Re: [Xpert]3D on ATI Mobility Radeon-M
On Fri, Sep 21, 2001 at 05:08:29PM +0200, Michel D?nzer wrote: > Brian Lavender wrote: > > > > On Fri, Sep 21, 2001 at 02:29:16PM +0200, Michel D?nzer wrote: > > > You probably need to build the radeon.o kernel module from XFree86 source > > > and use that. > > > > Do I need the agpgart module as well? > > Yes, PCI GART is still disabled in source for the Radeon AFAIK. I installed the agpgart module, and the following radeon modules, but when I start X, I don't seem to be getting the 3D support. I try running tuxracer, and I pretty much can tell 3D not working. When I quit X, I see a message (below) from the startup of X, that the dri was unloaded. Is DRI just not supported with this video card, or am I missing something. I have X 4.1.99 from cvs. I got the dri code from http://dri.sourceforge.net. I can't think what else there is. My video card is the ATI RAGE Mobility Radeon-M on a Sony VAIO PCG-GR170K [brian@localhost brian]$ /sbin/lsmod Module Size Used by radeon_state 21584 0 (unused) radeon_cp 12896 0 [radeon_state] radeon_drv 52336 0 [radeon_state radeon_cp] radeon 86064 0 [radeon_drv] agpgart26560 2 e100 62864 1 (autoclean) ide-scsi8032 0 scsi_mod 85040 1 [ide-scsi] Message I see after quiting X. Page up doesn't work on my laptop, so I can't see the rest either. 5 512x512 slots (II) RADEON(0): Acceleration enabled (II) RADEON(0): Using hardware cursor (scanline 2104) (II) RADEON(0): Largest offscreen area available: 1408 x 1924 (II) RADEON(0): Direct rendering disabled -- Brian Lavender http://www.brie.com/brian/ ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]XFree86 Version 4.1.0 on OSX/Darwin - Quartz Problem
I downloaded and installed the binaries for Xfree86 fro Mac OSX but when I run it it gives me an error: > Fatal server error: > Failed to connect as window server! > Quit the Mac OS X window server or use the -quartz option. > > OsVendorFatalError > AbortDDX > Quitting XDarwin... > giving up. > xinit: No such file or directory (errno 2): unable to connect to X server > xinit: No such process (errno 3): Server error. That's all fine and good but when I do that (-quartz) I get another error: > Fatal server error: > QuartzOsVendorInit called without Quartz support. > > OsVendorFatalError > AbortDDX > Quitting XDarwin... > > FatalError re-entered, aborting > QuartzGiveUp called without Quartz support. > > giving up. > xinit: No such file or directory (errno 2): unable to connect to X server > xinit: No such process (errno 3): Server error. Aaagghhh! When I run it in plain console mode (ie. login with >console and run X) it starts up fine. What am I doing wrong? ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]S3 ViRGE problems with Xfree86 v4.02
Mohan, I have an S3 Virge GX2 here. Does it lock up with certain colour-depths? Try running X at 8-bit and see if it locks up. In my case, if I try running it at 32-bit, all sorts of strange effects happen (like invisible windows and floating widgets). AFAIK, 16-bit is the highest your card can support. I would try running 4.0.3 at least. The fixes related to the gx2 partially fix the problems with the MX. The fixes were incorporated in 4.0.3. I'm now running 4.1.0 and have not had problems since 4.0.3. Before, the forsaken SVGA server would cause me to lock up within seconds of any line-stipling or fast-scrolling text. On Fri, 21 Sep 2001, Kevin Brosius wrote: |Mohan Rajagopalan wrote: |> |> |> |> I tried compiling and installing X on my Toshiba Satellite 490CDT laptop |> .. it compiles and installs. I have a S3 ViRGE /MX card with 2MB on the |> laptop which is identified by xf86config. | |4.0.2 is rather old, would you be able to install 4.1.0? | |> |> When I to start with any screen resolution > 320 X 200 the machine |> hangs. After spending hours trying to trace the cause, I discovered the S3 |> device driver would hang trying to update the MTRRs for Write Combining. |> Tha machine hangs even with all forms of acceleration disabled. |> The only fix I could use is to change the driver to VESA. |> |> Has anyone else had similar problems ? Any fixes ?? |> |> -mohan |> | |We've had some trouble with MX. What makes you think MTRRs are hanging |the machine? There are a couple updated drivers on my page at |http://www.user1.netcarrier.com/~kbrosius/pages/virge.html, although I |suspect even the 4.0.3 version will not run with 4.0.2. If you give it |a try let me know if it loads. | | -- ** Derek J Witt** * Email: mailto:[EMAIL PROTECTED] * * Home Page: http://www.flinthills.com/~djw/ * *** "Houston, the Eagle has landed and laid an egg!" -- Unknown ** ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]S3 ViRGE problems with Xfree86 v4.02
