* Re: [PATCH 0/5] arm64: dts: renesas: move I2C aliases to board files
[not found] <20220729164425.11062-1-wsa+renesas@sang-engineering.com>
@ 2022-08-18 13:27 ` Geert Uytterhoeven
2022-08-25 6:37 ` Wolfram Sang
0 siblings, 1 reply; 3+ messages in thread
From: Geert Uytterhoeven @ 2022-08-18 13:27 UTC (permalink / raw)
To: Wolfram Sang
Cc: Linux-Renesas, Rob Herring, Krzysztof Kozlowski,
open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS
Hi Wolfram,
CC devicetree
On Fri, Jul 29, 2022 at 6:50 PM Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> For R-Car Gen3 and Gen4, we had I2C aliases in per-SoC DTS include
> files. That doesn't really fit as bus naming is a board property. This
> series moves the aliases to the board files. The following procedure has
> been applied to avoid regressions:
>
> 1) move the aliases from SoC files to board files. Keep the empty
> alias-nodes in the SoC file and add the new aliases before existing
> ones. This ensured that identical binaries[1] were created.
>
> 2) Once all aliases were moved and all binaries[1] stayed identical,
> then the empty alias nodes from the SoC files were removed.
>
> The result is this series based on the renesas-drivers tag
> 'renesas-drivers-2022-07-19-v5.19-rc7'.
>
> It could be discussed if the aliases should be sorted alphabetically,
> but I'd think this is a seperate series then.
>
> I also leave it to Geert, if the commits should be squashed. At least
> for reviewing, I think this separation makes sense, though.
>
> Looking forward to comments,
Thanks for your series!
I have never been a fan of the i2c aliases in the SoC-specific .dtsi files,
as aliases are typically board-specific.
We also don't have physical connectors labeled "i2c<N>" on any of
the affected boards. But people like the i2c aliases, because i2c
exposes the full buses to userspace, and the aliases fix the userspace
naming of /dev/i2c-<N> (I believe there is no better way to identify
i2c buses from userspace?).
So moving the i2c aliases to the board files is definitely a step in
the good direction.
BTW, you missed r8a774a1.dtsi in your update.
> [1] For desired binaries, of course. For r8a77961 boards, the binaries
> did change. This was intended, though, because I wanted to add I2C
> aliases there which were not present before. Also, the r8a779m* boards
> now have I2C aliases, too.
>
>
> Wolfram Sang (5):
> arm64: dts: renesas: r8a779{51|60|65}: put I2C aliases to board files
> arm64: dts: renesas: r8a77970: put I2C aliases to board files
> arm64: dts: renesas: r8a77980: put I2C aliases to board files
> arm64: dts: renesas: r8a77990: put I2C aliases to board files
> arm64: dts: renesas: r8a779a0: put I2C aliases to board files
>
> arch/arm64/boot/dts/renesas/ebisu.dtsi | 8 ++++++++
> arch/arm64/boot/dts/renesas/r8a77951.dtsi | 11 -----------
> arch/arm64/boot/dts/renesas/r8a77960.dtsi | 11 -----------
> arch/arm64/boot/dts/renesas/r8a77965.dtsi | 11 -----------
> arch/arm64/boot/dts/renesas/r8a77970-eagle.dts | 5 +++++
> arch/arm64/boot/dts/renesas/r8a77970-v3msk.dts | 5 +++++
> arch/arm64/boot/dts/renesas/r8a77970.dtsi | 8 --------
> arch/arm64/boot/dts/renesas/r8a77980-condor.dts | 6 ++++++
> arch/arm64/boot/dts/renesas/r8a77980-v3hsk.dts | 6 ++++++
> arch/arm64/boot/dts/renesas/r8a77980.dtsi | 9 ---------
> arch/arm64/boot/dts/renesas/r8a77990.dtsi | 11 -----------
> arch/arm64/boot/dts/renesas/r8a779a0-falcon-cpu.dtsi | 7 +++++++
> arch/arm64/boot/dts/renesas/r8a779a0.dtsi | 10 ----------
> arch/arm64/boot/dts/renesas/salvator-common.dtsi | 8 ++++++++
> arch/arm64/boot/dts/renesas/ulcb.dtsi | 8 ++++++++
> 15 files changed, 53 insertions(+), 71 deletions(-)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 0/5] arm64: dts: renesas: move I2C aliases to board files
2022-08-18 13:27 ` [PATCH 0/5] arm64: dts: renesas: move I2C aliases to board files Geert Uytterhoeven
@ 2022-08-25 6:37 ` Wolfram Sang
2022-08-25 6:46 ` Geert Uytterhoeven
0 siblings, 1 reply; 3+ messages in thread
From: Wolfram Sang @ 2022-08-25 6:37 UTC (permalink / raw)
To: Geert Uytterhoeven
Cc: Linux-Renesas, Rob Herring, Krzysztof Kozlowski,
open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS
[-- Attachment #1: Type: text/plain, Size: 883 bytes --]
Hi Geert,
> I have never been a fan of the i2c aliases in the SoC-specific .dtsi files,
> as aliases are typically board-specific.
> We also don't have physical connectors labeled "i2c<N>" on any of
> the affected boards. But people like the i2c aliases, because i2c
The reasoning here was that the busses were named like this in the
schematics. Debugging was confusing if these numbers were mixed.
> exposes the full buses to userspace, and the aliases fix the userspace
> naming of /dev/i2c-<N> (I believe there is no better way to identify
> i2c buses from userspace?).
The proper way is udev rules.
> So moving the i2c aliases to the board files is definitely a step in
> the good direction.
>
> BTW, you missed r8a774a1.dtsi in your update.
Okay, so I will send v2 with the above SoC converted as well?
Thanks for your comments,
Wolfram
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 0/5] arm64: dts: renesas: move I2C aliases to board files
2022-08-25 6:37 ` Wolfram Sang
@ 2022-08-25 6:46 ` Geert Uytterhoeven
0 siblings, 0 replies; 3+ messages in thread
From: Geert Uytterhoeven @ 2022-08-25 6:46 UTC (permalink / raw)
To: Wolfram Sang
Cc: Linux-Renesas, Rob Herring, Krzysztof Kozlowski,
open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS
Hi Wolfram,
On Thu, Aug 25, 2022 at 8:37 AM Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> > I have never been a fan of the i2c aliases in the SoC-specific .dtsi files,
> > as aliases are typically board-specific.
> > We also don't have physical connectors labeled "i2c<N>" on any of
> > the affected boards. But people like the i2c aliases, because i2c
>
> The reasoning here was that the busses were named like this in the
> schematics. Debugging was confusing if these numbers were mixed.
OK.
> > exposes the full buses to userspace, and the aliases fix the userspace
> > naming of /dev/i2c-<N> (I believe there is no better way to identify
> > i2c buses from userspace?).
>
> The proper way is udev rules.
Great!
> > So moving the i2c aliases to the board files is definitely a step in
> > the good direction.
> >
> > BTW, you missed r8a774a1.dtsi in your update.
>
> Okay, so I will send v2 with the above SoC converted as well?
Yes, please. Thanks!
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-08-25 6:46 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20220729164425.11062-1-wsa+renesas@sang-engineering.com>
2022-08-18 13:27 ` [PATCH 0/5] arm64: dts: renesas: move I2C aliases to board files Geert Uytterhoeven
2022-08-25 6:37 ` Wolfram Sang
2022-08-25 6:46 ` Geert Uytterhoeven
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).