Re: RFR: 8333334: C2: Make result of `Node::dominates` more precise to enhance scalar replacement [v3]

2024-06-10 Thread MaxXing
> This patch changes the algorithm of `Node::dominates` to make the result more > precise, and allows the iterators of `ConcurrentHashMap` to be scalar > replaced. > > The previous algorithm will return a conservative result when encountering a > dead control flow, and only try the first two in

Re: RFR: 8333334: C2: Make result of `Node::dominates` more precise to enhance scalar replacement [v2]

2024-06-06 Thread MaxXing
> This patch changes the algorithm of `Node::dominates` to make the result more > precise, and allows the iterators of `ConcurrentHashMap` to be scalar > replaced. > > The previous algorithm will return a conservative result when encountering a > dead control flow, and only try the first two in

Re: RFR: 8333334: C2: Make result of `Node::dominates` more precise to enhance scalar replacement [v2]

2024-06-06 Thread MaxXing
On Wed, 5 Jun 2024 05:40:12 GMT, Tobias Hartmann wrote: >> MaxXing has updated the pull request incrementally with one additional >> commit since the last revision: >> >> Revert last commit, and push the `LoadNode` back to the worklist to wait >> for the dead code to be removed. > > Impressi

Re: RFR: 8333334: C2: Make result of `Node::dominates` more precise to enhance scalar replacement

2024-06-04 Thread Tobias Hartmann
On Fri, 31 May 2024 09:01:38 GMT, MaxXing wrote: > This patch changes the algorithm of `Node::dominates` to make the result more > precise, and allows the iterators of `ConcurrentHashMap` to be scalar > replaced. > > The previous algorithm will return a conservative result when encountering a

RFR: 8333334: C2: Make result of `Node::dominates` more precise to enhance scalar replacement

2024-05-31 Thread MaxXing
This patch changes the algorithm of `Node::dominates` to make the result more precise, and allows the iterators of `ConcurrentHashMap` to be scalar replaced. The previous algorithm will return a conservative result when encountering a dead control flow, and only try the first two input paths of