WARNING: at /build/buildd/linux-2.6.31/kernel/trace/ring_buffer.c:1393 rb_add_time_stamp+0x79/0x200()

------------[ cut here ]------------
WARNING: at /build/buildd/linux-2.6.31/kernel/trace/ring_buffer.c:1393 rb_add_time_stamp+0x79/0x200()
Hardware name: Inspiron 1011
Modules linked in: ufs qnx4 hfsplus hfs minix ntfs vfat msdos fat jfs xfs exportfs reiserfs tun bridge stp bnep btusb joydev snd_hda_codec_realtek usb_storage snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi lib80211_crypt_tkip snd_seq_midi_event snd_seq uvcvideo dell_wmi videodev v4l1_compat snd_timer dell_laptop snd_seq_device psmouse compal_laptop wl(P) dcdbas serio_raw lib80211 snd soundcore snd_page_alloc lp parport r8169 mii fbcon tileblit font bitblit softcursor i915 drm i2c_algo_bit video output intel_agp agpgart
Pid: 1454, comm: hald Tainted: P 2.6.31-10-generic #32-Ubuntu
Call Trace:
 [<c014082d>] warn_slowpath_common+0x6d/0xa0
 [<c0195809>] ? rb_add_time_stamp+0x79/0x200
 [<c0195809>] ? rb_add_time_stamp+0x79/0x200
 [<c0140875>] warn_slowpath_null+0x15/0x20
 [<c0195809>] rb_add_time_stamp+0x79/0x200
 [<c0108a18>] ? sched_clock+0x8/0x10
 [<c0195aa2>] rb_reserve_next_event+0x112/0x1b0
 [<c01eea6a>] ? do_filp_open+0x53a/0x890
 [<c0195cc3>] ring_buffer_lock_reserve+0x83/0x150
 [<c01eb7a6>] ? putname+0x26/0x40
 [<c0197f2c>] trace_buffer_lock_reserve+0x1c/0x50
 [<c01980ef>] trace_current_buffer_lock_reserve+0x1f/0x30
 [<c01e06ea>] ftrace_raw_event_do_sys_open+0x4a/0xb0
 [<c01e0b52>] do_sys_open+0x132/0x150

Note I am running with ~ubuntu-boot PPA packages as well

Ok, this warn actually is tripped due to this message below which occurs just before the stack dump. It seems to be a warning that we are taking a slow path in frace buffer insertion:

  [ 1086.779915] Delta way too big! 18446743427706068172 ts=18446744065310248178 write stamp = 637604180006

However this is actually benign, corrective action is taken and processing continues as normal. I don't think this can have been the cause of the network manager behaviour seen.

I am interested to know if you were using ftrace here.

I was not (knowingly) doing anything with ftrace. I simply suspended and later resumed, with a pretty standard handful of applications loaded.

If this warning isn't related to the resume failure, then let's track the resume failure over at bug 429009, and this one can be only about the warning.

Note that sreadahead uses ftrace

Looking at the nature of the ftrace in action it looks to be one we might expect to see enabled for sreadahead when it is in profiling mode. Perhaps this broke/was killed in the middle before the suspend/resume.

Ok this seems to be a benign message that one would expect to see if suspend was used in the middle of an ftrace, and that is used by sreadahead which may well have broken. As this was with pre-release and now updated boot packages I suspect we are not going to be able to figure out what happened to sreadahead. From the kernel side there should be nothing to worry about unless it is recurring. I will close this one out 'Won't Fix' so it remains searchable.