[llvm-commits] [PATCH] A fix for the inliner heuristics for recessive calls

[Duncan Sands]

Hi Nadav,

> I implemented the revised patch as part of the CallAnalyzer. The alloca-visitor collects the size of the statically allocated memory and we abort the inlining once we pass a threshold.  SROA is less of an issue because we only consider arrays, and the threshold is really high.

I didn't read the patch so I may be getting the wrong end of the stick here, but
this sounds odd.  Why only consider arrays?  A struct may contain arbitrarily
large arrays.  Also SROA works on arrays too.

Ciao, Duncan.