I thought there was a LogCompilation alternative to PrintCompilation, maybe it behaves better? I've never tried logging XML with it as I don't want XML in my life but maybe it works around that issue. https://wiki.openjdk.org/display/HotSpot/LogCompilation+overview
On Thu, Aug 21, 2025, 5:03 PM Anton Hägerstrand <an...@blunders.io> wrote: > Fwiw, what I saw when digging into inlining last year was that the outputs > were rather hard to parse, printed only to stdout(or stderr), and big. For > lucenebench I would guess at least tens of thousands of lines. Iirc the > formatting etc is hardcoded into the jvm. > > Anton H > > On Thu, 21 Aug 2025, 20:07 Robert Muir, <rcm...@gmail.com> wrote: > >> I'm pretty sure running things with the additional debug information >> will probably cause slowdowns. Do we really care as long as the >> overhead is reasonably "fair"? >> >> On Thu, Aug 21, 2025 at 8:12 AM Adrien Grand <jpou...@gmail.com> wrote: >> > >> > This would be great, do you know if this can be done without slowdown? >> > >> > On Fri, Aug 15, 2025 at 9:30 PM Robert Muir <rcm...@gmail.com> wrote: >> >> >> >> I think the nightly benchmarks may need to get beefed up with some >> >> compiler output so we can better see what is happening? openjdk >> >> supports many flags to print the compilation/inlining details. >> >> >> >> On Fri, Aug 15, 2025 at 3:08 PM Adrien Grand <jpou...@gmail.com> >> wrote: >> >> > >> >> > Nightly benchmarks caught up with this change, and got significant >> slowdowns as expected: >> https://benchmarks.mikemccandless.com/2025.08.13.20.38.43.html. >> >> > >> >> > It'd be good to better understand why it hurts so much and how it >> could be fixed, help is welcome. :) >> >> > >> >> > On Wed, Aug 13, 2025 at 8:39 PM Adrien Grand <jpou...@gmail.com> >> wrote: >> >> >> >> >> >> Hello all, >> >> >> >> >> >> For reference, I just merged this PR[0] to luceneutil, which >> "pollutes" call sites of DocIdSetIterator#nextDoc, >> DocIdSetIterator#advance, Bits#get (live docs), SimScorer#score and others >> to better reflect what performance may look like on a production system >> that handles diverse workloads. >> >> >> >> >> >> A consequence is that luceneutil now report worse performance. In >> particular, nightly benchmarks are expected to get noticeable drops on many >> tasks. This behavior in enabled by default but it is possible to disable it >> by passing pollute=False when initializing a Competitor. >> >> >> >> >> >> [0] https://github.com/mikemccand/luceneutil/pull/436 >> >> >> >> >> >> -- >> >> >> Adrien >> >> > >> >> > >> >> > >> >> > -- >> >> > Adrien >> >> >> >> --------------------------------------------------------------------- >> >> To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org >> >> For additional commands, e-mail: dev-h...@lucene.apache.org >> >> >> > >> > >> > -- >> > Adrien >> >> --------------------------------------------------------------------- >> To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org >> For additional commands, e-mail: dev-h...@lucene.apache.org >> >>
