Skip to contentSkip to conversation

bisected: i915 modeset broken in ac9b8236551d1177fd07b56aef9b565d1864420d

DISCUSSION Meelis Roos 3 participants 6 messages Dec 14, 2015 – Jan 7, 2016

lore
Share & tools
lore

Meelis Roos

author Reviewed-by Patrik Jakobsson
#0 lore raw

Between 4.4-rc3 and 4.4-rc4, i915 modesetting broke on my i5-2400 PC. Instead of seeing the new dense graphics mode, I see the last VGA text lines and no X appears either.

I saw something similar on I865G but have not had time to check if it is the same issue.

ac9b8236551d1177fd07b56aef9b565d1864420d is the first bad commit commit ac9b8236551d1177fd07b56aef9b565d1864420d

Author: Ville Syrjälä <[email protected]>
Date:   Fri Nov 27 18:55:26 2015 +0200

drm/i915: Introduce a gmbus power domain

Currently the gmbus code uses intel_aux_display_runtime_get/put in an effort to make sure the hardware is powered up sufficiently for gmbus. That function only takes the runtime PM reference which on VLV/CHV/BXT is not enough. We need the disp2d/pipe-a well on VLV/CHV and power well 2 on BXT. So add a new power domnain for gmbus and kill off the now unused intel_aux_display_runtime_get/put. And change intel_hdmi_set_edid() to use the gmbus power domain too since that's all we need there.

Also toss in a BUILD_BUG_ON() to catch problems if we run out of bits for power domains. We're already really close to the limit...

[Patrik: Add gmbus string to debugfs output]

:040000 040000 39379146d7e6dda8a4d5f8781ee3d307cce8c47e f4f09fae0485ad6263d31d425296fa9cd7de343b M drivers

Signature · 5 lines
Meelis Roos ([email protected])
_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel
#1 lore raw
Quoted 1 line · Meelis Roos

Between 4.4-rc3 and 4.4-rc4, i915 modesetting broke on my i5-2400 PC.

That would seem to be SNB.

Quoted 2 lines

Instead of seeing the new dense graphics mode, I see the last VGA text lines and no X appears either.

That's a bit weird. SNB has no power power wells, so only runtime PM could be a factor, but it should not kick in that fast during boot even if you enable it before loading the driver since we set the delay to 10 seconds.

And in any case the commit you list shouldn't really change anything for SNB. We used to grab a rpm reference for gmbus via intel_aux_display_runtime_get() and now we get it via the GMBUS power domain instead.

I would be more tempted to blame the live status check, except it was
already included in 4.4-rc1 I think. We didn't have a power well
reference for it until
commit 0f5a9be15797 ("drm/i915: take a power domain reference while checking the HDMI live status") but that's already in v4.4-rc4.

In any case the live status check was already reported to cause failures (at least on some SNB IIRC). There have been attempts to remedy it since:
http://lists.freedesktop.org/archives/intel-gfx/2015-December/082689.html
http://lists.freedesktop.org/archives/intel-gfx/2015-December/082932.html

So this bisect result is somewhat mysterious. A full dmesg with drm.debug=0xe with and without the offending patch reverted would be helpful. And might be best to attach those into a bug report
(https://bugs.freedesktop.org/ -> DRI -> DRM/Intel) so that we don't lose track of them.

Oh, are we even talking about HDMI/DVI here, or something else?

Quoted 37 lines

I saw something similar on I865G but have not had time to check if it is the same issue.

ac9b8236551d1177fd07b56aef9b565d1864420d is the first bad commit commit ac9b8236551d1177fd07b56aef9b565d1864420d

Author: Ville Syrjälä <[email protected]>
Date:   Fri Nov 27 18:55:26 2015 +0200

drm/i915: Introduce a gmbus power domain

Currently the gmbus code uses intel_aux_display_runtime_get/put in an effort to make sure the hardware is powered up sufficiently for gmbus. That function only takes the runtime PM reference which on VLV/CHV/BXT is not enough. We need the disp2d/pipe-a well on VLV/CHV and power well 2 on BXT. So add a new power domnain for gmbus and kill off the now unused intel_aux_display_runtime_get/put. And change intel_hdmi_set_edid() to use the gmbus power domain too since that's all we need there.

Also toss in a BUILD_BUG_ON() to catch problems if we run out of bits for power domains. We're already really close to the limit...

