[Xpert] Why won't the server accept connections from local host?
I run the XFree86 binary. I can't start a window manager from the console. Why is it doing this? I don't have this problem on another machine, but on this one it won't accept any connections at all. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86 snapshot 4.2.99.3
On Tue, 24 Dec 2002, Alessandro Astarita wrote: > Il 22:45, martedì 24 dicembre 2002, David Dawes ha scritto: > > 4.2.99.3 has most of the features that will be present in 4.3.0. > > The main exception is several radeon driver updates, which will be > > available in a future snapshot. > > How about GeForce4 support for PowerPC machines? > I'm not sure when 4.2.99.3 was cut, but GeForce4 should work on PPC after Fri Nov 8 22:00:14 2002 UTC. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Mouse no longer working in Quake3?
I just noticed that the mouse no longer works in Quake3. The only significant change I made to this system (with regards to the mouse, that is) was syncing and rebuilding XFree86 from CVS. I don't suppose anyone managed to get CVS (I synced at 18:31 on Dec 11) and Quake3 (ver. 1.32) to have working mouse support. At first I thought it was DGA stuff that was broken, but the mouse works fine in my DGA test app. Xev shows that Quake3 isn't getting any events in fullscreen or windowed mode, DGA mouse support or not. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86 -configure fails
On 7 Dec 2002, sashan wrote:
> Attached are the results of XFree86 -configure (configure.txt), scanpci
> -v (scanpci.txt) and the log file. Please can someone help or advise how
> to go about making changes to the header file mentioned (see
> configure.txt)
>
That card wasn't supported 4.2.0. It's supported in CVS.
If you don't want to build a CVS server, the module on my web
page: http://www.xfree86.org/~mvojkovi/nv_drv.o will work if
you start the server with option -ignoreABI ("startx -- -ignoreABI").
That module is a debug module built from CVS and you have to
pass ignoreABI because the version number won't match your server.
That module replaces /usr/X11R6/lib/modules/drivers/nv_drv.o.
Or, for Linux and FreeBSD you could use the binary linux drivers
on NVIDIA's web page.
Mark.
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Re: CRTL-ALT-BACKSPACE broken
On Thu, 28 Nov 2002, Mark Vojkovich wrote: > > As Alan mentioned earlier, this does seem to be broken. I just > synced up on PPC and find that none of the CRTL-ALT- keys work > anymore. Can't quit with CTRL-ALT-BS, can't switch VTs. > > There have been modifications by Marc and David to xf86Events.c > during the last week. David's specifically modifies CTRL-ALT behavior. > Marc's puts some #ifdefs around related code, but the logic > doesn't look suspect. > Are we the only ones seeing this? I just synced up and made World again, and made sure I got rid of my host.def so that I didn't have any stale configuration around. I still can't do any CTRL-ALT-XXX. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]CRTL-ALT-BACKSPACE broken
As Alan mentioned earlier, this does seem to be broken. I just synced up on PPC and find that none of the CRTL-ALT- keys work anymore. Can't quit with CTRL-ALT-BS, can't switch VTs. There have been modifications by Marc and David to xf86Events.c during the last week. David's specifically modifies CTRL-ALT behavior. Marc's puts some #defines around related code, but the logic doesn't look suspect. MArk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86's GDI functions
On Tue, 26 Nov 2002, Hua zhang wrote: > Yes. There is a project that needs to study the arithmetic of Window's GDI, but >there is no way to get its source code you know. We hope to study the GDI functions >of Linux first, that is XFree86. It is a huge code package and I have no too much >time to study it. So I want to know anybody who has stuided or implemented XFree86's >GDI functions and friendly enough can give me some useful information on it. > > Thanks The term GDI is used pretty loosely and what exactly GDI is could be open to interpretation. To me, GDI is the device independent part of Windows that's handles rendering, window placement, etc... It calls into drivers through a fixed API and drivers don't have access to this code. XFree86 drivers are designed with a different philosophy. All the major data structures are exposed and the driver itself actually provides most of what GDI does. Not that all that code is in the drivers, it's not. But the XFree86 drivers chose which routines to use and usually picks them from common helper functions (shared code that is electively, not forcibly, used by most of the drivers). But the driver had the option of providing its own code for most of this if it wanted to. There is a small part of code in the X-server that the driver doesn't have access to. This is the device independent code and actually doesn't do much. For the rendering requests you mentioned, they start at xc/programs/Xserver/dix/dispatch.c. This is where rendering request from the clients come in. They call into the driver fairly shortly after that. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86's GDI functions
On Tue, 26 Nov 2002, Hua zhang wrote: > Hi all: > >How to find the source code of XFree86's GDI functions, such as XDrawPoint(), >XDrawLine() etc? > Is the file /src/X410src-1/xc/programs/Xserver/Xext/panoramiXprocs.c implements all >of them? > No, that is the Xinerama extension. I'm not sure what the question is. The thing most closely resembling GDI in XFree86 would be XAA, and that actually has little to do with XDrawPoint(), etc... which are client library functions. Perhaps if you were to explain what you needed to do rather than what you think you are looking for. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Nvidia GEForce4 Mx440
On 26 Nov 2002, mike wrote: > On Sun, 2002-11-24 at 01:42, Mark Vojkovich wrote: > > On Wed, 20 Nov 2002, George Gopie wrote: > > > > > I have installed Linux Red Hat Advanced Server 2.4.9-e.3 on my intel > > > server with a 64MB nvidia #DGeForce4 Mx440 series video card. I am > > > unable to get it to work with X. I have down loaded XFRee86 4.0.2 and > > > installed it., but when I try to start the configuration with XFree86 > > > -configure i get that the Following: > > > Fatal Server error: > > > XFree86 has found a valid card configuration. > > > Unfortunately the appropriate data has not been added to the > > > xf86PciInfo.h. > > > > > > > > > This card is relatively new and there has been no official XFree86 release > > since that card has come out so there is no official release supporting > > it (certainly not 4.0.2 which is two years old). You either have > > to build XFree86 from CVS or install the binary drivers from nvidia's > > web site, or *maybe* replacing the /usr/X11R6/lib/modules/drivers/nv_drv.o > > with the one on my web page http://www.xfree86.org/~mvojkovi/nv_drv.o > > will work if you start the server with the -ignoreABI option (ie. > > "startx -- -ignoreABI") - that certainly works with 4.2.0, but I don't > > know about 4.0.2. > > > > Relatively new? - just use the nv driver > > I attach my XF86Config if you want to use it (I have a 440mx as well) > > It has worked with 4.1 4.2 and cvs - havn't checked 4.0.2 but the nv > driver has been around a while I added support for PCI ID 0x0171 "GeForce4 MX 440" on 2001/10/01. It can't possibly work with 4.1.0, which was released earlier in the year. While 0x0171 recognition was in 4.2.0, I only had prelease boards to run with it. I know that many modern NV17 boards (maybe most) do not work with 4.2.0. Everything should work fine in CVS. The "nv" driver doesn't officially support NV17 until after 4.2.0. MArk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Re: [NV driver] I2C bus(ses) question.
On Mon, 25 Nov 2002, Ducrot Bruno wrote: > Hi Mark, list. > > I am a bit confused with the code for the I2C bus(ses), especially > in a multiple heads environment. > > The current code use the extended registers 0x3e, 0x3f in order > to implement the I2C's functions I2CPutBits and I2CGetBits, but > I think that the second head should use actually the registers > 0x36 and 0x37. Not exactly. There is a difference between the CRT controller and the connector on the card. The I2C function is tied to a connector, not a CRTC. So it depends on the board layout. On NV11 CRTC 0 & 1 correspond to outputs 0 & 1, but on everything newer it could be either way. It may be the case that straight-across 0->0, 1->1 may work better overall, but it's not guaranteed. I will examine the success rate with my cards over the US holiday later this week. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Portrait Mode in X
On Mon, 25 Nov 2002, Mark Cuss wrote: > Mark, > > Thanks for the info. I noticed in your list below that the intel 815 / 830 > series chips aren't listed... Is it difficult to implement this > functionality in software? I don't currently have a copy of the X source, > but I could download it and give it a try... Rotation using the "shadow framebuffer" is easy. Rendering goes into system memory and then is copied rotated to video ram. The one driver that does this in hardware (silicon motion I think) has the two framebuffers in video ram. It renders to one in hardware and then updates the other rotated via hardware. > > The ultimate goal, however, is to have the rotate supported on the 815 / 830 > chips in hardware. The application that we plan to use displays decoded > MPEG2 that does a bunch of XvPutImage calls which would likely be pretty > slow if the X server had to rotate each frame in software. I'm thinking > that the hardware does support this functionality, as a lot of tablet > manufacturers are using this hardware in the new "Designed for Windows XP > Tablet Edition" computers... I was planning to prove them wrong by running > Linux on one :) Xv(Shm)PutImage doesn't render to the framebuffer on i815 - the data goes into the overlay and isn't handled by the shadow framebuffer code. You'd have to rotate the source for the overlay yourself in software (easy for 4:2:0, messy for 4:2:2). > > I've been in contact with one of the Tablet manufacturers, and they say that > the Intel video driver that they're using in Windows does the flipping for > them... this makes me assume that the Windows driver is flipping a 1 into a > register somewhere to rotate the image. However, I've looked through the > i830 data sheet and can't spot anything ... Unlikely. They probably render into video ram normally and then copy it with rotation using the texture engine. Video, if using the overlay, is probably done in software by rotating the source, or they just fall back to YUV->RGB blits in this case so they don't have to deal with the overlay. MARk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert][PATCH] let Xv work again with "nv" after a swsusp restore
On Sun, 24 Nov 2002, Pierre Lombard wrote: > * Mark Vojkovich <[EMAIL PROTECTED]> [2002-11-24 01:02]: > > On Wed, 13 Nov 2002, Pierre Lombard wrote: > > > > The attached patch lets Xv work again after restoring a swsusp-frozen > > > system. > > > > > > Please Cc if necessary as I am not subscribed to Xpert mailing list. > > >I'm not sure I understand the point of this patch. NVResetVideo > > doesn't really do anything other than reset the brightness and > > contrast of the video overlay. Is this an early revision NV11 you > > have? A laptop maybe? > > It's a laptop video card: > 01:00.0 VGA compatible controller: nVidia Corporation NV17 [GeForce4 440 > Go] (rev a3) > > It's just a port of http://cpbotha.net/dri_resume.html. > I only have a vague idea of what's happening but it has just worked for > me (maybe it simply workarounds some swsusp bug indeed...) It's more likely something strange that the combination system/video bios is doing, and probably only a problem on laptops by specific vendors. The patch seems harmless though, so I'll apply it (or a modification of it). Thanks, Mark. > > Without this patch xv does not work after you resume from a > swsusp'ended: the application are still able to use xv but the display > is no longer correct (for instance, mplayer only shows a blue window > after a swsusp restore, till I restart X). > > Some problems with nv driver and swsusp were reported there too: > http://lister.fornax.hu/pipermail/swsusp/2002-October/001003.html > http://lister.fornax.hu/pipermail/swsusp/2002-October/001021.html > > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Nvidia GEForce4 Mx440
On Wed, 20 Nov 2002, George Gopie wrote: > I have installed Linux Red Hat Advanced Server 2.4.9-e.3 on my intel > server with a 64MB nvidia #DGeForce4 Mx440 series video card. I am > unable to get it to work with X. I have down loaded XFRee86 4.0.2 and > installed it., but when I try to start the configuration with XFree86 > -configure i get that the Following: > Fatal Server error: > XFree86 has found a valid card configuration. > Unfortunately the appropriate data has not been added to the > xf86PciInfo.h. > This card is relatively new and there has been no official XFree86 release since that card has come out so there is no official release supporting it (certainly not 4.0.2 which is two years old). You either have to build XFree86 from CVS or install the binary drivers from nvidia's web site, or *maybe* replacing the /usr/X11R6/lib/modules/drivers/nv_drv.o with the one on my web page http://www.xfree86.org/~mvojkovi/nv_drv.o will work if you start the server with the -ignoreABI option (ie. "startx -- -ignoreABI") - that certainly works with 4.2.0, but I don't know about 4.0.2. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Portrait Mode in X
