https://bugzilla.novell.com/show_bug.cgi?id=270846 ms@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Add support for new RANDR |sax2: Add support for new RANDR 1.2 driver |1.2 driver configuration |configuration theme |theme | ------- Comment #2 from ms@novell.com 2007-05-03 03:49 MST ------- just a side note: I'm very busy with other tasks right now so this one here should be clarified and specified _exactly_ as soon as possible. I also like to ask for help concerning query of randr extension data (example implementation somewhere ?) Thanks -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=270846 mhopf@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|mhopf@novell.com | ------- Comment #4 from mhopf@novell.com 2007-05-03 11:12 MST ------- All monitor-specific data still comes from the driver :-( (II) intel(0): Output configuration: (II) intel(0): Pipe A is on (II) intel(0): Display plane A is now enabled and connected to pipe A. (II) intel(0): Pipe B is off (II) intel(0): Display plane B is now disabled and connected to pipe B. (II) intel(0): Output VGA is connected to pipe A Even worse, this full information is not included with -probeonly. With that option, only the previous short information is available: (II) intel(0): Output VGA connected (II) intel(0): Output VGA using initial mode 1152x768 Ok, now as Keith has added the "Enable" option, we can even have the same configuration as before, by adding a monitor section for each detected monitor, with the 'Option "Enable" "true"' included (tested, this works). However, strictly speaking this is no longer necessary as we already have a workaround / solution for that in /etc/sysconfig/displaymanager, setting this dynamically just before login. We should discuss which way to go in this case (laptop). For the general (fixed monitors, dual-head workstation) case: In sax, add monitor descriptions named "Monitor[<OUTPUT>]" as found, add Option "monitor-<OUTPUT>" "Monitor[<OUTPUT>]" to the driver section. Add Virtual <width> <height> with the *total* width and height of the complete monitor setup to each Display subsection of the Screen section. Add Option "LeftOf" "Monitor[<OTHER_OUTPUT>]" to the monitor section for the left monitor. Alternatively, use RightOf, Above, Below, or the Position option: Option "Position" "<x> <y>" with x, y being the offset of the origin of the combined monitor layout. The Position option would need to be set for all monitors. The <OUTPUT> names so far can only be determined by checking the logfile (driver dependent, but this is intel only ATM), or with xrandr with a running Xserver. If xrandr returns "default" as the only output, the driver is not RANDR 1.2 capable. I cannot estimate ATM how this interferes with the inner working of sax2, but I doubt this can be implemented with profiles alone. I have yet to verify the multimonitor-solution, the one laptop sitting here doesn't have os10.3 installed in a working condition, and the desktop has only a single output... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=270846 ------- Comment #6 from ms@novell.com 2007-05-04 02:30 MST ------- Created an attachment (id=137439) --> (https://bugzilla.novell.com/attachment.cgi?id=137439&action=view) randr 1.2 test coding -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=270846 ms@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |mhopf@novell.com ------- Comment #8 from ms@novell.com 2007-05-07 02:23 MST ------- Matthias if you find time could you check and adapt the code from comment #6 according to the needs of the configuration Thanks -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=270846 ms@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |casualprogrammer@yahoo.com ------- Comment #10 from ms@novell.com 2007-05-07 06:34 MST ------- *** Bug 271770 has been marked as a duplicate of this bug. *** -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=270846 ms@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |felix.rommel@web.de ------- Comment #12 from ms@novell.com 2007-06-03 13:33 MST ------- *** Bug 280336 has been marked as a duplicate of this bug. *** -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c14 --- Comment #14 from Marcus Schaefer <ms@novell.com> 2007-07-04 02:19:43 MST --- *** Bug 289445 has been marked as a duplicate of this bug. *** https://bugzilla.novell.com/show_bug.cgi?id=289445 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c18 --- Comment #18 from Matthias Hopf <mhopf@novell.com> 2007-07-20 12:36:07 MST --- Sorry for the delay (PSA). Ok, I can add the required information to this tool, but on the other hand why don't we use xrandr? If you think that the output of xrandr will change too much in the near future, it might be a good idea to use our own program. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c20 --- Comment #20 from Matthias Hopf <mhopf@novell.com> 2007-07-30 07:42:13 MST --- Created an attachment (id=153989) --> (https://bugzilla.novell.com/attachment.cgi?id=153989) Test case with added code. Ok, this is the "working" test case. You certainly will want to have a different output scheme. What we need is the names of the outputs (output->name), the connection state (output->connection, if you want to automatically select the right output selection in a multi-monitor layout), the number of preferred modes (output->npreferred, the first npreferred modes of the mode list are native/preferred otherwise by the monitor), and the actual modelist. Actually, if the mode list has at least one preferred mode, it should be taken (the first preferred mode is the most preferred (native) one). What we *really* only need for a trivial setup is the list of output names (and that could even be derived statically per chipset). But in order to make it more automatic, we need the connection state and the modelists. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c29 Marcus Schaefer <ms@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED --- Comment #29 from Marcus Schaefer <ms@novell.com> 2007-08-13 03:40:54 MST --- first version submitted... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c32 Matthias Hopf <mhopf@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED | --- Comment #32 from Matthias Hopf <mhopf@novell.com> 2007-08-17 10:21:48 MST --- Hm. This doesn't work correctly. Also, Stefan, I noticed that the xorg.conf (5) manpage is way too old. Why does it come from? git has a much newer manpage, including the valid monitor description for RandR 1.2. I'll attach SaX.log, xorg.conf, and Xorg.0.log for a multi-monitor setup. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c34 --- Comment #34 from Matthias Hopf <mhopf@novell.com> 2007-08-17 10:24:26 MST --- Created an attachment (id=158247) --> (https://bugzilla.novell.com/attachment.cgi?id=158247) xorg.conf The xorg.conf written by sax2. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c36 --- Comment #36 from Stefan Dirsch <sndirsch@novell.com> 2007-08-17 10:40:40 MST --- Thanks. I'll update the xorg.conf manual page. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c38 --- Comment #38 from Matthias Hopf <mhopf@novell.com> 2007-08-17 11:12:31 MST --- In Driver section: - Option "monitor-TV" "EXT" + Option "monitor-VGA" "EXT" + Option "monitor-LVDS" "INT" In existing monitor section: - Option "Position" "LeftOf" + Option "LeftOf" "INT" - VertRefresh 50-60 + VertRefresh 50-61 Additional monitor section: +Section "Monitor" + DisplaySize 246 184 + HorizSync 31-50 + Identifier "INT" + ModelName "1024X768@60HZ" + Option "PreferredMode" "1024x768" + VendorName "--> VESA" + VertRefresh 50-61 +EndSection Alternatively, you could use Option "Position" "0 0" in monitor "EXT", and Option "Position" "1024 0" in monitor "INT". That would also be independent of bugs (like in the LeftOf case) in RANDR - but of course prone to own bugs with calculating these values. Your choice. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c40 Marcus Schaefer <ms@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|REOPENED |NEEDINFO Info Provider| |mhopf@novell.com --- Comment #40 from Marcus Schaefer <ms@novell.com> 2007-08-20 02:48:47 MST --- sorry but this bunch of information is exactly the situation I'd like to avoid. One comment was:

SaX.log shows, that sysp is apparently not printing enough information, and that one is interpreted wrong: Card0 => RandR : TV disconnected 0 That is only one of 4 outputs, and it is disconnected, so it shouldn't be used.

That information is the result from your patch. I just rely on what has been changed there -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c42 Matthias Hopf <mhopf@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |REOPENED Info Provider|mhopf@novell.com | --- Comment #42 from Matthias Hopf <mhopf@novell.com> 2007-08-20 07:17:31 MST --- (In reply to comment #40 from Marcus Schaefer)

...

SaX.log shows, that sysp is apparently not printing enough information, and that one is interpreted wrong: Card0 => RandR : TV disconnected 0 That is only one of 4 outputs, and it is disconnected, so it shouldn't be used.

That information is the result from your patch. I just rely on what has been changed there

It is there. The patch just prints out 4 lines. This is just the first line. Also I noted that a "Preferred" option was added, while the '0' indicates that there is no preferred mode. But that is a minor issue. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c44 Matthias Hopf <mhopf@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |REOPENED Info Provider|mhopf@novell.com | --- Comment #44 from Matthias Hopf <mhopf@novell.com> 2007-08-20 10:42:34 MST --- I have an idea: Generally, the connected output only matters if we're doing a multihead setup. In that case I would order the outputs according to patterns. That would be LVDS LVDS-* DVI DVI-* TMDS TMDS-* VGA VGA-* TV <all_others> (just an example). Those not connected will not be used (you only want to create a multihead setup when monitors are actually connected). For a dualhead setup the first two entries in this list would be used. (In reply to comment #43 from Marcus Schaefer)

1) TV or not tv ?

If this cannot be handled like described above, I would like to have TV ignored, due to serveral serious issues (needs to be specified more (PAL/NTSC), low res, connection scanning fails often, etc.).

2) how to handle "disconnected" ?

They should IMHO not be configured.

3) Position setup is only possible in new Monitor section not in the existing section. So

Option "Position" "0 0" is equal to leftOf

==> what would be rightof, above and below ?

In that case we should use LeftOf etc., and fix the remaining issues in RandR (should do that anyway). AFAIK you cannot have negative position values. You only need LeftOf etc. in all but one monitor sections. With "position" you have to put it into each and every monitor section.

4) there are too many code paths for the setup of vsync in sax2 (DDC data CDB data, special profiles, etc...) if 60Hz is not valid anymore I

I understand. In that case, a safe margin in the Xserver is probably the best idea.

would suggest that the X-Server takes care for a not too restrictive check. And does this apply to intel only or all drivers ? If sufficient we can adapt the ranges in the CDB

That's probably for all RandR 1.2 capable drivers. As long as they use the infrastructure from RandR, and don't provide their own, that is... HTH, more tomorrow. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c46 Matthias Hopf <mhopf@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|FIXED | --- Comment #46 from Matthias Hopf <mhopf@novell.com> 2007-08-22 07:04:53 MST --- Reopening for my fix of the Xserver... -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846#c47 --- Comment #47 from Matthias Hopf <mhopf@novell.com> 2007-08-22 07:07:20 MST --- Created an attachment (id=159052) --> (https://bugzilla.novell.com/attachment.cgi?id=159052) Fix for Xserver being more fuzzy about mode validation. This fix has been astonishingly simple. I verified it on my laptop, it seems to work perfectly. I have to discuss upstream what the actual check should look like (+/- 0.5 or +/- 1%). For us this is good enough. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846 Stefan Dirsch <sndirsch@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P0 - Crit Sit -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846 User ms@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=270846#c50 Marcus Schaefer <ms@novell.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |jharmon@novell.com --- Comment #50 from Marcus Schaefer <ms@novell.com> 2008-01-09 01:41:25 MST --- *** Bug 352437 has been marked as a duplicate of this bug. *** https://bugzilla.novell.com/show_bug.cgi?id=352437 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=270846 User ms@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=270846#c52 --- Comment #52 from Marcus Schaefer <ms@novell.com> 2008-01-09 01:42:19 MST --- *** Bug 352474 has been marked as a duplicate of this bug. *** https://bugzilla.novell.com/show_bug.cgi?id=352474 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.