epoll_filefd {
 	struct file *file;
@@ -506,7 +508,7 @@
  */
 asmlinkage long sys_epoll_create(int size)
 {
-	int error, fd;
+	int error, fd = -1;
 	struct eventpoll *ep;
 	struct inode *inode;
 	struct file *file;
@@ -649,7 +651,6 @@
 	return error;
 }
 
-#define MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
 
 /*
  * Implement the event wait interface for the eventpoll file. It is the kernel
@@ -666,7 +667,7 @@
 		     current, epfd, events, maxevents, timeout));
 
 	/* The maximum number of event must be greater than zero */
-	if (maxevents <= 0 || maxevents > MAX_EVENTS)
+	if (maxevents <= 0 || maxevents > EP_MAX_EVENTS)
 		return -EINVAL;
 
 	/* Verify that the area passed by the user is writeable */
@@ -708,6 +709,54 @@
 }
 
 
+#ifdef TIF_RESTORE_SIGMASK
+
+/*
+ * Implement the event wait interface for the eventpoll file. It is the kernel
+ * part of the user space epoll_pwait(2).
+ */
+asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events,
+				int maxevents, int timeout, const sigset_t __user *sigmask,
+				size_t sigsetsize)
+{
+	int error;
+	sigset_t ksigmask, sigsaved;
+
+	/*
+	 * If the caller wants a certain signal mask to be set during the wait,
+	 * we apply it here.
+	 */
+	if (sigmask) {
+		if (sigsetsize != sizeof(sigset_t))
+			return -EINVAL;
+		if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask)))
+			return -EFAULT;
+		sigdelsetmask(&ksigmask, sigmask(SIGKILL) | sigmask(SIGSTOP));
+		sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved);
+	}
+
+	error = sys_epoll_wait(epfd, events, maxevents, timeout);
+
+	/*
+	 * If we changed the signal mask, we need to restore the original one.
+	 * In case we've got a signal while waiting, we do not restore the signal
+	 * mask yet, and we allow do_signal() to deliver the signal on the way back
+	 * to userspace, before the signal mask is restored.
+	 */
+	if (sigmask) {
+		if (error == -EINTR) {
+			memcpy(&current->saved_sigmask, &sigsaved, sizeof(sigsaved));
+			set_thread_flag(TIF_RESTORE_SIGMASK);			
+		} else
+			sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+	}
+
+	return error;
+}
+
+#endif /* #ifdef TIF_RESTORE_SIGMASK */
+
+
 /*
  * Creates the file descriptor to be used by the epoll interface.
  */
diff -Nru linux-2.6.15/include/asm-i386/unistd.h linux-2.6.15.mod/include/asm-i386/unistd.h
--- linux-2.6.15/include/asm-i386/unistd.h	2006-01-18 11:05:38.000000000 -0800
+++ linux-2.6.15.mod/include/asm-i386/unistd.h	2006-01-18 11:06:56.000000000 -0800
@@ -316,8 +316,9 @@
 #define __NR_pselect6		308
 #define __NR_ppoll		309
 #define __NR_unshare		310
+#define __NR_epoll_pwait	311
 
-#define NR_syscalls 311
+#define NR_syscalls 312
 
 /*
  * user-visible error numbers are in the range -1 - -128: see
diff -Nru linux-2.6.15/include/linux/syscalls.h linux-2.6.15.mod/include/linux/syscalls.h
--- linux-2.6.15/include/linux/syscalls.h	2006-01-18 11:05:38.000000000 -0800
+++ linux-2.6.15.mod/include/linux/syscalls.h	2006-01-18 11:07:11.000000000 -0800
@@ -428,6 +428,9 @@
 				struct epoll_event __user *event);
 asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events,
 				int maxevents, int timeout);
+asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events,
+				int maxevents, int timeout, const sigset_t __user *sigmask,
+				size_t sigsetsize);
 asmlinkage long sys_gethostname(char __user *name, int len);
 asmlinkage long sys_sethostname(char __user *name, int len);
 asmlinkage long sys_setdomainname(char __user *name, int len);