Re: canonical and non-canonical addresses
Does any body know the answer to this one? Now I am getting a bit frustrated. I have posted to this group only a couple of times and I never get an answer. All I get is people asking me to look at the archives. I did as much homework as possible before asking the group. ""Neil Desai"" [EMAIL PROTECTED] wrote in message 8n9llp$83c$[EMAIL PROTECTED]">news:8n9llp$83c$[EMAIL PROTECTED]... I am having a problem understanding the issues between canonical and non-canonical addressing. I understand that the bits are flipped within the byte. On page 32-33 of Interconnections Second Edition she gives the example of the address a2-41-42-59-31-51. Canonical: 10100010 0101 0110 01011001 00110001 01010001 Non-Canonical: 01000101 1010 0110 10011010 10001100 10001010 If you look at this you can clearly see that the address in canonical format is not a group address (last bit of first byte is zero) but in non-canonical format it is a group address. At this point I can see a big problem because she also states: ".the group bit in addresses was defined not as "the most significant bit" or the "least significant bit" but rather as "the first bit on the wire." Thus, an address that was a group address on 802.3 would not necessarily look like a group address when transmitted on 802.5 because a different bit would be transmitted first." Here is the confusion: In canonical format the least significant bit is transmitted first and in non-canonical format the most significant bit is transmitted first. So on the wire the 1's and 0's would be in the same order. Here is an excerpt from RFC 2469: The figure below illustrates the difference between canonical and non-canonical form using the canonical form address 12-34-56-78-9A-BC as an example: In memory, 12 34 56 78 9A BC canonical: 00010010 00110100 01010110 0000 10011010 1000 1st bit appearing on LAN (group address indicator) | On LAN: 01001000 00101100 01101010 0000 01011001 0001 In memory, MSB format: 01001000 00101100 01101010 0000 01011001 0001 48 2C 6A 1E 59 3D This shows that no matter how the information is stored in memory it looks the same on the wire. So if it looks the same on the wire wouldn't an adapter pickup the packet and flip the bits in the byte if it needed to. Since it on the wire it looks like the bits are in non-canonical format a canonical format media would automatically take the first byte and flip the bits and so on, or so I would think. If anyone can figure out where I am going wrong please let me know. If it would be best to talk, email me directly with a daytime phone number and I will call you. Thanks. Neil ___ UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html FAQ, list archives, and subscription info: http://www.groupstudy.com Report misconduct and Nondisclosure violations to [EMAIL PROTECTED] --- ___ UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html FAQ, list archives, and subscription info: http://www.groupstudy.com Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: canonical and non-canonical addresses
Neil, I'm not sure what your question is. The bits on the wire are transmitted the same 0100. Token-ring reads them one way (non-canonical) 0100, and Ethernet reads them another way (canonical)0010. It is how the network interface cards (NIC)/device drivers read the 0s and 1s that is different. I would be happy to help(since I have struggled with this myself) if you can clarify your question Regards, Drew R. Maness, CCNP, CCDA, CCSI, MCSE -Original Message- From: Neil Desai [mailto:[EMAIL PROTECTED]] Sent: Tuesday, August 15, 2000 12:47 PM To: [EMAIL PROTECTED] Subject: Re: canonical and non-canonical addresses Does any body know the answer to this one? Now I am getting a bit frustrated. I have posted to this group only a couple of times and I never get an answer. All I get is people asking me to look at the archives. I did as much homework as possible before asking the group. ""Neil Desai"" [EMAIL PROTECTED] wrote in message 8n9llp$83c$[EMAIL PROTECTED]">news:8n9llp$83c$[EMAIL PROTECTED]... I am having a problem understanding the issues between canonical and non-canonical addressing. I understand that the bits are flipped within the byte. On page 32-33 of Interconnections Second Edition she gives the example of the address a2-41-42-59-31-51. Canonical: 10100010 0101 0110 01011001 00110001 01010001 Non-Canonical: 01000101 1010 0110 10011010 10001100 10001010 If you look at this you can clearly see that the address in canonical format is not a group address (last bit of first byte is zero) but in non-canonical format it is a group address. At this point I can see a big problem because she also states: ".the group bit in addresses was defined not as "the most significant bit" or the "least significant bit" but rather as "the first bit on the wire." Thus, an address that was a group address on 802.3 would not necessarily look like a group address when transmitted on 802.5 because a different bit would be transmitted first." Here is the confusion: In canonical format the least significant bit is transmitted first and in non-canonical format the most significant bit is transmitted first. So on the wire the 1's and 0's would be in the same order. Here is an excerpt from RFC 2469: The figure below illustrates the difference between canonical and non-canonical form using the canonical form address 12-34-56-78-9A-BC as an example: In memory, 12 34 56 78 9A BC canonical: 00010010 00110100 01010110 0000 10011010 1000 1st bit appearing on LAN (group address indicator) | On LAN: 01001000 00101100 01101010 0000 01011001 0001 In memory, MSB format: 01001000 00101100 01101010 0000 01011001 0001 48 2C 6A 1E 59 3D This shows that no matter how the information is stored in memory it looks the same on the wire. So if it looks the same on the wire wouldn't an adapter pickup the packet and flip the bits in the byte if it needed to. Since it on the wire it looks like the bits are in non-canonical format a canonical format media would automatically take the first byte and flip the bits and so on, or so I would think. If anyone can figure out where I am going wrong please let me know. If it would be best to talk, email me directly with a daytime phone number and I will call you. Thanks. Neil ___ UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html FAQ, list archives, and subscription info: http://www.groupstudy.com Report misconduct and Nondisclosure violations to [EMAIL PROTECTED] --- ___ UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html FAQ, list archives, and subscription info: http://www.groupstudy.com Report misconduct and Nondisclosure violations to [EMAIL PROTECTED] ___ UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html FAQ, list archives, and subscription info: http://www.groupstudy.com Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
RE: canonical and non-canonical addresses
Neil, see if this helps...
