RFR (S) 8078438: Interpreter should support conditional card marks (UseCondCardMark)

[Andrew Haley]

On 24/04/15 23:18, Aleksey Shipilev wrote:
> It would be nice to have the same change on other platforms as well. I
> wanted to get some early feedback/implementation in to probe the ground.
> The potential hit on highly-threaded machines (so, SPARC) is probably
> much greater than on low-threaded ones.

Right.  I was thinking in particular of Cavium's ThunderX with its 48
cores.  It'll be interesting to see what effect that has.

Andrew.