[8u] RFR: 8218580: endpoint identification algorithm should be case-insensitive

[Hohensee, Paul]

Tip rev 52170 for 8208209 is http://hg.openjdk.java.net/jdk/jdk/rev/2990f1e1c325.
jdk11u rev 51320 for 820829 is http://hg.openjdk.java.net/jdk-updates/jdk11u/rev/781b5d8f2f75.
jdk8u-dev rev 13243 for 8202613 is http://hg.openjdk.java.net/jdk8u/jdk8u/jdk/rev/7f473886abb4.

PreSharedKeyExtension.java is part of the TLS 1.3 implementation, so the hunk I identified indeed doesn't apply.

Thanks,
Paul

On 11/22/19, 2:04 PM, "Verghese, Clive" <verghese at amazon.com> wrote: 

    Hi Paul,
    
    Thank you for reviewing the change. 
    
    The hunks in the TIP were added by JBS-Issue JDK-8208209 [rev 52170].
    8208209: Improve TLS connection stability again
    Reviewed-by: xuelei
    
    Though the same JBS issue was not backported,
    similar changeset exists in JDK-8 (JDK-8202613) [rev 13244].
    8202613: Improve TLS connections stability
    Reviewed-by: xuelei, wetmore
    
    It looks like the hunks on line 1036 does not apply to JDK-8
    
    Regards,
    Clive Verghese
    
    On 11/18/19, 3:56 PM, "Hohensee, Paul" <hohensee at amazon.com> wrote:
    
        If the code in the hunk that didn't apply is ever backported, context will be lost and incompatible code will result. Better to find the patch referenced by that hunk and backport it.
        
        Thanks,
        
        Paul
        
        On 10/31/19, 10:00 AM, "jdk8u-dev on behalf of Verghese, Clive" <jdk8u-dev-bounces at openjdk.java.net on behalf of verghese at amazon.com> wrote:
        
            Hi,
            
            I am requesting review for backport of JDK-8218580
            
            Webrev : http://cr.openjdk.java.net/~phh/8218580/webrev.8u.00/
            
            JBS Bug : https://bugs.openjdk.java.net/browse/JDK-8218580
            Original Change : http://hg.openjdk.java.net/jdk/jdk/rev/c34acb3a3330
            Change Set for JDK 11 : http://hg.openjdk.java.net/jdk-updates/jdk11u-dev/rev/148957581f5c
            
            The JBS Bug marks the issue as Introduced in 11. But checking the 8u code, it looks like the bug is effecting 8u as well.
            
            The backport to 8u did not apply cleanly. Other than the filename changes and line number changes, one of the notable changes is that only 2 of the 3 hunks applied in ClientHandshaker.java.
            
            The patch has passed tier-1 in Linux and no new regressions were found.
            
            Regards,
            Clive Verghese