Re: intel 537 with kernel 3.1.x

[Philippe Vouters <philippe.vouters@xxxxxxxxxxx>]

Patrick,

Can you really grant me this call stack is caused by the small test 
program which just opens a file descriptor to /dev/modem and immediately 
closes it ???? If you are indeed correct and unlike me during my tests, 
you are getting non reprodicible effects. Beleive my long IT customers 
support experience, when you are getting non reprodicible effects, 
especially when it happens that the very same binary executable produces 
various manisfestations, the root cause is almost always hardware and no 
longer software.

Just take time to compare the call stack you were getting with efax that 
you mailed me which obviously highlights a problem on close and this 
call stack. They can't strictly be compared although the small test 
program is a strong simplification of efax to better highlight the problem.

So I suggest you to only use this test program for the moment and to run 
it as many times as you can observing the call trace it produces each 
time. If, for the exact same sollicitation, you get various symptoms, 
either the problem is time dependant (a race condition) or the problem 
is hardware.

Are you still at IRQ #19 instead of #20 as I strongly recommended you ? 
If yes, do move again the modem to the other PCI slot prior to any tests 
with the test program. The other hardware interrupting at IRQ #19 is 
likely to interfere if also sollicited when your Tiger phone modem is in 
use.

Philippe

Le 06/01/2012 16:14, Patrick Boyadjian a écrit :
> Here is the output of your test code below...
>
>
> Jan  6 10:12:47 pb3000 kernel: [  538.561307] 537: Loaded
> Jan  6 10:12:49 pb3000 kernel: [  540.182152] BUG: scheduling while atomic: kworker/1:4/446/0x10000200
> Jan  6 10:12:49 pb3000 kernel: [  540.182204] Modules linked in: Intel537 8021q garp bridge stp llc it87 hwmon_vid nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack joydev snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm ppdev microcode parport_pc parport netjet isdnhdlc crc_ccitt r8169 mii mISDNipac mISDN_core i2c_i801 snd_timer iTCO_wdt serio_raw iTCO_vendor_support snd soundcore snd_page_alloc uinput usb_storage firewire_ohci firewire_core crc_itu_t i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: scsi_wait_scan]
> Jan  6 10:12:49 pb3000 kernel: [  540.182267] Pid: 446, comm: kworker/1:4 Not tainted 3.1.6-1.fc16.i686 #1
> Jan  6 10:12:49 pb3000 kernel: [  540.182270] Call Trace:
> Jan  6 10:12:49 pb3000 kernel: [  540.182280]  [<c090411b>] ? printk+0x2d/0x2f
> Jan  6 10:12:49 pb3000 kernel: [  540.182286]  [<c0903eb8>] __schedule_bug+0x5e/0x64
> Jan  6 10:12:49 pb3000 kernel: [  540.182292]  [<c090b4b6>] __schedule+0x876/0x8b0
> Jan  6 10:12:49 pb3000 kernel: [  540.182298]  [<c04e57bc>] ? __alloc_pages_nodemask+0xfc/0x6f0
> Jan  6 10:12:49 pb3000 kernel: [  540.182304]  [<c04d295f>] ? cpupri_set+0xbf/0x100
> Jan  6 10:12:49 pb3000 kernel: [  540.182310]  [<c041009f>] ? p4_pmu_enable_event+0xcf/0x130
> Jan  6 10:12:49 pb3000 kernel: [  540.182317]  [<c0651007>] ? __percpu_counter_add+0x67/0xa0
> Jan  6 10:12:49 pb3000 kernel: [  540.182323]  [<c090eae0>] ? do_debug+0x180/0x180
> Jan  6 10:12:49 pb3000 kernel: [  540.182328]  [<c04449ab>] __cond_resched+0x1b/0x30
> Jan  6 10:12:49 pb3000 kernel: [  540.182333]  [<c090b569>] _cond_resched+0x29/0x30
> Jan  6 10:12:49 pb3000 kernel: [  540.182339]  [<c0517a4d>] kmem_cache_alloc+0x11d/0x140
> Jan  6 10:12:49 pb3000 kernel: [  540.182344]  [<c0411a3c>] ? p4_pmu_disable_all+0x5c/0x70
> Jan  6 10:12:49 pb3000 kernel: [  540.182349]  [<c040b18f>] ? init_fpu+0x7f/0x120
> Jan  6 10:12:49 pb3000 kernel: [  540.182353]  [<c090eae0>] ? do_debug+0x180/0x180
> Jan  6 10:12:49 pb3000 kernel: [  540.182357]  [<c040b18f>] init_fpu+0x7f/0x120
> Jan  6 10:12:49 pb3000 kernel: [  540.182362]  [<c090eae0>] ? do_debug+0x180/0x180
> Jan  6 10:12:49 pb3000 kernel: [  540.182367]  [<c0404374>] math_state_restore+0x24/0x50
> Jan  6 10:12:49 pb3000 kernel: [  540.182371]  [<c090eaf0>] do_device_not_available+0x10/0x20
> Jan  6 10:12:49 pb3000 kernel: [  540.182376]  [<c090e0e7>] error_code+0x67/0x6c
> Jan  6 10:12:49 pb3000 kernel: [  540.182576]  [<f9f21c11>] ? w7_ippsFFTInitAlloc_C_16sc+0x3d/0x1d6 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.182689]  [<f9da4bd8>] ? Phase2InitAlloc+0xb8/0x100 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.182791]  [<f9d76039>] ? OmpInitAlloc+0x19/0x90 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.182889]  [<f9d7063d>] ? ModemInit+0x2ed/0x380 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.182988]  [<f9d6f39d>] ? ModemTasks+0x12d/0x160 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.183001]  [<f9d672a1>] ? ModemTask+0x11/0x20 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.183269]  [<f9d63af2>] ? linux_MT_AcquireLock_mt_dsp+0x12/0x20 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.183488]  [<f9d6585d>] ? InterruptDSP+0x1d/0x30 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.183785]  [<f9fbb676>] ? set_register+0x30/0x34 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.184061]  [<f9ffc688>] ? dspdrv_PutCRAM+0x74/0x7c [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.184240]  [<f9d63dfa>] ? linux_RSAGetExclusive+0xea/0x100 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.184487]  [<fa003b5d>] ? cp_dsp+0x3f/0xf2 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.184807]  [<f9fbd1d9>] ? BLK_shell+0x39/0xa4 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.185068]  [<fa000b45>] ? SendPacketOut+0x23/0x5a [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.185314]  [<fa000b0a>] ? send_pkt_to_dsp+0xe/0x26 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.185562]  [<fa00275d>] ? ResetDspIdle+0x87/0xde [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.185837]  [<f9ffce1a>] ? mt_init+0x32/0xf0 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.186087]  [<f9fd9bf6>] ? master_init+0x50/0x16a [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.186352]  [<f9d62892>] ? clm_bg+0x32/0x180 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.186543]  [<f9d62a2d>] ? RTS_Task+0x4d/0x60 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.186626]  [<c046567a>] ? process_one_work+0xfa/0x390
> Jan  6 10:12:49 pb3000 kernel: [  540.186682]  [<c04637b0>] ? do_work_for_cpu+0x20/0x20
> Jan  6 10:12:49 pb3000 kernel: [  540.186846]  [<f9d629e0>] ? clm_bg+0x180/0x180 [Intel537]
> Jan  6 10:12:49 pb3000 kernel: [  540.186903]  [<c0466371>] ? worker_thread+0x121/0x2d0
> Jan  6 10:12:49 pb3000 kernel: [  540.186935]  [<c0466250>] ? manage_workers+0x1f0/0x1f0
> Jan  6 10:12:49 pb3000 kernel: [  540.186967]  [<c046a1cd>] ? kthread+0x6d/0x80
> Jan  6 10:12:49 pb3000 kernel: [  540.187023]  [<c046a160>] ? kthread_worker_fn+0x140/0x140
> Jan  6 10:12:49 pb3000 kernel: [  540.187042]  [<c09148be>] ? kernel_thread_helper+0x6/0x10

--
Philippe Vouters (Fontainebleau/France)
URL: http://vouters.dyndns.org/