Mohan, I have an S3 Virge GX2 here. Does it lock up with certain colour-depths? Try running X at 8-bit and see if it locks up. In my case, if I try running it at 32-bit, all sorts of strange effects happen (like invisible windows and floating widgets). AFAIK, 16-bit is the highest your card can support. I would try running 4.0.3 at least. The fixes related to the gx2 partially fix the problems with the MX. The fixes were incorporated in 4.0.3. I'm now running 4.1.0 and have not had problems since 4.0.3. Before, the forsaken SVGA server would cause me to lock up within seconds of any line-stipling or fast-scrolling text. On Fri, 21 Sep 2001, Kevin Brosius wrote: |Mohan Rajagopalan wrote: |> |> |> |> I tried compiling and installing X on my Toshiba Satellite 490CDT laptop |> .. it compiles and installs. I have a S3 ViRGE /MX card with 2MB on the |> laptop which is identified by xf86config. | |4.0.2 is rather old, would you be able to install 4.1.0? | |> |> When I to start with any screen resolution > 320 X 200 the machine |> hangs. After spending hours trying to trace the cause, I discovered the S3 |> device driver would hang trying to update the MTRRs for Write Combining. |> Tha machine hangs even with all forms of acceleration disabled. |> The only fix I could use is to change the driver to VESA. |> |> Has anyone else had similar problems ? Any fixes ?? |> |> -mohan |> | |We've had some trouble with MX. What makes you think MTRRs are hanging |the machine? There are a couple updated drivers on my page at |http://www.user1.netcarrier.com/~kbrosius/pages/virge.html, although I |suspect even the 4.0.3 version will not run with 4.0.2. If you give it |a try let me know if it loads. | | -- ** Derek J Witt** * Email: mailto:[EMAIL PROTECTED] * * Home Page: http://www.flinthills.com/~djw/ * *** "Houston, the Eagle has landed and laid an egg!" -- Unknown ** ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
Gustavo Alberto Homem wrote: > So from what I read from you I would say that for an installer it would be > better to use the old XF86_FBDev since one does not need in this case to use > the new fancy features and from your words we get the impressions that the > old one could even run on some exotic video cards, right ? Probably, yes. I might be wrong though so I suggest trying both and maybe also Xfbdev, the 4.x lightweight fbdev server. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
Hello Michel: Thank you very much for your answeer. So from what I read from you I would say that for an installer it would be better to use the old XF86_FBDev since one does not need in this case to use the new fancy features and from your words we get the impressions that the old one could even run on some exotic video cards, right ? Thanks again Gustavo 2+2=5 for extremely large values of 2 .-``'. I.S.T - Instituto Superior Tecnico .` .`~ _.-' '._ Licenciatura em Engenharia Fisica e Tecnologica [EMAIL PROTECTED] http://mega.ist.utl.pt/~gach On Fri, 21 Sep 2001, Michel [iso-8859-1] Dänzer wrote: > Gustavo Carvalho Homem wrote: > > > One thing I can't find in the documentation. Are there advantages (or > > disadvantages) of using XF86_FBDev over XFree86 4.1 with fbdev driver ? Any > > other relevant comments ? > > The big disadvantage is that XF86_FBDev is ooold and lacks a lot of features > like shadow framebuffer, RENDER extension, ... > > The only possible advantages I see are for exotic cases like 1 or 4 bit depth > which might be broken in the 4.x fbdev driver (but should be easy to fix) and > that maybe XF86_FBDev offers some acceleration for some chipsets, but I'm not > even sure there. > > > -- > Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer > XFree86 and DRI project member / CS student, Free Software enthusiast > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]X11R6.3/6.4 'Xweb' and XFree86
Hello, Is the Internet solution 'Xweb' browser plugin implemented in XFree86 - if not, why not? http://www.opengroup.org/press/31mar98.htm http://www.google.com/search?q=cache:QMmnZSzLY6Y:www.broadwayinfo.com/+%22broadwayinfo.%2Bcom%22&hl=en http://www.google.com/search?q=cache:llSTaoAfKKU:www.broadwayinfo.com/bwfaq.htm+%22broadwayinfo.%2Bcom%22&hl=en TIA, Terje J. Hanssen ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
RE: [Xpert]Having trouble getting Xfree86 working on my laptop.