[Patrik: Add gmbus string to debugfs output]

:040000 040000 39379146d7e6dda8a4d5f8781ee3d307cce8c47e f4f09fae0485ad6263d31d425296fa9cd7de343b M drivers

Signature · 1 line
Meelis Roos ([email protected])
Signature · 6 lines
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#2 lore raw
Quoted 4 lines · Meelis Roos
Quoted 1 line

Between 4.4-rc3 and 4.4-rc4, i915 modesetting broke on my i5-2400 PC.

That would seem to be SNB.

Yes.

Quoted 8 lines
Quoted 2 lines

Instead of seeing the new dense graphics mode, I see the last VGA text lines and no X appears either.

That's a bit weird. SNB has no power power wells, so only runtime PM could be a factor, but it should not kick in that fast during boot even if you enable it before loading the driver since we set the delay to 10 seconds.

Quoted 4 lines

And in any case the commit you list shouldn't really change anything for SNB. We used to grab a rpm reference for gmbus via intel_aux_display_runtime_get() and now we get it via the GMBUS power domain instead.

I captured dmesg from failing boot, from system logs. gmbus has something to do with it:

[drm:i915_dump_device_info] i915 device info: gen=6, pciid=0x0102 rev=0x09 flags=need_gfx_hws,has_fbc,has_hotplug,has_llc, [drm:intel_detect_pch] Found CougarPoint PCH
[drm] Memory usable by graphics device = 2048M [drm:i915_gem_gtt_init] GMADR size = 256M [drm:i915_gem_gtt_init] GTT stolen size = 32M [drm:i915_gem_gtt_init] ppgtt mode: 1
[drm] Replacing VGA console driver
Console: switching to colour dummy device 80x25

