[Bug 60131] New: Problem with detecting HDMI plugging/unplagging (Samsung TV, Intel card)

Tue Jun 25 14:21:52 PDT 2013

https://bugzilla.kernel.org/show_bug.cgi?id=60131

           Summary: Problem with detecting HDMI plugging/unplagging
                    (Samsung TV, Intel card)
           Product: Drivers
           Version: 2.5
    Kernel Version: 3.9.7
          Platform: All
        OS/Version: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: Video(DRI - Intel)
        AssignedTo: intel-gfx-bugs at lists.freedesktop.org
        ReportedBy: gonciarz at gmail.com
                CC: intel-gfx-bugs at lists.freedesktop.org
        Regression: No

Hi,

I have a problem with plugging my Monitor/TV (Samsung SyncMaster T260HD) to my
Debian media server through HDMI / Display Port.
When I plug in either HDMI cable directly or DisplayPort adapter (+ the same
HDMI cable) to my motherboard, 
then system doesn't seem to register any udev event. D-SUB/VGA works fine.

I've checked that using the following command:
udevadm monitor

I've also checked xrandr (from VNC server):

Screen 0: minimum 320 x 200, current 1024 x 768, maximum 8192 x 8192
VGA1 disconnected (normal left inverted right x axis y axis)
HDMI1 disconnected (normal left inverted right x axis y axis)
DP1 disconnected (normal left inverted right x axis y axis)
HDMI2 disconnected (normal left inverted right x axis y axis)
DP2 disconnected (normal left inverted right x axis y axis)

and checked the drm modulu's status directly:
cat /sys/class/drm/card0-HDMI-A-1/status
cat /sys/class/drm/card0-HDMI-A-2/status

neither HDMI nor DP doesn't work.

I was able to connect my TV and multimedia server about a 1-1.5 year ago using
the same configuration and it was successfully discovered.
It just stopped working after a system upgrade (sorry I can't tell you exactly
the certain time).
I used to have Debian Wheezy (at that time it was Debian testing disto). Then I
migrated to Debian testing again (jessie).
But anyway, I'm sure that this hardware is able to detect my monitor connected
by HDMI cable.

More over, I've connected the same monitor and HDMI cable with different pair
of devices and operating system:
- Laptop1 (Core i5 + Nvidia1 card) + Display Port adapter + Arch Linux
- Laptop2 (Core i5 + Nvidia2 card) + Ubuntu12.10/Windows7
and everything works fine. So I assume it's not a monitor/TV or cable/adapter
issue.

When I connect Monitor to server using D-SUB (VGA) port, then everything works
fine (including booting process).
I've tried to connect the HDMI cable together with D-SUB during the boot time,
but that didn't help.

After booting with D-SUB/VGA cable connected, and activating VNC, then
disconnecting D-SUB and connecting HDMI
I've noticed that xrander shows some extra model lines even though it didn't
report any device connected to VGA/HDMI/DP port.

I was so desperate to solve this problem that I even flash my mainboard,
because I've seen some changes around HDMI in the bios changelog.
That didn't solve the problem

I fought I may force graphic card to see connected monitor like for Nvidia
card:
from xorg.conf:

Option         "ConnectedMonitor" "DP1"
Option         "CustomEDID" "DP1:/etc/X11/edid.bin"
Option         "IgnoreEDID" "false"
Option         "UseEDID" "true"

but these options don't work with intel driver. But anyway I've dumped edid
file.
I've noticed that there is an option in the DRM (DRM_LOAD_EDID_FIRMWARE) to
force using EDID data from file, but I don't think that may be the good path.

Then I updated the kernel to the newest one (3.9.7) (with more or less the
default debian configuration) but that also didn't helped.

My contact in Samsung told me that he isn't expert in this matter, but it might
be a problem with managing power on HDMI output.
He wasn't sure on which side. Monitor can't send me any EDID response after
connecting power supply.
As he remembers the source of power has to deliver 5V and 50mA on 18th Pin to
the monitor.
Then monitor returns with signal of high state on HPD (19th pin) and after that
sends EDID.

He guesses that monitor requires higher current strength than chipset is able
to deliver.
TV doesn't leave sleep state, HPD and EDID is silent.

He mentioned also that sometimes after switching TV/monitor to HDMI signal,
HPD's pin is set to high state 
which cause that graphic card can leave sleep state.

In my case I always connected TV/monitor after switching it to HDMI signal as
it is the same for other laptops I tested.

The Debian distro blocks me here a little bit. 
I prefer Debian for professional server solutions but except sid (unstable)
release it offers very old packages. 
Though it might be very difficult for me to upgrade xorg, libdrm, mesa, etc.
versions, but I may always switch the kernel.

Below you may find some information about my configuration.
If you need any more information please do not hesitate to contact me.

Best regards
Robert 

uname -a
Linux smithy 3.9.7 #1 SMP Sat Jun 22 22:04:11 CEST 2013 x86_64 GNU/Linux (now
it's the newest stable one)

lsb_release -a
Distributor ID: Debian
Description:    Debian GNU/Linux testing (jessie)
Release:        testing
Codename:       jessie

lshw -C display
  *-display
       description: VGA compatible controller
       product: 2nd Generation Core Processor Family Integrated Graphics
Controller
       vendor: Intel Corporation
       physical id: 2
       bus info: pci at 0000:00:02.0
       version: 09
       width: 64 bits
       clock: 33MHz
       capabilities: msi pm vga_controller bus_master cap_list rom
       configuration: driver=i915 latency=0
       resources: irq:52 memory:fb800000-fbbfffff memory:e0000000-efffffff
ioport:ff00(size=64)

lspci -nnk | grep -iA2 vga
00:02.0 VGA compatible controller [0300]: Intel Corporation 2nd Generation Core
Processor Family Integrated Graphics Controller [8086:0102] (rev 09)
        Subsystem: Gigabyte Technology Co., Ltd Device [1458:d000]
        Kernel driver in use: i915

dmidecode -t 1
# dmidecode 2.12
SMBIOS 2.4 present.

Handle 0x0001, DMI type 1, 27 bytes
System Information
        Manufacturer: Gigabyte Technology Co., Ltd.
        Product Name: H67MA-UD2H-B3
        Version:
        Serial Number:
        UUID: 00000000-0000-0000-0000-1C6F65D503D0
        Wake-up Type: Power Switch
        SKU Number:
        Family:

cat /var/log/Xorg.0.log | grep -e WW -e EE | grep -v '(II)'
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   387.576] (WW) The directory "/usr/share/fonts/X11/cyrillic" does not exist.
[   387.576] (WW) The directory "/usr/share/fonts/X11/100dpi/" does not exist.
[   387.576] (WW) The directory "/usr/share/fonts/X11/75dpi/" does not exist.
[   387.576] (WW) The directory "/usr/share/fonts/X11/100dpi" does not exist.
[   387.576] (WW) The directory "/usr/share/fonts/X11/75dpi" does not exist.
[   387.860] (WW) intel(0): No outputs definitely connected, trying again...
[   387.860] (WW) intel(0): Unable to find connected outputs - setting 1024x768
initial framebuffer
[   387.902] (WW) evdev: Logitech USB Receiver: ignoring absolute axes.

cat /etc/default/grub | grep LINUX_DEFAULT
GRUB_CMDLINE_LINUX_DEFAULT="quiet"

-- 
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
You are the assignee for the bug.