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: <87ed2o506g.fsf@intel.com>
Date: Tue, 03 Dec 2024 16:08:23 +0200
From: Jani Nikula <jani.nikula@...ux.intel.com>
To: Jocelyn Falempe <jfalempe@...hat.com>, Maarten Lankhorst
 <maarten.lankhorst@...ux.intel.com>, Maxime Ripard <mripard@...nel.org>,
 Thomas Zimmermann <tzimmermann@...e.de>, David Airlie <airlied@...il.com>,
 Daniel Vetter <daniel@...ll.ch>, John Ogness <john.ogness@...utronix.de>,
 Javier Martinez Canillas <javierm@...hat.com>, "Guilherme G . Piccoli"
 <gpiccoli@...lia.com>, bluescreen_avenger@...izon.net, Caleb Connolly
 <caleb.connolly@...aro.org>, Petr Mladek <pmladek@...e.com>,
 dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v8 1/6] drm/panic: Move drawing functions to drm_draw

On Tue, 03 Dec 2024, Jocelyn Falempe <jfalempe@...hat.com> wrote:
> On 03/12/2024 12:06, Jani Nikula wrote:
>> On Fri, 15 Nov 2024, Jocelyn Falempe <jfalempe@...hat.com> wrote:
>>> Move the color conversions, blit and fill functions to drm_draw.c,
>>> so that they can be re-used by drm_log.
>>> drm_draw is internal to the drm subsystem, and shouldn't be used by
>>> gpu drivers.
>> 
>> I started looking at this in patch 2:
>> 
>>> +#include "../drm_draw.h"
>> 
>> I think we should avoid #includes with ../ like this.
>
> Sure, I've added it in v8, after the clients moved to drm/clients/, but 
> I didn't think much about it.
>
>> 
>> Either drm_draw.h belongs in include/drm, or maybe clients/Makefile
>> needs to add subdir-ccflags-y += -I$(src)/.. or something like that?
>> 
>> If it's supposed to be internal, I guess the latter, but then the
>> current convention is to have _internal.h suffix. All drm headers under
>> drivers/ have that.
>
> ok, I can rename drm_draw.h to drm_draw_internal.h, and add the include 
> in the Makefile.
>> 
>> Is this the first drm subsystem internal thing that's a separate module?
>> Should we use EXPORT_SYMBOL_NS() and MODULE_IMPORT_NS() to enforce it
>> being internal?
>
> It's not really a separate module, as it's built in the core drm module. 
> (the reason is that it's used by drm_panic too, which must be in the 
> core drm module).

Right, sorry, I got confused and looked at this the other way round.

drm_draw is part of drm.ko, drm log is part of drm_client_lib.ko, and
uses drm_draw, right?

So is drm_draw really internal if drm client uses it?

I don't really deeply care either way, but it bugs me when it's
somewhere in between. :)

Either include/drm/drm_draw.h or drivers/gpu/drm/drm_draw_internal.h,
the latter *possibly* with symbol namespace, but not a big deal.


BR,
Jani.


>
> I don't know much about symbol namespace, but I can add that if needed.
>
> Best regards,

-- 
Jani Nikula, Intel

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