[src/netbsd-1-4]: src ross's pullups that I did last night

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

details:   https://anonhg.NetBSD.org/src/rev/7a133a0ee5bd
branches:  netbsd-1-4
changeset: 469218:7a133a0ee5bd
user:      cgd <cgd%NetBSD.org@localhost>
date:      Sun Aug 08 20:48:41 1999 +0000

description:
ross's pullups that I did last night

diffstat:

 CHANGES-1.4.1 |  21 ++++++++++++++++++++-
 1 files changed, 20 insertions(+), 1 deletions(-)

diffs (32 lines):

diff -r 0973998b34b6 -r 7a133a0ee5bd CHANGES-1.4.1
--- a/CHANGES-1.4.1     Sun Aug 08 07:19:05 1999 +0000
+++ b/CHANGES-1.4.1     Sun Aug 08 20:48:41 1999 +0000
@@ -1,4 +1,4 @@
-#      $NetBSD: CHANGES-1.4.1,v 1.1.2.1 1999/08/08 07:00:04 cgd Exp $
+#      $NetBSD: CHANGES-1.4.1,v 1.1.2.2 1999/08/08 20:48:41 cgd Exp $
 
 Woefully incomplete partial list of changes from NetBSD 1.4 to NetBSD 1.4.1:
 
@@ -387,3 +387,22 @@
   do proper charging of mmap()'d space against the process's data size
   at this time.  Prevents unreasonably large allocations from from
   succeeding.  This is a band-aid for PR#7897.  (cgd)
+
+sys/dev/rnd.c                                  patch
+
+  Correctly protect critical section in rnd_extract_data().  (ross)
+
+sys/arch/alpha/pci/sio_pic.c                   1.24
+
+  Fix the 16-bytes-of-death bug by generating specific-EOI cycles
+  during sio_intr_setup().  (ross)
+
+sys/arch/alpha/alpha/dec_3000_300.c            1.28
+sys/arch/alpha/alpha/dec_3000_500.c            1.27
+sys/arch/alpha/tc/scc.c                                1.46
+sys/arch/alpha/tc/sccvar.h                     1.6
+
+  DEC 3000-series machines always use a serial console, even if the
+  firmware was using the display.  Tell users what's happening when
+  we switch, otherwise almost everyone just thinks the installation
+  has locked up.  (ross)