[Haskell-cafe] rts nursery and stopping the world minor gc

[Brandon Allbery]

On Thu, Jun 8, 2017 at 3:45 PM, Brandon Allbery <allbery.b at gmail.com> wrote:

> If you now have multiple hardware threads, these can be running
> concurrently and allocating their own heap values which reference those
> shared values: for example, each is building its own list of cons cells
> referencing the value in different ways as specified by other values passed
> to the thread. Relocating the shared value from the nursery now requires
> updating all such thread-specific cons cells. Those heap values are also
> being promoted out of the nursery, but that doesn't change anything;
> copying them does not magically also update the copied things being pointed
> to, that is part of what GC does.
>

I should also point out that the fact that the cons cells for the other
thread are in a different nursery from the shared value *also* does not
change anything (aside from possibly complicating the implementation of the
garbage collector). They still must be updated to point to the new value,
so that thread must be quiescent/not actively accessing through the pointer
being updated.

This stuff *is* complicated; I've had to stop and think through the whole
thing a few times myself. And I might still have some things wrong. :/

-- 
brandon s allbery kf8nh                               sine nomine associates
allbery.b at gmail.com                                  ballbery at sinenomine.net
unix, openafs, kerberos, infrastructure, xmonad        http://sinenomine.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/haskell-cafe/attachments/20170608/3b7be693/attachment.html>