I wouldn't be surprised if most of the rotate drivers were broken in CVS since the resize and rotate extension went in - the "nv" driver is. But most of the drivers in 4.2.0 and 4.2.1 have support for rotate. Grepping CVS source code shows ROTATE options in the following drivers: chips and technology, cirrus logic, fbdev, cyrix, matrox, neomagic, rendition, S3 virge, S3 savage, siliconmotion, trident, sis, nv. Not sure which ones of those actually work though. Nearly all of these drivers actually implement this in software so it's slower than normal operation (no hardware features for the rotation). Mark. On Fri, 22 Nov 2002, Mark Cuss wrote: > Hello > > I was wondering if any of you folks had any information on running X in > "portrait" mode (ie - rotating the contents of the display 90 degrees) for > use on a tablet - style computer. > > I've noticed that the silicon motion driver has a "rotate" option, but I > haven't found any others that do. I'm assuming that this needs to be > supported by the hardware, as it would be an expensive operation in software > (or so I've seen under Windows, anyways)... > > If anyone has information on which hardware supports this mode of operation > and whether or not they're supported in XFree86, that would be great. My > main interest would probably be in the Intel integrated chipset video (i815, > 830, etc) as most tablet computers tend to use this as their graphics > processor. > > Thanks in advance, > > Mark > > Mark Cuss > Real Time Systems Analyst > CDL Systems Ltd. > Suite 230 > 3553 - 31 Street NW > Calgary, Alberta > T2L 2K7 > > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]nv driver crash on nForce
On Wed, 6 Nov 2002, Wheeler, Fred (Research) wrote: > I was experiencing this same lock up with XFree86 4.2.1, as shipped on the FreeBSD >4.7-RELEASE CDs. > I'm also using an Asus A7N266-VM with an Athlon XP1800+. The problem was fixed by >changing my > nv_drv.o to the one on Mark's web page (link is below). Thanks much for the fix. >Here are my simple > observations, in case they are helpful. > > I did in fact have to use -ignoreABI. I notice that Mark's nv_drv.o is about 1.5M >and the one it > replaces was about 50k. That's because it was built with -g so it has full debug info in it. When I tested the nForce support before 4.2.0 I only had my one early board to test with. It may be the case (sounds like it from the bug reports) that 4.2.0 (and 4.2.1 which was a bug fix version of 4.2.0, not a new CVS release) may not work with newer nForce boards. I'm glad there's a 4.3.0 release coming up which will have the support that's in CVS. Mark. > After nv_drv.o is replaced, "xf86cfg -textmode" no longer lists all of the > nVidia cards. I got a hang, and no log, even with -probeonly. I do not believe I'm >using a kernel > framebuffer. I did not specifically set one up, and nothing in the FreeBSD install >process mentioned > anything about a framebuffer. I have not yet been able to try Mozilla (just >installed FreeBSD for > the first time, coming up to speed). I have not seen anything unusual with any >other X applications. > > Thanks, > Fred Wheeler > > > > On 6 Oct 2002, Mikkel Lauritsen wrote: > > > On Thu, 2002-10-03 at 22:35, Mark Vojkovich wrote: > > > > > I don't recall having any problems with 4.2.0 on nForce. I'm > > > not sure that RedHat's "4.2.0" is really the same as a stock > > > XFree86 4.2.0. There have be cases in the past where they have > > > modified theirs. Try an official XFree86 4.2.0. > > > > I just did, and unfortunately the behavior is the exact same. > > > > Just to display my potential ignorance, what I did was to replace > > nv_drv.o from the RedHat package with that from the binary distribution > > at ftp.xfree86.org. I still get a hard lockup, and it seems as if the > > log files that I am able to obtain reveal nothing at all. I guess they > > are truncated at a more or less random point in time when the machine > > hangs. > >Copying the module should probably be fine. > > > > > So... is there any way that I might try to look deeper into this, > > preferably without having to rebuild X from scratch? > > > >Do you get a full /var/log/XFree86.0.log if you do > >"XFree86 -probeonly" > > I put an nv_drv.o compiled from CVS on my web page at: > > http://www.xfree86.org/~mvojkovi/nv_drv.o > > Since it's newer than your server, you may need to add -ignoreABI to your > server command line. (ie. "startx -- -ignoreABI"). > > Are you running some kernel framebuffer device like rivafb? It's > possible that rivafb is not compatible with the "nv" driver on nForce. > I never use something like rivafb. > > > Mark. > > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert][PATCH] let Xv work again with "nv" after a swsusp restore
On Wed, 13 Nov 2002, Pierre Lombard wrote: > Hi, > > The attached patch lets Xv work again after restoring a swsusp-frozen > system. > > Please Cc if necessary as I am not subscribed to Xpert mailing list. > I'm not sure I understand the point of this patch. NVResetVideo doesn't really do anything other than reset the brightness and contrast of the video overlay. Is this an early revision NV11 you have? A laptop maybe? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert][PATCH] Radeon M9 cursor fix
On 22 Nov 2002, Michel Dänzer wrote: > On Don, 2002-11-21 at 18:12, George Staikos wrote: > >Attached is a patch to uncorrupt the mouse pointer on PowerPC based > > laptops with Radeon M9 chipsets. I'm not sure if there are other chipsets > > that need this or not as this is the only one I have, but it fixes things > > for me. > > IMHO this makes ugly code (bit masks and shifts to compensate for the > framebuffer aperture byte swapping...) even uglier. I'd prefer disabling > that byte swapping (as is done in the radeon Xv code and I proposed for > ARGB cursors), using another aperture or a surface. > > Moreover, this is probably a bandaid for the real problem, which I > suspect also causes the color problems people are experiencing with > Radeon 9000s on big endian machines: the byte swapping doesn't seem to > work as expected. > Michel, note that Keith fixed a BigEndian client bug with ARGB cursors the other day. SW cursor works now. If your HW cursor support worked with the broken cursors it probably doesn't work anymore with the correct ones. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Sun(Sony)GDM-1662B
On 9 Nov 2002, Son of Zev wrote: > Hi Mark, > > Sorry about the late response. > > I did find this site...listed in earlier postings.. > > http://cvs.anu.edu.au/monitorconversion/ > > implying that it would be possible and had actually been done.. Yes, with external electronics you can get a sync-on-green or composite sync monitor to work with a separate sync VGA card. I made my own sync-on-green circuit for an IBM 19" workstation monitor once. Some PC video cards have an option for sync-on-green and/or composite sync, and some of the XFree86 drivers support options. Mark. > > The included modeline does generate a response from the monitor (it > flickers other than giving nothing at all... which is a start) but I > can't seem to edit the modeline succesfully to get it to work any > better. > > Considering I was given the monitor it's no big fuss but would be nice > to have it inuse as a second monitor. > > cheers > > Allan > > On Fri, 2002-10-18 at 05:28, Mark Vojkovich wrote: > > On 15 Oct 2002, Son of Zev wrote: > > > > > HI > > > > > > Has anyone actually got this one going... > > > > > > I've seen posts that claim to have working modelines on various websites > > > but none seem to work. > > > > > > >This is not just a fixed frequency monitor. It is a sync-on-green > > or composite sync monitor. Your video card must be capable of outputting > > sync-on-green or composite sync and the drivers for it must be > > capable of turning it on. It will likely not work with video cards > > without this capability. > > > > > > Mark. > > ___ > > Xpert mailing list > > [EMAIL PROTECTED] > > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Option "Rotate" doesn't work
Some drivers have Option "Rotate", they would swap the width and height passed to fbScreenInit, subsequently, this would result in the root window size being rotated. However, this no longer works. Swapping the width and height passed to fbScreenInit does not change the root window size in the ConnectionBlock, and subsequently "Rotate" capable drivers like "nv" break. How are drivers supposed to change the advertised root window size now? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Mouse Cursors
On Thu, 21 Nov 2002, Catherine Luedecker wrote: > Speaking of cursors.are there other cursors available for X and where > are they located? > The core cursors just use the bitmaps that get installed in /usr/X11R6/include/X11/bitmaps. The Alpha blended cursors with CVS (and soon 4.3) appear to get installed in /usr/X11R6/lib/X11/icons/ Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
RE: [Xpert]unresolved symbols
On Tue, 19 Nov 2002, Max Z. wrote: > I recompiled my kernel to 2.4.19 before I started on X. Could that be it? > Before I started configuring X I also upgraded it to 4.2.1 from 4.2.0. > Should I start over with 4.2.0 and a stock kernel? It's got nothing to do with the kernel. The upgrade from 4.2.0 to 4.2.1 was incomplete. It only installed the server, but not the modules. I'd try redoing that 4.2.1 install. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
RE: [Xpert]unresolved symbols
On Tue, 19 Nov 2002, Max Z. wrote: > I have also tried to run the "nv" driver you suggested, and got a lot of > errors: > > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol FindClientResourcesByType from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol ResourceNames from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol RegisterResourceName from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol lastResourceType from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol ResourceNames from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol lastResourceType from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol lastResourceType from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol FindAllClientResources from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol lastResourceType from module > /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved! > Symbol TypeMask from module /usr/X11R6/lib/modules/extensions/libextmod.a is > unresolved! > > > But then I tried something. I commented out 'Load "extmod"' and X started! > But I cant help but wander if extmod is necessary. What modules are > necessary? > Extmod isn't strictly necessary, but alot of people wouldn't be happy without the things it provides - The XVideo extension, The VidMode extension, the shaped window extension, etc... You should really get your installation fixed. Extmod is the most important module. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
RE: [Xpert]unresolved symbols
On Tue, 19 Nov 2002, Max Z. wrote: > There is a mismatch: > > XFree86 Version 4.2.1 / X Window System > ... > (II) LoadModule: "extmod" > (II) Loading /usr/X11R6/lib/modules/extensions/libextmod.a > (II) Module extmod: vendor="The XFree86 Project" > compiled for 4.2.0, module version = 1.0.0 > Module class: XFree86 Server Extension > ABI class: XFree86 Server Extension, version 0.1 > ... > (II) Loading extension X-Resource > > > Anything I can do? > Did you upgrade XFree86 or something? It looks like you have a 4.2.1 server but all of your modules are from 4.2.0. Looks like a botched XFree86 install/upgrade. Furthermore, since the module looks for the XResource extension functions and can't seem to find them, it would seem that the "4.2.0" modules were actually built with something newer than 4.2.0 (since there wasn't XResource support in 4.2.0) and the 4.2.1 server was built without support for it. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]unresolved symbols
On Tue, 19 Nov 2002, Max Z. wrote:
> I am trying to get XFree running under RH8 with a GeForce 4 440 go. X
> would not work with the nv driver, so i am trying to get the nvidia
The "nv" driver will work with GeForce 4 440 Go, but only if you
build the server from CVS. That hardware is newer than the last
official XFree86 release. Actually, there's a prebuilt nv_drv.o
on my web page at http://www.xfree86.org/~mvojkovi/nv_drv.o that
should work but you'll have to pass the -ignoreABI switch to the
server in order to get it to load a module that's newer than the
server ("startx -- -ignoreABI").
> driver to work. I compiled the nvidia drivers and it is running as a
> module perfectly fine, but X still goes down. with the following errors.
>
> (II) Loading extension NV-CONTROL
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol FindClientResourcesByType from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol ResourceNames from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol RegisterResourceName from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol lastResourceType from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol ResourceNames from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol lastResourceType from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol lastResourceType from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol FindAllClientResources from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol lastResourceType from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
> Symbol TypeMask from module
> /usr/X11R6/lib/modules/extensions/libextmod.a is unresolved!
I'm not sure how something like that can happen. Those
symbols are in the core server - they can't be unresolved.
Hmmm. Actually, it looks like that would only fail if your
extmod was built with XResExtension support but the server wasn't.
It looks like you have a module mismatch.
What version do you have for your exmod (ie. from the /var/log/XFree86.0.log)
(II) LoadModule: "extmod"
(II) Loading /usr/X11R6/lib/modules/extensions/libextmod.a
(II) Module extmod: vendor="The XFree86 Project"
compiled for 4.2.99.2, module version = 1.0.0
Module class: XFree86 Server Extension
ABI class: XFree86 Server Extension, version 0.1
Compared to the version of your X-server (ie from the /var/log/XFree86.0.log)
XFree86 Version 4.2.99.2 / X Window System
And after printing out that piece about the extmod module above,
do you see a line like the following:
(II) Loading extension X-Resource
Mark.
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Xserver Debug ...
