Return-Path: Date: Wed, 14 Oct 2015 13:56:10 -0700 From: Bjorn Andersson To: yfw CC: Marcel Holtmann , Gustavo Padovan , Johan Hedberg , Andy Gross , Srinivas Kandagatla , "linux-arm-msm@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "linux-bluetooth@vger.kernel.org" Subject: Re: [PATCH v2 7/7] Bluetooth: btqcomsmd: Qualcomm WCNSS HCI driver Message-ID: <20151014205610.GL24668@usrtlx11787.corpusers.net> References: <1444423685-14717-1-git-send-email-bjorn.andersson@sonymobile.com> <1444423685-14717-8-git-send-email-bjorn.andersson@sonymobile.com> <561E7067.2070007@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" In-Reply-To: <561E7067.2070007@linaro.org> List-ID: On Wed 14 Oct 08:10 PDT 2015, yfw wrote: [..] > > diff --git a/drivers/bluetooth/Makefile b/drivers/bluetooth/Makefile [..] > > +static int btqcomsmd_probe(struct qcom_smd_device *sdev) > > +{ > > + struct qcom_smd_channel *acl; > > + struct btqcomsmd *btq; > > + struct hci_dev *hdev; > > + int ret; > > + > > + acl = qcom_smd_open_channel(sdev, > > + "APPS_RIVA_BT_ACL", > > + btqcomsmd_acl_callback); > > + if (IS_ERR(acl)) > > + return PTR_ERR(acl); > Do we need to close acl channel in following failure path and > btqcomsmd_remove? > As the life cycle of the channels implicitly follow the life cycle of the qcom_smd_device the channel is supposed to be released by the caller of probe upon failure. I'll make sure this is documented in qcom_smd_open_channel() Thanks for the review. Regards, Bjorn