All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alexey Dobriyan <adobriyan@sw.ru>
To: akpm@osdl.org
Cc: linux-kernel@vger.kernel.org, devel@openvz.org, mingo@elte.hu,
	tglx@linutronix.de, viro@zeniv.linux.org.uk,
	rusty@rustcorp.com.au
Subject: [PATCH] Fix some kallsyms_lookup() vs rmmod races
Date: Thu, 15 Mar 2007 19:14:42 +0300	[thread overview]
Message-ID: <20070315161442.GA6812@localhost.sw.ru> (raw)

[cc'ing folks whose proc files are affected]

kallsyms_lookup() can call module_address_lookup() which iterates over
modules list without module_mutex taken. Comment at the top of
module_address_lookup() says it's for oops resolution so races are
irrelevant, but in some cases it's reachable from regular code:

BUG: unable to handle kernel paging request at virtual address f94fb19c
 printing eip:
c0138097
*pde = 33576067
Oops: 0000 [#1]
PREEMPT
Modules linked in: ohci_hcd af_packet e1000 ehci_hcd uhci_hcd usbcore
CPU:    0
EIP:    0060:[<c0138097>]    Not tainted VLI
EFLAGS: 00010246   (2.6.21-rc3-8b9909ded6922c33c221b105b26917780cfa497d #25)
EIP is at module_address_lookup+0x43/0x24c
eax: 00000000   ebx: 00000000   ecx: f94fb080   edx: f882e704
esi: 00000000   edi: 00000000   ebp: 00000000   esp: d30b8e78
ds: 007b   es: 007b   fs: 00d8  gs: 0033  ss: 0068
Process cat (pid: 13274, ti=d30b8000 task=d2b50af0 task.ti=d30b8000)
Stack: d30b8f44 d30b8f48 00008001 00000000 c039a23c d30b8ec4 00000000 d30b8f44
       d30b8f48 c013876c d30b8f4c 00000000 cf544000 fffffff4 d38ebb7c c01880ff
       d30b8f4c d30b8ec4 00000246 c0399600 00000001 00000001 00000000 c0399654
Call Trace:
 [<c013876c>] kallsyms_lookup+0x5e/0x7c
 [<c01880ff>] proc_pid_wchan+0x38/0x76
 [<c01404fe>] __alloc_pages+0x4f/0x2f9
 [<c01891c6>] proc_info_read+0x6f/0xa8
 [<c015bbfd>] sys_fstat64+0x1e/0x23
 [<c01592d4>] vfs_read+0x7d/0xb5
 [<c0189157>] proc_info_read+0x0/0xa8
 [<c0159621>] sys_read+0x41/0x6a
 [<c0103e72>] sysenter_past_esp+0x5f/0x99
 =======================
Code: 04 8b 51 04 0f 18 02 90 81 3d 4c 8d 39 c0 4c 8d 39 c0 74 3f 8b b9 18 01 00 00 8b 99 10 01 00 00 39 eb 77 07 8d 04 3b 39 c5 72 32 <8b> b1 1c 01 00 00 8b 81 14 01 00 00 39 c5 72 06 01 f0 39 c5 72
EIP: [<c0138097>] module_address_lookup+0x43/0x24c SS:ESP 0068:d30b8e78

Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
---

 fs/proc/base.c            |    7 +++++++
 kernel/time/timer_list.c  |    6 ++++++
 kernel/time/timer_stats.c |    6 ++++++
 mm/slab.c                 |    3 +++
 4 files changed, 22 insertions(+)

--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -61,6 +61,7 @@ #include <linux/seq_file.h>
 #include <linux/namei.h>
 #include <linux/mnt_namespace.h>
 #include <linux/mm.h>
+#include <linux/module.h>
 #include <linux/smp_lock.h>
 #include <linux/rcupdate.h>
 #include <linux/kallsyms.h>
@@ -282,7 +283,13 @@ static int proc_pid_wchan(struct task_st
 
 	wchan = get_wchan(task);
 
+	mutex_lock(&module_mutex);
 	sym_name = kallsyms_lookup(wchan, &size, &offset, &modname, namebuf);
+	/*
+	 * size, offset, modname aren't used, sym_name is copied into namebuf,
+	 * so drop it now.
+	 */
+	mutex_unlock(&module_mutex);
 	if (sym_name)
 		return sprintf(buffer, "%s", sym_name);
 	return sprintf(buffer, "%lu", wchan);
--- a/kernel/time/timer_list.c
+++ b/kernel/time/timer_list.c
@@ -44,7 +44,13 @@ static void print_name_offset(struct seq
 	const char *sym_name;
 	char *modname;
 
+	mutex_lock(&module_mutex);
 	sym_name = kallsyms_lookup(addr, &size, &offset, &modname, namebuf);
+	/*
+	 * size, offset, modname aren't used, sym_name is copied into namebuf,
+	 * so drop it now.
+	 */
+	mutex_unlock(&module_mutex);
 	if (sym_name)
 		SEQ_printf(m, "%s", sym_name);
 	else
--- a/kernel/time/timer_stats.c
+++ b/kernel/time/timer_stats.c
@@ -262,7 +262,13 @@ static void print_name_offset(struct seq
 	const char *sym_name;
 	char *modname;
 
+	mutex_lock(&module_mutex);
 	sym_name = kallsyms_lookup(addr, &size, &offset, &modname, namebuf);
+	/*
+	 * size, offset, modname aren't used, sym_name is copied into namebuf,
+	 * so drop it now.
+	 */
+	mutex_unlock(&module_mutex);
 	if (sym_name)
 		seq_printf(m, "%s", sym_name);
 	else
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -4385,14 +4385,17 @@ #ifdef CONFIG_KALLSYMS
 	unsigned long offset, size;
 	char namebuf[KSYM_NAME_LEN+1];
 
+	mutex_lock(&module_mutex);
 	name = kallsyms_lookup(address, &size, &offset, &modname, namebuf);
 
 	if (name) {
 		seq_printf(m, "%s+%#lx/%#lx", name, offset, size);
 		if (modname)
 			seq_printf(m, " [%s]", modname);
+		mutex_unlock(&module_mutex);
 		return;
 	}
+	mutex_unlock(&module_mutex);
 #endif
 	seq_printf(m, "%p", (void *)address);
 }


             reply	other threads:[~2007-03-15 16:07 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-03-15 16:14 Alexey Dobriyan [this message]
2007-03-15 16:53 ` [PATCH] Fix some kallsyms_lookup() vs rmmod races Paulo Marques
2007-03-15 18:26   ` Alexey Dobriyan

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=20070315161442.GA6812@localhost.sw.ru \
    --to=adobriyan@sw.ru \
    --cc=akpm@osdl.org \
    --cc=devel@openvz.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=rusty@rustcorp.com.au \
    --cc=tglx@linutronix.de \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.