RIOT-OS / RIOT

RIOT - The friendly OS for IoT
https://riot-os.org
GNU Lesser General Public License v2.1
4.93k stars 1.99k forks source link

Some I2C devices are not working on iotlab-m3 #15045

Closed aabadie closed 4 years ago

aabadie commented 4 years ago

Description

While testing #15044 on iotlab-m3, I noticed that for some I2C drivers the initialization is failing in examples/default application: l3d4200d, lps331ap, lsm303dlhc.

``` $ IOTLAB_NODE=auto-ssh make BOARD=iotlab-m3 -C examples/default clean flash term make: Entering directory '/work/riot/RIOT/examples/default' Building application "default" for "iotlab-m3" with MCU "stm32". "make" -C /work/riot/RIOT/boards/iotlab-m3 "make" -C /work/riot/RIOT/boards/common/iotlab "make" -C /work/riot/RIOT/core "make" -C /work/riot/RIOT/cpu/stm32 "make" -C /work/riot/RIOT/cpu/cortexm_common "make" -C /work/riot/RIOT/cpu/cortexm_common/periph "make" -C /work/riot/RIOT/cpu/stm32/bootloader "make" -C /work/riot/RIOT/cpu/stm32/periph "make" -C /work/riot/RIOT/cpu/stm32/stmclk "make" -C /work/riot/RIOT/cpu/stm32/vectors "make" -C /work/riot/RIOT/drivers "make" -C /work/riot/RIOT/drivers/at86rf2xx "make" -C /work/riot/RIOT/drivers/isl29020 "make" -C /work/riot/RIOT/drivers/l3g4200d "make" -C /work/riot/RIOT/drivers/lpsxxx "make" -C /work/riot/RIOT/drivers/lsm303dlhc "make" -C /work/riot/RIOT/drivers/netdev_ieee802154 "make" -C /work/riot/RIOT/drivers/periph_common "make" -C /work/riot/RIOT/drivers/saul "make" -C /work/riot/RIOT/drivers/saul/init_devs "make" -C /work/riot/RIOT/sys "make" -C /work/riot/RIOT/sys/auto_init "make" -C /work/riot/RIOT/sys/div "make" -C /work/riot/RIOT/sys/fmt "make" -C /work/riot/RIOT/sys/isrpipe "make" -C /work/riot/RIOT/sys/luid "make" -C /work/riot/RIOT/sys/net/gnrc "make" -C /work/riot/RIOT/sys/net/gnrc/netapi "make" -C /work/riot/RIOT/sys/net/gnrc/netif "make" -C /work/riot/RIOT/sys/net/gnrc/netif/hdr "make" -C /work/riot/RIOT/sys/net/gnrc/netif/ieee802154 "make" -C /work/riot/RIOT/sys/net/gnrc/netif/init_devs "make" -C /work/riot/RIOT/sys/net/gnrc/netreg "make" -C /work/riot/RIOT/sys/net/gnrc/pkt "make" -C /work/riot/RIOT/sys/net/gnrc/pktbuf "make" -C /work/riot/RIOT/sys/net/gnrc/pktbuf_static "make" -C /work/riot/RIOT/sys/net/gnrc/pktdump "make" -C /work/riot/RIOT/sys/net/link_layer/eui_provider "make" -C /work/riot/RIOT/sys/net/link_layer/ieee802154 "make" -C /work/riot/RIOT/sys/net/link_layer/l2util "make" -C /work/riot/RIOT/sys/net/netif "make" -C /work/riot/RIOT/sys/newlib_syscalls_default "make" -C /work/riot/RIOT/sys/od "make" -C /work/riot/RIOT/sys/phydat "make" -C /work/riot/RIOT/sys/pm_layered "make" -C /work/riot/RIOT/sys/ps "make" -C /work/riot/RIOT/sys/random "make" -C /work/riot/RIOT/sys/random/tinymt32 "make" -C /work/riot/RIOT/sys/saul_reg "make" -C /work/riot/RIOT/sys/shell "make" -C /work/riot/RIOT/sys/shell/commands "make" -C /work/riot/RIOT/sys/stdio_uart "make" -C /work/riot/RIOT/sys/tsrb "make" -C /work/riot/RIOT/sys/xtimer text data bss dec hex filename 47100 132 6028 53260 d00c /work/riot/RIOT/examples/default/bin/iotlab-m3/default.elf iotlab-node --jmespath='keys(@)[0]' --format='int' --list saclay,m3,1 --flash /work/riot/RIOT/examples/default/bin/iotlab-m3/default.bin | grep 0 0 ssh -t abadie@saclay.iot-lab.info 'socat - tcp:m3-1.saclay.iot-lab.info:20000' [auto_init_saul] error initializing l3g4200d #0 [auto_init_saul] error initializing lps331ap #0 [auto_init_saul] error initializing lsm303dlhc #0 main(): This is RIOT! (Version: 2020.10-devel-1454-gacf14) Welcome to RIOT! > ```

When run separately with their corresponding test applications (tests/lps331ap, tests/l3g4200d and tests/lsm303dlhc), only lsm303dlhc has issues.

Bisecting points to ac9afbe4f54f5085c45c57ba1ccc4e8cc5526a1c from #12910

This problem is present in 2020.07 and in master.

Steps to reproduce the issue

Build and flash examples/default on an iotlab-m3

Expected results

All I2C sensors are initialized with success

Actual results

Some I2C sensors are not initialized with success: lps331ap, l3g4200d and lsm303dlhc

Versions

aabadie commented 4 years ago

It seems that the issue is only there on m3-1 in Saclay. I tried on a few other nodes and can't reproduce the issue. Let's close this one.