BUG: unable to handle kernel NULL pointer dereference at           (null)
IP: [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
PGD 0 
Oops: 0002 [#1] SMP 
Modules linked in: i915(+) x86_pkg_temp_thermal kvm_intel kvm irqbypass video crc32c_intel i2c_algo_bit aesni_intel aes_x86_64 glue_helper lrw drm_kms_helper syscopyarea sysfillrect ablk_helper cryptd iTCO_wdt sysimgblt iTCO_vendor_support fb_sys_fops snd_hda_codec_realtek drm psmouse snd_hda_codec_generic e1000e xhci_pci xhci_hcd snd_hda_intel pcspkr snd_hda_codec snd_hwdep snd_hda_core snd_pcm_oss snd_mixer_oss i2c_i801 snd_pcm ehci_pci ehci_hcd nuvoton_cir usbcore snd_timer parport_pc ptp pps_core evdev rc_core parport snd usb_common soundcore tpm_tis tpm floppy lpc_ich mfd_core md_mod w83627ehf hwmon_vid coretemp hwmon eeprom i2c_core loop autofs4
CPU: 0 PID: 390 Comm: systemd-udevd Not tainted 4.4.0-rc2-00006-gac9b823 #185
Hardware name:                  /DQ67OW, BIOS SWQ6710H.86A.0066.2012.1105.1504 11/05/2012
task: ffff8800b7e48c40 ti: ffff880233420000 task.ti: ffff880233420000
RIP: 0010:[<ffffffff81519c74>]  [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
RSP: 0018:ffff880233423620  EFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff8800b6a594f8 RCX: ffff8800b7e48c40
RDX: 0000000000000001 RSI: ffff8800b7e48c40 RDI: ffff8800b6a594fc
RBP: ffff880233423670 R08: 0000000000000000 R09: 0000000000000001
R10: ffff8800b6a507e8 R11: 0000000000000013 R12: ffff8800b7e48c40
R13: ffff8800b6a594fc R14: 00000000ffffffff R15: ffff8800b6a59500
FS:  00007f58846428c0(0000) GS:ffff88023e200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 0000000233424000 CR4: 00000000000406f0
Stack:
 ffff8800b6a59500 0000000000000000 ffff8802353b7148 0000000000000206
 ffff8800b6a50000 ffff8800b6a594f8 000000000000001d ffff8800b6a594f8
 ffff8800b6a50000 ffff8800b6a50000 00000000fffeea06 ffffffff81519d1b
Call Trace:
 [<ffffffff81519d1b>] ? mutex_lock+0x1b/0x30
 [<ffffffffa0423859>] ? intel_display_power_get+0x29/0xe0 [i915]
 [<ffffffffa04af048>] ? gmbus_xfer+0x38/0x680 [i915]
 [<ffffffff81073113>] ? try_to_wake_up+0x43/0x320
 [<ffffffffa0015cc6>] ? __i2c_transfer+0x106/0x380 [i2c_core]
 [<ffffffffa0015fad>] ? i2c_transfer+0x6d/0xa0 [i2c_core]
 [<ffffffffa0016185>] ? i2c_smbus_xfer_emulated+0x105/0x4c0 [i2c_core]
 [<ffffffff81086f3e>] ? __wake_up_common+0x4e/0x90
 [<ffffffff812a7cab>] ? idr_get_empty_slot+0x18b/0x390
 [<ffffffffa0016658>] ? i2c_smbus_xfer+0x118/0x2e0 [i2c_core]
 [<ffffffffa00168e5>] ? i2c_default_probe+0xc5/0x110 [i2c_core]
 [<ffffffffa00154d9>] ? i2c_check_addr_busy+0x39/0x60 [i2c_core]
 [<ffffffffa0018339>] ? i2c_do_add_adapter+0x159/0x260 [i2c_core]
 [<ffffffffa0018440>] ? i2c_do_add_adapter+0x260/0x260 [i2c_core]
 [<ffffffff81382b85>] ? bus_for_each_drv+0x55/0x90
 [<ffffffffa0017fb6>] ? i2c_register_adapter+0x1c6/0x320 [i2c_core]
 [<ffffffffa04afa80>] ? intel_setup_gmbus+0x220/0x310 [i915]
 [<ffffffffa04ba1eb>] ? i915_driver_load+0x4eb/0x15e0 [i915]
 [<ffffffffa026493c>] ? drm_dev_register+0x9c/0xb0 [drm]
 [<ffffffffa0266c49>] ? drm_get_pci_dev+0x89/0x1d0 [drm]
 [<ffffffff812d8c71>] ? pci_device_probe+0x81/0xe0
 [<ffffffff81384857>] ? driver_probe_device+0x147/0x310
 [<ffffffff81384a9b>] ? __driver_attach+0x7b/0x80
 [<ffffffff81384a20>] ? driver_probe_device+0x310/0x310
 [<ffffffff81382ada>] ? bus_for_each_dev+0x5a/0x90
 [<ffffffff81383e34>] ? bus_add_driver+0x1a4/0x220
 [<ffffffffa033c000>] ? 0xffffffffa033c000
 [<ffffffff813851e7>] ? driver_register+0x57/0xc0
 [<ffffffff810003b1>] ? do_one_initcall+0x81/0x1b0
 [<ffffffff8116b601>] ? kmem_cache_alloc_trace+0x31/0x120
 [<ffffffff8111fb77>] ? do_init_module+0x5b/0x1dc
 [<ffffffff810c16c2>] ? load_module+0x1e52/0x2220
 [<ffffffff810be3b0>] ? __symbol_put+0x50/0x50
 [<ffffffff810c1c35>] ? SyS_finit_module+0x85/0x90
 [<ffffffff8151b9db>] ? entry_SYSCALL_64_fastpath+0x16/0x6a
Code: e8 c2 1a 00 00 8b 03 83 f8 01 0f 84 92 00 00 00 48 8b 43 10 4c 8d 7b 08 48 89 63 10 41 be ff ff ff ff 4c 89 3c 24 48 89 44 24 08 <48> 89 20 4c 89 64 24 10 eb 19 49 c7 04 24 02 00 00 00 c6 43 04 
RIP  [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
 RSP <ffff880233423620>
CR2: 0000000000000000
---[ end trace 5e2e7e41ffefe21d ]---
Quoted 5 lines

So this bisect result is somewhat mysterious. A full dmesg with drm.debug=0xe with and without the offending patch reverted would be helpful. And might be best to attach those into a bug report
(https://bugs.freedesktop.org/ -> DRI -> DRM/Intel) so that we don't lose track of them.

Full dmesgs at https://bugs.freedesktop.org/show_bug.cgi?id=93608

Quoted 2 lines

Oh, are we even talking about HDMI/DVI here, or something else?

DVI

Quoted 40 lines
Quoted 38 lines

I saw something similar on I865G but have not had time to check if it is the same issue.

ac9b8236551d1177fd07b56aef9b565d1864420d is the first bad commit commit ac9b8236551d1177fd07b56aef9b565d1864420d

Author: Ville Syrjälä <[email protected]>
Date:   Fri Nov 27 18:55:26 2015 +0200

drm/i915: Introduce a gmbus power domain

Currently the gmbus code uses intel_aux_display_runtime_get/put in an effort to make sure the hardware is powered up sufficiently for gmbus. That function only takes the runtime PM reference which on VLV/CHV/BXT is not enough. We need the disp2d/pipe-a well on VLV/CHV and power well 2 on BXT. So add a new power domnain for gmbus and kill off the now unused intel_aux_display_runtime_get/put. And change intel_hdmi_set_edid() to use the gmbus power domain too since that's all we need there.

