[195] in linux-net channel archive
comments/questions on Token Ring
daemon@ATHENA.MIT.EDU (Alvin Starr)
Tue Apr 11 17:20:23 1995
From: Alvin Starr <alvin@eyepoint.com>
To: linux-net@vger.rutgers.edu
Date: Tue, 11 Apr 1995 09:31:44 -0400
After a whole lot of fun with getting token ring up and running locally.
I had the unfortunate task of getting TR running over a Netronics router/bridge.
Conciderable time and not a few hairs later I think I have found some points
that may be of use to the rest of the TR users.
1) The Netronics TokenMaster 4000 seems to not like having the 0x80 bit in the
source address set if the route is to be handled in a non-source routing
format. If the 0x80 bit is set the router will not even respond to pings.
2) from one of the people at Netronics it was made quite clear that if the
0x80 bit is set then the message is to be source routed independant of
any information in the rcf field.
3) in tr_source_route I removed the last "trh->saddr[0]|=TR_RII" statement
and things started to work across teh bridge.
My question is: Am what I am doing and suggesting correct?
Please respond directly I am not a part of this group normaly.
My suggested changes are:
static void tr_source_route(struct trh_hdr *trh,struct device *dev) {
int i;
unsigned int hash;
rif_cache entry;
/* Broadcasts are single route as stated in RFC 1042 */
if(!memcmp(&(trh->daddr[0]),&(dev->broadcast[0]),TR_ALEN)) {
trh->rcf=htons((((sizeof(trh->rcf)) << 8) & TR_RCF_LEN_MASK)
| TR_RCF_FRAME2K | TR_RCF_LIMITED_BROADCAST);
trh->saddr[0]|=TR_RII;
---------------^^^^^^^^^^^^^^^^^^^^^
I am not sure but I think this line should be removed?
}
else {
for(i=0,hash=0;i<TR_ALEN;hash+=trh->daddr[i++]);
hash&=RIF_TABLE_SIZE-1;
for(entry=rif_table[hash];entry && memcmp(&(entry->addr[0]),&(trh->daddr[0]),TR_ALEN);entry=entry->next);
if(entry) {
#if 0
printk("source routing for %02X %02X %02X %02X %02X %02X\n",trh->daddr[0],
trh->daddr[1],trh->daddr[2],trh->daddr[3],trh->daddr[4],trh->daddr[5]);
#endif
if((ntohs(entry->rcf) & TR_RCF_LEN_MASK) >> 8) {
trh->rcf=entry->rcf;
memcpy(&trh->rseg[0],&entry->rseg[0],8*sizeof(un
signed short));
trh->rcf^=htons(TR_RCF_DIR_BIT);
trh->rcf&=htons(0x1fff); /* Issam Chehab<ichehab@madge1.demon.co.uk> */
trh->saddr[0]|=TR_RII;
entry->last_used=jiffies;
}
}
else {
trh->rcf=htons((((sizeof(trh->rcf)) << 8) & TR_RCF_LEN_MASK)
| TR_RCF_FRAME2K | TR_RCF_LIMITED_BROADCAST);
trh->saddr[0]|=TR_RII;
-----------------------^^^^^^^^^^^^^^^^^^^^^
removing this line seems to make thing work.
}
}
}
--
Alvin Starr || voice: (905)513-6717
Eyepoint Inc. || fax: (905)513-6718
alvin@eyepoint.com ||
