Created attachment 674339 [details] Full kernel warning Description of problem: abrtd reported a warning after booting to 3.8-rc2 kernel (abrt failed to report it though): atl1c 0000:03:00.0: DMA-API: device driver failed to check map error[device address=0x00000000cb8ca000] [size=90 bytes] [mapped as single] Version-Release number of selected component (if applicable): kernel-3.8.0-0.rc2.git1.1.fc19.x86_64 Call Trace: <IRQ> [<ffffffff8106aaef>] warn_slowpath_common+0x7f/0xc0 [<ffffffff8106abe6>] warn_slowpath_fmt+0x46/0x50 [<ffffffff81383367>] check_unmap+0x407/0x8a0 [<ffffffff81022e49>] ? sched_clock+0x9/0x10 [<ffffffff810af8c5>] ? sched_clock_cpu+0xc5/0x120 [<ffffffff81383a1e>] debug_dma_unmap_page+0x5e/0x70 [<ffffffffa02abd7c>] atl1c_intr+0x3cc/0x600 [atl1c] [<ffffffff810d75ed>] ? trace_hardirqs_off+0xd/0x10 [<ffffffff8111b59c>] handle_irq_event_percpu+0x6c/0x3c0 [<ffffffff8111b938>] handle_irq_event+0x48/0x70 [<ffffffff8111e2c7>] handle_edge_irq+0x77/0x110 [<ffffffff8101d2ff>] handle_irq+0xbf/0x150 Mapped at: [<ffffffff81383ad9>] debug_dma_map_page+0xa9/0x150 [<ffffffffa02acf95>] atl1c_xmit_frame+0x7b5/0xa90 [atl1c] [<ffffffff815e1010>] dev_hard_start_xmit+0x250/0x6e0 [<ffffffff81602f8e>] sch_direct_xmit+0xfe/0x2a0 [<ffffffff815e1c00>] dev_queue_xmit+0x1f0/0x920
I am seeing "device driver failed to check map error" with e100 and 3.8.0-0.rc kernels so this looks like a more widespread disease. Should that be reported as a separate bug? For now here: [ 55.395057] WARNING: at lib/dma-debug.c:933 check_unmap+0x407/0x8a0() [ 55.395057] Hardware name: To Be Filled By O.E.M. [ 55.395057] e100 0000:00:0e.0: DMA-API: device driver failed to check map error[device address=0x000000007a121292] [size=90 bytes] [mapped as single] [ 55.395057] Modules linked in: [ 55.395057] nf_conntrack_netbios_ns nf_conntrack_broadcast ipt_MASQUERADE ip6table_mangle ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 iptable_nat nf_nat_ipv4 nf_nat iptable_mangle nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ebtable_filter ebtables ip6table_filter ip6_tables be2iscsi iscsi_boot_sysfs bnx2i cnic uio cxgb4i cxgb4 cxgb3i cxgb3 mdio libcxgbi ib_iser rdma_cm ib_addr iw_cm ib_cm ib_sa ib_mad ib_core iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi w83627hf hwmon_vid snd_via82xx ppdev snd_ac97_codec ac97_bus snd_seq snd_pcm parport_pc snd_mpu401 snd_mpu401_uart snd_rawmidi ns558 gameport snd_seq_device parport e100 snd_page_alloc snd_timer snd soundcore shpchp edac_core skge k8temp i2c_viapro edac_mce_amd mii nfsd auth_rpcgss nfs_acl lockd uinput sunrpc binfmt_misc radeon [ 55.395057] ata_generic i2c_algo_bit drm_kms_helper pata_acpi ttm firewire_ohci drm firewire_core pata_via sata_via i2c_core sata_promise crc_itu_t [ 55.395057] Pid: 1086, comm: ifup-eth Not tainted 3.8.0-0.rc3.git1.1.fc19.x86_64 #1 [ 55.395057] Call Trace: [ 55.395057] <IRQ> [<ffffffff8106aaff>] warn_slowpath_common+0x7f/0xc0 [ 55.395057] [<ffffffff8106abf6>] warn_slowpath_fmt+0x46/0x50 [ 55.395057] [<ffffffff81383167>] check_unmap+0x407/0x8a0 [ 55.395057] [<ffffffff810ddb9d>] ? trace_hardirqs_on+0xd/0x10 [ 55.395057] [<ffffffff8138381e>] debug_dma_unmap_page+0x5e/0x70 [ 55.395057] [<ffffffffa030e9e8>] e100_tx_clean+0xe8/0x210 [e100] [ 55.395057] [<ffffffffa030fc81>] e100_poll+0x521/0x680 [e100] [ 55.395057] [<ffffffff810a5473>] ? __wake_up+0x53/0x70 [ 55.395057] [<ffffffff815dacf2>] net_rx_action+0x172/0x380 [ 55.395057] [<ffffffff81074800>] __do_softirq+0xf0/0x3d0 [ 55.395057] [<ffffffff817267bc>] call_softirq+0x1c/0x30 [ 55.395057] [<ffffffff8101d435>] do_softirq+0xa5/0xe0 [ 55.395057] [<ffffffff81074cd5>] irq_exit+0xd5/0xe0 [ 55.395057] [<ffffffff8172714e>] smp_apic_timer_interrupt+0x6e/0x99 [ 55.395057] [<ffffffff81726072>] apic_timer_interrupt+0x72/0x80 [ 55.395057] <EOI> [<ffffffff81712d3f>] ? __slab_alloc+0x525/0x58f [ 55.395057] [<ffffffff811bc871>] ? set_track+0x61/0x1b0 [ 55.395057] [<ffffffff81712d36>] ? __slab_alloc+0x51c/0x58f [ 55.395057] [<ffffffff810ddafd>] ? trace_hardirqs_on_caller+0x10d/0x1a0 [ 55.395057] [<ffffffff811a1a79>] ? anon_vma_clone+0x49/0x140 [ 55.395057] [<ffffffff810af8d8>] ? sched_clock_cpu+0xa8/0x120 [ 55.395057] [<ffffffff810d761d>] ? trace_hardirqs_off+0xd/0x10 [ 55.395057] [<ffffffff811a1a79>] ? anon_vma_clone+0x49/0x140 [ 55.395057] [<ffffffff811bfcc1>] kmem_cache_alloc+0x2d1/0x350 [ 55.395057] [<ffffffff811a1a79>] anon_vma_clone+0x49/0x140 [ 55.395057] [<ffffffff811a2938>] anon_vma_fork+0x38/0x110 [ 55.395057] [<ffffffff81068025>] dup_mm+0x275/0x660 [ 55.395057] [<ffffffff8106991f>] copy_process.part.21+0x14cf/0x1650 [ 55.395057] [<ffffffff81720374>] ? __do_page_fault+0x314/0x580 [ 55.395057] [<ffffffff81069b96>] do_fork+0xa6/0x440 [ 55.395057] [<ffffffff811915fc>] ? might_fault+0x5c/0xb0 [ 55.395057] [<ffffffff8171c398>] ? retint_swapgs+0x13/0x1b [ 55.395057] [<ffffffff81069fb6>] sys_clone+0x16/0x20 [ 55.395057] [<ffffffff817257f9>] stub_clone+0x69/0x90 [ 55.395057] [<ffffffff81725419>] ? system_call_fastpath+0x16/0x1b [ 55.395057] ---[ end trace 729c3927e69addd3 ]--- [ 55.395057] Mapped at: [ 55.395057] [<ffffffff813838d9>] debug_dma_map_page+0xa9/0x150 [ 55.395057] [<ffffffffa030e886>] e100_xmit_prepare+0x136/0x1b0 [e100] [ 55.395057] [<ffffffffa030c8e4>] e100_exec_cb+0x84/0x140 [e100] [ 55.395057] [<ffffffffa030ed45>] e100_xmit_frame+0x45/0x1b0 [e100] [ 55.395057] [<ffffffff815db9a0>] dev_hard_start_xmit+0x250/0x6e0 Full dmesg for 3.8.0-0.rc3.git1.1.fc19.x86_64 attached.
Created attachment 681983 [details] dmesg with "failed to check map error"
This bug appears to have been reported against 'rawhide' during the Fedora 19 development cycle. Changing version to '19'. (As we did not run this process for some time, it could affect also pre-Fedora 19 development cycle bugs. We are very sorry. It will help us with cleanup during Fedora 19 End Of Life. Thank you.) More information and reason for this action is here: https://fedoraproject.org/wiki/BugZappers/HouseKeeping/Fedora19
Is this still an issue with the 3.9 kernels in F19?
(In reply to comment #4) > Is this still an issue with the 3.9 kernels in F19? It sure is as far as e100 driver is concerned. See bug 907694. Looks pretty similar. You will need an info from OP about atl1c as I do not have this hardware. Neil Horman recently submitted upstream a patch for e100 driver. C.f. https://bugzilla.redhat.com/show_bug.cgi?id=907694#c10 . It should show up in Fedora kernels in a due course but this will be a while if that has to go through upstream. My guess would be that atl1c could use something like that too but this is only a guess.
