On Mon, Oct 02, 2023 at 04:39:42AM -0400, Ruslan Bilovol wrote:
On Mon, Oct 2, 2023 at 2:42 AM Michael Grzeschik <m.grzeschik@xxxxxxxxxxxxxx> wrote:When using uac2 with lockdep enabled, the kernel is throwing this message, due to uninitialized spinlocks. We initialize them now. [ 24.668867] The code is fine but needs lockdep annotation, or maybe [ 24.675878] you didn't initialize this object before use? [ 24.681910] turning off the locking correctness validator. [ 24.688038] CPU: 0 PID: 348 Comm: irq/43-dwc3 Tainted: G C 6.5.0-20230919-1+ #14 [ 24.697866] Hardware name: WolfVision PF5 (DT) [ 24.702831] Call trace: [ 24.705559] dump_backtrace+0xac/0x130 [ 24.709755] show_stack+0x30/0x48 [ 24.713456] dump_stack_lvl+0x60/0xb0 [ 24.717552] dump_stack+0x18/0x28 [ 24.721254] register_lock_class+0x4e8/0x4f8 [ 24.726029] __lock_acquire+0x88/0x2130 [ 24.730314] lock_acquire+0x17c/0x338 [ 24.734403] _raw_spin_lock_irqsave+0x60/0x90 [ 24.739274] u_audio_get_capture_srate+0x34/0x68 [ 24.744436] afunc_setup+0x2d8/0x538 [ 24.748431] composite_setup+0x1a8/0x1ba8 [ 24.752913] configfs_composite_setup+0x88/0xc0 [ 24.757974] dwc3_ep0_interrupt+0x5e8/0xab8 [ 24.762648] dwc3_process_event_buf+0x424/0xbb0 [ 24.767717] dwc3_thread_interrupt+0x4c/0x90 [ 24.772488] irq_thread_fn+0x34/0xb8 [ 24.776484] irq_thread+0x1a0/0x290 [ 24.780379] kthread+0x10c/0x120 [ 24.783985] ret_from_fork+0x10/0x20 Signed-off-by: Michael Grzeschik <m.grzeschik@xxxxxxxxxxxxxx> --- drivers/usb/gadget/function/u_audio.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/gadget/function/u_audio.c b/drivers/usb/gadget/function/u_audio.c index 4a42574b4a7feb..9d9f906adf7c00 100644 --- a/drivers/usb/gadget/function/u_audio.c +++ b/drivers/usb/gadget/function/u_audio.c @@ -1172,6 +1172,9 @@ int g_audio_setup(struct g_audio *g_audio, const char *pcm_name, g_audio->uac = uac; uac->audio_dev = g_audio; + spin_lock_init(&uac->c_prm.lock); + spin_lock_init(&uac->p_prm.lock);These locks are already initialized just a few lines below in this function.
It was probably to obvious, so I missed them.
It seems the issue is because uac2 does u_audio_get_capture_srate() while capture isn't enabled? What is your UAC2 gadget configuration?
Yes, so probably the c_chmask is not setup while we still try to work/configure the capture path. There is actually no application of ours involved in setting the srate. It might be possible that pipewire is changing things that should left alone. I will check that assumption. Thanks, Michael -- Pengutronix e.K. | | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
Attachment:
signature.asc
Description: PGP signature
