Skip to contentSkip to conversation

m68knommu: CONFIG_I2C_COLDFIRE and CONFIG_RTC_DRV_M5441x

DISCUSSION Paul Bolle 5 participants 11 messages May 20, 2014 – May 23, 2014

lore
Share & tools
lore

Paul Bolle

author
#0 lore raw

Hi Geert,

In v3.6 both a check for CONFIG_I2C_COLDFIRE and two checks for CONFIG_RTC_DRV_M5441x were added to the code for the coldfire platform (in m525x.c and in m5441x.c respectively). Neither of these macros can be matched to a Kconfig symbol.

Is a series that adds these Kconfig symbols pending?

Paul Bolle

#1 lore raw

Hi Paul,

Quoted 6 lines · Paul Bolle

In v3.6 both a check for CONFIG_I2C_COLDFIRE and two checks for CONFIG_RTC_DRV_M5441x were added to the code for the coldfire platform (in m525x.c and in m5441x.c respectively). Neither of these macros can be matched to a Kconfig symbol.

Is a series that adds these Kconfig symbols pending?

For the former, Steven King (cc) submitted patches in 2010 and 2012. For the latter, he submitted patches in 2012.

Seems like they were never integrated...

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

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

#2 lore raw
Quoted 8 lines · Geert Uytterhoeven

Hi Paul,

Quoted 5 lines · Paul Bolle

In v3.6 both a check for CONFIG_I2C_COLDFIRE and two checks for CONFIG_RTC_DRV_M5441x were added to the code for the coldfire platform (in m525x.c and in m5441x.c respectively). Neither of these macros can be matched to a Kconfig symbol.

Quoted 6 lines
Quoted 1 line

Is a series that adds these Kconfig symbols pending?

For the former, Steven King (cc) submitted patches in 2010 and 2012. For the latter, he submitted patches in 2012.

Seems like they were never integrated...

Indeed.

At this point, they're pretty much superfluous so if someone were to submit a patch removing them, they would not get any objections from me.

author
#3 lore raw
Quoted 3 lines · Steven King

At this point, they're pretty much superfluous so if someone were to submit a patch removing them, they would not get any objections from me.

Just to be clear: you mean the checks for these two macros (and the code hidden behind them), don't you?

Paul Bolle

#4 lore raw
Quoted 7 lines · Paul Bolle
Quoted 3 lines · Steven King

At this point, they're pretty much superfluous so if someone were to submit a patch removing them, they would not get any objections from me.

Just to be clear: you mean the checks for these two macros (and the code hidden behind them), don't you?

Yes.

#5 lore raw
In reply to Steven King

Hi all,

The MCF5441x support is quite a hot topic today for our company.
This Soc family is seducing some of our customers, especially for it's long term support from Freescale.

We have worked on this target since past few weeks, using a Tower Kit hardware and JTag, and just about to prepare a full integration of this device on recent kernel release. As a part of the open-source department, we have also planned to deliver this work to the community and propose some patches to refresh this target support.

Some informations we observe on this target to be refreshed:
- NAND controller needs to be integrated again,
- UART controller as well,

Also :
- This chip contains a MMU and can be unlink from the !MMU dependency

This is for the big lines, if the community is ok for a patch receipt, some other feature can be also pushed. BTW, what about this patches proposal ? Is there some time slot for patch proposal ?

Regards,

Quoted 21 lines · Steven King
Quoted 8 lines · Geert Uytterhoeven

Hi Paul,

Quoted 5 lines · Paul Bolle

In v3.6 both a check for CONFIG_I2C_COLDFIRE and two checks for CONFIG_RTC_DRV_M5441x were added to the code for the coldfire platform (in m525x.c and in m5441x.c respectively). Neither of these macros can be matched to a Kconfig symbol.

Quoted 5 lines
Quoted 1 line

Is a series that adds these Kconfig symbols pending?

For the former, Steven King (cc) submitted patches in 2010 and 2012. For the latter, he submitted patches in 2012.

Seems like they were never integrated...

Indeed.

At this point, they're pretty much superfluous so if someone were to submit a patch removing them, they would not get any objections from me. --
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

author
#6 lore raw

Yannick,

Quoted 15 lines · Yannick GICQUEL

The MCF5441x support is quite a hot topic today for our company. This Soc family is seducing some of our customers, especially for it's long term support from Freescale.

We have worked on this target since past few weeks, using a Tower Kit hardware and JTag, and just about to prepare a full integration of this device on recent kernel release.
As a part of the open-source department, we have also planned to deliver this work to the community and propose some patches to refresh this target support.

Some informations we observe on this target to be refreshed:
- NAND controller needs to be integrated again,
- UART controller as well,

Quoted 7 lines

Also :
- This chip contains a MMU and can be unlink from the !MMU dependency

This is for the big lines, if the community is ok for a patch receipt, some other feature can be also pushed.
BTW, what about this patches proposal ? Is there some time slot for patch proposal ?

The patch discussed (but not yet submitted) would remove all code currently hidden behind

    #if IS_ENABLED(CONFIG_RTC_DRV_M5441x)

in the kernel tree. Now, I'm guessing the (new) code you're working on doesn't care about CONFIG_RTC_DRV_M5441x. So chances are you could not care less if that code gets removed.

Paul Bolle

#7 lore raw
In reply to Yannick GICQUEL
Quoted 17 lines · Yannick GICQUEL

Hi all,

The MCF5441x support is quite a hot topic today for our company. This Soc family is seducing some of our customers, especially for it's long term support from Freescale.

We have worked on this target since past few weeks, using a Tower Kit hardware and JTag, and just about to prepare a full integration of this device on recent kernel release.
As a part of the open-source department, we have also planned to deliver this work to the community and propose some patches to refresh this target support.

Some informations we observe on this target to be refreshed:
- NAND controller needs to be integrated again,
- UART controller as well,

Quoted 8 lines

Also :
- This chip contains a MMU and can be unlink from the !MMU dependency

This is for the big lines, if the community is ok for a patch receipt, some other feature can be also pushed.
BTW, what about this patches proposal ? Is there some time slot for patch proposal ?

That interesting. As Paul mentioned the proposed patch would remove the unused code supporting the RTC on the m5441x, however, I do have an updated (v3.15-rc5) version of the RTC driver as well as updated versions of the 1wire driver that I had submiited a couple of years ago, also a much simplier version of the patches to enable the enet-fec that I had posted previously. I've also been looking at getting the eDMA working with an eye on adding support for the dspi.

As it didnt seem there was much interest in the platform I hadn't felt any need to try to get these merged, but if there is actually anyone besides myself interested in the m5441x I'd be a bit more motivated.

#8 lore raw

Hi Steven, Paul,

Quoted 21 lines · Steven King
Quoted 20 lines · Yannick GICQUEL

Hi all,

The MCF5441x support is quite a hot topic today for our company. This Soc family is seducing some of our customers, especially for it's long term support from Freescale.

We have worked on this target since past few weeks, using a Tower Kit hardware and JTag, and just about to prepare a full integration of this device on recent kernel release.
As a part of the open-source department, we have also planned to deliver this work to the community and propose some patches to refresh this target support.

Some informations we observe on this target to be refreshed:
- NAND controller needs to be integrated again,
- UART controller as well,

Also :
- This chip contains a MMU and can be unlink from the !MMU dependency

Quoted 8 lines
Quoted 5 lines

This is for the big lines, if the community is ok for a patch receipt, some other feature can be also pushed.
BTW, what about this patches proposal ? Is there some time slot for patch proposal ?

That interesting. As Paul mentioned the proposed patch would remove the unused code supporting the RTC on the m5441x, however, I do have an updated (v3.15-rc5) version of the RTC driver as well as updated versions of the 1wire driver that I had submiited a couple of years ago, also a much simplier version of the patches to enable the enet-fec that I had posted previously. I've also been looking at getting the eDMA working with an eye on adding support for the dspi.

As it didnt seem there was much interest in the platform I hadn't felt any need to try to get these merged, but if there is actually anyone besides myself interested in the m5441x I'd be a bit more motivated.

We agree these specifics defined are not handled in KConfig anymore and code cleanup is always a good thing. RTC support is not a big deal, we are more focusing on the Nand reintegration and MMU.

FYI, as a part of pre-study and POC activities for industrials requirements, the code we are working on is today based on initials BSP from Freescale (ported on 2.6.29 and 2.6.38) with features set hopefully qualified on them (nand ctrl, dspi, i2c, can, enet-fec, ...).

Regards,
Yannick

#9 lore raw
In reply to Yannick GICQUEL

Hi Yannick,

Quoted 9 lines · Yannick GICQUEL

The MCF5441x support is quite a hot topic today for our company.
This Soc family is seducing some of our customers, especially for it's long term support from Freescale.

We have worked on this target since past few weeks, using a Tower Kit hardware and JTag, and just about to prepare a full integration of this device on recent kernel release. As a part of the open-source department, we have also planned to deliver this work to the community and propose some patches to refresh this target support.

Some informations we observe on this target to be refreshed:
- NAND controller needs to be integrated again,
- UART controller as well,

What needs to be done to the mcf UART driver?
It seems all other ColdFire parts have used pretty much the same module.

Quoted 2 lines

Also :
- This chip contains a MMU and can be unlink from the !MMU dependency

The ColdFire v4e MMU has been supported for quite a while in main line kernels (from linux-3.3 onwards). I believe it will need some tweaks for the mcf5441x, but the heavy lifting is done.

Also keep in mind that for ColdFire that have an MMU we support them both with the MMU enabled and disabled.

Quoted 2 lines

This is for the big lines, if the community is ok for a patch receipt, some other feature can be also pushed. BTW, what about this patches proposal ? Is there some time slot for patch proposal ?

Early and often :-)
Post them here. If you are really unsure about some specific feature then you can discuss it here first. But if you have code already why not just send patches.

Regards
Greg

Quoted 30 lines

Regards,

Quoted 21 lines · Steven King
Quoted 8 lines · Geert Uytterhoeven

Hi Paul,

Quoted 5 lines · Paul Bolle

In v3.6 both a check for CONFIG_I2C_COLDFIRE and two checks for CONFIG_RTC_DRV_M5441x were added to the code for the coldfire platform (in m525x.c and in m5441x.c respectively). Neither of these macros can be matched to a Kconfig symbol.

Quoted 5 lines
Quoted 1 line

Is a series that adds these Kconfig symbols pending?

For the former, Steven King (cc) submitted patches in 2010 and 2012. For the latter, he submitted patches in 2012.

Seems like they were never integrated...

Indeed.

At this point, they're pretty much superfluous so if someone were to submit a patch removing them, they would not get any objections from me.

Signature · 3 lines
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 3 lines
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#10 lore raw

Hi Greg,

Quoted 9 lines · Greg Ungerer

Hi Yannick,

Quoted 6 lines · Yannick GICQUEL

The MCF5441x support is quite a hot topic today for our company.
This Soc family is seducing some of our customers, especially for it's long term support from Freescale.

We have worked on this target since past few weeks, using a Tower Kit hardware and JTag, and just about to prepare a full integration of this device on recent kernel release. As a part of the open-source department, we have also planned to deliver this work to the community and propose some patches to refresh this target support.

Quoted 6 lines
Quoted 3 lines

Some informations we observe on this target to be refreshed:
- NAND controller needs to be integrated again,
- UART controller as well,

What needs to be done to the mcf UART driver?
It seems all other ColdFire parts have used pretty much the same module.

You're right. After initials investigations, we thought there was a rework needed but after yesterday investigations, we confirm it's functional state is Ok. Please, just forget initial rework mentioned :-)

Quoted 9 lines
Quoted 2 lines

Also :
- This chip contains a MMU and can be unlink from the !MMU dependency

The ColdFire v4e MMU has been supported for quite a while in main line kernels (from linux-3.3 onwards). I believe it will need some tweaks for the mcf5441x, but the heavy lifting is done.

Also keep in mind that for ColdFire that have an MMU we support them both with the MMU enabled and disabled.

Ok, we will keep it in mind.
We had a look on this MMU activation and functional state is ko.
To state this, we had to unlink MMU activation to FPU activation, but this is not as simple and it need a deeper investigation.

Quoted 6 lines
Quoted 2 lines

This is for the big lines, if the community is ok for a patch receipt, some other feature can be also pushed. BTW, what about this patches proposal ? Is there some time slot for patch proposal ?

Early and often :-)
Post them here. If you are really unsure about some specific feature then you can discuss it here first. But if you have code already why not just send patches.

Our patches are not clean enough to be posted right now, but as people here are receptive to this, we will go on.

Quoted 34 lines

Regards
Greg

Quoted 29 lines

Regards,

Quoted 21 lines · Steven King
Quoted 8 lines · Geert Uytterhoeven

Hi Paul,

Quoted 5 lines · Paul Bolle

In v3.6 both a check for CONFIG_I2C_COLDFIRE and two checks for CONFIG_RTC_DRV_M5441x were added to the code for the coldfire platform (in m525x.c and in m5441x.c respectively). Neither of these macros can be matched to a Kconfig symbol.

Quoted 5 lines
Quoted 1 line

Is a series that adds these Kconfig symbols pending?

For the former, Steven King (cc) submitted patches in 2010 and 2012. For the latter, he submitted patches in 2012.

Seems like they were never integrated...

Indeed.

At this point, they're pretty much superfluous so if someone were to submit a patch removing them, they would not get any objections from me.

Signature · 3 lines
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 3 lines
To unsubscribe from this list: send the line "unsubscribe linux-m68k" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html