Created attachment 114078 [details] dmesg log ==System Environment== -------------------------- BIOS: V55 with 'dynamic memory self refresh' disabled Regression: No. We first run tests with DP-DVI adapter cable Non-working platforms: BSW ==kernel== -------------------------- -testing: drm-intel-testing-2015-02-27 + attachment 113995 [details] [review] (fails) ==Bug detailed description== ----------------------------- I'm not sure whether or not this is a real bug, anyway I submit it for tracking. When I connect DVI monitor with DP-DVI adapter cable to DP port of DUT, "testdisplay -i" shows that a "HDMI" port is connected. ==Reproduce steps== ---------------------------- 1. Connect DVI monitor with DP-DVI adapter cable to DP port of DUT and Boot 2. "testdisplay -i" shows: Connectors: id encoder status type size (mm) modes 41 40 connected eDP 280x160 1 modes: name refresh (Hz) hdisp hss hse htot vdisp vss vse vtot flags type clock [0] 1920x1080 60 1920 1966 1996 2080 1080 1082 1086 1112 0xa 0x48 138780 34 33 connected HDMI-A 510x290 22 modes: name refresh (Hz) hdisp hss hse htot vdisp vss vse vtot flags type clock [0] 1920x1080 60 1920 2008 2052 2200 1080 1084 1089 1125 0x5 0x48 148500 [1] 1680x1050 60 1680 1728 1760 1840 1050 1053 1059 1080 0x9 0x40 119000 [2] 1280x1024 75 1280 1296 1440 1688 1024 1025 1028 1066 0x5 0x40 135000 [3] 1280x1024 60 1280 1328 1440 1688 1024 1025 1028 1066 0x5 0x40 108000 ....... You can see system recognized a "HDMI" monitor.
What exactly is this DP->DVI cable of yours? This behaviour is perfectly normal if the cable is essentially a DP++ adapter (ie. just level shifters and direct wires).
This is a cable with DP at one end and DVI at the other end
It's a passive DP->DVI adapter that uses the HDMI encoder. How we name the outputs is very, very loose and has to be since we have no idea what the actual connector on the chasis is. If we want to be smart, we call the connector DDI and let it have both DP/HDMI modes (which is a connector property). This also gets around the presentation of both DP and HDMI for the same physical connector and the belief that we could drive both at the same time. Gone with WONTFIX as it is designed in behaviour at this point. Fortunately, this is not presented in GUIs which just list monitors and modes and configuration.
I have more questions: 1. Do I need have furthur testing for such combination (testing on DVI monitor with DP->DVI adapter)? I actually tried to run "testdisplay -a" and found some modes don't work. 2. I also tried a HDMI->DVI adapter, with which there was none connected. Do I need test such combination?
HDMI->DVI (even DP->HDMI->DVI should work, so long as it is not dvi duallink) as they are passive converters. So 1 and 2 definitely merit further investigation and reporting as bugs.
Got it. Thanks.
Closing old verified.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.