Opened 7 years ago

Closed 8 months ago

#8952 closed bug (duplicate)

Intel HD2000 HDMI problem

Reported by: dsjonny Owned by: nobody
Priority: normal Milestone: R1/beta2
Component: Drivers/Graphics/intel_extreme Version: R1/Development
Keywords: Cc:
Blocked By: #14301 Blocking: #9949
Has a Patch: no Platform: All

Description

I have integrated Intel HD2000 VGA in my Intel Core i3 2100T CPU. I use HDMI output, and Haiku can use HDMI only in VESA mode. The max resolution is: 1360x768 (cannot select full HD), and at the Screen preferences the Samsung 26" LED display identified as Samsung 7.2".

I think only the HDMI output is in VESA mode, because if I check the loaded driver in Terminal, there is the intel_extreme too:

/dev/graphics:
total 0
0 crw-r--r-- 1 user root        0 2012-09-08 00:44 intel_extreme_000200
0 crw-r--r-- 1 user root        0 2012-09-08 00:44 vesa

Device info:

  • device/flags: 1
  • device/id: 0x102
  • device/interface: 0x0
  • device/subtype: 0x0
  • device/type: 0x3
  • device/vendor: 0x8086
  • device name: 2nd Generation Core Processor Family Integrated Graphics Controller

Attachments (7)

SYSLOG01.TXT (93.4 KB ) - added by dsjonny 7 years ago.
screen.png (24.5 KB ) - added by dsjonny 7 years ago.
SAM_0684.JPG (538.4 KB ) - added by dsjonny 6 years ago.
SYSLOG00.TXT (61.0 KB ) - added by dsjonny 6 years ago.
SYSLOG02.TXT (108.4 KB ) - added by dsjonny 6 years ago.
hrev46033syslog.txt (100.7 KB ) - added by dsjonny 6 years ago.
hrev46033syslog-2.txt (123.4 KB ) - added by dsjonny 6 years ago.

Download all attachments as: .zip

Change History (35)

comment:1 by anevilyak, 7 years ago

For reference, vesa is always going to be listed there. However, if there's an accelerated driver present, then it will be used regardless, it's not possible for one port on the graphics card to be driven by VESA and the rest by the real driver.

comment:2 by mmadia, 7 years ago

Version: R1/alpha4R1/Development

by dsjonny, 7 years ago

Attachment: SYSLOG01.TXT added

comment:3 by dsjonny, 7 years ago

I have attached a syslog to the ticket. Maybe this will help. There are many entries fot the VGA.

by dsjonny, 7 years ago

Attachment: screen.png added

comment:4 by dsjonny, 7 years ago

I have checked the loaded drivers:

~> listimage | grep drivers/
...
 1061   /boot/system/add-ons/kernel/drivers/dev/graphics/intel_extreme 0x8111d000 0x81121000    0          0
 1112            /boot/system/add-ons/kernel/drivers/dev/graphics/vesa 0x81131000 0x81135000    0          0
...

The VESA and the Intel driver is still loaded, but the VESA is used.

comment:5 by dsjonny, 7 years ago

Found these lines in the syslog related to vesa and intel_extreme:

KERN: intel_extreme: init_hardware()
KERN: intel_extreme: init_driver()
KERN: intel_extreme: (0) SandyBridge Desktop GT1, revision = 0x9
KERN: intel_extreme: publish_devices()
KERN: intel_extreme: find_device()
KERN: loaded driver /boot/system/add-ons/kernel/drivers/dev/graphics/intel_extreme
KERN: vesa: init_hardware()
KERN: vesa: init_driver()
KERN: vesa: publish_devices()
KERN: vesa: find_device()
KERN: loaded driver /boot/system/add-ons/kernel/drivers/dev/graphics/vesa
KERN: intel_extreme: open(name = graphics/intel_extreme_000200)
KERN: vesa: vesa_init() completed successfully!
KERN: vesa: acc: vesa.accelerant
KERN: vesa: vesa_set_dpms_mode(): BIOS returned 0x014f

And these are the loaded image:

~/Desktop> listimage | grep driver
 1210   /boot/system/add-ons/kernel/drivers/dev/graphics/intel_extreme 0x827b3000 0x827b7000    0          0
 1218            /boot/system/add-ons/kernel/drivers/dev/graphics/vesa 0x827d6000 0x827da000    0          0
~/Desktop> listimage | grep acce
 1220                 /boot/system/add-ons/accelerants/vesa.accelerant 0x22a5000 0x22a9000    0          0
TEAM 45107 (/bin/grep --color=auto acce):
~/Desktop> 

comment:6 by dsjonny, 7 years ago

Any idea? Using 45341 nightly there is no change. Still only VESA driver using. Is there any other info what can I provide here to help to sovle the problem?

