RFR: 8173982: (fs) DefaultFileSystemProvider should be split into platform specific versions
Claes Redestad
claes.redestad at oracle.com
Mon Feb 6 21:56:04 UTC 2017
Hi Christoph,
oh, seems like the same pattern, yes.
I unfortunately pushed this just before I got your reply, and even
though it'd be a trivial fix to pile on, I think we must defer and do
this as an enhancement for 10 at this point since there is no
regression we could pin that on (AFAIK).
Filed: https://bugs.openjdk.java.net/browse/JDK-8174039
Thanks!
/Claes
On 2017-02-06 22:33, Langer, Christoph wrote:
> Hi Claes,
>
> this looks good.
>
> I'm wondering if the same should be done with sun/nio/ch/DefaultAsynchronousChannelProvider.java ?
>
> Best regards
> Christoph
>
>> -----Original Message-----
>> From: nio-dev [mailto:nio-dev-bounces at openjdk.java.net] On Behalf Of
>> Claes Redestad
>> Sent: Montag, 6. Februar 2017 14:09
>> To: Alan Bateman <Alan.Bateman at oracle.com>; nio-dev at openjdk.java.net
>> Subject: Re: RFR: 8173982: (fs) DefaultFileSystemProvider should be split into
>> platform specific versions
>>
>>
>>
>> On 02/06/2017 02:05 PM, Alan Bateman wrote:
>>> On 06/02/2017 13:02, Claes Redestad wrote:
>>>
>>>> Hi,
>>>>
>>>> sun.nio.fs.DefaultFileSystemProvider can now be implemented directly
>>>> for each platform, removing the need for using reflection.
>>>>
>>>> Since 9 the DefaultFileSystemProvider is initialized on startup, thus
>>>> this indirection can be shown to be cause for a small startup
>>>> regression on unix platforms relative to windows.
>>>>
>>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8173982
>>>> Webrev: http://cr.openjdk.java.net/~redestad/8173982/webrev.01/
>>> This looks okay to me.
>>>
>>
>> Thanks!
>>
>> //Claes
>>
>>> -Alan
>
More information about the nio-dev
mailing list