By: Etienne Lorrain (etienne_lorrain.delete@this.yahoo.fr), May 18, 2021 3:11 am
Room: Moderated Discussions
Little Horn (sink.delete@this.example.net) on May 17, 2021 5:03 pm wrote:
> Thoughts?
Looks like they are assuming all their hardware thread share the same virtual->physical translation table, the state of each threads will be a lot bigger if you have more than one application running.
They have to define how they handle a switch to a new thread with exactly the same virtual memory config, a completely independent virtual memory config, and a partially different one (has just forked, copy on write, access application memory from kernel space...)
> Thoughts?
Looks like they are assuming all their hardware thread share the same virtual->physical translation table, the state of each threads will be a lot bigger if you have more than one application running.
They have to define how they handle a switch to a new thread with exactly the same virtual memory config, a completely independent virtual memory config, and a partially different one (has just forked, copy on write, access application memory from kernel space...)