Wakeup on CAN/CAN-FD App Note

jmelnych_ford · ‎03-11-2024

Hi,

I am using an NXP TJA1044 as a CAN-FD transceiver for an S32G2. I am trying to see if there's any existing app notes on how to implement wake up on CAN.

Thanks,

--Jordan

jmelnych_ford · ‎03-19-2024

I made this modification (tried both `status="okay"` and `status="disabled"`) but was not able to stay asleep.

Snip of my modifications to arm-trusted-firmware s32cc.dtsi. I tried both status = "disabled" and status = "okay"

```

wkpu: wkpu@40090000 {

compatible = "nxp,s32cc-wkpu";

reg = <0x0 0x40090000 0x0 0x10000>, /* WKPU */

<0x0 0x4007cb04 0x0 0x4>; /* S32G_STDBY_GPR */

/*

* Enable RTC and GPIOs as wake-up source

*/

nxp,irqs =

<S32GEN1_WKPU_RTC_IRQ S32GEN1_WKPU_IRQ_RISING S32GEN1_WKPU_PULL_DIS>,

<S32GEN1_WKPU_EXT_IRQ(0) S32GEN1_WKPU_IRQ_FALLING S32GEN1_WKPU_PULL_DIS>, /* PC_11 */

<S32GEN1_WKPU_EXT_IRQ(1) S32GEN1_WKPU_IRQ_FALLING S32GEN1_WKPU_PULL_DIS>, /* PJ_02 */

<S32GEN1_WKPU_EXT_IRQ(2) S32GEN1_WKPU_IRQ_FALLING S32GEN1_WKPU_PULL_DIS>, /* PJ_04 */

<S32GEN1_WKPU_EXT_IRQ(4) S32GEN1_WKPU_IRQ_FALLING S32GEN1_WKPU_PULL_DIS>, /* PJ_06 */

<S32GEN1_WKPU_EXT_IRQ(9) S32GEN1_WKPU_IRQ_FALLING S32GEN1_WKPU_PULL_DIS>, /* PK_02 */

<S32GEN1_WKPU_EXT_IRQ(22) S32GEN1_WKPU_IRQ_FALLING S32GEN1_WKPU_PULL_DIS>; /* PA_10 */

nxp,warm-boot = <S32GEN1_WKPU_LONG_BOOT>;

status = "okay";

};