Also toss in a BUILD_BUG_ON() to catch problems if we run out of bits for power domains. We're already really close to the limit...

[Patrik: Add gmbus string to debugfs output]

:040000 040000 39379146d7e6dda8a4d5f8781ee3d307cce8c47e f4f09fae0485ad6263d31d425296fa9cd7de343b M drivers

Signature · 1 line
Meelis Roos ([email protected])
Signature · 5 lines
Meelis Roos ([email protected])
_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel
#3 lore raw
Quoted 64 lines · Meelis Roos
Quoted 4 lines · Meelis Roos
Quoted 1 line

Between 4.4-rc3 and 4.4-rc4, i915 modesetting broke on my i5-2400 PC.

That would seem to be SNB.

Yes.

Quoted 8 lines
Quoted 2 lines

Instead of seeing the new dense graphics mode, I see the last VGA text lines and no X appears either.

That's a bit weird. SNB has no power power wells, so only runtime PM could be a factor, but it should not kick in that fast during boot even if you enable it before loading the driver since we set the delay to 10 seconds.

Quoted 4 lines

And in any case the commit you list shouldn't really change anything for SNB. We used to grab a rpm reference for gmbus via intel_aux_display_runtime_get() and now we get it via the GMBUS power domain instead.

I captured dmesg from failing boot, from system logs. gmbus has something to do with it:

[drm:i915_dump_device_info] i915 device info: gen=6, pciid=0x0102 rev=0x09 flags=need_gfx_hws,has_fbc,has_hotplug,has_llc, [drm:intel_detect_pch] Found CougarPoint PCH
[drm] Memory usable by graphics device = 2048M [drm:i915_gem_gtt_init] GMADR size = 256M [drm:i915_gem_gtt_init] GTT stolen size = 32M [drm:i915_gem_gtt_init] ppgtt mode: 1
[drm] Replacing VGA console driver
Console: switching to colour dummy device 80x25

BUG: unable to handle kernel NULL pointer dereference at           (null)
IP: [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
PGD 0 
Oops: 0002 [#1] SMP 
Modules linked in: i915(+) x86_pkg_temp_thermal kvm_intel kvm irqbypass video crc32c_intel i2c_algo_bit aesni_intel aes_x86_64 glue_helper lrw drm_kms_helper syscopyarea sysfillrect ablk_helper cryptd iTCO_wdt sysimgblt iTCO_vendor_support fb_sys_fops snd_hda_codec_realtek drm psmouse snd_hda_codec_generic e1000e xhci_pci xhci_hcd snd_hda_intel pcspkr snd_hda_codec snd_hwdep snd_hda_core snd_pcm_oss snd_mixer_oss i2c_i801 snd_pcm ehci_pci ehci_hcd nuvoton_cir usbcore snd_timer parport_pc ptp pps_core evdev rc_core parport snd usb_common soundcore tpm_tis tpm floppy lpc_ich mfd_core md_mod w83627ehf hwmon_vid coretemp hwmon eeprom i2c_core loop autofs4
CPU: 0 PID: 390 Comm: systemd-udevd Not tainted 4.4.0-rc2-00006-gac9b823 #185
Hardware name:                  /DQ67OW, BIOS SWQ6710H.86A.0066.2012.1105.1504 11/05/2012
task: ffff8800b7e48c40 ti: ffff880233420000 task.ti: ffff880233420000
RIP: 0010:[<ffffffff81519c74>]  [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
RSP: 0018:ffff880233423620  EFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff8800b6a594f8 RCX: ffff8800b7e48c40
RDX: 0000000000000001 RSI: ffff8800b7e48c40 RDI: ffff8800b6a594fc
RBP: ffff880233423670 R08: 0000000000000000 R09: 0000000000000001
R10: ffff8800b6a507e8 R11: 0000000000000013 R12: ffff8800b7e48c40
R13: ffff8800b6a594fc R14: 00000000ffffffff R15: ffff8800b6a59500
FS:  00007f58846428c0(0000) GS:ffff88023e200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 0000000233424000 CR4: 00000000000406f0
Stack:
 ffff8800b6a59500 0000000000000000 ffff8802353b7148 0000000000000206
 ffff8800b6a50000 ffff8800b6a594f8 000000000000001d ffff8800b6a594f8
 ffff8800b6a50000 ffff8800b6a50000 00000000fffeea06 ffffffff81519d1b
Call Trace:
 [<ffffffff81519d1b>] ? mutex_lock+0x1b/0x30
 [<ffffffffa0423859>] ? intel_display_power_get+0x29/0xe0 [i915]
 [<ffffffffa04af048>] ? gmbus_xfer+0x38/0x680 [i915]
 [<ffffffff81073113>] ? try_to_wake_up+0x43/0x320
 [<ffffffffa0015cc6>] ? __i2c_transfer+0x106/0x380 [i2c_core]
 [<ffffffffa0015fad>] ? i2c_transfer+0x6d/0xa0 [i2c_core]
 [<ffffffffa0016185>] ? i2c_smbus_xfer_emulated+0x105/0x4c0 [i2c_core]
 [<ffffffff81086f3e>] ? __wake_up_common+0x4e/0x90
 [<ffffffff812a7cab>] ? idr_get_empty_slot+0x18b/0x390
Quoted 8 lines
 [<ffffffffa0016658>] ? i2c_smbus_xfer+0x118/0x2e0 [i2c_core]
 [<ffffffffa00168e5>] ? i2c_default_probe+0xc5/0x110 [i2c_core]
 [<ffffffffa00154d9>] ? i2c_check_addr_busy+0x39/0x60 [i2c_core]
 [<ffffffffa0018339>] ? i2c_do_add_adapter+0x159/0x260 [i2c_core]
 [<ffffffffa0018440>] ? i2c_do_add_adapter+0x260/0x260 [i2c_core]
 [<ffffffff81382b85>] ? bus_for_each_drv+0x55/0x90
 [<ffffffffa0017fb6>] ? i2c_register_adapter+0x1c6/0x320 [i2c_core]
 [<ffffffffa04afa80>] ? intel_setup_gmbus+0x220/0x310 [i915]

