message #4:
Code:
Section "Device"
Identifier "Card1"
Driver "intel"
BusID "PCI:0:0:2"
EndSection
You seem to have inadvertently changed the BusID into the one above:
Rich (BB code):
[....]
Section "Device"
Identifier "Card1"
Driver "intel"
BusID "pci0:0:2:0"
EndSection
(commenting the second section did nothing so I assume it's the BusID format)
Now I have two sections but the first format : both computer screens work together, but the TV is not shown in the Displays widget. Its screen is black,
but with a cursor, like when you first do startx to verify Xorg works. So it seems it is detected and connected to Xorg, but not in the graphical tool to move screens around, etc (I'm on Mate, but also have XFCE)
Here are the two cards :
Rich (BB code):
vgapci1@pci0:0:2:0: class=0x030000 rev=0x09 hdr=0x00 vendor=0x8086 device=0x0162 subvendor=0x8086 subdevice=0x2032
vendor = 'Intel Corporation'
device = 'IvyBridge GT2 [HD Graphics 4000]'
class = display
Any ideas?
[...] I see two things that may be related in my
Xorg.0.log :
Rich (BB code):
[ 15.537] (!!) More than one possible primary device found
[ 15.537] (--) PCI: (0@0:2:0) 8086:0162:8086:2032 rev 9, Mem @ 0xf7400000/4194304, 0xd0000000/268435456, I/O @ 0x0000f000/64, BIOS @ 0x????????/65536
According to
pciconf(8) and
xorg.conf(5), respectively:
Rich (BB code):
[...] Selectors identify a PCI device by its ad-
dress in PCI config space and can take one of the following forms:
• pcidomain:bus:device:function
Rich (BB code):
BusID "bus-id"
This specifies the bus location of the graphics card. For
PCI/AGP cards, the bus-id string has the form PCI:bus@domain:de-
vice:function (e.g., "PCI:1@0:0:0" might be appropriate for an
AGP card). The "@domain" part can be left out for PCI domain 0.
Thus, the pciconf output
vgapci1@pci0:0:2:0: translates[*] to:
BusID "PCI:0@0:2:0" (long form)
or
BusID "PCI:0:2:0" (short form)
(The different notation in the Handbook likely works as well but, I can't find that documented in the man pages, and I haven't tried it.)
From your Xorg log file I see that you are on 14.1-RELEASE; this will become EoL as of tomorrow (see:
Supported FreeBSD releases).
The 'problem' that you are facing is that normally you'd be advised to upgrade to 14.2-RELEASE. However, as also the the quarter will end tomorrow, and Q2 begins, all the packages on the remote FreeBSD servers will be (re)build against 14.2-RELEASE (this takes time, my guess is that within a week it will be completed). For almost all packages this won't hardly matter but, for a very small number of packages containing kernel modules like the package
graphics/drm-61-kmod (containing the
i915kms driver) it is rather essential (on 14.2-RELEASE instead of
drm-61-kmod-6.1.128.1401000_2 you need
drm-61-kmod-6.1.128.1402000_2). When all packages have been build against 14.2-RELEASE you can upgrade conventionally.
If you upgrade to 14.2-RELEASE now, you'd have to either build that very small number of packages containing kernel modules you actually need, like drm-{515|61}-kmod[**] or set up the kmods repository (see:
Possible solution to the drm-kmod kernel mismatch after upgrade from Bapt).
Let me emphasize: these extra efforts for the 'kmods kernel modules' (
very few relevant packages) will disappear when (with respect to 14.2-RELEASE) all packages have been rebuild against 14.2-RELEASE. Then you can use the latest or quarterly repsitories as usual; until this issue might appear again when 14.3-RELEASE will be released.
If you'd like to try using the drm-61-kmod on behalf of your mobo_HDMI-TV connection, you could consider waiting for a week before upgrading to 14.2-RELEASE; I presume you want to upgrade anyway as running an unsupported 14.1-RELEASE is not really what you want.
However, you could also consider trying the old Xorg intel driver (contained in
x11-drivers/xf86-video-intel)[***]; that might work as your intel GPU is not of a very recent date. For that you do not have to take the extra effort of getting a specific version tailored to 14.1-RELEASE or 14.2-RELEASE (because there just aren't any tailored ones).
Generally when testing such a non-trivial setup of graphics cards and drivers disable your display manager (like Lightdm) (comment out in rc.conf), boot into a virtual console to the command line see if everything seems to work (check 'kldload-ed' drivers), and start X manually.
___
[*]
Or in 'symbolic letters' (note the different order of 'domain' in the sequence):
vgapci1@pciW:X:Y:Z: to
BusID "PCI:X@W:Y:Z" (long form)
or
BusID "PCI:X:Y:Z" (short form)
[**]
On 14.2-RELEASE, with 'kmods' set up, on latest (quarterly differs slightly obviously) repository :
Code:
# pkg search '^drm-(515|61)-kmod' | sort
drm-515-kmod-5.15.160.1401000_3 DRM drivers modules
drm-515-kmod-5.15.160.1402000_3 DRM drivers modules
drm-61-kmod-6.1.128.1401000_2 DRM drivers modules
drm-61-kmod-6.1.128.1402000_2 DRM drivers modules
[***]
UMS (User Mode Setting) Xorg drivers are for some time now not longer being developed—they are being phased out when all development effort is concentrated on KMS (Kernel Mode Setting) drivers. This XOrg intel driver uses UMS and therefore it doesn't need to work with a DRM-KMS counterpart. Contrary to using the Xorg
modesetting(4) that
does need a DRM-KMS counterpart as for example
i915kms contained in
graphics/drm-61-kmod
