#5476: Cap memory usage for one thread without disrupting other threads
---------------------------------+------------------------------------------
Reporter: joeyadams | Owner:
Type: feature request | Status: new
Priority: normal | Milestone: _|_
Component: Runtime System | Version: 7.2.1
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: None/Unknown
---------------------------------+------------------------------------------
Changes (by simonmar):
* milestone: => _|_
Comment:
The only way I can see to do this would be to evacuate one thread at a
time, completely tracing the reachable data before proceeding to the next
thread.
One problem is that threads are not actually treated as roots by the GC,
because the GC detects when a thread is deadlocked by its reachability.
Another problem is that tracing threads one at a time would be terrible
for parallelism in the GC, so at the least it would have to be a RTS
option.
Personally I think the right way to handle this is to use Cloud Haskell
and start up a new RTS instance for each of the client threads. That way
you get much better sandboxing too.
So I'll leave this open with a _|_ milestone, just in case anyone comes up
with a cheap way to implement it.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5476#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
