efriedma-quic wrote: I'll restate:
On Windows, there are guard pages, so you have to grow the stack one page at a time. There are two ways to do this: __chkstk, or explicit memory accesses. __chkstk is faster for large allocations because it caches the size of the stack. LLVM currently relies explicit accesses for prologues that allocate less than one page, and __chkstk for all other allocations. But the ABI doesn't require that; we can use whatever mix we want. https://github.com/llvm/llvm-project/pull/149063 _______________________________________________ llvm-branch-commits mailing list llvm-branch-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
