[PATCH] D36199: [Inliner] Increase threshold for hot callsites without PGO.

[Chandler Carruth via Phabricator via llvm-commits]

chandlerc added a comment.

While the total size increase doesn't concern me much, the >10% code size growth in some benchmarks is a bit concerning. Do these benchmarks also improve performance? If so, then this might be fine in general (once the -Os behavior is fixed, see below). However, if the benchmarks that are growing in size by a lot aren't also getting faster, then it seems a hard sell at https://reviews.llvm.org/owners/package/2/ where we expect size increase to be at least generally associated with performance wins.

Naturally, this seems completely fine at https://reviews.llvm.org/owners/package/3/ either way.



================
Comment at: lib/Analysis/InlineCost.cpp:782-784
+        if (HotCallSiteThreshold) {
           DEBUG(dbgs() << "Hot callsite.\n");
+          Threshold = MaxIfValid(Threshold, HotCallSiteThreshold.getValue());
----------------
I think we need to avoid this for optForSize() functions.


https://reviews.llvm.org/D36199