Bug 87521

Summary:	i915 netbook boots into blank screen.
Product:	Drivers	Reporter:	Apostolos B. (barz621)
Component:	Video(DRI - Intel)	Assignee:	intel-gfx-bugs (intel-gfx-bugs)
Status:	RESOLVED CODE_FIX
Severity:	normal	CC:	alan, chris, darethehair, intel-gfx-bugs
Priority:	P1
Hardware:	All
OS:	Linux
Kernel Version:	3.17.1	Subsystem:
Regression:	Yes	Bisected commit-id:

Description Apostolos B. 2014-11-02 09:02:29 UTC

When booting with no kernel line parameters i cannot even ssh in it. It boots to a usable state (command prompt and can ssh) with i915.nomodeset=0. I tried acpi_osi=Linux and it boots to a blank screen but can ssh in it. There i got this:

Nov 02 10:51:12 testland kernel: BUG: unable to handle kernel NULL pointer dereference at 00000008
Nov 02 10:51:13 testland kernel: IP: [<f850fc7c>] drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm]
Nov 02 10:51:14 testland kernel: *pde = 00000000 
Nov 02 10:51:14 testland kernel: Oops: 0000 [#1] PREEMPT SMP 
Nov 02 10:51:14 testland kernel: Modules linked in: mac80211 cfg80211 iTCO_wdt iTCO_vendor_support i915(+) coretemp hp_wmi(+) hwmon sparse_keymap psmouse drm_kms_helper ecb pcspkr btusb uvcvid
Nov 02 10:51:14 testland kernel: 
Nov 02 10:51:14 testland kernel: CPU: 1 PID: 166 Comm: systemd-udevd Not tainted 3.17.1-1-ARCH #1
Nov 02 10:51:14 testland kernel: Hardware name: Hewlett-Packard HP Mini/361A, BIOS 361A0 Ver. F.10 12/25/2008
Nov 02 10:51:15 testland kernel: task: f59cddc0 ti: f4f92000 task.ti: f4f92000
Nov 02 10:51:15 testland kernel: EIP: 0060:[<f850fc7c>] EFLAGS: 00010286 CPU: 1
Nov 02 10:51:15 testland kernel: EIP is at drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm]
Nov 02 10:51:15 testland kernel: EAX: 00000000 EBX: fffffff8 ECX: 0012c000 EDX: f2885060
Nov 02 10:51:15 testland kernel: ESI: 00000000 EDI: f2ee40e4 EBP: f4f938a8 ESP: f4f9386c
Nov 02 10:51:15 testland kernel:  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Nov 02 10:51:15 testland kernel: CR0: 80050033 CR2: 00000008 CR3: 34f79000 CR4: 000007d0
Nov 02 10:51:15 testland kernel: Stack:
Nov 02 10:51:16 testland kernel:  f4f93878 f2885060 00000060 f94198dc 000080d0 00000000 0012c000 f2ee40e4
Nov 02 10:51:16 testland kernel:  f2ee40e4 f4f938a8 f94198dc ac88d68e 10000000 00010000 f4cee500 f4f93910
Nov 02 10:51:16 testland kernel:  f941e929 00010000 00000000 00000000 10000000 00000000 00000000 000020dc
Nov 02 10:51:16 testland kernel: Call Trace:
Nov 02 10:51:16 testland kernel:  [<f94198dc>] ? i915_gem_obj_lookup_or_create_vma+0x3c/0x130 [i915]
Nov 02 10:51:16 testland kernel:  [<f94198dc>] ? i915_gem_obj_lookup_or_create_vma+0x3c/0x130 [i915]
Nov 02 10:51:16 testland kernel:  [<f941e929>] i915_gem_object_pin+0x439/0x6d0 [i915]
Nov 02 10:51:16 testland kernel:  [<f9420b2c>] i915_gem_object_pin_to_display_plane+0x8c/0x1a0 [i915]
Nov 02 10:51:16 testland kernel:  [<f944a4c9>] intel_pin_and_fence_fb_obj+0xc9/0x150 [i915]
Nov 02 10:51:16 testland kernel:  [<f944ce3f>] __intel_set_mode+0x55f/0x14e0 [i915]
Nov 02 10:51:16 testland kernel:  [<f8510499>] ? drm_mode_object_get_reg+0x69/0x80 [drm]
Nov 02 10:51:16 testland kernel:  [<f9454db3>] intel_set_mode+0x23/0x40 [i915]
Nov 02 10:51:16 testland kernel:  [<f9454fb5>] intel_get_load_detect_pipe+0x1e5/0x480 [i915]
Nov 02 10:51:16 testland kernel:  [<f9456d91>] intel_modeset_setup_hw_state+0xb21/0xc10 [i915]
Nov 02 10:51:16 testland kernel:  [<f943840f>] ? gen4_read32+0x3f/0xd0 [i915]
Nov 02 10:51:16 testland kernel:  [<f9438730>] ? gen4_write64+0xd0/0xd0 [i915]
Nov 02 10:51:16 testland kernel:  [<f945754a>] intel_modeset_init+0x6ca/0x1430 [i915]
Nov 02 10:51:16 testland kernel:  [<f942bf15>] ? i915_enable_pipestat+0xa5/0x130 [i915]
Nov 02 10:51:16 testland kernel:  [<f942c4c2>] ? i915_irq_postinstall+0xf2/0x100 [i915]
Nov 02 10:51:16 testland kernel:  [<f8508bf9>] ? drm_irq_install+0xa9/0x180 [drm]
Nov 02 10:51:16 testland kernel:  [<f94841d9>] i915_driver_load+0xdd9/0xfc0 [i915]
Nov 02 10:51:16 testland kernel:  [<f9481130>] ? i915_switcheroo_set_state+0x90/0x90 [i915]
Nov 02 10:51:16 testland kernel:  [<c123eb82>] ? kobject_uevent_env+0x162/0x570
Nov 02 10:51:16 testland kernel:  [<c123eb82>] ? kobject_uevent_env+0x162/0x570
Nov 02 10:51:16 testland kernel:  [<c123e8e0>] ? kobj_ns_drop+0x60/0x60
Nov 02 10:51:16 testland kernel:  [<c11dd8e3>] ? sysfs_add_file+0x23/0x30
Nov 02 10:51:16 testland kernel:  [<c13116c4>] ? get_device+0x14/0x30
Nov 02 10:51:16 testland kernel:  [<c1316862>] ? klist_class_dev_get+0x12/0x20
Nov 02 10:51:16 testland kernel:  [<c1465e8e>] ? klist_node_init+0x2e/0x50
Nov 02 10:51:16 testland kernel:  [<c1465f24>] ? klist_add_tail+0x34/0x40
Nov 02 10:51:16 testland kernel:  [<c1312bd4>] ? device_add+0x1a4/0x600
Nov 02 10:51:16 testland kernel:  [<f850bcae>] drm_dev_register+0x8e/0xd0 [drm]
Nov 02 10:51:16 testland kernel:  [<f850e0b9>] drm_get_pci_dev+0xa9/0x1c0 [drm]
Nov 02 10:51:16 testland kernel:  [<c11db975>] ? kernfs_new_node+0x35/0x50
Nov 02 10:51:16 testland kernel:  [<f93fa405>] i915_pci_probe+0x35/0x60 [i915]
Nov 02 10:51:16 testland kernel:  [<c127530f>] pci_device_probe+0x6f/0xd0
Nov 02 10:51:16 testland kernel:  [<c11ddd05>] ? sysfs_create_link+0x25/0x50
Nov 02 10:51:16 testland kernel:  [<c1315983>] driver_probe_device+0x93/0x3c0
Nov 02 10:51:16 testland kernel:  [<c11dd9f7>] ? sysfs_create_dir_ns+0x37/0x90
Nov 02 10:51:16 testland kernel:  [<c127523a>] ? pci_match_device+0xca/0x100
Nov 02 10:51:16 testland kernel:  [<c1315d69>] __driver_attach+0x79/0x80
Nov 02 10:51:16 testland kernel:  [<c1315cf0>] ? __device_attach+0x40/0x40
Nov 02 10:51:16 testland kernel:  [<c1313c67>] bus_for_each_dev+0x57/0xa0
Nov 02 10:51:16 testland kernel:  [<c131547e>] driver_attach+0x1e/0x20
Nov 02 10:51:16 testland kernel:  [<c1315cf0>] ? __device_attach+0x40/0x40
Nov 02 10:51:16 testland kernel:  [<c13150c7>] bus_add_driver+0x157/0x240
Nov 02 10:51:16 testland kernel:  [<f94c1000>] ? 0xf94c1000
Nov 02 10:51:16 testland kernel:  [<f94c1000>] ? 0xf94c1000
Nov 02 10:51:16 testland kernel:  [<c131652d>] driver_register+0x5d/0xf0
Nov 02 10:51:16 testland kernel:  [<c1274b53>] __pci_register_driver+0x33/0x40
Nov 02 10:51:16 testland kernel:  [<f850e2bd>] drm_pci_init+0xed/0x110 [drm]
Nov 02 10:51:16 testland kernel:  [<f94c1000>] ? 0xf94c1000
Nov 02 10:51:16 testland kernel:  [<f94c1083>] i915_init+0x83/0x85 [i915]
Nov 02 10:51:16 testland kernel:  [<c1000496>] do_one_initcall+0xc6/0x200
Nov 02 10:51:16 testland kernel:  [<f94c1000>] ? 0xf94c1000
Nov 02 10:51:16 testland kernel:  [<c1151da5>] ? __vunmap+0xa5/0xf0
Nov 02 10:51:16 testland kernel:  [<c1151da5>] ? __vunmap+0xa5/0xf0
Nov 02 10:51:16 testland kernel:  [<c10c9d4b>] load_module+0x1b4b/0x2390
Nov 02 10:51:16 testland kernel:  [<f93ad440>] ? perf_trace_cfg80211_mgmt_tx_status+0x40/0x110 [cfg80211]
Nov 02 10:51:16 testland kernel:  [<f93ad418>] ? perf_trace_cfg80211_mgmt_tx_status+0x18/0x110 [cfg80211]
Nov 02 10:51:16 testland kernel:  [<f93aca90>] ? ftrace_raw_output_cfg80211_report_obss_beacon+0x60/0x60 [cfg80211]
Nov 02 10:51:16 testland kernel:  [<f92ec000>] ? 0xf92ec000
Nov 02 10:51:16 testland kernel:  [<c10ca63f>] SyS_init_module+0xaf/0x120
Nov 02 10:51:16 testland kernel:  [<f92ec000>] ? 0xf92ec000
Nov 02 10:51:16 testland kernel:  [<f93aca90>] ? ftrace_raw_output_cfg80211_report_obss_beacon+0x60/0x60 [cfg80211]
Nov 02 10:51:16 testland kernel:  [<f93ac834>] ? ftrace_raw_output_cfg80211_rx_evt+0x34/0x60 [cfg80211]
Nov 02 10:51:16 testland kernel:  [<c146f197>] sysenter_do_call+0x12/0x12
Nov 02 10:51:16 testland kernel: Code: db 0f 85 58 02 00 00 89 c7 89 f3 89 45 e4 83 e3 02 8b 40 04 0f 44 07 89 5d d8 89 4d dc 89 55 c8 39 c7 8d 58 f8 0f 84 84 02 00 00 <f6> 40 08 01 0f 84 da 0
Nov 02 10:51:17 testland kernel: EIP: [<f850fc7c>] drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm] SS:ESP 0068:f4f9386c
Nov 02 10:51:17 testland kernel: CR2: 0000000000000008
Nov 02 10:51:17 testland kernel: ---[ end trace 9a28cf6a5e1b519e ]---

& 

Nov 02 10:51:23 testland kernel: [drm:i915_irq_handler] *ERROR* pipe A underrun

Comment 1 Apostolos B. 2014-11-02 10:09:14 UTC

Managed to get to ssh without any boot parameters (with new kernel 3.17.2) and i get the same error as above.