On Tue, 19 Nov 2002, xpert wrote: > I assume this question to be asked lots of times in the past. > Unfortunately I tried searching the database and couldn't find an > answer. Can anyone Pl. let me know the steps involved in Debugging > Xserver. You need the version of gdb hacked for XFree86 modules. It's at ftp.xfree86.org in /pub/xpert/gdb. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Antialias with video overlay ?
On Thu, 14 Nov 2002, David Balazic wrote: > Video overlays perform some filtering , right ? > > antialias is usualy = supersampling + low-pass filtering. > supersampling : use higher resolution > low-pass : use overlay hardware > > Am I missing something ? > I admit I don't know how usually overlay filtering/resizing works. Here are some facts: Many video hardware don't support downscaling (too much bandwidth). Of the ones that do downscale, many don't filter (too much bandwidth). The filters, at least on newer hardware, are not linear filters. They are higher order filters, which might cause some ringing artifacts when trying to do this. I haven't thought about it much. I don't know of hardware that supports more than 2048 x 2048 overlay sources. That means for a 2x2 AA (which is pretty lame for text rendering) you'd be stuck with a maximum screen size of 1024x1024. Many hardware only support 1024x1024 and much of the older hardware don't have the bandwidth to display that. This sort of fullscreen AA would require that you expand all 2D rendering (including core primitives) to larger sizes so they downsample to the correct size. This would effect all rendering. How this interacts with things like OpenGL is complicated. Assuming you could solve the technical issues with OpenGL and 2D primitive expansion, I expect the quality of a 2x2 supersampling wouldn't be acceptable for text rendering, and the current video hardware wouldn't allow this to work for desktop sizes that people use. Besides, video overlays will likely be disappearing from hardware sometime in the not-so-distant future since the operating systems with majority market share are abandoning them. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Problems with an old monitor and XFree86
My guess would be that it's the adaptor. I've seen streaking and ghosting with poor cables/adaptors. Mark. On Wed, 13 Nov 2002, Frederic Marchand wrote: > Hello there! > > I'm having some issues with an old Apple "Two-page monochrome" monitor. It's a > 21" fixed frequency greyscale. It got a 13w3 to DB15 adapter + a DB15 to DB9. > > Here are my entry in my XF86Config: > This monitor got a 100Mhz dot clock > > Section "Monitor" > Identifier "secondaire" > VendorName "Apple" > ModelName "Twopage monochrome" > HorizSync 68.7 > VertRefresh 75 > Modeline "test" 100 1152 1208 1424 1456 870 870 872 916 -hsync -vsync > EndSection > > Section "Screen" > Identifier "pci/secondaire" > Device "pci" > Monitor "secondaire" > DefaultDepth 8 > SubSection "Display" > Depth 8 > Modes "test" > Visual "GrayScale" > EndSubSection > EndSection > > Section "Device" > Identifier "pci" > Driver "nv" > VendorName "NVidia" > BoardName "TNT 2 M64" > BusID "PCI:0:11:0" > EndSection > > By the way, this monitor work perfectly on a mac, so here is my problem: > > Look at this picture: http://www.mautadine.com/images/appletwopage.jpg > > Look at the titlebar of the xterm window, there is an horizontal band which is > whiter than the background... It's very annoying when a lot of windows are > opened. And i had the same problem with an AppleVision 1710 using the same > adapter DB15->DB9. > > I'm not sure if it can be a modeline problem or simply the adapter... Since my > video card work nicely with my main monitor. Anyone had similar problems? > > I forgot to mention that this is on linux and XFree86 4.2.1. > > Thanks a lot! > > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Antialias with video overlay ?
On Tue, 12 Nov 2002, David Balazic wrote: > Hi! > > I've been toying with this idea to use the video overlay feature > as an efficient anti alias method. Is this doable ? > > Here is how it would work : > > set the display to some resolution ( higher is better ) > render your picture ( at a lower , same or higher resolution than the > display ) > use the overlay to scale the picture to the physical screen > > cards with good scaling ( filtering etc. ) would make a nice > smoothed picture , right ? Blurrying pictures is not antialiasing them. Antialiasing is subpixel rendering implemented by calculating partial-pixel coverage and mixing the partial-pixel coverages to get an unaliased pixel. Video overlays don't do this. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]New xtest failure
We're getting worse... In 4.2.0 we shipped with three failures that are still present: Tests for XChangeKeyboardControl Test 9: FAIL Led mask was 80080004, expecting 8008000f Test 10: FAIL All leds were not set to correct value led_mask was 0xffe4, expecting 0x Tests for XRebindKeysym Test 1: FAIL XRebindKeysym() bound XK_A to "A" instead of "String to bind to modified key". The last one didn't appear if you didn't load the xkb module. We now have a new failure: Tests for XSync Test 2: FAIL --- Running test with pixmap depth 24 Missing error event Unexpected error BadPixmap BadPixmap (invalid Pixmap parameter) Protocol request was X_FreePixmap Unexpected Xlib error Can whoever broke these please fix them? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Option "Rotate" doesn't work
Some drivers have Option "Rotate", they would swap the width and height passed to fbScreenInit, subsequently, this would result in the root window size being rotated. However, this no longer works. Swapping the width and height passed to fbScreenInit does not change the root window size in the ConnectionBlock, and subsequently "Rotate" capable drivers like "nv" break. How are drivers supposed to change the advertised root window size now? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]"nv" driver on Linux PowerPC
If you've had problems with this, you might try this with CVS now. There was an issue where open firmware set the hardware to big endian mode, but the "nv" driver also set it to big endian mode and there was a hardware "feature" associated with setting the chip to big endian mode when it was already in big endian mode. This is also most likely the cause of the problem where you'd get a black screen when you run X twice. Mark. PS. alpha blended cursors don't work properly on big endian at the moment so don't be alarmed if they don't look right. It's expected behavior for the moment. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Alpha cursors broken on big endian
On 8 Nov 2002, Michel Dänzer wrote: > On Fre, 2002-11-08 at 22:24, Mark Vojkovich wrote: > > On 8 Nov 2002, Michel Dänzer wrote: > > > > > On Fre, 2002-11-08 at 21:53, Mark Vojkovich wrote: > > > > On Fri, 8 Nov 2002, Mark Vojkovich wrote: > > > > > > > > >Running top-of-tree on Linux PPC I'm finding that alpha blended > > > > > SW cursors do not seem to work at all. All cursors appear blank. > > > > > > > > > > > > >Actually depending on the color I do sometimes see stuff. The > > > > gumby cursor ends up being lavender. Also, the data passed for > > > > the HW alpha cursor isn't correct either. > > > > > > It's easy to make the radeon driver work with it, didn't you see the > > > patch I posted? > > > > > > >What did you patch? The SOFTWARE cursor is broken and it doesn't > > have anything to do with the drivers. Subsequently, the data passed to > > the driver for the HW interface is also incorrect. > > I don't follow. The fact that the software cursor is broken doesn't > imply that the data is incorrect (is its endianness defined?) - it might > be the blending code treating it incorrectly. Then perhaps we should define what format the HW cursor is supposed to be in. This data comes from the client doesn't it? Does the protocol specify the endian it's supposed to be in? At the moment I'm assuming the data passed to the HW cursor interface will change when the SW cursor gets fixed. Maybe not though. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Alpha cursors broken on big endian
On 8 Nov 2002, Michel Dänzer wrote: > On Fre, 2002-11-08 at 21:53, Mark Vojkovich wrote: > > On Fri, 8 Nov 2002, Mark Vojkovich wrote: > > > > >Running top-of-tree on Linux PPC I'm finding that alpha blended > > > SW cursors do not seem to work at all. All cursors appear blank. > > > > > > >Actually depending on the color I do sometimes see stuff. The > > gumby cursor ends up being lavender. Also, the data passed for > > the HW alpha cursor isn't correct either. > > It's easy to make the radeon driver work with it, didn't you see the > patch I posted? > What did you patch? The SOFTWARE cursor is broken and it doesn't have anything to do with the drivers. Subsequently, the data passed to the driver for the HW interface is also incorrect. If you've patched your driver to work with this incorrect data, that was the wrong solution to the problem. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Alpha cursors broken on big endian
On Fri, 8 Nov 2002, Mark Vojkovich wrote: >Running top-of-tree on Linux PPC I'm finding that alpha blended > SW cursors do not seem to work at all. All cursors appear blank. > Actually depending on the color I do sometimes see stuff. The gumby cursor ends up being lavender. Also, the data passed for the HW alpha cursor isn't correct either. Is it expected that this stuff shouldn't work on Big Endian? If so, we should disable it for BE. As it is now, things like X-term have mising cursors by default on Linux PPC. MArk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Alpha cursors broken on big endian
Running top-of-tree on Linux PPC I'm finding that alpha blended SW cursors do not seem to work at all. All cursors appear blank. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]How to disable doublescan modes ?
On 7 Nov 2002, Diego SANTA CRUZ wrote: > Hi there, > > Whenever I try a low res mode like 800x600 on my laptop, XFree86 > automatically selects a doublescan mode, at the expense of a higher > vertical frequency (I get doublescan @ 60 Hz, but I would prefer > singlescan @ 85 Hz or better). I think it is because XFree picks the > mode with the highest dot-clock that still fits the monitor and video > card specs, instead of the highest vertical frequency. It's probably an error that it picks a doublescan 60 Hz mode over a non-doublescan 85 Hz mode. We should probably remove the 60 Hz doublescan modes from the list and leave only the high frequency doublescan modes, or modify the logic that chooses the modes. > > Is there any way to disable doublescan modes? Or to modify the order of > preference for selecting modes at a given resolution? I looked on the > man pages but I did not find anything. The only way to override XFree86's internal mode is to provide a mode by the same name in your Monitor section. > > I know I could do it by listing all the desired modes in the monitor > section and then specifying those, but I would like to avoid duplicating > XFree's internal mode list. However, if this is the only currently > available solution, where can I get the definition of standard modes > used by XFree internally? The list is in the server tree at xc/programs/Xserver/hw/xfree86/common/modedefs Modeline "800x600" 56.3 800 832 896 1048 600 601 604 631 should do it. That's 85 Hz. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Re: But *why* no vblank?
On 6 Nov 2002, Michel Dänzer wrote: > On Mit, 2002-11-06 at 17:39, Billy Biggs wrote: > > Michel Dänzer ([EMAIL PROTECTED]): > > > > > > It would be preferable in general for video apps, though, to provide > > > > a DRM-based api to use the overlay buffer, too. Like, a DRM-Xv. > > > > For desktop use, the X11 context switch may be fairly acceptable > > > > with something like XSYNC, but to achieve really excellent quality > > > > (eg, suitable for output to a TV/broadcast/etc.) in, say, a video > > > > player, a direct API would be nicer. > > > > > > If I'm not mistaken that's what XvMC is for. > > > > No, XvMC is an API to hardware motion compensation, basically for > > hardware MPEG decoding. > > Don't let the name mislead you. Sure, motion compensation was probably > the initial motivation, but my understanding from reading posts about it > on Xpert (Mark or someone please correct me if I'm wrong) is that it > supports non-MC surfaces, so it's basically a beefed up Xv which > supports MC and direct rendering. It is direct rendered, but the API allows only acceleration of mpeg decoding - either at the mocomp of IDCT level. The implementation details of overlays, image storage, etc... are opaque. Moreso than Xv. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Shadowfb rotation broken?
I've noticed that Option "Rotate" mostly doesn't work anymore in the "nv" driver, and code related to that hasn't been touched since before 4.2. It seems that behavior related to viewports has changed. Is this a universal problem with "Rotate" drivers? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]nv driver crash on nForce
On 4 Nov 2002, Mikkel Lauritsen wrote: > On Sun, 2002-11-03 at 22:26, Mark Vojkovich wrote: > > On 3 Nov 2002, Mikkel Lauritsen wrote: > > --- snip --- > > > > As mentioned in RedHat bug 75018, Mike Harris has a new build from CVS > > > at ftp://people.redhat.com/mharris . This build fixes the nForce crash > > > but introduces a new bug where (at least) the Mozilla throbber and > > > scrollbars become garbled. Still, it eliminates the need for using the > > > -ignoreABI parameter. > > > >What's this? I know of no rendering problems on nForce. Did > > it detect the amount of video ram correctly? Ie. the server reports > > the amount configured in the bios? > > Yup, looks like it. The weird bit is that so far Mozilla is the > only application that seems to be affected by the bug; Phoenix > and other programs that I have tried have no problems at all. > > I have created a bug in Redhat's Bugzilla including a log file and > a screenshot showing the corruption; it's bug #77300 - > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=77300 > That just looks like a bad pitch. Maybe a client bug with an XImage. Does this app use the RENDER extension? If so the "nv" driver is out of the picture since it doesn't accelerate anything related to RENDER. If it's very reproducible, try adding Option "NoAccel". If it doesn't go away, then it's a client bug or a bug in XFree86's software rendering code, which would be unlikely unless maybe the render extension is involved. If it does go away with "NoAccel", try with acceleration enabled but with either Options "XaaNoOffscreenPixmaps" or "XaaNoPixmapCache". If either of those fix it send me your /var/log/XFree86.0.log. That would indicate that offscreen memory was getting corrupted, which a likely cause of could be the driver probing the memory size incorrectly. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]XFree86 gdb for Linux PPC?
