On Mon, 25 May 2026 04:09:10 GMT, Benjamin Peterson <[email protected]> wrote:

>> StubRoutines blobs bundle many individually-named stubs into a single 
>> BufferBlob, so perf profiles previously attributed all samples inside those 
>> blobs to the enclosing blob name (e.g. "StubRoutines (compiler stubs)"). 
>> Walk the StubCodeDesc list and emit a line per descriptor, so perf can name 
>> the actual stub.
>> 
>> 
>> 
>> 
>> ---------
>> - [x] I confirm that I make this contribution in accordance with the 
>> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai).
>
> Benjamin Peterson has updated the pull request with a new target base due to 
> a merge or a rebase. The incremental webrev excludes the unrelated changes 
> brought in by the merge/rebase. The pull request contains three additional 
> commits since the last revision:
> 
>  - use space instead of ::
>  - Merge branch 'master' into perfmap-stub-entries
>  - 8383709: Linux perf map should record individual stubs
>    
>    StubRoutines blobs bundle many individually-named stubs into a single 
> BufferBlob, so perf profiles previously attributed all samples inside those 
> blobs to the enclosing blob name (e.g. "StubRoutines (compiler stubs)"). Walk 
> the StubCodeDesc list and emit a line per descriptor, so perf can name the 
> actual stub.

src/hotspot/share/code/codeCache.cpp line 1980:

> 1978:     CodeBlob *cb = iter.method();
> 1979:     if (is_stub_code_blob(cb)) {
> 1980:       continue;

Please add some comments that stub code would be handled latter.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/31007#discussion_r3296330191

Reply via email to