[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20180305221833.GJ4197@saruman>
Date: Mon, 5 Mar 2018 22:37:36 +0000
From: James Hogan <jhogan@...nel.org>
To: Vasyl Gomonovych <gomonovych@...il.com>
Cc: ralf@...ux-mips.org, paul.gortmaker@...driver.com,
linux-mips@...ux-mips.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] MIPS: TXx9: Add missing iounmap
On Thu, Nov 23, 2017 at 10:49:55PM +0100, Vasyl Gomonovych wrote:
> Add the missing iounmap() before put_device and
> return from txx9_sramc_init().
>
> Signed-off-by: Vasyl Gomonovych <gomonovych@...il.com>
> ---
> arch/mips/txx9/generic/setup.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/arch/mips/txx9/generic/setup.c b/arch/mips/txx9/generic/setup.c
> index 1791a44ee570..6ef5edb85d68 100644
> --- a/arch/mips/txx9/generic/setup.c
> +++ b/arch/mips/txx9/generic/setup.c
> @@ -965,6 +965,8 @@ void __init txx9_sramc_init(struct resource *r)
> }
> return;
> exit_put:
> + if (dev->base)
> + iounmap(dev->base);
I think if dev->base was NULL then a different error path would have
been taken already.
More concerning to be honest though is the sysfs_create_bin_file() error
handling just above here, which seems to iounmap() and kfree() *after*
device_unregister(). txx9_device_release() already kfrees it.
The iounmap() call should prresumably move before the kfree() in
txx9_device_release(), and the kfree() after device_unregister() should
be removed?
Cheers
James
> put_device(&dev->dev);
> return;
> }
> --
> 1.9.1
>
Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)
Powered by blists - more mailing lists