linux-kernel - Re: [PATCH] greybus: constify the struct device

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]

Date: Mon, 26 Feb 2024 17:21:37 -0300
From: "Ricardo B. Marliere" <ricardo@...liere.net>
To: Alex Elder <elder@...e.org>
Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>, 
	Johan Hovold <johan@...nel.org>, Alex Elder <elder@...nel.org>, greybus-dev@...ts.linaro.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] greybus: constify the struct device_type usage

On 26 Feb 13:50, Alex Elder wrote:
> On 2/25/24 5:04 AM, Ricardo B. Marliere wrote:
> >>>> On another subject:
> >>>>
> >>>> Johan might disagree, but I think it would be nice to make
> >>>> the definitions of the Greybus device types as static (private)
> >>>> and make the is_gb_host_device() etc. functions real functions
> >>>> rather than static inlines in <linux/greybus.h>.
> >>>>
> >>>> It turns out that all of the is_gb_*() functions are called only
> >>>> from drivers/greybus/core.c; they could all be moved there rather
> >>>> than advertising them in <linux/greybus.h>.
> >>> I guess it depends whether they would be used somewhere else in the
> >>> future. Perhaps it was left there with that intention when it was first
> >>> being developed? I agree, though. Will happily send a patch with this if
> >>> desired.
> >> Let's clean the code up for what we have today.  If it's needed in the
> >> future, we can move the structures then.
> > Sounds good to me, will send a v2 then!
> 
> I might be misinterpreting Greg's response; I *think* he
> agrees with my suggestion.

That's what I thought too.

> 
> In any case, please do *not* send v2 with the purpose of
> including my suggestion.
> 
> If you send a v2, keep it focused on this original patch.
> You can then implement the other suggestion as a follow-on
> patch (or series).

Indeed, this one is good as is but I thought of converting it into a
series so that they can be taken with no dependency on this one. So it
would look like:

Patch 1: move "is_gb_*()" into drivers/greybus/core.c
Patch 2: move "device_type greybus_*" into drivers/greybus/core.c
Patch 3: make "device_type greybus_*" const

But you're right. I could simply send 1 and 2 after this one has been
applied. If I were to send them separately, how would I communicate that
there's a dependency? Something like:

---
This series depends on [1].
[1]: lore://link-to-this-patch 

?

Thanks and sorry for the noobishness

> 
> 					-Alex