[PATCH] Post-b34 Fixes

[Andrew Hughes]

----- Original Message -----
> * Andrew Hughes <gnu.andrew at redhat.com> [2015-01-29 16:12]:
> > * OPENJDK6-51: Sync patch for 4873188 with 7 version
> > 
> > The backport of 4873188 (TLS 1.1 support) accidentally incorporated part
> > of 7031830 (see below) and placed a chunk of code in a slightly different
> > position to where it is in 7. Neither issue results in a problem, but
> > this changeset just makes the diff between 6 & 7 smaller.
> 
> I looked at the current code in jdk7 and it looks like this:
> 
>         // eventually compress as well.
>         HandshakeStatus hsStatus =
>                 writer.writeRecord(eor, ea, writeMAC, writeCipher);
> 
>         /*
>          * We only need to check the sequence number state for
>          * non-handshaking record.
>          *
>          * Note that in order to maintain the handshake status
>          * properly, we check the sequence number after the last
>          * record writing process. As we request renegotiation
>          * or close the connection for wrapped sequence number
>          * when there is enough sequence number space left to
>          * handle a few more records, so the sequence number
>          * of the last record cannot be wrapped.
>          */
>         if (connectionState < cs_ERROR && !isOutboundDone() &&
>                 (hsStatus == HandshakeStatus.NOT_HANDSHAKING)) {
>             if (checkSequenceNumber(writeMAC, eor.contentType())) {
>                 hsStatus = getHSStatus(null);
>             }
>         }
> 
>         /*
>          * turn off the flag of the first application record if we really
>          * consumed at least byte.
>          */
>         if (isFirstAppOutputRecord && ea.deltaApp() > 0) {
>             isFirstAppOutputRecord = false;
>         }
> 
>         return hsStatus;
> 
> The patch, however, changes the current placement in jdk6:
> 
>                  writer.writeRecord(eor, ea, writeMAC, writeCipher);
>  
>          /*
> +         * turn off the flag of the first application record if we really
> +         * consumed at least byte.
> +         */
> +        if (isFirstAppOutputRecord && ea.deltaApp() > 0) {
> +            isFirstAppOutputRecord = false;
> +        }
> +
> +        /*
>           * We only need to check the sequence number state for
>           * non-handshaking record.
>           *
> @@ -1279,14 +1287,6 @@
>              }
>          }
>  
> -        /*
> -         * turn off the flag of the first application record if we really
> -         * consumed at least byte.
> -         */
> -        if (isFirstAppOutputRecord && ea.deltaApp() > 0) {
> -            isFirstAppOutputRecord = false;
> -        }
> -
>          return hsStatus;
> 
> That looks like the current code in jdk6 matches what's in jdk7u and the
> patch
> makes it diverge. Is that right?
> 
> Everything else looks fine.
> 
> Thanks,
> Omair
> 
> --
> PGP Key: 66484681 (http://pgp.mit.edu/)
> Fingerprint = F072 555B 0A17 3957 4E95  0056 F286 F14F 6648 4681
> 

No. Good catch.

It seems IcedTea has both, thanks to the original security patch putting
it in one place, then the upstream changeset placing it in the other.

At least our original was right :) I'll post another webrev later.
-- 
Andrew :)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222

PGP Key: rsa4096/248BDC07 (hkp://keys.gnupg.net)
Fingerprint = EC5A 1F5E C0AD 1D15 8F1F  8F91 3B96 A578 248B DC07