linux-kernel - Re: [PATCH 2/5] Input: mtk-pmic-keys - Use regmap_{set,clear}

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: <You9SjGt4UjJk17/@google.com>
Date:   Mon, 23 May 2022 09:58:50 -0700
From:   Dmitry Torokhov <dmitry.torokhov@...il.com>
To:     AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@...labora.com>
Cc:     matthias.bgg@...il.com, mkorpershoek@...libre.com,
        linux-input@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        linux-mediatek@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/5] Input: mtk-pmic-keys - Use regmap_{set,clear}_bits
 where possible

On Mon, May 23, 2022 at 10:58:47AM +0200, AngeloGioacchino Del Regno wrote:
> Il 23/05/22 06:51, Dmitry Torokhov ha scritto:
> > On Fri, May 20, 2022 at 02:51:29PM +0200, AngeloGioacchino Del Regno wrote:
> > > Instead of always using regmap_update_bits(), let's go for the shorter
> > > regmap_set_bits() and regmap_clear_bits() where possible.
> > > 
> > > No functional change.
> > > 
> > > Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>
> > > ---
> > >   drivers/input/keyboard/mtk-pmic-keys.c | 24 ++++++------------------
> > >   1 file changed, 6 insertions(+), 18 deletions(-)
> > > 
> > > diff --git a/drivers/input/keyboard/mtk-pmic-keys.c b/drivers/input/keyboard/mtk-pmic-keys.c
> > > index 8e4fa7cd16e6..83d0b90cc8cb 100644
> > > --- a/drivers/input/keyboard/mtk-pmic-keys.c
> > > +++ b/drivers/input/keyboard/mtk-pmic-keys.c
> > > @@ -157,28 +157,16 @@ static void mtk_pmic_keys_lp_reset_setup(struct mtk_pmic_keys *keys,
> > >   	switch (long_press_mode) {
> > >   	case LP_ONEKEY:
> > > -		regmap_update_bits(keys->regmap, pmic_rst_reg,
> > > -				   MTK_PMIC_PWRKEY_RST,
> > > -				   MTK_PMIC_PWRKEY_RST);
> > > -		regmap_update_bits(keys->regmap, pmic_rst_reg,
> > > -				   MTK_PMIC_HOMEKEY_RST,
> > > -				   0);
> > > +		regmap_set_bits(keys->regmap, pmic_rst_reg, MTK_PMIC_PWRKEY_RST);
> > > +		regmap_clear_bits(keys->regmap, pmic_rst_reg, MTK_PMIC_HOMEKEY_RST);
> > 
> > Why not combine this into a single update instead? I.e. assuming
> > 
> 
> All downstream kernels (at least, I checked 4 different kernel versions for 4
> different SoCs) are doing these updates one-at-a-time, never combining them.

It is not like drivers in these downstream kernels were developed
separately and each of them discovered this as a requirement. It was
written once by someone and then either copied as is, or maybe
additional key handling was added later.

> 
> Even though I agree with you about one single update being simply more logical,
> I am afraid that (on some SoCs) the IP will not like that so - since I don't have
> any *clear* documentation saying that this is possible, or that this is not, I
> would leave it like that.

If we go with that we will never be able to touch any of the hardware
interfaces, as I do not recall when spec would explicitly document every
register and call out that individual bits can be changed together in
one write.

Thanks.

-- 
Dmitry