Title: RE: [Xpert]Having trouble getting Xfree86 working on my laptop. It should be the same as my current laptop (IBM 760XD), if so, then you need to use the SVGA server @ 1024x768x16bpp with the monitor as Generic LCD panel 1024x768 (personally, I specify one of the IBM TFT Panels (something revision b)). George E. Jones IV Store Systems Support Borders Group Inc. -Original Message- From: David Pendell [mailto:[EMAIL PROTECTED]] Sent: Friday, September 21, 2001 6:43 PM To: [EMAIL PROTECTED] Subject: [Xpert]Having trouble getting Xfree86 working on my laptop. I have an IBM Thinkpad 760ED with a trident TGUI chipset (not sure which one I can't get a definative answer.) Under 3.3.6 Everything worked great. Under 4.1.0 I only get to results under the vga driver I get a 320x200 screen that has been magnified by about 2x, or under the trident driver all I get is a garbled screen. I have already tried every Option that is listed in the trident man page. I have even tried compiling framebuffer support into the kernel and using the framebuffer driver. Nothing has worked. I really don't want to move back back to 3.3.6. I have to believe that SOMEONE out there has seen this problem and knows the fix. Anybody? d.p. = David Pendell Information Technology Consultant [EMAIL PROTECTED] __ Do You Yahoo!? Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger. http://im.yahoo.com ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XVideo and vsync
On Fri, Sep 21, 2001 at 02:41:29PM -0700, Mark Vojkovich wrote: >Hardware should automatically synchronize stuff like Xv(Shm)PutImage > with the retrace. If you want a VSync pulse for some custom application > you're out of luck unless you write a kernel module to do it. It's > a hardware specific thing done with interrupts. Hmm then my tvout patch fsckes it up, because it clearly isn't synced on my tv. Or could it be that aviplay or sdl don't call xvshmputimage with sync on? will check > Mark. Bye, Peter Surda (Shurdeek) <[EMAIL PROTECTED]>, ICQ 10236103, +436505122023 -- funny thats the third time since ive been on IRC today that i seen the word "slave" and "microsoft" in the same sentence PGP signature
Re: [Xpert]Trident Cyberblade XP XV support
Does the changes fix xv and or dga on the Cyberblade XP? On Fri, 2001-09-21 at 19:09, Anders Rune Jensen wrote: Hi Im running a Toshiba Tecra 8200 laptop with a trident cyberblade xp gfx card. The card works in Xfree but lacks xv and dga. I saw that there were some unofficial patches and maybe some of them fixed the problem. Could anyone put them in cvs or tell me how to install them. Im running debian sid. Med venlig Hilsen: - Anders Rune Jensen icq : 52926571 Markusgade 7 4. TH email : [EMAIL PROTECTED] 9000 Ålborg web : http://www.gnulinux.dk Med venlig Hilsen: - Anders Rune Jensen icq : 52926571 Markusgade 7 4. TH email : [EMAIL PROTECTED] 9000 Ålborg web : http://www.gnulinux.dk PGP signature
[Xpert]Patch for wheel emulation mode
Hello,
I've tried the current CVS version to test the wheel emulation mode that
Henry T. So implemented and I've found a problem.
