You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@harmony.apache.org by "Mikhail Fursov (JIRA)" <ji...@apache.org> on 2007/03/12 13:36:09 UTC
[jira] Created: (HARMONY-3365) [drlvm][win64]
Double.doubleToLongBits(double) produces incorrect results.
[drlvm][win64] Double.doubleToLongBits(double) produces incorrect results.
--------------------------------------------------------------------------
Key: HARMONY-3365
URL: https://issues.apache.org/jira/browse/HARMONY-3365
Project: Harmony
Issue Type: Bug
Components: DRLVM
Reporter: Mikhail Fursov
Priority: Critical
While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
Here is the test:
public static void main(String[] args) {
System.out.println(Double.doubleToLongBits(1));
}
RI prints:
4607182418800017408
Harmony (both JET and OPT)
-3689348814741910324
I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HARMONY-3365) [drlvm][win64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Pavel Pervov (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12480147 ]
Pavel Pervov commented on HARMONY-3365:
---------------------------------------
The source of "some benchmarks" not passing successfully is somewhere else.
> [drlvm][win64] Double.doubleToLongBits(double) produces incorrect results.
> --------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Priority: Critical
> Attachments: H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Assigned: (HARMONY-3365) [drlvm][win64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Gregory Shimansky (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Gregory Shimansky reassigned HARMONY-3365:
------------------------------------------
Assignee: Gregory Shimansky
> [drlvm][win64] Double.doubleToLongBits(double) produces incorrect results.
> --------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HARMONY-3365) [drlvm][win64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Mikhail Fursov (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12480074 ]
Mikhail Fursov commented on HARMONY-3365:
-----------------------------------------
Just checked that Linix EM64T build is not affected by this issue.
> [drlvm][win64] Double.doubleToLongBits(double) produces incorrect results.
> --------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Priority: Critical
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Pavel Rebriy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Pavel Rebriy updated HARMONY-3365:
----------------------------------
Attachment: add-on-fix.patch
Updated patch.
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: add-on-fix.patch, add-on-fix.patch, add-on-fix.patch, H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Closed: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Gregory Shimansky (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Gregory Shimansky closed HARMONY-3365.
--------------------------------------
VERIFIED
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: add-on-fix.patch, add-on-fix.patch, add-on-fix.patch, H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Gregory Shimansky (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Gregory Shimansky updated HARMONY-3365:
---------------------------------------
Summary: [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results. (was: [drlvm][win64] Double.doubleToLongBits(double) produces incorrect results.)
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HARMONY-3365) [drlvm][win64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Pavel Pervov (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Pavel Pervov updated HARMONY-3365:
----------------------------------
Attachment: H3365.patch
The patch which fixes passing float parameters to native methods.
> [drlvm][win64] Double.doubleToLongBits(double) produces incorrect results.
> --------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Priority: Critical
> Attachments: H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Pavel Rebriy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Pavel Rebriy updated HARMONY-3365:
----------------------------------
Attachment: add-on-fix.patch
For parsing variable index number doesn't need to read POINTER_SIZE_INT value.
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: add-on-fix.patch, H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Pavel Rebriy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Pavel Rebriy updated HARMONY-3365:
----------------------------------
Attachment: add-on-fix.patch
Grant license.
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: add-on-fix.patch, add-on-fix.patch, H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Pavel Rebriy (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12480667 ]
Pavel Rebriy commented on HARMONY-3365:
---------------------------------------
All changes are valid.
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: add-on-fix.patch, add-on-fix.patch, add-on-fix.patch, H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (HARMONY-3365) [drlvm][winx64]
Double.doubleToLongBits(double) produces incorrect results.
Posted by "Gregory Shimansky (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HARMONY-3365?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Gregory Shimansky resolved HARMONY-3365.
----------------------------------------
Resolution: Fixed
Patches applied at 517820. I've used the latest version of add-on-fix.patch. Please check that the patch is applied as expected.
> [drlvm][winx64] Double.doubleToLongBits(double) produces incorrect results.
> ---------------------------------------------------------------------------
>
> Key: HARMONY-3365
> URL: https://issues.apache.org/jira/browse/HARMONY-3365
> Project: Harmony
> Issue Type: Bug
> Components: DRLVM
> Reporter: Mikhail Fursov
> Assigned To: Gregory Shimansky
> Priority: Critical
> Attachments: add-on-fix.patch, add-on-fix.patch, add-on-fix.patch, H3365.patch
>
>
> While debugging DRLVM on WIN64 platform I've found that Double.doubleToLongBits(double) method produces incorect results and prevents some benchmarks to pass.
> Here is the test:
> public static void main(String[] args) {
> System.out.println(Double.doubleToLongBits(1));
> }
> RI prints:
> 4607182418800017408
> Harmony (both JET and OPT)
> -3689348814741910324
> I think the cause is a problem in native stubs and calling conventions on windows in DRLVM.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.