intel_setup_gmbus registers the i2c adapters, which does transfers on the i2c bus on probe, and this happens before intel_power_domains_init which initializes the power domain lock.

The bisect and backtrace make sense and are not mysterious at all.

Not sure of the fix though, are we better off changing the init order, or making sure the probes don't happen or don't screw us up.

BR,
Jani.

Quoted 78 lines

[<ffffffffa04ba1eb>] ? i915_driver_load+0x4eb/0x15e0 [i915] [<ffffffffa026493c>] ? drm_dev_register+0x9c/0xb0 [drm] [<ffffffffa0266c49>] ? drm_get_pci_dev+0x89/0x1d0 [drm] [<ffffffff812d8c71>] ? pci_device_probe+0x81/0xe0 [<ffffffff81384857>] ? driver_probe_device+0x147/0x310 [<ffffffff81384a9b>] ? __driver_attach+0x7b/0x80 [<ffffffff81384a20>] ? driver_probe_device+0x310/0x310 [<ffffffff81382ada>] ? bus_for_each_dev+0x5a/0x90 [<ffffffff81383e34>] ? bus_add_driver+0x1a4/0x220 [<ffffffffa033c000>] ? 0xffffffffa033c000 [<ffffffff813851e7>] ? driver_register+0x57/0xc0 [<ffffffff810003b1>] ? do_one_initcall+0x81/0x1b0 [<ffffffff8116b601>] ? kmem_cache_alloc_trace+0x31/0x120 [<ffffffff8111fb77>] ? do_init_module+0x5b/0x1dc [<ffffffff810c16c2>] ? load_module+0x1e52/0x2220 [<ffffffff810be3b0>] ? __symbol_put+0x50/0x50 [<ffffffff810c1c35>] ? SyS_finit_module+0x85/0x90 [<ffffffff8151b9db>] ? entry_SYSCALL_64_fastpath+0x16/0x6a

Code: e8 c2 1a 00 00 8b 03 83 f8 01 0f 84 92 00 00 00 48 8b 43 10 4c 8d 7b 08 48 89 63 10 41 be ff ff ff ff 4c 89 3c 24 48 89 44 24 08 <48> 89 20 4c 89 64 24 10 eb 19 49 c7 04 24 02 00 00 00 c6 43 04 
RIP  [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
 RSP <ffff880233423620>
CR2: 0000000000000000
---[ end trace 5e2e7e41ffefe21d ]---
Quoted 5 lines

So this bisect result is somewhat mysterious. A full dmesg with drm.debug=0xe with and without the offending patch reverted would be helpful. And might be best to attach those into a bug report
(https://bugs.freedesktop.org/ -> DRI -> DRM/Intel) so that we don't lose track of them.

