Review Request for bug# 8006395 (original) (raw)
Edvard Wendelin edvard.wendelin at oracle.com
Wed Feb 6 07:38:12 PST 2013
- Previous message: Review Request for bug# 8006395
- Next message: [7u14] Request for approval for CR 7077259 - [TEST_BUG] [macosx] Test work correctly only when default L&F is Metal.
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Approved.
On Feb 6, 2013, at 12:41 PM, Chris Hegarty <chris.hegarty at oracle.com> wrote:
The changes look good to me.
-Chris. On 06/02/2013 11:21, mala bankal wrote: HI,
Request reviews for backport of bug# 8006395 to jdk7u : http://cr.openjdk.java.net/~mbankal/8006395/webrev.00/ Brief Description : Race in the socket asynchronous close mechanism on Linux which can lead to blocking socket operations issue is with closefd which first sends wakeup signal to all the threads blocked on the fd , then closes/dup2 the fd. But since signal may arrive in between these , the thread not blocked needs to see that fd is closed on entry, else it will be woken up by the signal.
JDK8 changeset is here : http://hg.openjdk.java.net/jdk8/tl/jdk/rev/17b643956999 rgds mala
- Previous message: Review Request for bug# 8006395
- Next message: [7u14] Request for approval for CR 7077259 - [TEST_BUG] [macosx] Test work correctly only when default L&F is Metal.
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]