soc/same54: Initial support #14685

benpicco · 2019-03-19T17:27:42Z

The Atmel SAME54 is a Cortex M4F with Ethernet that shares it's peripherals with the sam0 family. (This seems to be true for all new Atmel parts, it certainly is for the new SAML1x Cortex M23 parts).

The MCU is part of the SAMD5x/SAME5x family - the E parts are identical to the D ones, but come with an additional GMAC Ethernet peripheral.

This adds support for the MCU and the SAM E54 Xplained Pro development board.

I ported all existing sam0 peripherals use the new clock setup if available.

What works:

requires zephyrproject-rtos/hal_atmel#1

zephyrbot · 2019-03-19T17:35:43Z

Found the following issues, please fix and resubmit:

Identity/Emails issues

9b3747865efa9f07bffc0742228ca0a8dbfaee70: author email (Benjamin Valentin [email protected]) needs to match one of the signed-off-by entries.

8841465af732dc80cf4179cd59f1bd8346ebf87a: author email (Benjamin Valentin [email protected]) needs to match one of the signed-off-by entries.

License issues

In most cases you do not need to do anything here, especially if the files
reported below are going into ext/ and if license was approved for inclusion
into ext/ already. Fix any missing license/copyright issues. The license
exception if a JFYI for the maintainers and can be overriden when merging the
pull request.

ext/hal/atmel/asf/sam0/include/samd51/component/can.h is not apache-2.0 licensed: bsd-new
ext/hal/atmel/asf/sam0/include/samd51/instance/can0.h is not apache-2.0 licensed: bsd-new
ext/hal/atmel/asf/sam0/include/samd51/instance/can1.h is not apache-2.0 licensed: bsd-new
ext/hal/atmel/asf/sam0/include/same51/instance/sdhc1.h is not apache-2.0 licensed: bsd-new
ext/hal/atmel/asf/sam0/include/same54/component/picop.h is not apache-2.0 licensed: bsd-new
ext/hal/atmel/asf/sam0/include/same54/instance/picop.h is not apache-2.0 licensed: bsd-new

Sizurka

The parenthesis trick is very neat and I'm going to use it :) But it might be better to put it with the other macro voodoo in the shared header so it can be reused:

include/misc/util.h:387

#define UTIL_IS_PARENTHESIZED(x) UTIL_CHECK(UTIL_IS_PARENTHESIZED_ x, 0)
#define UTIL_IS_PARENTHESIZED_(...) 1, 1

You can also use some of the other macros in there to simply the usage later. For example, I'm planning on changing the GPIO handling to something like:

#define SAM0_EIC_IRQ_CONNECT(n, _)					\
COND_CODE_1(UTIL_IS_PARENTHESIZED(DT_EIC_SAM0_IRQ(n)), ( {		\
	IRQ_CONNECT(DT_EIC_SAM0_IRQ(n), DT_EIC_SAM0_IRQ_PRIORITY(n),	\
		    sam0_eic_isr, DEVICE_GET(sam0_eic), 0);		\
	irq_enable(DT_EIC_SAM0_IRQ(n));					\
} ), () )

...

UTIL_LISTIFY(16, SAM0_EIC_IRQ_CONNECT, _);

So if you don't mind I'd like to add the DTS changing and however you decide to handle the macro definitions to the GPIO PR and I can just drop it if yours gets merged first. Alternatively, I can just strip my current multi IRQ handling out and you can add something like the above to your changes for the EIC, whichever way you prefer.

benpicco · 2019-03-21T13:17:16Z

Nice! I suppose it takes a super worm moon for this kind of macro magic to come forth.
Maybe it's easier to get this included as a separate pull request.

edit: making a separate pull request revealed that the change to extract_dts_includes.py breaks the macro magic performed by nrf52, e.g. zephyr/samples/boards/bbc_microbit/sound that depends on there being no parenthesis.

A solution is to enclose all variables in parenthesizes in dts_fixup.h, but that's a lot less elegant :/

edit2: @galak suggested not to do anything weird with dts_fixup.h as that file should be removed entirely at some point. But having all the interrupt setup code capsuled in SAM0_UART_IRQ_CONNECT already saves a lot of the ugliness.

Sizurka · 2019-03-21T23:04:19Z

Oh well, it was a nice dream. I'll pull the parametric macros for the EIC out of dts_fixup.h as well, since that probably qualifies as "something weird". Since the EIC only has once instance (i.e. it won't re reordered or anything), it's probably safe to use the DT_ATMEL_SAM0_EIC_0_ macros directly, so it shouldn't need anything fixups for any of the variants.

Really, it'd be nice if scripts/dts/extract_dts_includes.py (extract/default.py ?) generated kconfig style defines for the presence of all properties in the device, so there'd be a #define DT_ATMEL_SAM0_UART_42001000_IRQ_0_PRESENT 1 or something. That's probably too specialized to the SAM0 code generation needs, though.

benpicco · 2019-03-21T23:42:54Z

Yes, I will indeed try to reduce the use of dts_fixup.h and switch to DT_ATMEL_SAM0_UART_0_IRQ_0 et al. Since any re-ordering would be consistent when sticking to the defines from generated_dts_board_unfixed.h, I don't see how it would be a problem.

