On 02/14/06 Thong Nguyen wrote: > I've been getting some memory corruption running an application after the GC > runs. I've tracked it down to the allocation of primitive arrays (in > metadata/object.c) using GC_ALLOC_ATOMIC (or the mono equivalent). Now, > theoretically it should be fine, but changing it to a normal GC_ALLOC seems > to fix (or maybe hide) the issue.
It is very likely only hiding the issue. > Do mono objects contain links to any GC allocated memory? A primitive array > might not contain any managed references but the object header *may* contain > some pointers. I know at one point mono objects used to heap allocate > object monitors but that doesn't appear to be the case anymore so that > doesn't appear to be the issue... Right, the object header doesn't contain references to objects. > Any ideas? A test case that replicates the issue would be the best way to allow us to debug and see where the real problem is. lupus -- ----------------------------------------------------------------- [EMAIL PROTECTED] debian/rules [EMAIL PROTECTED] Monkeys do it better _______________________________________________ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list