Full dmesgs at https://bugs.freedesktop.org/show_bug.cgi?id=93608

Quoted 2 lines

Oh, are we even talking about HDMI/DVI here, or something else?

DVI

Quoted 40 lines
Quoted 38 lines

I saw something similar on I865G but have not had time to check if it is the same issue.

ac9b8236551d1177fd07b56aef9b565d1864420d is the first bad commit commit ac9b8236551d1177fd07b56aef9b565d1864420d

Author: Ville Syrjälä <[email protected]>
Date:   Fri Nov 27 18:55:26 2015 +0200

drm/i915: Introduce a gmbus power domain

Currently the gmbus code uses intel_aux_display_runtime_get/put in an effort to make sure the hardware is powered up sufficiently for gmbus. That function only takes the runtime PM reference which on VLV/CHV/BXT is not enough. We need the disp2d/pipe-a well on VLV/CHV and power well 2 on BXT. So add a new power domnain for gmbus and kill off the now unused intel_aux_display_runtime_get/put. And change intel_hdmi_set_edid() to use the gmbus power domain too since that's all we need there.

Also toss in a BUILD_BUG_ON() to catch problems if we run out of bits for power domains. We're already really close to the limit...

[Patrik: Add gmbus string to debugfs output]

:040000 040000 39379146d7e6dda8a4d5f8781ee3d307cce8c47e f4f09fae0485ad6263d31d425296fa9cd7de343b M drivers

Signature · 1 line
Meelis Roos ([email protected])
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#4 lore raw
Quoted 74 lines · Jani Nikula
Quoted 64 lines · Meelis Roos
Quoted 4 lines · Meelis Roos
Quoted 1 line

Between 4.4-rc3 and 4.4-rc4, i915 modesetting broke on my i5-2400 PC.

That would seem to be SNB.

Yes.

Quoted 8 lines
Quoted 2 lines

Instead of seeing the new dense graphics mode, I see the last VGA text lines and no X appears either.

That's a bit weird. SNB has no power power wells, so only runtime PM could be a factor, but it should not kick in that fast during boot even if you enable it before loading the driver since we set the delay to 10 seconds.

Quoted 4 lines

And in any case the commit you list shouldn't really change anything for SNB. We used to grab a rpm reference for gmbus via intel_aux_display_runtime_get() and now we get it via the GMBUS power domain instead.

I captured dmesg from failing boot, from system logs. gmbus has something to do with it:

[drm:i915_dump_device_info] i915 device info: gen=6, pciid=0x0102 rev=0x09 flags=need_gfx_hws,has_fbc,has_hotplug,has_llc, [drm:intel_detect_pch] Found CougarPoint PCH
[drm] Memory usable by graphics device = 2048M [drm:i915_gem_gtt_init] GMADR size = 256M [drm:i915_gem_gtt_init] GTT stolen size = 32M [drm:i915_gem_gtt_init] ppgtt mode: 1
[drm] Replacing VGA console driver
Console: switching to colour dummy device 80x25

BUG: unable to handle kernel NULL pointer dereference at           (null)
IP: [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
PGD 0 
Oops: 0002 [#1] SMP 
Modules linked in: i915(+) x86_pkg_temp_thermal kvm_intel kvm irqbypass video crc32c_intel i2c_algo_bit aesni_intel aes_x86_64 glue_helper lrw drm_kms_helper syscopyarea sysfillrect ablk_helper cryptd iTCO_wdt sysimgblt iTCO_vendor_support fb_sys_fops snd_hda_codec_realtek drm psmouse snd_hda_codec_generic e1000e xhci_pci xhci_hcd snd_hda_intel pcspkr snd_hda_codec snd_hwdep snd_hda_core snd_pcm_oss snd_mixer_oss i2c_i801 snd_pcm ehci_pci ehci_hcd nuvoton_cir usbcore snd_timer parport_pc ptp pps_core evdev rc_core parport snd usb_common soundcore tpm_tis tpm floppy lpc_ich mfd_core md_mod w83627ehf hwmon_vid coretemp hwmon eeprom i2c_core loop autofs4
CPU: 0 PID: 390 Comm: systemd-udevd Not tainted 4.4.0-rc2-00006-gac9b823 #185
Hardware name:                  /DQ67OW, BIOS SWQ6710H.86A.0066.2012.1105.1504 11/05/2012
task: ffff8800b7e48c40 ti: ffff880233420000 task.ti: ffff880233420000
RIP: 0010:[<ffffffff81519c74>]  [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
RSP: 0018:ffff880233423620  EFLAGS: 00010282
RAX: 0000000000000000 RBX: ffff8800b6a594f8 RCX: ffff8800b7e48c40
RDX: 0000000000000001 RSI: ffff8800b7e48c40 RDI: ffff8800b6a594fc
RBP: ffff880233423670 R08: 0000000000000000 R09: 0000000000000001
R10: ffff8800b6a507e8 R11: 0000000000000013 R12: ffff8800b7e48c40
R13: ffff8800b6a594fc R14: 00000000ffffffff R15: ffff8800b6a59500
FS:  00007f58846428c0(0000) GS:ffff88023e200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 0000000233424000 CR4: 00000000000406f0
Stack:
 ffff8800b6a59500 0000000000000000 ffff8802353b7148 0000000000000206
 ffff8800b6a50000 ffff8800b6a594f8 000000000000001d ffff8800b6a594f8
 ffff8800b6a50000 ffff8800b6a50000 00000000fffeea06 ffffffff81519d1b
Call Trace:
 [<ffffffff81519d1b>] ? mutex_lock+0x1b/0x30
 [<ffffffffa0423859>] ? intel_display_power_get+0x29/0xe0 [i915]
 [<ffffffffa04af048>] ? gmbus_xfer+0x38/0x680 [i915]
 [<ffffffff81073113>] ? try_to_wake_up+0x43/0x320
 [<ffffffffa0015cc6>] ? __i2c_transfer+0x106/0x380 [i2c_core]
 [<ffffffffa0015fad>] ? i2c_transfer+0x6d/0xa0 [i2c_core]
 [<ffffffffa0016185>] ? i2c_smbus_xfer_emulated+0x105/0x4c0 [i2c_core]
 [<ffffffff81086f3e>] ? __wake_up_common+0x4e/0x90
 [<ffffffff812a7cab>] ? idr_get_empty_slot+0x18b/0x390
Quoted 8 lines
 [<ffffffffa0016658>] ? i2c_smbus_xfer+0x118/0x2e0 [i2c_core]
 [<ffffffffa00168e5>] ? i2c_default_probe+0xc5/0x110 [i2c_core]
 [<ffffffffa00154d9>] ? i2c_check_addr_busy+0x39/0x60 [i2c_core]
 [<ffffffffa0018339>] ? i2c_do_add_adapter+0x159/0x260 [i2c_core]
 [<ffffffffa0018440>] ? i2c_do_add_adapter+0x260/0x260 [i2c_core]
 [<ffffffff81382b85>] ? bus_for_each_drv+0x55/0x90
 [<ffffffffa0017fb6>] ? i2c_register_adapter+0x1c6/0x320 [i2c_core]
 [<ffffffffa04afa80>] ? intel_setup_gmbus+0x220/0x310 [i915]
Quoted 8 lines

intel_setup_gmbus registers the i2c adapters, which does transfers on the i2c bus on probe, and this happens before intel_power_domains_init which initializes the power domain lock.

The bisect and backtrace make sense and are not mysterious at all.

Not sure of the fix though, are we better off changing the init order, or making sure the probes don't happen or don't screw us up.

So I started wondering why we are not seeing this. To reproduce, looks like you'll need to have an i2c driver with class I2C_CLASS_DDC for the i2c detect (and the bug) to happen. In tree, the only ones seem to be

drivers/misc/eeprom/eeprom.c
drivers/staging/olpc_dcon/olpc_dcon.c

I presume you have one or the other.

No matter what, userspace can access the adapter right away when we register it, so this needs to be fixed along the lines of [1].

BR,
Jani.

[1] http://patchwork.freedesktop.org/patch/msgid/[email protected]

Quoted 84 lines

BR,
Jani.

