Hi David,
Thank you for the quick look and the fix!
Naoto
On 10/6/11 10:09 AM, David Holmes wrote:
Hi Chris,
Thanks. Here's the bug:
private List getJRELocales() {
if (availableJRELocales == null) {
synchronized (LocaleServiceProviderPool.class) {
if (availableJRELocales == null) {
Locale[]
D'oh! I should have seen this. Thanks David
-Chris
David Holmes wrote:
>Hi Chris,
>
>Thanks. Here's the bug:
>
>private List getJRELocales() {
> if (availableJRELocales == null) {
> synchronized (LocaleServiceProviderPool.class) {
> if (availableJRELocale
Hi Chris,
Thanks. Here's the bug:
private List getJRELocales() {
if (availableJRELocales == null) {
synchronized (LocaleServiceProviderPool.class) {
if (availableJRELocales == null) {
Locale[] allLocales = LocaleData.getAvailableLocales(
I filed CR 7098394 against this.
Solaris 10 x64 build fails since the integration of CR 7092679.
"../../../src/solaris/native/java/util/TimeZone_md.c", line 128:
prototype mismatch: 2 args passed, 3 expected
"../../../src/solaris/native/java/util/TimeZone_md.c", line 128:
warning: improper po
David,
Expanding (more threads and more runs) and running this test on one of
our dual core Linux x64 boxes, reproduce the CME about one in every ten
runs.
I instrumented where the CME was being created to determine the
expected/actual modcount:
final void checkForComodification() {
