linux-kernel - Re: [RESEND PATCH] staging: rtl8188eu: Fix else after return WARNING (checkpatch)

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] [day] [month] [year] [list]

Message-ID: <2f037e34bc5a32af66b2a306cc93dde2338cb960.camel@perches.com>
Date:   Tue, 15 Sep 2020 08:25:41 -0700
From:   Joe Perches <joe@...ches.com>
To:     Dan Carpenter <dan.carpenter@...cle.com>,
        Sebastian Arriola <sebdeveloper6952@...il.com>
Cc:     Sohom Datta <sohom.datta@...rner.manipal.edu>,
        devel@...verdev.osuosl.org,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-kernel@...r.kernel.org,
        Larry Finger <Larry.Finger@...inger.net>
Subject: Re: [RESEND PATCH] staging: rtl8188eu: Fix else after return
 WARNING (checkpatch)

On Tue, 2020-09-15 at 14:20 +0300, Dan Carpenter wrote:
> On Mon, Sep 14, 2020 at 09:42:49AM -0700, Joe Perches wrote:
> > On Mon, 2020-09-14 at 17:57 +0300, Dan Carpenter wrote:
> > > On Sun, Sep 13, 2020 at 12:19:50PM +0530, Sohom Datta wrote:
> > > > > From 4c8c8f3ff7f4d711daea4ac3bb987fcecc7ef1ed Mon Sep 17 00:00:00 2001
> > > > From: Sohom <sohom.datta@...rner.manipal.edu>
> > > > Date: Sat, 12 Sep 2020 18:04:56 +0530
> > > > Subject: [RESEND PATCH] staging: rtl8188eu: Fix else after return WARNING
> > > >  (checkpatch)
> > > > 
> > > > Fixed:
> > > > WARNING: else is not generally useful after a break or return
> > > > 1636: FILE: ./rtw_recv.c:1636:
> > > > +           return false;
> > > > +       else
> > > > 
> > > > Separated the return statement into a separate block since
> > > > it doesn't seem to depend on the SN_LESS explicity being false.
> > > > 
> > > > Signed-off-by: Sohom <sohom.datta@...rner.manipal.edu>
> > > > ---
> > > >  drivers/staging/rtl8188eu/core/rtw_recv.c | 5 +++--
> > > >  1 file changed, 3 insertions(+), 2 deletions(-)
> > > > 
> > > > diff --git a/drivers/staging/rtl8188eu/core/rtw_recv.c b/drivers/staging/rtl8188eu/core/rtw_recv.c
> > > > index 5fe7a0458dd2..5e81134ffb6d 100644
> > > > --- a/drivers/staging/rtl8188eu/core/rtw_recv.c
> > > > +++ b/drivers/staging/rtl8188eu/core/rtw_recv.c
> > > > @@ -1629,10 +1629,11 @@ static int enqueue_reorder_recvframe(struct recv_reorder_ctrl *preorder_ctrl,
> > > >  		hdr = list_entry(plist, struct recv_frame, list);
> > > >  		pnextattrib = &hdr->attrib;
> > > >  
> > > > +		if (SN_EQUAL(pnextattrib->seq_num, pattrib->seq_num))
> > > > +			return false;
> > > > +
> > > >  		if (SN_LESS(pnextattrib->seq_num, pattrib->seq_num))
> > > >  			plist = plist->next;
> > > > -		else if (SN_EQUAL(pnextattrib->seq_num, pattrib->seq_num))
> > > > -			return false;
> > > >  		else
> > > >  			break;
> > > >  	}
> > > 
> > > Checkpatch is just wrong here.  Ignore it when it's wrong.
> > 
> > It's not "wrong" here.  It's making a suggestion.
> > 
> > Perhaps read the SN_EQUAL and SN_LESS macros.
> > 
> > a and b are both u16's here.
> > 
> > drivers/staging/rtl8188eu/include/rtw_recv.h:#define SN_LESS(a, b)              (((a - b) & 0x800) != 0)
> > drivers/staging/rtl8188eu/include/rtw_recv.h:#define SN_EQUAL(a, b)     (a == b)
> > 
> > Reordering works, perhaps it's just a question of
> > whether it's the most likely result of the test.
> > 
> > This is in a while loop.
> > 
> > If the expected test is really the most likely that
> > SN_LESS is true, then perhaps this loop could be
> > something like:
> > 
> > 		if (SN_LESS(pnextattrib->seq_num, pattrib->seq_num)) {
> > 			plist = plist->next;
> > 			continue;
> > 		}
> > 		if (SN_EQUAL(pnextattrib->seq_num, pattrib->seq_num))
> > 			return false;
> > 		break;
> > 	}
> > 
> > The real question is whether or not that's more readable.
> > 
> 
> It's not clear to me that any of these are more readable than the other.
> 
> I see that someone broke the staging/rtl8712 version of this driver in
> June.  See commit 98fe05e21a6e ("staging: rtl8712: Remove unnecesary
> else after return statement.").  That patch went through LKML instead of
> going through the driver-devel list...  :/

That's sad.

Then another question is whether either is more prone
to unintentional breakage by novice code readers being
guided by brainless scripts...

A bit of a pity as the original intent of the checkpatch
test was somewhat useful.  Maybe it's outlived its value
though when used with -f files.

There aren't many of these left in the kernel.

Maybe it should be changed to work only on patches.