I'm using the fourth button as wheel emulation button, but then, when the
it's going to generate a fourth button press/release events (wheel up),
it chooses not to do so because "the real" 4th button is already pressed.
So it doesn't generate wheel up events !
The attached patch makes it work and also lowers a lot the Inertia default
value, as I think 50 is too high . I was going to suggest to set it to "1"
as I've done in my configuration, but I suppose 1 is too sensible to be
the default. In any case, even if 50 is ok for normal mice, it's too high
for trackballs (and remember that this feature is mainly for trackball's
users), so I'd say to set the default to 10.
I would also like to ask for Button6 and Button7 (and related) to be
defined in X.h . Would this be possible ? If we want to support horizontal
wheels as good as we support vertical wheels (and I can't find a reason
not to do so), then these definitions should be done in order to allow
applications to use nice definitions instead of simple values (I mean,
instead of using plain 6 and 7 numbers when checking the button being
pressed). In other way, how should the apps check for horizontal wheel's
left and right events ? (the ones in XAxisMapping)
Greetings,
--
Antonio Larrosa Jimenez
KDE Core developer - [EMAIL PROTECTED]
SuSE Labs developer - [EMAIL PROTECTED]
http://perso.wanadoo.es/antlarr
KDE - The development framework of the future, today.
Index: mouse.c
===
RCS file: /cvs/xc/programs/Xserver/hw/xfree86/input/mouse/mouse.c,v
retrieving revision 1.47
diff -u -p -r1.47 mouse.c
--- mouse.c 2001/09/14 19:12:17 1.47
+++ mouse.c 2001/09/21 14:49:55
@@ -511,7 +511,7 @@ MouseCommonOptions(InputInfoPtr pInfo)
pMse->wheelButtonMask = 1 << (wheelButton - 1);
pMse->wheelInertia = xf86SetIntOption(pInfo->options,
- "EmulateWheelInertia", 50);
+ "EmulateWheelInertia", 10);
if (pMse->wheelInertia <= 0) {
xf86Msg(X_WARNING, "%s: Invalid EmulateWheelInertia value: %d\n",
pInfo->name, pMse->wheelInertia);
@@ -2004,7 +2004,7 @@ MouseDoPostEvent(InputInfoPtr pInfo, int
MouseDevPtr pMse;
int truebuttons, emulateButtons;
int id, change;
-int emuWheelDelta, emuWheelButton;
+int emuWheelDelta, emuWheelButton, emuWheelButtonMask;
pMse = pInfo->private;
@@ -2026,6 +2026,7 @@ MouseDoPostEvent(InputInfoPtr pInfo, int
emuWheelDelta = pMse->wheelInertia;
emuWheelButton = pMse->positiveY;
}
+ emuWheelButtonMask = 1 << (emuWheelButton - 1);
while (abs(pMse->wheelYDistance) > pMse->wheelInertia) {
pMse->wheelYDistance -= emuWheelDelta;
@@ -2033,7 +2034,8 @@ MouseDoPostEvent(InputInfoPtr pInfo, int
* Synthesize the press and release, but not when the button.
* to be synthesized is already pressed "for real".
*/
- if (!((1 << (emuWheelButton - 1)) & buttons)) {
+ if (!(emuWheelButtonMask & buttons)
+ || (emuWheelButtonMask & pMse->wheelButtonMask) ) {
xf86PostButtonEvent(pInfo->dev, 0, emuWheelButton, 1, 0, 0);
xf86PostButtonEvent(pInfo->dev, 0, emuWheelButton, 0, 0, 0);
}
@@ -2057,7 +2059,8 @@ MouseDoPostEvent(InputInfoPtr pInfo, int
* Synthesize the press and release, but not when the button.
* to be synthesized is already pressed "for real".
