linux-kernel - Re: [tpmdd-devel] [BUG, bisect, PATCH 04/10] tpm: move the PPI attributes to character device directory.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20151105164758.GA2557@newt.localdomain>
Date:	Thu, 5 Nov 2015 08:47:58 -0800
From:	Jeremiah Mahler <jmmahler@...il.com>
To:	Jarkko Sakkinen <jarkko.sakkinen@...ux.intel.com>
Cc:	tpmdd-devel@...ts.sourceforge.net, linux-kernel@...r.kernel.org,
	peterhuewe@....de, gregkh@...uxfoundation.org,
	jgunthorpe@...idianresearch.com, dhowells@...hat.com,
	artem.bityutskiy@...ux.intel.com,
	Marcel Selhorst <tpmdd@...horst.net>
Subject: Re: [tpmdd-devel] [BUG, bisect, PATCH 04/10] tpm: move the PPI
 attributes to character device directory.

Jarkko,

On Thu, Nov 05, 2015 at 01:05:45PM +0200, Jarkko Sakkinen wrote:
> On Thu, Nov 05, 2015 at 11:22:55AM +0200, Jarkko Sakkinen wrote:
> > On Wed, Nov 04, 2015 at 10:17:05AM -0800, Jeremiah Mahler wrote:
> > > Jarkko, all,
> > > 
[...]
> > > 
> > > The commit for this patch (9b774d5cf2db4) present in the latest
> > > linux-next (20151101+) breaks suspend/resume on an Acer C720 Chromebook.
> > > The computer will successfully suspend but when a resume is attempted
> > > a blank screen is displayed for a few seconds and then it reboots.
> > 
> > I think I have this same model at home (have to check). If it is, I can
> > try to reproduce it today when I get back to home.
> 
> Yup, it's C720P!
> 
> > What kind of environment are you running?
> > Are you able to acquire kernel logs?
> 
> Both of these questions still apply. Even if I create the same
> environment, your logs might have something useful embedded.
> 
> Thank you for reporting this issue! I'll try to find a way to reproduce
> and fix it at latest for -rc2.
> 
> > > -- 
> > > - Jeremiah Mahler
> 
> /Jarkko

I have attached the full dmesg.  The following snippets looked
particularly interesting.

  ...
  [    2.423070] input: PC Speaker as /devices/platform/pcspkr/input/input10
  [    2.453071] ------------[ cut here ]------------
  [    2.453078] WARNING: CPU: 0 PID: 237 at kernel/irq/manage.c:1379 __free_irq+0x90/0x1e0()
  [    2.453079] Trying to free already-free IRQ 6
  [    2.453110] Modules linked in: evdev serio_raw pcspkr cfg80211 sg i915 lpc_ich mfd_core i2c_i801 battery ac snd_hda_codec_realtek ath3k snd_hda_codec_generic i2c_algo_bit btusb drm_kms_helper btrtl btbcm btintel snd_hda_intel bluetooth drm snd_hda_codec snd_hwdep snd_hda_core video snd_pcm rfkill tpm_tis(+) snd_timer tpm snd button i2c_designware_pci i2c_designware_core shpchp i2c_core soundcore autofs4 ext4 crc16 mbcache jbd2 sd_mod ahci libahci fan sdhci_acpi sdhci libata mmc_core xhci_pci scsi_mod xhci_hcd thermal usbcore usb_common
  [    2.453113] CPU: 0 PID: 237 Comm: systemd-udevd Tainted: G        W       4.3.0-rc4+ #271
  [    2.453114] Hardware name: Acer Peppy, BIOS          04/30/2014
  [    2.453117]  ffffffff817288a1 ffffffff812c37a7 ffff880035a6fad0 ffffffff8106d8bd
  [    2.453119]  ffff880100421800 ffff880035a6fb20 ffff880100421800 0000000000000006
  [    2.453121]  ffff8801004218d8 ffffffff8106d93c ffffffff8171e8c8 ffff880000000020
  [    2.453122] Call Trace:
  [    2.453127]  [<ffffffff812c37a7>] ? dump_stack+0x40/0x59
  [    2.453132]  [<ffffffff8106d8bd>] ? warn_slowpath_common+0x7d/0xb0
  [    2.453135]  [<ffffffff8106d93c>] ? warn_slowpath_fmt+0x4c/0x50
  [    2.453140]  [<ffffffffa01dce22>] ? tpm_chip_register+0x142/0x190 [tpm]
  [    2.453142]  [<ffffffff810bd780>] ? __free_irq+0x90/0x1e0
  [    2.453144]  [<ffffffff810bd965>] ? free_irq+0x45/0xb0
  [    2.453148]  [<ffffffff813d8675>] ? release_nodes+0xf5/0x1c0
  [    2.453152]  [<ffffffff813d4c3d>] ? driver_probe_device+0xcd/0x480
  [    2.453155]  [<ffffffff813d506b>] ? __driver_attach+0x7b/0x80
  [    2.453158]  [<ffffffff813d4ff0>] ? driver_probe_device+0x480/0x480
  [    2.453161]  [<ffffffff813d2c4a>] ? bus_for_each_dev+0x5a/0x90
  [    2.453164]  [<ffffffff813d413f>] ? bus_add_driver+0x1df/0x270
  [    2.453166]  [<ffffffffa00d1000>] ? 0xffffffffa00d1000
  [    2.453169]  [<ffffffff813d5807>] ? driver_register+0x57/0xc0
  [    2.453173]  [<ffffffffa00d1025>] ? init_tis+0x25/0x1000 [tpm_tis]
  [    2.453178]  [<ffffffff811510c9>] ? free_pcppages_bulk+0xc9/0x450
  [    2.453179]  [<ffffffffa00d1000>] ? 0xffffffffa00d1000
  [    2.453182]  [<ffffffff81002122>] ? do_one_initcall+0xb2/0x200
  [    2.453185]  [<ffffffff8114953d>] ? do_init_module+0x5b/0x1dc
  [    2.453188]  [<ffffffff810e5637>] ? load_module+0x2197/0x27a0
  [    2.453190]  [<ffffffff810e1ed0>] ? __symbol_put+0x30/0x30
  [    2.453194]  [<ffffffff810e5e20>] ? SyS_finit_module+0x90/0xc0
  [    2.453198]  [<ffffffff815360b6>] ? entry_SYSCALL_64_fastpath+0x16/0x75
  [    2.453200] ---[ end trace 0835dfb15879b441 ]---
  [    2.453212] tpm_tis: probe of 00:08 failed with error -2
  [    2.456012] tpm_inf_pnp 00:08: Found TPM with ID IFX0102
  [    2.544707] usb 1-4: new full-speed USB device number 4 using xhci_hcd
  [    2.571036] __add_probed_i2c_device failed to register device 1-4a
  ...

Notice the "tpm_tis: probe of ... failed with error".  With a working
kernel this probe succeeds.  However, the probe fails, which causes it
to try unload the module which produces the "free already-free IRQ"
trace.  The IRQ message is a secondary issue.  Interrupts were disabled
due to a firmware bug and it is trying to free them anyway.

  ...
  [    2.244779] tpm_tis 00:08: [Firmware Bug]: TPM interrupt not working, polling instead
  ...

So the patch changed something which causes the probe to fail.  I
haven't figured out what that is yet.

-- 
- Jeremiah Mahler

View attachment "dmesg.bad" of type "text/plain" (50702 bytes)