On 2019-07-23 7:55 p.m., Guenter Roeck wrote:
Hi Doug,
On Tue, Jul 23, 2019 at 07:26:34PM -0400, Douglas Gilbert wrote:
Hi,
Apologies if this is not code you maintain or the wrong medium.
No worries. You should copy the respective mailing lists, but I'll do that
for you.
I'm experimenting with a NXP OM 13588 board (USB Type C Arduino shield)
on a Atmel SAMA5D2_XPLAINED board running lk 5.1.18 and Debian 10 .
When I did the "echo" command in the subject line to sysfs I saw the
following:
[28988.750000] Unable to handle kernel NULL pointer dereference at virtual
address 00000028
[28988.750000] pgd = f69149ad
[28988.760000] [00000028] *pgd=00000000
[28988.760000] Internal error: Oops: 5 [#1] THUMB2
[28988.760000] Modules linked in: tcpci tcpm
[28988.760000] CPU: 0 PID: 1882 Comm: bash Not tainted 5.1.18-sama5-armv7-r2 #4
[28988.760000] Hardware name: Atmel SAMA5
[28988.760000] PC is at tcpm_try_role+0x3a/0x4c [tcpm]
[28988.760000] LR is at tcpm_try_role+0x15/0x4c [tcpm]
[28988.760000] pc : [<bf8000e2>] lr : [<bf8000bd>] psr: 60030033
[28988.760000] sp : dc1a1e88 ip : c03fb47d fp : 00000000
[28988.760000] r10: dc216190 r9 : dc1a1f78 r8 : 00000001
[28988.760000] r7 : df4ae044 r6 : dd032e90 r5 : dd1ce340 r4 : df4ae054
[28988.760000] r3 : 00000000 r2 : 00000000 r1 : 00000000 r0 : df4ae044
[28988.760000] Flags: nZCv IRQs on FIQs on Mode SVC_32 ISA Thumb Segment none
[28988.760000] Control: 50c53c7d Table: 3efec059 DAC: 00000051
[28988.760000] Process bash (pid: 1882, stack limit = 0x6a6d4aa5)
[28988.760000] Stack: (0xdc1a1e88 to 0xdc1a2000)
[28988.760000] 1e80: dd05d808 dd1ce340 00000001 00000007
dd1ce340 c03fb4a7
[28988.760000] 1ea0: 00000007 00000007 dc216180 00000000 00000000 c01e1e03
00000000 00000000
[28988.760000] 1ec0: c0907008 dee98b40 c01e1d5d c06106c4 00000000 00000000
00000007 c0194e8b
[28988.760000] 1ee0: 0000000a 00000400 00000000 c01a97db dc22bf00 ffffe000
df4b6a00 df745900
[28988.760000] 1f00: 00000001 00000001 000000dd c01a9c2f 7aeab3be c0907008
00000000 dc22bf00
[28988.760000] 1f20: c0907008 00000000 00000000 00000000 00000000 7aeab3be
00000007 dee98b40
[28988.760000] 1f40: 005dc318 dc1a1f78 00000000 00000000 00000007 c01969f7
0000000a c01a20cb
[28988.760000] 1f60: dee98b40 c0907008 dee98b40 005dc318 00000000 c0196b9b
00000000 00000000
[28988.760000] 1f80: dee98b40 7aeab3be 00000074 005dc318 b6f3bdb0 00000004
c0101224 dc1a0000
[28988.760000] 1fa0: 00000004 c0101001 00000074 005dc318 00000001 005dc318
00000007 00000000
[28988.760000] 1fc0: 00000074 005dc318 b6f3bdb0 00000004 00000007 00000007
00000000 00000000
[28988.760000] 1fe0: 00000004 be800880 b6ed35b3 b6e5c746 60030030 00000001
00000000 00000000
[28988.760000] [<bf8000e2>] (tcpm_try_role [tcpm]) from [<c03fb4a7>]
(preferred_role_store+0x2b/0x5c)
[28988.760000] [<c03fb4a7>] (preferred_role_store) from [<c01e1e03>]
(kernfs_fop_write+0xa7/0x150)
[28988.760000] [<c01e1e03>] (kernfs_fop_write) from [<c0194e8b>]
(__vfs_write+0x1f/0x104)
[28988.760000] [<c0194e8b>] (__vfs_write) from [<c01969f7>] (vfs_write+0x6b/0x104)
[28988.760000] [<c01969f7>] (vfs_write) from [<c0196b9b>] (ksys_write+0x43/0x94)
[28988.760000] [<c0196b9b>] (ksys_write) from [<c0101001>]
(ret_fast_syscall+0x1/0x62)
[28988.760000] Exception stack(0xdc1a1fa8 to 0xdc1a1ff0)
[28988.760000] 1fa0: 00000074 005dc318 00000001 005dc318
00000007 00000000
[28988.760000] 1fc0: 00000074 005dc318 b6f3bdb0 00000004 00000007 00000007
00000000 00000000
[28988.760000] 1fe0: 00000004 be800880 b6ed35b3 b6e5c746
[28988.760000] Code: 4628 e8bd 81f0 6833 (f893) 5028
[28989.020000] ---[ end trace d842183af2e975d9 ]---
The dts file is attached. It is not quite right since OF complains it
can't find that port (but there really isn't a "port" from the
"xplained" board's Point of view with my set up). The full oops is
also attached.
Doug Gilbert
Can you try the following ?
Thanks,
Guenter
---
diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c
index fba32d84e578..6edacf60a226 100644
--- a/drivers/usb/typec/tcpm/tcpm.c
+++ b/drivers/usb/typec/tcpm/tcpm.c
@@ -4114,7 +4114,7 @@ static int tcpm_try_role(const struct typec_capability *cap, int role)
mutex_lock(&port->lock);
if (tcpc->try_role)
ret = tcpc->try_role(tcpc, role);
- if (!ret && !tcpc->config->try_role_hw)
+ if (!ret && (!tcpc->config || !tcpc->config->try_role_hw))
port->try_role = role;
port->try_src_count = 0;
port->try_snk_count = 0;
Tested by: Douglas Gilbert <dgilbert@xxxxxxxxxxxx>
That fixed the oops.
