Menu
▾
▴
vizstack-users — Mailing list for everyone who uses VizStack
You can subscribe to this list here.
| 2009 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(1) |
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2010 |
Jan
|
Feb
(3) |
Mar
(8) |
Apr
(26) |
May
(41) |
Jun
(3) |
Jul
(19) |
Aug
|
Sep
|
Oct
|
Nov
(2) |
Dec
|
| 2011 |
Jan
(17) |
Feb
(5) |
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2012 |
Jan
|
Feb
|
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: <Sim...@cs...> - 2010-05-15 07:40:31
|
I'll give that version a shot, but I think it's one of the ones we've tried - it may be something to do with the GLX issues, if that version has worked as intended for you before. Simon ________________________________________ From: Kumar, Shree [shr...@hp...] Sent: Saturday, May 15, 2010 5:02 PM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch I checked the version of the nvidia driver on my laptop. Turns out its not old. The version number is : 195.36.15 . You may want to try with this driver, but YMMV. I tried on my machine with this, but faced some other problem. The configure script just hangs with this driver. Investigation revealed that the X server started by the configure Script had not exited. The configure script starts/stops X servers multiple times. If this driver is installed, the X server does not exit at all ! So, I have added a check in the current code - if it finds that the X server doesn't exit in 30 seconds, it flags you to change the driver ! HTH -- Shree -----Original Message----- From: Kumar, Shree Sent: Friday, May 14, 2010 3:27 PM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch > Which version did you find worked? It might be okay for us to move to an older > driver version as a temporary workaround for this. My laptop is at home. I'll check the version and let you know on Monday. That driver wasn't perfect however : when extra pixels were configured, the X server would crash sometimes. > We also have some contacts at nvidia that we're going to talk to about this issue. Any influence helps :-) > Feel free to point anyone who's interested in our user experience with vizstack > at me - the more people using it, the easier it is for us to sell it as a viable > solution ;-) Thanks ! Regards -- Shree -----Original Message----- From: Sim...@cs... [mailto:Sim...@cs...] Sent: Friday, May 14, 2010 12:00 PM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch > -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Friday, 14 May 2010 4:20 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch > > > > Okay. It's quite frustrating to be so close to exactly what > we want, > > but falling short on something we don't have much power over . . . > > I am not happy with this too. I checked this on my laptop, > which also has nvidia graphics. The laptop has some older > drivers installed, and the bezel compensation does seem to > work well there. Let me hope nvidia can fix this. > Which version did you find worked? It might be okay for us to move to an older driver version as a temporary workaround for this. We also have some contacts at nvidia that we're going to talk to about this issue. > > I was thinking in terms of an app that draws a cross through each > > monitor, and lets you adjust each cross until it lines up with the > > corresponding cross on adjacent monitors - that would give you the > > offsets, which you could then add to the display config > file. I think > > you'd have to assume that monitors weren't tilted - unless > you really > > want to try and apply a totally arbitrary rotation to part of the > > display (which, given that most displays will have multiple > monitors driven by a single GPU, would be a real pain). > > Seems like a workable approach for desktops. This can be made > to setup using VizStack's Python API. > I'll check out the docs and see if I can come up with something. > >>> Oh, another thing that would be nice: some way to set up the > >>> remapping of outputs in the vs-manage-tiled-displays script. > >> > >> It really should have been there. Will fix this. > > > >Cool. > > I have put this in. If you update SVN, you will find the > needed changes. I have been modifying some interfaces > (chiefly the process launching ones), so if anything that was > working two days ago stops working now, do let me know ASAP ! > Testing now - I'll let you know. > I am looking forward to pushing out release 1.1 next week, so > any feedback would be very useful. Note that the > documentation is being updated & there is a user guide and > dev guide too. > > Post release 1.1, I'm going to try giving VizStack some more > publicity. I believe users of Paraview, VirtualGL/TurboVNC, > Equalizer and OpenSG will find Vizstack very useful ! > Feel free to point anyone who's interested in our user experience with vizstack at me - the more people using it, the easier it is for us to sell it as a viable solution ;-) Simon ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users |
|
From: Kumar, S. <shr...@hp...> - 2010-05-15 07:06:35
|
I checked the version of the nvidia driver on my laptop. Turns out its not old. The version number is : 195.36.15 . You may want to try with this driver, but YMMV. I tried on my machine with this, but faced some other problem. The configure script just hangs with this driver. Investigation revealed that the X server started by the configure Script had not exited. The configure script starts/stops X servers multiple times. If this driver is installed, the X server does not exit at all ! So, I have added a check in the current code - if it finds that the X server doesn't exit in 30 seconds, it flags you to change the driver ! HTH -- Shree -----Original Message----- From: Kumar, Shree Sent: Friday, May 14, 2010 3:27 PM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch > Which version did you find worked? It might be okay for us to move to an older > driver version as a temporary workaround for this. My laptop is at home. I'll check the version and let you know on Monday. That driver wasn't perfect however : when extra pixels were configured, the X server would crash sometimes. > We also have some contacts at nvidia that we're going to talk to about this issue. Any influence helps :-) > Feel free to point anyone who's interested in our user experience with vizstack > at me - the more people using it, the easier it is for us to sell it as a viable > solution ;-) Thanks ! Regards -- Shree -----Original Message----- From: Sim...@cs... [mailto:Sim...@cs...] Sent: Friday, May 14, 2010 12:00 PM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch > -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Friday, 14 May 2010 4:20 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch > > > > Okay. It's quite frustrating to be so close to exactly what > we want, > > but falling short on something we don't have much power over . . . > > I am not happy with this too. I checked this on my laptop, > which also has nvidia graphics. The laptop has some older > drivers installed, and the bezel compensation does seem to > work well there. Let me hope nvidia can fix this. > Which version did you find worked? It might be okay for us to move to an older driver version as a temporary workaround for this. We also have some contacts at nvidia that we're going to talk to about this issue. > > I was thinking in terms of an app that draws a cross through each > > monitor, and lets you adjust each cross until it lines up with the > > corresponding cross on adjacent monitors - that would give you the > > offsets, which you could then add to the display config > file. I think > > you'd have to assume that monitors weren't tilted - unless > you really > > want to try and apply a totally arbitrary rotation to part of the > > display (which, given that most displays will have multiple > monitors driven by a single GPU, would be a real pain). > > Seems like a workable approach for desktops. This can be made > to setup using VizStack's Python API. > I'll check out the docs and see if I can come up with something. > >>> Oh, another thing that would be nice: some way to set up the > >>> remapping of outputs in the vs-manage-tiled-displays script. > >> > >> It really should have been there. Will fix this. > > > >Cool. > > I have put this in. If you update SVN, you will find the > needed changes. I have been modifying some interfaces > (chiefly the process launching ones), so if anything that was > working two days ago stops working now, do let me know ASAP ! > Testing now - I'll let you know. > I am looking forward to pushing out release 1.1 next week, so > any feedback would be very useful. Note that the > documentation is being updated & there is a user guide and > dev guide too. > > Post release 1.1, I'm going to try giving VizStack some more > publicity. I believe users of Paraview, VirtualGL/TurboVNC, > Equalizer and OpenSG will find Vizstack very useful ! > Feel free to point anyone who's interested in our user experience with vizstack at me - the more people using it, the easier it is for us to sell it as a viable solution ;-) Simon ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users |
|
From: Kumar, S. <shr...@hp...> - 2010-05-14 09:58:08
|
> Which version did you find worked? It might be okay for us to move to an older > driver version as a temporary workaround for this. My laptop is at home. I'll check the version and let you know on Monday. That driver wasn't perfect however : when extra pixels were configured, the X server would crash sometimes. > We also have some contacts at nvidia that we're going to talk to about this issue. Any influence helps :-) > Feel free to point anyone who's interested in our user experience with vizstack > at me - the more people using it, the easier it is for us to sell it as a viable > solution ;-) Thanks ! Regards -- Shree -----Original Message----- From: Sim...@cs... [mailto:Sim...@cs...] Sent: Friday, May 14, 2010 12:00 PM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch > -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Friday, 14 May 2010 4:20 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch > > > > Okay. It's quite frustrating to be so close to exactly what > we want, > > but falling short on something we don't have much power over . . . > > I am not happy with this too. I checked this on my laptop, > which also has nvidia graphics. The laptop has some older > drivers installed, and the bezel compensation does seem to > work well there. Let me hope nvidia can fix this. > Which version did you find worked? It might be okay for us to move to an older driver version as a temporary workaround for this. We also have some contacts at nvidia that we're going to talk to about this issue. > > I was thinking in terms of an app that draws a cross through each > > monitor, and lets you adjust each cross until it lines up with the > > corresponding cross on adjacent monitors - that would give you the > > offsets, which you could then add to the display config > file. I think > > you'd have to assume that monitors weren't tilted - unless > you really > > want to try and apply a totally arbitrary rotation to part of the > > display (which, given that most displays will have multiple > monitors driven by a single GPU, would be a real pain). > > Seems like a workable approach for desktops. This can be made > to setup using VizStack's Python API. > I'll check out the docs and see if I can come up with something. > >>> Oh, another thing that would be nice: some way to set up the > >>> remapping of outputs in the vs-manage-tiled-displays script. > >> > >> It really should have been there. Will fix this. > > > >Cool. > > I have put this in. If you update SVN, you will find the > needed changes. I have been modifying some interfaces > (chiefly the process launching ones), so if anything that was > working two days ago stops working now, do let me know ASAP ! > Testing now - I'll let you know. > I am looking forward to pushing out release 1.1 next week, so > any feedback would be very useful. Note that the > documentation is being updated & there is a user guide and > dev guide too. > > Post release 1.1, I'm going to try giving VizStack some more > publicity. I believe users of Paraview, VirtualGL/TurboVNC, > Equalizer and OpenSG will find Vizstack very useful ! > Feel free to point anyone who's interested in our user experience with vizstack at me - the more people using it, the easier it is for us to sell it as a viable solution ;-) Simon ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users |
|
From: Kumar, S. <shr...@hp...> - 2010-05-14 09:54:25
|
Hi Simon, If you run a desktop which uses up _all_ the GPUs on the system, then you will not face such issues. What you are noticing is a contention for input devices between the X server and your console. When you run a desktop which does not use all GPUs on a system, you need a dedicated keyboard and a mouse for the session to avoid such conflicts. Let me explain: - your console is running all the time on a VT(virtual terminal). The console has access to the keyboard and mouse. - Using viz-desktop, when you start desktop on all the GPUs, the VT gets switched. This is similar to what happens when you run GDM, or if press CTRL-ALT-F7. The VT running the X server now has control of the keyboard and the mouse. - When you start a desktop on one or more (but not all) GPUs, the VT does not get switched. Not switching the VT is a requirement to run multiple X servers, which in turn is needed to share the node with multiple X servers. Since the VT does not switch, your console still has access to the keyboard and the mouse. That accounts for all the weirdness you see. In your case, the X server pops up on the same screen where your console is, causing even more confusion. To get one or more physical desktops working the way you want, you will need a separate physical keyboard/mouse per user. I have had best results with USB Keyboards and USB Mice. These need to be added into node_config.xml. Let me know if I need to describe the nitty gritty involved in this ! Regards -- Shree -----Original Message----- From: Sim...@cs... [mailto:Sim...@cs...] Sent: Friday, May 14, 2010 12:46 PM To: viz...@li... Subject: [vizstack-users] Yet another odd problem . . . A very odd problem . . . I've been running viz-desktop from an ssh session all the time, because when I run it from a console session, the first time I try to enter anything via the keyboard the display locks up - the mouse is still rendered normally, but nothing responds to it. I just found, while doing a bit of testing, that when I'm logged in at the console and I run a viz-desktop session via ssh the things that were typed into the session show up on the console as if they were typed in. So, I fire up viz-desktop, bring up a terminal, start up a vnc session, type in my password, and so on, and all those lines show up on the console. I just did a test where the only thing I brought up was firefox - I typed in a URL, hit enter, and the URL was in the bash history of the console. Just for fun I brought up a terminal and changed the prompt - when I closed the session the prompt in the console was also changed. Presumably this is some kind of issue with stdin/stdout not being redirected correctly . . . Simon Fowler Technical Specialist eResearch Visualisation team CSIRO IM&T Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871 ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users |
