Re: bin/56471: ntpd dies at startup on macppc/current, sshd randomly dies

[Rin Okuyama <rokuyama.rk%gmail.com@localhost>]

The following reply was made to PR bin/56471; it has been noted by GNATS.

From: Rin Okuyama <rokuyama.rk%gmail.com@localhost>
To: Jason Thorpe <thorpej%me.com@localhost>
Cc: "gnats-bugs%NetBSD.org@localhost" <gnats-bugs%NetBSD.org@localhost>
Subject: Re: bin/56471: ntpd dies at startup on macppc/current, sshd randomly
 dies
Date: Sun, 31 Oct 2021 00:12:57 +0900

 On 2021/10/30 23:41, Jason Thorpe wrote:
 > 
 >> On Oct 30, 2021, at 7:25 AM, Rin Okuyama <rokuyama.rk%gmail.com@localhost> wrote:
 >>
 >> "Sane" value for MSR varies between oea/40x/booke. Userland should not
 >> be bothered, and kernel should not accept any values from userland.
 > 
 > Well, I think it needs to at least accept round-tripping the FP modes and the tracing bits.  But yah, for initial value, just set to 0 and let the kernel initialize is the Right Thing.
 
 For FP mode bits, we emulate m[ft]msr instructions for fenv(3). These
 insns raise privilege exceptions if executed from userland. We catch
 the exceptions and emulate m[ft]msr in powerpc_machdep.c:emulate_mxmsr(),
 by which only FP mode bits can be retrieved/modified.
 
 Tracing bit (PSL_DE) for oea cannot be set directly from userland. It
 can be set only via ptrace(2). Also, 40x/booke do not have PSL_DE bit,
 and we use software single stepping for these subarchs.
 
 >> However, IIRC, once I tried to remove this macro, KASSERT fired for oea.
 > 
 > Which KASSERT fired?
 
 Hmm, sorry, I cannot find it. Maybe I added by myself for debugging in
 my local tree.
 
 Thanks,
 rin