Code review request: 7105940, Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore

2011-10-28 Thread Xuelei Fan
Hi Weijun, Would you please review my fix for 7105940(Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore)? webrev: http://cr.openjdk.java.net/~xuelei/7105940/webrev.00/ Thanks, Xuelei

Re: Code review request: 7105940, Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore

2011-10-28 Thread Weijun Wang
CipherTest looks fine. What is the change in ClientJSSEServerJSSE for? Maybe you meant to remove the @ignore line? Thanks Weijun On 10/28/2011 09:19 PM, Xuelei Fan wrote: Hi Weijun, Would you please review my fix for 7105940(Test regression: KeyStore must be from provider SunPKCS11-NSSKeySto

Re: Code review request: 7105940, Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore

2011-10-28 Thread Xuelei Fan
Move forward the @run tag, so that it can be really ignored by @ignore tag. Otherwise, it will be run as normal. That's also why I can catch the exception. Thanks, Xuelei On 10/28/2011 9:53 PM, Weijun Wang wrote: > CipherTest looks fine. What is the change in ClientJSSEServerJSSE for? > Maybe you

Re: Code review request: 7105940, Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore

2011-10-28 Thread Weijun Wang
Oh, really? On my Linux it's always ignored. What did you mean "When run the test manually"? javac and java it? Anyway, the fix is correct. You can putback it. Thanks Max On 10/28/2011 09:55 PM, Xuelei Fan wrote: Move forward the @run tag, so that it can be really ignored by @ignore tag. Othe

Re: Code review request: 7105940, Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore

2011-10-28 Thread Xuelei Fan
On 10/28/2011 10:05 PM, Weijun Wang wrote: > Oh, really? On my Linux it's always ignored. > Yes, it's true. If I'm correct, I also address similar issues on other CRs. > What did you mean "When run the test manually"? javac and java it? > By remove @ignore tag, and jtreg. > Anyway, the fix is c

hg: jdk8/tl/jdk: 7105940: Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore

2011-10-28 Thread xuelei . fan
Changeset: 6e59c482e9b8 Author:xuelei Date: 2011-10-28 07:18 -0700 URL: http://hg.openjdk.java.net/jdk8/tl/jdk/rev/6e59c482e9b8 7105940: Test regression: KeyStore must be from provider SunPKCS11-NSSKeyStore Reviewed-by: weijun ! test/sun/security/pkcs11/fips/CipherTest.java ! test/

Review: 7105780: Add SSLSocket client/SSLEngine server to templates directory

2011-10-28 Thread Brad Wetmore
Hi Andrew, Wrapping up some loose ends. I was thinking it would be a good idea to put the test case for the recent Bad MAC error into the JSSE test template directory. It might be useful to have a SSLSocket client that can easily talk to a SSLEngine server. If we keep this test buried down

Review 7105792: Remove sun/security/pkcs11/Provider/Absolute.java from JPRT testing. No windows-x64 impl.

2011-10-28 Thread Brad Wetmore
Hi Valerie, http://cr.openjdk.java.net/~wetmore/7105792/ 7105792: Remove sun/security/pkcs11/Provider/Absolute.java from JPRT testing. No windows-x64 impl. As you know, there is a broken JDK test on windows-x64 under the jdk_security3 target. It doesn't compile due to a missing library P

Re: Review: 7105780: Add SSLSocket client/SSLEngine server to templates directory

2011-10-28 Thread Xuelei Fan
It's a good idea to move it to template directory. I did quick look at the code, looks fine to me. I did not read the code line by line carefully, hopefully, there is no significant changes from previous test. Just a minor suggest, it would be better if moving the @run tag at the bottom of the com

Code review request: 7106277 Brokenness in the seqNumberOverflow of MAC

2011-10-28 Thread Xuelei Fan
Hi, Would you please review my fix for 7106277 (Brokenness in the seqNumberOverflow of MAC)? webrev: http://cr.openjdk.java.net/~xuelei/7106277/webrev.00/ Thanks, Xuelei

Re: Code review request: 7106277 Brokenness in the seqNumberOverflow of MAC

2011-10-28 Thread Bradford Wetmore
Looks good. As you may remember, my personal preference is to use lots of parens to clearly show what you intended, but up to you as it's pretty clear without return ((block != null) && (mac != null) && (block[0] == (byte)0xFF) && (block[1] == (byte)0xFF) && ...

Re: Code review request: 7106277 Brokenness in the seqNumberOverflow of MAC

2011-10-28 Thread Xuelei Fan
On 10/29/2011 1:04 PM, Bradford Wetmore wrote: > Looks good. As you may remember, my personal preference is to use lots > of parens to clearly show what you intended, but up to you as it's > pretty clear without > > return ((block != null) && (mac != null) && > (block[0] == (b