|
From: <Sim...@cs...> - 2010-05-14 07:15:54
|
A very odd problem . . . I've been running viz-desktop from an ssh session all the time, because when I run it from a console session, the first time I try to enter anything via the keyboard the display locks up - the mouse is still rendered normally, but nothing responds to it. I just found, while doing a bit of testing, that when I'm logged in at the console and I run a viz-desktop session via ssh the things that were typed into the session show up on the console as if they were typed in. So, I fire up viz-desktop, bring up a terminal, start up a vnc session, type in my password, and so on, and all those lines show up on the console. I just did a test where the only thing I brought up was firefox - I typed in a URL, hit enter, and the URL was in the bash history of the console. Just for fun I brought up a terminal and changed the prompt - when I closed the session the prompt in the console was also changed. Presumably this is some kind of issue with stdin/stdout not being redirected correctly . . . Simon Fowler Technical Specialist eResearch Visualisation team CSIRO IM&T Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871 |
|
From: <Sim...@cs...> - 2010-05-14 06:35:39
|
> -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Friday, 14 May 2010 4:20 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch <snipped> > >>> Oh, another thing that would be nice: some way to set up the > >>> remapping of outputs in the vs-manage-tiled-displays script. > >> > >> It really should have been there. Will fix this. > > > >Cool. > > I have put this in. If you update SVN, you will find the > needed changes. I have been modifying some interfaces > (chiefly the process launching ones), so if anything that was > working two days ago stops working now, do let me know ASAP ! > Tested and working properly - thanks yet again for the promptness of your response. Simon |
|
From: <Sim...@cs...> - 2010-05-14 06:30:25
|
> -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Friday, 14 May 2010 4:20 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch > > > > Okay. It's quite frustrating to be so close to exactly what > we want, > > but falling short on something we don't have much power over . . . > > I am not happy with this too. I checked this on my laptop, > which also has nvidia graphics. The laptop has some older > drivers installed, and the bezel compensation does seem to > work well there. Let me hope nvidia can fix this. > Which version did you find worked? It might be okay for us to move to an older driver version as a temporary workaround for this. We also have some contacts at nvidia that we're going to talk to about this issue. > > I was thinking in terms of an app that draws a cross through each > > monitor, and lets you adjust each cross until it lines up with the > > corresponding cross on adjacent monitors - that would give you the > > offsets, which you could then add to the display config > file. I think > > you'd have to assume that monitors weren't tilted - unless > you really > > want to try and apply a totally arbitrary rotation to part of the > > display (which, given that most displays will have multiple > monitors driven by a single GPU, would be a real pain). > > Seems like a workable approach for desktops. This can be made > to setup using VizStack's Python API. > I'll check out the docs and see if I can come up with something. > >>> Oh, another thing that would be nice: some way to set up the > >>> remapping of outputs in the vs-manage-tiled-displays script. > >> > >> It really should have been there. Will fix this. > > > >Cool. > > I have put this in. If you update SVN, you will find the > needed changes. I have been modifying some interfaces > (chiefly the process launching ones), so if anything that was > working two days ago stops working now, do let me know ASAP ! > Testing now - I'll let you know. > I am looking forward to pushing out release 1.1 next week, so > any feedback would be very useful. Note that the > documentation is being updated & there is a user guide and > dev guide too. > > Post release 1.1, I'm going to try giving VizStack some more > publicity. I believe users of Paraview, VirtualGL/TurboVNC, > Equalizer and OpenSG will find Vizstack very useful ! > Feel free to point anyone who's interested in our user experience with vizstack at me - the more people using it, the easier it is for us to sell it as a viable solution ;-) Simon |
|
From: Kumar, S. <shr...@hp...> - 2010-05-14 06:21:59
|
> Okay. It's quite frustrating to be so close to exactly what we want, but falling > short on something we don't have much power over . . . I am not happy with this too. I checked this on my laptop, which also has nvidia graphics. The laptop has some older drivers installed, and the bezel compensation does seem to work well there. Let me hope nvidia can fix this. > I was thinking in terms of an app that draws a cross through each monitor, and > lets you adjust each cross until it lines up with the corresponding cross on > adjacent monitors - that would give you the offsets, which you could then add > to the display config file. I think you'd have to assume that monitors weren't > tilted - unless you really want to try and apply a totally arbitrary rotation to > part of the display (which, given that most displays will have multiple monitors > driven by a single GPU, would be a real pain). Seems like a workable approach for desktops. This can be made to setup using VizStack's Python API. >>> Oh, another thing that would be nice: some way to set up >>> the remapping of outputs in the vs-manage-tiled-displays script. >> >> It really should have been there. Will fix this. > >Cool. I have put this in. If you update SVN, you will find the needed changes. I have been modifying some interfaces (chiefly the process launching ones), so if anything that was working two days ago stops working now, do let me know ASAP ! I am looking forward to pushing out release 1.1 next week, so any feedback would be very useful. Note that the documentation is being updated & there is a user guide and dev guide too. Post release 1.1, I'm going to try giving VizStack some more publicity. I believe users of Paraview, VirtualGL/TurboVNC, Equalizer and OpenSG will find Vizstack very useful ! Regards -- Shree -----Original Message----- From: Sim...@cs... [mailto:Sim...@cs...] Sent: Thursday, May 13, 2010 9:31 AM To: viz...@li... Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch > -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Thursday, 13 May 2010 1:30 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch > > > > The only problem now is that as the mouse moves through the hidden > > pixels on the edges of each X screen the screen still > scrolls - this > > doesn't make it totally unusable, but if it's at all > possible it should probably be avoided . . . > > This really is the nvidia driver issue that I mentioned in > passing earlier. It really shouldn't be panning the display - > but it is. I'll report this to nvidia and see what they think of this. > Okay. It's quite frustrating to be so close to exactly what we want, but falling short on something we don't have much power over . . . > > On the subject of misalignment of the monitors, it'd be > nice to have > > some way to specify small offsets on a per-monitor basis, to adjust > > for those minor physical alignment issues. It's probably > right at the > > bottom of the list of priorities, but for real-world tiled displays > > (particularly ones that aren't bolted to walls) it'd be a > useful additional feature. > > Yeah, we have this problem too; our tiled display isn't > perfect. But that said, I'm not sure what the best way to > support this would be. > > VizStack aims to make it possible to run any software. > Unfortunately, they all have different capabilities. Some > assume a strict rectangular arrangement of tiles, some cannot > handle mullions at all, some assume all tiles have the same size. > > E.g. ParaView can handle mullions on its own. However, it assumes that > 1. the horizontal mullions are equal to each other > 2. the vertical mullions are equal to each other > > Note that (2) is violated in your case. So the only way to > setup ParaView on your kind of environment is by the kind of > mechanism I've provided now. If you run ParaView on your > tiled display, you'd find that it compensates for bezels > appropriately. (Ok, I do remember GLX is broken in your > site, so ParaView may not run) > All the monitors I've looked at have different bezel widths all around - some are consistent across the vertical/horizontal pairs, but there generally seems to be variations all round. > Another thing I've noticed with our tiled display is that > some tiles have a little bit of tilt. That can really > complicate calculations. Overall, this is perhaps quite low > in the list of priorities. > I was thinking in terms of an app that draws a cross through each monitor, and lets you adjust each cross until it lines up with the corresponding cross on adjacent monitors - that would give you the offsets, which you could then add to the display config file. I think you'd have to assume that monitors weren't tilted - unless you really want to try and apply a totally arbitrary rotation to part of the display (which, given that most displays will have multiple monitors driven by a single GPU, would be a real pain). > > Oh, another thing that would be nice: some way to set up > the remapping > > of outputs in the vs-manage-tiled-displays script. > > It really should have been there. Will fix this. > Cool. Simon ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users |
|
From: <Sim...@cs...> - 2010-05-13 04:01:36
|
> -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Thursday, 13 May 2010 1:30 PM > To: viz...@li... > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem > running vs-configure-standalone from shree branch > > > > The only problem now is that as the mouse moves through the hidden > > pixels on the edges of each X screen the screen still > scrolls - this > > doesn't make it totally unusable, but if it's at all > possible it should probably be avoided . . . > > This really is the nvidia driver issue that I mentioned in > passing earlier. It really shouldn't be panning the display - > but it is. I'll report this to nvidia and see what they think of this. > Okay. It's quite frustrating to be so close to exactly what we want, but falling short on something we don't have much power over . . . > > On the subject of misalignment of the monitors, it'd be > nice to have > > some way to specify small offsets on a per-monitor basis, to adjust > > for those minor physical alignment issues. It's probably > right at the > > bottom of the list of priorities, but for real-world tiled displays > > (particularly ones that aren't bolted to walls) it'd be a > useful additional feature. > > Yeah, we have this problem too; our tiled display isn't > perfect. But that said, I'm not sure what the best way to > support this would be. > > VizStack aims to make it possible to run any software. > Unfortunately, they all have different capabilities. Some > assume a strict rectangular arrangement of tiles, some cannot > handle mullions at all, some assume all tiles have the same size. > > E.g. ParaView can handle mullions on its own. However, it assumes that > 1. the horizontal mullions are equal to each other > 2. the vertical mullions are equal to each other > > Note that (2) is violated in your case. So the only way to > setup ParaView on your kind of environment is by the kind of > mechanism I've provided now. If you run ParaView on your > tiled display, you'd find that it compensates for bezels > appropriately. (Ok, I do remember GLX is broken in your > site, so ParaView may not run) > All the monitors I've looked at have different bezel widths all around - some are consistent across the vertical/horizontal pairs, but there generally seems to be variations all round. > Another thing I've noticed with our tiled display is that > some tiles have a little bit of tilt. That can really > complicate calculations. Overall, this is perhaps quite low > in the list of priorities. > I was thinking in terms of an app that draws a cross through each monitor, and lets you adjust each cross until it lines up with the corresponding cross on adjacent monitors - that would give you the offsets, which you could then add to the display config file. I think you'd have to assume that monitors weren't tilted - unless you really want to try and apply a totally arbitrary rotation to part of the display (which, given that most displays will have multiple monitors driven by a single GPU, would be a real pain). > > Oh, another thing that would be nice: some way to set up > the remapping > > of outputs in the vs-manage-tiled-displays script. > > It really should have been there. Will fix this. > Cool. Simon |
|
From: Kumar, S. <shr...@hp...> - 2010-05-13 03:32:16
|
> The only problem now is that as the mouse moves through the hidden
> pixels on the edges of each X screen the screen still scrolls - this
> doesn't make it totally unusable, but if it's at all possible it should probably be avoided . . .
