linux-kernel - [PATCH v2] oprofile: Fix uninitialized memory access when writing to

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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20111219153830.GH16765@erda.amd.com>
Date:	Mon, 19 Dec 2011 16:38:30 +0100
From:	Robert Richter <robert.richter@....com>
To:	Ingo Molnar <mingo@...e.hu>
CC:	LKML <linux-kernel@...r.kernel.org>,
	oprofile-list <oprofile-list@...ts.sourceforge.net>
Subject: [PATCH v2] oprofile: Fix uninitialized memory access when writing to

On 19.12.11 15:17:03, Robert Richter wrote:
> On 19.12.11 15:00:49, Ingo Molnar wrote:
> 
> > +	if (!count)
> > +		return 0;
> > +
> >  	retval = oprofilefs_ulong_from_user(&value, buf, count);
> >  	if (retval)
> >  		return retval;
> > 
> > See the ugly and fragile pattern?
> > 
> > This should *really* be solved via the 
> > oprofilefs_ulong_from_user() helper function, not by sprinkling 
> > the !count checks in half a dozen places ...
> 
> Hmm, I thought there was no way to leave the code path with count == 0
> and retval. But thinking about it it would be possible with returning
> count or errors < 0. Will improve the patch.

Ingo, see my updated version below. I am fine with appling it directly
on tip/perf/urgent.

Thanks,

-Robert

--


>From f07214bc05560f63da8404d31aa83034a16c6229 Mon Sep 17 00:00:00 2001
From: Robert Richter <robert.richter@....com>
Date: Fri, 16 Dec 2011 15:45:31 +0100
Subject: [PATCH] oprofile: Fix uninitialized memory access when writing to
 oprofilefs

If oprofilefs_ulong_from_user() is called with count equals zero, *val
remains unchanged. Depending on the implementation it might be
uninitialized.

Change oprofilefs_ulong_from_user()'s interface to return count on
success. Thus, we are able to return early if count equals zero which
avoids using *val uninitialized. Fixing all users of oprofilefs_ulong_
from_user().

This follows write syscall implementation when count is zero: "If
count is zero ... [and if] no errors are detected, 0 will be returned
without causing any other effect." (man 2 write)

Reported-By: Mike Waychison <mikew@...gle.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: stable@...r.kernel.org
Signed-off-by: Robert Richter <robert.richter@....com>
---
 arch/s390/oprofile/init.c         |    2 +-
 drivers/oprofile/oprofile_files.c |    7 ++++---
 drivers/oprofile/oprofilefs.c     |   11 +++++++++--
 3 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/arch/s390/oprofile/init.c b/arch/s390/oprofile/init.c
index 6efc18b..bd58b72 100644
--- a/arch/s390/oprofile/init.c
+++ b/arch/s390/oprofile/init.c
@@ -88,7 +88,7 @@ static ssize_t hwsampler_write(struct file *file, char const __user *buf,
 		return -EINVAL;
 
 	retval = oprofilefs_ulong_from_user(&val, buf, count);
-	if (retval)
+	if (retval <= 0)
 		return retval;
 
 	if (oprofile_started)
diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c
index 89f6345..84a208d 100644
--- a/drivers/oprofile/oprofile_files.c
+++ b/drivers/oprofile/oprofile_files.c
@@ -45,7 +45,7 @@ static ssize_t timeout_write(struct file *file, char const __user *buf,
 		return -EINVAL;
 
 	retval = oprofilefs_ulong_from_user(&val, buf, count);
-	if (retval)
+	if (retval <= 0)
 		return retval;
 
 	retval = oprofile_set_timeout(val);
@@ -84,7 +84,7 @@ static ssize_t depth_write(struct file *file, char const __user *buf, size_t cou
 		return -EINVAL;
 
 	retval = oprofilefs_ulong_from_user(&val, buf, count);
-	if (retval)
+	if (retval <= 0)
 		return retval;
 
 	retval = oprofile_set_ulong(&oprofile_backtrace_depth, val);
@@ -141,9 +141,10 @@ static ssize_t enable_write(struct file *file, char const __user *buf, size_t co
 		return -EINVAL;
 
 	retval = oprofilefs_ulong_from_user(&val, buf, count);
-	if (retval)
+	if (retval <= 0)
 		return retval;
 
+	retval = 0;
 	if (val)
 		retval = oprofile_start();
 	else
diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c
index d0de6cc..2f0aa0f 100644
--- a/drivers/oprofile/oprofilefs.c
+++ b/drivers/oprofile/oprofilefs.c
@@ -60,6 +60,13 @@ ssize_t oprofilefs_ulong_to_user(unsigned long val, char __user *buf, size_t cou
 }
 
 
+/*
+ * Note: If oprofilefs_ulong_from_user() returns 0, then *val remains
+ * unchanged and might be uninitialized. This follows write syscall
+ * implementation when count is zero: "If count is zero ... [and if]
+ * no errors are detected, 0 will be returned without causing any
+ * other effect." (man 2 write)
+ */
 int oprofilefs_ulong_from_user(unsigned long *val, char const __user *buf, size_t count)
 {
 	char tmpbuf[TMPBUFSIZE];
@@ -79,7 +86,7 @@ int oprofilefs_ulong_from_user(unsigned long *val, char const __user *buf, size_
 	raw_spin_lock_irqsave(&oprofilefs_lock, flags);
 	*val = simple_strtoul(tmpbuf, NULL, 0);
 	raw_spin_unlock_irqrestore(&oprofilefs_lock, flags);
-	return 0;
+	return count;
 }
 
 
@@ -99,7 +106,7 @@ static ssize_t ulong_write_file(struct file *file, char const __user *buf, size_
 		return -EINVAL;
 
 	retval = oprofilefs_ulong_from_user(&value, buf, count);
-	if (retval)
+	if (retval <= 0)
 		return retval;
 
 	retval = oprofile_set_ulong(file->private_data, value);
-- 
1.7.7




-- 
Advanced Micro Devices, Inc.
Operating System Research Center

--
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/