Canonical / non-Canonical, LSB / MSB address format.
The physical bitstream always, independent from the topology,
starts with the I/G-bit ("group"-bit).
The destination address of a SRF-frame, for example
1000 0001 0010 0011 0100 0101 0110 0111 1000 1001
is displayed in MSB-/ non-canonical format as
80 01 23 45 67 89
and in LSB- / canonical format as
01 80 C8 A2 E6 91
The source and the destination addresses are usually displayed by
FDDI and Ethernet in LSB-/canonical format, by Token-Ring in
MSB-/non-canonical format.
A typical MAC-address of a SK Concentrator II (FDDI), for example
00 00 5A 4F 01 89(LSB / canonical)
is in MSB / non-canonical
00 00 5A F2 80 91
and is physically transmitted like this:
0101 1010 0010 1000 1001 0001
|
first Bit at the cable
The various topologies handle this as shown in the table:
| Ethernet |FDDI | Token Ring
--
phys. transmission | canonical | non-canonical | non-canonical
protocol Layer | canonical | canonical | non-canonical
Attention: the MAC-address you can see on Ethernet and FDDI
is normally LSB-/canonical format.
The MAC-address you can see on Token-Ring is
normally MSB-/non-canonical format.
How to convert from canonical to non-canonical format?
Trick: Read the single bytes bitwise from right to left.
Example 1
change this (which is LSB- / canonical)
00 00 5A 41 1E C4
in Bits:
0101 1010 0100 0001 0001 1110 1100 0100
to (in Bits):
0101 1010 1000 0010 0111 1000 0010 0011
Hex (which is MSB- / non-canonical)
00 00 5A 82 78 23
Example 2
change this (which is LSB- / canonical)
00 00 EF 03 16 20
in Bits:
1110 0011 0001 0110 0010
to (in Bits):
0111 1100 0110 1000 0100
Hex (which is MSB- / non-canonical)
00 00 F7 C0 68 04
regards,
mohan
-Original Message-
From: Maness, Drew [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, August 16, 2000 8:59 AM
To: [EMAIL PROTECTED]
Subject: RE: canonical and non-canonical addresses
Neil,
I'm not sure what your question is. The bits on the wire are transmitted
the same 0100. Token-ring reads them one way (non-canonical) 0100,
and Ethernet reads them another way (canonical)0010. It is how the
network interface cards (NIC)/device drivers read the 0s and 1s that is
different.
I would be happy to help(since I have struggled with this myself) if you can
clarify your question
Regards,
Drew R. Maness, CCNP, CCDA, CCSI, MCSE
-Original Message-
From: Neil Desai [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, August 15, 2000 12:47 PM
To: [EMAIL PROTECTED]
Subject: Re: canonical and non-canonical addresses
Does any body know the answer to this one? Now I am getting a bit
frustrated. I have posted to this group only a couple of times and I never
get an answer. All I get is people asking me to look at the archives. I did
as much homework as possible before asking the group.
""Neil Desai"" [EMAIL PROTECTED] wrote in message
8n9llp$83c$[EMAIL PROTECTED]">news:8n9llp$83c$[EMAIL PROTECTED]...
I am having a problem understanding the issues between canonical and
non-canonical addressing. I understand that the bits are flipped within
the
byte. On page 32-33 of Interconnections Second Edition she gives the
example
of the address a2-41-42-59-31-51.
Canonical:
10100010 0101 0110 01011001 00110001 01010001
Non-Canonical:
01000101 1010 0110 10011010 10001100 10001010
If you look at this you can clearly see that the address in canonical
format
is not a group address (last bit of first byte is zero) but in
non-canonical
format it is a group address. At this point I can see a big problem
because
she also states:
".the group bit in addresses was defined not as "the most significant
bit"
or the "least significant bit" but rather as "the first bit on the wire."
Thus, an address that was a group address on 802.3 would not necessarily
look like a group address when transmitted on 802.5 because a different
bit
would be transmitted first."
Here is the confusion: In canonical format the least significant bit is
transmitted first and in non-canonical format the most significant bit is
transmitted first. So on the wire the 1's and 0's would be in the same
order. Here is an excerpt from RFC 2469:
The figure be