Is there an XFree86 module-aware version of gdb for Linux PPC someplace? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]nv driver crash on nForce
On 3 Nov 2002, Mikkel Lauritsen wrote: > On Fri, 2002-11-01 at 12:33, Vasco Alexandre Da Silva Costa wrote: > > > The XFree86 drivers in the XFree86-4.2.0-72 RPM package which comes with > > RedHat 8.0 do not work with nForce properly. > > As mentioned in RedHat bug 75018, Mike Harris has a new build from CVS > at ftp://people.redhat.com/mharris . This build fixes the nForce crash > but introduces a new bug where (at least) the Mozilla throbber and > scrollbars become garbled. Still, it eliminates the need for using the > -ignoreABI parameter. What's this? I know of no rendering problems on nForce. Did it detect the amount of video ram correctly? Ie. the server reports the amount configured in the bios? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]But *why* no vblank?
On Fri, 1 Nov 2002, Keith Packard wrote: > Around 15 o'clock on Nov 1, Mark Vojkovich wrote: > > > > How about a signal when vertical blanking arrives? (1st choice) That, or > > > something we can select on? (2nd choice) > > > > What are you trying to solve? The problem of syncing XCopyArea > > to the retrace? If so, how does this lead to a solution? > > Hook the signal up to an XSYNC counter, block the client on the counter > with the XCopyArea request sitting in the request buffer. Signal fires, > the client is scheduled and the CopyArea executes. What? Did you just say you're going to delay flushing the data to the server until some time after the retrace happens (there's a significant latency there) and you think the X-server is actually going to execute that almost immediately? If so, that will totally not work. The X-server might not even be scheduled anytime soon. Even the X-server blocking until an ioctl doesn't work very well. Not to mention that my hardware doesn't even work this way. The burden of retrace syncing needs to be left entirely to the driver. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]But *why* no vblank?
On Fri, 1 Nov 2002 [EMAIL PROTECTED] wrote: > I've just chatted with Keith Packard on this. > > This interface (an ioctl that blocks) isn't a good one. > > How about a signal when vertical blanking arrives? (1st choice) That, or > something we can select on? (2nd choice) > > - Jim Gettys What are you trying to solve? The problem of syncing XCopyArea to the retrace? If so, how does this lead to a solution? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]But *why* no vblank?
On Fri, 1 Nov 2002, Scott Long wrote: > I guess I'm proposing some sort of X extension that allows a > particular X request to be tagged as "synced to vblank." For example a > XCopyArea request could be preceded by another request (call it > XRequestVerticalSync) that indicates to the server that the operation > shouldn't be performed until a vblank period. > > This is a very simple change. If no driver support exists for vertical > sync, the server can just politely ignore the request, and do the > operation whenever it feels like it. This would at the very least > provide a framework for individual driver writers to start working on > a kernel interface. > Your "XRequestVerticalSync" implies that this would be applied to only this client's next request. This would mean that the dispatch level code would have to make note of this, and make a call into the driver before the next request from the same client. It can't call into the driver as soon as it gets it, because the next requests it gets may be from different clients. Anyhow. If there was such a entry point into the driver, and the top-level dispatch code could delay calling it until the correct point, I could support this in NVIDIA's binary drivers. You'd have to define which request XRequestVerticalSync acted on. Ie. any request, or only certain requests. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Vertical retrace and Pixmaps
On Thu, 31 Oct 2002 [EMAIL PROTECTED] wrote: > > > > 2.If both the Xserver and the client are on the same machine then which > > one > > > gives better performance ie in terms of speed ...Pixmaps or shared memory > > > pixmaps.Are pixmaps stored directly in the video RAM? > > > >That's too open-ended to answer. It all depends what you are > > trying to do. > > probably i was not clear on this. > I have to display a number of pic files one after the other.So i want to > optimise the delay for the display. > Right now i am just creating images and then making a shared memory pixmap. > I do a Xputimage and Xcopyarea to display a picture. I assume you mean XShmPutImage. XShmPutImage is considerably faster than XPutImage when the connection is local. > For this kind of application is Pixmap better or shared memory pixmap. > The fastest way to move image data to a Window is to do XShmPutImage of the XImage into the Window. Making the XImage data into a shared memory pixmap and then calling XCopyArea is a round-about way of doing the same thing, only more complicated and probably slower. The OpenGL equivalent would be to do glDrawPixels into the Window. For double buffering, it would be glDrawPixels into the back buffer and then glXSwapBuffers. Some implementations may sync glXSwapBuffers to the vertical retrace. With NVIDIA's binary Linux drivers setting the environment variable __GL_SYNC_TO_VBLANK = 1 would turn vblank on syncing. You could double buffer in X by XShmPutImage into a Pixmap and then XCopyArea from the Pixmap into the Window. Having the Pixmap NOT be in shared memory would be fastest there. Unfortunately, this will not sync to the vblank and there is no way to make it so. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Vertical retrace and Pixmaps
On Thu, 31 Oct 2002 [EMAIL PROTECTED] wrote: > Hi, > > I am a student working on X am i am pretty new to it. > I have a few questions > > 1. Is there any way to detect vertical retrace in X so that the tearing down of > he image can be prevented ? You mean for XPutImage or something? No. The only things that can sync to the retrace are some X-Video extension implementations and some OpenGL implementations. > > 2.If both the Xserver and the client are on the same machine then which one > gives better performance ie in terms of speed ...Pixmaps or shared memory > pixmaps.Are pixmaps stored directly in the video RAM? That's too open-ended to answer. It all depends what you are trying to do. MArk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86 and LCd
On Wed, 30 Oct 2002, Yongming Wei wrote:
> My LCD is philips 150S3f.
>
> accurately, the flickering is not serious, although
> characters can be seen a little messed up, I can still
> recognize them.
>
> Now I want inscreasing DotClock to see if it is
> because the DotClock is too low.
Flickering is generally because the refresh is too low. If
you increase the dot clock you will increase the refresh.
According to the PDF for this monitor at Philips web site
1024x786@75Hz IS a supported mode. So it doesn't sound
like there was anything wrong with the original mode you
were using.
Mark.
>
>
>
> --- Mark Vojkovich <[EMAIL PROTECTED]> wrote:
> > On Wed, 30 Oct 2002, Yongming Wei wrote:
> >
> > > Thanks, I works after change VertRefresh to 60,
> > but
> > > after that, my monitor begins to flicker, and I
> > found
> > > the dotrate degrade from 75 to 60, maybe it is the
> > > reason of this problem.
> > >
> > > How to avoid it?
> >
> >60 Hz Is expected to flicker on a CRT, but
> > usually not on an LCD.
> > Which LCD panel is this?
> >
> > Mark.
> >
> > >
> > > Regards,
> > >
> > > --- Mark Vojkovich <[EMAIL PROTECTED]> wrote:
> > > > On Tue, 29 Oct 2002, Yongming Wei wrote:
> > > >
> > > > > I changed my CRT to LCD, it works well in my
> > > > redhat
> > > > > 8.0, but I found a problem: the manual for
> > this
> > > > LCD
> > > > > says the refresh rate is recommended set to
> > 60HZ
> > > > at
> > > > > 1024*768, but XFree86 will automaticly search
> > for
> > > > the
> > > > > highest refresh rate possiblely which is 75HZ
> > in
> > > > my
> > > > > box.
> > > > >
> > > > > Can somebody point me the way to change the
> > > > refresh
> > > > > rate to 60HZ?
> > > > >
> > > >
> > > >Just specify in the Section "Monitor" of the
> > > > XF86Config file:
> > > >
> > > > VertRefresh 60
> > > >
> > > >and you'll only get 60 Hz modes.
> > > >
> > > >
> > > > Mark.
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]can't restart X
On Wed, 30 Oct 2002, Fred Heitkamp wrote: > I have a snakebite dual G4 Powermac. > I have Debian installed with XFree from CVS that is > just a week or so old. The problem I have has occurred > with all versions of X that I have tried. > My Mac has the GForce 2MX video card. > X seems to run OK but if I "log out" of X and try > to restart my display goes black. > I have to log in from another computer and reboot. > A wild guess is that the video card is left > in a weird state. Anyone else had a similar > problem? > I've seen that before. I was under the impression it had something to do with the framebuffer device. Are you using Offb or rivafb? Not that there aren't problems with both as far as I can tell. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86 and LCd
On Wed, 30 Oct 2002, Yongming Wei wrote: > Thanks, I works after change VertRefresh to 60, but > after that, my monitor begins to flicker, and I found > the dotrate degrade from 75 to 60, maybe it is the > reason of this problem. > > How to avoid it? 60 Hz Is expected to flicker on a CRT, but usually not on an LCD. Which LCD panel is this? Mark. > > Regards, > > --- Mark Vojkovich <[EMAIL PROTECTED]> wrote: > > On Tue, 29 Oct 2002, Yongming Wei wrote: > > > > > I changed my CRT to LCD, it works well in my > > redhat > > > 8.0, but I found a problem: the manual for this > > LCD > > > says the refresh rate is recommended set to 60HZ > > at > > > 1024*768, but XFree86 will automaticly search for > > the > > > highest refresh rate possiblely which is 75HZ in > > my > > > box. > > > > > > Can somebody point me the way to change the > > refresh > > > rate to 60HZ? > > > > > > >Just specify in the Section "Monitor" of the > > XF86Config file: > > > > VertRefresh 60 > > > >and you'll only get 60 Hz modes. > > > > > > Mark. > > ___ > > Xpert mailing list > > [EMAIL PROTECTED] > > http://XFree86.Org/mailman/listinfo/xpert > > > __ > Do you Yahoo!? > HotJobs - Search new jobs daily now > http://hotjobs.yahoo.com/ > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]xReq and the client!!!!!
On Wed, 30 Oct 2002, Preetham wrote: > What does the client->index point to. You mean in the ClientRec in the server? That is the client number. It ranges from 0 to (MAXCLIENTS - 1). > The "algorithm" that you referred to, is it to do with the rendering > requests clients send to the Xserver. No. Each client has a separate socket connection to the X-server. The X-server knows which client the request comes from because it knows which socket it came from, therefore there is no need for any client identifcation in the protocol stream itself. The X-server identifies clients by their sockets. > Can i use the record extension to get to know all clients that are connected > to the Xserver display. I sorry, but I don't know anything about the record extension. The X-Resource extension which I added into CVS will tell this, but of course, since it's only in CVS it's not widely available on most peoples systems. Mark. > Thanks, > Preetham. > > - Original Message - > From: "Mark Vojkovich" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Friday, October 25, 2002 12:41 PM > Subject: Re: [Xpert]xReq and the client! > > > > On Fri, 25 Oct 2002, Preetham wrote: > > > > > Hi All: > > > i would like to know the Client that sent a particular xReq to the > > > server. When the X Client sends a request to the XServer, the xRqq > > > contains the Type and the Data. Is the X Client ID(or some kind of > > > identification) present in the data. > > > Appreciate the help. > > > Thanks, > > > Preetham. > > > >I don't think the client really even knows its client ID, > > though it can be derived from the resources it creates because > > we know the algorithm. I believe the X-server has associated a > > client ID with a socket and that info resides on its end of the pipe. > > Even when deriving the client ID from the resources, this is not > > foolproof because not all requests have XIDs in them and clients > > can reference XIDs created by other clients. Also, the algorithm > > can differ between XFree86 and other servers. > > > > > > Mark. > > ___ > > Xpert mailing list > > [EMAIL PROTECTED] > > http://XFree86.Org/mailman/listinfo/xpert > > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XFree86 and LCd
On Tue, 29 Oct 2002, Yongming Wei wrote: > I changed my CRT to LCD, it works well in my redhat > 8.0, but I found a problem: the manual for this LCD > says the refresh rate is recommended set to 60HZ at > 1024*768, but XFree86 will automaticly search for the > highest refresh rate possiblely which is 75HZ in my > box. > > Can somebody point me the way to change the refresh > rate to 60HZ? > Just specify in the Section "Monitor" of the XF86Config file: VertRefresh 60 and you'll only get 60 Hz modes. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Geforce4 support for big endian in nv driver ?
