[PATCH] r8152: Avoid memcpy() over-reading of ETH_SS_STATS

Subsystems: networking drivers, the rest, usb networking drivers

STALE1878d

5 messages, 3 authors, 2021-06-17 · open the first message on its own page

[PATCH] r8152: Avoid memcpy() over-reading of ETH_SS_STATS

From: Kees Cook <hidden>
Date: 2021-06-16 19:53:20

In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array.
Adjust the source argument so the compiler can do appropriate bounds
checking.

Signed-off-by: Kees Cook <redacted>
---
 drivers/net/usb/r8152.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 85039e17f4cd..5f08720bf1c9 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8678,7 +8678,7 @@ static void rtl8152_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *rtl8152_gstrings, sizeof(rtl8152_gstrings));
+		memcpy(data, rtl8152_gstrings, sizeof(rtl8152_gstrings));
 		break;
 	}
 }
-- 
2.25.1

Re: [PATCH] r8152: Avoid memcpy() over-reading of ETH_SS_STATS

From: Andrew Lunn <andrew@lunn.ch>
Date: 2021-06-16 20:02:59

On Wed, Jun 16, 2021 at 12:53:03PM -0700, Kees Cook wrote:
quoted hunk
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array.
Adjust the source argument so the compiler can do appropriate bounds
checking.

Signed-off-by: Kees Cook <redacted>
---
 drivers/net/usb/r8152.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 85039e17f4cd..5f08720bf1c9 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8678,7 +8678,7 @@ static void rtl8152_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *rtl8152_gstrings, sizeof(rtl8152_gstrings));
+		memcpy(data, rtl8152_gstrings, sizeof(rtl8152_gstrings));
 		break;
Is this correct? The call is supposed to return all the statistic
strings, which would be the entire structure.

	  Andrew

Re: [PATCH] r8152: Avoid memcpy() over-reading of ETH_SS_STATS

From: patchwork-bot+netdevbpf@kernel.org
Date: 2021-06-16 20:10:09

Hello:

This patch was applied to netdev/net.git (refs/heads/master):

On Wed, 16 Jun 2021 12:53:03 -0700 you wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array.
Adjust the source argument so the compiler can do appropriate bounds
checking.

[...]
Here is the summary with links:
  - r8152: Avoid memcpy() over-reading of ETH_SS_STATS
    https://git.kernel.org/netdev/net/c/99718abdc00e

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH] r8152: Avoid memcpy() over-reading of ETH_SS_STATS

From: Andrew Lunn <andrew@lunn.ch>
Date: 2021-06-16 20:11:11

On Wed, Jun 16, 2021 at 10:02:43PM +0200, Andrew Lunn wrote:
On Wed, Jun 16, 2021 at 12:53:03PM -0700, Kees Cook wrote:
quoted
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array.
Adjust the source argument so the compiler can do appropriate bounds
checking.

Signed-off-by: Kees Cook <redacted>
---
 drivers/net/usb/r8152.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 85039e17f4cd..5f08720bf1c9 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8678,7 +8678,7 @@ static void rtl8152_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *rtl8152_gstrings, sizeof(rtl8152_gstrings));
+		memcpy(data, rtl8152_gstrings, sizeof(rtl8152_gstrings));
 		break;
Is this correct? The call is supposed to return all the statistic
strings, which would be the entire structure.
Ah! now i think i get it.

Although *rtl8152_gstrings == rtl8152_gstrings in terms of addresses,
the compiler sees that *rtl8152_gstrings is sizeof(ETH_GSTRING_LEN),
but we are copying sizeof(rtl8152_gstrings), so it will issue a
warning. So you remove the * to indicate we are interesting in the
whole structure of arrays.

      Andrew

Re: [PATCH] r8152: Avoid memcpy() over-reading of ETH_SS_STATS

From: Kees Cook <hidden>
Date: 2021-06-17 03:21:11

On Wed, Jun 16, 2021 at 10:10:40PM +0200, Andrew Lunn wrote:
On Wed, Jun 16, 2021 at 10:02:43PM +0200, Andrew Lunn wrote:
quoted
On Wed, Jun 16, 2021 at 12:53:03PM -0700, Kees Cook wrote:
quoted
In preparation for FORTIFY_SOURCE performing compile-time and run-time
field bounds checking for memcpy(), memmove(), and memset(), avoid
intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array.
Adjust the source argument so the compiler can do appropriate bounds
checking.

Signed-off-by: Kees Cook <redacted>
---
 drivers/net/usb/r8152.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/usb/r8152.c b/drivers/net/usb/r8152.c
index 85039e17f4cd..5f08720bf1c9 100644
--- a/drivers/net/usb/r8152.c
+++ b/drivers/net/usb/r8152.c
@@ -8678,7 +8678,7 @@ static void rtl8152_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *rtl8152_gstrings, sizeof(rtl8152_gstrings));
+		memcpy(data, rtl8152_gstrings, sizeof(rtl8152_gstrings));
 		break;
Is this correct? The call is supposed to return all the statistic
strings, which would be the entire structure.
Ah! now i think i get it.

Although *rtl8152_gstrings == rtl8152_gstrings in terms of addresses,
the compiler sees that *rtl8152_gstrings is sizeof(ETH_GSTRING_LEN),
but we are copying sizeof(rtl8152_gstrings), so it will issue a
warning. So you remove the * to indicate we are interesting in the
whole structure of arrays.
Right! Sorry if that wasn't more clear. :)

-- 
Kees Cook
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help