Hi Guennadi, On Thu, Apr 05, 2007 at 12:59:40PM +0200, Guennadi Liakhovetski wrote: > Ok, a simple analysis reveals the recursive spinlock: > > On Thu, 5 Apr 2007, Guennadi Liakhovetski wrote: > > > [<bf12b220>] (ppp_channel_push+0x0/0xc8 [ppp_generic]) from [<bf12bf98>] (ppp_output_wakeup+0x18/0x1c [ppp_generic]) > ===> ^^^^^^^^^^^^^^^^ > > r7 = C38F42BC r6 = C38F4200 r5 = C38F4200 r4 = 00000000 > > ===> spin_lock_bh(&pch->downl); > > > [<bf12bf80>] (ppp_output_wakeup+0x0/0x1c [ppp_generic]) from [<bf132c98>] (irnet_flow_indication+0x38/0x3c [irnet]) > > [<bf132c60>] (irnet_flow_indication+0x0/0x3c [irnet]) from [<bf104e4c>] (irttp_run_tx_queue+0x1c0/0x1d4 [irda]) > > [<bf104c8c>] (irttp_run_tx_queue+0x0/0x1d4 [irda]) from [<bf104f88>] (irttp_data_request+0x128/0x4f8 [irda]) > > r8 = BF121560 r7 = 00000002 r6 = C38F4200 r5 = C21418B8 > > r4 = C21418B8 > > [<bf104e60>] (irttp_data_request+0x0/0x4f8 [irda]) from [<bf1321bc>] (ppp_irnet_send+0x134/0x238 [irnet]) > > [<bf132088>] (ppp_irnet_send+0x0/0x238 [irnet]) from [<bf12a600>] (ppp_push+0x80/0xb8 [ppp_generic]) > > r7 = C3A436E0 r6 = 00000000 r5 = C21418B8 r4 = C1489600 > > [<bf12a580>] (ppp_push+0x0/0xb8 [ppp_generic]) from [<bf12a8d8>] (ppp_xmit_process+0x34/0x50c [ppp_generic]) > ===> ^^^^^^^^ > > r7 = 00000021 r6 = C21418B8 r5 = C1489600 r4 = 00000000 > > ===> spin_lock_bh(&pch->downl); > > > [<bf12a8a4>] (ppp_xmit_process+0x0/0x50c [ppp_generic]) from [<bf12aed8>] (ppp_start_xmit+0x128/0x254 [ppp_generic]) > > [<bf12adb0>] (ppp_start_xmit+0x0/0x254 [ppp_generic]) from [<c0186fa4>] (dev_hard_start_xmit+0x170/0x268) > > [<c0186e34>] (dev_hard_start_xmit+0x0/0x268) from [<c01979b8>] (__qdisc_run+0x60/0x270) > > r8 = C1BBC914 r7 = C21418B8 r6 = 00000000 r5 = C21418B8 > > r4 = C1BBC800 > > [<c0197958>] (__qdisc_run+0x0/0x270) from [<c0187250>] (dev_queue_xmit+0x1b4/0x25c) > > Now, does anyone have an idea how best to fix it? > > 1. Should re-entrance in ppp_channel_push() be prevented, and if yes - at > which level? Or > > 2. Should re-entrance be allowed and only recursive spin_lock_bh() > avoided? IMHO, irnet_flow_indication() should be called asynchronously by irttp_run_tx_queue(), through some bottom-half mechanism. That would fix your locking issues, and that would reduce the time we spend in the IrDA code with this lock taken. I will try to come up with some patches for you later this weekend. Cheers, Samuel. - To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
