Re: [PATCH 1/2] [v2] lib/hexdump: introduce DUMP_PREFIX_UNHASHED for unhashed addresses

[Andy Shevchenko]

On Mon, Jan 18, 2021 at 09:57:55AM -0600, Timur Tabi wrote:
> On 1/18/21 4:03 AM, Andy Shevchenko wrote:
> > On Sun, Jan 17, 2021 at 12:12 AM Timur Tabi <timur@xxxxxxxxxx> wrote:

...

> > Any user of this? (For the record, I don't see any other mail except this one)

> It's patch #2 of this set.

I haven't got that one.

> They were all sent together.

Apparently not to me.

> http://lkml.iu.edu/hypermail/linux/kernel/2101.2/00245.html
> 
> Let me know what you think.

Makes sense. Hint: use lore.kernel.org references as they are much better in
terms of provided features and patch representation.

...

> > >          DUMP_PREFIX_NONE,
> > >          DUMP_PREFIX_ADDRESS,
> > > -       DUMP_PREFIX_OFFSET
> > > +       DUMP_PREFIX_OFFSET,
> > > +       DUMP_PREFIX_UNHASHED,
> > 
> > Since it's an address, I would like to group them together, i.e. put
> > after DUMP_PREFIX_ADDRESS.
> 
> I didn't want to change the numbering of any existing enums, just in case
> there are users that accidentally hard-code the values.  I'm trying to make
> this patch as unobtrusive as possible.

But isn't it good to expose those issues (and fix them)?

...

> > Perhaps even add _ADDRESS to DUMP_PREFIX_UNHASHED, but this maybe too
> long.
> 
> I think DUMP_PREFIX_ADDRESS_UNHASHED is too long.

What about introducing new two like these:

	DUMP_PREFIX_OFFSET,
	DUMP_PREFIX_ADDRESS,
	DUMP_PREFIX_ADDR_UNHASHED,
	DUMP_PREFIX_ADDR_HASHED,

and allow people step-by-step move to them?

-- 
With Best Regards,
Andy Shevchenko