Created attachment 736258 [details] dmesg output Present in Fedora 19 RC3 Live LXDE DVD Traceback portion is: [ 120.708318] ------------[ cut here ]------------ [ 120.711396] WARNING: at lib/dma-debug.c:937 check_unmap+0x493/0x950() [ 120.714518] Hardware name: AOD270 [ 120.717626] bcma-pci-bridge 0000:02:00.0: DMA-API: device driver failed to check map error[device address=0x00000000124300ec] [size=1876 bytes] [mapped as single] [ 120.721115] Modules linked in: [ 120.724482] ip6table_nat nf_nat_ipv6 ip6table_mangle ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 iptable_nat nf_nat_ipv4 nf_nat iptable_mangle nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ebtable_filter ebtables ip6table_filter ip6_tables arc4 brcmsmac cordic brcmutil mac80211 cfg80211 rtsx_pci_sdmmc rtsx_pci_ms mmc_core acer_wmi iTCO_wdt memstick sparse_keymap iTCO_vendor_support rfkill snd_hda_codec_realtek snd_hda_codec_hdmi uvcvideo snd_hda_intel acpi_cpufreq snd_hda_codec mperf videobuf2_vmalloc videobuf2_memops snd_hwdep videobuf2_core coretemp snd_seq snd_seq_device videodev snd_pcm media microcode joydev snd_page_alloc bcma serio_raw snd_timer rtsx_pci snd soundcore i2c_i801 lpc_ich of_i2c uinput isofs squashfs usb_storage gma500_gfx i2c_algo_bit drm_kms_helper drm r8169 mii [ 120.749714] i2c_core wmi video sunrpc [ 120.754234] Pid: 0, comm: swapper/2 Not tainted 3.9.0-0.rc6.git2.3.fc19.i686 #1 [ 120.758931] Call Trace: [ 120.763582] [<c0441e63>] warn_slowpath_common+0x63/0x80 [ 120.768297] [<c06ed113>] ? check_unmap+0x493/0x950 [ 120.773012] [<c06ed113>] ? check_unmap+0x493/0x950 [ 120.777685] [<c0441eb3>] warn_slowpath_fmt+0x33/0x40 [ 120.782365] [<c06ed113>] check_unmap+0x493/0x950 [ 120.787077] [<c04a65cb>] ? trace_hardirqs_off+0xb/0x10 [ 120.791793] [<c06ed637>] debug_dma_unmap_page+0x67/0x70 [ 120.796559] [<f96a264a>] dma64_getnextrxp+0xda/0x130 [brcmsmac] [ 120.801376] [<f96a31f9>] dma_rx+0x59/0x1d0 [brcmsmac] [ 120.806226] [<c04a0000>] ? clocksource_suspend+0x20/0x40 [ 120.811057] [<c06de95c>] ? __raw_spin_lock_init+0x2c/0x50 [ 120.815896] [<f967271e>] brcms_c_dpc+0x17e/0x8a0 [brcmsmac] [ 120.820709] [<c04a65cb>] ? trace_hardirqs_off+0xb/0x10 [ 120.825446] [<c047e4c5>] ? local_clock+0x55/0x60 [ 120.830118] [<f96687a8>] brcms_dpc+0x38/0xd0 [brcmsmac] [ 120.834762] [<c04a94ae>] ? trace_hardirqs_on_caller+0x9e/0x170 [ 120.839404] [<c044b3b8>] tasklet_action+0x78/0x1b0 [ 120.843999] [<c044baa9>] __do_softirq+0xc9/0x350 [ 120.848648] [<c0409348>] ? sched_clock+0x8/0x10 [ 120.853437] [<c047e39d>] ? sched_clock_cpu+0xed/0x150 [ 120.857957] [<c044bec5>] irq_exit+0xa5/0xb0 [ 120.862424] [<c0404c65>] do_IRQ+0x45/0xb0 [ 120.866793] [<c0409348>] ? sched_clock+0x8/0x10 [ 120.871124] [<c0a38df8>] common_interrupt+0x38/0x40 [ 120.875419] [<c04a007b>] ? clocksource_touch_watchdog+0x1b/0x20 [ 120.879716] [<c08c7c6c>] ? cpuidle_wrap_enter+0x3c/0x90 [ 120.883874] [<c08c7cd2>] cpuidle_enter_tk+0x12/0x20 [ 120.887872] [<c08c7290>] ? intel_pstate_timer_func+0x420/0x420 [ 120.891744] [<c08c7841>] cpuidle_enter_state+0x11/0x50 [ 120.895455] [<c08c790a>] cpuidle_idle_call+0x8a/0x2e0 [ 120.898989] [<c040adb5>] cpu_idle+0xa5/0xe0 [ 120.902345] [<c0a1ea23>] start_secondary+0x26d/0x272 [ 120.905563] ---[ end trace 518717d935617dd8 ]--- [ 120.908624] Mapped at: [ 120.911485] [<c06ebc93>] debug_dma_map_page+0x63/0x130 [ 120.914356] [<f96a3554>] dma_rxfill+0x1e4/0x300 [brcmsmac] [ 120.917213] [<f96735dd>] brcms_c_init+0x79d/0xf50 [brcmsmac] [ 120.920056] [<f966869f>] brcms_init+0x2f/0x40 [brcmsmac] [ 120.922886] [<f96720cb>] brcms_c_up+0x13b/0x3d0 [brcmsmac]
Looks like this got fixed in commit ac574804d412047e52de4dd887615cc88f58aeb0. You'll want separate bugs for the other stack traces as the fixes are all driver specific.