From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751449AbeBVT6x (ORCPT ); Thu, 22 Feb 2018 14:58:53 -0500 Received: from esa4.dell-outbound.iphmx.com ([68.232.149.214]:48154 "EHLO esa4.dell-outbound.iphmx.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750816AbeBVT6v (ORCPT ); Thu, 22 Feb 2018 14:58:51 -0500 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A2ErAwAoII9ah2Oa6ERcGgEBAQEBAgEBA?= =?us-ascii?q?QEIAQEBAYUlKH0DmwmDGIN2kliCFgo3hH2CMFYWAQIBAQEBAQECAQIQAQEBCgs?= =?us-ascii?q?JCCgvgjgkAYJ1Uk9vARKKI61MOoh3gkmFGYIngVeBZ4srgyIFinKIeJBVCYJOk?= =?us-ascii?q?0AClEACmCqBPCYCggJwgxKCBAFPEAyCJVmMaAEBAQ?= X-IPAS-Result: =?us-ascii?q?A2ErAwAoII9ah2Oa6ERcGgEBAQEBAgEBAQEIAQEBAYUlKH0?= =?us-ascii?q?DmwmDGIN2kliCFgo3hH2CMFYWAQIBAQEBAQECAQIQAQEBCgsJCCgvgjgkAYJ1U?= =?us-ascii?q?k9vARKKI61MOoh3gkmFGYIngVeBZ4srgyIFinKIeJBVCYJOk0AClEACmCqBPCY?= =?us-ascii?q?CggJwgxKCBAFPEAyCJVmMaAEBAQ?= X-LoopCount0: from 10.208.86.39 X-IronPort-AV: E=Sophos;i="5.47,378,1515477600"; d="scan'208";a="621359774" X-DLP: DLP_GlobalPCIDSS From: Mario Limonciello To: dvhart@infradead.org, Andy Shevchenko Cc: LKML , platform-driver-x86@vger.kernel.org, Jeremy Cline , Alex Hung , Mario Limonciello Subject: [PATCH] platform/x86: Only activate tablet mode switch on 2-in-1's Date: Thu, 22 Feb 2018 13:58:42 -0600 Message-Id: <1519329522-16919-1-git-send-email-mario.limonciello@dell.com> X-Mailer: git-send-email 2.7.4 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Some laptops such as the XPS 9360 support the intel-vbtn INT33D6 interface but don't initialize the bit that intel-vbtn uses to represent switching tablet mode. By running this only on real 2-in-1's it shouldn't cause false positives. Fixes: 30323fb6d5 ("Support tablet mode switch") Reported-by: Jeremy Cline Signed-off-by: Mario Limonciello --- Note: I also moved the code for detecting tablet into it's own function. To keep things under 80 lines with if statements was making the code very unreadable otherwise. drivers/platform/x86/intel-vbtn.c | 46 ++++++++++++++++++++++++--------------- 1 file changed, 29 insertions(+), 17 deletions(-) diff --git a/drivers/platform/x86/intel-vbtn.c b/drivers/platform/x86/intel-vbtn.c index b703d6f..8173307 100644 --- a/drivers/platform/x86/intel-vbtn.c +++ b/drivers/platform/x86/intel-vbtn.c @@ -7,6 +7,7 @@ */ #include +#include #include #include #include @@ -97,9 +98,35 @@ static void notify_handler(acpi_handle handle, u32 event, void *context) dev_dbg(&device->dev, "unknown event index 0x%x\n", event); } -static int intel_vbtn_probe(struct platform_device *device) +static void detect_tablet_mode(struct platform_device *device) { + const char *chassis_type = dmi_get_system_info(DMI_CHASSIS_TYPE); + struct intel_vbtn_priv *priv = dev_get_drvdata(&device->dev); + acpi_handle handle = ACPI_HANDLE(&device->dev); struct acpi_buffer vgbs_output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object *obj; + acpi_status status; + int m; + + if (!(chassis_type && strcmp(chassis_type, "31") == 0)) + goto out; + + status = acpi_evaluate_object(handle, "VGBS", NULL, &vgbs_output); + if (ACPI_FAILURE(status)) + goto out; + + obj = vgbs_output.pointer; + if (!(obj && obj->type == ACPI_TYPE_INTEGER)) + goto out; + + m = !(obj->integer.value & TABLET_MODE_FLAG); + input_report_switch(priv->input_dev, SW_TABLET_MODE, m); +out: + kfree(vgbs_output.pointer); +} + +static int intel_vbtn_probe(struct platform_device *device) +{ acpi_handle handle = ACPI_HANDLE(&device->dev); struct intel_vbtn_priv *priv; acpi_status status; @@ -122,22 +149,7 @@ static int intel_vbtn_probe(struct platform_device *device) return err; } - /* - * VGBS being present and returning something means we have - * a tablet mode switch. - */ - status = acpi_evaluate_object(handle, "VGBS", NULL, &vgbs_output); - if (ACPI_SUCCESS(status)) { - union acpi_object *obj = vgbs_output.pointer; - - if (obj && obj->type == ACPI_TYPE_INTEGER) { - int m = !(obj->integer.value & TABLET_MODE_FLAG); - - input_report_switch(priv->input_dev, SW_TABLET_MODE, m); - } - } - - kfree(vgbs_output.pointer); + detect_tablet_mode(device); status = acpi_install_notify_handler(handle, ACPI_DEVICE_NOTIFY, -- 2.7.4