[B.A.T.M.A.N.] [PATCH 1/2] batctl: Fix static buffer for resolved mac adddress

[B.A.T.M.A.N.] [PATCH 1/2] batctl: Fix static buffer for resolved mac adddress

[Sven Eckelmann]

The resolve_mac_* functions are currently not reentrant-safe because they are
using ether_aton. Thus all functions returning a mac address in a similar way
must also ensure that their data is in a statically allocated buffer and not on
a stack.

resolve_mac_from_cache did that for the wrong buffer.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
 functions.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/functions.c b/functions.c
index 9e03e72..959784b 100644
--- a/functions.c
+++ b/functions.c
@@ -630,8 +630,8 @@ static int resolve_mac_from_cache_parse(struct ndmsg *ndmsg, size_t len_payload,
 static struct ether_addr *resolve_mac_from_cache(int ai_family,
 						 const void *l3addr)
 {
-	static uint8_t l3addr_tmp[16];
-	struct ether_addr mac_tmp;
+	uint8_t l3addr_tmp[16];
+	static struct ether_addr mac_tmp;
 	struct ether_addr *mac_result = NULL;
 	void *buf = NULL;
 	size_t buflen;
-- 
1.8.4.rc3

[B.A.T.M.A.N.] [PATCH 2/2] batctl: Finish resolving of mac address early

[Sven Eckelmann]

It is not necessary to check the remaining msg buffer coming from the kernel
when already the correct mac address was found.

Signed-off-by: Sven Eckelmann <sven@narfation.org>
---
 functions.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/functions.c b/functions.c
index 959784b..f2b23f2 100644
--- a/functions.c
+++ b/functions.c
@@ -683,6 +683,7 @@ static struct ether_addr *resolve_mac_from_cache(int ai_family,
 			if (parsed) {
 				if (memcmp(&l3addr_tmp, l3addr, l3_len) == 0) {
 					mac_result = &mac_tmp;
+					finished = 1;
 					break;
 				}
 			}
-- 
1.8.4.rc3

Re: [B.A.T.M.A.N.] [PATCH 1/2] batctl: Fix static buffer for resolved mac adddress

[Marek Lindner]

[-- Attachment #1: Type: text/plain, Size: 552 bytes --]

On Tuesday 15 October 2013 18:01:08 Sven Eckelmann wrote:
> The resolve_mac_* functions are currently not reentrant-safe because they
> are using ether_aton. Thus all functions returning a mac address in a
> similar way must also ensure that their data is in a statically allocated
> buffer and not on a stack.
> 
> resolve_mac_from_cache did that for the wrong buffer.
> 
> Signed-off-by: Sven Eckelmann <sven@narfation.org>
> ---
>  functions.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

Applied in revision ef41a8d.

Thanks,
Marek

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

Re: [B.A.T.M.A.N.] [PATCH 2/2] batctl: Finish resolving of mac address early

[Marek Lindner]

[-- Attachment #1: Type: text/plain, Size: 349 bytes --]

On Tuesday 15 October 2013 18:01:09 Sven Eckelmann wrote:
> It is not necessary to check the remaining msg buffer coming from the kernel
> when already the correct mac address was found.
> 
> Signed-off-by: Sven Eckelmann <sven@narfation.org>
> ---
>  functions.c | 1 +
>  1 file changed, 1 insertion(+)

Applied in revision 7cd9934.

Thanks,
Marek

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 490 bytes --]