code review 170520043: hash/crc32: fix comment that the IEEE polynomial applie...

hash/crc32: fix comment that the IEEE polynomial applies to MPEG-2.

[nigeltao, 2014-11-12 05:48:12]

Hello adg@golang.org (cc: golang-codereviews@googlegroups.com),

I'd like you to review this change to
https://code.google.com/p/go/

[minux, 2014-11-12 07:10:56]

LGTM.

Isn't the CRC32 used in MPEG-2 just a reflected version of IEEE CRC32
(and no negate at the end)?

[nigeltao, 2014-11-12 07:47:15]

On Wed, Nov 12, 2014 at 6:10 PM, minux <minux@golang.org> wrote:
> Isn't the CRC32 used in MPEG-2 just a reflected version of IEEE CRC32
> (and no negate at the end)?

The motivating discussion is at
https://groups.google.com/forum/#!topic/golang-nuts/6HAJOaJOT9c

I think that the bit-reflection means we can't just use crc32.IEEE out
of the box, since we'd have to change
if crc&1 == 1 {
to
if crc&0x80000000 != 0 {
in the inner loop, amongst a few other things. It seems easier to make
any MPEG-2 Go code to write their own (trivial) CRC implementation.

[nigeltao, 2014-11-12 07:48:14]

*** Submitted as https://code.google.com/p/go/source/detail?r=775e14cff140 ***

hash/crc32: fix comment that the IEEE polynomial applies to MPEG-2.

LGTM=minux
R=adg, minux
CC=golang-codereviews
https://codereview.appspot.com/170520043