From your post above, I thought the "read in 100MB chunks into the
big array" failed to work around the issue? Ie that's why you
allocated a tiny intermediate buffer? Did that turn out not to be
I verified that again with a 1.5.0_08 as well as the latest JVM on linux 32bit. Did not run the tests on windows yet but will do so tomorrow. Haven't had time to install the JVMs (or at least one) listed in the Sun Bug Report. Reading in chunks into a big array is fine.
How about, instead of ignoring the chunk size on 64bit jre, we
conditionally default it to Long.MAX_VALUE?
Hmm, would remove "code duplication" - good point. The readBytes method seems to be critical but I guess there won't be any performance impact by assigning one or two extra variables.
Can you link also to the Sun VM bug in the javadocs for these
Sure, I will also add CHANGES.TXT in the patch.
will do as soon as I get into the office tomorrow.