On Tue, 29 Oct 2002 [EMAIL PROTECTED] wrote: > Hi ! > > I am currently experimenting with the nv driver on the ppc platform. My hardware >target is an Apple imac G4 with a Geforce4 mx card. > > All linux ppc distros report this setup as unsupported with XFree 4.2.x, but XFree86 >cvs code seems to support this card on x86. So I build a fresh cvs on the imac, with >no success (color strips..) The card is recognized as a Geforce 4 440 Go by the way, >is this card supported on x86 ? > It works on x86. I'm not aware of any endian problems in the current "nv" driver. I think I even got an NV17 working at one time, but it was highly dependent on the fcode. The offb driver I was using clearly had bugs that would prevent nvidia cards with newer fcode (with child nodes for the individual heads) from working properly. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]stopping screen blanks
On Tue, 29 Oct 2002 [EMAIL PROTECTED] wrote: > How can I stop X from blanking the screen after a period of inactivity? I have > a laptop with a builtin mouse and keyboard, and an external USB mouse. When I > use the external mouse, it doesn't count it as being "activity", so I'll be in > the middle of browsing a web page and the screen will blank. > > Can I disable this? I tried startx -- -s 0 and that didn't seem to help. > You could do it with xset. 'xset s off'. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]X freezes
Add Option "NoRenderAccel" to the Section "Device" for the nvidia card. There's no test suite for the render extension and the driver is attempting to accelerate some stuff that only now has applications using it. Mark. On Thu, 24 Oct 2002, patrick rutka wrote: > hi, > > I have some problemes with X on Mandrake 9.0 (with kernel recompil for athon), > and i never have this before. > I don't explain why (it's seem to be randomly), but my X freez often (keyboard > don't work but sound and network work) and i must reboot the computer. I > have a another pc and i try to connect wtih ssh to kill X (i before killall > other application (like konqueror kde) over X to unlock the situation > but nothing worked...) and when i try to kill X my machin (this frozen) that > lock my ssh session from the second computeur and the other ssh call not work > so my linux seems to be dead when i try to kill X by another pc when it > freez. When my X freez, I have the mouse fonctional but the backgroud is lock > (the clock is froezn and all graphical applications too) and i don't goto > the second screen (i have 2 screens). > > X freez locks in all cases : (1 screen or 2 screen, nv or nvidia driver too) > and for now it only lock on my first card (nvidia geforce, but i don't be > often on my second screen...) > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Only 16 bpp mode via DVI, when 24 bpp is shown
On Sat, 26 Oct 2002, Holger Isenberg wrote:
> Ok, I checked again with starting the server in 16bpp mode where
> 32 of the 64 graylevels looked greenish as described.
> However, in 24bpp there are still these 64 graylevels, but all really gray.
> I also tried the GATOS ATI.2 radeon/r128 module with same result.
>
> So it seems, that the DAC is set into 6bit per channel mode, and displayed
> wrong as:
> (==) RADEON(0): RGB weight 888
> (II) RADEON(0): Using 8 bits per RGB (8 bit DAC)
You guys are likely misinterpreting what you are seeing. Most
digital flat panels are only 6 bits per component! It's not
the video card that's truncating the data, but the flat panel.
Many video cards have an option to dither 8 down to 6 before the
flat panel truncates it. I just added such an option to the "nv"
driver recently ("FPDither"). It was my impression that only
the really expensive panels are 8 bit, and that pretty much all
of the laptop panels are only 6 bit.
MArk.
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]xReq and the client!!!!!
On Fri, 25 Oct 2002, Preetham wrote: > Hi All: > i would like to know the Client that sent a particular xReq to the > server. When the X Client sends a request to the XServer, the xRqq > contains the Type and the Data. Is the X Client ID(or some kind of > identification) present in the data. > Appreciate the help. > Thanks, > Preetham. I don't think the client really even knows its client ID, though it can be derived from the resources it creates because we know the algorithm. I believe the X-server has associated a client ID with a socket and that info resides on its end of the pipe. Even when deriving the client ID from the resources, this is not foolproof because not all requests have XIDs in them and clients can reference XIDs created by other clients. Also, the algorithm can differ between XFree86 and other servers. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]xsetroot for multiple screens?
On Fri, 25 Oct 2002, Michael wrote: > I'm wondering if there exists something like xsetroot for multiple screens. > Like when you run multiple graphics cards without xinerama, running xsetroot on the >screens individually works fine, but i'd like to pass the commands in .xinitrc for >diffrent images on each screen. > It takes a -display arguement. So perhaps "xsetroot -display :0.0", "xsetroot -display :0.1", etc... will work from the .xinitrc. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]DVI output with 'nv' driver
On Mon, 21 Oct 2002, Richard Weber wrote: > Hello, > > I've been having trouble getting the digital output of > my NVidia GeForce 2 to work with the 'nv' driver > provided with XFree86 4.2.0. When the DVI cable is > connected, all I get is fuzzy blue scrolling down the > screen. DVI doesn't work in 4.2.0. You have to use XFree86 CVS. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]VBlank interrupt?
On Tue, 22 Oct 2002, Tony Wu wrote: > Does XFree86 video driver support VBlank(Vsync) interrupt? If it does, > how to enable it? XFree86 drivers are user-space drivers, and as such cannot service interrupts. Only the kernel can handle interrupts. Since most XFree86 drivers do not have a kernel component they need to disable interrupts (because there's nothing to service them). If you had a kernel module for your video card you could have a device file that clients could use to do a non-spinning block until the retrace. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Re: [Dri-devel] r200 and libxaa
On Sun, 20 Oct 2002, Jens Owen wrote: > Mark Vojkovich wrote: > > On 21 Oct 2002, Michel Dänzer wrote: > >> > >>The problem appeared when I merged radeon driver fixes by Kevin E. > >>Martin from XFree86 CVS to DRI CVS. They involved adding new fields to > >>the XaaInfoRec struct; the fact that they were added in the middle of > >>that struct caused the DRI binary snapshots, which were built against > >>the new version of that struct but ran with an XAA module built against > >>the old version, to crash and burn. I moved the new fields to the end of > >>the struct, bumped the minor XAA version and added code to the radeon > >>driver to only access the new fields if it's dealing with the new XAA. > >>That worked as expected when I tested it here by copying libxaa.a from > >>my Debian 4.2 installation into my DRI tree, which makes me suspect that > >>the remaining problem with the snapshots is due to toolchain > >>incompatibilities or something along those lines. > >> > > > > > >I don't know why people do this. This change just broke NVIDIA's > > and ATI's binary drivers too. Any changes to structures that are > > referenced by drivers need to go at the end of the structures. > > Mark, > > I'd like to see driver level binary compatability continue, too. I'm > not certain if Michael's changes have made it back to XFree86's CVS. I > posted to this thread (and copied Xpert) because I read some posts to > dri-devel that implied Michel's fixes weren't enough to restore driver > compatability (old drivers, new XFree86 core and modules). If I'm > mistaken, please let us know. I just changed the header file in CVS to move the changes to the end. I think that's all that's needed to restore binary compatibility with the old XAA module. I don't see anything else in there that has changed recently that would break things. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Overriding a graphics card
On Sun, 20 Oct 2002, John Tapsell wrote: > > The card needs an interrupt for graphics mode. Does one show up in > /proc/pci? > > If it doesn't have one, that would be a motherboard bios bug. Perhaps > > there's a newer bios available for it? Which motherboard was this? > > I will check in a minute.. > > It's a PC-Chips motherboard with sis chips. There are bios updates, but > I > don't have a dos boot disk to boot into to flash the bios heh. > > Can I flash an ami bios from linux? :) Unfortunately, the very unreal-time behavior of Linux is unsuited to bios flashing. You could write a flasher, but it would probably ruin them much of the time. If this is a common problem, I'd expect google to have some answers. In particular a search for the mother board model, TNT2 and "interrupt" or something like that. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Re: [Dri-devel] r200 and libxaa
On 21 Oct 2002, Michel Dänzer wrote: > On Mon, 2002-10-21 at 02:05, Jens Owen wrote: > > José Fonseca wrote: > > > On Wed, Oct 16, 2002 at 04:52:59PM +1000, Simon Bland wrote: > > > > > >> I've seen ppl mention the libxaa.a from Fonseca seems to fix the problem > > >> with the screen blanking. Thing is I can't seem to find this libxaa.. > > >> Can anyone point me in the right direction for this? And can anyone > > >> confirm that this does seem to fix the screen blanking problem with the > > >> Radeon 8500. > > >> > > > > > > Initially I hosted on my website because there was no certain that it > > > would help. But now that I know it does I've moved it to > > > http://dri.sf.net/snapshots/extras/libxaa.a.bz2 and wrote a README.Sig11 > > > on http://dri.sf.net/snapshots/ . > > > > Can anyone on Xpert comment on XAA binary incompatability? Does this > > just affect Forward compatability, i.e. older X environments with newer > > binary drivers? > > The problem appeared when I merged radeon driver fixes by Kevin E. > Martin from XFree86 CVS to DRI CVS. They involved adding new fields to > the XaaInfoRec struct; the fact that they were added in the middle of > that struct caused the DRI binary snapshots, which were built against > the new version of that struct but ran with an XAA module built against > the old version, to crash and burn. I moved the new fields to the end of > the struct, bumped the minor XAA version and added code to the radeon > driver to only access the new fields if it's dealing with the new XAA. > That worked as expected when I tested it here by copying libxaa.a from > my Debian 4.2 installation into my DRI tree, which makes me suspect that > the remaining problem with the snapshots is due to toolchain > incompatibilities or something along those lines. > I don't know why people do this. This change just broke NVIDIA's and ATI's binary drivers too. Any changes to structures that are referenced by drivers need to go at the end of the structures. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Re: [Dri-devel] r200 and libxaa
On Sun, 20 Oct 2002, Jens Owen wrote: > José Fonseca wrote: > > On Wed, Oct 16, 2002 at 04:52:59PM +1000, Simon Bland wrote: > > > >> I've seen ppl mention the libxaa.a from Fonseca seems to fix the problem > >> with the screen blanking. Thing is I can't seem to find this libxaa.. > >> Can anyone point me in the right direction for this? And can anyone > >> confirm that this does seem to fix the screen blanking problem with the > >> Radeon 8500. > >> > > > > Initially I hosted on my website because there was no certain that it > > would help. But now that I know it does I've moved it to > > http://dri.sf.net/snapshots/extras/libxaa.a.bz2 and wrote a README.Sig11 > > on http://dri.sf.net/snapshots/ . > > Can anyone on Xpert comment on XAA binary incompatability? Does this > just affect Forward compatability, i.e. older X environments with newer > binary drivers? I don't know why you'd be using a different XAA module in the first place. There haven't been any known bugs in XAA in a long time. I'm not sure we have any compatibility rules for anything other than driver interfaces. XAA interacts with more than the driver, it interacts with other modules and the core server. There may be some explicit assumptions that you are not mixing core modules. Precautions have not been made to allow mixing of an XAA module with a server of a different vintage. As for drivers, you can use a driver older than the server and it will generally work. You can use a driver newer than the server only if the driver doesn't use any entry points or exports newer than the server, and even if that part is safe you usually need to pass the -ignoreABI flag. But I don't think we have any goals of binary compability between core modules. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Overriding a graphics card
On Sun, 20 Oct 2002, John Tapsell wrote: > On Sun, Oct 20, 2002 at 02:56:00PM -0700, Mark Vojkovich wrote: > > On Sun, 20 Oct 2002, John Tapsell wrote: > > > > > On Sun, Oct 20, 2002 at 02:33:05PM -0700, Mark Vojkovich wrote: > > > >Are you able to disable the SiS card from the bios? You > > > > should try that if you haven't already. > > > > > > Being crappy SiS - you can't. And no, there's no jumpers on the > mobo. > > > >I assume you got it to post to the nvidia card at least (console > > comes up on the nvidia card)? You could try moving the slot that > > the nvidia card is in, or maybe just reserve the interrupt that it > > is using. I'm assuming you have a shared interrupt or something > > that is making things unhappy. > > The bios comes up on the nvidia card. Linux boots on the nvidia card. > The framebuffer works on the nvidia card. But X doesn't. (I haven't > tried > a framebuffered X) > > When I cat /proc/interrupts, I see the SiS is sharing an interrupt with > the > usb and ethernet card - 11. But I see no mention of the nvidia card. The card needs an interrupt for graphics mode. Does one show up in /proc/pci? If it doesn't have one, that would be a motherboard bios bug. Perhaps there's a newer bios available for it? Which motherboard was this? > > How do I assign an interrupt to the nvidia card? > > My bios has no options for this. The option for "plug and play OS" is > set > to "N", and "reserver IRQ for PCI card" is set to "Y". That's about how I'd expect you to go about doing it. Have you tried changing which PCI slot it's in? That helps get a different interrupt sometimes. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Changing -- to --
On Thu, 17 Oct 2002, Boris wrote:
> I have found the file, What line would I have to change and what do I have
> to change it too?
