> From: Mark Thomas [mailto:ma...@apache.org]
> Subject: Re: getCurrentTimeMillis definition in JDBCAccessValve
>
> Note that there are some tests in the test tree you can use to see what
> impact your changes have on performance.
I would expect almost nil performance benefit fro
On 02/08/2010 23:33, Caldarale, Charles R wrote:
>> From: Mark Thomas [mailto:ma...@apache.org]
>> Subject: Re: getCurrentTimeMillis definition in JDBCAccessValve
>>
>> Speed. This way we only generate a new date string every 1s.
>
> But why generate th
> From: Mark Thomas [mailto:ma...@apache.org]
> Subject: Re: getCurrentTimeMillis definition in JDBCAccessValve
>
> Speed. This way we only generate a new date string every 1s.
But why generate the Date instance at all? Since System.currentTimeMillis() is
being called every t
On 02/08/2010 23:12, Channing Benson wrote:
> Hi,
>
> What am I missing here?
>
> In JDBCAccessValve, the method getCurrentTimeMillis is defined as
>
> public long getCurrentTimeMillis() {
> long systime = System.currentTimeMillis();
> if ((systime - currentTimeMillis) > 10
