RFR: [6904367]: (coll) IdentityHashMap is resized before exceeding the expected maximum size (original) (raw)

Peter Levart peter.levart at gmail.com
Tue Jul 8 10:12:30 UTC 2014

On 07/08/2014 09:33 AM, Martin Buchholz wrote:

I've updated the webrev http://cr.openjdk.java.net/~martin/webrevs/openjdk9/IdentityHashMap-capacity/ It now has all my TODOs done. The test case has been testng-ified.

Hi Martin,

What happened to the desire that when OOME is thrown on resizing, IHM is left unchanged?

Regards, Peter

On Mon, Jul 7, 2014 at 6:54 PM, Ivan Gerasimov <ivan.gerasimov at oracle.com> wrote:

Unfortunately, x + x << 1 causes the same overflow bug as 3*x: x + (x << 1) is merely supposed to be possibly more efficient than 3*x.

(int)(1431655766 + 1431655766 << 1) == 2 OK, I think my latest version doesn't have any overflow bugs.

More information about the core-libs-dev mailing list