[src/trunk]: src/sys/arch/xen/xen Set frame = NULL in XENPV whether DIAGNOSTI...

[riastradh <riastradh%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/b8fba6496e41
branches:  trunk
changeset: 933219:b8fba6496e41
user:      riastradh <riastradh%NetBSD.org@localhost>
date:      Fri May 22 17:44:05 2020 +0000

description:
Set frame = NULL in XENPV whether DIAGNOSTIC or not.

For Xen PV, the interrupt logic stores the relevant information in
the CPU's struct cpu_info, and CLKF_USERMODE and CLKF_PC don't use
the frame argument at all.  (For Xen PVH, we behave like normal x86
and use the frame as passed.)

DIAGNOSTIC shouldn't change the data flows that affect non-DIAGNOSTIC
logic, so let's just unconditionally set frame = NULL for XENPV.  If
this makes a difference, that's a bug -- and it's a bug that should
not be concealed by enabling DIAGNOSTIC.

ok bouyer

diffstat:

 sys/arch/xen/xen/xen_clock.c |  6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diffs (27 lines):

diff -r e181e4ba657e -r b8fba6496e41 sys/arch/xen/xen/xen_clock.c
--- a/sys/arch/xen/xen/xen_clock.c      Fri May 22 16:17:42 2020 +0000
+++ b/sys/arch/xen/xen/xen_clock.c      Fri May 22 17:44:05 2020 +0000
@@ -1,4 +1,4 @@
-/*     $NetBSD: xen_clock.c,v 1.6 2020/05/21 21:12:31 ad Exp $ */
+/*     $NetBSD: xen_clock.c,v 1.7 2020/05/22 17:44:05 riastradh Exp $  */
 
 /*-
  * Copyright (c) 2017, 2018 The NetBSD Foundation, Inc.
@@ -36,7 +36,7 @@
 #endif
 
 #include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: xen_clock.c,v 1.6 2020/05/21 21:12:31 ad Exp $");
+__KERNEL_RCSID(0, "$NetBSD: xen_clock.c,v 1.7 2020/05/22 17:44:05 riastradh Exp $");
 
 #include <sys/param.h>
 #include <sys/types.h>
@@ -681,7 +681,7 @@
        KASSERT(cpu_intr_p());
        KASSERT(cookie == ci);
 
-#if defined(DIAGNOSTIC) && defined(XENPV)
+#if defined(XENPV)
        frame = NULL; /* We use values cached in curcpu()  */
 #endif
 again: