Hi Alan
There's been more than a week bug
https://jbs.oracle.com/bugs/browse/JDK-8000439
was reassigned to hotspot and the test still fails for latest nightly at
/net/sqenfs-1/export1/comp/vm/jdk/nightly/fastdebug/comp_baseline/solaris-sparcv9/
Therefore, I request it be added into Proble
On 22/10/2012 09:34, Weijun Wang wrote:
and here is webrev:
http://cr.openjdk.java.net/~weijun/8000624/webrev.00/
It make sense to exclude the test until the hotspot issue is resolved.
The change looks okay, assuming that "solaris-sparcv9" does indeed
exclude it.
-Alan
John,
Sorry for being later. Again, it's not to your changes but as well as
you are touching this code.
88, 103, 118: Type-o - we are checking for uid,gid,groups field,
but exception says "invalid field: username" in all cases.
It's better to fix it as well.
??: Does it make sense to ab
On 10/22/2012 04:38 PM, Alan Bateman wrote:
On 22/10/2012 09:34, Weijun Wang wrote:
and here is webrev:
http://cr.openjdk.java.net/~weijun/8000624/webrev.00/
It make sense to exclude the test until the hotspot issue is resolved.
The change looks okay, assuming that "solaris-sparcv9" doe
Changeset: e19dc885da0d
Author:weijun
Date: 2012-10-22 17:01 +0800
URL: http://hg.openjdk.java.net/jdk8/tl/jdk/rev/e19dc885da0d
8000624: Move MaxRetries.java to ProblemList for the moment
Reviewed-by: alanb
! test/ProblemList.txt
Dmitry:
I see what you mean, these error messages are quite misleading. I will change
all three of them.
I propose the following changes:
1.) line 88, change: "invalid field: username" to "invalid field: user ID". (In
Unix a user ID is a numerical value, not a "name" or character string.)
2.) li
Looks fine to me. I would suggest you add a few lines about the prefix
of KRB5CCNAME in the method comment.
Xuelei
On 10/22/2012 2:38 PM, Weijun Wang wrote:
> Please take a look at
>
>http://cr.openjdk.java.net/~weijun/8001208/webrev.00/
>
> An old test is enhanced to check for the fix.
>
Thanks. Will add some comments.
-Max
在 Oct 22, 2012，7:30 PM，Xuelei Fan 写道：
> Looks fine to me. I would suggest you add a few lines about the prefix
> of KRB5CCNAME in the method comment.
>
> Xuelei
>
> On 10/22/2012 2:38 PM, Weijun Wang wrote:
>> Please take a look at
>>
>> http://cr.open
> When "file ID equals zero" we return before the "ThrowNew()" call is
made. Should this call be moved to immediately before the "goto"
statement? I.e.
> if (fid == 0) {
> jclass newExcCls =
> (*env)->FindClass(env,
"java/lang/IllegalArgumentException");
So, the most good we can do here is to make the reasons for errors correct...
when the exceptions are thrown, be sure that the corresponding explanations
makes sense and are not misleading?
John
- Original Message -
From: chris.hega...@oracle.com
To: john.zavg...@oracle.com
Cc: dmitry.
On 22/10/2012 14:42, John Zavgren wrote:
So, the most good we can do here is to make the reasons for errors correct...
when the exceptions are thrown, be sure that the corresponding explanations
makes sense and are not misleading?
This is common practice where a Java library class some native
On Oct 22, 1:06pm, weijun.w...@oracle.com (Weijun Wang) wrote:
-- Subject: Re: bug fix for native kerberos libraries
| But are you using MEMORY: type ccache in your case? If I understand
| correctly, the substring after MEMORY: is not a normal file name. It's a
| tag that links to a block of by
On Oct 22, 2:38pm, weijun.w...@oracle.com (Weijun Wang) wrote:
-- Subject: Code review request: 8001208: Fix for KRB5CCNAME not complete
| Please take a look at
|
| http://cr.openjdk.java.net/~weijun/8001208/webrev.00/
|
| An old test is enhanced to check for the fix.
|
| Honestly, it migh
Chris,
On 2012-10-22 17:11, Chris Hegarty wrote:
> We should simply do:
> fid = (*env)->GetFieldID(env, cls, "uid", "J");
> if (fid == 0)
> goto cleanUpAndReturn;
>
> .. and forget the IAE lookup, etc..
I'm second for simple code above if now spec requires IAE here.
Also it's
Dmitry:
You recommend that this procedure check the return value of the GetField() call
and if there is an error (the return value is zero), clean up and return
immediately and not try to create an instance of the exception class?
The code that we have now will accept every possible return valu
John,
The existing/old code is actually wrong and would crash if the uid field
did not exist. What Dmitry said in his last mail is the best approach.
-Chris
On 22/10/2012 15:35, John Zavgren wrote:
Dmitry:
You recommend that this procedure check the return value of the GetField() call
and
Changeset: a1e77f7ed52b
Author:weijun
Date: 2012-10-23 10:02 +0800
URL: http://hg.openjdk.java.net/jdk8/tl/jdk/rev/a1e77f7ed52b
8001208: Fix for KRB5CCNAME not complete
Reviewed-by: xuelei
! src/share/classes/sun/security/krb5/internal/ccache/FileCredentialsCache.java
! test/sun/se
17 matches
Mail list logo