You'll have to figure out how things work and investigate that.
You can see where the current is handled. It looks
like:
if ((ModifierDown(ControlMask | AltMask)) ||
(ModifierDown(ControlMask | AltLangMask)))
{
switch (specialkey) {
case KEY_BackSpace:
if (!xf86Info.dontZap) {
#ifdef XFreeXDGA
DGAShutdown();
#endif
GiveUp(0);
}
break;
[...]
It *might* be as easy as adding or changing the case statement to
include the delete key.
There may be other ways to do this via configuring the XKB
extension. But I know next to nothing about XKB.
Mark.
>
> - Original Message -----
> From: "Mark Vojkovich" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent: Thursday, October 17, 2002 2:37 PM
> Subject: Re: [Xpert]Changing -- to --
>
>
> > On Thu, 17 Oct 2002, Boris wrote:
> >
> > > Does anyone know how to change exiting XFree from
> --
> > > to --? Any help would be greatful. Even if I have to
> > > hardcode it.
> > >
> >
> > If you wish to hack the source see xf86PostKbdEvent() in
> common/xf86Events.c
> >
> >
> > MArk.
> > ___
> > Xpert mailing list
> > [EMAIL PROTECTED]
> > http://XFree86.Org/mailman/listinfo/xpert
> >
>
> ___
> Xpert mailing list
> [EMAIL PROTECTED]
> http://XFree86.Org/mailman/listinfo/xpert
>
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Nvidia and Suspend
On Fri, 18 Oct 2002, Brad Hards wrote: > On Fri, 18 Oct 2002 05:35, Mark Vojkovich wrote: > >The "nv" driver doesn't know (and can't know) anything about > > suspend events. It's handled entirely by the bios and there is > > no mechanism for XFree86 to get these ACPI events from the kernel. > > Subsequently, the bios will mess up the "nv" driver's state and > > the "nv" driver won't know that it needs to be reinitialized. > > You have to VT switch to clean things up. I think the only solution > > to this problem is to have ACPI support in the kernel and > > have the events routed to /dev/apm (which XFree86 supports) or > > to some other device and have XFree86 add support for that device. > > Tim Hockin wrote a acpid (on sourceforge.net) that can take ACPI events from > the kernel ( via /proc/acpi/event ) and runs things in userspace. That is a > potentially useful approach in this case. > > I see something like this (generalised to handle many other events), along > with the current Linux hotplug style approach, as the path to make X work in > dynamic hardware and networking environments. > I'm wondering if it's feasible for the kernel to route ACPI power management events to the corresponding APM events through /dev/apm (optionally, of course) for some sort of ACPI to APM backwards compatibility for some apps that use /dev/apm (like XFree86). Otherwise somebody will have to add /dev/acpi support to XFree86. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Changing -- to --
On Thu, 17 Oct 2002, Boris wrote: > Does anyone know how to change exiting XFree from -- > to --? Any help would be greatful. Even if I have to > hardcode it. > If you wish to hack the source see xf86PostKbdEvent() in common/xf86Events.c MArk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Nvidia and Suspend
The "nv" driver doesn't know (and can't know) anything about suspend events. It's handled entirely by the bios and there is no mechanism for XFree86 to get these ACPI events from the kernel. Subsequently, the bios will mess up the "nv" driver's state and the "nv" driver won't know that it needs to be reinitialized. You have to VT switch to clean things up. I think the only solution to this problem is to have ACPI support in the kernel and have the events routed to /dev/apm (which XFree86 supports) or to some other device and have XFree86 add support for that device. Mark. On Wed, 16 Oct 2002 [EMAIL PROTECTED] wrote: > I am using Gentoo 1.4 with Xfree 4.2.1, with the patch for the > Geforce2Go on > a dell 8100. > > I am running kernel 2.4.19 patched with the swsusp patch, but have seen > the > same effect on 2.4.18. > > When I resume from a suspend I find that the Gtk theme is not applied > properly, in fact certain widgets do not have any theming applied to > them. > If I switch to another theme and back to the original, the theme is > then > displayed properly. However, after another suspend/resume cycle I > cannot > switch to this theme and back again I have to use another one. > > I looks as though Gtk caches the themes via the driver and the driver > is not restoring or saving its memory properly. > > This effect has been confirmed by a number of people on the swsusp list, > > including one person reporting problems with KDE. > > > Phil > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Sun(Sony)GDM-1662B
On 15 Oct 2002, Son of Zev wrote: > HI > > Has anyone actually got this one going... > > I've seen posts that claim to have working modelines on various websites > but none seem to work. > This is not just a fixed frequency monitor. It is a sync-on-green or composite sync monitor. Your video card must be capable of outputting sync-on-green or composite sync and the drivers for it must be capable of turning it on. It will likely not work with video cards without this capability. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]How is modeline chosen?
On Tue, 15 Oct 2002, Spammers Must Die wrote: > I just got my new 1280x1024 17" flatscreen set up, and > it looked like hell, with moires, aliased vertical > lines, etc.. As seems to be the case after running > XF86Config originally, I had several 1280x1024 > ModeLines in my XF86Config file. The closest modeline > looked like this: > > # 1280x1024 @ 61 Hz, 64.2 kHz hsync > Modeline "1280x1024" 110 1280 1328 1512 1712 1024 1025 > 1028 1054 > > Since the flat screen really wants to have *exactly* > the right pixel > frequency, as a shot in the dark I added these lines > right above that > one: > Modeline "1280x1024" 108 1280 1328 1512 1712 1024 1025 > 1028 1054 > Modeline "1280x1024" 108.5 1280 1328 1512 1712 1024 > 1025 1028 1054 > Modeline "1280x1024" 109 1280 1328 1512 1712 1024 1025 > 1028 1054 > Modeline "1280x1024" 109.5 1280 1328 1512 1712 1024 > 1025 1028 1054 > > I changed only the pixel clock - it's close enough > that the other > parameters didn't seem to need adjustment - and things > look very good now.. > right now my display says it's getting 63.9 kHz > horzontal sync - almost > perfect. (The display wants 63.98 H and 60.02 V.) > > My question is: In the presence of several 1280x1024 > modelines, how is > the modeline chosen? Is XF86 even using the XF86Config > file? I think the selection criteria is as follows: Use the mode in the XF86Config file unless it falls outside of the monitor specs listed in the XF86Config. If no such mode exists, then it goes to the pool of modes internal to XFree86. From that pool, it picks the highest refresh rate mode that falls within the monitor specs. If you have multiple modes by the same name in the XF86Config file, it probably uses the last one. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Double mouse pointer (nv geforce3ti dvi)
On Tue, 15 Oct 2002, Björn Stenberg wrote: > With 4.2.1, the nv driver *finally* produces a correct image on my setup. > > Only one small bug remains: The mouse pointer. It is duplicated and half the right >size. > > Actually, the duplication effect looks more to be a case of every other bit being >drawn 16 pixels too far to the right. The "duplicates" are not exact copies. > > Is this known/fixed? Is there a workaround? Is there anything I can do to help? > > Driver "nv" > Option "FlatPanel" ??? If there is flat panel support in 4.2.1, I don't know how it got there. I didn't put it there. The flat panel support I added was in CVS only. The artifact you describe sounds like what you get when you run the "nv" driver after running the "nvidia" driver - a problem which doesn't exist in CVS (or if you don't run the "nvidia" driver). But as far as I know, DFP isn't supported in 4.2.1 in the first place. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]libXxf86vm.a vs. libXxf86vm.so
On Tue, 15 Oct 2002, Alexey Gromov wrote: > I've tried to run wine with X support and have got a following error: > x11drv.dll.so: undefined symbol: XF86VidModeSetGammaRamp > > I've use ldd to find out the libraries x11drv.dll.so uses... > ... libXxf86vm.so.1 was in the list. > I've made > > string libXxf86vm.so.1 | grep XF86VidModeSetGamma > > The result was only > XF86VidModeSetGamma > > !!! There's really no such symbol XF86VidModeSetGammaRamp in this > library... I was just curious and looked at the output of > > string libXxf86vm.so.1 | grep XF86VidModeSetGamma > > XF86VidModeSetGamma > XF86VidModeSetGammaRamp > XF86VidModeSetGamma > XF86VidModeSetGammaRamp > > So, the XF86VidModeSetGammaRamp symbol is IN the static version of the > library but not in the dynamic one... Have I misconfigured XFree > somehow, or is it a feature, not a bug? > > P.S. I'm using XFree86 4.2.1 compiled from source under i686-gnu-linux. > I don't think we build a dynamic version of the library, at least not by default. The libXxf86vm.so.1 you have most likely was on your system from before and not installed by from 4.2.1. XF86VidModeSetGammaRamp was added in 4.1.0. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Nvidia libGL + Xfree libGL
On Sun, 13 Oct 2002, Jens Owen wrote: > Mark Vojkovich wrote: > > On Mon, 14 Oct 2002, Michael wrote: > > > > > >>I'm running my system with two graphic cards, and i'm not using xinerama. > >>one of the cards is a nvidia and the other is a banshee, individually i can get >them both glx accelerated, i just need to copy the appropriate libGL depending on if >i want my banshee or nvidia accelerated, i'm wondering if it's possible in any way to >use multiple libGL's so i can get both my cards accelerated at the same time. > >> > > > > > >I don't think so. The separate libGL's could probably be solved with > > some LD_PRELOAD magic, but there can only be one GLX extension in > > the server (extensions aren't per-screen, they're server-wide). > > Mark, > > Have you, or anyone else at NVidia, considered reworking the device > independent pieces of the GLX extension and OpenGL client side library > from XFree86 to work with NVidia's binary drivers? > Not in a long time and never very seriously. I think our implementations have diverged quite a bit (we didn't start with the public SI and we hacked it quite a bit). Seems like alot of work, especially since the XFree86 GLX doesn't support dlopening libGLcore (correct?). Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Nvidia libGL + Xfree libGL
On Mon, 14 Oct 2002, Michael wrote: > I'm running my system with two graphic cards, and i'm not using xinerama. > one of the cards is a nvidia and the other is a banshee, individually i can get them >both glx accelerated, i just need to copy the appropriate libGL depending on if i >want my banshee or nvidia accelerated, i'm wondering if it's possible in any way to >use multiple libGL's so i can get both my cards accelerated at the same time. > I don't think so. The separate libGL's could probably be solved with some LD_PRELOAD magic, but there can only be one GLX extension in the server (extensions aren't per-screen, they're server-wide). Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]fastest refresh rate card
On Fri, 11 Oct 2002, Daniel Sheltraw wrote: > Hello Xpert list > > I realize this is not really a XF86 question but do you > of any graphics cards with vertical refresh frequencies > greater than 100/sec. Also what monitors or projectors > exist to support high refresh rates. > Vertical refresh is not going to be your limiting factor. The video cards don't care, and most CRTs will have no problem with 100 Hz. It's the other parameters that go along with high vertical referesh rates that will be the limitations, that is the pixel clock for the video card and the horizontal sync rate for the monitor. So like Andrew mentioned, it's the resolution in addition to the refresh rate. I don't think you'll find flat panels or projectors that can go that high though. Those have inherently slow refresh rates. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Re: expert's opinion wished.
