To fix a minor malleability found by Sergio Lerner (reported here:
https://bitcointalk.org/index.php?topic=8392.msg1245898#msg1245898)
The problem is that if (R,S) is a valid ECDSA signature for a given
message and public key, (R,-S) is also valid. Modulo N (the order
of the secp256k1 curve), this means that both (R,S) and (R,N-S) are
valid. Given that N is odd, S and N-S have a different lowest bit.
We solve the problem by forcing signatures to have an even S value,
excluding one of the alternatives.
This commit just changes the signing code to always produce even S
values, and adds a verification mode to check it. This code is not
enabled anywhere yet. Existing tests in key_tests.cpp verify that
the produced signatures are still valid.
The length of vectors, maps, sets, etc are serialized using
Write/ReadCompactSize -- which, unfortunately, do not use a
unique encoding.
So deserializing and then re-serializing a transaction (for example)
can give you different bits than you started with. That doesn't
cause any problems that we are aware of, but it is exactly the type
of subtle mismatch that can lead to exploits.
With this pull, reading a non-canonical CompactSize throws an
exception, which means nodes will ignore 'tx' or 'block' or
other messages that are not properly encoded.
Please check my logic... but this change is safe with respect to
causing a network split. Old clients that receive
non-canonically-encoded transactions or blocks deserialize
them into CTransaction/CBlock structures in memory, and then
re-serialize them before relaying them to peers.
And please check my logic with respect to causing a blockchain
split: there are no CompactSize fields in the block header, so
the block hash is always canonical. The merkle root in the block
header is computed on a vector<CTransaction>, so
any non-canonical encoding of the transactions in 'tx' or 'block'
messages is erased as they are read into memory by old clients,
and does not affect the block hash. And, as noted above, old
clients re-serialize (with canonical encoding) 'tx' and 'block'
messages before relaying to peers.
Fixes issue#2838; this is a tweaked version of pull#2845 that
should not leak the length of the password and is more generic,
in case we run into other situations where we need
timing-attack-resistant comparisons.
Orphan transactions were stored as a CDataStream pointer;
this changes the mapOrphanTransactions data structures to
store orphans as a CTransaction.
This also fixes CVE-2013-4627 by always re-serializing
transactions before relaying them.
Gavin Andresen