Firstly, the bitmap we use for replay protection was ending up with zero
authorLutz Jänicke <jaenicke@openssl.org>
Mon, 13 Oct 2008 06:43:06 +0000 (06:43 +0000)
committerLutz Jänicke <jaenicke@openssl.org>
Mon, 13 Oct 2008 06:43:06 +0000 (06:43 +0000)
commit4db3e88459e8f5e742f414e49eab50e5be53ebca
tree43e720d73defd36f30217c445341736cd021a94a
parentab073bad4fb950f84c02e8660a9c36647d7f476e
Firstly, the bitmap we use for replay protection was ending up with zero
length, so a _single_ pair of packets getting switched around would
cause one of them to be 'dropped'.

Secondly, it wasn't even _dropping_ the offending packets, in the
non-blocking case. It was just returning garbage instead.
PR: #1752
Submitted by: David Woodhouse <dwmw2@infradead.org>
ssl/d1_lib.c
ssl/d1_pkt.c