Comment 2 Apostolos B. 2014-11-03 14:22:17 UTC

I tested with 3.14.23 and it seems to work fine (was able to boot to a command line). No errors.

Comment 3 Apostolos B. 2014-11-03 19:10:07 UTC

Sorry i rushed with the previous comment.

in 3.14.23 i get:Nov 03 21:03:39 testland kernel: random: nonblocking pool is initialized
Nov 03 21:03:39 testland kernel: ------------[ cut here ]------------
Nov 03 21:03:39 testland kernel: WARNING: CPU: 0 PID: 162 at drivers/gpu/drm/i915/intel_display.c:1133 assert_pipe+0xc1/0xe0 [i915]()
Nov 03 21:03:39 testland kernel: pipe A assertion failure (expected on, current off)
Nov 03 21:03:39 testland kernel: Modules linked in: uvcvideo psmouse btusb bluetooth evdev mac_hid snd_hda_intel(+) i915(+) serio_raw 6lowpan_iphc videobuf2_vmalloc videobuf2_memops videobuf2_core snd_hda_codec rfkill led_class vide
Nov 03 21:03:39 testland kernel: CPU: 0 PID: 162 Comm: systemd-udevd Not tainted 3.14.23-1-lts #1
Nov 03 21:03:39 testland kernel: Hardware name: Hewlett-Packard HP Mini/361A, BIOS 361A0 Ver. F.10 12/25/2008
Nov 03 21:03:39 testland kernel:  00000000 52d7c05a 00000000 f4a118f4 c1442d7a f4a11938 f4a11928 c104cbb2
Nov 03 21:03:39 testland kernel:  f84d7708 f4a11958 000000a2 f84d5fe8 0000046d f847d2a1 0000046d f847d2a1
Nov 03 21:03:39 testland kernel:  00000001 f3d70000 00000041 f4a11944 c104cc0e 00000009 f4a11938 f84d7708
Nov 03 21:03:39 testland kernel: Call Trace:
Nov 03 21:03:39 testland kernel:  [<c1442d7a>] dump_stack+0x41/0x52
Nov 03 21:03:39 testland kernel:  [<c104cbb2>] warn_slowpath_common+0x82/0xa0
Nov 03 21:03:39 testland kernel:  [<f847d2a1>] ? assert_pipe+0xc1/0xe0 [i915]
Nov 03 21:03:39 testland kernel:  [<f847d2a1>] ? assert_pipe+0xc1/0xe0 [i915]
Nov 03 21:03:39 testland kernel:  [<c104cc0e>] warn_slowpath_fmt+0x3e/0x60
Nov 03 21:03:39 testland kernel:  [<f847d2a1>] assert_pipe+0xc1/0xe0 [i915]
Nov 03 21:03:39 testland kernel:  [<f8481f7c>] i9xx_crtc_enable+0xcc/0x3b0 [i915]
Nov 03 21:03:39 testland kernel:  [<f8484656>] __intel_set_mode+0x726/0x1430 [i915]
Nov 03 21:03:39 testland kernel:  [<f813e170>] ? drm_framebuffer_init+0x80/0x90 [drm]
Nov 03 21:03:39 testland kernel:  [<f84878f3>] intel_set_mode+0x23/0x40 [i915]
Nov 03 21:03:39 testland kernel:  [<f8488d54>] intel_get_load_detect_pipe+0x254/0x3d0 [i915]
Nov 03 21:03:39 testland kernel:  [<f848a98d>] intel_modeset_setup_hw_state+0xafd/0xb40 [i915]
Nov 03 21:03:39 testland kernel:  [<f848aa05>] intel_modeset_gem_init+0x35/0x40 [i915]
Nov 03 21:03:39 testland kernel:  [<f8453ea3>] i915_driver_load+0xc03/0xd90 [i915]
Nov 03 21:03:39 testland kernel:  [<f8450f50>] ? i915_switcheroo_set_state+0xa0/0xa0 [i915]
Nov 03 21:03:39 testland kernel:  [<f8138baa>] drm_dev_register+0x6a/0x140 [drm]
Nov 03 21:03:39 testland kernel:  [<f813ae41>] drm_get_pci_dev+0xc1/0x1e0 [drm]
Nov 03 21:03:39 testland kernel:  [<c11ce852>] ? kernfs_create_link+0x62/0xa0
Nov 03 21:03:39 testland kernel:  [<f84505f5>] i915_pci_probe+0x35/0x60 [i915]
Nov 03 21:03:39 testland kernel:  [<c125ce7f>] pci_device_probe+0x6f/0xd0
Nov 03 21:03:39 testland kernel:  [<c11cb615>] ? sysfs_create_link+0x25/0x50
Nov 03 21:03:39 testland kernel:  [<c12face3>] driver_probe_device+0x93/0x3c0
Nov 03 21:03:39 testland kernel:  [<c11cb337>] ? sysfs_create_dir_ns+0x37/0x90
Nov 03 21:03:39 testland kernel:  [<c125c6fe>] ? pci_match_device+0x9e/0xb0
Nov 03 21:03:39 testland kernel:  [<c12fb0c9>] __driver_attach+0x79/0x80
Nov 03 21:03:39 testland kernel:  [<c12fb050>] ? __device_attach+0x40/0x40
Nov 03 21:03:39 testland kernel:  [<c12f8f57>] bus_for_each_dev+0x57/0xa0
Nov 03 21:03:39 testland kernel:  [<c12fa78e>] driver_attach+0x1e/0x20
Nov 03 21:03:39 testland kernel:  [<c12fb050>] ? __device_attach+0x40/0x40
Nov 03 21:03:39 testland kernel:  [<c12fa3a7>] bus_add_driver+0x157/0x240
Nov 03 21:03:39 testland kernel:  [<c12fb6dd>] driver_register+0x5d/0xf0
Nov 03 21:03:39 testland kernel:  [<c125c803>] __pci_register_driver+0x33/0x40
Nov 03 21:03:39 testland kernel:  [<f813b06d>] drm_pci_init+0x10d/0x120 [drm]
Nov 03 21:03:39 testland kernel:  [<f84f7000>] ? 0xf84f6fff
Nov 03 21:03:39 testland kernel:  [<f84f7062>] i915_init+0x62/0x64 [i915]
Nov 03 21:03:39 testland kernel:  [<c100049a>] do_one_initcall+0xca/0x1a0
Nov 03 21:03:39 testland kernel:  [<c10d3221>] ? tracepoint_update_probe_range+0x41/0x130
Nov 03 21:03:39 testland kernel:  [<c10d3c57>] ? tracepoint_module_notify+0x177/0x1a0
Nov 03 21:03:39 testland kernel:  [<c144af09>] ? notifier_call_chain+0x49/0x70
Nov 03 21:03:39 testland kernel:  [<c1070252>] ? __blocking_notifier_call_chain+0x42/0x60
Nov 03 21:03:39 testland kernel:  [<c10c1126>] load_module+0x1b86/0x2210
Nov 03 21:03:39 testland kernel:  [<c113f3bc>] ? map_vm_area+0x2c/0x40
Nov 03 21:03:39 testland kernel:  [<c10c185f>] SyS_init_module+0xaf/0x120
Nov 03 21:03:39 testland kernel:  [<c144f411>] sysenter_do_call+0x12/0x12
Nov 03 21:03:39 testland kernel: ---[ end trace 8297882c8bd72624 ]---

