linux-kernel - Re: [PATCH 2/3] staging: r8188eu: simplify c2h_evt

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YSwVwKdJaj3EcW5V@equinox>
Date:   Mon, 30 Aug 2021 00:18:24 +0100
From:   Phillip Potter <phil@...lpotter.co.uk>
To:     Pavel Skripkin <paskripkin@...il.com>
Cc:     gregkh@...uxfoundation.org, straube.linux@...il.com,
        fmdefrancesco@...il.com, Larry.Finger@...inger.net,
        linux-staging@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3] staging: r8188eu: simplify c2h_evt_hdl function

On Sun, Aug 29, 2021 at 02:54:14PM +0300, Pavel Skripkin wrote:
> On 8/29/21 12:24 AM, Phillip Potter wrote:
> > Simplify c2h_evt_hdl function by removing majority of its code. The
> > function always returns _FAIL anyway, due to the wrapper function it
> > calls always returning _FAIL. For this reason, it is better to just
> > return _FAIL directly.
> > 
> > Leave the call to c2h_evt_read in place, as without it, event handling
> > semantics of the driver would be changed, despite nothing actually being
> > done with the event.
> > 
> > Signed-off-by: Phillip Potter <phil@...lpotter.co.uk>
> > ---
> >   drivers/staging/r8188eu/core/rtw_cmd.c | 21 +++------------------
> >   1 file changed, 3 insertions(+), 18 deletions(-)
> > 
> > diff --git a/drivers/staging/r8188eu/core/rtw_cmd.c b/drivers/staging/r8188eu/core/rtw_cmd.c
> > index ce73ac7cf973..b520c6b43c03 100644
> > --- a/drivers/staging/r8188eu/core/rtw_cmd.c
> > +++ b/drivers/staging/r8188eu/core/rtw_cmd.c
> > @@ -1854,27 +1854,12 @@ u8 rtw_c2h_wk_cmd(struct adapter *padapter, u8 *c2h_evt)
> >   static s32 c2h_evt_hdl(struct adapter *adapter, struct c2h_evt_hdr *c2h_evt, c2h_id_filter filter)
> >   {
> > -	s32 ret = _FAIL;
> >   	u8 buf[16];
> > -	if (!c2h_evt) {
> > -		/* No c2h event in cmd_obj, read c2h event before handling*/
> > -		if (c2h_evt_read(adapter, buf) == _SUCCESS) {
> > -			c2h_evt = (struct c2h_evt_hdr *)buf;
> > +	if (!c2h_evt)
> > +		c2h_evt_read(adapter, buf);
> > -			if (filter && !filter(c2h_evt->id))
> > -				goto exit;
> > -
> > -			ret = rtw_hal_c2h_handler(adapter, c2h_evt);
> > -		}
> > -	} else {
> > -		if (filter && !filter(c2h_evt->id))
> > -			goto exit;
> > -
> > -		ret = rtw_hal_c2h_handler(adapter, c2h_evt);
> > -	}
> > -exit:
> > -	return ret;
> > +	return _FAIL;
> 
> 
> Hi, Phillip!
> 
> 
> Do we really need to return _FAIL every time? The only one caller of
> c2h_evt_hdl() does not rely on it's return value. Shouldn't we make this
> function return void to simplify the code?

Dear Pavel,

Thanks for the review. Good point on the return type, I'll publish a v2
series.

> 
> 
> BTW, this function does nothing now, as I understand. It reads to local
> buffer and returns. I think, it can be removed
> 
> 
> 
> >   }
> >   static void c2h_wk_callback(struct work_struct *work)
> > 

Not sure if you're referring to c2h_wk_callback or c2h_evt_hdl, but
either way, they both call (indirectly or otherwise) c2h_evt_read and
c2h_evt_clear as well as rtw_c2h_wk_cmd in the case of c2h_wk_callback.
To just delete them wholesale therefore would be unsafe I think, due to
the effect on event semantics.

Certainly, it is possible to handle this c2h stuff by just reading from
register and ignoring though - another series is a better place for that
though I think. Admittedly, I may be talking nonsense here :-)

Regards,
Phil