Nice done. Thanks for finding the bug. I write most part of ideapad-laptop.c
though the maintainer is still Matthew.
I think the patch attached below is much simple and we can avoid unnecessary
read_ec_data if blightdev is not set. Since you are the one who find the bug
and have the first patch, I would like to keep you as author and your sign
off. Of course I need your agree. (Hoping I do not break the sign-off rule)
Please review the patch below and reply if you agree with this or disagree.
Hi Matthew,
When acpi_backlight=video, priv->blightdev will be NULL and use it without
checking will cause kernel oops. I suggest we shall have the fix in v3.1
cycle.
====== 8< ======
>From 8bd37cd5da31abaca4b906bc84eb45a0834678ea Mon Sep 17 00:00:00 2001
From: Rene Bolldorf <xsec...@googlemail.com>
Date: Wed, 24 Aug 2011 09:56:24 +0800
Subject: [PATCH] ideapad: Check if acpi already handle backlight power in
'ideapad_backlight_notify_power' to avoid a page fault
This patch avoid a page fault in the ideapad-laptop extras when
turning the backlight power on or off.
Signed-off-by: Rene Bolldorf <xsec...@googlemail.com>
Signed-off-by: Ike Panhc <ike....@canonical.com>
---
drivers/platform/x86/ideapad-laptop.c | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/drivers/platform/x86/ideapad-laptop.c
b/drivers/platform/x86/ideapad-laptop.c
index 0c59541..0d94eec 100644
--- a/drivers/platform/x86/ideapad-laptop.c
+++ b/drivers/platform/x86/ideapad-laptop.c
@@ -493,6 +493,8 @@ static void ideapad_backlight_notify_power(struct
ideapad_private *priv)
unsigned long power;
struct backlight_device *blightdev = priv->blightdev;
+ if (!blightdev)
+ return;
if (read_ec_data(ideapad_handle, 0x18, &power))
return;
blightdev->props.power = power ? FB_BLANK_UNBLANK : FB_BLANK_POWERDOWN;
--
1.7.5.4
--
To unsubscribe from this list: send the line "unsubscribe platform-driver-x86"
in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
