From: Jia He <jiakernel@gmail.com>
To: linux-kernel@vger.kernel.org
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>,
Andrew Morton <akpm@linux-foundation.org>,
Rik van Riel <riel@redhat.com>,
Manfred Spraul <manfred@colorfullife.com>,
Al Viro <viro@zeniv.linux.org.uk>, Jia He <jiakernel@gmail.com>
Subject: [PATCH] ipc/sem.c: fix update sem_otime when calling sem_op in semaphore initialization
Date: Sun, 22 Sep 2013 10:11:24 +0800 [thread overview]
Message-ID: <1379815884-11035-1-git-send-email-jiakernel@gmail.com> (raw)
In commit 0a2b9d4c,the update of semaphore's sem_otime(last semop time)
was removed because he wanted to move setting sem->sem_otime to one
place. But after that, the initial semop() will not set the otime
because its sem_op value is 0(in semtimedop,will not change
otime if alter == 1).
the error case:
process_a(server) process_b(client)
semget()
semctl(SETVAL)
semop()
semget()
setctl(IP_STAT)
for(;;) { <--not successful here
check until sem_otime > 0
}
provide test codes here:
$cat server.c
int semid;
int main()
{
int key;
struct semid64_ds sem_info;
union semun arg;
struct sembuf sop;
key = ftok(SEM_PATH,'a');
semid = semget(key,1,IPC_CREAT|IPC_EXCL|00666);
if(semid < 0)
perror("server:semget");
arg.val = 0;
if(semctl(semid,0,SETVAL,arg) == -1)
perror("semctl setval error");
sop.sem_num = 0;
sop.sem_op = 0;
sop.sem_flg = 0;
if (semop(semid, &sop, 1) == -1)
perror("semop error");
sleep(30);
if(semctl(semid, 0, IPC_RMID) == -1)
perror("semctl IPC_RMID");
else printf("remove sem ok\n");
}
$cat client.c
int semid;
int main()
{
int i;
int key;
union semun arg;
struct semid64_ds sem_info;
key = ftok(SEM_PATH,'a');
semid = semget(key,1,IPC_CREAT|00666);
if(semid < 0)
perror("client:semget");
for (i = 0; i < MAX_TRIES; i++)
{
arg.buf = &sem_info;
semctl(semid, 0, IPC_STAT, arg);
if (sem_info.sem_otime != 0) break;
sleep(1);
}
if(MAX_TRIES == i)
printf("error in opening a existed sem\n");
else
printf("open exsited sem sucessfully\n");
return 0;
}
the steps to test:
touch /tmp/my_sem
./server &
sleep 1
./client &
With the patch
1.test output:
error in opening a existed sem
2.cat /proc/sysvipc/sem
the field sem_otime is always zero
Without this patch
1.test output:
open exsited sem sucessfully
2.cat /proc/sysvipc/sem
the field sem_otime is not zero
Signed-off-by: Jia He <jiakernel@gmail.com>
---
ipc/sem.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/ipc/sem.c b/ipc/sem.c
index 69b6a21..8e01e76 100644
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -590,6 +590,7 @@ static int perform_atomic_semop(struct sem_array *sma, struct sembuf *sops,
sop--;
}
+ sma->sem_base[sops[0].sem_num].sem_otime = get_seconds();
return 0;
out_of_range:
--
1.8.1.2
next reply other threads:[~2013-09-22 2:12 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-22 2:11 Jia He [this message]
2013-09-22 8:17 ` [PATCH] ipc/sem.c: fix update sem_otime when calling sem_op in semaphore initialization Mike Galbraith
2013-09-22 8:26 ` Mike Galbraith
2013-09-22 9:34 ` Jia He
2013-09-22 10:00 ` Mike Galbraith
2013-09-22 12:44 ` Jia He
2013-09-22 10:42 ` Manfred Spraul
2013-09-22 12:53 ` Jia He
2013-09-22 15:14 ` Jia He
2013-09-24 21:09 ` Manfred Spraul
2013-09-25 3:05 ` Jia He
2013-09-25 6:55 ` Manfred Spraul
2013-09-25 7:49 ` Jia He
2013-09-23 1:08 ` Mike Galbraith
2013-09-23 2:24 ` Jia He
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1379815884-11035-1-git-send-email-jiakernel@gmail.com \
--to=jiakernel@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=davidlohr.bueso@hp.com \
--cc=linux-kernel@vger.kernel.org \
--cc=manfred@colorfullife.com \
--cc=riel@redhat.com \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).