From: Andrew Vagin <avagin@parallels.com>
To: Oleg Nesterov <oleg@redhat.com>
Cc: Pavel Emelyanov <xemul@parallels.com>,
David Howells <dhowells@redhat.com>,
<linux-kernel@vger.kernel.org>, <criu@openvz.org>,
Cyrill Gorcunov <gorcunov@openvz.org>,
Andrey Wagin <avagin@gmail.com>,
Alexander Viro <viro@zeniv.linux.org.uk>,
<linux-fsdevel@vger.kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
Michael Kerrisk <mtk.manpages@gmail.com>
Subject: Re: [CRIU] [PATCH 1/4] signalfd: add ability to return siginfo in a raw format
Date: Wed, 26 Dec 2012 18:47:51 +0400 [thread overview]
Message-ID: <20121226144751.GA18767@paralelels.com> (raw)
In-Reply-To: <20121225165803.GA9066@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 1129 bytes --]
On Tue, Dec 25, 2012 at 05:58:03PM +0100, Oleg Nesterov wrote:
> On 12/25, Pavel Emelyanov wrote:
> >
> > On 12/25/2012 07:27 PM, Oleg Nesterov wrote:
> > >
> > > I guess that probably you actually need DUMP, not DEQUEUE. but the
> > > latter is not trivial. However, perhaps we can do this assuming that
> > > all other threads are sleeping and nobody can do dequeue_signal().
> > > Say, we can play with ppos/llseek. If *ppos is not zero,
> > > signalfd_dequeue() could dump the nth entry from list or return 0.
> >
> > This would be perfect, but isn't it better to preserve the pos
> > semantics -- we do know size of entry we're about to copy, we can
> > treat pos as offset in bytes, not in elements.
>
> nr-of-records looks better (more flexible) than nr-of-bytes to me. And
> perhaps we can also encode private-or-shared into ppos. But I will not
> argue in any case.
Oleg and Pavel, could you look at these two patches. I implemented in them,
what you described here.
>
> Oleg.
>
> _______________________________________________
> CRIU mailing list
> CRIU@openvz.org
> http://lists.openvz.org/mailman/listinfo/criu
[-- Attachment #2: 0001-signal-add-helper-to-get-siginfo-without-removing-fr.patch --]
[-- Type: text/plain, Size: 1944 bytes --]
>From 2b8b475b39f41ca65e623905b2f7d2d9348bfa84 Mon Sep 17 00:00:00 2001
From: Andrey Vagin <avagin@openvz.org>
Date: Thu, 29 Nov 2012 20:51:49 +0400
Subject: [PATCH 1/2] signal: add helper to get siginfo without removing from
the queue
Signed-off-by: Andrey Vagin <avagin@openvz.org>
---
include/linux/sched.h | 2 ++
kernel/signal.c | 28 ++++++++++++++++++++++++++++
2 files changed, 30 insertions(+)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 206bb08..a907854 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2222,6 +2222,8 @@ extern void __flush_signals(struct task_struct *);
extern void ignore_signals(struct task_struct *);
extern void flush_signal_handlers(struct task_struct *, int force_default);
extern int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info);
+extern int peek_signal(struct task_struct *tsk, sigset_t *mask,
+ siginfo_t *info, int offset, bool group);
static inline int dequeue_signal_lock(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
{
diff --git a/kernel/signal.c b/kernel/signal.c
index ac5f5e7..aa71213 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -536,6 +536,34 @@ unblock_all_signals(void)
spin_unlock_irqrestore(¤t->sighand->siglock, flags);
}
+int peek_signal(struct task_struct *tsk, sigset_t *mask,
+ siginfo_t *info, int offset, bool group)
+{
+ struct sigpending *pending;
+ struct sigqueue *q;
+ int i = 0, ret = 0;
+
+ if (group)
+ pending = &tsk->signal->shared_pending;
+ else
+ pending = &tsk->pending;
+
+ list_for_each_entry(q, &pending->list, list) {
+ if (sigismember(mask, q->info.si_signo))
+ continue;
+
+ if (i == offset) {
+ copy_siginfo(info, &q->info);
+ ret = info->si_signo;
+ break;
+ }
+
+ i++;
+ }
+
+ return ret;
+}
+
static void collect_signal(int sig, struct sigpending *list, siginfo_t *info)
{
struct sigqueue *q, *first = NULL;
--
1.7.11.7
[-- Attachment #3: 0002-signalfd-add-ability-to-get-signal-without-removing-.patch --]
[-- Type: text/plain, Size: 3410 bytes --]
>From 6bf48dd6a77261af0daaa9be19cd13a9a11a008d Mon Sep 17 00:00:00 2001
From: Andrey Vagin <avagin@openvz.org>
Date: Wed, 26 Dec 2012 13:45:54 +0400
Subject: [PATCH 2/2] signalfd: add ability to get signal without removing
from the queue
lseek sets a sequence number of signal in a queue, then read() returns
siginfo if a signal is exists, otherwise it returns 0. All signals
remain in a queue.
If lseek sets a positive position, signals are taken from a shared queue.
If lseek sets a negative position, signals are taken from a private queue.
If ppos is zero (default), signalfd dequeues signals.
Signed-off-by: Andrey Vagin <avagin@openvz.org>
---
fs/signalfd.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++--
1 file changed, 46 insertions(+), 2 deletions(-)
diff --git a/fs/signalfd.c b/fs/signalfd.c
index ee60d5f..0ce6fb3 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -48,8 +48,21 @@ void signalfd_cleanup(struct sighand_struct *sighand)
struct signalfd_ctx {
sigset_t sigmask;
+ loff_t peek_offset;
};
+static ssize_t signalfd_peek(struct signalfd_ctx *ctx, siginfo_t *info)
+{
+ int ret;
+
+ spin_lock_irq(¤t->sighand->siglock);
+ ret = peek_signal(current, &ctx->sigmask, info,
+ abs(ctx->peek_offset) - 1, ctx->peek_offset > 0);
+ spin_unlock_irq(¤t->sighand->siglock);
+
+ return ret;
+}
+
static int signalfd_release(struct inode *inode, struct file *file)
{
kfree(file->private_data);
@@ -230,7 +243,11 @@ static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count,
siginfo = (struct signalfd_siginfo __user *) buf;
do {
- ret = signalfd_dequeue(ctx, &info, nonblock);
+ if (ctx->peek_offset == 0)
+ ret = signalfd_dequeue(ctx, &info, nonblock);
+ else
+ ret = signalfd_peek(ctx, &info);
+
if (unlikely(ret <= 0))
break;
@@ -242,6 +259,13 @@ static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count,
if (ret < 0)
break;
+ if (ctx->peek_offset) {
+ if (ctx->peek_offset > 0)
+ ctx->peek_offset++;
+ else
+ ctx->peek_offset--;
+ }
+
siginfo++;
total += ret;
nonblock = 1;
@@ -264,6 +288,24 @@ static int signalfd_show_fdinfo(struct seq_file *m, struct file *f)
}
#endif
+loff_t signalfd_llseek(struct file *f, loff_t offset, int whence)
+{
+ struct signalfd_ctx *ctx = f->private_data;
+
+ switch (whence) {
+ case SEEK_SET:
+ ctx->peek_offset = offset;
+ break;
+ case SEEK_CUR:
+ ctx->peek_offset += offset;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return ctx->peek_offset;
+}
+
static const struct file_operations signalfd_fops = {
#ifdef CONFIG_PROC_FS
.show_fdinfo = signalfd_show_fdinfo,
@@ -271,7 +313,7 @@ static const struct file_operations signalfd_fops = {
.release = signalfd_release,
.poll = signalfd_poll,
.read = signalfd_read,
- .llseek = noop_llseek,
+ .llseek = signalfd_llseek,
};
SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask,
@@ -300,6 +342,7 @@ SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask,
return -ENOMEM;
ctx->sigmask = sigmask;
+ ctx->peek_offset = 0;
ufd = get_unused_fd_flags(flags);
if (ufd < 0) {
@@ -321,6 +364,7 @@ SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask,
}
file->f_flags |= flags & SFD_RAW;
+ file->f_mode |= FMODE_LSEEK;
fd_install(ufd, file);
} else {
--
1.7.11.7
next prev parent reply other threads:[~2012-12-26 14:48 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-24 8:13 [PATCH 0/4] signalfd: a kernel interface for dumping/restoring pending signals Andrey Vagin
2012-12-24 8:13 ` [PATCH 1/4] signalfd: add ability to return siginfo in a raw format Andrey Vagin
2012-12-24 16:53 ` Oleg Nesterov
2012-12-25 8:29 ` Andrey Wagin
2012-12-25 14:30 ` Oleg Nesterov
2012-12-25 15:27 ` Oleg Nesterov
2012-12-25 15:40 ` Pavel Emelyanov
2012-12-25 16:58 ` Oleg Nesterov
2012-12-26 14:47 ` Andrew Vagin [this message]
2012-12-26 16:31 ` [CRIU] " Oleg Nesterov
2012-12-27 14:36 ` Andrey Wagin
2012-12-27 15:30 ` Oleg Nesterov
2012-12-27 18:40 ` Andrey Wagin
2012-12-28 14:12 ` Oleg Nesterov
2012-12-28 14:28 ` Andrey Wagin
2012-12-28 14:46 ` Oleg Nesterov
2012-12-28 14:48 ` Andrey Wagin
2012-12-28 14:56 ` Oleg Nesterov
2012-12-24 8:13 ` [PATCH 2/4] signal: add a helper for dequeuing signals from a specified queue Andrey Vagin
2012-12-24 20:52 ` Michael Kerrisk
2012-12-24 8:13 ` [PATCH 3/4] signalfd: add ability to choose a private or shared queue Andrey Vagin
2012-12-24 17:05 ` Oleg Nesterov
2012-12-24 20:53 ` Michael Kerrisk
2012-12-24 8:13 ` [PATCH 4/4] signal: allow to send any siginfo to itself Andrey Vagin
2012-12-24 20:51 ` [PATCH 0/4] signalfd: a kernel interface for dumping/restoring pending signals Michael Kerrisk
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=20121226144751.GA18767@paralelels.com \
--to=avagin@parallels.com \
--cc=avagin@gmail.com \
--cc=criu@openvz.org \
--cc=dhowells@redhat.com \
--cc=gorcunov@openvz.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mtk.manpages@gmail.com \
--cc=oleg@redhat.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=tglx@linutronix.de \
--cc=viro@zeniv.linux.org.uk \
--cc=xemul@parallels.com \
/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).