Created attachment 21877 [details] My .config Nothing dangerous, I hope, but i saw this warning twice. Adding 399992k swap on /swap.file. Priority:-1 extents:728 across:2152036k ------------[ cut here ]------------ WARNING: at lib/dma-debug.c:576 check_for_illegal_area+0xd3/0xfd() Hardware name: MS-6380E VIA_IDE 0000:00:11.1: DMA-API: device driver maps memory from kernel text or rodata [addr=c0ffc000] [size=16384] Modules linked in: snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss sg sd_mod ipv6 nfsd lockd nfs_acl auth_rpcgss sunrpc usb_storage scsi_mod usb_libusual tuner_simple tuner_types tda9887 tda8290 nouveau ppdev snd_cs4236 snd_mpu401 ttm snd_wavefront uhci_hcd fb saa7134 snd_emu10k1 snd_via82xx snd_ac97_codec snd_wss_lib ac97_bus ehci_hcd ir_common snd_opl3_lib snd_mpu401_uart snd_pcm videobuf_dma_sg snd_timer snd_page_alloc snd_rawmidi videobuf_core tveeprom snd_util_mem snd_seq_device snd_hwdep rtc_cmos drm usbcore i2c_algo_bit cfbcopyarea emu10k1_gp cfbimgblt cfbfillrect rtc_core parport_pc rtc_lib ns558 snd 8139too floppy parport soundcore gameport shpchp via_agp pci_hotplug xfs exportfs ufs agpgart tuner v4l2_common videodev v4l1_compat w83627hf hwmon_vid hwmon i2c_viapro i2c_dev Pid: 87, comm: kblockd/0 Not tainted 2.6.30-i486 #4 Call Trace: [<c1031c86>] warn_slowpath_common+0x65/0x95 [<c1031cf4>] warn_slowpath_fmt+0x29/0x2c [<c11611df>] check_for_illegal_area+0xd3/0xfd [<c11622a4>] debug_dma_map_sg+0x13b/0x147 [<c11cfae1>] ide_dma_prepare+0xc2/0x10b [<c11cbaba>] do_rw_taskfile+0x221/0x261 [<c11d9990>] ide_do_rw_disk+0x23d/0x2b4 [<c114ee9b>] ? delay_tsc+0x47/0x6c [<c11d9077>] ide_gd_do_request+0xf/0x12 [<c11c8cc8>] do_ide_request+0x345/0x4a1 [<c105374a>] ? trace_hardirqs_off_caller+0x18/0xa3 [<c10537e0>] ? trace_hardirqs_off+0xb/0xd [<c128f8e1>] ? _spin_unlock_irqrestore+0x34/0x41 [<c103ae8f>] ? del_timer+0x4a/0x52 [<c113d670>] ? blk_remove_plug+0x109/0x114 [<c113d736>] __generic_unplug_device+0x46/0x4a [<c113d77b>] blk_start_queueing+0x41/0x45 [<c1146460>] cfq_kick_queue+0x21/0x2f [<c10412b2>] worker_thread+0x1d4/0x2ce [<c1041265>] ? worker_thread+0x187/0x2ce [<c114643f>] ? cfq_kick_queue+0x0/0x2f [<c1044d0c>] ? autoremove_wake_function+0x0/0x38 [<c10410de>] ? worker_thread+0x0/0x2ce [<c1044a09>] kthread+0x4a/0x70 [<c10449bf>] ? kthread+0x0/0x70 [<c1003a87>] kernel_thread_helper+0x7/0x10 ---[ end trace fa41507272d4fe04 ]--- and second time today reconnect_path: npd != pd mtrr: no MTRR for c0000000,10000000 found ------------[ cut here ]------------ WARNING: at lib/dma-debug.c:576 check_for_illegal_area+0xd3/0xfd() Hardware name: MS-6380E VIA_IDE 0000:00:11.1: DMA-API: device driver maps memory from kernel text or rodata [addr=c0fff000] [size=4096] Modules linked in: sg sd_mod snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss ipv6 nfsd lockd nfs_acl auth_rpcgss sunrpc usb_storage scsi_mod usb_libusual tuner_simple tuner_types tda9887 snd_cs4236 tda8290 snd_mpu401 ppdev snd_via82xx snd_emu10k1 snd_wavefront snd_wss_lib snd_ac97_codec ac97_bus saa7134 uhci_hcd snd_pcm ehci_hcd ir_common snd_opl3_lib videobuf_dma_sg snd_mpu401_uart snd_timer snd_page_alloc videobuf_core snd_rawmidi snd_util_mem tveeprom snd_seq_device usbcore shpchp snd_hwdep 8139too emu10k1_gp rtc_cmos pci_hotplug via_agp parport_pc rtc_core ns558 rtc_lib snd parport floppy gameport soundcore xfs exportfs ufs agpgart tuner v4l2_common videodev v4l1_compat w83627hf hwmon_vid hwmon i2c_viapro i2c_dev [last unloaded: cfbfillrect] Pid: 1542, comm: kjournald Not tainted 2.6.30-i486 #4 Call Trace: [<c1031c86>] warn_slowpath_common+0x65/0x95 [<c1031cf4>] warn_slowpath_fmt+0x29/0x2c [<c11611df>] check_for_illegal_area+0xd3/0xfd [<c11622a4>] debug_dma_map_sg+0x13b/0x147 [<c11cfae1>] ide_dma_prepare+0xc2/0x10b [<c11cbaba>] do_rw_taskfile+0x221/0x261 [<c11d9990>] ide_do_rw_disk+0x23d/0x2b4 [<c114ee9b>] ? delay_tsc+0x47/0x6c [<c11d9077>] ide_gd_do_request+0xf/0x12 [<c11c8cc8>] do_ide_request+0x345/0x4a1 [<c105374a>] ? trace_hardirqs_off_caller+0x18/0xa3 [<c10537e0>] ? trace_hardirqs_off+0xb/0xd [<c128f8e1>] ? _spin_unlock_irqrestore+0x34/0x41 [<c103ae8f>] ? del_timer+0x4a/0x52 [<c113d670>] ? blk_remove_plug+0x109/0x114 [<c113d736>] __generic_unplug_device+0x46/0x4a [<c113de7c>] __make_request+0x3d5/0x3ea [<c113ba4d>] ? trace_block_remap+0x2f/0x5f [<c113c5ab>] generic_make_request+0x262/0x2a5 [<c1093ac0>] ? mempool_alloc+0x5f/0x111 [<c113c6b7>] submit_bio+0xc9/0xd1 [<c10df797>] ? bio_alloc_bioset+0x8d/0x10e [<c10dafcc>] submit_bh+0x10d/0x12b [<c10dc87d>] sync_dirty_buffer+0x72/0xaf [<c11195bc>] journal_commit_transaction+0xb0b/0xf4f [<c10537e0>] ? trace_hardirqs_off+0xb/0xd [<c103ae3d>] ? try_to_del_timer_sync+0x44/0x4c [<c111c806>] kjournald+0xcc/0x1ed [<c1044d0c>] ? autoremove_wake_function+0x0/0x38 [<c111c73a>] ? kjournald+0x0/0x1ed [<c1044a09>] kthread+0x4a/0x70 [<c10449bf>] ? kthread+0x0/0x70 [<c1003a87>] kernel_thread_helper+0x7/0x10 ---[ end trace a9c5ff347ea44ae3 ]---
I currently think this is a driver bug. But to be sure can you please upload the vmlinux file of the failing kernel somewhere so I can have a look at the addresses of the segment boundary symbols?
http://url.file.am/?DJvfr md5sum: dc6da35c982542b8c083edae790ceb8c kernel-2.6.30.bz2
Ok, this was a dma-debug bug. The two warnings above are for memory areas which exactly ends at the beginning of the kernel text segment. I queued a fix in my tree and send it out when -rc1 is released. I also marked the patch for stable backporting.
On Tuesday 16 June 2009 12:28:01 bugzilla-daemon@bugzilla.kernel.org wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=13520 > > > > > > --- Comment #3 from Joerg Roedel <joro@8bytes.org> 2009-06-16 10:27:59 --- > Ok, this was a dma-debug bug. The two warnings above are for memory areas > which > exactly ends at the beginning of the kernel text segment. I queued a fix in > my > tree and send it out when -rc1 is released. I also marked the patch for > stable > backporting. Could you please post a link to the fix and then close the bug? Thanks.