comment:7 by dsjonny, 7 years ago

Still exists in 45385 nightly.

comment:8 by korli, 6 years ago

Resolution: fixed
Status: newclosed

This might be fixed in hrev45822 or newer. Please reopen if needed and provides a current syslog.

by dsjonny, 6 years ago

Attachment: SAM_0684.JPG added

comment:9 by dsjonny, 6 years ago

Resolution: fixed
Status: closedreopened

Sorry, but using 45824 nightly, I can boot Haiku only in safe video mode. I got KDL on normal boot. Please check the attached SAM_0684.jpg file.

comment:10 by korli, 6 years ago

At least it crashs somewhere :) Please test again with hrev45828. It should fail a little further, could you possibly post the generated syslog?

by dsjonny, 6 years ago

Attachment: SYSLOG00.TXT added

comment:11 by dsjonny, 6 years ago

I have attached the SYSLOG00.TXT now. Sorry, but before I had no FAT32 drive.. I will try the latest image tomorrow when it will be available.

comment:12 by dsjonny, 6 years ago

Sorry, but does not working.. :( Haiku boot successful, but after the boot screen I got only a blank screen (but no KDL). I will try to get the syslog..

by dsjonny, 6 years ago

Attachment: SYSLOG02.TXT added

comment:13 by dsjonny, 6 years ago

OK, attached SYSLOG02.TXT.

Found these for the VGA:

intel_extreme: CALLED status_t init_hardware()
intel_extreme: CALLED status_t init_driver()
[36mAGP:[0m bus manager init
[36mAGP:[0m found 0 AGP devices
intel_extreme: (0) SandyBridge Desktop GT1, revision = 0x9
intel_extreme: CALLED const char ** publish_devices()
intel_extreme: CALLED struct device_hooks * find_device(const char *)
loaded driver /boot/system/add-ons/kernel/drivers/dev/graphics/intel_extreme
Radeon - init_hardware: Version: 5.1.6.0
Radeon - Radeon_CardDetect: no supported devices found
radeon_hd: init_hardware
S3: init_hardware() - no supported devices
vesa: init_hardware()
vesa: init_driver()
vesa: publish_devices()
vesa: find_device()
loaded driver /boot/system/add-ons/kernel/drivers/dev/graphics/vesa
intel_extreme: CALLED status_t device_open(const char *, long unsigned int, void **)
intel_extreme: intel_extreme: CALLED status_t intel_extreme_init(intel_info &)
add_memory_type_range(4345, 0xf7800000, 0x400000, 0)
set MTRRs to:
  mtrr:  0: base: 0xbe000000, size:   0x400000, type: 0
  mtrr:  1: base: 0xbe400000, size:   0x200000, type: 0
  mtrr:  2: base: 0xf0000000, size: 0x10000000, type: 0
  mtrr:  3: base: 0xc0000000, size: 0x40000000, type: 1
add_memory_type_range(4347, 0xf7a00000, 0x200000, 0)
set MTRRs to:
  mtrr:  0: base: 0xbe000000, size:   0x400000, type: 0
  mtrr:  1: base: 0xbe400000, size:   0x200000, type: 0
  mtrr:  2: base: 0xf0000000, size: 0x10000000, type: 0
  mtrr:  3: base: 0xc0000000, size: 0x40000000, type: 1
intel_gart: detected 512 MB of stolen memory, aperture size 256 MB, GTT size 2048 KB
intel_gart: GTT base = 0xf7a00000
intel_gart: MMIO base = 0xf7800000
intel_gart: GMR base = 0xe0000000
add_memory_type_range(4348, 0xe0000000, 0x10000000, 1)
set MTRRs to:
  mtrr:  0: base: 0xbe000000, size:   0x400000, type: 0
  mtrr:  1: base: 0xbe400000, size:   0x200000, type: 0
  mtrr:  2: base: 0xf0000000, size: 0x10000000, type: 0
  mtrr:  3: base: 0xc0000000, size: 0x40000000, type: 1
add_memory_type_range(4350, 0xf7800000, 0x400000, 0)
set MTRRs to:
  mtrr:  0: base: 0xbe000000, size:   0x400000, type: 0
  mtrr:  1: base: 0xbe400000, size:   0x200000, type: 0
  mtrr:  2: base: 0xf0000000, size: 0x10000000, type: 0
  mtrr:  3: base: 0xc0000000, size: 0x40000000, type: 1
[36mAGP:[0m create memory 0x82ecc6e0, base e0000000, size 10000, flags 0
[36mAGP:[0m allocation is made of reserved memory
[36mAGP:[0m reserved memory already bound
intel_extreme:CALLED status_t intel_en_gating(intel_info &)
intel_extreme:SandyBridge clock gating
intel_extreme:CALLED status_t intel_en_downclock(intel_info &)
intel_extreme:intel_en_downclock: Skip GPU downclocking on non-mobile device.
intel_extreme: error: could not allocate overlay memory! Out of memory
intel_extreme: intel_extreme_init: completed successfully!
intel_extreme: accelerant: intel_extreme.accelerant
slab memory manager: created area 0xdf001000 (4351)
intel_extreme accelerant:CALLED status_t intel_init_accelerant(int)
intel_extreme accelerant:head detected: 0x1
intel_extreme accelerant:adpa: 00c40000, dova: 0000001c, dovb: 0000001c, lvds: 00000000
DDC: ddc2_read: DDC information read failure
Last message repeated 3 times.
intel_extreme accelerant:getting EDID on port A (analog) failed : No Error (8). Trying on port C (lvds)
DDC: ddc2_read: DDC information read failure
Last message repeated 3 times.
intel_extreme accelerant:getting EDID on port C failed : No Error (8)
intel_extreme accelerant:CALLED status_t intel_get_edid_info(void *, long unsigned int, uint32 *)
intel_extreme accelerant:CALLED status_t intel_get_edid_info(void *, long unsigned int, uint32 *)
intel_extreme accelerant:CALLED uint32 intel_accelerant_mode_count()
intel_extreme accelerant:CALLED status_t intel_get_mode_list(display_mode *)
intel_extreme accelerant:intel_set_display_mode(1024x768)

