[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200422135215.GA25950@nuc8i5>
Date: Wed, 22 Apr 2020 21:52:15 +0800
From: Dejin Zheng <zhengdejin5@...il.com>
To: Andy Shevchenko <andy.shevchenko@...il.com>
Cc: kraxel@...hat.com, David Airlie <airlied@...ux.ie>,
Daniel Vetter <daniel@...ll.ch>,
virtualization@...ts.linux-foundation.org,
dri-devel <dri-devel@...ts.freedesktop.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v1] drm/bochs: fix an issue of ioremap() leak
On Tue, Apr 21, 2020 at 08:24:24PM +0300, Andy Shevchenko wrote:
> On Tue, Apr 21, 2020 at 7:45 PM Dejin Zheng <zhengdejin5@...il.com> wrote:
> >
> > It forgot to call bochs_hw_fini() to release related resources when
> > bochs_pci_probe() fail. eg: io virtual address get by ioremap().
>
> Good start, although I think the best is to switch this driver to use
> pcim_*() functions and drop tons of legacy code.
>
Andy, thanks for your encouragement, I think we might be able to fix this
issue first, after that, drop tons of legacy code by pcim_*() functions.
Do you think it is ok?
BR,
Dejin
> > Fixes: 81da8c3b8d3df6 ("drm/bochs: add drm_driver.release callback.")
> > CC: Andy Shevchenko <andy.shevchenko@...il.com>
> > Signed-off-by: Dejin Zheng <zhengdejin5@...il.com>
> > ---
> > drivers/gpu/drm/bochs/bochs_drv.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/drivers/gpu/drm/bochs/bochs_drv.c b/drivers/gpu/drm/bochs/bochs_drv.c
> > index addb0568c1af..210a60135c8a 100644
> > --- a/drivers/gpu/drm/bochs/bochs_drv.c
> > +++ b/drivers/gpu/drm/bochs/bochs_drv.c
> > @@ -138,6 +138,7 @@ static int bochs_pci_probe(struct pci_dev *pdev,
> > return ret;
> >
> > err_unload:
> > + bochs_hw_fini(dev);
> > bochs_unload(dev);
> > err_free_dev:
> > drm_dev_put(dev);
> > --
> > 2.25.0
> >
>
>
> --
> With Best Regards,
> Andy Shevchenko
Powered by blists - more mailing lists