Re: README: ports switching to GCC 8.

[Jason Thorpe <thorpej%me.com@localhost>]

> On Dec 3, 2019, at 10:39 AM, Michael L. Hitch <mhitch%NetBSD.org@localhost> wrote:
> 
>  OK, I was able to track down the commit that causes this failure, and it
> was not obvious.  The commit was revision 1.69 of sys/dev/i2c/i2c.c.  This
> resulted in attempting to configure amdtemp0 instead of gpio0.  Reverting
> this change results in a working kernel - both netbsd-9 (gcc7) and
> -current (gcc8).

I'm not sure what the motivation for that change was:

----------------------------
revision 1.69
date: 2019-03-26 09:20:38 +0000;  author: mlelstv;  state: Exp;  lines: +8 -4;  commitid: NzJPOJHDHtCW7RgB;
touch match_result only on success.
----------------------------

It's clearly not correct.  The "true" or "false" return means "you should use the match_result that's returned".

-- thorpej