Add BAM DMUX driver as network interface to the modem #103
Conversation
This allows getting the current state reported by a remoteproc from the SMSM data.
In some configurations, we are required to configure the BAM, but the BAM clk is managed by a remote processor. It is only turned on after coordination with the remote processor, and while turned off accessing the registers results in a bus stall. Attempt to support this mode by only accessing BAM registers when a DMA channel is actually requested. Otherwise, keep the BAM in reset.
BAM DMUX is used as network interface to the modem on some older Qualcomm SoCs like MSM8916 or MSM8974.
BAM DMUX is used as the network interface to the modem.
Add BAM DMUX as network interface to the modem.
Enable BAM DMUX as network interface to the modem.
|
Missing dt bindings for remote |
Yes, I did not work on dt bindings at all yet :) |
|
|
|
On harpia with this kernel I was able to get ofono working for voice/sms. rmnet0 is present and I can do I tried |
You might need to set up the proper APN. @z3ntu wrote on Matrix:
You need a special patched version of ModemManager from https://gitlab.com/Minecrell/pmaports/-/commits/modemmanager-rpmsg. |
Actually. Maybe it failed due to ofono being activated. I rebooted and now mmcli -L returns a modem. So I will proceed with these tips as well. I do have the patched ModemManager. |
|
@Minecrell I pulled the firmware from my actual device "modem" partition and then things started working just fine using ModemManager. It definitely did not work with the default firmware in pmos or the firmware I had been using which worked for voice/sms with ofono: https://gitlab.com/unrznbl/vendor_motorola_harpia/-/tree/xt1607-retus-6.0.1. So good job and now I'll try and get MMS working. Ofono works also. The crux was my firmware. :) |
… lwt hook xmit_check_hhlen() observes the dst for getting the device hard header length to make sure a modified packet can fit. When a helper which changes the dst - such as bpf_skb_set_tunnel_key() - is called as part of the xmit program the accessed dst is no longer valid. This leads to the following splat: BUG: kernel NULL pointer dereference, address: 00000000000000de #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 0 PID: 798 Comm: ping Not tainted 5.18.0-rc2+ #103 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-2 04/01/2014 RIP: 0010:bpf_xmit+0xfb/0x17f Code: c6 c0 4d cd 8e 48 c7 c7 7d 33 f0 8e e8 42 09 fb ff 48 8b 45 58 48 8b 95 c8 00 00 00 48 2b 95 c0 00 00 00 48 83 e0 fe 48 8b 00 <0f> b7 80 de 00 00 00 39 c2 73 22 29 d0 b9 20 0a 00 00 31 d2 48 89 RSP: 0018:ffffb148c0bc7b98 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000240008 RCX: 0000000000000000 RDX: 0000000000000010 RSI: 00000000ffffffea RDI: 00000000ffffffff RBP: ffff922a828a4e00 R08: ffffffff8f1350e8 R09: 00000000ffffdfff R10: ffffffff8f055100 R11: ffffffff8f105100 R12: 0000000000000000 R13: ffff922a828a4e00 R14: 0000000000000040 R15: 0000000000000000 FS: 00007f414e8f0080(0000) GS:ffff922afdc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000000000de CR3: 0000000002d80006 CR4: 0000000000370ef0 Call Trace: <TASK> lwtunnel_xmit.cold+0x71/0xc8 ip_finish_output2+0x279/0x520 ? __ip_finish_output.part.0+0x21/0x130 Fix by fetching the device hard header length before running the BPF code. Fixes: 3a0af8f ("bpf: BPF for lightweight tunnel infrastructure") Signed-off-by: Eyal Birger <eyal.birger@gmail.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Link: https://lore.kernel.org/bpf/20220420165219.1755407-1-eyal.birger@gmail.com
This adds a new, (imo) mainline-quality driver for the BAM DMUX interface that provides the network interface to the modem on MSM8916 and MSM8974. It replaces the old 20k+ driver I ported from downstream a while ago.
I spent almost the entire last week to write it from scratch. It lacks some performance optimizations of the downstream driver but otherwise it has much better code quality. Overall it seems to work just fine. I have already tested it on:
@TravMurav, @Sorixelle Would be great if you could give it a try on wileyfox-cracking/motorola-harpia.
I'm also looking for someone willing to test it on MSM8974. I'm hoping someone will volunteer :)
The instructions for using it are still largely the same, only the kernel config option is different. It's documented in this gist: https://gist.github.com/Minecrell/4cc2bfb9fcae18e294386b0a213907d1