*/
- if (!((1 << (emuWheelButton - 1)) & buttons)) {
+ if (!(emuWheelButtonMask & buttons)
+ || (emuWheelButtonMask & pMse->wheelButtonMask) ) {
xf86PostButtonEvent(pInfo->dev, 0, emuWheelButton, 1, 0, 0);
xf86PostButtonEvent(pInfo->dev, 0, emuWheelButton, 0, 0, 0);
}
[Xpert]Having trouble getting Xfree86 working on my laptop.
I have an IBM Thinkpad 760ED with a trident TGUI chipset (not sure which one I can't get a definative answer.) Under 3.3.6 Everything worked great. Under 4.1.0 I only get to results under the vga driver I get a 320x200 screen that has been magnified by about 2x, or under the trident driver all I get is a garbled screen. I have already tried every Option that is listed in the trident man page. I have even tried compiling framebuffer support into the kernel and using the framebuffer driver. Nothing has worked. I really don't want to move back back to 3.3.6. I have to believe that SOMEONE out there has seen this problem and knows the fix. Anybody? d.p. = David Pendell Information Technology Consultant [EMAIL PROTECTED] __ Do You Yahoo!? Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger. http://im.yahoo.com ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]S3 ViRGE problems with Xfree86 v4.02
Mohan Rajagopalan wrote: > > > > I tried compiling and installing X on my Toshiba Satellite 490CDT laptop > .. it compiles and installs. I have a S3 ViRGE /MX card with 2MB on the > laptop which is identified by xf86config. 4.0.2 is rather old, would you be able to install 4.1.0? > > When I to start with any screen resolution > 320 X 200 the machine > hangs. After spending hours trying to trace the cause, I discovered the S3 > device driver would hang trying to update the MTRRs for Write Combining. > Tha machine hangs even with all forms of acceleration disabled. > The only fix I could use is to change the driver to VESA. > > Has anyone else had similar problems ? Any fixes ?? > > -mohan > We've had some trouble with MX. What makes you think MTRRs are hanging the machine? There are a couple updated drivers on my page at http://www.user1.netcarrier.com/~kbrosius/pages/virge.html, although I suspect even the 4.0.3 version will not run with 4.0.2. If you give it a try let me know if it loads. -- Kevin ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]XVideo and vsync
On Fri, Sep 21, 2001 at 10:12:54PM +0200, Matthias Dahl wrote: > Hey there... ]) powered by Linux 2.4.8 ([ hi > I have run into a problem and I hope some of you guys can help me out. :) > Thanks in advance for your help and time! np > Is it possible to do some vsync synchronization with the XVideo (or any other) > extension? I vaguely remember that for mga you can use /dev/mga_vid. I think there is a generic interface somewhere in X but don't know where. > I need this for proper video output. Yet I could not find any hint > about any function in the X documentation. If this isn't implemented yet, is > there any other way to synchronize the video output with the vsync? What card do you have? > So long, Matt ]) [EMAIL PROTECTED], GPG key available at public keyservers ([ gpg --send-key [EMAIL PROTECTED] --keyserver certserver.pgp.com or better yet, define a keyserver in ~/.gnupg/options, then gpg will download automagically any keys if necessary, works well under mutt too. Bye, Peter Surda (Shurdeek) <[EMAIL PROTECTED]>, ICQ 10236103, +436505122023 -- PGP signature
Re: [Xpert]XVideo and vsync
On Fri, 21 Sep 2001, Matthias Dahl wrote: > Hey there... ]) powered by Linux 2.4.8 ([ > > I have run into a problem and I hope some of you guys can help me out. :) > Thanks in advance for your help and time! > > Is it possible to do some vsync synchronization with the XVideo (or any other) > extension? I need this for proper video output. Yet I could not find any > hint about any function in the X documentation. If this isn't implemented yet, > is there any other way to synchronize the video output with the vsync? > > Thanks again. Hardware should automatically synchronize stuff like Xv(Shm)PutImage with the retrace. If you want a VSync pulse for some custom application you're out of luck unless you write a kernel module to do it. It's a hardware specific thing done with interrupts. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]XVideo and vsync
Hey there... ]) powered by Linux 2.4.8 ([ I have run into a problem and I hope some of you guys can help me out. :) Thanks in advance for your help and time! Is it possible to do some vsync synchronization with the XVideo (or any other) extension? I need this for proper video output. Yet I could not find any hint about any function in the X documentation. If this isn't implemented yet, is there any other way to synchronize the video output with the vsync? Thanks again. -- So long, Matt ]) [EMAIL PROTECTED], GPG key available at public keyservers ([ ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]RECORD extension stability question.
