From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id A6E79C4332F for ; Fri, 11 Nov 2022 12:27:23 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233015AbiKKM1W (ORCPT ); Fri, 11 Nov 2022 07:27:22 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:48340 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231615AbiKKM1V (ORCPT ); Fri, 11 Nov 2022 07:27:21 -0500 Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [45.249.212.187]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id CB77E1FFAB for ; Fri, 11 Nov 2022 04:27:19 -0800 (PST) Received: from kwepemi500024.china.huawei.com (unknown [172.30.72.55]) by szxga01-in.huawei.com (SkyGuard) with ESMTP id 4N7yd21fRrzmVpH; Fri, 11 Nov 2022 20:27:02 +0800 (CST) Received: from [10.174.179.163] (10.174.179.163) by kwepemi500024.china.huawei.com (7.221.188.100) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Fri, 11 Nov 2022 20:27:17 +0800 Message-ID: <8c484562-e1ae-346e-efd3-72dc2d450d41@huawei.com> Date: Fri, 11 Nov 2022 20:27:16 +0800 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101 Thunderbird/102.1.2 Subject: Re: [PATCH] iio: fix kobject_put warning in iio_device_register Content-Language: en-US From: Zeng Heng To: Jonathan Cameron CC: , , , , References: <20221110132615.331454-1-zengheng4@huawei.com> <20221111113141.00000917@Huawei.com> In-Reply-To: Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 8bit X-Originating-IP: [10.174.179.163] X-ClientProxiedBy: dggems703-chm.china.huawei.com (10.3.19.180) To kwepemi500024.china.huawei.com (7.221.188.100) X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: linux-iio@vger.kernel.org On 2022/11/11 20:05, Zeng Heng wrote: > > On 2022/11/11 19:31, Jonathan Cameron wrote: >> On Thu, 10 Nov 2022 21:26:15 +0800 >> Zeng Heng wrote: >> >>> There is warning reported by kobject lib in kobject_put(): >>> >>> kobject: '(null)' (00000000be81a546): is not initialized, yet >>> kobject_put() is being called. >>> WARNING: CPU: 0 PID: 535 at lib/kobject.c:718 kobject_put+0x12c/0x180 >>> Call Trace: >>>   cdev_device_add >>>   __iio_device_register >>>   __devm_iio_device_register >>>   tmp117_probe >>> >>> If don't need to register chardev for most of IIO devices, >>> we just register them with device_add() only, and use device_del() >>> to unregister them. >> >> >>> Otherwise, when device_add() fails in internal and calls kobject_put() >>> in error handling path, it would report warning because the device >>> never be registered as chardev and there is no release function for it. >>> >>> Fixes: 8ebaa3ff1e71 ("iio: core: register chardev only if needed") >>> Signed-off-by: Zeng Heng >> Interesting corner case. The cdev_device_add() call is fine with >> !dev->devt which is what this code was taking advantage of. The >> exception >> as you have highlighted is the error path of device_add(). >> >> So I think it should also cope with unwinding if device_add() fails >> and not be calling cdev_del()  Note that cdev_device_del() has the >> appropriate guards to be safe whether or not (dev->devt) is true. >> >> Perhaps change cdev_device_add() to have >> >>     rc = device_add(dev); >>     if (rc && dev->devt) >>         cdev_del(cdev); >> >>     return rc; > yes, I agree with your opinion about cdev_device_add(), which would be more flexible for callers. And I *just find* there is existing patch trying to fix this out. https://lore.kernel.org/lkml/Y1fmgCS7fuf%2FLQBc@kroah.com/ So just pass the patch. Best Regards, Zeng Heng >>> --- >>>   drivers/iio/industrialio-core.c | 11 +++++++++-- >>>   1 file changed, 9 insertions(+), 2 deletions(-) >>> >>> diff --git a/drivers/iio/industrialio-core.c >>> b/drivers/iio/industrialio-core.c >>> index 151ff3993354..f4f48bda07f7 100644 >>> --- a/drivers/iio/industrialio-core.c >>> +++ b/drivers/iio/industrialio-core.c >>> @@ -1982,7 +1982,11 @@ int __iio_device_register(struct iio_dev >>> *indio_dev, struct module *this_mod) >>>       /* assign device groups now; they should be all registered now */ >>>       indio_dev->dev.groups = iio_dev_opaque->groups; >>>   -    ret = cdev_device_add(&iio_dev_opaque->chrdev, &indio_dev->dev); >>> +    if (iio_dev_opaque->attached_buffers_cnt || >>> iio_dev_opaque->event_interface) >>> +        ret = cdev_device_add(&iio_dev_opaque->chrdev, >>> &indio_dev->dev); >>> +    else >>> +        ret = device_add(&indio_dev->dev); >>> + >>>       if (ret < 0) >>>           goto error_unreg_eventset; >>>   @@ -2008,7 +2012,10 @@ void iio_device_unregister(struct iio_dev >>> *indio_dev) >>>   { >>>       struct iio_dev_opaque *iio_dev_opaque = >>> to_iio_dev_opaque(indio_dev); >>>   -    cdev_device_del(&iio_dev_opaque->chrdev, &indio_dev->dev); >>> +    if (iio_dev_opaque->chrdev.kobj.state_initialized) >>> +        cdev_device_del(&iio_dev_opaque->chrdev, &indio_dev->dev); >>> +    else >>> +        device_del(&indio_dev->dev); >>>         mutex_lock(&iio_dev_opaque->info_exist_lock);