```

Error that I get when trying to sleep (it's a little jumbled but it doesn't sleep and I get an i/o error)

```

root@s32g274aucimdev1:/sys/devices/platform/soc/40060000.rtc/rtc/rtc0/power# echo mem > /sys/power/state
[ 294.077880] PM: suspend entry (deep)
[ 294.080347] Filesystems sync: 0.002 seconds
[ 294.081136] Freezing user space processes ... (elapsed 0.001 seconds) done.
[ 294.082458] OOM killer disabled.
[ 294.082462] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
[ 294.083617] printk: Suspending console(s) (use no_console_suspend to debug)
[ 294.197500] pfeng 46000000.pfe: Suspending driver
[ 294.197866] pfeng 46000000.pfe pfe0: Link is Down
[ 294.197972] pfeng 46000000.pfe pfe0: Link is Up - 1Gbps/Full - flow control off
[ 294.198025] pfeng 46000000.pfe pfe0: Link is Down
[ 294.198427] pfeng 46000000.pfe pfe1: Link is Down
[ 294.198488] pfeng 46000000.pfe pfe1: Link is Up - 1Gbps/Full - flow control off
[ 294.198527] pfeng 46000000.pfe pfe1: Link is Down
[ 294.198879] pfeng 46000000.pfe pfe2: Link is Down
[ 294.198897] pfeng 46000000.pfe pfe2: Link is Up - 100Mbps/Full - flow control off
[ 294.198937] pfeng 46000000.pfe pfe2: Link is Down
[ 294.199102] pfeng 46000000.pfe: HIF3 not enabled, skipped
[ 294.199114] pfeng 46000000.pfe: HIF2 stopped
[ 294.199316] pfeng 46000000.pfe: HIF2 disabled
[ 294.199326] pfeng 46000000.pfe: HIF1 stopped
[ 294.199566] pfeng 46000000.pfe: HIF1 disabled
[ 294.199576] pfeng 46000000.pfe: HIF0 stopped
[ 294.199841] pfeng 46000000.pfe: HIF0 disabled
[ 295.173469] pfeng 46000000.pfe: Global poller finished
[ 295.173550] pfeng 46000000.pfe: Removing default logical interface (pfe0, parent: emac0)
[ 295.173618] pfeng 46000000.pfe: pfe0 (p0x00000000384bf720) removed from emac0 (p0x00000000f100ce35)
[ 295.173781] pfeng 46000000.pfe: Removing default logical interface (hif0-logif, parent: hif0)
[ 295.173839] pfeng 46000000.pfe: hif0-logif (p0x0000000079b9b168) removed from hif0 (p0x0000000009947d5a)
[ 295.173998] pfeng 46000000.pfe: Removing default logical interface (pfe1, parent: emac1)
[ 295.174056] pfeng 46000000.pfe: pfe1 (p0x00000000bff17df6) removed from emac1 (p0x00000000b8a05c86)
[ 295.174211] pfeng 46000000.pfe: Removing default logical interface (hif1-logif, parent: hif1)
[ 295.174268] pfeng 46000000.pfe: hif1-logif (p0x000000006dd447ee) removed from hif1 (p0x000000007e3d42d0)
[ 295.174427] pfeng 46000000.pfe: Removing default logical interface (pfe2, parent: emac2)
[ 295.174485] pfeng 46000000.pfe: pfe2 (p0x0000000068dd0b02) removed from emac2 (p0x00000000158146c3)
[ 295.174642] pfeng 46000000.pfe: Removing default logical interface (hif2-logif, parent: hif2)
[ 295.174700] pfeng 46000000.pfe: hif2-logif (p0x00000000ad97132a) removed from hif2 (p0x000000007be1b77a)
[ 296.029709] pfeng 46000000.pfe: Stopping rtable worker...
[ 296.029763] pfeng 46000000.pfe: rtable worker stopped
[ 296.030165] pfeng 46000000.pfe: PFE Platform stopped
[ 296.061606] s32cc-rtc 40060000.rtc: RTC timer expired before entering in suspend
[ 296.061617] PM: dpm_run_callback(): platform_pm_suspend+0x0/0x70 returns -5
[ 296.061642] s32cc-rtc 40060000.rtc: PM: failed to suspend: error -5
[ 296.061653] PM: Some devices failed to suspend, or early wake event detected
