Kernel Crash in ath driver

[sgaud-quic]

Hi,
Crashdump in ath driver with kernel version 6.15 while running tests on below PR :
#29

Issue is seen intermittently
Device : qcs6490-rb3gen2

[ 309.498171] Workqueue: ath11k_wq ath11k_core_restart [ath11k]
[ 309.504092] pstate: 604000c5 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 309.511244] pc : __enable_irq+0x4c/0x7c
[ 309.515206] lr : __enable_irq+0x4c/0x7c
[ 309.519170] sp : ffff8000811abcb0
[ 309.522593] x29: ffff8000811abcb0 x28: 0000000000000000 x27: 0000000000000000
[ 309.529930] x26: ffff319ec0028028 x25: ffff319ec038b100 x24: ffff319ec10a6605
[ 309.537269] x23: 0000000000000000 x22: 0000000000002348 x21: ffff319fad9b8000
[ 309.544615] x20: 000000000000007d x19: ffff319ece01de00 x18: 0000000000000006
[ 309.551954] x17: 000000040044ffff x16: 005000f2b5503510 x15: ffff8000811ab6c0
[ 309.559289] x14: 0000000000000000 x13: ffffdae2fbda7138 x12: 0000000000000e04
[ 309.566626] x11: 00000000000004ac x10: ffffdae2fbdff138 x9 : ffffdae2fbda7138
[ 309.573970] x8 : 00000000ffffefff x7 : ffffdae2fbdff138 x6 : 80000000fffff000
[ 309.581303] x5 : ffff319fbc6c2508 x4 : 0000000000000000 x3 : ffff56bcc11e9000
[ 309.588640] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff319ec5af2300
tionalArea/baseport/Probe_Failure_Check/Probe_Failure_Check.res
[ 309.595982] Call trace:
./suites/Kernel/FunctionalArea/baseport/GIC/GIC.res
./suites/K[ 309.604242] __enable_irq+0x4c/0x7c (P)
[ 309.613928] enable_irq+0x50/0xa8
[ 309.617352] ath11k_pcic_ce_irqs_enable+0x74/0xb0 [ath11k]
[ 309.623008] ath11k_pcic_start+0x30/0x64 [ath11k]
ernel/FunctionalArea/baseport/storage/storage.res
./suites/Kern[ 309.627858] ath11k_core_qmi_firmware_ready+0xe4/0x5a0 [ath11k]
el/FunctionalArea/baseport/kaslr/kaslr.res
./suites/Kernel/Func[ 309.639695] ath11k_core_restart+0xc0/0x198 [ath11k]
[ 309.650547] process_one_work+0x148/0x28c
[ 309.654681] worker_thread+0x2d0/0x3d8
[ 309.658546] kthread+0x12c/0x204
[ 309.661876] ret_from_fork+0x10/0x20
[ 309.665564] ---[ end trace 0000000000000000 ]---
[ 309.670343] ------------[ cut here ]------------

[idlethread]

Is this reproducible with v6.16-rc1 based tree?

[miaoqing-quic]

Yes, but I can't reproduce this issue. Seems CI is using outdated firmware.

[shashim-quic]

Yes, but I can't reproduce this issue. Seems CI is using outdated firmware.

Salendarsingh Gaud (@sgaud-quic) can you please check this and accordingly close this issue if NA.

[ricardosalveti]

Yes, but I can't reproduce this issue. Seems CI is using outdated firmware.

What is the base build that includes the firmware that is being used here? Is the upstream linux-firmware that is outdated or the based image?

[miaoqing-quic]

I used https://snapshots.linaro.org/member-builds/qcomlt/testimages/arm64/1379/initramfs-test-image-qemuarm64-20230321073831-1379.rootfs.cpio.gz as the ramdisk directly. And the firmware is from the latest META.

[shashim-quic]

I used https://snapshots.linaro.org/member-builds/qcomlt/testimages/arm64/1379/initramfs-test-image-qemuarm64-20230321073831-1379.rootfs.cpio.gz as the ramdisk directly. And the firmware is from the latest META.

I think we can conclude this issue if kernel CI and meta-qcom consumption are
all aligned to public firmware available and we don't see any issues.

Let me know. Salendarsingh Gaud (@sgaud-quic) just confirm if firmware usage is aligned.

[shashim-quic]

I used https://snapshots.linaro.org/member-builds/qcomlt/testimages/arm64/1379/initramfs-test-image-qemuarm64-20230321073831-1379.rootfs.cpio.gz as the ramdisk directly. And the firmware is from the latest META.

I think we can conclude this issue if kernel CI and meta-qcom consumption are all aligned to public firmware available and we don't see any issues.

Let me know. @sgaud-quic just confirm if firmware usage is aligned.

this is no more observed and firmware usage in kernel CI is aligned to meta-qcom