On 28 April 2011 12:01, Alan Bateman <alan.bate...@oracle.com> wrote: > I skimmed through the webrev and I agree this is a better approach. I will > try to do a detailed review before Monday. It would be great if others on > the list could jump in and help too as we are running out of time. > > Neil - I don't know if you've had a chance to look at Sherman's changes but > I think it's better than checking if mUTF-8 can be used. If you agree then > would you have time to run your tests that altered you to this performance > regression? There's a patch file in the webrev.
Hi Sherman, Thank you for responding so quickly to my post. I realise that my suggested change was quite restricted in its scope - it was designed so it could be considered easily (in the hope it might make it through the OpenJDK 7 gates before they finally clang shut). However, it's evident from your response that you have been thinking about possible enhancements in this area, and I agree that your proposed change addresses the issue in a more comprehensive - and strategic - manner. >From the testing I've been able to look at thus far, your proposed change seems to address the performance penalty which had caused me to raise the matter. Given this, I'd be delighted if it is adopted into OpenJDK 7, Thanks once again, Neil -- Unless stated above: IBM email: neil_richards at uk.ibm.com IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
