On Saturday 23 July 2011 11:13:30 Thomas Gleixner wrote:
> I'm pleased to announce the 3.0-rt2 release.
>
> Changes versus 3.0-rt1:
>
> * mqueue fix (Yoshitake Kobayashi)
>
> * dca locking fix (Mike Galbraith)
>
> * trace updates (Carsten Emde)
>
> * Fixed vprintk wreckage and rcutorture (tglx)
>
> Known issues:
>
> * Some weird "console=..." commandline + config dependent
> interactions which have been not yet investigated down to their
> root cause. Result in a boot hang. YMMV
Got this during boot with rt2. The kernel survives and seems to work (eg X starts and kmail can sent this).
[ 1.142427] pci 0000:01:05.0: Boot video device
[ 1.142483] PCI: CLS 64 bytes, default 64
[ 1.145555] PCI-DMA: Disabling AGP.
[ 1.145758] PCI-DMA: aperture base @ b4000000 size 65536 KB
[ 1.145759] PCI-DMA: using GART IOMMU.
[ 1.145764] PCI-DMA: Reserving 64MB of IOMMU area in the AGP aperture
[ 1.176417]
[ 1.176419] =================================
[ 1.176420] [ INFO: inconsistent lock state ]
[ 1.176422] 3.0.0-rt2-crc+ #351
[ 1.176424] ---------------------------------
[ 1.176425] inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage.
[ 1.176428] swapper/0 [HC1[1]:SC0[18446612140904153088]:HE0:SE1] takes:
[ 1.176430] (&(&(&(*({ do { const void *__vpp_verify = (typeof((&(slab_lock))))((void *)0); (void)__vpp_verify; } while (0); ({ unsigned long __ptr; __asm__ ("" : "=r"(__ptr) : "0"((typeof(*(&(slab_lock))) *)(&(slab_lock)))); (typeof((typeof(*(&(slab_lock))) *)(&(slab_lock)))) (__ptr + (((__per_cpu_offset[__cpu])))); }); })).lock)->lock)->wait_lock){?.+...}, at: [<ffffffff815820db>] rt_spin_lock_slowlock+0x4b/0x250
[ 1.176440] {HARDIRQ-ON-W} state was registered at:
[ 1.176442] [<ffffffff8108c3ed>] __lock_acquire+0x5ed/0x16a0
[ 1.176446] [<ffffffff8108daa5>] lock_acquire+0x95/0x140
[ 1.176449] [<ffffffff8158351b>] _raw_spin_lock+0x3b/0x50
[ 1.176452] [<ffffffff815820db>] rt_spin_lock_slowlock+0x4b/0x250
[ 1.176455] [<ffffffff815829ac>] rt_spin_lock+0x2c/0x60
[ 1.176458] [<ffffffff81146cde>] kmem_cache_alloc_trace+0x7e/0x3f0
[ 1.176461] [<ffffffff811475e7>] do_tune_cpucache+0x47/0x580
[ 1.176463] [<ffffffff81147d00>] enable_cpucache+0x30/0xc0
[ 1.176466] [<ffffffff81bbc879>] kmem_cache_init_late+0x2f/0x82
[ 1.176470] [<ffffffff81ba1ac7>] start_kernel+0x267/0x325
[ 1.176473] [<ffffffff81ba12eb>] x86_64_start_reservations+0xfb/0xff
[ 1.176476] [<ffffffff81ba13dc>] x86_64_start_kernel+0xed/0xf4
[ 1.176478] irq event stamp: 125372
[ 1.176479] hardirqs last enabled at (125371): [<ffffffff8100ac4d>] default_idle+0x19d/0x1b0
[ 1.176483] hardirqs last disabled at (125372): [<ffffffff81583b67>] save_args+0x67/0x70
[ 1.176486] softirqs last enabled at (0): [< (null)>] (null)
[ 1.176487] softirqs last disabled at (0): [< (null)>] (null)
[ 1.176489]
[ 1.176490] other info that might help us debug this:
[ 1.176491] Possible unsafe locking scenario:
[ 1.176491]
[ 1.176492] CPU0
[ 1.176493] ----
[ 1.176493] lock(&(&(&(*({ do { const void *__vpp_verify = (typeof((&(slab_lock))))((void *)0); (void)__vpp_verify; } while (0); ({ unsigned long __ptr; __asm__ ("" : "=r"(__ptr) : "0"((typeof(*(&(slab_lock))) *)(&(slab_lock)))); (typeof((typeof(*(&(slab_lock))) *)(&(slab_lock)))) (__ptr + (((__per_cpu_offset[__cpu])))); }); })).lock)->lock)->wait_lock);
[ 1.176498] <Interrupt>
[ 1.176498] lock(&(&(&(*({ do { const void *__vpp_verify = (typeof((&(slab_lock))))((void *)0); (void)__vpp_verify; } while (0); ({ unsigned long __ptr; __asm__ ("" : "=r"(__ptr) : "0"((typeof(*(&(slab_lock))) *)(&(slab_lock)))); (typeof((typeof(*(&(slab_lock))) *)(&(slab_lock)))) (__ptr + (((__per_cpu_offset[__cpu])))); }); })).lock)->lock)->wait_lock);
[ 1.176502]
[ 1.176503] *** DEADLOCK ***
[ 1.176503]
[ 1.176504] no locks held by swapper/0.
