linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Julian Calaby <julian.calaby@gmail.com>
To: "Justin P. Mattock" <justinmattock@gmail.com>
Cc: linux-wireless@vger.kernel.org, j@w1.fi, linville@tuxdriver.com,
	mcgrof@gmail.com
Subject: Re: [PATCH v3]hostap:hostap_ioctl.c Fix variable 'ret' set but not used
Date: Tue, 3 Aug 2010 13:31:39 +1000	[thread overview]
Message-ID: <AANLkTinbSvn=kjkvf1VceH3oLsScT5KxtwhSdfGuy5_a@mail.gmail.com> (raw)
In-Reply-To: <4C5783AD.4060405@gmail.com>

On Tue, Aug 3, 2010 at 12:49, Justin P. Mattock <justinmattock@gmail.com> wrote:
> On 08/02/2010 07:31 PM, Julian Calaby wrote:
>>
>> On Tue, Aug 3, 2010 at 12:23, Justin P. Mattock<justinmattock@gmail.com>
>>  wrote:
>>>
>>> wait im lost.. looking at your patch looks the same(number wise) except I
>>> see different parts of code in there:
>>>
>>> hostap_set_word(dev, HFA384X_RID_CNFROAMINGMODE,
>>>                                HFA384X_ROAMING_FIRMWARE);
>>>
>>> -       return 0;
>>> +       return ret;
>>>
>>>
>>>
>>> mine is:
>>>
>>> printk(KERN_DEBUG "SCANREQUEST failed\n");
>>> -               ret = -EINVAL;
>>> +               return ret;
>>>
>>>
>>> The tree Im using is Linus's tree.
>>
>> I'm using wireless-testing, and the line you're changing is #1692, I'm
>> changing #1699.
>>
>> Thanks,
>>
>
>
> ahh.. so there's two return ret's that need to be added then..
> (if so then go ahead and take my patch and add it to yours.. I'm off to
> other things)

No.

The function works like this:

1. set up scan_req
2. if local->host_roaming is not set, set host roaming on the hardware
3. set scan_req on the hardware
4. if local->host_roaming is not set, set the hardware to do firmware roaming
5. If set_req could not be set on the hardware, return -EINVAL,
otherwise return 0.

The significant bits are steps 2 and 4. These must *both* be executed,
regardless of whether setting scan_req on the hardware succeeds or
not.

As such, the return value is stored in the ret variable, which is
initialised to 0 initially, then set to -EINVAL if setting scan_req on
the hardware fails. Hence, step 5 can be performed by returning ret.

As such, all of your patches are wrong as they are either returning
the wrong value when setting scan_req on the hardware fails, or
failing to execute step 4. The *only* correct place to return a value
is once the roaming status of the hardware has been set properly -
i.e. at the end of the function.

I hope this has cleared up your confusion.

Thanks,

-- 

Julian Calaby

Email: julian.calaby@gmail.com
.Plan: http://sites.google.com/site/juliancalaby/

  reply	other threads:[~2010-08-03  3:32 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-03  1:46 [PATCH v3]hostap:hostap_ioctl.c Fix variable 'ret' set but not used Justin P. Mattock
2010-08-03  1:58 ` Julian Calaby
2010-08-03  2:23   ` Justin P. Mattock
2010-08-03  2:31     ` Julian Calaby
2010-08-03  2:49       ` Justin P. Mattock
2010-08-03  3:31         ` Julian Calaby [this message]
2010-08-03  3:44           ` Justin P. Mattock

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='AANLkTinbSvn=kjkvf1VceH3oLsScT5KxtwhSdfGuy5_a@mail.gmail.com' \
    --to=julian.calaby@gmail.com \
    --cc=j@w1.fi \
    --cc=justinmattock@gmail.com \
    --cc=linux-wireless@vger.kernel.org \
    --cc=linville@tuxdriver.com \
    --cc=mcgrof@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).