Request for review: 7110152 assert(size_in_words <= (julong)max_jint) failed: no overflow
Vladimir Kozlov
vladimir.kozlov at oracle.com
Wed Nov 9 22:06:49 UTC 2011
Why not change it in all CollectedHeap methods?
Vladimir
Bengt Rutisson wrote:
>
> Hi all,
>
> Could I have a couple of reviews for this small change?
> http://cr.openjdk.java.net/~brutisso/7110152/webrev.01/
>
> The problem is that CollectedHeap::array_allocate() takes its size as an
> int instead of a size_t. This was exposed by the fix I made to 7102044
> recently.
>
> That change allows arrays to be max_jint long on 64 bit systems. This is
> correct since we can handle this size in a 64 bit size_t even when we
> add the object header. However, it will overflow an int.
>
> Here is a small reproducer that will trigger the assert mentioned in
> 7110152 when it is run on a 64 bit VM:
>
> public class MaxJIntArray {
> public static void main(String[] args) {
> final int MAX_JINT = 2147483647;
> double[] a = new double[MAX_JINT];
> System.out.println("Allocated a[" + a.length + "]");
> }
> }
>
> The fix is to remove the assert (since it is too strict) and changing
> CollectedHeap::array_allocate() to take a size_t instead of an int.
>
> CR:
> 7110152 assert(size_in_words <= (julong)max_jint) failed: no overflow
> http://monaco.sfbay.sun.com/detail.jsf?cr=7110152
>
> Bengt
More information about the hotspot-gc-dev
mailing list