From patchwork Wed Nov 28 16:57:55 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hans de Goede X-Patchwork-Id: 1004673 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=linux-gpio-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=fail (p=none dis=none) header.from=redhat.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 434n086skFz9s1c for ; Thu, 29 Nov 2018 03:58:00 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728883AbeK2EAR (ORCPT ); Wed, 28 Nov 2018 23:00:17 -0500 Received: from mx1.redhat.com ([209.132.183.28]:16123 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728833AbeK2EAR (ORCPT ); Wed, 28 Nov 2018 23:00:17 -0500 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 346CE88E50; Wed, 28 Nov 2018 16:57:59 +0000 (UTC) Received: from shalem.localdomain.com (ovpn-116-176.ams2.redhat.com [10.36.116.176]) by smtp.corp.redhat.com (Postfix) with ESMTP id 9BF9B5C7A1; Wed, 28 Nov 2018 16:57:57 +0000 (UTC) From: Hans de Goede To: Mika Westerberg , Andy Shevchenko , Linus Walleij Cc: Hans de Goede , linux-gpio@vger.kernel.org, linux-acpi@vger.kernel.org, stable@vger.kernel.org Subject: [PATCH v2 1/2] gpiolib-acpi: Only defer request_irq for GpioInt ACPI event handlers Date: Wed, 28 Nov 2018 17:57:55 +0100 Message-Id: <20181128165756.20908-1-hdegoede@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.11 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.25]); Wed, 28 Nov 2018 16:57:59 +0000 (UTC) Sender: linux-gpio-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-gpio@vger.kernel.org Commit 78d3a92edbfb ("gpiolib-acpi: Register GpioInt ACPI event handlers from a late_initcall") deferred the entire acpi_gpiochip_request_interrupt call for each event resource. This means it also delays the gpiochip_request_own_desc(..., "ACPI:Event") call. This is a problem if some AML code reads the GPIO pin before we run the deferred acpi_gpiochip_request_interrupt, because in that case acpi_gpio_adr_space_handler() will already have called gpiochip_request_own_desc(..., "ACPI:OpRegion") causing the call from acpi_gpiochip_request_interrupt to fail with -EBUSY and we will fail to register an event handler. acpi_gpio_adr_space_handler is prepared for acpi_gpiochip_request_interrupt already having claimed the pin, but the other way around does not work. One example of a problem this causes, is the event handler for the OTG ID pin on a Prowise PT301 tablet not registering, keeping the port stuck in whatever mode it was in during boot and e.g. only allowing charging after a reboot. This commit fixes this by only deferring the request_irq call and the initial run of edge-triggered IRQs instead of deferring all of acpi_gpiochip_request_interrupt. Cc: stable@vger.kernel.org Fixes: 78d3a92edbfb ("gpiolib-acpi: Register GpioInt ACPI event ...") Signed-off-by: Hans de Goede Reviewed-by: Andy Shevchenko Acked-by: Mika Westerberg --- Changes in v2: -add kernel-doc describing struct acpi_gpio_event members (requested by Mika) -s/acpi_gpiochip_request_interrupt_desc/acpi_gpiochip_alloc_event/ (req Mika) -minor style fixups (requested by Andy) --- drivers/gpio/gpiolib-acpi.c | 144 +++++++++++++++++++++--------------- 1 file changed, 84 insertions(+), 60 deletions(-) diff --git a/drivers/gpio/gpiolib-acpi.c b/drivers/gpio/gpiolib-acpi.c index 55b72fbe1631..7f93954c58ea 100644 --- a/drivers/gpio/gpiolib-acpi.c +++ b/drivers/gpio/gpiolib-acpi.c @@ -19,11 +19,28 @@ #include "gpiolib.h" +/** + * struct acpi_gpio_event - ACPI GPIO event handler data + * + * @node: list-entry of the events list of the struct acpi_gpio_chip + * @handle: handle of ACPI method to execute when the IRQ triggers + * @handler: irq_handler to pass to request_irq when requesting the IRQ + * @pin: GPIO pin number on the gpio_chip + * @irq: Linux IRQ number for the event, for request_ / free_irq + * @irqflags: flags to pass to request_irq when requesting the IRQ + * @irq_is_wake: If the ACPI flags indicate the IRQ is a wakeup source + * @is_requested: True if request_irq has been done + * @desc: gpio_desc for the GPIO pin for this event + */ struct acpi_gpio_event { struct list_head node; acpi_handle handle; + irq_handler_t handler; unsigned int pin; unsigned int irq; + unsigned long irqflags; + bool irq_is_wake; + bool irq_requested; struct gpio_desc *desc; }; @@ -49,10 +66,10 @@ struct acpi_gpio_chip { /* * For gpiochips which call acpi_gpiochip_request_interrupts() before late_init - * (so builtin drivers) we register the ACPI GpioInt event handlers from a + * (so builtin drivers) we register the ACPI GpioInt IRQ handlers from a * late_initcall_sync handler, so that other builtin drivers can register their * OpRegions before the event handlers can run. This list contains gpiochips - * for which the acpi_gpiochip_request_interrupts() has been deferred. + * for which the acpi_gpiochip_request_irqs() call has been deferred. */ static DEFINE_MUTEX(acpi_gpio_deferred_req_irqs_lock); static LIST_HEAD(acpi_gpio_deferred_req_irqs_list); @@ -133,8 +150,42 @@ bool acpi_gpio_get_irq_resource(struct acpi_resource *ares, } EXPORT_SYMBOL_GPL(acpi_gpio_get_irq_resource); -static acpi_status acpi_gpiochip_request_interrupt(struct acpi_resource *ares, - void *context) +static void acpi_gpiochip_request_irq(struct acpi_gpio_chip *acpi_gpio, + struct acpi_gpio_event *event) +{ + int ret, value; + + ret = request_threaded_irq(event->irq, NULL, event->handler, + event->irqflags, "ACPI:Event", event); + if (ret) { + dev_err(acpi_gpio->chip->parent, + "Failed to setup interrupt handler for %d\n", + event->irq); + return; + } + + if (event->irq_is_wake) + enable_irq_wake(event->irq); + + event->irq_requested = true; + + /* Make sure we trigger the initial state of edge-triggered IRQs */ + value = gpiod_get_raw_value_cansleep(event->desc); + if (((event->irqflags & IRQF_TRIGGER_RISING) && value == 1) || + ((event->irqflags & IRQF_TRIGGER_FALLING) && value == 0)) + event->handler(event->irq, event); +} + +static void acpi_gpiochip_request_irqs(struct acpi_gpio_chip *acpi_gpio) +{ + struct acpi_gpio_event *event; + + list_for_each_entry(event, &acpi_gpio->events, node) + acpi_gpiochip_request_irq(acpi_gpio, event); +} + +static acpi_status acpi_gpiochip_alloc_event(struct acpi_resource *ares, + void *context) { struct acpi_gpio_chip *acpi_gpio = context; struct gpio_chip *chip = acpi_gpio->chip; @@ -143,8 +194,7 @@ static acpi_status acpi_gpiochip_request_interrupt(struct acpi_resource *ares, struct acpi_gpio_event *event; irq_handler_t handler = NULL; struct gpio_desc *desc; - unsigned long irqflags; - int ret, pin, irq, value; + int ret, pin, irq; if (!acpi_gpio_get_irq_resource(ares, &agpio)) return AE_OK; @@ -175,8 +225,6 @@ static acpi_status acpi_gpiochip_request_interrupt(struct acpi_resource *ares, gpiod_direction_input(desc); - value = gpiod_get_value_cansleep(desc); - ret = gpiochip_lock_as_irq(chip, pin); if (ret) { dev_err(chip->parent, "Failed to lock GPIO as interrupt\n"); @@ -189,64 +237,42 @@ static acpi_status acpi_gpiochip_request_interrupt(struct acpi_resource *ares, goto fail_unlock_irq; } - irqflags = IRQF_ONESHOT; + event = kzalloc(sizeof(*event), GFP_KERNEL); + if (!event) + goto fail_unlock_irq; + + event->irqflags = IRQF_ONESHOT; if (agpio->triggering == ACPI_LEVEL_SENSITIVE) { if (agpio->polarity == ACPI_ACTIVE_HIGH) - irqflags |= IRQF_TRIGGER_HIGH; + event->irqflags |= IRQF_TRIGGER_HIGH; else - irqflags |= IRQF_TRIGGER_LOW; + event->irqflags |= IRQF_TRIGGER_LOW; } else { switch (agpio->polarity) { case ACPI_ACTIVE_HIGH: - irqflags |= IRQF_TRIGGER_RISING; + event->irqflags |= IRQF_TRIGGER_RISING; break; case ACPI_ACTIVE_LOW: - irqflags |= IRQF_TRIGGER_FALLING; + event->irqflags |= IRQF_TRIGGER_FALLING; break; default: - irqflags |= IRQF_TRIGGER_RISING | - IRQF_TRIGGER_FALLING; + event->irqflags |= IRQF_TRIGGER_RISING | + IRQF_TRIGGER_FALLING; break; } } - event = kzalloc(sizeof(*event), GFP_KERNEL); - if (!event) - goto fail_unlock_irq; - event->handle = evt_handle; + event->handler = handler; event->irq = irq; + event->irq_is_wake = agpio->wake_capable == ACPI_WAKE_CAPABLE; event->pin = pin; event->desc = desc; - ret = request_threaded_irq(event->irq, NULL, handler, irqflags, - "ACPI:Event", event); - if (ret) { - dev_err(chip->parent, - "Failed to setup interrupt handler for %d\n", - event->irq); - goto fail_free_event; - } - - if (agpio->wake_capable == ACPI_WAKE_CAPABLE) - enable_irq_wake(irq); - list_add_tail(&event->node, &acpi_gpio->events); - /* - * Make sure we trigger the initial state of the IRQ when using RISING - * or FALLING. Note we run the handlers on late_init, the AML code - * may refer to OperationRegions from other (builtin) drivers which - * may be probed after us. - */ - if (((irqflags & IRQF_TRIGGER_RISING) && value == 1) || - ((irqflags & IRQF_TRIGGER_FALLING) && value == 0)) - handler(event->irq, event); - return AE_OK; -fail_free_event: - kfree(event); fail_unlock_irq: gpiochip_unlock_as_irq(chip, pin); fail_free_desc: @@ -283,6 +309,9 @@ void acpi_gpiochip_request_interrupts(struct gpio_chip *chip) if (ACPI_FAILURE(status)) return; + acpi_walk_resources(handle, "_AEI", + acpi_gpiochip_alloc_event, acpi_gpio); + mutex_lock(&acpi_gpio_deferred_req_irqs_lock); defer = !acpi_gpio_deferred_req_irqs_done; if (defer) @@ -293,8 +322,7 @@ void acpi_gpiochip_request_interrupts(struct gpio_chip *chip) if (defer) return; - acpi_walk_resources(handle, "_AEI", - acpi_gpiochip_request_interrupt, acpi_gpio); + acpi_gpiochip_request_irqs(acpi_gpio); } EXPORT_SYMBOL_GPL(acpi_gpiochip_request_interrupts); @@ -331,10 +359,13 @@ void acpi_gpiochip_free_interrupts(struct gpio_chip *chip) list_for_each_entry_safe_reverse(event, ep, &acpi_gpio->events, node) { struct gpio_desc *desc; - if (irqd_is_wakeup_set(irq_get_irq_data(event->irq))) - disable_irq_wake(event->irq); + if (event->irq_requested) { + if (event->irq_is_wake) + disable_irq_wake(event->irq); + + free_irq(event->irq, event); + } - free_irq(event->irq, event); desc = event->desc; if (WARN_ON(IS_ERR(desc))) continue; @@ -1200,23 +1231,16 @@ bool acpi_can_fallback_to_crs(struct acpi_device *adev, const char *con_id) return con_id == NULL; } -/* Run deferred acpi_gpiochip_request_interrupts() */ -static int acpi_gpio_handle_deferred_request_interrupts(void) +/* Run deferred acpi_gpiochip_request_irqs() */ +static int acpi_gpio_handle_deferred_request_irqs(void) { struct acpi_gpio_chip *acpi_gpio, *tmp; mutex_lock(&acpi_gpio_deferred_req_irqs_lock); list_for_each_entry_safe(acpi_gpio, tmp, &acpi_gpio_deferred_req_irqs_list, - deferred_req_irqs_list_entry) { - acpi_handle handle; - - handle = ACPI_HANDLE(acpi_gpio->chip->parent); - acpi_walk_resources(handle, "_AEI", - acpi_gpiochip_request_interrupt, acpi_gpio); - - list_del_init(&acpi_gpio->deferred_req_irqs_list_entry); - } + deferred_req_irqs_list_entry) + acpi_gpiochip_request_irqs(acpi_gpio); acpi_gpio_deferred_req_irqs_done = true; mutex_unlock(&acpi_gpio_deferred_req_irqs_lock); @@ -1224,4 +1248,4 @@ static int acpi_gpio_handle_deferred_request_interrupts(void) return 0; } /* We must use _sync so that this runs after the first deferred_probe run */ -late_initcall_sync(acpi_gpio_handle_deferred_request_interrupts); +late_initcall_sync(acpi_gpio_handle_deferred_request_irqs); From patchwork Wed Nov 28 16:57:56 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hans de Goede X-Patchwork-Id: 1004674 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=linux-gpio-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=fail (p=none dis=none) header.from=redhat.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 434n0B0M47z9s3Z for ; Thu, 29 Nov 2018 03:58:02 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728982AbeK2EAS (ORCPT ); Wed, 28 Nov 2018 23:00:18 -0500 Received: from mx1.redhat.com ([209.132.183.28]:34626 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728833AbeK2EAS (ORCPT ); Wed, 28 Nov 2018 23:00:18 -0500 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id BC5E530820F2; Wed, 28 Nov 2018 16:58:00 +0000 (UTC) Received: from shalem.localdomain.com (ovpn-116-176.ams2.redhat.com [10.36.116.176]) by smtp.corp.redhat.com (Postfix) with ESMTP id 75DF660123; Wed, 28 Nov 2018 16:57:59 +0000 (UTC) From: Hans de Goede To: Mika Westerberg , Andy Shevchenko , Linus Walleij Cc: Hans de Goede , linux-gpio@vger.kernel.org, linux-acpi@vger.kernel.org Subject: [PATCH v2 2/2] gpiolib-acpi: Remove unnecessary WARN_ON from acpi_gpiochip_free_interrupts Date: Wed, 28 Nov 2018 17:57:56 +0100 Message-Id: <20181128165756.20908-2-hdegoede@redhat.com> In-Reply-To: <20181128165756.20908-1-hdegoede@redhat.com> References: <20181128165756.20908-1-hdegoede@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.11 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.47]); Wed, 28 Nov 2018 16:58:00 +0000 (UTC) Sender: linux-gpio-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-gpio@vger.kernel.org acpi_gpiochip_alloc_event only continues allocating an event and adding it to the list if gpiochip_request_own_desc does not return an error. So events with an error desc are never placed on the events list and this check is really not necessary. Suggested-by: Andy Shevchenko Signed-off-by: Hans de Goede Acked-by: Mika Westerberg --- drivers/gpio/gpiolib-acpi.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/gpio/gpiolib-acpi.c b/drivers/gpio/gpiolib-acpi.c index 7f93954c58ea..091a3784720d 100644 --- a/drivers/gpio/gpiolib-acpi.c +++ b/drivers/gpio/gpiolib-acpi.c @@ -357,8 +357,6 @@ void acpi_gpiochip_free_interrupts(struct gpio_chip *chip) mutex_unlock(&acpi_gpio_deferred_req_irqs_lock); list_for_each_entry_safe_reverse(event, ep, &acpi_gpio->events, node) { - struct gpio_desc *desc; - if (event->irq_requested) { if (event->irq_is_wake) disable_irq_wake(event->irq); @@ -366,11 +364,8 @@ void acpi_gpiochip_free_interrupts(struct gpio_chip *chip) free_irq(event->irq, event); } - desc = event->desc; - if (WARN_ON(IS_ERR(desc))) - continue; gpiochip_unlock_as_irq(chip, event->pin); - gpiochip_free_own_desc(desc); + gpiochip_free_own_desc(event->desc); list_del(&event->node); kfree(event); }