On 16/12/2015 6:23 AM, Martin Buchholz wrote:
On Mon, Dec 14, 2015 at 8:44 PM, David Holmes <david.hol...@oracle.com> wrote:
On 15/12/2015 6:53 AM, Martin Buchholz wrote:
Thread local storage is trouble.
java stack sizes should be in _addition_ to any OS overhead, which
includes TLS.
TLS shouldn't be coming out the stack of the thread AFAIK - I see nothing
about that in "ELF Handling for Thread Local Storage". That is why I want to
know more about the test, the compilation environment and the execution
environment.
I have also struggled to understand "ELF Handling for Thread Local Storage".
Do you mean that the value specified at the Java level should be rounded up
to accommodate guard pages etc at the native level?
Yes, I think so. All of the -Xss size should be available for actual
stack frames.
Alas no nice way to know what that value should be. I cringe at taking
the same approach that Rust had to do.
It's kind of interesting we've never hit this before. It's very
frustrating that we hit it on a thread that couldn't care less about the
TLS variables that have claimed space on its stack. I guess they need a
notion of "inheritable thread-locals" too.
David
