[src/trunk]: src/doc Add hacks for GCC8 vs m68k ports:

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

details:   https://anonhg.NetBSD.org/src/rev/35f9fd4a6229
branches:  trunk
changeset: 974761:35f9fd4a6229
user:      rin <rin%NetBSD.org@localhost>
date:      Mon Aug 10 06:46:27 2020 +0000

description:
Add hacks for GCC8 vs m68k ports:

compile aes_ccm_tag() with -O0 for GCC8
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/crypto/aes/aes_ccm.c#rev1.5

compile kernel with -fno-omit-frame-pointer for GCC8
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/arch/sun2/conf/Makefile.sun2#rev1.25

diffstat:

 doc/HACKS |  28 +++++++++++++++++++++++++++-
 1 files changed, 27 insertions(+), 1 deletions(-)

diffs (39 lines):

diff -r 109853cf6fbb -r 35f9fd4a6229 doc/HACKS
--- a/doc/HACKS Mon Aug 10 06:33:45 2020 +0000
+++ b/doc/HACKS Mon Aug 10 06:46:27 2020 +0000
@@ -1,4 +1,4 @@
-# $NetBSD: HACKS,v 1.207 2020/05/10 06:38:24 maxv Exp $
+# $NetBSD: HACKS,v 1.208 2020/08/10 06:46:27 rin Exp $
 #
 # This file is intended to document workarounds for currently unsolved
 # (mostly) compiler bugs.
@@ -975,3 +975,29 @@
        only for earmv7hf{,eb} as far as I can see. Neither earmv6hf{,eb} nor
        earmv7{,eb} (softfloat) are affected.
 kcah
+
+port   m68k
+hack   compile aes_ccm_tag() with -O0 for GCC8
+cdate  Mon Aug 10 06:27:29 UTC 2020
+who    rin
+file   src/sys/crypto/aes/aes_ccm.c: 1.5
+descr  GCC 8.4 miscompiles aes_ccm_tag() for m68k with optimization level
+       -O[12], which results in failure in aes_ccm_selftest().
+       This is observed for amiga (A1200, 68060), mac68k (Quadra 840AV,
+       68040), and luna68k (nono, 68030 emulator). However, it is not for
+       sun3 (TME, 68020 emulator) and sun2 (TME, 68010 emulator).
+       At the moment, it is unclear whether this is due to differences b/w
+       68010-20 vs 68030-60, or something wrong with TME.
+kcah
+
+port   sun2
+hack   compile kernel with -fno-omit-frame-pointer for GCC8
+cdate  Mon Aug 10 06:28:42 UTC 2020
+who    rin
+file   src/sys/arch/sun2/conf/Makefile.sun2: 1.25
+descr  Work around for reproducible kernel freezes just after ``Starting
+       postfix.'', where I cannot even enter DDB nor obtain crash dump.
+       I still haven't figured out why. Possibly something wrong with -Os
+       optimization level for GCC/m68k, cf.,
+       http://mail-index.netbsd.org/port-sun3/2020/07/19/msg000166.html
+kcah