linux-kernel - [PATCH] Input: serio_raw - ensure we don't block in non-blocking read

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 3 Apr 2012 13:24:33 -0700
From:	Dmitry Torokhov <dmitry.torokhov@...il.com>
To:	linux-input@...r.kernel.org
Cc:	Wanlong Gao <gaowanlong@...fujitsu.com>,
	Che-Liang Chiou <clchiou@...omium.org>,
	David Herrmann <dh.herrmann@...glemail.com>,
	linux-kernel@...r.kernel.org
Subject: [PATCH] Input: serio_raw - ensure we don't block in non-blocking read

Avoid calling wait_event_interruptible() if client requested non-blocking
read, since it is not guaranteed that another thread will not consume
event after we checked if serio_raw->head != serio_raw->tail.

Also ensure we do not return 0 but keep waiting instead in blocking case,
when another thread steals "our" byte.

Signed-off-by: Dmitry Torokhov <dtor@...l.ru>
---
 drivers/input/serio/serio_raw.c |   43 ++++++++++++++++++++++----------------
 1 files changed, 25 insertions(+), 18 deletions(-)

diff --git a/drivers/input/serio/serio_raw.c b/drivers/input/serio/serio_raw.c
index 4494233..094d6fb 100644
--- a/drivers/input/serio/serio_raw.c
+++ b/drivers/input/serio/serio_raw.c
@@ -165,31 +165,38 @@ static ssize_t serio_raw_read(struct file *file, char __user *buffer,
 	struct serio_raw *serio_raw = client->serio_raw;
 	char uninitialized_var(c);
 	ssize_t read = 0;
-	int retval;
+	int error = 0;
 
-	if (serio_raw->dead)
-		return -ENODEV;
+	do {
+		if (serio_raw->dead)
+			return -ENODEV;
 
-	if (serio_raw->head == serio_raw->tail && (file->f_flags & O_NONBLOCK))
-		return -EAGAIN;
+		if (serio_raw->head == serio_raw->tail &&
+		    (file->f_flags & O_NONBLOCK))
+			return -EAGAIN;
 
-	retval = wait_event_interruptible(serio_raw->wait,
-			serio_raw->head != serio_raw->tail || serio_raw->dead);
-	if (retval)
-		return retval;
+		if (count == 0)
+			break;
 
-	if (serio_raw->dead)
-		return -ENODEV;
+		while (read < count && serio_raw_fetch_byte(serio_raw, &c)) {
+			if (put_user(c, buffer++)) {
+				error = -EFAULT;
+				goto out;
+			}
+			read++;
+		}
 
-	while (read < count && serio_raw_fetch_byte(serio_raw, &c)) {
-		if (put_user(c, buffer++)) {
-			retval = -EFAULT;
+		if (read)
 			break;
-		}
-		read++;
-	}
 
-	return read ?: retval;
+		if (!(file->f_flags & O_NONBLOCK))
+			error = wait_event_interruptible(serio_raw->wait,
+					serio_raw->head != serio_raw->tail ||
+					serio_raw->dead);
+	} while (!error);
+
+out:
+	return read ?: error;
 }
 
 static ssize_t serio_raw_write(struct file *file, const char __user *buffer,
-- 
1.7.7.6


-- 
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/