[ 296.062626] s32cc_fccu 4030c000.fccu: FCCU status is 0 (normal)
[ 296.081200] phy-s32cc-serdes 40480000.serdes: Using mode 0 for SerDes subsystem
[ 296.081370] phy-s32cc-serdes 44180000.serdes: Using mode 3 for SerDes subsystem
[ 296.108669] phy-s32cc-serdes 44180000.serdes: Unstable RX detected on XPCS1
[ 296.109299] phy-s32cc-serdes 44180000.serdes: Unstable RX detected on XPCS0
[ 296.109446] pfeng 46000000.pfe: Resuming driver
[ 296.109494] pfeng 46000000.pfe: PFE port coherency enabled, mask 0x1e
[ 296.109646] pfeng 46000000.pfe: Interface selected: EMAC0: 0x4 EMAC1: 0x4 EMAC2: 0x7
[ 296.110331] pfeng 46000000.pfe: PFE controller reset done
[ 296.110550] pfeng 46000000.pfe: PFE CBUS p0x46000000 mapped @ v0xffffffc011000000 (0x1000000 bytes)
[ 296.110561] pfeng 46000000.pfe: HW version 0x50300
[ 296.110569] pfeng 46000000.pfe: Silicon S32G2
[ 296.111674] pfeng 46000000.pfe: PFE_ERRORS:Parity instance created
[ 296.111684] pfeng 46000000.pfe: PFE_ERRORS:Watchdog instance created
[ 296.111694] pfeng 46000000.pfe: BMU1 buffer base: p0xc0000000
[ 296.111799] pfeng 46000000.pfe: BMU2 buffer base: p0x34000000 (0x80000 bytes)
[ 296.113083] pfeng 46000000.pfe: register IRQ 71 by name 'PFE BMU IRQ'
[ 296.113273] pfeng 46000000.pfe: BMU_EMPTY_INT (BMU @ p0x00000000a966d5e7). Pool ready.
[ 296.113280] pfeng 46000000.pfe: Firmware .elf detected
[ 296.113288] pfeng 46000000.pfe: BMU_EMPTY_INT (BMU @ p0x000000005a631beb). Pool ready.
[ 296.113290] pfeng 46000000.pfe: Uploading CLASS firmware
[ 296.113299] pfeng 46000000.pfe: Selected FW loading OPs to load 8 PEs in parallel
[ 296.133028] pfeng 46000000.pfe: pfe_ct.h file version"92367c0e25f21f49217a9b08168ad2c8"
[ 296.149800] pfeng 46000000.pfe: [FW VERSION] 1.6.0, Build: Mar 15 2023, 12:37:54 (), ID: 0x31454650
[ 296.150171] pfeng 46000000.pfe: EMAC timestamp external mode bitmap: 0
[ 296.150222] pfeng 46000000.pfe: Uploading UTIL firmware
[ 296.150227] pfeng 46000000.pfe: Selected FW loading OPs to load 1 PEs in parallel
[ 296.152699] pfeng 46000000.pfe: pfe_ct.h file version"92367c0e25f21f49217a9b08168ad2c8"
[ 296.160448] pfeng 46000000.pfe: FW feature: drv_run_on_g3
[ 296.160456] pfeng 46000000.pfe: FW feature: jumbo_frames
-sh[ 296.160464] pfeng 46000000.pfe: FW feature: software_vlan_table
: [ 296.160471] pfeng 46000000.pfe: FW feature: timestamping
e[ 296.160476] pfeng 46000000.pfe: FW feature: qos_mapping
c[ 296.160482] pfeng 46000000.pfe: FW feature: core_functionality
h[ 296.160488] pfeng 46000000.pfe: FW feature: extended_features
o[ 296.160495] pfeng 46000000.pfe: FW feature: flexible_router
:[ 296.160500] pfeng 46000000.pfe: FW feature: validate_hif_csum
[ 296.160507] pfeng 46000000.pfe: FW feature: err051211_workaround
w[ 296.160513] pfeng 46000000.pfe: FW feature: IPsec
r[ 296.160519] pfeng 46000000.pfe: FW feature: l2_bridge_aging
i[ 296.160526] pfeng 46000000.pfe: FW feature: receive_malformed
t[ 296.160532] pfeng 46000000.pfe: FW feature: ptp_conf_check
e[ 296.160538] pfeng 46000000.pfe: FW feature: vlan_conf_check
[ 296.160545] pfeng 46000000.pfe: FW feature: hash_load_spread
e[ 296.160552] pfeng 46000000.pfe: FW feature: egress_vlan
r[ 296.160558] pfeng 46000000.pfe: FW feature: ingress_vlan
r[ 296.160564] pfeng 46000000.pfe: FW feature: safety
