Re: Poss. bug in tulip driver since 2.4.7

[Andrew Morton]

Prasanna Meda <pmeda@xxxxxxxxxx> wrote:
>
> The inner for loop shown below was not
>  supposed to be  inside  the  outside  loop.
>  They  also use  the  same index i.
>  Due to  this, when mc_count is more than
>  14,  with non ASIX chips,  panics, corruptions
>  and  denial of services to multicast addresses
>  can  result!
> 
>  http://lxr.linux.no/source/drivers/net/tulip/tulip_core.c#L1055

So can you confirm that the driver works correctly with this change?

--- 25/drivers/net/tulip/tulip_core.c~tulip-hash-fix	2003-11-11 18:51:52.000000000 -0800
+++ 25-akpm/drivers/net/tulip/tulip_core.c	2003-11-11 18:52:31.000000000 -0800
@@ -979,12 +979,13 @@ static void build_setup_frame_hash(u16 *
 	for (i = 0, mclist = dev->mc_list; mclist && i < dev->mc_count;
 	     i++, mclist = mclist->next) {
 		int index = ether_crc_le(ETH_ALEN, mclist->dmi_addr) & 0x1ff;
+		int j;
 
 		set_bit_le(index, hash_table);
 
-		for (i = 0; i < 32; i++) {
-			*setup_frm++ = hash_table[i];
-			*setup_frm++ = hash_table[i];
+		for (j = 0; j < 32; j++) {
+			*setup_frm++ = hash_table[j];
+			*setup_frm++ = hash_table[j];
 		}
 		setup_frm = &tp->setup_frame[13*6];
 	}

_

-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html