Re: 2.6.30-rc2: WARNING at i915_gem.c for i915_gem_idle

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 21 Apr 2009 23:35:26 +0200
Niel Lambrechts <niel.lambrechts@gmail.com> wrote:

> Hi there,
> 
> I get the following when I switch from runlevel 5 to 3 using the latest
> git kernel. I haven't noticed any other side-effects and X started up
> fine, but I'm assuming something is misbehaving and the trace might be
> meaningful to someone. I also frequently still notice this "*ERROR*
> trying to get vblank count for disabled pipe 0" message when I exit
> Xorg. (BTW, I am still using EXA acceleration due to issues with UXA+KMS).
> 
> The hardware is an Lenovo Thinkpad W500 with:
>      00:02.1 Display controller: Intel Corporation Mobile 4 Series
> Chipset Integrated Graphics Controller (rev 07)
> 
> messages:
> Apr 21 00:44:00 linux-7vph init: Switching to runlevel: 3
> Apr 21 00:44:06 linux-7vph kernel: ------------[ cut here ]------------
> Apr 21 00:44:06 linux-7vph kernel: WARNING: at
> drivers/gpu/drm/i915/i915_gem.c:3823 i915_gem_idle+0x17a/0x28c [i915]()
> Apr 21 00:44:06 linux-7vph kernel: Hardware name: 40622XG
> Apr 21 00:44:06 linux-7vph kernel: Modules linked in: iwlagn rndis_wlan
> rndis_host cdc_ether usbnet cdc_acm mii usbhid hid cpufreq_stats tun
> i915 drm i2c_algo_bit af_packet ip6t_LOG xt_tcpudp xt_pkttype ipt_LOG
> xt_limit snd_pcm_oss snd_mixer_oss snd_seq binfmt_misc ip6t_REJECT
> nf_conntrack_ipv6 ip6table_raw xt_NOTRACK ipt_REJECT xt_state
> iptable_raw iptable_filter ip6table_mangle cpufreq_conservative
> cpufreq_userspace nf_conntrack_netbios_ns cpufreq_powersave
> nf_conntrack_ipv4 acpi_cpufreq nf_conntrack nf_defrag_ipv4 ip_tables
> speedstep_lib ip6table_filter ip6_tables x_tables ipv6 microcode fuse
> loop dm_mod snd_usb_audio snd_usb_lib snd_rawmidi snd_seq_device
> snd_hda_codec_conexant arc4 snd_hda_intel ecb snd_hda_codec snd_hwdep
> snd_pcm iwlcore snd_timer thinkpad_acpi snd mac80211 soundcore rtc_cmos
> video rfkill led_class intel_agp iTCO_wdt output rtc_core snd_page_alloc
> ohci1394 i2c_i801 button sr_mod wmi battery ac e1000e rtc_lib agpgart
> cfg80211 nvram joydev i2c_core cdrom iTCO_vendor_support sg ieee1394
> sd_mod us
> Apr 21 00:44:06 linux-7vph kernel: bcore ext4 jbd2 crc16 edd ext3
> mbcache jbd fan ahci libata scsi_mod thermal processor [last unloaded:
> iwlagn]
> Apr 21 00:44:06 linux-7vph kernel: Pid: 3193, comm: Xorg Not tainted
> 2.6.30-rc2-pae #24
> Apr 21 00:44:06 linux-7vph kernel: Call Trace:
> Apr 21 00:44:06 linux-7vph kernel:  [<c0127768>] warn_slowpath+0x71/0xa0
> Apr 21 00:44:06 linux-7vph kernel:  [<c012f66f>] ? del_timer_sync+0xd/0x18
> Apr 21 00:44:06 linux-7vph kernel:  [<c030573c>] ?
> schedule_timeout+0x154/0x16a
> Apr 21 00:44:06 linux-7vph kernel:  [<c01354f3>] ?
> __cancel_work_timer+0x124/0x162
> Apr 21 00:44:06 linux-7vph kernel:  [<f82df3ba>] ?
> i915_gem_retire_requests+0x141/0x15b [i915]
> Apr 21 00:44:06 linux-7vph kernel:  [<f82e07e2>]
> i915_gem_idle+0x17a/0x28c [i915]
> Apr 21 00:44:06 linux-7vph kernel:  [<f82e0947>]
> i915_gem_leavevt_ioctl+0x1f/0x2e [i915]
> Apr 21 00:44:06 linux-7vph kernel:  [<f828574c>] drm_ioctl+0x1f6/0x27c [drm]
> Apr 21 00:44:06 linux-7vph kernel:  [<f82e0928>] ?
> i915_gem_leavevt_ioctl+0x0/0x2e [i915]
> Apr 21 00:44:06 linux-7vph kernel:  [<c0305dc0>] ? mutex_lock+0xe/0x28
> Apr 21 00:44:06 linux-7vph kernel:  [<f7ee5640>] ?
> agpioc_release_wrap+0x57/0x5b [agpgart]
> Apr 21 00:44:06 linux-7vph kernel:  [<f7ee6279>] ? agp_ioctl+0x3b6/0x3d9
> [agpgart]
> Apr 21 00:44:06 linux-7vph kernel:  [<c013ef0f>] ? getnstimeofday+0x51/0xda
> Apr 21 00:44:06 linux-7vph kernel:  [<c0196b3a>] vfs_ioctl+0x50/0x69
> Apr 21 00:44:06 linux-7vph kernel:  [<c0196fac>] do_vfs_ioctl+0x459/0x492
> Apr 21 00:44:06 linux-7vph kernel:  [<c01047fe>] ? timer_interrupt+0x3a/0x41
> Apr 21 00:44:06 linux-7vph kernel:  [<c0158a2a>] ?
> handle_IRQ_event+0x8c/0x142
> Apr 21 00:44:06 linux-7vph kernel:  [<c0306e6f>] ? _spin_lock+0x8/0xb
> Apr 21 00:44:06 linux-7vph kernel:  [<c0159d5a>] ?
> handle_edge_irq+0x108/0x111
> Apr 21 00:44:06 linux-7vph kernel:  [<c012bcaa>] ? __do_softirq+0x12b/0x133
> Apr 21 00:44:06 linux-7vph kernel:  [<c0197025>] sys_ioctl+0x40/0x5a
> Apr 21 00:44:06 linux-7vph kernel:  [<c01029d4>] sysenter_do_call+0x12/0x28
> Apr 21 00:44:06 linux-7vph kernel: ---[ end trace e40fb1797bec2bee ]---
> Apr 21 00:44:06 linux-7vph kernel: [drm:gm45_get_vblank_counter] *ERROR*
> trying to get vblank count for disabled pipe 0
> Apr 21 00:44:10 linux-7vph bonobo-activation-server (niella-19973):

That is

		WARN_ON(!list_empty(&dev_priv->mm.flushing_list));

Does the warning persist in 2.6.30-rc3 or later?  If so, we should add
it to the post-2.6.29 regression list (if there's any room left).

Thanks.