Re: [PATCH 1/5] wifi: rtl8xxxu: Add central frequency offset tracking

From: Johannes Berg <johannes@sipsolutions.net>
To: Bitterblue Smith <rtl8821cerfe2@gmail.com>,
	Ping-Ke Shih <pkshih@realtek.com>,
	"linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>
Cc: Jes Sorensen <Jes.Sorensen@gmail.com>
Subject: Re: [PATCH 1/5] wifi: rtl8xxxu: Add central frequency offset tracking
Date: Fri, 21 Oct 2022 22:22:14 +0200	[thread overview]
Message-ID: <d23613213568545ec808599a33c04a2e77ac7889.camel@sipsolutions.net> (raw)
In-Reply-To: <c3540f87-b124-2642-f53e-aa15704a54e8@gmail.com>

On Fri, 2022-10-21 at 21:32 +0300, Bitterblue Smith wrote:
> > 
> > #define XTAL1 GENMASK(22, 17)
> > #define XTAL0 GENMASK(16, 11)
> > 
> > val32 &= ~(XTAL1 | XTAL2)
> > val32 |= FIELD_PREP(XTAL1, crystal_cap) |
> >          FIELD_PREP(XTAL0, crystal_cap);
> > 
> 
> Ah, so that's what FIELD_PREP does.

FWIW, personally, I've come to prefer u32_encode_bits() and friends, or,
in this case perhaps

	u32p_replace_bits(&val32, crystal_cap, XTAL0);
	u32p_replace_bits(&val32, crystal_cap, XTAL1);

or maybe writing that as

val32 &= ~(XTAL0 | XTAL1);
val32 |= u32_encode_bits(crystal_cap, XTAL0);
val32 |= u32_encode_bits(crystal_cap, XTAL1);

instead.

That's partially because there are also le32 variants of it etc. that
are simpler to use with endian-fixed firmware interfaces or similar.

But YMMV (as might that of the maintainers of this driver) :-)

johannes