linux-kernel - Re: [PATCH 12/36] tty/vt: consolemap: use sizeof(*pointer) instead of sizeof(type)

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 for Android: free password hash cracker in your pocket

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

Message-ID: <b2df9240-2b47-6fe3-4a65-3bdfcb961e4@linux.intel.com>
Date:   Wed, 8 Jun 2022 10:23:45 +0300 (EEST)
From:   Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>
To:     Jiri Slaby <jslaby@...e.cz>
cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        linux-serial <linux-serial@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 12/36] tty/vt: consolemap: use sizeof(*pointer) instead
 of sizeof(type)

On Tue, 7 Jun 2022, Jiri Slaby wrote:

> On 07. 06. 22, 16:00, Ilpo Järvinen wrote:
> > On Tue, 7 Jun 2022, Jiri Slaby wrote:
> > 
> > > It is preferred to use sizeof(*pointer) instead of sizeof(type). First,
> > > the type of the variable can change and one needs not change the former
> > > (unlike the latter). Second, the latter is error-prone due to (u16),
> > > (u16 *), and (u16 **) mixture here.
> > > 
> > > Signed-off-by: Jiri Slaby <jslaby@...e.cz>
> > 
> > Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>
> > 
> > This seems fine but see the comments below which are not directly related
> > to the change itself.
> > 
> > > ---
> > >   drivers/tty/vt/consolemap.c | 23 ++++++++++++-----------
> > >   1 file changed, 12 insertions(+), 11 deletions(-)
> > > 
> > > diff --git a/drivers/tty/vt/consolemap.c b/drivers/tty/vt/consolemap.c
> > > index 097ab7d01f8b..79a62dcca046 100644
> > > --- a/drivers/tty/vt/consolemap.c
> > > +++ b/drivers/tty/vt/consolemap.c
> > > @@ -251,12 +251,12 @@ static void set_inverse_trans_unicode(struct vc_data
> > > *conp,
> > >   		return;
> > >   	q = p->inverse_trans_unicode;
> > >   	if (!q) {
> > > -		q = p->inverse_trans_unicode =
> > > -			kmalloc_array(MAX_GLYPH, sizeof(u16), GFP_KERNEL);
> > > +		q = p->inverse_trans_unicode = kmalloc_array(MAX_GLYPH,
> > > +				sizeof(*q), GFP_KERNEL);
> > >   		if (!q)
> > >   			return;
> > >   	}
> > > -	memset(q, 0, MAX_GLYPH * sizeof(u16));
> > > +	memset(q, 0, MAX_GLYPH * sizeof(*q));
> > 
> > Convert kmalloc_array into kcalloc and place memset() into else branch?
> 
> IMO, the way it is now is more obvious.

Fair enough.

> > > @@ -514,11 +514,12 @@ con_insert_unipair(struct uni_pagedict *p, u_short
> > > unicode, u_short fontpos)
> > >   	n = UNI_ROW(unicode);
> > >   	p2 = p1[n];
> > >   	if (!p2) {
> > > -		p2 = p1[n] = kmalloc_array(UNI_ROW_GLYPHS, sizeof(u16),
> > > GFP_KERNEL);
> > > +		p2 = p1[n] = kmalloc_array(UNI_ROW_GLYPHS, sizeof(*p2),
> > > +				GFP_KERNEL);
> > >   		if (!p2)
> > >   			return -ENOMEM;
> > >   		/* No glyphs for the characters (yet) */
> > > -		memset(p2, 0xff, UNI_ROW_GLYPHS * sizeof(u16));
> > > +		memset(p2, 0xff, UNI_ROW_GLYPHS * sizeof(*p2));
> > 
> > This could have been kcalloc'ed.
> 
> Why would you zero it before setting it to 0xff?

Yes, nevermind that. I obviously completely missed the fact it wasn't 
memsetting to zero.

-- 
 i.