This really is the nvidia driver issue that I mentioned in passing earlier. It really
shouldn't be panning the display - but it is. I'll report this to nvidia and see what
they think of this.
> On the subject of misalignment of the monitors, it'd be nice to have some way to
> specify small offsets on a per-monitor basis, to adjust for those minor physical
> alignment issues. It's probably right at the bottom of the list of priorities, but
> for real-world tiled displays (particularly ones that aren't bolted to walls) it'd
> be a useful additional feature.
Yeah, we have this problem too; our tiled display isn't perfect. But that said, I'm
not sure what the best way to support this would be.
VizStack aims to make it possible to run any software. Unfortunately, they all have
different capabilities. Some assume a strict rectangular arrangement of tiles, some
cannot handle mullions at all, some assume all tiles have the same size.
E.g. ParaView can handle mullions on its own. However, it assumes that
1. the horizontal mullions are equal to each other
2. the vertical mullions are equal to each other
Note that (2) is violated in your case. So the only way to setup ParaView on your
kind of environment is by the kind of mechanism I've provided now. If you run ParaView
on your tiled display, you'd find that it compensates for bezels appropriately. (Ok,
I do remember GLX is broken in your site, so ParaView may not run)
Another thing I've noticed with our tiled display is that some tiles have a little
bit of tilt. That can really complicate calculations. Overall, this is perhaps
quite low in the list of priorities.
> Oh, another thing that would be nice: some way to set up the remapping of outputs
> in the vs-manage-tiled-displays script.
It really should have been there. Will fix this.
> Of course - I imagine Dell will like that, too ;-)
Thanks. Will add this.
Cheers
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Thursday, May 13, 2010 6:12 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Well, having brought it up for testing it looks like the bezel calculcations are all good now - I'm getting the correct resolution, and visually it looks good (barring minor misalignment of the monitors, which needs to be fixed physically rather than in software).
The only problem now is that as the mouse moves through the hidden pixels on the edges of each X screen the screen still scrolls - this doesn't make it totally unusable, but if it's at all possible it should probably be avoided . . .
On the subject of misalignment of the monitors, it'd be nice to have some way to specify small offsets on a per-monitor basis, to adjust for those minor physical alignment issues. It's probably right at the bottom of the list of priorities, but for real-world tiled displays (particularly ones that aren't bolted to walls) it'd be a useful additional feature.
Simon
> -----Original Message-----
> From: Kumar, Shree [mailto:shr...@hp...]
> Sent: Thursday, 13 May 2010 12:33 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Hi Simon,
>
> I think I have fixed this problem. I have manually tested
> your scenario and verified the results.
>
> Please test out the 'shree' branch again. Your desktop should now be
> - 7680x3200 pixels if pixel skipping is disabled. This is
> the default
> - 8104x3422 pixels if pixel skipping is enabled. You may do
> this by passing "-B" on the command line to viz-desktop
>
> Note that I have added a template for the Quadro NVS 420 into
> VizStack itself. I am also interested to add the template for
> you Dell display device. Would you be fine with me including
> your template into VizStack ?
>
> Thanks!
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, April 28, 2010 11:40 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Here it is.
>
> I'll drop back to the trunk for the moment - plenty of stuff
> I can work on that doesn't rely on the new features. Let me
> know when you need some more testing.
>
> Simon
> ________________________________________
> From: Kumar, Shree [shr...@hp...]
> Sent: Wednesday, April 28, 2010 2:52 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Mistake in calculations. Perhaps I made a hasty change
> without considering all cases right. Please send me the EDID
> file for your monitor also. That allows me to test without
> having your monitor.
>
> Give me a week; there are other things I need to sort. I
> would recommend you shift to the main branch while I do some
> more testing !
>
> Thanks
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, April 28, 2010 9:31 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Attached the template - I built it by editing the LP3065
> template, so it should be fine. The screen dimensions come
> from the EDID, and the bezels were directly measured. The 3x2
> display ends up being 8072x3298. I've also attached the
> generated xorg.conf file.
>
> Aha, after a bit more careful testing it looks like the
> scrolling happens across all the bezels - the same thing I'd
> expect if it was three virtual desktops sized to take the
> bezels into account.
>
> A command line option to viz-desktop would be fine, or an
> option for the tiled display config so that you could have a
> 3x2-full and 3x2-no-bezels.
>
> Simon
> ________________________________________
> From: Kumar, Shree [shr...@hp...]
> Sent: Wednesday, April 28, 2010 1:35 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Hi Simon,
>
> Good to know things work for you. Now to fix the bezel problem.
>
> > Two issues: firstly, the bezel width calculations are off - they're
> > hiding significantly less pixels than they should be (maybe
> around three quarters - could it be a 75dpi vs 100dpi thing?).
>
> Can you send me your display template ? Please note that all
> dimensions are needed in millimeters. If in doubt, look at
> the template for the "HP LP3065". This is our 30" monitor,
> and you would have comparable dimensions.
>
> > Secondly, it's not taking into account the outside bezels - they
> > shouldn't hide any pixels, so that there's no scrolling of
> the virtual desktop when you move the pointer to the edges of
> the tiled display.
>
> The code is supposed to handle these cases.
>
> > Oh, and it would be nice if we could have a way to define a tiled
> > display that didn't hide any pixels - there are some types
> of data where you want to see every pixel.
>
> Good catch. Should make this a command line option to the
> viz-desktop script. (that's what you are using, right?)
>
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, April 28, 2010 7:58 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Actually, I tell a lie - after fixing up the template for my
> monitors it's working, after a fashion.
>
> Two issues: firstly, the bezel width calculations are off -
> they're hiding significantly less pixels than they should be
> (maybe around three quarters - could it be a 75dpi vs 100dpi
> thing?). Secondly, it's not taking into account the outside
> bezels - they shouldn't hide any pixels, so that there's no
> scrolling of the virtual desktop when you move the pointer to
> the edges of the tiled display.
>
> Oh, and it would be nice if we could have a way to define a
> tiled display that didn't hide any pixels - there are some
> types of data where you want to see every pixel.
>
> Those issues aside it's great ;-) Thanks for the very prompt fixes.
>
> Simon
>
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Wednesday, 28 April 2010 12:09 PM
> > To: viz...@li...
> > Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> > vs-configure-standalone from shree branch
> >
> > With that workaround I can bring up the desktop as before,
> though it's
> > not handling the bezels with this setup - is there
> something I have to
> > do to get that working?
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 5:14 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > >
> > > You won't be able to run 3D applications if GLX doesn't work.
> > >
> > > For now, you can replace the lines
> > > ----
> > > limitData = open(temp_result_file,'r').read() limitData =
> > > limitData.lstrip().rstrip().split(' ')
> > > ----
> > > With
> > > ----
> > > try:
> > > limitData = open(temp_result_file,'r').read()
> > > limitData = limitData.lstrip().rstrip().split(' ')
> > > except:
> > > limitData = [4096, 4096]
> > > ----
> > > Remember to keep the indentation right :-)
> > >
> > > Also note one more thing : the display template is slightly
> > changed.
> > > So please adjust your template appropriately.
> > >
> > > HTH
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 12:36 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > As I said in an early post, GLX is misbehaving somehow -
> > the extension
> > > loads and is apparently available, but any attempt to
> create a GLX
> > > context fails. With earlier versions of VizStack that
> > didn't cause any
> > > problems, but now that you're relying on GLX to figure out
> > things like
> > > this it's causing problems.
> > >
> > > I'm trying to get it working now.
> > >
> > > Simon
> > >
> > > > -----Original Message-----
> > > > From: Kumar, Shree [mailto:shr...@hp...]
> > > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > > To: viz...@li...
> > > > Subject: Re: [vizstack-users] Problem running
> > > vs-configure-standalone
> > > > from shree branch
> > > >
> > > > Hi Simon,
> > > >
> > > > Does GLX not work at all at your site ?
> > > >
> > > > vs-get-limits is trying to find the maximum frambuffer
> > size that is
> > > > supported by your GPU.
> > > >
> > > > Regards
> > > > -- Shree
> > > >
> > > > -----Original Message-----
> > > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > > To: viz...@li...
> > > > Subject: [vizstack-users] Problem running
> > > vs-configure-standalone from
> > > > shree branch
> > > >
> > > > I'm trying out the head of the shree branch, and finding that
> > > > vs-configure-standalone fails with the following output:
> > > >
> > > > root@smlopt1-display:~#
> /opt/vizstack/sbin/vs-configure-standalone
> > > >
> > > > Processing Node 'localhost'...
> > > > Errors happened while trying to get the configuration of node
> > > > 'localhost'. Reason:
> > > >
> > > > Traceback (most recent call last):
> > > > File "/opt/vizstack/sbin/vs-generate-node-config",
> line 489, in
> > > > <module>
> > > > limitData = open(temp_result_file,'r').read()
> > > > IOError: [Errno 2] No such file or directory:
> > > > '/tmp/temp-vs-configure-system/temp-result-file'
> > > >
> > > >
> > > > Please fix the above errors & run this tool again Failed to
> > > configure
> > > > standalone configuration
> > > >
> > > >
> > > > Looking at the code it looks like vs-get-limits is failing
> > > to produce
> > > > an output file. I'm pretty sure this is due to the issue I
> > > noted early
> > > > on about GLX not working on this setup (there's no output from
> > > > vs-get-limits, but looking at the code it's not going to
> > > work unless
> > > > GLX is behaving itself).
> > > >
> > > > What's odd is that vs-get-limits apparently isn't returning
> > > an error -
> > > > I put a few debugging prints in vs-generate-node-config,
> > > and it looks
> > > > like it's following the 'if ret == 0:' path every time it's
> > > run to get
> > > > to the point it fails (line 489).
> > > >
> > > > I'll have another shot at getting GLX to behave, but this
> > > failure mode
> > > > is somewhat problematic . . .
> > > >
> > > > Simon Fowler
> > > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > > --------------------------------------------------------------
> > > > ----------------
> > > > _______________________________________________
> > > > vizstack-users mailing list
> > > > viz...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > >
> > > > --------------------------------------------------------------
> > > > ----------------
> > > > _______________________________________________
> > > > vizstack-users mailing list
> > > > viz...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
>
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: <Sim...@cs...> - 2010-05-13 02:37:25
|
Oh, another thing that would be nice: some way to set up the remapping of outputs in the vs-manage-tiled-displays script.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Thursday, 13 May 2010 10:42 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Well, having brought it up for testing it looks like the
> bezel calculcations are all good now - I'm getting the
> correct resolution, and visually it looks good (barring minor
> misalignment of the monitors, which needs to be fixed
> physically rather than in software).
>
> The only problem now is that as the mouse moves through the
> hidden pixels on the edges of each X screen the screen still
> scrolls - this doesn't make it totally unusable, but if it's
> at all possible it should probably be avoided . . .
>
> On the subject of misalignment of the monitors, it'd be nice
> to have some way to specify small offsets on a per-monitor
> basis, to adjust for those minor physical alignment issues.
> It's probably right at the bottom of the list of priorities,
> but for real-world tiled displays (particularly ones that
> aren't bolted to walls) it'd be a useful additional feature.
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Thursday, 13 May 2010 12:33 AM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running
> > vs-configure-standalone from shree branch
> >
> > Hi Simon,
> >
> > I think I have fixed this problem. I have manually tested your
> > scenario and verified the results.
> >
> > Please test out the 'shree' branch again. Your desktop should now be
> > - 7680x3200 pixels if pixel skipping is disabled. This is the
> > default
> > - 8104x3422 pixels if pixel skipping is enabled. You may
> do this by
> > passing "-B" on the command line to viz-desktop
> >
> > Note that I have added a template for the Quadro NVS 420
> into VizStack
> > itself. I am also interested to add the template for you
> Dell display
> > device. Would you be fine with me including your template into
> > VizStack ?
> >
> > Thanks!
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Wednesday, April 28, 2010 11:40 AM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running
> > vs-configure-standalone from shree branch
> >
> > Here it is.
> >
> > I'll drop back to the trunk for the moment - plenty of stuff I can
> > work on that doesn't rely on the new features. Let me know when you
> > need some more testing.
> >
> > Simon
> > ________________________________________
> > From: Kumar, Shree [shr...@hp...]
> > Sent: Wednesday, April 28, 2010 2:52 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> > running vs-configure-standalone from shree branch
> >
> > Mistake in calculations. Perhaps I made a hasty change without
> > considering all cases right. Please send me the EDID file for your
> > monitor also. That allows me to test without having your monitor.
> >
> > Give me a week; there are other things I need to sort. I would
> > recommend you shift to the main branch while I do some more
> testing !
> >
> > Thanks
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Wednesday, April 28, 2010 9:31 AM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running
> > vs-configure-standalone from shree branch
> >
> > Attached the template - I built it by editing the LP3065
> template, so
> > it should be fine. The screen dimensions come from the
> EDID, and the
> > bezels were directly measured. The 3x2 display ends up being
> > 8072x3298. I've also attached the generated xorg.conf file.
> >
> > Aha, after a bit more careful testing it looks like the scrolling
> > happens across all the bezels - the same thing I'd expect if it was
> > three virtual desktops sized to take the bezels into account.
> >
> > A command line option to viz-desktop would be fine, or an
> option for
> > the tiled display config so that you could have a 3x2-full and
> > 3x2-no-bezels.
> >
> > Simon
> > ________________________________________
> > From: Kumar, Shree [shr...@hp...]
> > Sent: Wednesday, April 28, 2010 1:35 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> > running vs-configure-standalone from shree branch
> >
> > Hi Simon,
> >
> > Good to know things work for you. Now to fix the bezel problem.
> >
> > > Two issues: firstly, the bezel width calculations are off
> - they're
> > > hiding significantly less pixels than they should be (maybe
> > around three quarters - could it be a 75dpi vs 100dpi thing?).
> >
> > Can you send me your display template ? Please note that all
> > dimensions are needed in millimeters. If in doubt, look at the
> > template for the "HP LP3065". This is our 30" monitor, and
> you would
> > have comparable dimensions.
> >
> > > Secondly, it's not taking into account the outside bezels - they
> > > shouldn't hide any pixels, so that there's no scrolling of
> > the virtual desktop when you move the pointer to the edges of the
> > tiled display.
> >
> > The code is supposed to handle these cases.
> >
> > > Oh, and it would be nice if we could have a way to define a tiled
> > > display that didn't hide any pixels - there are some types
> > of data where you want to see every pixel.
> >
> > Good catch. Should make this a command line option to the
> viz-desktop
> > script. (that's what you are using, right?)
> >
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Wednesday, April 28, 2010 7:58 AM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running
> > vs-configure-standalone from shree branch
> >
> > Actually, I tell a lie - after fixing up the template for
> my monitors
> > it's working, after a fashion.
> >
> > Two issues: firstly, the bezel width calculations are off - they're
> > hiding significantly less pixels than they should be (maybe around
> > three quarters - could it be a 75dpi vs 100dpi thing?).
> Secondly, it's
> > not taking into account the outside bezels - they shouldn't
> hide any
> > pixels, so that there's no scrolling of the virtual desktop
> when you
> > move the pointer to the edges of the tiled display.
> >
> > Oh, and it would be nice if we could have a way to define a tiled
> > display that didn't hide any pixels - there are some types of data
> > where you want to see every pixel.
> >
> > Those issues aside it's great ;-) Thanks for the very prompt fixes.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Wednesday, 28 April 2010 12:09 PM
> > > To: viz...@li...
> > > Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> > > vs-configure-standalone from shree branch
> > >
> > > With that workaround I can bring up the desktop as before,
> > though it's
> > > not handling the bezels with this setup - is there
> > something I have to
> > > do to get that working?
> > >
> > > Simon
> > >
> > > > -----Original Message-----
> > > > From: Kumar, Shree [mailto:shr...@hp...]
> > > > Sent: Tuesday, 27 April 2010 5:14 PM
> > > > To: viz...@li...
> > > > Subject: Re: [vizstack-users] Problem running
> > > vs-configure-standalone
> > > > from shree branch
> > > >
> > > >
> > > > You won't be able to run 3D applications if GLX doesn't work.
> > > >
> > > > For now, you can replace the lines
> > > > ----
> > > > limitData = open(temp_result_file,'r').read() limitData =
> > > > limitData.lstrip().rstrip().split(' ')
> > > > ----
> > > > With
> > > > ----
> > > > try:
> > > > limitData = open(temp_result_file,'r').read()
> > > > limitData = limitData.lstrip().rstrip().split(' ')
> > > > except:
> > > > limitData = [4096, 4096]
> > > > ----
> > > > Remember to keep the indentation right :-)
> > > >
> > > > Also note one more thing : the display template is slightly
> > > changed.
> > > > So please adjust your template appropriately.
> > > >
> > > > HTH
> > > > -- Shree
> > > >
> > > > -----Original Message-----
> > > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > > Sent: Tuesday, April 27, 2010 12:36 PM
> > > > To: viz...@li...
> > > > Subject: Re: [vizstack-users] Problem running
> > > vs-configure-standalone
> > > > from shree branch
> > > >
> > > > As I said in an early post, GLX is misbehaving somehow -
> > > the extension
> > > > loads and is apparently available, but any attempt to
> > create a GLX
> > > > context fails. With earlier versions of VizStack that
> > > didn't cause any
> > > > problems, but now that you're relying on GLX to figure out
> > > things like
> > > > this it's causing problems.
> > > >
> > > > I'm trying to get it working now.
> > > >
> > > > Simon
> > > >
> > > > > -----Original Message-----
> > > > > From: Kumar, Shree [mailto:shr...@hp...]
> > > > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > > > To: viz...@li...
> > > > > Subject: Re: [vizstack-users] Problem running
> > > > vs-configure-standalone
> > > > > from shree branch
> > > > >
> > > > > Hi Simon,
> > > > >
> > > > > Does GLX not work at all at your site ?
> > > > >
> > > > > vs-get-limits is trying to find the maximum frambuffer
> > > size that is
> > > > > supported by your GPU.
> > > > >
> > > > > Regards
> > > > > -- Shree
> > > > >
> > > > > -----Original Message-----
> > > > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > > > To: viz...@li...
> > > > > Subject: [vizstack-users] Problem running
> > > > vs-configure-standalone from
> > > > > shree branch
> > > > >
> > > > > I'm trying out the head of the shree branch, and finding that
> > > > > vs-configure-standalone fails with the following output:
> > > > >
> > > > > root@smlopt1-display:~#
> > /opt/vizstack/sbin/vs-configure-standalone
> > > > >
> > > > > Processing Node 'localhost'...
> > > > > Errors happened while trying to get the configuration of node
> > > > > 'localhost'. Reason:
> > > > >
> > > > > Traceback (most recent call last):
> > > > > File "/opt/vizstack/sbin/vs-generate-node-config",
> > line 489, in
> > > > > <module>
> > > > > limitData = open(temp_result_file,'r').read()
> > > > > IOError: [Errno 2] No such file or directory:
> > > > > '/tmp/temp-vs-configure-system/temp-result-file'
> > > > >
> > > > >
> > > > > Please fix the above errors & run this tool again Failed to
> > > > configure
> > > > > standalone configuration
> > > > >
> > > > >
> > > > > Looking at the code it looks like vs-get-limits is failing
> > > > to produce
> > > > > an output file. I'm pretty sure this is due to the issue I
> > > > noted early
> > > > > on about GLX not working on this setup (there's no
> output from
> > > > > vs-get-limits, but looking at the code it's not going to
> > > > work unless
> > > > > GLX is behaving itself).
> > > > >
> > > > > What's odd is that vs-get-limits apparently isn't returning
> > > > an error -
> > > > > I put a few debugging prints in vs-generate-node-config,
> > > > and it looks
> > > > > like it's following the 'if ret == 0:' path every time it's
> > > > run to get
> > > > > to the point it fails (line 489).
> > > > >
> > > > > I'll have another shot at getting GLX to behave, but this
> > > > failure mode
> > > > > is somewhat problematic . . .
> > > > >
> > > > > Simon Fowler
> > > > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > > > --------------------------------------------------------------
> > > > > ----------------
> > > > > _______________________________________________
> > > > > vizstack-users mailing list
> > > > > viz...@li...
> > > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > > >
> > > > > --------------------------------------------------------------
> > > > > ----------------
> > > > > _______________________________________________
> > > > > vizstack-users mailing list
> > > > > viz...@li...
> > > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > > >
> > > > --------------------------------------------------------------
> > > > ----------------
> > > > _______________________________________________
> > > > vizstack-users mailing list
> > > > viz...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > >
> > > > --------------------------------------------------------------
> > > > ----------------
> > > > _______________________________________________
> > > > vizstack-users mailing list
> > > > viz...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> >
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
>
> --------------------------------------------------------------
> ----------------
>
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
|
|
From: <Sim...@cs...> - 2010-05-13 00:42:41
|
Well, having brought it up for testing it looks like the bezel calculcations are all good now - I'm getting the correct resolution, and visually it looks good (barring minor misalignment of the monitors, which needs to be fixed physically rather than in software).
The only problem now is that as the mouse moves through the hidden pixels on the edges of each X screen the screen still scrolls - this doesn't make it totally unusable, but if it's at all possible it should probably be avoided . . .
On the subject of misalignment of the monitors, it'd be nice to have some way to specify small offsets on a per-monitor basis, to adjust for those minor physical alignment issues. It's probably right at the bottom of the list of priorities, but for real-world tiled displays (particularly ones that aren't bolted to walls) it'd be a useful additional feature.
Simon
> -----Original Message-----
> From: Kumar, Shree [mailto:shr...@hp...]
> Sent: Thursday, 13 May 2010 12:33 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Hi Simon,
>
> I think I have fixed this problem. I have manually tested
> your scenario and verified the results.
>
> Please test out the 'shree' branch again. Your desktop should now be
> - 7680x3200 pixels if pixel skipping is disabled. This is
> the default
> - 8104x3422 pixels if pixel skipping is enabled. You may do
> this by passing "-B" on the command line to viz-desktop
>
> Note that I have added a template for the Quadro NVS 420 into
> VizStack itself. I am also interested to add the template for
> you Dell display device. Would you be fine with me including
> your template into VizStack ?
>
> Thanks!
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, April 28, 2010 11:40 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Here it is.
>
> I'll drop back to the trunk for the moment - plenty of stuff
> I can work on that doesn't rely on the new features. Let me
> know when you need some more testing.
>
> Simon
> ________________________________________
> From: Kumar, Shree [shr...@hp...]
> Sent: Wednesday, April 28, 2010 2:52 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Mistake in calculations. Perhaps I made a hasty change
> without considering all cases right. Please send me the EDID
> file for your monitor also. That allows me to test without
> having your monitor.
>
> Give me a week; there are other things I need to sort. I
> would recommend you shift to the main branch while I do some
> more testing !
>
> Thanks
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, April 28, 2010 9:31 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Attached the template - I built it by editing the LP3065
> template, so it should be fine. The screen dimensions come
> from the EDID, and the bezels were directly measured. The 3x2
> display ends up being 8072x3298. I've also attached the
> generated xorg.conf file.
>
> Aha, after a bit more careful testing it looks like the
> scrolling happens across all the bezels - the same thing I'd
> expect if it was three virtual desktops sized to take the
> bezels into account.
>
> A command line option to viz-desktop would be fine, or an
> option for the tiled display config so that you could have a
> 3x2-full and 3x2-no-bezels.
>
> Simon
> ________________________________________
> From: Kumar, Shree [shr...@hp...]
> Sent: Wednesday, April 28, 2010 1:35 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Hi Simon,
>
> Good to know things work for you. Now to fix the bezel problem.
>
> > Two issues: firstly, the bezel width calculations are off - they're
> > hiding significantly less pixels than they should be (maybe
> around three quarters - could it be a 75dpi vs 100dpi thing?).
>
> Can you send me your display template ? Please note that all
> dimensions are needed in millimeters. If in doubt, look at
> the template for the "HP LP3065". This is our 30" monitor,
> and you would have comparable dimensions.
>
> > Secondly, it's not taking into account the outside bezels - they
> > shouldn't hide any pixels, so that there's no scrolling of
> the virtual desktop when you move the pointer to the edges of
> the tiled display.
>
> The code is supposed to handle these cases.
>
> > Oh, and it would be nice if we could have a way to define a tiled
> > display that didn't hide any pixels - there are some types
> of data where you want to see every pixel.
>
> Good catch. Should make this a command line option to the
> viz-desktop script. (that's what you are using, right?)
>
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, April 28, 2010 7:58 AM
> To: viz...@li...
> Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem
> running vs-configure-standalone from shree branch
>
> Actually, I tell a lie - after fixing up the template for my
> monitors it's working, after a fashion.
>
> Two issues: firstly, the bezel width calculations are off -
> they're hiding significantly less pixels than they should be
> (maybe around three quarters - could it be a 75dpi vs 100dpi
> thing?). Secondly, it's not taking into account the outside
> bezels - they shouldn't hide any pixels, so that there's no
> scrolling of the virtual desktop when you move the pointer to
> the edges of the tiled display.
>
> Oh, and it would be nice if we could have a way to define a
> tiled display that didn't hide any pixels - there are some
> types of data where you want to see every pixel.
>
> Those issues aside it's great ;-) Thanks for the very prompt fixes.
>
> Simon
>
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Wednesday, 28 April 2010 12:09 PM
> > To: viz...@li...
> > Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> > vs-configure-standalone from shree branch
> >
> > With that workaround I can bring up the desktop as before,
> though it's
> > not handling the bezels with this setup - is there
> something I have to
> > do to get that working?
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 5:14 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > >
> > > You won't be able to run 3D applications if GLX doesn't work.
> > >
> > > For now, you can replace the lines
> > > ----
> > > limitData = open(temp_result_file,'r').read() limitData =
> > > limitData.lstrip().rstrip().split(' ')
> > > ----
> > > With
> > > ----
> > > try:
> > > limitData = open(temp_result_file,'r').read()
> > > limitData = limitData.lstrip().rstrip().split(' ')
> > > except:
> > > limitData = [4096, 4096]
> > > ----
> > > Remember to keep the indentation right :-)
> > >
> > > Also note one more thing : the display template is slightly
> > changed.
> > > So please adjust your template appropriately.
> > >
> > > HTH
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 12:36 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > As I said in an early post, GLX is misbehaving somehow -
> > the extension
> > > loads and is apparently available, but any attempt to
> create a GLX
> > > context fails. With earlier versions of VizStack that
> > didn't cause any
> > > problems, but now that you're relying on GLX to figure out
> > things like
> > > this it's causing problems.
> > >
> > > I'm trying to get it working now.
> > >
> > > Simon
> > >
> > > > -----Original Message-----
> > > > From: Kumar, Shree [mailto:shr...@hp...]
> > > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > > To: viz...@li...
> > > > Subject: Re: [vizstack-users] Problem running
> > > vs-configure-standalone
> > > > from shree branch
> > > >
> > > > Hi Simon,
> > > >
> > > > Does GLX not work at all at your site ?
> > > >
> > > > vs-get-limits is trying to find the maximum frambuffer
> > size that is
> > > > supported by your GPU.
> > > >
> > > > Regards
> > > > -- Shree
> > > >
> > > > -----Original Message-----
> > > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > > To: viz...@li...
> > > > Subject: [vizstack-users] Problem running
> > > vs-configure-standalone from
> > > > shree branch
> > > >
> > > > I'm trying out the head of the shree branch, and finding that
> > > > vs-configure-standalone fails with the following output:
> > > >
> > > > root@smlopt1-display:~#
> /opt/vizstack/sbin/vs-configure-standalone
> > > >
> > > > Processing Node 'localhost'...
> > > > Errors happened while trying to get the configuration of node
> > > > 'localhost'. Reason:
> > > >
> > > > Traceback (most recent call last):
> > > > File "/opt/vizstack/sbin/vs-generate-node-config",
> line 489, in
> > > > <module>
> > > > limitData = open(temp_result_file,'r').read()
> > > > IOError: [Errno 2] No such file or directory:
> > > > '/tmp/temp-vs-configure-system/temp-result-file'
> > > >
> > > >
> > > > Please fix the above errors & run this tool again Failed to
> > > configure
> > > > standalone configuration
> > > >
> > > >
> > > > Looking at the code it looks like vs-get-limits is failing
> > > to produce
> > > > an output file. I'm pretty sure this is due to the issue I
> > > noted early
> > > > on about GLX not working on this setup (there's no output from
> > > > vs-get-limits, but looking at the code it's not going to
> > > work unless
> > > > GLX is behaving itself).
> > > >
> > > > What's odd is that vs-get-limits apparently isn't returning
> > > an error -
> > > > I put a few debugging prints in vs-generate-node-config,
> > > and it looks
> > > > like it's following the 'if ret == 0:' path every time it's
> > > run to get
> > > > to the point it fails (line 489).
> > > >
> > > > I'll have another shot at getting GLX to behave, but this
> > > failure mode
> > > > is somewhat problematic . . .
> > > >
> > > > Simon Fowler
> > > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > > --------------------------------------------------------------
> > > > ----------------
> > > > _______________________________________________
> > > > vizstack-users mailing list
> > > > viz...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > >
> > > > --------------------------------------------------------------
> > > > ----------------
> > > > _______________________________________________
> > > > vizstack-users mailing list
> > > > viz...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
>
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
|
|
From: <Sim...@cs...> - 2010-05-12 22:42:08
|
Of course - I imagine Dell will like that, too ;-)
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Thursday, May 13, 2010 12:32 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Hi Simon,
I think I have fixed this problem. I have manually tested your scenario and verified the results.
Please test out the 'shree' branch again. Your desktop should now be
- 7680x3200 pixels if pixel skipping is disabled. This is the default
- 8104x3422 pixels if pixel skipping is enabled. You may do this by passing "-B" on the command line to viz-desktop
Note that I have added a template for the Quadro NVS 420 into VizStack itself. I am also interested to add the template for you Dell display device. Would you be fine with me including your template into VizStack ?
Thanks!
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 11:40 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Here it is.
I'll drop back to the trunk for the moment - plenty of stuff I can work on that doesn't rely on the new features. Let me know when you need some more testing.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 2:52 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Mistake in calculations. Perhaps I made a hasty change without considering all cases right. Please send me the EDID file for your monitor also. That allows me to test without having your monitor.
Give me a week; there are other things I need to sort. I would recommend you shift to the main branch while I do some more testing !
Thanks
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 9:31 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Attached the template - I built it by editing the LP3065 template, so it should be fine. The screen dimensions come from the EDID, and the bezels were directly measured. The 3x2 display ends up being 8072x3298. I've also attached the generated xorg.conf file.
Aha, after a bit more careful testing it looks like the scrolling happens across all the bezels - the same thing I'd expect if it was three virtual desktops sized to take the bezels into account.
A command line option to viz-desktop would be fine, or an option for the tiled display config so that you could have a 3x2-full and 3x2-no-bezels.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 1:35 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Hi Simon,
Good to know things work for you. Now to fix the bezel problem.
> Two issues: firstly, the bezel width calculations are off - they're
> hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?).
Can you send me your display template ? Please note that all dimensions are needed in millimeters. If in doubt, look at the template for the "HP LP3065". This is our 30" monitor, and you would have comparable dimensions.
> Secondly, it's not taking into account the outside bezels - they
> shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
The code is supposed to handle these cases.
> Oh, and it would be nice if we could have a way to define a tiled
> display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Good catch. Should make this a command line option to the viz-desktop script. (that's what you are using, right?)
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 7:58 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before, though it's
> not handling the bezels with this setup - is there something I have to
> do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: Kumar, S. <shr...@hp...> - 2010-05-12 14:33:53
|
Hi Simon,
I think I have fixed this problem. I have manually tested your scenario and verified the results.
Please test out the 'shree' branch again. Your desktop should now be
- 7680x3200 pixels if pixel skipping is disabled. This is the default
- 8104x3422 pixels if pixel skipping is enabled. You may do this by passing "-B" on the command line to viz-desktop
Note that I have added a template for the Quadro NVS 420 into VizStack itself. I am also interested to add the template for you Dell display device. Would you be fine with me including your template into VizStack ?
Thanks!
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 11:40 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Here it is.
I'll drop back to the trunk for the moment - plenty of stuff I can work on that doesn't rely on the new features. Let me know when you need some more testing.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 2:52 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Mistake in calculations. Perhaps I made a hasty change without considering all cases right. Please send me the EDID file for your monitor also. That allows me to test without having your monitor.
Give me a week; there are other things I need to sort. I would recommend you shift to the main branch while I do some more testing !
Thanks
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 9:31 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Attached the template - I built it by editing the LP3065 template, so it should be fine. The screen dimensions come from the EDID, and the bezels were directly measured. The 3x2 display ends up being 8072x3298. I've also attached the generated xorg.conf file.
Aha, after a bit more careful testing it looks like the scrolling happens across all the bezels - the same thing I'd expect if it was three virtual desktops sized to take the bezels into account.
A command line option to viz-desktop would be fine, or an option for the tiled display config so that you could have a 3x2-full and 3x2-no-bezels.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 1:35 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Hi Simon,
Good to know things work for you. Now to fix the bezel problem.
> Two issues: firstly, the bezel width calculations are off - they're
> hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?).
Can you send me your display template ? Please note that all dimensions are needed in millimeters. If in doubt, look at the template for the "HP LP3065". This is our 30" monitor, and you would have comparable dimensions.
> Secondly, it's not taking into account the outside bezels - they
> shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
The code is supposed to handle these cases.
> Oh, and it would be nice if we could have a way to define a tiled
> display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Good catch. Should make this a command line option to the viz-desktop script. (that's what you are using, right?)
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 7:58 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before, though it's
> not handling the bezels with this setup - is there something I have to
> do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: <Sim...@cs...> - 2010-04-28 06:13:32
|
Here it is.
I'll drop back to the trunk for the moment - plenty of stuff I can work on that doesn't rely on the new features. Let me know when you need some more testing.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 2:52 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Mistake in calculations. Perhaps I made a hasty change without considering all cases right. Please send me the EDID file for your monitor also. That allows me to test without having your monitor.
Give me a week; there are other things I need to sort. I would recommend you shift to the main branch while I do some more testing !
Thanks
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 9:31 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Attached the template - I built it by editing the LP3065 template, so it should be fine. The screen dimensions come from the EDID, and the bezels were directly measured. The 3x2 display ends up being 8072x3298. I've also attached the generated xorg.conf file.
Aha, after a bit more careful testing it looks like the scrolling happens across all the bezels - the same thing I'd expect if it was three virtual desktops sized to take the bezels into account.
A command line option to viz-desktop would be fine, or an option for the tiled display config so that you could have a 3x2-full and 3x2-no-bezels.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 1:35 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Hi Simon,
Good to know things work for you. Now to fix the bezel problem.
> Two issues: firstly, the bezel width calculations are off - they're
> hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?).
Can you send me your display template ? Please note that all dimensions are needed in millimeters. If in doubt, look at the template for the "HP LP3065". This is our 30" monitor, and you would have comparable dimensions.
> Secondly, it's not taking into account the outside bezels - they
> shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
The code is supposed to handle these cases.
> Oh, and it would be nice if we could have a way to define a tiled
> display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Good catch. Should make this a command line option to the viz-desktop script. (that's what you are using, right?)
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 7:58 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before, though it's
> not handling the bezels with this setup - is there something I have to
> do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: Kumar, S. <shr...@hp...> - 2010-04-28 04:54:39
|
Mistake in calculations. Perhaps I made a hasty change without considering all cases right. Please send me the EDID file for your monitor also. That allows me to test without having your monitor.
Give me a week; there are other things I need to sort. I would recommend you shift to the main branch while I do some more testing !
Thanks
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 9:31 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Attached the template - I built it by editing the LP3065 template, so it should be fine. The screen dimensions come from the EDID, and the bezels were directly measured. The 3x2 display ends up being 8072x3298. I've also attached the generated xorg.conf file.
Aha, after a bit more careful testing it looks like the scrolling happens across all the bezels - the same thing I'd expect if it was three virtual desktops sized to take the bezels into account.
A command line option to viz-desktop would be fine, or an option for the tiled display config so that you could have a 3x2-full and 3x2-no-bezels.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 1:35 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Hi Simon,
Good to know things work for you. Now to fix the bezel problem.
> Two issues: firstly, the bezel width calculations are off - they're
> hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?).
Can you send me your display template ? Please note that all dimensions are needed in millimeters. If in doubt, look at the template for the "HP LP3065". This is our 30" monitor, and you would have comparable dimensions.
> Secondly, it's not taking into account the outside bezels - they
> shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
The code is supposed to handle these cases.
> Oh, and it would be nice if we could have a way to define a tiled
> display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Good catch. Should make this a command line option to the viz-desktop script. (that's what you are using, right?)
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 7:58 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before, though it's
> not handling the bezels with this setup - is there something I have to
> do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: <Sim...@cs...> - 2010-04-28 04:01:29
|
Attached the template - I built it by editing the LP3065 template, so it should be fine. The screen dimensions come from the EDID, and the bezels were directly measured. The 3x2 display ends up being 8072x3298. I've also attached the generated xorg.conf file.
Aha, after a bit more careful testing it looks like the scrolling happens across all the bezels - the same thing I'd expect if it was three virtual desktops sized to take the bezels into account.
A command line option to viz-desktop would be fine, or an option for the tiled display config so that you could have a 3x2-full and 3x2-no-bezels.
Simon
________________________________________
From: Kumar, Shree [shr...@hp...]
Sent: Wednesday, April 28, 2010 1:35 PM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Hi Simon,
Good to know things work for you. Now to fix the bezel problem.
> Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than
> they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?).
Can you send me your display template ? Please note that all dimensions are needed in millimeters. If in
doubt, look at the template for the "HP LP3065". This is our 30" monitor, and you would have comparable
dimensions.
> Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels,
> so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
The code is supposed to handle these cases.
> Oh, and it would be nice if we could have a way to define a tiled display that didn't hide
> any pixels - there are some types of data where you want to see every pixel.
Good catch. Should make this a command line option to the viz-desktop script. (that's what you are using, right?)
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 7:58 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before,
> though it's not handling the bezels with this setup - is
> there something I have to do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: Kumar, S. <shr...@hp...> - 2010-04-28 03:37:14
|
Hi Simon,
Good to know things work for you. Now to fix the bezel problem.
> Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than
> they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?).
Can you send me your display template ? Please note that all dimensions are needed in millimeters. If in
doubt, look at the template for the "HP LP3065". This is our 30" monitor, and you would have comparable
dimensions.
> Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels,
> so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
The code is supposed to handle these cases.
> Oh, and it would be nice if we could have a way to define a tiled display that didn't hide
> any pixels - there are some types of data where you want to see every pixel.
Good catch. Should make this a command line option to the viz-desktop script. (that's what you are using, right?)
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Wednesday, April 28, 2010 7:58 AM
To: viz...@li...
Subject: Re: [vizstack-users] [ExternalEmail] Re: Problem running vs-configure-standalone from shree branch
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before,
> though it's not handling the bezels with this setup - is
> there something I have to do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: <Sim...@cs...> - 2010-04-28 02:28:01
|
Actually, I tell a lie - after fixing up the template for my monitors it's working, after a fashion.
Two issues: firstly, the bezel width calculations are off - they're hiding significantly less pixels than they should be (maybe around three quarters - could it be a 75dpi vs 100dpi thing?). Secondly, it's not taking into account the outside bezels - they shouldn't hide any pixels, so that there's no scrolling of the virtual desktop when you move the pointer to the edges of the tiled display.
Oh, and it would be nice if we could have a way to define a tiled display that didn't hide any pixels - there are some types of data where you want to see every pixel.
Those issues aside it's great ;-) Thanks for the very prompt fixes.
Simon
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Wednesday, 28 April 2010 12:09 PM
> To: viz...@li...
> Subject: [ExternalEmail] Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> With that workaround I can bring up the desktop as before,
> though it's not handling the bezels with this setup - is
> there something I have to do to get that working?
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 5:14 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> >
> > You won't be able to run 3D applications if GLX doesn't work.
> >
> > For now, you can replace the lines
> > ----
> > limitData = open(temp_result_file,'r').read() limitData =
> > limitData.lstrip().rstrip().split(' ')
> > ----
> > With
> > ----
> > try:
> > limitData = open(temp_result_file,'r').read()
> > limitData = limitData.lstrip().rstrip().split(' ')
> > except:
> > limitData = [4096, 4096]
> > ----
> > Remember to keep the indentation right :-)
> >
> > Also note one more thing : the display template is slightly
> changed.
> > So please adjust your template appropriately.
> >
> > HTH
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 12:36 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > As I said in an early post, GLX is misbehaving somehow -
> the extension
> > loads and is apparently available, but any attempt to create a GLX
> > context fails. With earlier versions of VizStack that
> didn't cause any
> > problems, but now that you're relying on GLX to figure out
> things like
> > this it's causing problems.
> >
> > I'm trying to get it working now.
> >
> > Simon
> >
> > > -----Original Message-----
> > > From: Kumar, Shree [mailto:shr...@hp...]
> > > Sent: Tuesday, 27 April 2010 4:49 PM
> > > To: viz...@li...
> > > Subject: Re: [vizstack-users] Problem running
> > vs-configure-standalone
> > > from shree branch
> > >
> > > Hi Simon,
> > >
> > > Does GLX not work at all at your site ?
> > >
> > > vs-get-limits is trying to find the maximum frambuffer
> size that is
> > > supported by your GPU.
> > >
> > > Regards
> > > -- Shree
> > >
> > > -----Original Message-----
> > > From: Sim...@cs... [mailto:Sim...@cs...]
> > > Sent: Tuesday, April 27, 2010 9:58 AM
> > > To: viz...@li...
> > > Subject: [vizstack-users] Problem running
> > vs-configure-standalone from
> > > shree branch
> > >
> > > I'm trying out the head of the shree branch, and finding that
> > > vs-configure-standalone fails with the following output:
> > >
> > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> > >
> > > Processing Node 'localhost'...
> > > Errors happened while trying to get the configuration of node
> > > 'localhost'. Reason:
> > >
> > > Traceback (most recent call last):
> > > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > > <module>
> > > limitData = open(temp_result_file,'r').read()
> > > IOError: [Errno 2] No such file or directory:
> > > '/tmp/temp-vs-configure-system/temp-result-file'
> > >
> > >
> > > Please fix the above errors & run this tool again Failed to
> > configure
> > > standalone configuration
> > >
> > >
> > > Looking at the code it looks like vs-get-limits is failing
> > to produce
> > > an output file. I'm pretty sure this is due to the issue I
> > noted early
> > > on about GLX not working on this setup (there's no output from
> > > vs-get-limits, but looking at the code it's not going to
> > work unless
> > > GLX is behaving itself).
> > >
> > > What's odd is that vs-get-limits apparently isn't returning
> > an error -
> > > I put a few debugging prints in vs-generate-node-config,
> > and it looks
> > > like it's following the 'if ret == 0:' path every time it's
> > run to get
> > > to the point it fails (line 489).
> > >
> > > I'll have another shot at getting GLX to behave, but this
> > failure mode
> > > is somewhat problematic . . .
> > >
> > > Simon Fowler
> > > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > > --------------------------------------------------------------
> > > ----------------
> > > _______________________________________________
> > > vizstack-users mailing list
> > > viz...@li...
> > > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> > >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
|
|
From: <Sim...@cs...> - 2010-04-28 02:09:53
|
With that workaround I can bring up the desktop as before, though it's not handling the bezels with this setup - is there something I have to do to get that working?
Simon
> -----Original Message-----
> From: Kumar, Shree [mailto:shr...@hp...]
> Sent: Tuesday, 27 April 2010 5:14 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
>
> You won't be able to run 3D applications if GLX doesn't work.
>
> For now, you can replace the lines
> ----
> limitData = open(temp_result_file,'r').read() limitData =
> limitData.lstrip().rstrip().split(' ')
> ----
> With
> ----
> try:
> limitData = open(temp_result_file,'r').read()
> limitData = limitData.lstrip().rstrip().split(' ')
> except:
> limitData = [4096, 4096]
> ----
> Remember to keep the indentation right :-)
>
> Also note one more thing : the display template is slightly
> changed. So please adjust your template appropriately.
>
> HTH
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Tuesday, April 27, 2010 12:36 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> As I said in an early post, GLX is misbehaving somehow - the
> extension loads and is apparently available, but any attempt
> to create a GLX context fails. With earlier versions of
> VizStack that didn't cause any problems, but now that you're
> relying on GLX to figure out things like this it's causing problems.
>
> I'm trying to get it working now.
>
> Simon
>
> > -----Original Message-----
> > From: Kumar, Shree [mailto:shr...@hp...]
> > Sent: Tuesday, 27 April 2010 4:49 PM
> > To: viz...@li...
> > Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone
> > from shree branch
> >
> > Hi Simon,
> >
> > Does GLX not work at all at your site ?
> >
> > vs-get-limits is trying to find the maximum frambuffer size that is
> > supported by your GPU.
> >
> > Regards
> > -- Shree
> >
> > -----Original Message-----
> > From: Sim...@cs... [mailto:Sim...@cs...]
> > Sent: Tuesday, April 27, 2010 9:58 AM
> > To: viz...@li...
> > Subject: [vizstack-users] Problem running
> vs-configure-standalone from
> > shree branch
> >
> > I'm trying out the head of the shree branch, and finding that
> > vs-configure-standalone fails with the following output:
> >
> > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
> >
> > Processing Node 'localhost'...
> > Errors happened while trying to get the configuration of node
> > 'localhost'. Reason:
> >
> > Traceback (most recent call last):
> > File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in
> > <module>
> > limitData = open(temp_result_file,'r').read()
> > IOError: [Errno 2] No such file or directory:
> > '/tmp/temp-vs-configure-system/temp-result-file'
> >
> >
> > Please fix the above errors & run this tool again Failed to
> configure
> > standalone configuration
> >
> >
> > Looking at the code it looks like vs-get-limits is failing
> to produce
> > an output file. I'm pretty sure this is due to the issue I
> noted early
> > on about GLX not working on this setup (there's no output from
> > vs-get-limits, but looking at the code it's not going to
> work unless
> > GLX is behaving itself).
> >
> > What's odd is that vs-get-limits apparently isn't returning
> an error -
> > I put a few debugging prints in vs-generate-node-config,
> and it looks
> > like it's following the 'if ret == 0:' path every time it's
> run to get
> > to the point it fails (line 489).
> >
> > I'll have another shot at getting GLX to behave, but this
> failure mode
> > is somewhat problematic . . .
> >
> > Simon Fowler
> > Technical Specialist eResearch Visualisation team CSIRO IM&T
> > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> > --------------------------------------------------------------
> > ----------------
> > _______________________________________________
> > vizstack-users mailing list
> > viz...@li...
> > https://lists.sourceforge.net/lists/listinfo/vizstack-users
> >
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
|
|
From: Kumar, S. <shr...@hp...> - 2010-04-27 07:15:37
|
You won't be able to run 3D applications if GLX doesn't work.
For now, you can replace the lines
----
limitData = open(temp_result_file,'r').read()
limitData = limitData.lstrip().rstrip().split(' ')
----
With
----
try:
limitData = open(temp_result_file,'r').read()
limitData = limitData.lstrip().rstrip().split(' ')
except:
limitData = [4096, 4096]
----
Remember to keep the indentation right :-)
Also note one more thing : the display template is slightly changed. So please adjust your template appropriately.
HTH
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Tuesday, April 27, 2010 12:36 PM
To: viz...@li...
Subject: Re: [vizstack-users] Problem running vs-configure-standalone from shree branch
As I said in an early post, GLX is misbehaving somehow - the extension loads and is apparently available, but any attempt to create a GLX context fails. With earlier versions of VizStack that didn't cause any problems, but now that you're relying on GLX to figure out things like this it's causing problems.
I'm trying to get it working now.
Simon
> -----Original Message-----
> From: Kumar, Shree [mailto:shr...@hp...]
> Sent: Tuesday, 27 April 2010 4:49 PM
> To: viz...@li...
> Subject: Re: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> Hi Simon,
>
> Does GLX not work at all at your site ?
>
> vs-get-limits is trying to find the maximum frambuffer size
> that is supported by your GPU.
>
> Regards
> -- Shree
>
> -----Original Message-----
> From: Sim...@cs... [mailto:Sim...@cs...]
> Sent: Tuesday, April 27, 2010 9:58 AM
> To: viz...@li...
> Subject: [vizstack-users] Problem running
> vs-configure-standalone from shree branch
>
> I'm trying out the head of the shree branch, and finding that
> vs-configure-standalone fails with the following output:
>
> root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
>
> Processing Node 'localhost'...
> Errors happened while trying to get the configuration of node
> 'localhost'. Reason:
>
> Traceback (most recent call last):
> File "/opt/vizstack/sbin/vs-generate-node-config", line
> 489, in <module>
> limitData = open(temp_result_file,'r').read()
> IOError: [Errno 2] No such file or directory:
> '/tmp/temp-vs-configure-system/temp-result-file'
>
>
> Please fix the above errors & run this tool again Failed to
> configure standalone configuration
>
>
> Looking at the code it looks like vs-get-limits is failing to
> produce an output file. I'm pretty sure this is due to the
> issue I noted early on about GLX not working on this setup
> (there's no output from vs-get-limits, but looking at the
> code it's not going to work unless GLX is behaving itself).
>
> What's odd is that vs-get-limits apparently isn't returning
> an error - I put a few debugging prints in
> vs-generate-node-config, and it looks like it's following the
> 'if ret == 0:' path every time it's run to get to the point
> it fails (line 489).
>
> I'll have another shot at getting GLX to behave, but this
> failure mode is somewhat problematic . . .
>
> Simon Fowler
> Technical Specialist eResearch Visualisation team CSIRO IM&T
> Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
> --------------------------------------------------------------
> ----------------
> _______________________________________________
> vizstack-users mailing list
> viz...@li...
> https://lists.sourceforge.net/lists/listinfo/vizstack-users
>
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: <Sim...@cs...> - 2010-04-27 07:05:50
|
As I said in an early post, GLX is misbehaving somehow - the extension loads and is apparently available, but any attempt to create a GLX context fails. With earlier versions of VizStack that didn't cause any problems, but now that you're relying on GLX to figure out things like this it's causing problems. I'm trying to get it working now. Simon > -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Tuesday, 27 April 2010 4:49 PM > To: viz...@li... > Subject: Re: [vizstack-users] Problem running > vs-configure-standalone from shree branch > > Hi Simon, > > Does GLX not work at all at your site ? > > vs-get-limits is trying to find the maximum frambuffer size > that is supported by your GPU. > > Regards > -- Shree > > -----Original Message----- > From: Sim...@cs... [mailto:Sim...@cs...] > Sent: Tuesday, April 27, 2010 9:58 AM > To: viz...@li... > Subject: [vizstack-users] Problem running > vs-configure-standalone from shree branch > > I'm trying out the head of the shree branch, and finding that > vs-configure-standalone fails with the following output: > > root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone > > Processing Node 'localhost'... > Errors happened while trying to get the configuration of node > 'localhost'. Reason: > > Traceback (most recent call last): > File "/opt/vizstack/sbin/vs-generate-node-config", line > 489, in <module> > limitData = open(temp_result_file,'r').read() > IOError: [Errno 2] No such file or directory: > '/tmp/temp-vs-configure-system/temp-result-file' > > > Please fix the above errors & run this tool again Failed to > configure standalone configuration > > > Looking at the code it looks like vs-get-limits is failing to > produce an output file. I'm pretty sure this is due to the > issue I noted early on about GLX not working on this setup > (there's no output from vs-get-limits, but looking at the > code it's not going to work unless GLX is behaving itself). > > What's odd is that vs-get-limits apparently isn't returning > an error - I put a few debugging prints in > vs-generate-node-config, and it looks like it's following the > 'if ret == 0:' path every time it's run to get to the point > it fails (line 489). > > I'll have another shot at getting GLX to behave, but this > failure mode is somewhat problematic . . . > > Simon Fowler > Technical Specialist eResearch Visualisation team CSIRO IM&T > Yarralumla, 2600 Desk 02 6124 1453 Mob 0409 245 871 > -------------------------------------------------------------- > ---------------- > _______________________________________________ > vizstack-users mailing list > viz...@li... > https://lists.sourceforge.net/lists/listinfo/vizstack-users > > -------------------------------------------------------------- > ---------------- > _______________________________________________ > vizstack-users mailing list > viz...@li... > https://lists.sourceforge.net/lists/listinfo/vizstack-users > |
|
From: Kumar, S. <shr...@hp...> - 2010-04-27 06:50:49
|
Hi Simon,
Does GLX not work at all at your site ?
vs-get-limits is trying to find the maximum frambuffer size that is supported by your GPU.
Regards
-- Shree
-----Original Message-----
From: Sim...@cs... [mailto:Sim...@cs...]
Sent: Tuesday, April 27, 2010 9:58 AM
To: viz...@li...
Subject: [vizstack-users] Problem running vs-configure-standalone from shree branch
I'm trying out the head of the shree branch, and finding that vs-configure-standalone fails with the following output:
root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
Processing Node 'localhost'...
Errors happened while trying to get the configuration of node 'localhost'. Reason:
Traceback (most recent call last):
File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in <module>
limitData = open(temp_result_file,'r').read()
IOError: [Errno 2] No such file or directory: '/tmp/temp-vs-configure-system/temp-result-file'
Please fix the above errors & run this tool again
Failed to configure standalone configuration
Looking at the code it looks like vs-get-limits is failing to produce an output file. I'm pretty sure this is due to the issue I noted early on about GLX not working on this setup (there's no output from vs-get-limits, but looking at the code it's not going to work unless GLX is behaving itself).
What's odd is that vs-get-limits apparently isn't returning an error - I put a few debugging prints in vs-generate-node-config, and it looks like it's following the 'if ret == 0:' path every time it's run to get to the point it fails (line 489).
I'll have another shot at getting GLX to behave, but this failure mode is somewhat problematic . . .
Simon Fowler
Technical Specialist eResearch Visualisation team
CSIRO IM&T
Yarralumla, 2600
Desk 02 6124 1453
Mob 0409 245 871
------------------------------------------------------------------------------
_______________________________________________
vizstack-users mailing list
viz...@li...
https://lists.sourceforge.net/lists/listinfo/vizstack-users
|
|
From: <Sim...@cs...> - 2010-04-27 04:28:20
|
I'm trying out the head of the shree branch, and finding that vs-configure-standalone fails with the following output:
root@smlopt1-display:~# /opt/vizstack/sbin/vs-configure-standalone
Processing Node 'localhost'...
Errors happened while trying to get the configuration of node 'localhost'. Reason:
Traceback (most recent call last):
File "/opt/vizstack/sbin/vs-generate-node-config", line 489, in <module>
limitData = open(temp_result_file,'r').read()
IOError: [Errno 2] No such file or directory: '/tmp/temp-vs-configure-system/temp-result-file'
Please fix the above errors & run this tool again
Failed to configure standalone configuration
Looking at the code it looks like vs-get-limits is failing to produce an output file. I'm pretty sure this is due to the issue I noted early on about GLX not working on this setup (there's no output from vs-get-limits, but looking at the code it's not going to work unless GLX is behaving itself).
What's odd is that vs-get-limits apparently isn't returning an error - I put a few debugging prints in vs-generate-node-config, and it looks like it's following the 'if ret == 0:' path every time it's run to get to the point it fails (line 489).
I'll have another shot at getting GLX to behave, but this failure mode is somewhat problematic . . .
Simon Fowler
Technical Specialist eResearch Visualisation team
CSIRO IM&T
Yarralumla, 2600
Desk 02 6124 1453
Mob 0409 245 871
|
|
From: <Sim...@cs...> - 2010-04-23 23:21:22
|
(**) NVIDIA(0): Option "UseDisplayDevice" set to "none"; enabling NoScanout (**) NVIDIA(0): mode (--) NVIDIA(0): Not mapping the primary surface by default. (WW) NVIDIA(0): NoScanout is only supported on Tesla or Quadro hardware. (WW) NVIDIA(0): Disabling NoScanout. (II) NVIDIA(0): NVIDIA GPU Quadro NVS 420 (G98GL) at PCI:9:0:0 (GPU-0) (--) NVIDIA(0): Memory: 524288 kBytes (II) NVIDIA(0): GPU RAM Type: GDDR3 (--) NVIDIA(0): VideoBIOS: 62.98.68.00.06 (--) NVIDIA(0): Found 2 CRTCs on board (II) NVIDIA(0): Supported display device(s): DFP-0, DFP-1, DFP-2, DFP-3 (II) NVIDIA(0): Bus detected as PCI Express (II) NVIDIA(0): Detected PCI Express Link width: 4X (--) NVIDIA(0): Interlaced video modes are supported on this GPU (II) NVIDIA(0): DFP scaling is supported on this GPU (II) NVIDIA(0): SPS : 1 (II) NVIDIA(0): (II) NVIDIA(0): Mode timing constraints for : Quadro NVS 420 (II) NVIDIA(0): Maximum mode timing values : (II) NVIDIA(0): Horizontal Visible Width : 32767 (II) NVIDIA(0): Horizontal Blank Start : 32767 (II) NVIDIA(0): Horizontal Blank Width : 32767 (II) NVIDIA(0): Horizontal Sync Start : 32767 (II) NVIDIA(0): Horizontal Sync Width : 32767 (II) NVIDIA(0): Horizontal Total Width : 32767 (II) NVIDIA(0): Vertical Visible Height : 32767 (II) NVIDIA(0): Vertical Blank Start : 32767 (II) NVIDIA(0): Vertical Blank Width : 32767 (II) NVIDIA(0): Veritcal Sync Start : 32767 (II) NVIDIA(0): Vertical Sync Width : 32767 (II) NVIDIA(0): Vertical Total Height : 32767 (II) NVIDIA(0): (II) NVIDIA(0): Minimum mode timing values : (II) NVIDIA(0): Horizontal Total Width : 8 (II) NVIDIA(0): Vertical Total Height : 5 (II) NVIDIA(0): (II) NVIDIA(0): Mode timing alignment : (II) NVIDIA(0): Horizontal Visible Width : multiples of 1 (II) NVIDIA(0): Horizontal Blank Start : multiples of 1 (II) NVIDIA(0): Horizontal Blank Width : multiples of 1 (II) NVIDIA(0): Horizontal Sync Start : multiples of 1 (II) NVIDIA(0): Horizontal Sync Width : multiples of 1 (II) NVIDIA(0): Horizontal Total Width : multiples of 1 After that it goes onto the monitors (on DFP-2/3). Simon ________________________________________ From: Kumar, Shree [shr...@hp...] Sent: Friday, April 23, 2010 3:31 PM To: viz...@li... Subject: Re: [vizstack-users] Issue when changing between DVI and displayport on NVS420 No problem; you can test it at your convenience. I need one more piece of information to solve this problem. If you look in your Xorg log file, you will find a line which says "Supported display device(s)". Can you send me a few Lines around it ? Thanks -- Shree -----Original Message----- From: Sim...@cs... [mailto:Sim...@cs...] Sent: Friday, April 23, 2010 10:17 AM To: viz...@li... Subject: Re: [vizstack-users] Issue when changing between DVI and displayport on NVS420 I probably won't be able to test it until Tuesday Australian time (ANZAC day public holiday Monday), but I'll get onto it as soon as I can. Thanks for the quick response ;-) Simon > -----Original Message----- > From: Kumar, Shree [mailto:shr...@hp...] > Sent: Friday, 23 April 2010 2:25 PM > To: viz...@li... > Subject: Re: [vizstack-users] Issue when changing between DVI > and displayport on NVS420 > > Hi Simon, > > Thanks for the lspci dump and the info about the display > outputs. I'm glad this won't break any of my internal designs :-) > > Give me a day to remove the hardcoding. Then I'll let you > know so you can test. > > Thanks > -- Shree ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users ------------------------------------------------------------------------------ _______________________________________________ vizstack-users mailing list viz...@li... https://lists.sourceforge.net/lists/listinfo/vizstack-users |
