[PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

['Simon Horman']

From: Sam Sun <samsun1006219@gmail.com>

In function bond_option_arp_ip_targets_set(), if newval->string is an
empty string, newval->string+1 will point to the byte after the
string, causing an out-of-bound read.

BUG: KASAN: slab-out-of-bounds in strlen+0x7d/0xa0 lib/string.c:418
Read of size 1 at addr ffff8881119c4781 by task syz-executor665/8107
CPU: 1 PID: 8107 Comm: syz-executor665 Not tainted 6.7.0-rc7 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
 print_address_description mm/kasan/report.c:364 [inline]
 print_report+0xc1/0x5e0 mm/kasan/report.c:475
 kasan_report+0xbe/0xf0 mm/kasan/report.c:588
 strlen+0x7d/0xa0 lib/string.c:418
 __fortify_strlen include/linux/fortify-string.h:210 [inline]
 in4_pton+0xa3/0x3f0 net/core/utils.c:130
 bond_option_arp_ip_targets_set+0xc2/0x910
drivers/net/bonding/bond_options.c:1201
 __bond_opt_set+0x2a4/0x1030 drivers/net/bonding/bond_options.c:767
 __bond_opt_set_notify+0x48/0x150 drivers/net/bonding/bond_options.c:792
 bond_opt_tryset_rtnl+0xda/0x160 drivers/net/bonding/bond_options.c:817
 bonding_sysfs_store_option+0xa1/0x120 drivers/net/bonding/bond_sysfs.c:156
 dev_attr_store+0x54/0x80 drivers/base/core.c:2366
 sysfs_kf_write+0x114/0x170 fs/sysfs/file.c:136
 kernfs_fop_write_iter+0x337/0x500 fs/kernfs/file.c:334
 call_write_iter include/linux/fs.h:2020 [inline]
 new_sync_write fs/read_write.c:491 [inline]
 vfs_write+0x96a/0xd80 fs/read_write.c:584
 ksys_write+0x122/0x250 fs/read_write.c:637
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0x40/0x110 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x63/0x6b
---[ end trace ]---

Fix it by adding a check of string length before using it.

Fixes: f9de11a16594 ("bonding: add ip checks when store ip target")
Signed-off-by: Yue Sun <samsun1006219@gmail.com>
Signed-off-by: Simon Horman <horms@kernel.org>
---
Changes in v5 (Simon):
- Remove stray 'I4' from netdev_err() string. Thanks to Hangbin Liu.
- Sorry for the long delay between v4 and v5, this completely slipped my
  mind.
- Link to v4: https://lore.kernel.org/r/20240419-bond-oob-v4-1-69dd1a66db20@kernel.org

Changes in v4 (Simon):
- Correct  whitespace mangled patch; posting as requested by Sam Sun
- Link to v3: https://lore.kernel.org/r/CAEkJfYOnsLLiCrtgOpq2Upr+_W0dViYVHU8YdjJOi-mxD8H9oQ@mail.gmail.com

Changes in v3 (Sam Sun):
- According to Hangbin's opinion, change Fixes tag from 4fb0ef585eb2
  ("bonding: convert arp_ip_target to use the new option API") to
  f9de11a16594 ("bonding: add ip checks when store ip target").
- Link to v2: https://lore.kernel.org/r/CAEkJfYMdDQKY1C-wBZLiaJ=dCqfM9r=rykwwf+J-XHsFp7D9Ag@mail.gmail.com/

Changes in v2 (Sam Sun):
- According to Jay and Hangbin's opinion, remove target address in
  netdev_err message since target is not initialized in error path and
  will not provide useful information.
- Link to v1: https://lore.kernel.org/r/CAEkJfYPYF-nNB2oiXfXwjPG0VVB2Bd8Q8kAq+74J=R+4HkngWw@mail.gmail.com/
---
 drivers/net/bonding/bond_options.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index 0cacd7027e35..228128e727c2 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -1214,9 +1214,9 @@ static int bond_option_arp_ip_targets_set(struct bonding *bond,
 	__be32 target;
 
 	if (newval->string) {
-		if (!in4_pton(newval->string+1, -1, (u8 *)&target, -1, NULL)) {
-			netdev_err(bond->dev, "invalid ARP target %pI4 specified\n",
-				   &target);
+		if (!(strlen(newval->string)) ||
+		    !in4_pton(newval->string + 1, -1, (u8 *)&target, -1, NULL)) {
+			netdev_err(bond->dev, "invalid ARP target specified\n");
 			return ret;
 		}
 		if (newval->string[0] == '+')

Re: [PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[Jay Vosburgh]

'Simon Horman' <horms@kernel.org> wrote:

>From: Sam Sun <samsun1006219@gmail.com>
>
>In function bond_option_arp_ip_targets_set(), if newval->string is an
>empty string, newval->string+1 will point to the byte after the
>string, causing an out-of-bound read.
>
>BUG: KASAN: slab-out-of-bounds in strlen+0x7d/0xa0 lib/string.c:418
>Read of size 1 at addr ffff8881119c4781 by task syz-executor665/8107
>CPU: 1 PID: 8107 Comm: syz-executor665 Not tainted 6.7.0-rc7 #1
>Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
>Call Trace:
> <TASK>
> __dump_stack lib/dump_stack.c:88 [inline]
> dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
> print_address_description mm/kasan/report.c:364 [inline]
> print_report+0xc1/0x5e0 mm/kasan/report.c:475
> kasan_report+0xbe/0xf0 mm/kasan/report.c:588
> strlen+0x7d/0xa0 lib/string.c:418
> __fortify_strlen include/linux/fortify-string.h:210 [inline]
> in4_pton+0xa3/0x3f0 net/core/utils.c:130
> bond_option_arp_ip_targets_set+0xc2/0x910
>drivers/net/bonding/bond_options.c:1201
> __bond_opt_set+0x2a4/0x1030 drivers/net/bonding/bond_options.c:767
> __bond_opt_set_notify+0x48/0x150 drivers/net/bonding/bond_options.c:792
> bond_opt_tryset_rtnl+0xda/0x160 drivers/net/bonding/bond_options.c:817
> bonding_sysfs_store_option+0xa1/0x120 drivers/net/bonding/bond_sysfs.c:156
> dev_attr_store+0x54/0x80 drivers/base/core.c:2366
> sysfs_kf_write+0x114/0x170 fs/sysfs/file.c:136
> kernfs_fop_write_iter+0x337/0x500 fs/kernfs/file.c:334
> call_write_iter include/linux/fs.h:2020 [inline]
> new_sync_write fs/read_write.c:491 [inline]
> vfs_write+0x96a/0xd80 fs/read_write.c:584
> ksys_write+0x122/0x250 fs/read_write.c:637
> do_syscall_x64 arch/x86/entry/common.c:52 [inline]
> do_syscall_64+0x40/0x110 arch/x86/entry/common.c:83
> entry_SYSCALL_64_after_hwframe+0x63/0x6b
>---[ end trace ]---
>
>Fix it by adding a check of string length before using it.
>
>Fixes: f9de11a16594 ("bonding: add ip checks when store ip target")
>Signed-off-by: Yue Sun <samsun1006219@gmail.com>
>Signed-off-by: Simon Horman <horms@kernel.org>

Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com>

>---
>Changes in v5 (Simon):
>- Remove stray 'I4' from netdev_err() string. Thanks to Hangbin Liu.
>- Sorry for the long delay between v4 and v5, this completely slipped my
>  mind.
>- Link to v4: https://lore.kernel.org/r/20240419-bond-oob-v4-1-69dd1a66db20@kernel.org
>
>Changes in v4 (Simon):
>- Correct  whitespace mangled patch; posting as requested by Sam Sun
>- Link to v3: https://lore.kernel.org/r/CAEkJfYOnsLLiCrtgOpq2Upr+_W0dViYVHU8YdjJOi-mxD8H9oQ@mail.gmail.com
>
>Changes in v3 (Sam Sun):
>- According to Hangbin's opinion, change Fixes tag from 4fb0ef585eb2
>  ("bonding: convert arp_ip_target to use the new option API") to
>  f9de11a16594 ("bonding: add ip checks when store ip target").
>- Link to v2: https://lore.kernel.org/r/CAEkJfYMdDQKY1C-wBZLiaJ=dCqfM9r=rykwwf+J-XHsFp7D9Ag@mail.gmail.com/
>
>Changes in v2 (Sam Sun):
>- According to Jay and Hangbin's opinion, remove target address in
>  netdev_err message since target is not initialized in error path and
>  will not provide useful information.
>- Link to v1: https://lore.kernel.org/r/CAEkJfYPYF-nNB2oiXfXwjPG0VVB2Bd8Q8kAq+74J=R+4HkngWw@mail.gmail.com/
>---
> drivers/net/bonding/bond_options.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
>diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
>index 0cacd7027e35..228128e727c2 100644
>--- a/drivers/net/bonding/bond_options.c
>+++ b/drivers/net/bonding/bond_options.c
>@@ -1214,9 +1214,9 @@ static int bond_option_arp_ip_targets_set(struct bonding *bond,
> 	__be32 target;
> 
> 	if (newval->string) {
>-		if (!in4_pton(newval->string+1, -1, (u8 *)&target, -1, NULL)) {
>-			netdev_err(bond->dev, "invalid ARP target %pI4 specified\n",
>-				   &target);
>+		if (!(strlen(newval->string)) ||
>+		    !in4_pton(newval->string + 1, -1, (u8 *)&target, -1, NULL)) {
>+			netdev_err(bond->dev, "invalid ARP target specified\n");
> 			return ret;
> 		}
> 		if (newval->string[0] == '+')
>

Re: [PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[Hangbin Liu]

On Sun, Jun 30, 2024 at 02:20:55PM +0100, 'Simon Horman' wrote:
> From: Sam Sun <samsun1006219@gmail.com>
> 
> In function bond_option_arp_ip_targets_set(), if newval->string is an
> empty string, newval->string+1 will point to the byte after the
> string, causing an out-of-bound read.
> 
> BUG: KASAN: slab-out-of-bounds in strlen+0x7d/0xa0 lib/string.c:418
> Read of size 1 at addr ffff8881119c4781 by task syz-executor665/8107
> CPU: 1 PID: 8107 Comm: syz-executor665 Not tainted 6.7.0-rc7 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
> Call Trace:
>  <TASK>
>  __dump_stack lib/dump_stack.c:88 [inline]
>  dump_stack_lvl+0xd9/0x150 lib/dump_stack.c:106
>  print_address_description mm/kasan/report.c:364 [inline]
>  print_report+0xc1/0x5e0 mm/kasan/report.c:475
>  kasan_report+0xbe/0xf0 mm/kasan/report.c:588
>  strlen+0x7d/0xa0 lib/string.c:418
>  __fortify_strlen include/linux/fortify-string.h:210 [inline]
>  in4_pton+0xa3/0x3f0 net/core/utils.c:130
>  bond_option_arp_ip_targets_set+0xc2/0x910
> drivers/net/bonding/bond_options.c:1201
>  __bond_opt_set+0x2a4/0x1030 drivers/net/bonding/bond_options.c:767
>  __bond_opt_set_notify+0x48/0x150 drivers/net/bonding/bond_options.c:792
>  bond_opt_tryset_rtnl+0xda/0x160 drivers/net/bonding/bond_options.c:817
>  bonding_sysfs_store_option+0xa1/0x120 drivers/net/bonding/bond_sysfs.c:156
>  dev_attr_store+0x54/0x80 drivers/base/core.c:2366
>  sysfs_kf_write+0x114/0x170 fs/sysfs/file.c:136
>  kernfs_fop_write_iter+0x337/0x500 fs/kernfs/file.c:334
>  call_write_iter include/linux/fs.h:2020 [inline]
>  new_sync_write fs/read_write.c:491 [inline]
>  vfs_write+0x96a/0xd80 fs/read_write.c:584
>  ksys_write+0x122/0x250 fs/read_write.c:637
>  do_syscall_x64 arch/x86/entry/common.c:52 [inline]
>  do_syscall_64+0x40/0x110 arch/x86/entry/common.c:83
>  entry_SYSCALL_64_after_hwframe+0x63/0x6b
> ---[ end trace ]---
> 
> Fix it by adding a check of string length before using it.
> 
> Fixes: f9de11a16594 ("bonding: add ip checks when store ip target")
> Signed-off-by: Yue Sun <samsun1006219@gmail.com>
> Signed-off-by: Simon Horman <horms@kernel.org>

Reviewed-by: Hangbin Liu <liuhangbin@gmail.com>

Re: [PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[Jakub Kicinski]

On Sun, 30 Jun 2024 14:20:55 +0100 'Simon Horman' wrote:
> +		if (!(strlen(newval->string)) ||

The extra brackets made me look, this really feels like it wants to be
written as:

	if (!newval->string[0] ||

or:

	if (strlen(newval->string) < 1 ||

Re: [PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[Jay Vosburgh]

Jakub Kicinski <kuba@kernel.org> wrote:

>On Sun, 30 Jun 2024 14:20:55 +0100 'Simon Horman' wrote:
>> +		if (!(strlen(newval->string)) ||
>
>The extra brackets made me look, this really feels like it wants to be
>written as:
>
>	if (!newval->string[0] ||
>
>or:
>
>	if (strlen(newval->string) < 1 ||

	I find the second option clearer, FWIW.  This isn't in a hot
path, and including strlen() in there makes it more obvious to my
reading what the intent is.  The size_t return from strlen() is
unsigned, so we really want to test the return value for zero-ness.

	-J

---
	-Jay Vosburgh, jay.vosburgh@canonical.com

Re: [PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[Jakub Kicinski]

On Mon, 01 Jul 2024 16:07:46 -0700 Jay Vosburgh wrote:
> >	if (strlen(newval->string) < 1 ||  
> 
> 	I find the second option clearer, FWIW.  This isn't in a hot
> path, and including strlen() in there makes it more obvious to my
> reading what the intent is.  The size_t return from strlen() is
> unsigned, so we really want to test the return value for zero-ness.

True, I picked < 1 because the line below offsets by 1.
Either way is fine, but lets make sure we drop the parenthesis.

Re: [PATCH net v5] bonding: Fix out-of-bounds read in bond_option_arp_ip_targets_set()

[Simon Horman]

On Mon, Jul 01, 2024 at 07:59:08PM -0700, Jakub Kicinski wrote:
> On Mon, 01 Jul 2024 16:07:46 -0700 Jay Vosburgh wrote:
> > >	if (strlen(newval->string) < 1 ||  
> > 
> > 	I find the second option clearer, FWIW.  This isn't in a hot
> > path, and including strlen() in there makes it more obvious to my
> > reading what the intent is.  The size_t return from strlen() is
> > unsigned, so we really want to test the return value for zero-ness.
> 
> True, I picked < 1 because the line below offsets by 1.
> Either way is fine, but lets make sure we drop the parenthesis.

Thanks, I will go with < 1.