aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorH. Peter Anvin <hpa@zytor.com>2008-06-27 13:23:00 -0700
committerH. Peter Anvin <hpa@zytor.com>2008-06-27 13:29:01 -0700
commitaa60d13fb04f6d5ce72e4da508a4048b934ebd24 (patch)
tree85f58c1735c71f41579d942e9fdb15d86b15e0ad
parent23deb06821442506615f34bd92ccd6a2422629d7 (diff)
x86: setup: issue a null command after enabling A20 via KBC
Apparently, DOS and possibly other legacy operating systems issued a null command to the keyboard controller after toggling A20, specifically "pulse output pins" with no output pins specified. This was presumably done for synchronization reasons. This has made it into at least the UHCI spec, and it has been found to cause compatibility problems when "legacy USB" is enabled (which it almost always is) to not have this byte sent. It is *NOT* clear if any of these compatibility problems has any effect on Linux. However, for maximum compatibility, issue this null command after togging A20 through the KBC. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
-rw-r--r--arch/x86/boot/a20.c5
1 files changed, 4 insertions, 1 deletions
diff --git a/arch/x86/boot/a20.c b/arch/x86/boot/a20.c
index 90943f83e84..ef6e142a6b4 100644
--- a/arch/x86/boot/a20.c
+++ b/arch/x86/boot/a20.c
@@ -1,7 +1,7 @@
/* -*- linux-c -*- ------------------------------------------------------- *
*
* Copyright (C) 1991, 1992 Linus Torvalds
- * Copyright 2007 rPath, Inc. - All Rights Reserved
+ * Copyright 2007-2008 rPath, Inc. - All Rights Reserved
*
* This file is part of the Linux kernel, and is made available under
* the terms of the GNU General Public License version 2.
@@ -95,6 +95,9 @@ static void enable_a20_kbc(void)
outb(0xdf, 0x60); /* A20 on */
empty_8042();
+
+ outb(0xff, 0x64); /* Null command, but UHCI wants it */
+ empty_8042();
}
static void enable_a20_fast(void)