RFR: 8031639: make dependency management (mostly) ci independent (original) (raw)
Doug Simon doug.simon at oracle.com
Thu Jan 23 07:14:16 PST 2014
- Previous message: RFR: 8031639: make dependency management (mostly) ci independent
- Next message: RFR(XS): 8030662: "assert(counter_changed) failed: failed dependencies, but counter didn't change" still fails
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Jan 23, 2014, at 4:08 PM, Roland Westrelin <roland.westrelin at oracle.com> wrote:
I assume that lines 106 and 107 above are incomplete. Yes.
How would you compute an ident for these values and ensure the indents for each of the 3 DepValue types don’t clash with each other? Or are you relying on a Dependencies client to either be using depciobject (e.g., c1 and c2) or depmetadata and depjobject (e.g., Graal) and the client ensures that ident values are unique for all DepValues added to a Dependencies object? Yes, I would be relying on the fact that it’s either c1/c2 or graal.
Ok.
Maybe you could hum a few bars on the advantages of this approach? Is it mainly about avoiding all VMENTRY concerns? It tries to use the CI objects without breaking into them. Anyway, I don’t find my solution pretty either. I’d like someone else from the team to take a look at this.
Sure. Just let me know if there’s any further input I can give. Thanks.
-Doug
- Previous message: RFR: 8031639: make dependency management (mostly) ci independent
- Next message: RFR(XS): 8030662: "assert(counter_changed) failed: failed dependencies, but counter didn't change" still fails
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the hotspot-compiler-dev mailing list