On Fri, 11 Oct 2002, Franco Bevilacqua wrote: > Yes, I have already thought to this solution > But I need to display at least 30 frame per seconds in a image format of 640x512 > without overloading the CPU. > My idea consists to write the data in "the video driver" like V4L and then > display the data in the overlay of the graphic card. Overlays aren't grayscale, they're YCbCr. V4L devices like TV cards dump 4:2:2 YCbCr to the framebuffer. You'd need to be able to get your hardware to dump only to the luma portions (every other byte) and leave the chroma alone (which you've set up to black beforehand) or somehow have your hardware only dump black to the luma portions. I'm guessing that your capture hardware probably can't do that. > I suppose that I can directly show the overlay's image in a XWindow. > I need apply some transformations like magnifying, inverting, gamma correction > and so on. I think do that by using the graphics card features, but I don't know The video overlay can scale, but can't do image processing. If you had the data in system memory you could use 4:2:0 (YV12) XvShmPutImage. 4:2:0 is a planar format so luma is all in one chunk and never have to touch the chroma planes, just initialize them to black. You'd have to do any image processing on the luma data with the CPU before the XvShmPutImage request, but XvShmPutImage would handle the scaling with high quality filtering. I can probably modify one of my test apps to show how to display grayscale data using 4:2:0 XvShmPutImage. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]how Can I get GC from GContext
On Fri, 11 Oct 2002, Martin Ding wrote:
> How Can I get GC information only in client side? I mean not searching
> in Server side
In theory the client GC is an opaque handle and you're not
supposed to be able to get any info from it on the client side.
However, Xlib does cache all of these values as an optimization.
It compresses state changes and sends them down in blocks only
when they are needed. So if you look at a function like
XSetLineAttributes it will show how the line values are stored:
#include "Xlibint.h"
int
XSetLineAttributes(dpy, gc, linewidth, linestyle, capstyle, joinstyle)
register Display *dpy;
GC gc;
unsigned int linewidth; /* CARD16 */
int linestyle;
int capstyle;
int joinstyle;
{
XGCValues *gv = &gc->values;
LockDisplay(dpy);
if (linewidth != gv->line_width) {
gv->line_width = linewidth;
gc->dirty |= GCLineWidth;
}
if (linestyle != gv->line_style) {
gv->line_style = linestyle;
gc->dirty |= GCLineStyle;
}
if (capstyle != gv->cap_style) {
gv->cap_style = capstyle;
gc->dirty |= GCCapStyle;
}
if (joinstyle != gv->join_style) {
gv->join_style = joinstyle;
gc->dirty |= GCJoinStyle;
}
UnlockDisplay(dpy);
SyncHandle();
return 1;
}
You can see where it stores it and how it marks what state it
has that it hasn't flushed to the server yet. When the GC is actually
used to draw something, it will flush the dirty state back to the
server.
Mark.
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]How to debug extension?
On Fri, 11 Oct 2002, Martin Ding wrote: > I met a problem about one extension, which contains server side and > client side. I just doubt the extension has some problem. But reading > source code make me sick and faid to find anything I need. > So, Is there any way to debug the extension? For example, XTEST > extension. What sort of problem are you having? If you have the XFree86 modified version of gdb you can debug XFree86 modules. But a description of the problems you are having may allow us to give you some hints. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Need clues
On Thu, 10 Oct 2002, Franco Bevilacqua wrote:
>
> The format acquired from the DSP board are gray levels (0..255).
> The data come from the DSP board's driver continuously.
>
I'm assuming you mostly just want to take a raw greyscale
image and display it on the screen with scaling.
I'd recommend using OpenGL. Draw a textured quad. That would
be a LUMINANCE texture that you could update with glTexSubImage2D.
Mark.
>
> Mark Vojkovich wrote:
>
> > On Wed, 9 Oct 2002, Franco Bevilacqua wrote:
> >
> > > Hi all,
> > > I'm a newbie in developpment of video under linux.
> > > Where can I find docs and tutorials to program video under linux ??
> > > I need to display video on a XWindow starting from raw data (not
> > > compressed ) acquired on DSP card.
> > > I need also to apply some filters on the raw data, like interpolation.
> > > Does anybody has can give me some inputs ?
> > > In advance thanks.
> > > -Franco
> >
> >What format is your data in? YCbCr, I assume. Planar? Packed?
> >
> > Mark.
> > ___
> > Xpert mailing list
> > [EMAIL PROTECTED]
> > http://XFree86.Org/mailman/listinfo/xpert
>
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Monitor refresh rates
On Thu, 10 Oct 2002, Miroslav Silovic wrote: > While RandR probably has nothing to do with this (correct me if I'm > wrong here :) ), is it possible to dynamically change the monitor > refresh rates? Some people I'm working with consider this absolutely > essential (although the applications are exotic. Think PAL playback). Is > there already any tool that does this? Is there anything that needs to > be added to the X server to support this? You can have multiple modelines in the XF86Config and a client can switch through them with the XF86VidMode extension. > > I know that just dropping the proper set of modelines into XF86Config > would probably solve this, but this solution is far from user-friendly. > And it also leaves the problem of autogenerating these modelines from > some more general monitor profile. > All modelines need validation before the root window comes up so they need to be in the XF86Config file from the start. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
RE: [Xpert]resolution 1280x1024 ---> cursor disappears
On Thu, 10 Oct 2002, Berge, Harry ten wrote: > Hmmm... > > Yesterday I installed SuSE 8.1 Pro with a geforce2 MX400. Guess what, I had > the same problem! > The problem is fixed if you move the screen to the left somewhat (with > xvidtune). Don't know what the problem is, but it works now :-) > > regards Harry Can you send me the problematic modeline? Maybe there's some horizontal timing restriction that I wasn't aware of. Moving to the left would be shortening HSyncStart. Mark. > > > -Original Message- > > From: Mark Vojkovich [SMTP:[EMAIL PROTECTED]] > > Sent: Wednesday, October 09, 2002 8:08 PM > > To: [EMAIL PROTECTED] > > Subject:Re: [Xpert]resolution 1280x1024 ---> cursor disappears > > > > On Wed, 9 Oct 2002, Sven.Sanwald wrote: > > > > > Hello, > > > > > > the problem appears, when I change > > > the resolution from 1024x768 to > > > 1280x1024 (Strg +). > > > The screen looks ok, but the cursor > > > disappears or better it is only a one > > > pixel x 1cm long line and hard to find. > > > I'm using the XF 4.2.0 and the gcard > > > is a geforce2 MX400 with the "nv" > > > driver. > > > If the log-file ist of interrest I'll send it > > > :-) > > > > > > sven > > > >I've never heard of this problem. Cursor operation doesn't > > change when switching modes so I can't think why it would do > > something like that. > > > > > > Mark. > > ___ > > Xpert mailing list > > [EMAIL PROTECTED] > > http://XFree86.Org/mailman/listinfo/xpert > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]what will client report to server about window's framechanging?
On Thu, 10 Oct 2002, Martin Ding wrote: > If just switching 2 windows by clicking, we only see these 2 windows > changing their frame(such as title bar become bright or dim), while > their content changing nothing. > So, How does client make server do this? I mean how do make server only > change window's frame? and which request sent from > client(X_ConfigureWindow does not did this), or Event server will return > to client? > Or it is just controled by window manager in local side? not through > XServer? The client window is sitting on top of (is the child of) a window owned by the window manager. The window manager's window is usually larger than the client's window and the extra space around the client window holds the window decorations (title bar, etc...). When you click on the window frame you are clicking on the window manager's window and the window manager gets the events and takes the appropriate action. The client doesn't know anything about it and was not involved in the process at all. The client doesn't own the decorations and has nothing to do with them. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]keyboard and mouse lockup
Can the system live through "XFree86 -probeonly" and produce a /var/log/XFree86.0.log file? Mark. On Wed, 9 Oct 2002, Kre?imir Peharda wrote: > Hello all, > > I am experiencing a problem from which others have suffered as well, at least > according to the mailing lists. However, so far I did not see an answer on how > to solve it. > > What happens is that after starting X, both keyboard and mouse completely stop > responding. Mouse pointer is not moving, keyboard is completely dead, including > keys like num-lock, caps-lock that should switch on/off LED's. It is obvious > that the machine is still running. If I start X only, it will lock after > starting X, if I run startx script it will execute window manager as well. > Other people have reported that one can telnet to the machine and kill X, but I > cannot test that. > > This appears to be a frequent problem. Can anyone remember what has he/she > done to fix/circumvent it. > > If I don't fix this one quick, my wife will kill me. Although some might think > that the world would benefit from it, I would, on the other hand, be very > disappointed. > > Thanks for help, > Kresimir > > __ > Do you Yahoo!? > Faith Hill - Exclusive Performances, Videos & More > http://faith.yahoo.com > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]memory leak
On Wed, 9 Oct 2002, Frederic Heem wrote: > Hi experts, > Do you mind telling me your methods to detect memory leaks in X ? > I wonder why X is eating more and more RAM, every week I need to restart > X, otherwise, it's taking more than 64Mb. > Regards, > Frederic Heem > More often, these are client resource leaks rather than X server memory leaks. If you are running XFree86 built from CVS you can use the X-Resource extension to discover client resource usage. This is only in CVS and not in public release versions. There's a sample app on my web page at: http://www.xfree86.org/~mvojkovi/ bottom of the page. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Xinerama operation
On Wed, 9 Oct 2002, Verduzco Jesus Alberto wrote: > Mark > > Thanks for your answer, neverthless I have other > doubts: > > Let us retake the previous scene, xinerama with > two screens D1 and D2. > > "All rendering requests are duplicated and sent to > both screens > all the time. If the window is not visible, rendering > simply > clipped away as it normally would when rendering to a > window that > is not visible." > > * What is exactly the clipped operation? I'm not sure what you're asking. If the window is not visible, it has an empty cliplist. There is simply nothing to draw. > * On the Screen D2 (where the xterm window is not > visible), how > the xterm window is stored (a set of rendering > request or > a set of pixels)? Where these data are stored? > > I was reading the Xinerama documentation : " A region Documentation? Where? > ( XineramaScreenRegiosn[i] is initialized for each > physical screen, and single region ( > PanoramiXScreenRegion) is initialized > to be the union of the screen regions." > > What the regions are used for ? > There is a "PanoramiXScreenRegion" in the server. It doesn't look like it's used for anything and is just leftover from DEC's original implementation (I guess I should remove it). XineramaScreenRegions are the parts of the desktop that each individual screen takes up. This is really only used in GetImage calculations. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]resolution 1280x1024 ---> cursor disappears
On Wed, 9 Oct 2002, Sven.Sanwald wrote: > Hello, > > the problem appears, when I change > the resolution from 1024x768 to > 1280x1024 (Strg +). > The screen looks ok, but the cursor > disappears or better it is only a one > pixel x 1cm long line and hard to find. > I'm using the XF 4.2.0 and the gcard > is a geforce2 MX400 with the "nv" > driver. > If the log-file ist of interrest I'll send it > :-) > > sven I've never heard of this problem. Cursor operation doesn't change when switching modes so I can't think why it would do something like that. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Need clues
On Wed, 9 Oct 2002, Franco Bevilacqua wrote: > Hi all, > I'm a newbie in developpment of video under linux. > Where can I find docs and tutorials to program video under linux ?? > I need to display video on a XWindow starting from raw data (not > compressed ) acquired on DSP card. > I need also to apply some filters on the raw data, like interpolation. > Does anybody has can give me some inputs ? > In advance thanks. > -Franco What format is your data in? YCbCr, I assume. Planar? Packed? Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]command line options
On Mon, 19 Aug 2002, Kyle Balling wrote: > Is there a way to define a screen resolution for x in the command > line via startx or xinit? The purpose for this is a script that lets the > user choose a resolution for X to start with before starting X. I would > like to avoid appending the XF86Config file. > You can't specify the resolution on the command line but you can have different layout sections in the XF86Config and specify the layout you want to use from the commandline. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Setting the DPI so that images display at the correctsize on screen
On Tue, 13 Aug 2002, J. Grant wrote: > Hi, > > Hopefully someone can point me in the right direction on this. > > I have a 45cm [17.7 inch] (diaganal), 36cm x 27.2cm CRT that I would > like to configure with a screen mode that has the correct dpi. Currently > displaying a 10cm x 10cm square at 72dpi ends up 7cm x 7.5cm. I could > use the on-screen-display options to stretch the screen bigger, but then > the whole screen is not visible. > > Currently i have a 1400x1050 modeline in use. > I added: > > DisplaySize 360 272 > > but that only changed it from (96, 98) to (98,98) Which is exactly correct. 360x272 mm @ 1400x1050 is 98x98 dpi. > > In the XFree86.0.log file > it says: > > (**) NV(0): DPI set to (98, 98) > > Is this generated by the driver? is it possible for me to set it to > 72x72 as this is the conventional value... > I think "startx -- -dpi 72" probably works. You'd proably not use the DisplaySize though, because there's an obvious conflict with that display size and the dpi you want. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Xinerama operation
On Tue, 8 Oct 2002, Verduzco Jesus Alberto wrote: > > Hello > >I am interested in the operation of xinerama >Let us suppose the following, xinerama with two > monitor D1 and D2. > >That it happens? > >* When a xterm window is created and it only is > showed on D1? When a window is created, it is created on both screens but with different locations. It is at some location outside of the visible screen on D2, and subsequently clipped away. > >* the xterm window will be created only on D1 ? > > Later time the xterm window is modified ... > >* Moved to D2? It is moved on both. It moves offscreen on D1 and onscreen on D2. > >* Resized, and now the xterm window is showed on D1 > and D2? > > >* all rendering requests are sent to D1 and D2 ? All rendering requests are duplicated and sent to both screens all the time. If the window is not visible, rendering simply clipped away as it normally would when rendering to a window that is not visible. There were optimizations made to XAA to try to determine the fully-clipped case early on, since this is expected to be a common case with Xinerama. > >* How XGetImage() subdivides the image transfers > along screen boundaries ? Starting with screen 0. It gets as much as it can from screen 0. If it hasn't gotten it all, it moves along to the next screen to see if there is more there and so on. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Invisible mouse cursor on second head
On Tue, 8 Oct 2002, Russell wrote: > Mark Vojkovich wrote: > > > > On Mon, 7 Oct 2002, Russell wrote: > > > > > Hi all, > > > > > > I'm trying to get an identical display on two monitors > > > using separate video cards. If i set up XF86Config-4 to > > > have the RightOf or LeftOf screen option, then i get the > > > usual xinerama effect and can move the mouse between the > > > two monitors. > > > > > > If i specify absolute 0 0 position for both screens, > > > i get an identical fvwm desktop, but the mouse cursor > > > is invisible on the second monitor. > > > >Due to the way Xinerama is implemented (on top of a normal > > multihead server), there will only ever be one cursor. This > > is the expected behavior and not likely to change. > > You'd think that with the "clonedisplay" option that xfree86 > would be smart enough to know everything should just be replicated. > Without the mouse cursor, clonedisplay shouldn't even be implemented. It's not "clonedisplay" - it's Xinerama. Replicating the cursor is an architectural issue not a configuration issue. X11 doesn't support more than one cursor, hence there is no infrastructure in the server to support that. This is a deep-seated assumption that makes cloning the cursor non-trivial. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Invisible mouse cursor on second head
On Mon, 7 Oct 2002, Russell wrote: > Hi all, > > I'm trying to get an identical display on two monitors > using separate video cards. If i set up XF86Config-4 to > have the RightOf or LeftOf screen option, then i get the > usual xinerama effect and can move the mouse between the > two monitors. > > If i specify absolute 0 0 position for both screens, > i get an identical fvwm desktop, but the mouse cursor > is invisible on the second monitor. Due to the way Xinerama is implemented (on top of a normal multihead server), there will only ever be one cursor. This is the expected behavior and not likely to change. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]nv driver crash on nForce
On 6 Oct 2002, Mikkel Lauritsen wrote: > On Thu, 2002-10-03 at 22:35, Mark Vojkovich wrote: > > > I don't recall having any problems with 4.2.0 on nForce. I'm > > not sure that RedHat's "4.2.0" is really the same as a stock > > XFree86 4.2.0. There have be cases in the past where they have > > modified theirs. Try an official XFree86 4.2.0. > > I just did, and unfortunately the behavior is the exact same. > > Just to display my potential ignorance, what I did was to replace > nv_drv.o from the RedHat package with that from the binary distribution > at ftp.xfree86.org. I still get a hard lockup, and it seems as if the > log files that I am able to obtain reveal nothing at all. I guess they > are truncated at a more or less random point in time when the machine > hangs. Copying the module should probably be fine. > > So... is there any way that I might try to look deeper into this, > preferably without having to rebuild X from scratch? > Do you get a full /var/log/XFree86.0.log if you do "XFree86 -probeonly" I put an nv_drv.o compiled from CVS on my web page at: http://www.xfree86.org/~mvojkovi/nv_drv.o Since it's newer than your server, you may need to add -ignoreABI to your server command line. (ie. "startx -- -ignoreABI"). Are you running some kernel framebuffer device like rivafb? It's possible that rivafb is not compatible with the "nv" driver on nForce. I never use something like rivafb. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Updating of windows during window move/resize is slow
On Sat, 3 Aug 2002, Jouni Tulkki wrote: > > I have noticed that moving and resizing windows is very slow when using > window managers that update the window position/size immediately. > The reason for this slowness (I think) is that a huge number of > exposure events are sent to the windows that need to be updated. > On my machine (PentiumII 233 with 128MB memory and Matrox G400 with 16MB > memory) this causes updating "chains" that can take many seconds to finish > even when using relatively fast and well optimized programs like Netscape > 4.77. > > The problem is, I think, that exposure events are sent to windows > even when the previous exposure events have not been processed. This > causes exposure events to stack up. A smart client could ofcourse > check if there are other exposure events in the event queue and remove > them. Still I don't think this is the client's job and I haven't seen > any program that would do so. It is the client's job and any client that doesn't compress events is poortly written. There's nothing the server can do about it, it has to present the motion and exposure events in the proper order. In the case of window moves the event is already sent to the client and the X-server goes back to waiting on its file descriptors before it gets the next move that would cause a new exposure. There aren't multiple exposes sitting around in the server. There are rendering schemes where clients don't even need exposures. Consider rendering everything into a pixmap and setting that as a window background. You wouldn't even get exposures in that case. Some current apps are so dumb that they redraw the whole scene for each rectangle of exposed area - they're not even compressing rects generated from the same move. I don't think the server can be faulted here. It's just that there are too many people writing applications when they don't fully know what they're doing. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Can't turn dpms off with nv driver
On 4 Oct 2002, James D Strandboge wrote: > I have been having screen locks since switching to ACPI. In trying to > locate the problem, I am trying to disable DPMS when using the nv > driver. However, no matter what I try, it always is enabled: > > ... > (II) Loading extension XFree86-Misc > (II) Loading extension XFree86-DGA > (II) Loading extension DPMS > (II) Loading extension FontCache > (II) Loading extension TOG-CUP > ... > (==) NV(0): Backing store disabled > (==) NV(0): Silken mouse enabled > (**) Option "dpms" "no" > (**) NV(0): DPMS enabled > (II) Initializing built-in extension MIT-SHM > (II) Initializing built-in extension XInputExtension > ... > > I have set this option in both the Moniter section as well as the Device > section for the driver, but neither work. Is this a bug or am I missing > something? Poke around in common/xf86DPMS.c:xf86DPMSInit(). It looks like it always prints that out if you have any sort of "dpms" option. At a minimum, the code printing out the message is wrong. You might check that the other parameters in that code are getting set to expected values. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]nv driver crash on nForce
I don't recall having any problems with 4.2.0 on nForce. I'm not sure that RedHat's "4.2.0" is really the same as a stock XFree86 4.2.0. There have be cases in the past where they have modified theirs. Try an official XFree86 4.2.0. Mark. On 3 Oct 2002, Mikkel Lauritsen wrote: > Hi all, > > this bug report has been posted by me before, but I unfortunately > got sidetracked and never got to follow up on it. My apologies. > > Whenever I try to run the nv driver in 4.2.0 on my nForce-based > PC the X server crashes and locks the PC, requiring a hard reset > to continue, while the closed-source nvidia driver from NVidia > works perfectly. Because of the hard crash I'm not sure whether > the log file is complete, but it does not seem to contain any > irregularities. > > I'm running RedHat Linux 7.3 on an Asus A7N266-VM with an Athlon > XP1800+ and 1 GB ram. The behavior is exactly the same under RedHat > 8.0, where the X package according to the release notes basically > is 4.2.1, even if the version number is still 4.2.0. > > This problem might of course be a RedHat-ism, so I have reported > the problem there as well. Other people seem to have experienced > the same behavior on different nForce-based boards. > > Do you have any suggestions as to how I might try to track this > one down? In case there is a need for it, I'd be more than happy > to provide log files and any other relevant information. > > Thanks, > Mikkel Lauritsen > > > ___ > Xpert mailing list > [EMAIL PROTECTED] > http://XFree86.Org/mailman/listinfo/xpert > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Where can I get source
On Wed, 2 Oct 2002, Larry wrote: > > Where can I get the source for xset? I've looked all over the web site and > can't find anything but the sources for X, which I really don't need. It's in the server tree at xc/programs/xset. > What I would really like is the code to blank and unblank the screen much like > the xset program sets up. I want to add the blanking and unblanking to a > screensaver program I am writing so that the gui won't take up resources when > the screensaver is running. But I can't find much on the subject. > See the man page on XSetScreenSaver. MArk. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]dvi output
On Tue, 1 Oct 2002, SweeTLeaF wrote: > Hello gentlemen / ladies > > I am running freebsd 4.7 with a "Pine" GEforce 4 card that has > standard vga output, s-video output and DVI output. I am currently > using the vga output but i would like to know how to use the DVI > output and connect to my digital lcd monitor. > There is DVI support for GeForce4 in the "nv" driver in XFree86 CVS. You'll need to add Option "FlatPanel" to the Section "Device". I don't know what patchlevel of XFree86 comes with FreeBSD, but I'd guess that the version that it ships with doesn't support it so you'd need to build something newer. There's a little bit about "FlatPanel" and another option you might need "CrtcNumber" in the "nv" man page. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]colour mask values question
On 1 Oct 2002, Morten Sylvest Olsen wrote:
> Hi..
>
> Could anybody enlighten me on the proper use and understanding of the
> values {red,green,blue}_mask in the Visual and XImage structures. I
> assumed that X would convert between fx. RGB and BGR color ordering from
> image to server by setting these masks according but they seem to make
> absolutely no difference (also I cannot find where in the source this
> conversion would take place either).
>
I never understood why these were in the XImage structures.
The Visual masks indicate how a pixel breaks up into RGB components.
This is necessary for you to know when filling in XImage data.
But the point of the masks in the XImage structure escape me
since there is no conversion going on during PutImage.
mark.
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Lockups!
On Mon, 30 Sep 2002, Staci wrote: > > Things to try: > > 1) Don't use a font server. If you have, in your XF86Config, in > >the Section "Files", a FontPath entry that looks like: > > FontPath "unix/:some number" > > Nope, nothing like that. > > >remove it. And instead take the paths from your font server config > > (often in /etc/X11/fs/config) and put those in the XF86Config instead. > > Eg. > >FontPath "/usr/X11R6/lib/X11/fonts/misc:unscaled" > >FontPath "/usr/X11R6/lib/X11/fonts/75dpi:unscaled" > >FontPath "/usr/X11R6/lib/X11/fonts/100dpi:unscaled" > >FontPath "/usr/X11R6/lib/X11/fonts/Type1" > >FontPath "/usr/X11R6/lib/X11/fonts/Speedo" > >FontPath "/usr/X11R6/lib/X11/fonts/TrueType" > > I did so, and commented out the only one on the list (that was already in > XF86Config) that wasn't in /etc/X11/fs/config, which was > /usr/X11R6/lib/X11/fonts/local/. > > > 2) If that doesn't help, comment out the TrueType line. That may change > > the appearance of the fonts severely, but it's a good test to try > > to narrow down the problem. > > Hehe, I don't even HAVE a TrueType line. My fonts don't look funny to ME! ;) > > sl > > PS Opera hasn't crashed yet, knock on wood. I'm just covering known X-server problems. There are some problems with the font server (which you don't appear to be running), and historically there have been some problems with TrueType renderers choking on some broken fonts. I only guessed that your TrueType fonts were in a TrueType directory. Another way to ensure that true type fonts aren't being used is to not load the modules for them. ie. remove the "freetype" and "xtt" modules from the Section "Module" in the XF86Config file. Font related problems have traditionally show up when running browers, which is why your crash is a familiar scenario. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