comment:14 by anevilyak, 6 years ago

Blocking: 9949 added

(In #9949) Possibly the same underlying issue as #8952.

comment:15 by korli, 6 years ago

Please attach a syslog with hrev46033 or newer. Thanks!

by dsjonny, 6 years ago

Attachment: hrev46033syslog.txt added

comment:16 by dsjonny, 6 years ago

I have attached the syslog from hrev46033 nightly. Please chek it.

comment:17 by korli, 6 years ago

Interesting... the line "intel_extreme: error: could not allocate overlay memory! Out of memory" doesn't appear anymore in your syslog.

by dsjonny, 6 years ago

Attachment: hrev46033syslog-2.txt added

comment:18 by dsjonny, 6 years ago

I have attached a new syslog, and there is the memory problem. I have changed the BIOS settings, and added more memory (512MB), and used the DVI port in analog and in digital mode. And there are some interesting lines (for me) in the log:

intel_extreme accelerant:getting EDID on port A (analog) failed : No Error (8). Trying on port C (lvds)
intel_extreme accelerant:getting EDID on port C failed : No Error (8)

intel_extreme accelerant:intel_set_display_mode(1024x768)

Maybe there is the problem with the ports? I can use only HDMI/DVI.

And here are all related lines:

intel_extreme: CALLED status_t init_hardware()
intel_extreme: CALLED status_t init_driver()
[36mAGP:[0m bus manager init
[36mAGP:[0m found 0 AGP devices
intel_extreme: (0) SandyBridge Desktop GT1, revision = 0x9
intel_extreme: CALLED const char ** publish_devices()
intel_extreme: CALLED struct device_hooks * find_device(const char *)
loaded driver /boot/system/add-ons/kernel/drivers/dev/graphics/intel_extreme
vesa: init_hardware()
vesa: init_driver()
vesa: publish_devices()
vesa: find_device()
loaded driver /boot/system/add-ons/kernel/drivers/dev/graphics/vesa
intel_extreme: CALLED status_t device_open(const char *, long unsigned int, void **)
intel_extreme: intel_extreme: CALLED status_t intel_extreme_init(intel_info &)
intel_gart: detected 512 MB of stolen memory, aperture size 256 MB, GTT size 2048 KB
intel_gart: GTT base = 0xf7a00000
intel_gart: MMIO base = 0xf7800000
intel_gart: GMR base = 0xe0000000
[36mAGP:[0m create memory 0x82e54a40, base e0000000, size 10000, flags 0
[36mAGP:[0m allocation is made of reserved memory
[36mAGP:[0m reserved memory already bound
intel_extreme:CALLED status_t intel_en_gating(intel_info &)
intel_extreme:SandyBridge clock gating
intel_extreme:CALLED status_t intel_en_downclock(intel_info &)
intel_extreme:intel_en_downclock: Skip GPU downclocking on non-mobile device.
[36mAGP:[0m Aperture::CreateMemory(): did not find a free space large for this memory object
intel_extreme: error: could not allocate overlay memory! Out of memory
intel_extreme: using message signaled interrupts
intel_extreme: intel_extreme_init: completed successfully!
intel_extreme: accelerant: intel_extreme.accelerant
intel_extreme accelerant:CALLED status_t intel_init_accelerant(int)
intel_extreme accelerant:head detected: 0x3
intel_extreme accelerant:adpa: 00c40000, dova: 80000a44, dovb: 80000a44, lvds: 00000000
DDC: ddc2_read: DDC information read failure
Last message repeated 3 times.