o[ 296.162875] pfeng 46000000.pfe: VLAN ID incorrect or not set. Using default VLAN ID = 0x01.
r[ 296.162881] pfeng 46000000.pfe: VLAN stats size incorrect or not set. Using default VLAN stats size = 20.
:[ 296.162976] pfeng 46000000.pfe: Software vlan hash table @ p0x20001208
[ 296.163162] pfeng 46000000.pfe: Fall-back bridge domain @ 0x20000a44 (class)
I[ 296.163169] pfeng 46000000.pfe: Default bridge domain @ 0x20000a3c (class)
n[ 296.163523] pfeng 46000000.pfe: Routing table created, Hash Table @ p0x34080000, Pool @ p0x34088000 (65536 bytes)
p[ 296.163801] pfeng 46000000.pfe: Feature err051211_workaround: DISABLED
u[ 296.165477] pfeng 46000000.pfe: HIF0 enabled
t[ 296.165703] pfeng 46000000.pfe: HIF1 enabled
/[ 296.165969] pfeng 46000000.pfe: HIF2 enabled
o[ 296.165975] pfeng 46000000.pfe: HIF3 not configured, skipped
u[ 296.166014] pfeng 46000000.pfe: TX clocks on EMAC0 restarted
t[ 296.166042] pfeng 46000000.pfe: RX clocks on EMAC0 restarted
p[ 296.166052] pfeng 46000000.pfe pfe0: Host LLTX disabled
u[ 296.166476] pfeng 46000000.pfe pfe0: Enable HIF0
t[ 296.166644] pfeng 46000000.pfe pfe0: setting MAC addr: 00:01:be:be:ef:11
[ 296.166688] pfeng 46000000.pfe pfe0: PTP HW addend 0x80000000, max_adj configured to 46566128 ppb
e[ 296.166699] pfeng 46000000.pfe: IEEE1588: Input Clock: 200000000Hz, Output: 100000000Hz, Accuracy: 10.0ns
r[ 296.167297] pfeng 46000000.pfe pfe0: Registered PTP HW clock successfully on EMAC0
r[ 296.167310] pfeng 46000000.pfe: HIF0 started
o[ 296.167663] pfeng 46000000.pfe pfe0: configuring for fixed/sgmii link mode
r[ 296.167821] pfeng 46000000.pfe pfe0: Link is Up - 1Gbps/Full - flow control off
[ 296.167894] pfeng 46000000.pfe: TX clocks on EMAC1 restarted

[ 296.167922] pfeng 46000000.pfe: RX clocks on EMAC1 restarted
[ 296.167930] pfeng 46000000.pfe pfe1: Host LLTX disabled
[ 296.168345] pfeng 46000000.pfe pfe1: Enable HIF1
[ 296.168513] pfeng 46000000.pfe pfe1: setting MAC addr: 00:01:be:be:ef:22
[ 296.168547] pfeng 46000000.pfe pfe1: PTP HW addend 0x80000000, max_adj configured to 46566128 ppb
[ 296.168556] pfeng 46000000.pfe: IEEE1588: Input Clock: 200000000Hz, Output: 100000000Hz, Accuracy: 10.0ns
[ 296.169034] pfeng 46000000.pfe pfe1: Registered PTP HW clock successfully on EMAC1
[ 296.169045] pfeng 46000000.pfe: HIF1 started
[ 296.169388] pfeng 46000000.pfe pfe1: configuring for fixed/sgmii link mode
r[ 296.169517] pfeng 46000000.pfe pfe1: Link is Up - 1Gbps/Full - flow control off
o[ 296.169554] pfeng 46000000.pfe pfe2: Host LLTX disabled
o[ 296.169961] pfeng 46000000.pfe pfe2: Enable HIF2
t[ 296.170128] pfeng 46000000.pfe pfe2: setting MAC addr: 00:01:be:be:ef:33
@[ 296.170151] pfeng 46000000.pfe pfe2: PTP HW addend 0x80000000, max_adj configured to 46566128 ppb
s[ 296.170162] pfeng 46000000.pfe: IEEE1588: Input Clock: 200000000Hz, Output: 100000000Hz, Accuracy: 10.0ns
3[ 296.170611] pfeng 46000000.pfe pfe2: Registered PTP HW clock successfully on EMAC2
2[ 296.170621] pfeng 46000000.pfe: HIF2 started
g[ 296.170964] pfeng 46000000.pfe pfe2: configuring for fixed/rmii link mode
2[ 296.170995] pfeng 46000000.pfe pfe2: Link is Up - 100Mbps/Full - flow control off
7[ 296.278873] OOM killer enabled.
4[ 296.278882] Restarting tasks ... done.
a[ 296.289699] PM: suspend exit

