On Fri, 23 Apr 2021 20:51:19 GMT, Alexey Bakhtin <abakh...@openjdk.org> wrote:
> Hello All, > > Could you please review the fix for the JDK-8241248? > The issue happens during the TLSv1.3 handshake without server stateless > session resumption in case of server receives several parallel requests with > the same pre_shared_key. > The main idea of the fix is to remove resuming session from the session cache > in the early stage. > > JBS: https://bugs.openjdk.java.net/browse/JDK-8241248 > Webrev: http://cr.openjdk.java.net/~abakhtin/8241248/webrev.v0/ > > The test from the bug report using OpenSSL is passed ( > -Djdk.tls.server.enableSessionTicketExtension=false ) > javax/net/ssl and sun/security/ssl jtreg tests passed > > Regards > Alexey src/java.base/share/classes/sun/security/ssl/PreSharedKeyExtension.java line 377: > 375: // If we are keeping state, see if the identity is > in the cache > 376: if (requestedId.identity.length == > SessionId.MAX_LENGTH) { > 377: synchronized (sessionCache) { Did you have a test if there is a performance regression by introducing the synchronization here? ------------- PR: https://git.openjdk.java.net/jdk/pull/3664
