Messages in this thread | | | From | Kay Sievers <> | Date | Thu, 12 Jul 2012 15:52:17 +0200 | Subject | Re: 3.5-rc6 printk formatting problem during oom-kill. |
| |
On Thu, Jul 12, 2012 at 2:54 AM, Dave Jones <davej@redhat.com> wrote: > On Mon, Jul 09, 2012 at 08:48:51PM +0200, Kay Sievers wrote:
> > It looks fine here with the above mentioned patch: > > Now that that patch is in Linus tree, I've hit what's probably a different case. > Look at the modules list in this oops.. > > [10016.460020] BUG: soft lockup - CPU#1 stuck for 22s! [trinity-child1:24295] > [10016.470008] rose<4>[10016.470008] ip_set_bitmap_ipmac<4>[10016.470008]
> Also, I have no idea how the hell the 'Modules linked in:' line (9th line) ended up being printed /after/ the > module listing began (2nd line).
It's the output of 'dmesg' you pasted, right?
I tried to force all sorts of racy print_modules() calls, and kept your trinity tool from git for hours, it looks all fine here:
[ 54.664301] NET: Registered protocol family 24 [ 54.684721] Netfilter messages via NETLINK v0.30. [ 63.103847] ------------[ cut here ]------------ [ 63.103857] WARNING: at kernel/futex.c:2452 sys_get_robust_list+0x28e/0x2a0() [ 63.103859] Hardware name: 4174NEG [ 63.103861] deprecated: get_robust_list will be deleted in 2013. [ 63.103862] Modules linked in: binfmt_misc nfnetlink pppoe pppox ppp_generic slhc atm fuse arc4 i915 hid_generic i2c_algo_bit drm_kms_helper drm snd_usb_audio snd_usbmidi_lib snd_rawmidi snd_seq_device bnep bluetooth snd_hda_codec_hdmi snd_hda_codec_conexant uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core videodev media coretemp kvm_intel kvm crc32c_intel microcode i2c_i801 i2c_core iwlwifi snd_hda_intel snd_hda_codec mac80211 snd_hwdep lpc_ich snd_pcm mfd_core sdhci_pci sdhci mmc_core snd_page_alloc cfg80211 snd_timer thinkpad_acpi snd e1000e soundcore rfkill video cdc_ncm wmi usbnet mii cdc_wdm cdc_acm [ 63.103921] Pid: 837, comm: trinity-child1 Not tainted 3.5.0-0.rc6.git2.1.fc18.x86_64 #1 [ 63.103923] Call Trace: [ 63.103930] [<ffffffff8106310f>] warn_slowpath_common+0x7f/0xc0 [ 63.103934] [<ffffffff81063206>] warn_slowpath_fmt+0x46/0x50 [ 63.103938] [<ffffffff810d253d>] ? trace_hardirqs_on_caller+0x10d/0x1a0 [ 63.103941] [<ffffffff810d6ffe>] sys_get_robust_list+0x28e/0x2a0 [ 63.103946] [<ffffffff816d7d29>] system_call_fastpath+0x16/0x1b [ 63.103948] ---[ end trace 796993e76a8397be ]--- [ 63.126701] NET: Registered protocol family 23 [ 63.397208] trinity-child3 (839): Using mlock ulimits for SHM_HUGETLB is deprecated
Can you easily reproduce the issue you pasted? If, could you give me the /dev/kmsg output?
Thanks, Kay
| |