Date: Fri, 15 Dec 2017 05:34:01 +0000
From: Al Viro <viro@ZenIV.linux.org.uk>
To: Jakub Kicinski <kubakici@wp.pl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [RFC][PATCH] Add primitives for manipulating bitfields both in
 host- and fixed-endian.
Message-ID: <20171215053401.GH21978@ZenIV.linux.org.uk>
References: <20171213003659.GA21978@ZenIV.linux.org.uk>
 <20171212170437.4b129e50@cakuba.netronome.com>
 <20171213013056.GB21978@ZenIV.linux.org.uk>
 <20171212173528.340cd002@cakuba.netronome.com>
 <20171213015125.GC21978@ZenIV.linux.org.uk>
 <20171212184400.13b27cf8@cakuba.netronome.com>
 <20171213142212.GD21978@ZenIV.linux.org.uk>
 <20171213174554.GE21978@ZenIV.linux.org.uk>
 <20171215023343.GG21978@ZenIV.linux.org.uk>
 <20171214210713.6be03d43@cakuba.netronome.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20171214210713.6be03d43@cakuba.netronome.com>
User-Agent: Mutt/1.9.0 (2017-09-02)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1656
Lines: 40

On Thu, Dec 14, 2017 at 09:07:13PM -0800, Jakub Kicinski wrote:
> Looks great to me!
> 
> On Fri, 15 Dec 2017 02:33:43 +0000, Al Viro wrote:
> > The following primitives are defined in linux/bitfield.h:
> > 
> > * u32 le32_get_bits(__le32 val, u32 field) extracts the contents of the
> >   bitfield specified by @field in little-endian 32bit value @val and
> >   converts it to host-endian.
> > 
> > * void le32p_replace_bits(__le32 *p, u32 v, u32 field) replaces
> >   the contents of the bitfield specified by @field in little-endian
> >   32bit object pointet to by *p with the value of @v.  New value is
> >   given in host-endian and stored as little-endian.
> > 
> > * __le32 le32_replace_bits(__le32 old, u32 v, u32 field) is equivalent to
> >   ({__le32 tmp = old; le32p_replace_bits(&old, v, field); tmp;})
> >   In other words, instead of modifying an object in memory, it takes
> >   the initial value and returns the modified one.
> 
> the current macros take filed/mask as first param, not sure if it's
> worth maintaining the order

Umm...  For something like Haskell that would be more natural (as in
replace_foo = replace_field foo), but it's C - no partially applied
functions here...

While we are at it, to cover the FIELD_PREP users it might make sense to
add

__le32 le32_encode_bits(u32 v, u32 field)
{
        if (__builtin_constant_p(v) &&
                    (v & ~(field/mask_to_multiplier(field))))
                                    __field_overflow();
	return cpu_to_le32((v * mask_to_multiplier(field)) & field);
}

turning the body of le32_replace_bits into
	return (old & ~cpu_to_le32(field)) | le32_encode_bits(v, field);