[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <570FBC50.1000603@gmail.com>
Date: Thu, 14 Apr 2016 21:20:40 +0530
From: Sudip Mukherjee <sudipm.mukherjee@...il.com>
To: Felipe Balbi <balbi@...nel.org>,
Yoshihiro Shimoda <yoshihiro.shimoda.uh@...esas.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>
Subject: Re: [PATCH] usb: renesas_usbhs: fix signed-unsigned return
On Thursday 14 April 2016 04:25 PM, Felipe Balbi wrote:
>
> Hi,
>
> Yoshihiro Shimoda <yoshihiro.shimoda.uh@...esas.com> writes:
>>> From: Sudip Mukherjee
>>> Sent: Saturday, April 09, 2016 12:05 AM
>>>
>>> The return type of usbhsp_setup_pipecfg() was u16 but it was returning
>>> a negative value (-EINVAL). Instead lets return a pointer to u16 which
>>> will hold the value to be returned or in case of error, return the
>>> error code in ERR_PTR.
>>
>> Thank you for the patch!
>> I also think this usbhsp_setup_pipecfg() should return error code using correct variable type.
>>
>> However, I would like to avoid to use ERR_PTR and kmalloc() somehow because
>> I feel this patch is complex a little.
>> How about the usbhsp_setup_pipecfg() prototype is changed like the following?
>>
>> static int usbhsp_setup_pipecfg(struct usbhs_pipe *pipe,
>> int is_host, int dir_in, u16 *pipecfg);
>
> IMO, this makes much more sense.
Infact, I thought about both these ways while making the patch but
somehow I thought this one is a better but ofcourse the other way is
much simpler.
I will post v2.
regards
sudip
Powered by blists - more mailing lists