[ 1.176505]
[ 1.176506] stack backtrace:
[ 1.176508] Pid: 0, comm: swapper Not tainted 3.0.0-rt2-crc+ #351
[ 1.176509] Call Trace:
[ 1.176510] <IRQ> [<ffffffff8108ae1f>] print_usage_bug+0x21f/0x270
[ 1.176514] [<ffffffff8108b742>] mark_lock+0x332/0x400
[ 1.176517] [<ffffffff8108c55e>] __lock_acquire+0x75e/0x16a0
[ 1.176519] [<ffffffff8108b608>] ? mark_lock+0x1f8/0x400
[ 1.176521] [<ffffffff815820db>] ? rt_spin_lock_slowlock+0x4b/0x250
[ 1.176523] [<ffffffff8108daa5>] lock_acquire+0x95/0x140
[ 1.176525] [<ffffffff815820db>] ? rt_spin_lock_slowlock+0x4b/0x250
[ 1.176528] [<ffffffff81010294>] ? kzalloc+0x14/0x20
[ 1.176530] [<ffffffff8158351b>] _raw_spin_lock+0x3b/0x50
[ 1.176533] [<ffffffff815820db>] ? rt_spin_lock_slowlock+0x4b/0x250
[ 1.176535] [<ffffffff815820db>] rt_spin_lock_slowlock+0x4b/0x250
[ 1.176537] [<ffffffff81010294>] ? kzalloc+0x14/0x20
[ 1.176539] [<ffffffff815829ac>] rt_spin_lock+0x2c/0x60
[ 1.176541] [<ffffffff81147149>] __kmalloc+0xf9/0x4b0
[ 1.176544] [<ffffffff81010294>] kzalloc+0x14/0x20
[ 1.176546] [<ffffffff81576999>] cpuid4_cache_lookup_regs+0x1bc/0x33a
[ 1.176549] [<ffffffff81079257>] ? __hrtimer_start_range_ns+0x157/0x4a0
[ 1.176551] [<ffffffff81077b48>] ? hrtimer_get_next_event+0xe8/0x110
[ 1.176553] [<ffffffff81581c38>] ? rt_spin_lock_slowunlock+0x78/0x80
[ 1.176555] [<ffffffff815775da>] get_cpu_leaves+0x88/0x318
[ 1.176558] [<ffffffff81095586>] generic_smp_call_function_single_interrupt+0x136/0x190
[ 1.176562] [<ffffffff8101e087>] smp_call_function_single_interrupt+0x27/0x40
[ 1.176565] [<ffffffff8158bb33>] call_function_single_interrupt+0x13/0x20
[ 1.176566] <EOI> [<ffffffff8100ac4f>] ? default_idle+0x19f/0x1b0
[ 1.176570] [<ffffffff8100ac4d>] ? default_idle+0x19d/0x1b0
[ 1.176572] [<ffffffff8100acd0>] amd_e400_idle+0x70/0x120
[ 1.176575] [<ffffffff8100123f>] cpu_idle+0xdf/0x120
[ 1.176577] [<ffffffff8156a7dc>] rest_init+0xbc/0xd0
[ 1.176579] [<ffffffff8156a720>] ? csum_partial_copy_generic+0x170/0x170
[ 1.176582] [<ffffffff81ba1b7a>] start_kernel+0x31a/0x325
[ 1.176584] [<ffffffff81ba12eb>] x86_64_start_reservations+0xfb/0xff
[ 1.176586] [<ffffffff81ba13dc>] x86_64_start_kernel+0xed/0xf4
[ 1.182619] audit: initializing netlink socket (disabled)
[ 1.182648] type=2000 audit(1311440614.181:1): initialized
[ 1.211068] HugeTLB registered 2 MB page size, pre-allocated 0 pages
[ 1.246054] VFS: Disk quotas dquot_6.5.2
[ 1.246104] Dquot-cache hash table entries: 512 (order 0, 4096 bytes)
[ 1.246873] msgmni has been set to 15445
Hope this helps.
Ed
--
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
