Skip to contentSkip to conversation

Question regarding back compatibility for DT

DISCUSSION Daniel Lezcano 3 participants 5 messages Mar 21, 2017 2 messages not in the archive

lore
Share & tools
lore
#0 lore raw

Hi Rob, dt people,

is it acceptable a driver is changed using a new DT binding but becomes incompatible with the old DT files ?

IOW, is it possible to change the DT and the driver without supporting the old DT format, forcing an update of the DT and the kernel ?

Thanks.

-- Daniel

Signature · 9 lines
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#1 lore raw
Quoted 7 lines · Daniel Lezcano

Hi Rob, dt people,

is it acceptable a driver is changed using a new DT binding but becomes incompatible with the old DT files ?

IOW, is it possible to change the DT and the driver without supporting the old DT format, forcing an update of the DT and the kernel ?

I've seen new properties many times added, even though the new driver wouldn't work quite right on old DT files. I've also seen people create a new hardware compatibility name for a totally new binding. Finally, I've seen the bindings just randomly change for no reason with no concern for backwards compat. The last really sucks if you have a driver that's not in the Linux tree (say, because it's a FreeBSD driver).

Warner

Quoted 8 lines

Thanks.

-- Daniel

--

<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Quoted 7 lines

Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#4 lore raw
In reply to Daniel Lezcano
Quoted 7 lines · Daniel Lezcano

Hi Rob, dt people,

is it acceptable a driver is changed using a new DT binding but becomes incompatible with the old DT files ?

IOW, is it possible to change the DT and the driver without supporting the old DT format, forcing an update of the DT and the kernel ?

If you want a global rule, then no it is not okay. However, it depends on the platform and it's users. I just ask that breaking compatibility and why is highlighted in the commit message and the platform maintainer is okay with it.

Rob
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#3 lore raw
In reply to a message not in the archive
Quoted 10 lines · Warner Losh
Quoted 7 lines · Daniel Lezcano

Hi Rob, dt people,

is it acceptable a driver is changed using a new DT binding but becomes incompatible with the old DT files ?

IOW, is it possible to change the DT and the driver without supporting the old DT format, forcing an update of the DT and the kernel ?

Quoted 7 lines

I've seen new properties many times added, even though the new driver wouldn't work quite right on old DT files. I've also seen people create a new hardware compatibility name for a totally new binding. Finally, I've seen the bindings just randomly change for no reason with no concern for backwards compat. The last really sucks if you have a driver that's not in the Linux tree (say, because it's a FreeBSD driver).

Hopefully that is getting better? If not, how do we make it better? I try to prevent it through reviews, but I have have no visibility into what FreeBSD does and doesn't care about.

The common suggestion is moving the bindings out of the kernel. The biggest issue I have with that is that we would lose the review of all the kernel subsystem maintainers. They may be biased toward Linux, but they know their class of h/w and no other project has the breath of h/w support.

Rob
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

#6 lore raw
In reply to a message not in the archive
Quoted 7 lines · Rob Herring
Quoted 5 lines · Daniel Lezcano

Hi Rob, dt people,

is it acceptable a driver is changed using a new DT binding but becomes incompatible with the old DT files ?

Quoted 7 lines
Quoted 2 lines

IOW, is it possible to change the DT and the driver without supporting the old DT format, forcing an update of the DT and the kernel ?

If you want a global rule, then no it is not okay. However, it depends on the platform and it's users. I just ask that breaking compatibility and why is highlighted in the commit message and the platform maintainer is okay with it.

Thanks for the clarification. I will ask the maintainer if breaking the compatibility is ok.

That shows how important is to wisely choose the right binding.

-- Daniel

Signature · 9 lines
 <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html