```

Any idea why I can't go to sleep?

Thanks,

--Jordan

Daniel-Aguirre · ‎03-19-2024

Hi,

Can you confirm which BSP version are you using? Are you using any NXP reference board? Also, did you replicate successfully the example shown under the BSP User Manual?

If the platform is waking up, it should be that the wake-up routine was triggered, hence it wakes up.

Please, let us know.

jmelnych_ford · ‎03-19-2024

BSP 36. This is on a custom board so I can't use the same pin in the example. I did check that the pins are wkpu compatible in the io-mux spreadsheet

Daniel-Aguirre · ‎03-19-2024

Hi,

Thanks for your feedback. Can you confirm that your I/O's are not being turned down? Do you have any pull-ups on the signals?

You could change the IRQ type to RISING, and apply voltage to the signal.

But still, we can recommend replicating as close as possible the example shown under the BSP User Manual, then continue applying modifications.

Please, let us know.

jmelnych_ford · ‎03-20-2024

Hey Daniel,

I tracked it down to a yocto issue. Looks like the patches weren't getting applied..... I'm now able to wakeup on falling edge as expected. However when I wake back up I get these errors:

```

[ 611.391812] Restarting tasks ... done.
[ 611.409130] PM: suspend exit
[ 612.389538] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 613.413533] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 614.437538] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 615.461534] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 616.485530] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 617.509535] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 618.533527] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 618.533544] pfeng 46000000.pfe: HM event storage is full, no further events will be stored.
[ 619.557528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 620.581529] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 621.605530] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 622.629529] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 623.653529] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 624.677530] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 625.701528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 626.725528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 627.749525] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 628.773528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 629.797529] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 630.821529] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 631.845528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 632.869528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 633.893528] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 634.917529] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 635.941572] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 636.965493] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 637.993492] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 639.013526] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 640.037547] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 641.061491] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 642.085526] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 643.109493] pfeng 46000000.pfe: ERR: (PFENG_DEV) event 15 - EMAC2 GPI Watchdog trigered: [pfe_wdt_csr.c:126]
[ 643.109518] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 644.133493] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]
[ 645.157491] pfeng 46000000.pfe: ERR: (EMAC0) event 42 - APP FSM timeout error: [pfe_emac_csr.c:1635]

```

Daniel-Aguirre · ‎03-21-2024

Hi,

Thanks for your feedback. On regards of those messages, we see the following information was provided:

"The root cause of this issue is that TF-A clocking framework is not able to inherit state after MCAL previously running on M7 when M7 perform a functional rest. The destructive reset put HW into defined state, so there is no issue."

Please, let us know.

jmelnych_ford · ‎03-12-2024

Thanks for the response Daniel,

I have the LLCE driver enabled for CAN. Does that mean it should "just work" already or do I need to make any changes to the device tree or .config?

Daniel-Aguirre · ‎03-12-2024

Hi,

Thanks for your feedback. The LLCE-FW does not relate to the WKUP functionality of a specific pin. Since we understand that you are working with Linux (since you are mentioning the usage of a device tree), you should enable the WKUP functionality by adding the specific pin to the ATF device tree.

Some steps should be shown under the specific BSP version User Manual. Below will be a capture of the BSP38.0 User Manual:

This may vary from BSP version to BSP version, hence we recommend looking into your specific BSP version User Manual.

Please, let us know.

Daniel-Aguirre · ‎03-11-2024

Hi,

For TJA1044 support, help us opening a ticket under the NXP online services, since these PHY's are not supported under the public community. We do apologize for the inconvenience.

As for an application note talking about wake on CAN for S32G2 processor, we don't see any information. We do apologize. There is though some steps available that might be useful:

"

1. The CAN phy receives the remote wake-up frame
2. The CAN phy RXD pin outputs a low level.

3. S32G LLCE_CANx_RX pin as the wake-up source to wake up S32G. (S32G LLCE_CANx_RXD connected to CAN PHY RXD pin)

4. S32G makes the MODEB pin output high

5. VR5510 wake-up and power supply on.

6. S32G execution application

"

Please, let us know.