Comment 4 Apostolos B. 2014-11-26 15:32:29 UTC

Still present in 3.17.4.

Comment 5 Apostolos B. 2015-01-15 13:27:58 UTC

This is still present in 3.18.2 

Jan 15 15:18:38 testland kernel: BUG: unable to handle kernel NULL pointer dereference at 00000008
Jan 15 15:18:38 testland kernel: IP: [<f90d7d6c>] drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm]
Jan 15 15:18:39 testland kernel: *pde = 00000000 
Jan 15 15:18:39 testland kernel: Oops: 0000 [#1] PREEMPT SMP 
Jan 15 15:18:39 testland kernel: Modules linked in: bluetooth i2c_i801(+) evdev i915(+) ssb sky2 uvcvideo mac_hid serio_raw videobuf2_vmalloc drm_kms_helper videobuf2_memops videobuf2_core v4l2_common rfkill mmc_core drm videodev media i2c
Jan 15 15:18:39 testland kernel: CPU: 1 PID: 170 Comm: systemd-udevd Not tainted 3.18.2-2-ARCH #1
Jan 15 15:18:39 testland kernel: Hardware name: Hewlett-Packard HP Mini/361A, BIOS 361A0 Ver. F.10 12/25/2008
Jan 15 15:18:40 testland kernel: task: f5144b00 ti: f4f72000 task.ti: f4f72000
Jan 15 15:18:40 testland kernel: EIP: 0060:[<f90d7d6c>] EFLAGS: 00010286 CPU: 1
Jan 15 15:18:40 testland kernel: EIP is at drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm]
Jan 15 15:18:40 testland kernel: EAX: 00000000 EBX: fffffff8 ECX: 0012c000 EDX: f4f8b6c0
Jan 15 15:18:40 testland kernel: ESI: 00000000 EDI: f370c1b4 EBP: f4f738a8 ESP: f4f7386c
Jan 15 15:18:40 testland kernel:  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Jan 15 15:18:40 testland kernel: CR0: 80050033 CR2: 00000008 CR3: 34f61000 CR4: 000007d0
Jan 15 15:18:40 testland kernel: Stack:
Jan 15 15:18:40 testland kernel:  00000001 f4f8b6c0 00000293 00000060 f924b88c 00000000 0012c000 00000000
Jan 15 15:18:40 testland kernel:  f370c1b4 f4e94d00 f4f738a8 47643ad2 10000000 00010000 f4e94d00 f4f73910
Jan 15 15:18:40 testland kernel:  f9251181 00010000 00000000 00000000 10000000 00000000 00000000 f926dc9f
Jan 15 15:18:40 testland kernel: Call Trace:
Jan 15 15:18:40 testland kernel:  [<f924b88c>] ? i915_gem_obj_lookup_or_create_vma+0x3c/0x1b0 [i915]
Jan 15 15:18:40 testland kernel:  [<f9251181>] i915_gem_object_pin+0x431/0x6d0 [i915]
Jan 15 15:18:40 testland kernel:  [<f926dc9f>] ? gen4_read32+0x3f/0xd0 [i915]
Jan 15 15:18:40 testland kernel:  [<f925285c>] i915_gem_object_pin_to_display_plane+0x11c/0x1f0 [i915]
Jan 15 15:18:40 testland kernel:  [<f9280179>] intel_pin_and_fence_fb_obj+0xc9/0x150 [i915]
Jan 15 15:18:40 testland kernel:  [<f9282f7f>] __intel_set_mode+0x55f/0x14e0 [i915]
Jan 15 15:18:40 testland kernel:  [<f90d8489>] ? drm_mode_object_get_reg+0x69/0x80 [drm]
Jan 15 15:18:40 testland kernel:  [<f928b3a3>] intel_set_mode+0x23/0x40 [i915]
Jan 15 15:18:40 testland kernel:  [<f928b5a5>] intel_get_load_detect_pipe+0x1e5/0x480 [i915]
Jan 15 15:18:40 testland kernel:  [<f928d371>] intel_modeset_setup_hw_state+0xb21/0xc10 [i915]
Jan 15 15:18:40 testland kernel:  [<f926dfc0>] ? gen4_write64+0xd0/0xd0 [i915]
Jan 15 15:18:40 testland kernel:  [<f90e7801>] ? __drm_modeset_lock_all+0xa1/0xf0 [drm]
Jan 15 15:18:40 testland kernel:  [<f928dbb2>] intel_modeset_init+0x752/0x1530 [i915]
Jan 15 15:18:40 testland kernel:  [<f925e5b5>] ? i915_enable_pipestat+0xa5/0x130 [i915]
Jan 15 15:18:40 testland kernel:  [<f925eb62>] ? i915_irq_postinstall+0xf2/0x100 [i915]
Jan 15 15:18:40 testland kernel:  [<f90d0cd9>] ? drm_irq_install+0xa9/0x180 [drm]
Jan 15 15:18:40 testland kernel:  [<f92bb00d>] i915_driver_load+0xddd/0xf80 [i915]
Jan 15 15:18:40 testland kernel:  [<f92b7f60>] ? i915_switcheroo_set_state+0x90/0x90 [i915]
Jan 15 15:18:40 testland kernel:  [<c1242db2>] ? kobject_uevent_env+0x162/0x560
Jan 15 15:18:40 testland kernel:  [<c1242db2>] ? kobject_uevent_env+0x162/0x560
Jan 15 15:18:40 testland kernel:  [<c1242b10>] ? kobj_ns_drop+0x60/0x60
Jan 15 15:18:40 testland kernel:  [<c11e0b83>] ? sysfs_add_file+0x23/0x30
Jan 15 15:18:40 testland kernel:  [<c1316b04>] ? get_device+0x14/0x30
Jan 15 15:18:40 testland kernel:  [<c131bd62>] ? klist_class_dev_get+0x12/0x20
Jan 15 15:18:40 testland kernel:  [<c1470f9e>] ? klist_node_init+0x2e/0x50
Jan 15 15:18:41 testland kernel:  [<c1471034>] ? klist_add_tail+0x34/0x40
Jan 15 15:18:41 testland kernel:  [<c13180d4>] ? device_add+0x1a4/0x600
Jan 15 15:18:41 testland kernel:  [<f90d3bbe>] drm_dev_register+0x8e/0xd0 [drm]
Jan 15 15:18:41 testland kernel:  [<f90d6219>] drm_get_pci_dev+0xa9/0x1c0 [drm]
Jan 15 15:18:41 testland kernel:  [<c11dec15>] ? kernfs_new_node+0x35/0x50
Jan 15 15:18:41 testland kernel:  [<f922b2f5>] i915_pci_probe+0x35/0x60 [i915]
Jan 15 15:18:41 testland kernel:  [<c127a2df>] pci_device_probe+0x6f/0xd0
Jan 15 15:18:41 testland kernel:  [<c11e0fa5>] ? sysfs_create_link+0x25/0x50
Jan 15 15:18:41 testland kernel:  [<c131ae83>] driver_probe_device+0x93/0x3c0
Jan 15 15:18:41 testland kernel:  [<c11e0c97>] ? sysfs_create_dir_ns+0x37/0x90
Jan 15 15:18:41 testland kernel:  [<c127a20a>] ? pci_match_device+0xca/0x100
Jan 15 15:18:41 testland kernel:  [<c131b269>] __driver_attach+0x79/0x80
Jan 15 15:18:41 testland kernel:  [<c131b1f0>] ? __device_attach+0x40/0x40
Jan 15 15:18:41 testland kernel:  [<c1319167>] bus_for_each_dev+0x57/0xa0
Jan 15 15:18:41 testland kernel:  [<c131a97e>] driver_attach+0x1e/0x20
Jan 15 15:18:41 testland kernel:  [<c131b1f0>] ? __device_attach+0x40/0x40
Jan 15 15:18:41 testland kernel:  [<c131a5c7>] bus_add_driver+0x157/0x240
Jan 15 15:18:41 testland kernel:  [<f92fa000>] ? 0xf92fa000
Jan 15 15:18:41 testland kernel:  [<f92fa000>] ? 0xf92fa000
Jan 15 15:18:41 testland kernel:  [<c131ba2d>] driver_register+0x5d/0xf0
Jan 15 15:18:41 testland kernel:  [<c1279b23>] __pci_register_driver+0x33/0x40
Jan 15 15:18:41 testland kernel:  [<f90d640d>] drm_pci_init+0xdd/0x100 [drm]
Jan 15 15:18:41 testland kernel:  [<f92fa000>] ? 0xf92fa000
Jan 15 15:18:41 testland kernel:  [<f92fa083>] i915_init+0x83/0x85 [i915]
Jan 15 15:18:41 testland kernel:  [<c1000496>] do_one_initcall+0xc6/0x200
Jan 15 15:18:41 testland kernel:  [<f92fa000>] ? 0xf92fa000
Jan 15 15:18:41 testland kernel:  [<c1153b65>] ? __vunmap+0xa5/0xf0
Jan 15 15:18:41 testland kernel:  [<c1153b65>] ? __vunmap+0xa5/0xf0
Jan 15 15:18:41 testland kernel:  [<c10cb4da>] load_module+0x1aea/0x24a0
Jan 15 15:18:41 testland kernel:  [<f9113000>] ? 0xf9113000
Jan 15 15:18:41 testland kernel:  [<c10cbf3f>] SyS_init_module+0xaf/0x120
Jan 15 15:18:41 testland kernel:  [<f9113000>] ? 0xf9113000
Jan 15 15:18:41 testland kernel:  [<c1479cd7>] sysenter_do_call+0x12/0x12
Jan 15 15:18:41 testland kernel: Code: db 0f 85 58 02 00 00 89 c7 89 f3 89 45 e4 83 e3 02 8b 40 04 0f 44 07 89 5d d8 89 4d dc 89 55 c8 39 c7 8d 58 f8 0f 84 84 02 00 00 <f6> 40 08 01 0f 84 da 00 00 00 83 e6 01 c7 45 cc 00 00 00 00 c7
Jan 15 15:18:41 testland kernel: EIP: [<f90d7d6c>] drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm] SS:ESP 0068:f4f7386c
lines 726-798/798 (END)

Comment 6 Apostolos B. 2015-01-28 14:34:19 UTC

3.18.4 had quite a few i915 related commits/fixes but this damn thing is still present.

Comment 7 Apostolos B. 2015-02-02 18:48:57 UTC

3.18.5 crashes on boot.

Feb 02 20:43:34 testland kernel: BUG: unable to handle kernel NULL pointer dereference at 00000008
Feb 02 20:43:35 testland kernel: IP: [<f9123d7c>] drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm]
Feb 02 20:43:35 testland kernel: *pde = 00000000 
Feb 02 20:43:35 testland kernel: Oops: 0000 [#1] PREEMPT SMP 
Feb 02 20:43:35 testland kernel: Modules linked in: coretemp cfg80211 hwmon iTCO_wdt iTCO_vendor_support btusb i915(+) hp_wmi(+) sparse_keymap bluetooth pcspkr evdev uvcvideo drm_kms_helper psmouse mac_hid serio_raw i2c_i801 videobuf2_
Feb 02 20:43:35 testland kernel: CPU: 0 PID: 173 Comm: systemd-udevd Not tainted 3.18.5-1-ARCH #1
Feb 02 20:43:36 testland kernel: Hardware name: Hewlett-Packard HP Mini/361A, BIOS 361A0 Ver. F.10 12/25/2008
Feb 02 20:43:36 testland kernel: task: f524c4c0 ti: f441c000 task.ti: f441c000
Feb 02 20:43:36 testland kernel: EIP: 0060:[<f9123d7c>] EFLAGS: 00010286 CPU: 0

Feb 02 20:43:36 testland kernel: EIP is at drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm]
Feb 02 20:43:36 testland kernel: EAX: 00000000 EBX: fffffff8 ECX: 0012c000 EDX: f2ec8300
Feb 02 20:43:36 testland kernel: ESI: 00000000 EDI: f289c1b4 EBP: f441d8a8 ESP: f441d86c
Feb 02 20:43:36 testland kernel:  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Feb 02 20:43:36 testland kernel: CR0: 80050033 CR2: 00000008 CR3: 34b64000 CR4: 000007d0
Feb 02 20:43:36 testland kernel: Stack:
Feb 02 20:43:36 testland kernel:  f441dfec f2ec8300 00000293 00000060 f9342a0c 00000000 0012c000 00000000
Feb 02 20:43:36 testland kernel:  f289c1b4 f52fd400 f441d8a8 d5df9221 10000000 00010000 f52fd400 f441d910
Feb 02 20:43:36 testland kernel:  f9348301 00010000 00000000 00000000 10000000 00000000 00000000 f9364e3f
Feb 02 20:43:36 testland kernel: Call Trace:
Feb 02 20:43:37 testland kernel:  [<f9342a0c>] ? i915_gem_obj_lookup_or_create_vma+0x3c/0x1b0 [i915]
Feb 02 20:43:37 testland kernel:  [<f9348301>] i915_gem_object_pin+0x431/0x6d0 [i915]
Feb 02 20:43:37 testland kernel:  [<f9364e3f>] ? gen4_read32+0x3f/0xd0 [i915]
Feb 02 20:43:37 testland kernel:  [<f93499dc>] i915_gem_object_pin_to_display_plane+0x11c/0x1f0 [i915]
Feb 02 20:43:37 testland kernel:  [<f9377319>] intel_pin_and_fence_fb_obj+0xc9/0x150 [i915]
Feb 02 20:43:37 testland kernel:  [<f937a11f>] __intel_set_mode+0x55f/0x14e0 [i915]
Feb 02 20:43:37 testland kernel:  [<f9124499>] ? drm_mode_object_get_reg+0x69/0x80 [drm]
Feb 02 20:43:37 testland kernel:  [<f9014580>] ? video_usercopy+0x100/0x5d0 [videodev]
Feb 02 20:43:37 testland kernel:  [<f9382543>] intel_set_mode+0x23/0x40 [i915]
Feb 02 20:43:37 testland kernel:  [<f9382745>] intel_get_load_detect_pipe+0x1e5/0x480 [i915]
Feb 02 20:43:37 testland kernel:  [<f9384511>] intel_modeset_setup_hw_state+0xb21/0xc10 [i915]
Feb 02 20:43:37 testland kernel:  [<f9365160>] ? gen4_write64+0xd0/0xd0 [i915]
Feb 02 20:43:37 testland kernel:  [<f9133801>] ? __drm_modeset_lock_all+0x91/0xf0 [drm]
Feb 02 20:43:37 testland kernel:  [<f9384d1a>] intel_modeset_init+0x71a/0x14f0 [i915]
Feb 02 20:43:37 testland kernel:  [<f9355745>] ? i915_enable_pipestat+0xa5/0x130 [i915]
Feb 02 20:43:37 testland kernel:  [<f9355cf2>] ? i915_irq_postinstall+0xf2/0x100 [i915]
Feb 02 20:43:37 testland kernel:  [<f911ccd9>] ? drm_irq_install+0xa9/0x180 [drm]
Feb 02 20:43:37 testland kernel:  [<f93b216d>] i915_driver_load+0xddd/0xf80 [i915]
Feb 02 20:43:37 testland kernel:  [<f93af0c0>] ? i915_switcheroo_set_state+0x90/0x90 [i915]
Feb 02 20:43:37 testland kernel:  [<c1242ed2>] ? kobject_uevent_env+0x162/0x560
Feb 02 20:43:37 testland kernel:  [<c1242ed2>] ? kobject_uevent_env+0x162/0x560
Feb 02 20:43:37 testland kernel:  [<c1242c30>] ? kobj_ns_drop+0x60/0x60
Feb 02 20:43:37 testland kernel:  [<c11e0cd3>] ? sysfs_add_file+0x23/0x30
Feb 02 20:43:37 testland kernel:  [<c1317044>] ? get_device+0x14/0x30
Feb 02 20:43:37 testland kernel:  [<c131c2b2>] ? klist_class_dev_get+0x12/0x20
Feb 02 20:43:37 testland kernel:  [<c14715de>] ? klist_node_init+0x2e/0x50
Feb 02 20:43:37 testland kernel:  [<c1471674>] ? klist_add_tail+0x34/0x40
Feb 02 20:43:37 testland kernel:  [<c1318614>] ? device_add+0x1a4/0x600
Feb 02 20:43:37 testland kernel:  [<f911fbce>] drm_dev_register+0x8e/0xd0 [drm]
Feb 02 20:43:37 testland kernel:  [<f9122229>] drm_get_pci_dev+0xa9/0x1c0 [drm]
Feb 02 20:43:37 testland kernel:  [<c11ded65>] ? kernfs_new_node+0x35/0x50
Feb 02 20:43:37 testland kernel:  [<f93222f5>] i915_pci_probe+0x35/0x60 [i915]
Feb 02 20:43:37 testland kernel:  [<c127a6af>] pci_device_probe+0x6f/0xd0
Feb 02 20:43:37 testland kernel:  [<c11e10f5>] ? sysfs_create_link+0x25/0x50
Feb 02 20:43:37 testland kernel:  [<c131b3d3>] driver_probe_device+0x93/0x3c0
Feb 02 20:43:37 testland kernel:  [<c11e0de7>] ? sysfs_create_dir_ns+0x37/0x90
Feb 02 20:43:37 testland kernel:  [<c127a5da>] ? pci_match_device+0xca/0x100
Feb 02 20:43:37 testland kernel:  [<c131b7b9>] __driver_attach+0x79/0x80
Feb 02 20:43:37 testland kernel:  [<c131b740>] ? __device_attach+0x40/0x40
Feb 02 20:43:37 testland kernel:  [<c13196a7>] bus_for_each_dev+0x57/0xa0
Feb 02 20:43:37 testland kernel:  [<c131aece>] driver_attach+0x1e/0x20
Feb 02 20:43:37 testland kernel:  [<c131b740>] ? __device_attach+0x40/0x40
Feb 02 20:43:37 testland kernel:  [<c131ab17>] bus_add_driver+0x157/0x240
Feb 02 20:43:37 testland kernel:  [<f93f1000>] ? 0xf93f1000
Feb 02 20:43:37 testland kernel:  [<f93f1000>] ? 0xf93f1000
Feb 02 20:43:37 testland kernel:  [<c131bf7d>] driver_register+0x5d/0xf0
Feb 02 20:43:37 testland kernel:  [<c1279ef3>] __pci_register_driver+0x33/0x40
Feb 02 20:43:37 testland kernel:  [<f912241d>] drm_pci_init+0xdd/0x100 [drm]
Feb 02 20:43:37 testland kernel:  [<f93f1000>] ? 0xf93f1000
Feb 02 20:43:37 testland kernel:  [<f93f1083>] i915_init+0x83/0x85 [i915]
Feb 02 20:43:37 testland kernel:  [<c1000496>] do_one_initcall+0xc6/0x200
Feb 02 20:43:37 testland kernel:  [<f93f1000>] ? 0xf93f1000
Feb 02 20:43:37 testland kernel:  [<c1167135>] ? kfree+0x165/0x170
Feb 02 20:43:37 testland kernel:  [<c1153d35>] ? __vunmap+0xa5/0xf0
Feb 02 20:43:37 testland kernel:  [<c1153d35>] ? __vunmap+0xa5/0xf0
Feb 02 20:43:37 testland kernel:  [<c1153d35>] ? __vunmap+0xa5/0xf0
Feb 02 20:43:37 testland kernel:  [<c10cb62a>] load_module+0x1aea/0x24a0
Feb 02 20:43:37 testland kernel:  [<c10cc08f>] SyS_init_module+0xaf/0x120
Feb 02 20:43:37 testland kernel:  [<f92d1e1c>] ? perf_trace_rdev_join_mesh+0x10c/0x1e0 [cfg80211]
Feb 02 20:43:37 testland kernel:  [<c147a317>] sysenter_do_call+0x12/0x12
Feb 02 20:43:37 testland kernel: Code: db 0f 85 58 02 00 00 89 c7 89 f3 89 45 e4 83 e3 02 8b 40 04 0f 44 07 89 5d d8 89 4d dc 89 55 c8 39 c7 8d 58 f8 0f 84 84 02 00 00 <f6> 40 08 01 0f 84 da 00 00 00 83 e6 01 c7 45 cc 00 00 00 00 c7
Feb 02 20:43:37 testland kernel: EIP: [<f9123d7c>] drm_mm_insert_node_in_range_generic+0x4c/0x2e0 [drm] SS:ESP 0068:f441d86c

Comment 8 Jani Nikula 2015-02-04 13:55:06 UTC

We seem to have neglected this bug, apologies.

Please attach dmesg with drm.debug=14, from boot, to get some details of your machine.

Also, please try drm-intel-nightly branch from http://cgit.freedesktop.org/drm-intel

Chris, any ideas off the top of your head?

Comment 9 Chris Wilson 2015-02-04 14:16:12 UTC

It's quite simple really. We try to use GEM inside intel_modeset_init() prior to GEM itself being initialised, which stems from:

commit 7fad798e16fecddd41c6a91728a09f0b9507e40c
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Wed Jul 4 17:51:47 2012 +0200

    drm/i915: ensure the force pipe A quirk is actually followed
    
    Many BIOSen forget to turn on the pipe A after resume (because they
    actually don't turn on anything), so we have to do that ourselves when
    sanitizing the hw state.
    
    I've discovered this due to the recent addition of a pipe WARN that
    takes the force quirk into account.
    
    v2: Actually try to enable the pipe with a proper configuration instead
    of simpyl switching it on with whatever random state the bios left it
    in after resume.
    
    v3: Fixup rebase conflict - the load_detect functions have lost their
    encoder argument.
    
    Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
    Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>


That netbook should not be using the PIPE A quirk!, but 830/845 are still prone to this issue.

Comment 10 Jani Nikula 2015-02-04 14:47:36 UTC

Apostolos, please provide 'lspci -vmmnn -s 00:02' output.

Comment 11 Apostolos B. 2015-02-04 15:05:24 UTC

lspci -vmmnn -s 00:02
Slot:   00:02.0
Class:  VGA compatible controller [0300]
Vendor: Intel Corporation [8086]
Device: Mobile 945GSE Express Integrated Graphics Controller [27ae]
SVendor:        Hewlett-Packard Company [103c]
SDevice:        Device [361a]
Rev:    03
 
Slot:   00:02.1
Class:  Display controller [0380]
Vendor: Intel Corporation [8086]
Device: Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller [27a6]
SVendor:        Hewlett-Packard Company [103c]
SDevice:        Device [361a]
Rev:    03


Didn't chris fixed this with the commit on the above comment?

Comment 12 Chris Wilson 2015-02-04 15:09:04 UTC

No, that's the commit that introduced the use of GEM withing intel_modeset_init(), i.e. the commit that I think introduced the bug.

Comment 13 Chris Wilson 2015-02-04 15:12:30 UTC

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d8324c69fa86..d2d36a014518 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -12468,9 +12468,6 @@ static const struct intel_dmi_quirk intel_dmi_quirks[] = {
 };
 
 static struct intel_quirk intel_quirks[] = {
-       /* HP Mini needs pipe A force quirk (LP: #322104) */
-       { 0x27ae, 0x103c, 0x361a, quirk_pipea_force },
-
        /* Toshiba Protege R-205, S-209 needs pipe A force quirk */
        { 0x2592, 0x1179, 0x0001, quirk_pipea_force },

Comment 14 Apostolos B. 2015-02-04 15:15:01 UTC

Ok. Do you need more info i can provide to fix this??

Comment 15 Jani Nikula 2015-02-05 10:33:15 UTC

Chris, that HP mini quirk has been there forever. How do we know it's safe to toss out?

Apostolos, please try Chris' patch from comment #13. Testing is always welcome.

Comment 16 Chris Wilson 2015-02-05 10:41:07 UTC

The origin lies in https://bugs.freedesktop.org/show_bug.cgi?id=21960, so has nothing to do with PIPEA it seems.

Comment 17 Apostolos B. 2015-02-08 21:24:53 UTC

Jani this machine is so slow that testing kernels is out of the question. Sorry. :(

I'll have to wait for Chris's patch to hit a point release or 3.19.

Comment 18 Jani Nikula 2015-02-09 14:05:25 UTC

(In reply to Apostolos B. from comment #17)
> Jani this machine is so slow that testing kernels is out of the question.
> Sorry. :(
> 
> I'll have to wait for Chris's patch to hit a point release or 3.19.

Catch-22. It won't hit a stable release unless it is known and tested to fix a real world issue...

Comment 19 Jani Nikula 2015-02-09 14:06:38 UTC

You know you don't have to build the kernel on *that* machine, right?

Comment 20 Apostolos B. 2015-02-09 14:38:59 UTC

Yes but it is a bit of a hassle to set up something for building 32bit packages at the moment. Sorry. The only thing i can do at the moment is provide info as needed.

Comment 21 Jani Nikula 2015-03-03 14:05:20 UTC

You only need to configure the kernel build for 32-bit, nothing more.

Comment 22 Apostolos B. 2015-03-03 14:26:03 UTC

No time for stuff like that at the moment. I'd appreciate it if it was fixed. It is a real world issue after all. I don't just type randomly and generate kernel crashes.

Comment 23 Alan 2015-03-03 14:29:33 UTC

If you "just want it fixed" and don't have time to help you need to talk to your support vendor, or distribution supplier.

Comment 24 Apostolos B. 2015-03-03 14:35:34 UTC

Sorry didn't mean this to come out the wrong way. Its just that i can't build/test/patch kernels atm. If there is something more that can be done to help you pinpoint the bug without the build/test/patch thing let me know.

Comment 25 Apostolos B. 2015-03-03 21:15:24 UTC

@Alan 

Just wanted to reply to your support vendor/distribution supplier comment. Distributions will normally tell you to report it upstream. As for support from the vendor there aren't many companies that support linux. If you try to get them to help the first answer you will get is that "We don't support linux /we only support the software it came with (read windows)". I had it happen even with intel on one of its motherboards when i faced firmware problems (still broken more or less). So like it or not the kernel bugzilla is the only support someone can get on linux. Also this means that you will also get bug reports from less technically skilled people or people that can't help much. Of course you have every right to ignore those bug reports.
Regards.

Comment 26 Apostolos B. 2015-04-19 13:54:41 UTC

Got some time and tested Chris's patch. 


On a 4.0 kernel it seems to work. Didn't test extensively but it boots to a command line and the journal doesn't report anything weird.


Please apply. :)

Comment 27 Apostolos B. 2015-05-06 23:08:23 UTC

Any update on this??

I've been using it for a few weeks and it works just fine.

Comment 28 Jani Nikula 2015-05-07 12:29:18 UTC

Fixed by

commit 736a69ca8c99a595a523d2fece66491b89168da6
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu May 7 11:16:26 2015 +0100

    drm/i915: Drop PIPE-A quirk for 945GSE HP Mini

in drm-intel-fixes. Thanks for the report and testing.

Comment 29 darethehair 2015-05-20 22:44:17 UTC

[forgive if this question is out of place or inappropriate]

Just a normal Linux user here with the misfortune of owning one of the HP Mini netbooks affected by this Intel video bug -- which messed up my Linux Mint LMDE upgrade very nicely :)

Is there a simple way for me to plop in patched binary file to fix this, or are we waiting for a new kernel with this fix in it?  If so, when/where would I look for it?

Thanks

Comment 30 Apostolos B. 2015-05-21 06:38:07 UTC

It will be in 4.1. What you can do now is either build your own kernel with the patch applied, bug the maintainer of your distro to include the patch or use an lts kernel if your distro packs it.

Comment 31 darethehair 2015-05-21 18:04:10 UTC

(In reply to darethehair from comment #29)
> [forgive if this question is out of place or inappropriate]
> 
> Just a normal Linux user here with the misfortune of owning one of the HP
> Mini netbooks affected by this Intel video bug -- which messed up my Linux
> Mint LMDE upgrade very nicely :)
> 
> Is there a simple way for me to plop in patched binary file to fix this, or
> are we waiting for a new kernel with this fix in it?  If so, when/where
> would I look for it?
> 
> Thanks

Followup: According to change logs, it looks like this fix made it into v4.1-rc3-unstable.  I opted to go for v4.1-rc4-unstable (I had nothing to lose) by following these instructions:

http://media-opensource.blogspot.ca/2015/05/how-to-upgrade-to-linux-kernel-41-rc4.html

Yes, it definitely fixed the 'black screen' issue.  Thanks to all who helped get to this point.