GPIO IRQ assignment failure

[cxrider]

Before I file a bug for this, I want to make sure I'm not doing something wrong.

I am trying to enable PPS on GPIO pin PD14 of an Orange Pi PC2.   I've enabled the pps-gpio and uart overlays in my /boot/armbianEnv.txt file and rebooted.  

 

root@ntp0:~# cat /boot/armbianEnv.txt 
verbosity=1
bootlogo=false
console=both
disp_mode=1920x1080p60
overlay_prefix=sun50i-h5
rootdev=UUID=20dbbd2a-5783-4f98-b467-62fa62990ac8
rootfstype=ext4
overlays=pps-gpio uart1 
param_pps_pin=PD14
usbstoragequirks=0x2537:0x1066:u,0x2537:0x1068:u
root@ntp0:~# 

 

On reboot I see the following problem in dmesg:

 

[    5.542593] pps-gpio pps@0: failed to map GPIO to IRQ: -22
[    5.542626] pps-gpio: probe of pps@0 failed with error -22

 

If I 'rmmod/modprobe' the pps-gpio/pps-ldisc modules I see pps0 created:

 

[   54.983158] pps pps0: new PPS source serial1
[   54.983284] pps pps0: source "/dev/ttyS1" added

 

Yet I am not getting PPS data on pps0 even with the gps locked as measured by cgps/gpsmon.

root@ntp0:~# ppstest /dev/pps0
trying PPS source "/dev/pps0"
found PPS source "/dev/pps0"
ok, found 1 source(s), now start fetching data...
time_pps_fetch() error -1 (Connection timed out)
time_pps_fetch() error -1 (Connection timed out)
time_pps_fetch() error -1 (Connection timed out)
^C
root@ntp0:~# 

lqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqklqqqqqqqqqqqqqqqqqqSeen 12/Used  9k
x Time:        2022-04-17T18:35:49.000Z (0) xxGNSS   PRN  Elev   Azim   SNR Usex
x Latitude:         xx.xxxxxxxx N           xxGP  1    1  26.0   49.0  15.0  Y x
x Longitude:        yy.yyyyyyyy W           xxGP  6    6  23.0  169.0  18.0  Y x
x Alt (HAE, MSL):   6353.675,   6407.480 ft xxGP 13   13  28.0  235.0  18.0  Y x
x Speed:             0.11 mph               xxGP 14   14  54.0   72.0  21.0  Y x
x Track (true, var):   332.6,  11.0     deg xxGP 15   15  18.0  270.0  23.0  Y x
x Climb:             0.00 ft/min            xxGP 17   17  84.0   23.0  22.0  Y x
x Status:         3D FIX (9 secs)           xxGP 19   19  70.0  231.0  21.0  Y x
x Long Err  (XDOP, EPX):  0.61, +/- 30.1 ft xxGP 24   24  21.0  314.0  28.0  Y x
x Lat Err   (YDOP, EPY):  0.64, +/- 31.6 ft xxGP 30   30  26.0  151.0  16.0  Y x
x Alt Err   (VDOP, EPV):  0.81, +/- 61.1 ft xxGP  7    7   2.0  146.0   0.0  N x
x 2D Err    (HDOP, CEP):  0.87, +/- 54.2 ft xxGP 21   21   3.0   32.0   0.0  N x
x 3D Err    (PDOP, SEP):  1.19, +/- 74.2 ft xxSB134   47   0.0    0.0   0.0  N x
x Time Err  (TDOP):       0.88              xx                                 x
x Geo Err   (GDOP):       1.87              xx                                 x
x ECEF X, VX:              n/a    n/a       xx                                 x
x ECEF Y, VY:              n/a    n/a       xx                                 x
x ECEF Z, VZ:              n/a    n/a       xx                                 x
x Speed Err (EPS):       +/- 43.0 mph       xx                                 x
x Track Err (EPD):        n/a               xx                                 x
x Time offset:            0.551504007 s     xx                                 x
x Grid Square:            AABBCCDD          xx                                 x
mqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqjmqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqj

(location/grid square redacted)

 

Looking at what the kernel thinks it has for gpio pins enabled I see this:

 

root@ntp0:~# cat /sys/kernel/debug/gpio 
gpiochip1: GPIOs 0-223, parent: platform/1c20800.pinctrl, 1c20800.pinctrl:
 gpio-20  (                    |orangepi:red:status ) out lo 
 gpio-102 (                    |gmac-3v3            ) out hi 
 gpio-166 (                    |cd                  ) in  lo IRQ ACTIVE LOW
 gpio-204 (                    |usb0_id_det         ) in  hi IRQ 

gpiochip0: GPIOs 352-383, parent: platform/1f02c00.pinctrl, 1f02c00.pinctrl:
 gpio-354 (                    |usb0-vbus           ) out lo 
 gpio-355 (                    |sw4                 ) in  hi IRQ ACTIVE LOW
 gpio-362 (                    |orangepi:green:pwr  ) out hi 
root@ntp0:~# 

 

Is this indeed a case of the pin not being assigned an interrupt or is the pps-gpio IRQ mapping message unrelated to the problem?

 

 

Thanks

[Werner]

Providing logs with 

armbianmonitor -u

 helps with troubleshooting and significantly raises chances that issue gets addressed.

[cxrider]

On 4/17/2022 at 1:03 PM, Werner said:

Providing logs with 

armbianmonitor -u

 helps with troubleshooting and significantly raises chances that issue gets addressed.

Thanks for the reply

While I wasn't necessarily asking for assistance I won't say no to help

http://ix.io/3Vo9

It seems as if I am only able to post one message a day here so please understand the delayed replies.    I walked the log file and noticed that it has three different reboots captured; two are with the overlay included and the last is without it.   I wondered whether or not there may have been a conflict with r-gpio-keys so I removed pps-gpio; no change.

I also downloaded kernel source but I haven't quite figured out where I should start debugging the code.   From what I can tell the message in dmesg is coming from line 177 of drivers/pps/client/pps-gpio.c.  That seems to be the result of a return from gpiod_to_irq which seems to be defined in the include file linux/gpio/consumer.h.   It seems the only thing this function does is warn and return -EINVAL which I see EINVAL is defined as 22 in uapi/asm-generic/errno-base.h.     We drop into pps-gpio.c from a call to pps_gpio_probe.    I'm guessing that this is called when kernel loads the pps-gpio module from a modprobe.    Other than looking at the overlays in arch/arm64/boot/dts/allwinner/* I haven't made much headway in figuring out the problem.