Quoted 78 lines

[<ffffffffa04ba1eb>] ? i915_driver_load+0x4eb/0x15e0 [i915] [<ffffffffa026493c>] ? drm_dev_register+0x9c/0xb0 [drm] [<ffffffffa0266c49>] ? drm_get_pci_dev+0x89/0x1d0 [drm] [<ffffffff812d8c71>] ? pci_device_probe+0x81/0xe0 [<ffffffff81384857>] ? driver_probe_device+0x147/0x310 [<ffffffff81384a9b>] ? __driver_attach+0x7b/0x80 [<ffffffff81384a20>] ? driver_probe_device+0x310/0x310 [<ffffffff81382ada>] ? bus_for_each_dev+0x5a/0x90 [<ffffffff81383e34>] ? bus_add_driver+0x1a4/0x220 [<ffffffffa033c000>] ? 0xffffffffa033c000 [<ffffffff813851e7>] ? driver_register+0x57/0xc0 [<ffffffff810003b1>] ? do_one_initcall+0x81/0x1b0 [<ffffffff8116b601>] ? kmem_cache_alloc_trace+0x31/0x120 [<ffffffff8111fb77>] ? do_init_module+0x5b/0x1dc [<ffffffff810c16c2>] ? load_module+0x1e52/0x2220 [<ffffffff810be3b0>] ? __symbol_put+0x50/0x50 [<ffffffff810c1c35>] ? SyS_finit_module+0x85/0x90 [<ffffffff8151b9db>] ? entry_SYSCALL_64_fastpath+0x16/0x6a

Code: e8 c2 1a 00 00 8b 03 83 f8 01 0f 84 92 00 00 00 48 8b 43 10 4c 8d 7b 08 48 89 63 10 41 be ff ff ff ff 4c 89 3c 24 48 89 44 24 08 <48> 89 20 4c 89 64 24 10 eb 19 49 c7 04 24 02 00 00 00 c6 43 04 
RIP  [<ffffffff81519c74>] __mutex_lock_slowpath+0x74/0x100
 RSP <ffff880233423620>
CR2: 0000000000000000
---[ end trace 5e2e7e41ffefe21d ]---
Quoted 5 lines

So this bisect result is somewhat mysterious. A full dmesg with drm.debug=0xe with and without the offending patch reverted would be helpful. And might be best to attach those into a bug report
(https://bugs.freedesktop.org/ -> DRI -> DRM/Intel) so that we don't lose track of them.

Full dmesgs at https://bugs.freedesktop.org/show_bug.cgi?id=93608

Quoted 2 lines

Oh, are we even talking about HDMI/DVI here, or something else?

DVI

Quoted 40 lines
Quoted 38 lines

I saw something similar on I865G but have not had time to check if it is the same issue.

ac9b8236551d1177fd07b56aef9b565d1864420d is the first bad commit commit ac9b8236551d1177fd07b56aef9b565d1864420d

Author: Ville Syrjälä <[email protected]>
Date:   Fri Nov 27 18:55:26 2015 +0200

drm/i915: Introduce a gmbus power domain

Currently the gmbus code uses intel_aux_display_runtime_get/put in an effort to make sure the hardware is powered up sufficiently for gmbus. That function only takes the runtime PM reference which on VLV/CHV/BXT is not enough. We need the disp2d/pipe-a well on VLV/CHV and power well 2 on BXT. So add a new power domnain for gmbus and kill off the now unused intel_aux_display_runtime_get/put. And change intel_hdmi_set_edid() to use the gmbus power domain too since that's all we need there.

Also toss in a BUILD_BUG_ON() to catch problems if we run out of bits for power domains. We're already really close to the limit...

[Patrik: Add gmbus string to debugfs output]

:040000 040000 39379146d7e6dda8a4d5f8781ee3d307cce8c47e f4f09fae0485ad6263d31d425296fa9cd7de343b M drivers

Signature · 1 line
Meelis Roos ([email protected])
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
#5 lore raw
Quoted 6 lines
Quoted 6 lines

intel_setup_gmbus registers the i2c adapters, which does transfers on the i2c bus on probe, and this happens before intel_power_domains_init which initializes the power domain lock.

The bisect and backtrace make sense and are not mysterious at all.

Quoted 8 lines
Quoted 2 lines

Not sure of the fix though, are we better off changing the init order, or making sure the probes don't happen or don't screw us up.

So I started wondering why we are not seeing this. To reproduce, looks like you'll need to have an i2c driver with class I2C_CLASS_DDC for the i2c detect (and the bug) to happen. In tree, the only ones seem to be

drivers/misc/eeprom/eeprom.c

Yes, I have that for DIMM SPD information access.

Quoted 12 lines

drivers/staging/olpc_dcon/olpc_dcon.c

I presume you have one or the other.

No matter what, userspace can access the adapter right away when we register it, so this needs to be fixed along the lines of [1].

BR,
Jani.

[1] http://patchwork.freedesktop.org/patch/msgid/[email protected]

Signature · 5 lines
Meelis Roos ([email protected])
_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel