Am Freitag, den 07.11.2014, 16:08 +0100 schrieb Sebastian Andrzej Siewior: > On 11/07/2014 03:45 PM, Sebastian Reimers wrote: > > $ /usr/bin/jackd -R -P89 -dalsa -d hw:0,0 -r48000 -p1024 -n3 > > > > With: kernel.sched_rt_runtime_us = -1 > uhhh. Why do you need this? I need low audio latency and with musb in pio mode this was required to get lower frames per period (-p < 512). Otherwise I get: musb_ep_program 896: broken !rx_reinit, ep11 csr 0003 I tested this now with musb DMA mode (default sched_rt_runtime_us value) and in current mainline this didn`t happen anymore, only the cpu load is 2-3x higher compared to PIO (Kernel 3.17.1). Testing PIO and mainline is on my todo list now ;-) But with your patch #2 and default sched_rt_runtime_us the cppi41_irq occur. > There should be two back traces. You are right, sorry, here are a fresh complete log. [ 135.909622] ------------[ cut here ]------------ [ 135.914521] WARNING: CPU: 0 PID: 272 at drivers/dma/cppi41.c:320 cppi41_irq+0x140/0x18c() [ 135.923096] Modules linked in: [ 135.926312] CPU: 0 PID: 272 Comm: jackd Not tainted 3.18.0-rc3 #9 [ 135.932727] [<c0013ba8>] (unwind_backtrace) from [<c00117a0>] (show_stack+0x10/0x14) [ 135.940855] [<c00117a0>] (show_stack) from [<c0036324>] (warn_slowpath_common+0x6c/0x84) [ 135.949345] [<c0036324>] (warn_slowpath_common) from [<c00363d8>] (warn_slowpath_null+0x1c/0x24) [ 135.958562] [<c00363d8>] (warn_slowpath_null) from [<c02ca18c>] (cppi41_irq+0x140/0x18c) [ 135.967055] [<c02ca18c>] (cppi41_irq) from [<c006c768>] (handle_irq_event_percpu+0x38/0x140) [ 135.975908] [<c006c768>] (handle_irq_event_percpu) from [<c006c8ac>] (handle_irq_event+0x3c/0x5c) [ 135.985217] [<c006c8ac>] (handle_irq_event) from [<c006ef10>] (handle_level_irq+0xbc/0x134) [ 135.993979] [<c006ef10>] (handle_level_irq) from [<c006bf70>] (generic_handle_irq+0x2c/0x3c) [ 136.002830] [<c006bf70>] (generic_handle_irq) from [<c006c1e0>] (__handle_domain_irq+0x70/0xc8) [ 136.011956] [<c006c1e0>] (__handle_domain_irq) from [<c0008670>] (omap_intc_handle_irq+0xc8/0xd0) [ 136.021264] [<c0008670>] (omap_intc_handle_irq) from [<c00125a4>] (__irq_usr+0x44/0x60) [ 136.029657] Exception stack(0xdc18ffb0 to 0xdc18fff8) [ 136.034955] ffa0: 000000f0 000000f0 00000000 00000021 [ 136.043533] ffc0: 000000f0 000000f0 081979e7 00000000 00000039 00000000 0001ece8 00000000 [ 136.052109] ffe0: b155e0f8 b1540db8 b1551c84 b6d7f658 40000010 ffffffff [ 136.059044] ---[ end trace 66d4bc7aceea6dda ]--- [ 136.063884] cppi41_irq() q 138 desc 00000000 [ 138.965737] ------------[ cut here ]------------ [ 138.970632] WARNING: CPU: 0 PID: 272 at drivers/dma/cppi41.c:612 cppi41_dma_control+0x2ac/0x2d8() [ 138.979936] Modules linked in: [ 138.983151] CPU: 0 PID: 272 Comm: jackd Tainted: G W 3.18.0-rc3 #9 [ 138.990755] [<c0013ba8>] (unwind_backtrace) from [<c00117a0>] (show_stack+0x10/0x14) [ 138.998885] [<c00117a0>] (show_stack) from [<c0036324>] (warn_slowpath_common+0x6c/0x84) [ 139.007374] [<c0036324>] (warn_slowpath_common) from [<c00363d8>] (warn_slowpath_null+0x1c/0x24) [ 139.016591] [<c00363d8>] (warn_slowpath_null) from [<c02c9790>] (cppi41_dma_control+0x2ac/0x2d8) [ 139.025813] [<c02c9790>] (cppi41_dma_control) from [<c0391068>] (cppi41_dma_channel_abort+0xcc/0x140) [ 139.035496] [<c0391068>] (cppi41_dma_channel_abort) from [<c038c758>] (musb_cleanup_urb.isra.13+0x58/0x11c) [ 139.045717] [<c038c758>] (musb_cleanup_urb.isra.13) from [<c038ca5c>] (musb_urb_dequeue+0xe8/0x128) [ 139.055210] [<c038ca5c>] (musb_urb_dequeue) from [<c0374ff0>] (unlink1+0x2c/0x110) [ 139.063151] [<c0374ff0>] (unlink1) from [<c0375a68>] (usb_hcd_unlink_urb+0x4c/0x84) [ 139.071184] [<c0375a68>] (usb_hcd_unlink_urb) from [<c03fc634>] (deactivate_urbs+0xd8/0xf4) [ 139.079947] [<c03fc634>] (deactivate_urbs) from [<c03fd718>] (snd_usb_endpoint_stop+0x30/0x70) [ 139.088985] [<c03fd718>] (snd_usb_endpoint_stop) from [<c0404b9c>] (stop_endpoints+0x70/0x88) [ 139.097929] [<c0404b9c>] (stop_endpoints) from [<c0404cfc>] (snd_usb_substream_playback_trigger+0x9c/0xb8) [ 139.108058] [<c0404cfc>] (snd_usb_substream_playback_trigger) from [<c03ee79c>] (snd_pcm_do_stop+0x54/0x58) [ 139.118277] [<c03ee79c>] (snd_pcm_do_stop) from [<c03eda30>] (snd_pcm_action_group+0x190/0x20c) [ 139.127401] [<c03eda30>] (snd_pcm_action_group) from [<c03ee604>] (snd_pcm_action+0x54/0xfc) [ 139.136253] [<c03ee604>] (snd_pcm_action) from [<c03f0918>] (snd_pcm_common_ioctl1+0xc84/0xf94) [ 139.145379] [<c03f0918>] (snd_pcm_common_ioctl1) from [<c03f1290>] (snd_pcm_capture_ioctl1+0x1b8/0x464) [ 139.155237] [<c03f1290>] (snd_pcm_capture_ioctl1) from [<c00eeed4>] (do_vfs_ioctl+0x3fc/0x5f0) [ 139.164274] [<c00eeed4>] (do_vfs_ioctl) from [<c00ef134>] (SyS_ioctl+0x6c/0x7c) [ 139.171943] [<c00ef134>] (SyS_ioctl) from [<c000e440>] (ret_fast_syscall+0x0/0x48) [ 139.179881] ---[ end trace 66d4bc7aceea6ddb ]--- > and the transfer is canceled because jack assumed it took too long to > complete? You didn't stop it on your own right? > Correct. -- ALSA: poll time out, polled for 7500118 usecs JackAudioDriver::ProcessAsync: read error, stopping... --
Attachment:
signature.asc
Description: This is a digitally signed message part