The parametric macros then can stay in the corresponding .c file

Sizurka · 2019-03-21T23:48:09Z

Yeah I think it should be fine as long as you only reference devices by the label, it's just a bit odd when "SERCOM3" ends up as DT_ATMEL_SAM0_UART_0 (since it was the first UART instance) while "SERCOM1" is also DT_ATMEL_SAM0_SPI_0. As long as it's consistent everything should be assigned correctly. So I'll change the I2C code as well.

codecov-io · 2019-03-25T11:15:20Z

Codecov Report

❗ No coverage uploaded for pull request base (master@56c4004). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master   #14685   +/-   ##
=========================================
  Coverage          ?   52.92%           
=========================================
  Files             ?      309           
  Lines             ?    45268           
  Branches          ?    10451           
=========================================
  Hits              ?    23956           
  Misses            ?    16544           
  Partials          ?     4768

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 56c4004...e24d9cb. Read the comment docs.

dts/arm/atmel/samd5x.dtsi

dbkinder

+1 for docs

It shares the same peripheral with other sam0 parts while differing in the clock config. Signed-off-by: Benjamin Valentin <[email protected]>

The RTC peripheral found in the SAMD5x/SAME5x MCUs is very simmilar to the one found in existing sam0 devices with only a few changes to register names and the clock source selection. Signed-off-by: Benjamin Valentin <[email protected]>

This adds a driver for the True Random Number Generator found in some Atmel SAM0 SoCs. The Code is based on the driver for the SAM TRNG, but uses different register and clock definitions. Signed-off-by: Benjamin Valentin <[email protected]>

This adds supoprt for the Atmel SAMD51 SoC. The SAME5x/SAMD5x is a line of Cortex-M4F MCUs that share peripherals with the sam0 Cortex-M0+ and saml1x Cortex-M23 parts. Signed-off-by: Benjamin Valentin <[email protected]>

This adds supoprt for the Atmel SAME51 SoC. The SAME5x/SAMD5x is a line of Cortex-M4F MCUs that share peripherals with the sam0 Cortex-M0+ and saml1x Cortex-M23 parts. Signed-off-by: Benjamin Valentin <[email protected]>

This adds supoprt for the Atmel SAME53 SoC. The SAME5x/SAMD5x is a line of Cortex-M4F MCUs that share peripherals with the sam0 Cortex-M0+ and saml1x Cortex-M23 parts. Signed-off-by: Benjamin Valentin <[email protected]>

We need to select TRIGACT_BLOCK for DMA to peripheral to work. Signed-off-by: Benjamin Valentin <[email protected]>

The amount of lock regions differs between different sam0 MCUs. saml10: 2 saml11: 4 samd2x: 16 saml2x: 16 samd5x: 32 ASF does not provide a definition for this, so create a new one in dts. Signed-off-by: Benjamin Valentin <[email protected]>

SAMD5x/SAME5x header files do not provide this define anymore. On SAMD2x it was 0, this is still valid. Signed-off-by: Benjamin Valentin <[email protected]>

This adds support for the Atmel SAM E54 Xplained Pro Evaluation Kit. Only basic functionality has been tested so far. Signed-off-by: Benjamin Valentin <[email protected]>

benpicco · 2019-10-23T11:02:41Z

@galak I've split it into individual PRs as requested.

stephanosio · 2019-11-22T05:14:42Z

Ethernet GMAC driver not implemented/ported yet

@benpicco Do you plan to/are you already working on this?

benpicco · 2019-11-22T05:46:31Z

@stephanosio no i don't have any concrete plans for that.

stephanosio · 2019-11-22T06:35:16Z

@stephanosio no i don't have any concrete plans for that.

Ok, I will look into it then. At first glance, it seems to be using the same GMAC core as SAM7.

benpicco force-pushed the same54 branch 2 times, most recently from d3ed798 to 165dced Compare March 20, 2019 16:50

Sizurka reviewed Mar 20, 2019

View reviewed changes

benpicco force-pushed the same54 branch from 165dced to cb757bf Compare March 21, 2019 13:05

benpicco mentioned this pull request Mar 21, 2019

misc: util: Allow conditional macros to depend on whether a property in DTS exists #14788

Closed

benpicco force-pushed the same54 branch from cb757bf to b22901d Compare March 21, 2019 17:56

benpicco force-pushed the same54 branch 3 times, most recently from 8f09e97 to fdc1a5b Compare March 25, 2019 16:51

Sizurka reviewed Mar 26, 2019

View reviewed changes

dts/arm/atmel/samd5x.dtsi Outdated Show resolved Hide resolved

benpicco force-pushed the same54 branch 4 times, most recently from f294804 to d1191d1 Compare March 27, 2019 15:07

benpicco mentioned this pull request Mar 28, 2019

drivers: hwinfo: add driver support for Atmel SAM0 device ID #14908

Merged

benpicco force-pushed the same54 branch 2 times, most recently from add0e67 to 7c1be9b Compare April 2, 2019 13:20

dbkinder approved these changes Apr 11, 2019

View reviewed changes

benpicco force-pushed the same54 branch 4 times, most recently from d3887de to d08c678 Compare April 29, 2019 15:43

benpicco added the platform: Microchip SAM Microchip SAM Platform (formerly Atmel SAM) label May 2, 2019

benpicco force-pushed the same54 branch from d08c678 to 7705c2e Compare May 2, 2019 16:52

benpicco and others added 10 commits October 23, 2019 12:20

i2c: sam0: Add support for SAME54

0916126

It shares the same peripheral with other sam0 parts while differing in the clock config. Signed-off-by: Benjamin Valentin <[email protected]>

timer: sam0_rtc_timer: Add support for SAME54

f873256

The RTC peripheral found in the SAMD5x/SAME5x MCUs is very simmilar to the one found in existing sam0 devices with only a few changes to register names and the clock source selection. Signed-off-by: Benjamin Valentin <[email protected]>

drivers: entropy: Add SAM0 entropy driver

afa5973

This adds a driver for the True Random Number Generator found in some Atmel SAM0 SoCs. The Code is based on the driver for the SAM TRNG, but uses different register and clock definitions. Signed-off-by: Benjamin Valentin <[email protected]>

soc: atmel_sam0: Add SAMD51

353a665

This adds supoprt for the Atmel SAMD51 SoC. The SAME5x/SAMD5x is a line of Cortex-M4F MCUs that share peripherals with the sam0 Cortex-M0+ and saml1x Cortex-M23 parts. Signed-off-by: Benjamin Valentin <[email protected]>

soc: atmel_sam0: Add SAME51

ba8cb76

This adds supoprt for the Atmel SAME51 SoC. The SAME5x/SAMD5x is a line of Cortex-M4F MCUs that share peripherals with the sam0 Cortex-M0+ and saml1x Cortex-M23 parts. Signed-off-by: Benjamin Valentin <[email protected]>

soc: atmel_sam0: Add SAME53

df51276

This adds supoprt for the Atmel SAME53 SoC. The SAME5x/SAMD5x is a line of Cortex-M4F MCUs that share peripherals with the sam0 Cortex-M0+ and saml1x Cortex-M23 parts. Signed-off-by: Benjamin Valentin <[email protected]>

drivers: dma: sam0: fix DMA to peripheral transfer on SAMD5x

060878b

We need to select TRIGACT_BLOCK for DMA to peripheral to work. Signed-off-by: Benjamin Valentin <[email protected]>

flash: sam0: Define LOCK_REGIONS in dts

78df765

The amount of lock regions differs between different sam0 MCUs. saml10: 2 saml11: 4 samd2x: 16 saml2x: 16 samd5x: 32 ASF does not provide a definition for this, so create a new one in dts. Signed-off-by: Benjamin Valentin <[email protected]>

gpio: sam0: provide PORT_PMUX_PMUXE_A_Val define

972c8d4

SAMD5x/SAME5x header files do not provide this define anymore. On SAMD2x it was 0, this is still valid. Signed-off-by: Benjamin Valentin <[email protected]>

boards: atsame54_xpro: Add SAM E54 Xplained Pro Evaluation Kit

f3544a4

This adds support for the Atmel SAM E54 Xplained Pro Evaluation Kit. Only basic functionality has been tested so far. Signed-off-by: Benjamin Valentin <[email protected]>

benpicco force-pushed the same54 branch from e24d9cb to f3544a4 Compare October 23, 2019 10:20

benpicco closed this Dec 21, 2019

stephanosio mentioned this pull request Feb 21, 2020

Implement SAM E5X GMAC support #23001

Closed

benpicco deleted the same54 branch February 21, 2020 07:49

soc/same54: Initial support #14685

soc/same54: Initial support #14685

Uh oh!

Conversation

benpicco commented Mar 19, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

zephyrbot commented Mar 19, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Identity/Emails issues

License issues

Uh oh!

Sizurka left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

benpicco commented Mar 21, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Sizurka commented Mar 21, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

benpicco commented Mar 21, 2019

Uh oh!

Sizurka commented Mar 21, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov-io commented Mar 25, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Uh oh!

dbkinder left a comment

Choose a reason for hiding this comment

Uh oh!

benpicco commented Oct 23, 2019

Uh oh!

stephanosio commented Nov 22, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

benpicco commented Nov 22, 2019

Uh oh!

stephanosio commented Nov 22, 2019

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

benpicco commented Mar 19, 2019 •

edited

Loading

zephyrbot commented Mar 19, 2019 •

edited

Loading

Sizurka left a comment •

edited

Loading

benpicco commented Mar 21, 2019 •

edited

Loading

Sizurka commented Mar 21, 2019 •

edited

Loading

Sizurka commented Mar 21, 2019 •

edited

Loading

codecov-io commented Mar 25, 2019 •

edited

Loading

stephanosio commented Nov 22, 2019 •

edited

Loading