RFR: 8335708: C2: Compile::verify_graph_edges must start at root and safepoints, just like CCP traversal [v2]

Marc Chevalier duke at openjdk.org
Tue Mar 18 13:57:34 UTC 2025


On Mon, 17 Mar 2025 06:40:38 GMT, Emanuel Peter <epeter at openjdk.org> wrote:

>> Marc Chevalier has updated the pull request incrementally with one additional commit since the last revision:
>> 
>>   various fixes
>
> test/hotspot/jtreg/compiler/loopopts/Test8335708.java line 1:
> 
>> 1: /*
> 
> Can you rename the test file? I think the new common practice is to give it a descriptive name, rather than just the bug number which is already tracked under `@bug 8335708` anyway ;)

Done. I think it's more explicit.

> test/hotspot/jtreg/compiler/loopopts/Test8335708.java line 28:
> 
>> 26:  * @bug 8335708
>> 27:  * @summary Crash Compile::verify_graph_edges
>> 28:  * @requires vm.debug == true & vm.flavor == "server"
> 
> Can we find a way not to have this restriction? It could make sense to still execute this in product, or with other compilers.
> 
> If the issues is with vm flags, then you can always use `-XX:+IgnoreUnrecognizedVMOptions`.

Done.

> test/hotspot/jtreg/compiler/loopopts/Test8335708.java line 35:
> 
>> 33:  *       -XX:+StressIGVN -Xcomp
>> 34:  *       -XX:CompileCommand=compileonly,compiler.loopopts.Test8335708::mainTest
>> 35:  *       compiler.loopopts.Test8335708
> 
> Can you please add a run without any flags? Sometimes that allows other bugs to trigger, because it can then be used without any flags, or other flag combinations.

Done. I hope it's the right way.

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/23977#discussion_r2001109017
PR Review Comment: https://git.openjdk.org/jdk/pull/23977#discussion_r2001106171
PR Review Comment: https://git.openjdk.org/jdk/pull/23977#discussion_r2001107606


More information about the hotspot-compiler-dev mailing list