intel_extreme accelerant:getting EDID on port A (analog) failed : No Error (8). Trying on port C (lvds)
DDC: ddc2_read: DDC information read failure
Last message repeated 3 times.
intel_extreme accelerant:getting EDID on port C failed : No Error (8)
intel_extreme accelerant:CALLED status_t intel_get_edid_info(void *, long unsigned int, uint32 *)
intel_extreme accelerant:CALLED status_t intel_get_edid_info(void *, long unsigned int, uint32 *)
intel_extreme accelerant:CALLED uint32 intel_accelerant_mode_count()
intel_extreme accelerant:CALLED status_t intel_get_mode_list(display_mode *)
intel_extreme accelerant:intel_set_display_mode(1024x768)
[36mAGP:[0m create memory 0x82e055a0, base e0010000, size 300000, flags 0
[36mAGP:[0m allocation is made of reserved memory
[36mAGP:[0m reserved memory already bound
intel_extreme accelerant:PLL limits, min: p 5 (p1 1, p2 10), n 1, m 79 (m1 12, m2 5)
intel_extreme accelerant:PLL limits, max: p 80 (p1 8, p2 5), n 5, m 127 (m1 22, m2 9)
intel_extreme accelerant:compute_pll_divisors: required MHz: 65
intel_extreme accelerant:compute_pll_divisors: found: 65.1429 MHz, p = 42 (p1 = 3, p2 = 14), n = 4, m = 114 (m1 = 21, m2 = 9)
intel_extreme accelerant:PLL limits, min: p 5 (p1 1, p2 10), n 1, m 79 (m1 12, m2 5)
intel_extreme accelerant:PLL limits, max: p 80 (p1 8, p2 5), n 5, m 127 (m1 22, m2 9)
intel_extreme accelerant:compute_pll_divisors: required MHz: 65
intel_extreme accelerant:compute_pll_divisors: found: 64.8 MHz, p = 40 (p1 = 4, p2 = 10), n = 4, m = 108 (m1 = 20, m2 = 8)
intel_extreme accelerant:CALLED status_t intel_get_frame_buffer_config(frame_buffer_config *)

Unfortunatelly now I can only boot Haiku in safe video mode.

comment:19 by dsjonny, 6 years ago

Is there any way to list all video output and to select another?

comment:20 by dsjonny, 6 years ago

Using hrev46281 the status is the same. I can boot Haiku only using safe video mode (selected from the boot menu). But I need to do this in every boot. :( I tried to remove the intel_extreme drom the drivers, but I can't (I got a readonly volume alert).

comment:21 by dsjonny, 6 years ago

Priority: normalhigh

I have changed the priority to high, because I cannot boot Haiku until I logged into the boot menu and set the safe video mode.

comment:22 by anevilyak, 6 years ago

Priority: highnormal

comment:23 by axeld, 6 years ago

dsjonny: if it works, it's not high priority :-) Seriously, though, you can make this setting permanent using the kernel settings file by adding the following line to it:

fail_safe_video_mode true

comment:24 by dsjonny, 6 years ago

Thank you! I will try it.

comment:25 by dsjonny, 6 years ago

It works fine now (automatically boot into safe video mode). Sorry, I did not know this option.

However, it does not solve the main problem. So, I hope somebody will find the solution.

comment:26 by axeld, 3 years ago

Owner: changed from axeld to nobody
Status: reopenedassigned

comment:27 by pulkomandy, 13 months ago

Milestone: R1R1/beta2

comment:28 by waddlesplash, 8 months ago

Blocked By: 14301 added
Resolution: duplicate
Status: assignedclosed

#14301 is a dupe but has more relevant info.

Note: See TracTickets for help on using tickets.