Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S937079AbdIZLmN (ORCPT ); Tue, 26 Sep 2017 07:42:13 -0400 Received: from mail-wr0-f193.google.com ([209.85.128.193]:36476 "EHLO mail-wr0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751450AbdIZLmL (ORCPT ); Tue, 26 Sep 2017 07:42:11 -0400 X-Google-Smtp-Source: AOwi7QBRjqbJWGb7ZQkV/ixbaAsTZnC9ubam1iSVyO5xbpgXDoDB2qCXZxPvfYLXHz7TIR8dks1nOPO0Nq0Hl2UvMBg= MIME-Version: 1.0 In-Reply-To: <20170926113923.18181-1-net147@gmail.com> References: <20170926113923.18181-1-net147@gmail.com> From: Jonathan Liu Date: Tue, 26 Sep 2017 21:42:09 +1000 Message-ID: Subject: Re: [PATCH] usb: musb: sunxi: Explicitly release USB PHY on exit To: Bin Liu , Greg Kroah-Hartman , Maxime Ripard , Chen-Yu Tsai Cc: linux-usb@vger.kernel.org, linux-arm-kernel , linux-kernel , linux-sunxi , Jonathan Liu Content-Type: text/plain; charset="UTF-8" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2321 Lines: 65 Hi, On 26 September 2017 at 21:39, Jonathan Liu wrote: > This fixes a kernel oops when unloading the driver due to usb_put_phy > being called after usb_phy_generic_unregister when the device is > detached. Calling usb_phy_generic_unregister causes x->dev->driver to > be NULL in usb_put_phy and results in a NULL pointer dereference. > > As we are now explicitly managing the lifetime of usb_phy, > devm_usb_get_phy is changed to usb_get_phy and we call usb_put_phy in > the probe error path. > > Cc: stable@vger.kernel.org # v4.3+ > Signed-off-by: Jonathan Liu > --- > drivers/usb/musb/sunxi.c | 8 ++++++-- > 1 file changed, 6 insertions(+), 2 deletions(-) > > diff --git a/drivers/usb/musb/sunxi.c b/drivers/usb/musb/sunxi.c > index c9a09b5bb6e5..8eed61042103 100644 > --- a/drivers/usb/musb/sunxi.c > +++ b/drivers/usb/musb/sunxi.c > @@ -297,6 +297,8 @@ static int sunxi_musb_exit(struct musb *musb) > if (test_bit(SUNXI_MUSB_FL_HAS_SRAM, &glue->flags)) > sunxi_sram_release(musb->controller->parent); > > + usb_put_phy(glue->xceiv); > + > return 0; > } > > @@ -781,7 +783,7 @@ static int sunxi_musb_probe(struct platform_device *pdev) > return PTR_ERR(glue->usb_phy); > } > > - glue->xceiv = devm_usb_get_phy(&pdev->dev, USB_PHY_TYPE_USB2); > + glue->xceiv = usb_get_phy(USB_PHY_TYPE_USB2); > if (IS_ERR(glue->xceiv)) { > ret = PTR_ERR(glue->xceiv); > dev_err(&pdev->dev, "Error getting usb-phy %d\n", ret); > @@ -803,11 +805,13 @@ static int sunxi_musb_probe(struct platform_device *pdev) > if (IS_ERR(glue->musb_pdev)) { > ret = PTR_ERR(glue->musb_pdev); > dev_err(&pdev->dev, "Error registering musb dev: %d\n", ret); > - goto err_unregister_usb_phy; > + goto err_put_usb_phy; > } > > return 0; > > +err_put_usb_phy: > + usb_put_phy(glue->xceiv); > err_unregister_usb_phy: > usb_phy_generic_unregister(glue->usb_phy); > return ret; > -- > 2.13.2 > Another way I could fix this would be to check if x->dev->driver is NULL before dereferencing in usb_put_phy but I am not sure if this breaks any expectations in the USB subsystem... Regards, Jonathan