| |||||||||||||
|
| |||||||||||||
Why not just use the SHA1 only?Why not just use the SHA1 only?Posted Nov 16, 2008 13:39 UTC (Sun) by nix (subscriber, #2304)In reply to: Why not just use the SHA1 only? by nevets Parent article: /dev/ksm: dynamic memory sharing
It only generates a hash of the first, what, 128 bytes of the page, so any
pages with the same leading 128 bytes will 'collide' (in the sense that the first 128 bytes are identical, but the rest are not).
(Log in to post comments)
Why not just use the SHA1 only? Posted Nov 18, 2008 4:30 UTC (Tue) by nevets (subscriber, #11875) [Link] From your friends at Wikipedia, there is an article on SHA-1SHA-1 (as well as SHA-0) produces a 160-bit digest from a message with a maximum length of (2^64 − 1) bits This looks like it can be much bigger than 128 bytes.
Why not just use the SHA1 only? Posted Nov 18, 2008 16:01 UTC (Tue) by nix (subscriber, #2304) [Link]
Yeah, but for speed reasons they're only hashing the first N bytes (I
think it's 128), rather than the whole page. It's a sensible tradeoff, I think.
Why not just use the SHA1 only? Posted May 23, 2010 9:47 UTC (Sun) by rafal.maj (guest, #66508) [Link]
This doesn't make sense, they are for sure hashing entire page.
Why not just use the SHA1 only? Posted May 23, 2010 10:57 UTC (Sun) by johill (subscriber, #25196) [Link] Well, but you clearly didn't read the code:
+#define PAGECMP_OFFSET 128
+#define PAGEHASH_SIZE (PAGECMP_OFFSET ? PAGECMP_OFFSET : PAGE_SIZE)
+/* hash the page */
+static void page_hash(struct page *page, unsigned char *digest)
+{
+ struct scatterlist sg;
+ struct hash_desc desc;
+
+ sg_init_table(&sg, 1);
+ sg_set_page(&sg, page, PAGEHASH_SIZE, 0);
+ desc.tfm = tfm;
+ desc.flags = 0;
+ crypto_hash_digest(&desc, &sg, PAGEHASH_SIZE, digest);
+}
and it does "for sure" make sense since it's just a way to speed up matching, and just using 128 bytes means much less loading from memory. Tuning the 128 might make sense, potentially.
|
|
||||||||||||