[PATCH] Add a pass for inserting safepoints into (nearly) arbitrary IR
listmail at philipreames.com
Wed Jan 14 15:01:47 PST 2015
Hi whitequark, sanjoy, atrick, ributzka,
This pass is responsible for figuring out where to place call safepoints and safepoint polls. It doesn't actually make the relocations explicit; that's the job of the RewriteStatepointsForGC pass (http://reviews.llvm.org/D6975).
Points for review:
- I'd like to get this landed relatively quickly. This code is actually changing a bit as we work through optimization issues. For example, the SplitBackedge option was just added yesterday.
- I welcome suggestions on how to get rid of the nasty embedded pass manager trick. I think at this point that can be replaced with a more normal analysis dependency, but I don't believe having a loop analysis pass as a dependency for a Module pass works.
- Once landed, I plan on restructuring the statepoint rewrite to use the functions add to the IRBuilder a while back. I would prefer to do this as a separate change against TOT.
- In our version of this pass, it also handles inserting deoptimization state for the inserted safepoints. Since that code is not yet ready to be upstreamed, I ripped it out. This may get added at a later point.
- In the current pass, the function "gc.safepoint_poll" is treated specially but is not an intrinsic. I plan to make identifying the poll function a property of the GCStrategy at some point in the near future.
- It's not explicit in the code, but these two patches are introducing a new state for a statepoint which looks a lot like a patchpoint. There's no a transient form which doesn't yet have the relocations explicitly represented, but does prevent reordering of memory operations. Once this is in, I need to update actually make this explicit by reserving the 'unused' argument of the statepoint as a flag, updating the docs, and making the code explicitly check for such a thing. This wasn't really planned, but once I split the two passes - which was done for other reasons - the intermediate state fell out. Just reminds us once again that we need to merge statepoints and patchpoints at some point in the not that distant future.
Future directions planned:
- Identifying more cases where a backedge safepoint isn't required to ensure timely execution of a safepoint poll.
- Tweaking the insertion process to generate easier to optimize IR. (For example, investigating making SplitBackedge) the default.
- Adding opt-in flags for a GCStrategy to use this pass. Once done, add this pass to the actual pass ordering.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 43297 bytes
Desc: not available
More information about the llvm-commits