-
Enhancement
-
Resolution: Fixed
-
P3
-
None
-
b18
-
itanium
-
generic
Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8008056 | 8 | Unassigned | P3 | Resolved | Fixed | b77 |
JDK-8180969 | openjdk7u | Unassigned | P3 | Resolved | Fixed | master |
There are about 40 #ifdef directives which mention IA64. (See attachment.)
Recently a probable bug was noticed around line 150 of oop.inline.hpp,
in IA64-specific code: It looks like the length value is garbage.
On the other hand, another bug was fixed around line 1500 in graphKit.cpp,
as part of 6470497, which had earlier (2002/11/05) been fixed only in an IA64 ifdef.
Failure to backport this fix led to a customer VM crash.
In both cases, it would probably have been better to eliminate or
at least minimize the number and volume of IA64-specific #ifdefs.
(And other #ifdefs for other platforms, too!)
Somebody needs to make a pass through the sources and vigorously
rethink, rewrite, and remove the platform-specific #ifdefs in our code.
Recently a probable bug was noticed around line 150 of oop.inline.hpp,
in IA64-specific code: It looks like the length value is garbage.
On the other hand, another bug was fixed around line 1500 in graphKit.cpp,
as part of 6470497, which had earlier (2002/11/05) been fixed only in an IA64 ifdef.
Failure to backport this fix led to a customer VM crash.
In both cases, it would probably have been better to eliminate or
at least minimize the number and volume of IA64-specific #ifdefs.
(And other #ifdefs for other platforms, too!)
Somebody needs to make a pass through the sources and vigorously
rethink, rewrite, and remove the platform-specific #ifdefs in our code.
- backported by
-
JDK-8008056 cleanup IA64 specific code in Hotspot
- Resolved
-
JDK-8180969 cleanup IA64 specific code in Hotspot
- Resolved
- relates to
-
JDK-8000412 Clean SA code on ia64
- Closed
-
JDK-8340731 Cleanup remaining IA64 references in hotspot code
- Open