linux-fbdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sam Ravnborg <sam@ravnborg.org>
To: Helge Deller <deller@gmx.de>
Cc: Thomas Zimmermann <tzimmermann@suse.de>,
	javierm@redhat.com, linux-media@vger.kernel.org,
	linux-fbdev@vger.kernel.org, dri-devel@lists.freedesktop.org,
	linux-arm-kernel@lists.infradead.org,
	linux-geode@lists.infradead.org, linux-omap@vger.kernel.org,
	kvm@vger.kernel.org
Subject: Re: [PATCH 00/47] fbdev: Use I/O helpers
Date: Fri, 28 Jul 2023 23:01:27 +0200	[thread overview]
Message-ID: <20230728210127.GA1156027@ravnborg.org> (raw)
In-Reply-To: <c1a4b7c9-50f2-c43f-277d-c2af9ccc0b50@gmx.de>

Hi Helge,

On Fri, Jul 28, 2023 at 08:46:59PM +0200, Helge Deller wrote:
> On 7/28/23 18:39, Thomas Zimmermann wrote:
> > Most fbdev drivers operate on I/O memory.
> 
> Just nitpicking here:
> What is I/O memory?
> Isn't it either memory, or I/O ?
> I mean, I would never think of the cfb* draw functions under I/O.
> 
> > And most of those use the
> > default implementations for file I/O and console drawing. Convert all
> > these low-hanging fruits to the fb_ops initializer macro and Kconfig
> > token for fbdev I/O helpers.
> 
> I do see the motivation for your patch, but I think the
> macro names are very misleading.
> 
> You have:
> #define __FB_DEFAULT_IO_OPS_RDWR \
>         .fb_read        = fb_io_read, \
>         .fb_write       = fb_io_write
> 
> #define __FB_DEFAULT_IO_OPS_DRAW \
>         .fb_fillrect    = cfb_fillrect, \
>         .fb_copyarea    = cfb_copyarea, \
>         .fb_imageblit   = cfb_imageblit
> 
> #define __FB_DEFAULT_IO_OPS_MMAP \
>         .fb_mmap        = NULL /* default implementation */
> 
> #define FB_DEFAULT_IO_OPS \
>         __FB_DEFAULT_IO_OPS_RDWR, \
>         __FB_DEFAULT_IO_OPS_DRAW, \
>         __FB_DEFAULT_IO_OPS_MMAP
> 
> I think FB_DEFAULT_IO_OPS is OK for read/write/mmap.
> But I would suggest to split out __FB_DEFAULT_IO_OPS_DRAW.
> Something like:
> #define FB_DEFAULT_IO_OPS \
>         __FB_DEFAULT_IO_OPS_RDWR, \
>         __FB_DEFAULT_IO_OPS_MMAP


> #define FB_DEFAULT_CFB_OPS \
>         .fb_fillrect    = cfb_fillrect, \
>         .fb_copyarea    = cfb_copyarea, \
>         .fb_imageblit   = cfb_imageblit

The prefix cfb, I have recently learned, equals color frame buffer.
They are named such for purely historical reasons.

What is important is where the data are copied as we have two
implementations of for example copyarea - one using system memory, the
other using IO memory.

The naming FB_DEFAULT_IO_OPS says this is the defaults to IO memory
operations, which tell what they do and avoid the strange cfb acronym.

Reserve cfb for color frame buffers only - and maybe in the end rename
the three cfbcopyarea, cfbfillrect, cfbimgblt to use the io prefix.
Which is much simpler to do after this series - and nice extra benefit.

I hope this properly explains why I like the current naming and
acked it when the macros were introduced.

	Sam

  reply	other threads:[~2023-07-28 21:01 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-28 16:39 [PATCH 00/47] fbdev: Use I/O helpers Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 01/47] media/vivid: Use fbdev " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 02/47] fbdev/acornfb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 03/47] fbdev/asiliantfb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 04/47] fbdev/atmel_lcdfb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 05/47] fbdev/aty128fb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 06/47] fbdev/carminefb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 07/47] fbdev/chipsfb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 08/47] fbdev/da8xx-fb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 09/47] fbdev/efifb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 10/47] fbdev/fm2fb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 11/47] fbdev/fsl-diu-fb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 12/47] fbdev/g364fb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 13/47] fbdev/geode/gx1fb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 14/47] fbdev/geode/gxfb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 15/47] fbdev/geode/lxfb: " Thomas Zimmermann
2023-07-28 16:39 ` [PATCH 16/47] fbdev/goldfishfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 17/47] fbdev/grvga: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 18/47] fbdev/gxt4500: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 19/47] fbdev/i740fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 20/47] fbdev/imxfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 21/47] fbdev/kyro: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 22/47] fbdev/macfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 23/47] fbdev/maxinefb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 24/47] fbdev/mb862xxfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 25/47] fbdev/mmpfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 26/47] fbdev/mx3fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 27/47] fbdev/ocfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 28/47] fbdev/offb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 29/47] fbdev/omapfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 30/47] fbdev/platinumfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 31/47] fbdev/pmag-aa-fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 32/47] fbdev/pmag-ba-fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 33/47] fbdev/pmag-b-fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 34/47] fbdev/pxa168fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 35/47] fbdev/pxafb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 36/47] fbdev/q40fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 37/47] fbdev/s3cfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 38/47] fbdev/sh7760fb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 39/47] fbdev/simplefb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 40/47] fbdev/sstfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 41/47] fbdev/sunxvr1000: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 42/47] fbdev/sunxvr2500: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 43/47] fbdev/uvesafb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 44/47] fbdev/valkyriefb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 45/47] fbdev/vesafb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 46/47] fbdev/xilinxfb: " Thomas Zimmermann
2023-07-28 16:40 ` [PATCH 47/47] vfio-dev/mdpy-fb: " Thomas Zimmermann
2023-07-28 18:35 ` [PATCH 00/47] fbdev: Use " Sam Ravnborg
2023-07-30  8:55   ` Sam Ravnborg
2023-07-28 18:39 ` Sam Ravnborg
2023-07-29 13:13   ` Thomas Zimmermann
2023-08-01  8:59   ` Thomas Zimmermann
2023-07-28 18:46 ` Helge Deller
2023-07-28 21:01   ` Sam Ravnborg [this message]
2023-07-29  6:51     ` Helge Deller
2023-07-29 13:21       ` Thomas Zimmermann
2023-07-29 13:53         ` Helge Deller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230728210127.GA1156027@ravnborg.org \
    --to=sam@ravnborg.org \
    --cc=deller@gmx.de \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=javierm@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-fbdev@vger.kernel.org \
    --cc=linux-geode@lists.infradead.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=tzimmermann@suse.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).