Received: by 2002:ac0:98c7:0:0:0:0:0 with SMTP id g7-v6csp168030imd; Fri, 26 Oct 2018 06:50:27 -0700 (PDT) X-Google-Smtp-Source: AJdET5ec/rokiZEmuE1cQLJv/P25V2U4Hd9E17s9/XNFfBoMbf3xMNQCZDRe/5Xrz1wQgAg4AsGz X-Received: by 2002:a63:6dc8:: with SMTP id i191mr3454759pgc.215.1540561827421; Fri, 26 Oct 2018 06:50:27 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1540561827; cv=none; d=google.com; s=arc-20160816; b=RIr6zygm2Q/OCBsNn96pPIOxa5ki/8v4t0eGXVV9kIBu9NGqC1Og3fK55Zr+KSukNt k+JCaCFcU715nL3a/p6w/UnOzRG+pHWFdLOTthiW3y1eXOIw5/W7PaHGrHuLkpRtqgVc 2DSisrM2xDZUeKLfVZYA/IIrXyTOJPhlboRjxU6iOGEApnGt9yJ9NEs7ySN4C1tIBfnQ VOfjck+sYcnVad3qzBhCYSN0VJhlR0EM4YkAVGRvIepXrun/g8jOwbtPsK+B3/7jPA87 0XunOD9Lpk50MCJSglOuw7qHlYERJ/2zd10ThBndDNbVzLK6Mit52vOmvLY3GOzff4JA vshg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:content-transfer-encoding:mime-version :references:in-reply-to:message-id:date:subject:cc:to:from; bh=izXGjfY1sFCR6wR+g1aaCXDfniDFkjgLOTNIpFnmeHI=; b=JHrhOHKHTpl4EHecr5t5cnSnWcG6VWRqhFTGRDSyErIvpHKuIpWps12DYN+4OKS4DG 8vNdE1CY2Gzb4ZrUg6GpFjX0xNz/CjNIJhU8UMmv1AJjjC2amFb5hvkZw9ySWg74B7JE W3ar2PHRRjt7EwhDB9g1QtKtkWZYGD2o7bSOS3pQXaYKMFU2UHOgFAcz79D3UhuKpJEJ F+pIgznhSngwYesgF7TKzGiCgWNRgDOeYxjCzTu/Bz8X0txCnS3g5zQ9RWcZkhcG/FQ6 58gy5mYy66jsAPUZIzCMWpcnNQYWnzd1yBUUVn9gMzri5+prHcD9vkAQfBUsLgyXTl8t 7NZg== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id g11-v6si11228379plp.128.2018.10.26.06.50.11; Fri, 26 Oct 2018 06:50:27 -0700 (PDT) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727581AbeJZWZq (ORCPT + 99 others); Fri, 26 Oct 2018 18:25:46 -0400 Received: from mx2.suse.de ([195.135.220.15]:44840 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1727520AbeJZWZp (ORCPT ); Fri, 26 Oct 2018 18:25:45 -0400 X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (unknown [195.135.220.254]) by mx1.suse.de (Postfix) with ESMTP id 73EE8B052; Fri, 26 Oct 2018 13:48:36 +0000 (UTC) From: Nicolas Saenz Julienne To: stefan.wahren@i2se.com, eric@anholt.net, dave.stevenson@raspberrypi.org Cc: nsaenzjulienne@suse.de, linux-rpi-kernel@lists.infradead.org, gregkh@linuxfoundation.org, linux-arm-kernel@lists.infradead.org, devel@driverdev.osuosl.org, linux-kernel@vger.kernel.org Subject: [PATCH RFC 11/18] staging: vchiq_arm: use completions instead of semaphores Date: Fri, 26 Oct 2018 15:48:06 +0200 Message-Id: <20181026134813.7775-12-nsaenzjulienne@suse.de> X-Mailer: git-send-email 2.19.1 In-Reply-To: <20181026134813.7775-1-nsaenzjulienne@suse.de> References: <20181026134813.7775-1-nsaenzjulienne@suse.de> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org It is preferred in the kernel to avoid using semaphores to wait for events, as they are optimised for the opposite situation; where the common case is that they are available and may block only occasionally. FYI see this thread: https://lkml.org/lkml/2008/4/11/323. Also completions are semantically more explicit in this case. Signed-off-by: Nicolas Saenz Julienne --- .../interface/vchiq_arm/vchiq_arm.c | 58 ++++++++++--------- 1 file changed, 31 insertions(+), 27 deletions(-) diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c index 1cdfdb714abc..463d0eec2e96 100644 --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c @@ -44,7 +44,7 @@ #include #include #include -#include +#include #include #include #include @@ -121,9 +121,9 @@ typedef struct user_service_struct { int message_available_pos; int msg_insert; int msg_remove; - struct semaphore insert_event; - struct semaphore remove_event; - struct semaphore close_event; + struct completion insert_event; + struct completion remove_event; + struct completion close_event; VCHIQ_HEADER_T * msg_queue[MSG_QUEUE_SIZE]; } USER_SERVICE_T; @@ -138,8 +138,8 @@ struct vchiq_instance_struct { VCHIQ_COMPLETION_DATA_T completions[MAX_COMPLETIONS]; int completion_insert; int completion_remove; - struct semaphore insert_event; - struct semaphore remove_event; + struct completion insert_event; + struct completion remove_event; struct mutex completion_mutex; int connected; @@ -562,7 +562,8 @@ add_completion(VCHIQ_INSTANCE_T instance, VCHIQ_REASON_T reason, vchiq_log_trace(vchiq_arm_log_level, "%s - completion queue full", __func__); DEBUG_COUNT(COMPLETION_QUEUE_FULL_COUNT); - if (down_interruptible(&instance->remove_event) != 0) { + if (wait_for_completion_interruptible( + &instance->remove_event)) { vchiq_log_info(vchiq_arm_log_level, "service_callback interrupted"); return VCHIQ_RETRY; @@ -600,7 +601,7 @@ add_completion(VCHIQ_INSTANCE_T instance, VCHIQ_REASON_T reason, insert++; instance->completion_insert = insert; - up(&instance->insert_event); + complete(&instance->insert_event); return VCHIQ_SUCCESS; } @@ -673,7 +674,8 @@ service_callback(VCHIQ_REASON_T reason, VCHIQ_HEADER_T *header, } DEBUG_TRACE(SERVICE_CALLBACK_LINE); - if (down_interruptible(&user_service->remove_event) + if (wait_for_completion_interruptible( + &user_service->remove_event) != 0) { vchiq_log_info(vchiq_arm_log_level, "%s interrupted", __func__); @@ -705,7 +707,7 @@ service_callback(VCHIQ_REASON_T reason, VCHIQ_HEADER_T *header, } spin_unlock(&msg_queue_spinlock); - up(&user_service->insert_event); + complete(&user_service->insert_event); header = NULL; } @@ -745,7 +747,7 @@ static void close_delivered(USER_SERVICE_T *user_service) unlock_service(user_service->service); /* Wake the user-thread blocked in close_ or remove_service */ - up(&user_service->close_event); + complete(&user_service->close_event); user_service->close_pending = 0; } @@ -867,7 +869,7 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) if (status == VCHIQ_SUCCESS) { /* Wake the completion thread and ask it to exit */ instance->closing = 1; - up(&instance->insert_event); + complete(&instance->insert_event); } break; @@ -948,9 +950,9 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) instance->completion_remove - 1; user_service->msg_insert = 0; user_service->msg_remove = 0; - sema_init(&user_service->insert_event, 0); - sema_init(&user_service->remove_event, 0); - sema_init(&user_service->close_event, 0); + init_completion(&user_service->insert_event); + init_completion(&user_service->remove_event); + init_completion(&user_service->close_event); if (args.is_open) { status = vchiq_open_service_internal @@ -1007,7 +1009,8 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) has been closed until the client library calls the CLOSE_DELIVERED ioctl, signalling close_event. */ if (user_service->close_pending && - down_interruptible(&user_service->close_event)) + wait_for_completion_interruptible( + &user_service->close_event)) status = VCHIQ_RETRY; break; } @@ -1182,7 +1185,8 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) DEBUG_TRACE(AWAIT_COMPLETION_LINE); mutex_unlock(&instance->completion_mutex); - rc = down_interruptible(&instance->insert_event); + rc = wait_for_completion_interruptible( + &instance->insert_event); mutex_lock(&instance->completion_mutex); if (rc != 0) { DEBUG_TRACE(AWAIT_COMPLETION_LINE); @@ -1310,7 +1314,7 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) } if (ret != 0) - up(&instance->remove_event); + complete(&instance->remove_event); mutex_unlock(&instance->completion_mutex); DEBUG_TRACE(AWAIT_COMPLETION_LINE); } break; @@ -1350,8 +1354,8 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) do { spin_unlock(&msg_queue_spinlock); DEBUG_TRACE(DEQUEUE_MESSAGE_LINE); - if (down_interruptible( - &user_service->insert_event) != 0) { + if (wait_for_completion_interruptible( + &user_service->insert_event)) { vchiq_log_info(vchiq_arm_log_level, "DEQUEUE_MESSAGE interrupted"); ret = -EINTR; @@ -1373,7 +1377,7 @@ vchiq_ioctl(struct file *file, unsigned int cmd, unsigned long arg) user_service->msg_remove++; spin_unlock(&msg_queue_spinlock); - up(&user_service->remove_event); + complete(&user_service->remove_event); if (header == NULL) ret = -ENOTCONN; else if (header->size <= args.bufsize) { @@ -1979,8 +1983,8 @@ vchiq_open(struct inode *inode, struct file *file) vchiq_debugfs_add_instance(instance); - sema_init(&instance->insert_event, 0); - sema_init(&instance->remove_event, 0); + init_completion(&instance->insert_event); + init_completion(&instance->remove_event); mutex_init(&instance->completion_mutex); mutex_init(&instance->bulk_waiter_list_mutex); INIT_LIST_HEAD(&instance->bulk_waiter_list); @@ -2033,12 +2037,12 @@ vchiq_release(struct inode *inode, struct file *file) /* Wake the completion thread and ask it to exit */ instance->closing = 1; - up(&instance->insert_event); + complete(&instance->insert_event); mutex_unlock(&instance->completion_mutex); /* Wake the slot handler if the completion queue is full. */ - up(&instance->remove_event); + complete(&instance->remove_event); /* Mark all services for termination... */ i = 0; @@ -2047,7 +2051,7 @@ vchiq_release(struct inode *inode, struct file *file) USER_SERVICE_T *user_service = service->base.userdata; /* Wake the slot handler if the msg queue is full. */ - up(&user_service->remove_event); + complete(&user_service->remove_event); vchiq_terminate_service_internal(service); unlock_service(service); @@ -2103,7 +2107,7 @@ vchiq_release(struct inode *inode, struct file *file) /* Wake any blocked user-thread */ if (instance->use_close_delivered) - up(&user_service->close_event); + complete(&user_service->close_event); unlock_service(service); } instance->completion_remove++; -- 2.19.1