Arthur O'Dwyer

> I thought you added this term in this PR? Well, yes :) but only as a replacement for the phrase "the alternate search path" which has the same question...

> "include search path" is ambiguous; 19.2.2 and 19.2.3 both refer to a "search" being conducted for both and "" forms. I think the term could work if we qualify...

@apenn-msft: "...as you noted, there are two paths: _for the two lookups_ this is why the term "include search path" is ambiguous unless we know which form is being used....

Ah, I see that Xcode (and Clang via `-iquote`) support putting more things in the "user header search path" than just `.` (and in fact Clang has a dead codepath...

> These aren't new to this PR so they don't have to be addressed here. If we can fit them in great IIUC, this sentence referred only to the two...

@cubbimew: I really think "thread2 _could miss_ the notification" implies the wrong intuition. I want to avoid the implication that a notification could somehow "fail to arrive" thread2, as if...

https://quuxplusone.github.io/blog/2019/02/18/knightmare-of-initialization/#simple-guidelines-for-variable-i In particular, please don't write `for (auto i{ 0 }; i < 10; ++i)` when you could write `for (int i=0; i < 10; ++i)`.

Here's what I get (on Mac OSX): ``` $ make gcc -std=c99 -O3 -pedantic -W -Wall -Wextra \ -Wno-switch -Wno-unused-result \ advent.c -o advent ../vbcc/bin/vc +z -c99 -O1 -DZ_MACHINE -DSAVE_AND_RESTORE...

Interesting/scary. It would be nice to find out more about why it's being miscompiled by GCC. FWIW, I would not be terribly surprised by miscompilation on 32-bit platforms and/or big-endian...

@apthorpe: Thanks very much for this patch! It's _right at the boundary_ between "patches I want to accept" and "patches I don't." On the one hand it's kind of invasive...