linux-kernel - RE: [PATCH v1] ethernet: marvell: Use min macro

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: <7cfe24b82098487e9b1d35f964bf652f@AcuMS.aculab.com>
Date: Sat, 31 Aug 2024 12:39:28 +0000
From: David Laight <David.Laight@...LAB.COM>
To: 'Simon Horman' <horms@...nel.org>, Yan Zhen <yanzhen@...o.com>
CC: "marcin.s.wojtas@...il.com" <marcin.s.wojtas@...il.com>,
	"davem@...emloft.net" <davem@...emloft.net>, "edumazet@...gle.com"
	<edumazet@...gle.com>, "kuba@...nel.org" <kuba@...nel.org>,
	"pabeni@...hat.com" <pabeni@...hat.com>, "netdev@...r.kernel.org"
	<netdev@...r.kernel.org>, "linux-kernel@...r.kernel.org"
	<linux-kernel@...r.kernel.org>, "opensource.kernel@...o.com"
	<opensource.kernel@...o.com>
Subject: RE: [PATCH v1] ethernet: marvell: Use min macro

From: Simon Horman
> Sent: 27 August 2024 18:58
> 
> On Tue, Aug 27, 2024 at 06:54:08PM +0100, Simon Horman wrote:
> > On Tue, Aug 27, 2024 at 07:58:48PM +0800, Yan Zhen wrote:
> > > Using the real macro is usually more intuitive and readable,
> > > When the original file is guaranteed to contain the minmax.h header file
> > > and compile correctly.
> > >
> > > Signed-off-by: Yan Zhen <yanzhen@...o.com>
> > > ---
> > >  drivers/net/ethernet/marvell/mvneta.c | 3 +--
> > >  1 file changed, 1 insertion(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
> > > index d72b2d5f96db..415d2b9e63f9 100644
> > > --- a/drivers/net/ethernet/marvell/mvneta.c
> > > +++ b/drivers/net/ethernet/marvell/mvneta.c
> > > @@ -4750,8 +4750,7 @@ mvneta_ethtool_set_ringparam(struct net_device *dev,
> > >
> > >  	if ((ring->rx_pending == 0) || (ring->tx_pending == 0))
> > >  		return -EINVAL;
> > > -	pp->rx_ring_size = ring->rx_pending < MVNETA_MAX_RXD ?
> > > -		ring->rx_pending : MVNETA_MAX_RXD;
> > > +	pp->rx_ring_size = min(ring->rx_pending, MVNETA_MAX_RXD);
> >
> > Given that the type of ring->rx_pending is __32, and MVNETA_MAX_RXD is
> > a positive value.
> 
> Sorry, I hit send to soon. What I wanted to say is:
> 
> I think that it is appropriate to use umin() here.
> Because:
> 1) As I understand things, the type of MVNETA_MAX_RXD is signed,
>    but it always holds a positive value
> 2) ring->rx_pending is unsigned

Provided MVNETA_MAX_RXD is constant it is fine.
umin() is only needed for signed variables that can only contain
non-negative values.

You only need to use it is the compiler bleats...

umin(x, y) is safer than min_t(unsigned_type, x, y) because you can't
get the type wrong.
If will also generate better code since it never sign extends a
32bit value to 64bits (expensive on 32bit).

	David

> 
> > See: 80fcac55385c ("minmax: add umin(a, b) and umax(a, b)")
> >      https://git.kernel.org/torvalds/c/80fcac55385c

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)