Skip to contentSkip to conversation

Ignore garbage in dell-wmi events

PATCH In review Mario Limonciello 3 participants 5 messages Apr 28, 2009 – Apr 29, 2009

Subsystems (2)
ACPI x86 platform drivers
lore
Share & tools
lore

Mario Limonciello

author
#0 lore raw

PATCH 1/1 Ignore garbage in dell-wmi events

Hi:

In debugging with some future machines that actually contain BIOS level support for dell-wmi, I've determined that the upper half of the data that comes back from wmi_get_event_data may sometimes contain extra information that isn't currently relevant when pulling scan codes out of the data. This causes dell-wmi to improperly respond to these events.

I'm attaching a patch (Exchange would munge up the line breaks otherwise).

Thanks,

Signature · 3 lines
Mario Limonciello
*Dell | Linux Engineering*
[email protected]
drivers/platform/x86/dell-wmi.c renamed from drivers/platform/x86/dell-wmi.c~ +2 -2
@@ -182,7 +182,7 @@
 
 	if (obj && obj->type == ACPI_TYPE_BUFFER) {
 		int *buffer = (int *)obj->buffer.pointer;
-		key = dell_wmi_get_entry_by_scancode(buffer[1]);
+		key = dell_wmi_get_entry_by_scancode(0xFFFF & buffer[1]);
 		if (key) {
 			input_report_key(dell_wmi_input_dev, key->keycode, 1);
 			input_sync(dell_wmi_input_dev);

@@ -190,7 +190,7 @@
 			input_sync(dell_wmi_input_dev);
 		} else
 			printk(KERN_INFO "dell-wmi: Unknown key %x pressed\n",
-			       buffer[1]);
+			       0xFFFF & buffer[1]);
 	}
 }
author
#2 lore raw

Sure. Do you want me to resend all of them to add that?

Quoted 6 lines · Matthew Garrett

Hi Mario,

Can you add your signed-off-by: for this?

Thanks,

Regards,

Signature · 3 lines
Mario Limonciello
*Dell | Linux Engineering*
[email protected]
Signed-off-by Mario Limonciello <[email protected]>

---

drivers/platform/x86/dell-wmi.c renamed from drivers/platform/x86/dell-wmi.c~ +2 -2
@@ -182,7 +182,7 @@
 
 	if (obj && obj->type == ACPI_TYPE_BUFFER) {
 		int *buffer = (int *)obj->buffer.pointer;
-		key = dell_wmi_get_entry_by_scancode(buffer[1]);
+		key = dell_wmi_get_entry_by_scancode(0xFFFF & buffer[1]);
 		if (key) {
 			input_report_key(dell_wmi_input_dev, key->keycode, 1);
 			input_sync(dell_wmi_input_dev);

@@ -190,7 +190,7 @@
 			input_sync(dell_wmi_input_dev);
 		} else
 			printk(KERN_INFO "dell-wmi: Unknown key %x pressed\n",
-			       buffer[1]);
+			       0xFFFF & buffer[1]);
 	}
 }
#4 lore raw
In reply to Mario Limonciello
Quoted 7 lines · On Tue, 28 Apr 2009 12:40:34 -0500 Mario Limonciello…

In debugging with some future machines that actually contain BIOS level support for dell-wmi, I've determined that the upper half of the data that comes back from wmi_get_event_data may sometimes contain extra information that isn't currently relevant when pulling scan codes out of the data. This causes dell-wmi to improperly respond to these events.

I'm attaching a patch (Exchange would munge up the line breaks otherwise).

Can you please send along a Signed-off-by: for this patch? Documentation/SubmittingPatches explains what it's all about.

Thanks.