Kernel Warning when running PRU codes

I am running some PRU codes from this repository:
https://github.com/PierrickRauby/BBAI-PRU-I2C

The code is using the PRU to control I2C4 and read the WHO_AM_I register of and I2C accelerometer on the bus, this register value is then send back to user space via rpmsg

The code behaves as expected and the user space receive a message from the PRU, however, inspecting dmesg -Hw I see this trace:
[Sep 1 13:23] remoteproc remoteproc5: powering up 4b238000.pru
[ +0.005504] remoteproc remoteproc5: Booting fw image am57xx-pru1_1-fw, size 124524
[ +0.004561] pruss 4b200000.pruss: configured system_events[63-0] = 0x00000000.000c0000
[ +0.000030] pruss 4b200000.pruss: configured intr_channels = 0x0000000a host_intr = 0x0000000a
[ +0.023871] virtio_rpmsg_bus virtio4: creating channel rpmsg-pru addr 0x1e
[ +0.000393] virtio_rpmsg_bus virtio4: rpmsg host is online
[ +0.000146] remoteproc remoteproc5: registered virtio4 (type 7)
[ +0.000023] remoteproc remoteproc5: remote processor 4b238000.pru is now up
[ +0.230683] rpmsg_pru virtio4.rpmsg-pru.-1.30: new rpmsg_pru device: /dev/rpmsg_pru30
[Sep 1 13:24] ------------[ cut here ]------------
[ +0.000051] WARNING: CPU: 0 PID: 2274 at drivers/bus/omap_l3_noc.c:147 l3_interrupt_handler+0x368/0x3a4
[ +0.000022] 44000000.ocp:L3 Custom Error: MASTER PRUSS1 PRU2 TARGET L4_PER1_P3 (Read): Data Access in User mode during Functional access
[ +0.000014] Modules linked in: rpmsg_pru pru_rproc pruss_intc rpmsg_rpc pruss rpmsg_proto bnep pruss_soc_bus uio_pruss_shmem ti_vpe ti_sc videobuf2_dma_contig ti_csc ti_vpdma v4l2_mem2mem v4l2_common videobuf2_memops videobuf2_v4l2 videobuf2_core videodev btsdio bluetooth media joydev ecdh_generic brcmfmac stmpe_adc brcmutil cfg80211 omap_remoteproc virtio_rpmsg_bus rpmsg_core usb_f_ncm usb_f_mass_storage usb_f_rndis u_ether uio_pdrv_genirq uio libcomposite cmemk(O)
[ +0.000494] CPU: 0 PID: 2274 Comm: user_space Tainted: G W O 4.14.108-ti-r143 #1buster
[ +0.000015] Hardware name: Generic DRA74X (Flattened Device Tree)
[ +0.000048] [] (unwind_backtrace) from [] (show_stack+0x20/0x24)
[ +0.000033] [] (show_stack) from [] (dump_stack+0x80/0x94)
[ +0.000032] [] (dump_stack) from [] (__warn+0xec/0x114)
[ +0.000027] [] (__warn) from [] (warn_slowpath_fmt+0x58/0x74)
[ +0.000033] [] (warn_slowpath_fmt) from [] (l3_interrupt_handler+0x368/0x3a4)
[ +0.000037] [] (l3_interrupt_handler) from [] (__handle_irq_event_percpu+0x84/0x2d0)
[ +0.000030] [] (__handle_irq_event_percpu) from [] (handle_irq_event_percpu+0x3c/0x90)
[ +0.000028] [] (handle_irq_event_percpu) from [] (handle_irq_event+0x48/0x6c)
[ +0.000031] [] (handle_irq_event) from [] (handle_fasteoi_irq+0xd0/0x178)
[ +0.000031] [] (handle_fasteoi_irq) from [] (generic_handle_irq+0x34/0x44)
[ +0.000027] [] (generic_handle_irq) from [] (__handle_domain_irq+0x8c/0xfc)
[ +0.000028] [] (__handle_domain_irq) from [] (gic_handle_irq+0x4c/0x88)
[ +0.000027] [] (gic_handle_irq) from [] (__irq_svc+0x6c/0xa8)
[ +0.000017] Exception stack(0xd06ddf60 to 0xd06ddfa8)
[ +0.000027] df60: d06ddfec 0048a748 b6e55d24 80070010 b6e55d24 80070010 ffffffff 10c5387d
[ +0.000029] df80: 10c5387d c0ccfa8c 0049b000 00000000 01919150 d06ddfb0 b6e0e746 c0ccf888
[ +0.000017] dfa0: 40070013 ffffffff
[ +0.000028] [] (__irq_svc) from [] (__und_usr+0x48/0x6c)
[ +0.000018] —[ end trace 328dd8d280ee6328 ]—
Is there anything I should change in the device trees to avoid DRA74X complaining? I have tried to disable I2C4 from in the dtb but it does not fix the problem.
Does anyone have some inputs,
Best
Pierrick