Hello, I have written an application which uses the RECORD extension in the Xtst library. I have noticed that the application while appearing to function correctly has a severe stability problem. When recording is enabled the X session is closed whenever a window is closed or maximized. I don't know if I am describing the problem correctly by saying the "X Session" is closing, but what happens is that the screen flickers for a few minutes than I am greeted with the normal gdm login screen. Has anyone out there successfully used the RECORD extension? Or does anyone know if it is stable? I read some posts from around 1996 which claimed is was unstable then. I'm hoping over the years it has matured, but I haven't found any info on it. I upgraded to the latest XFree86 4.1.0, but I am having some difficulty in getting the RECORD extension to operate. It appears that the extension isn't being loaded. I'd appreciate any help on this topic too. BACKGROUND INFO: Platform: Red Hat Linux 6.2 on Intel Pentium II X Windows System: XFree86 3.3.6 The application is requesting to be notified of "delivered events" ButtonPress and ButtonRelease. I am than looking at the root and event window ID's to determine which widget was clicked. So far it appears to be working correctly, except for the stability problem. I haven't found any documentation which says that the root and event fields will contain valid data, but so far they appear to be correct. The purpose is to integrate this event monitoring functionality into a training simulator. Thanks in advance, John [EMAIL PROTECTED] ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Xv card list and Vesa 2.0 compliance
On Fri, 21 Sep 2001, Alex Deucher wrote: > I actually just got a response from matrox, and they > said that all cards from Millennium I onward have > hardware colorspace conversion, but only the G200 and > up have hardware scaling. They didn't have overlay hardware before the G200. And it required that alot of data be passed to the hardware. My impression was that on a modern computer it would be a video decellerator. Maybe Matrox knows some tricks, but I gave up on it when I was looking into it a few years ago. Mark. > > Alex > > --- Mark Vojkovich <[EMAIL PROTECTED]> wrote: > > On Fri, 21 Sep 2001, Alex Deucher wrote: > > > > > Just out of curiosity, did the older millennium I > > > (2064w) and millennium II's (2164w) have Xv > > hardware > > > on them as well, or did that not get added till > > later? > > > > > > > Only G200 and newer. > > > > Mark. > > > > > __ > Terrorist Attacks on U.S. - How can you help? > Donate cash, emergency relief information > http://dailynews.yahoo.com/fc/US/Emergency_Information/ > ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Xv card list and Vesa 2.0 compliance
I actually just got a response from matrox, and they said that all cards from Millennium I onward have hardware colorspace conversion, but only the G200 and up have hardware scaling. Alex --- Mark Vojkovich <[EMAIL PROTECTED]> wrote: > On Fri, 21 Sep 2001, Alex Deucher wrote: > > > Just out of curiosity, did the older millennium I > > (2064w) and millennium II's (2164w) have Xv > hardware > > on them as well, or did that not get added till > later? > > > > Only G200 and newer. > > Mark. > __ Terrorist Attacks on U.S. - How can you help? Donate cash, emergency relief information http://dailynews.yahoo.com/fc/US/Emergency_Information/ ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Xv card list and Vesa 2.0 compliance
On Fri, 21 Sep 2001, Alex Deucher wrote: > Just out of curiosity, did the older millennium I > (2064w) and millennium II's (2164w) have Xv hardware > on them as well, or did that not get added till later? > Only G200 and newer. Mark. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]lockup in keyboard initialization when no keyboard is connected
> Hello,
>
> I decided to try XFree86 4.1.0, but it won't start; the X server hangs
> during keyboard initialization. The following backtrace is from a
> recent CVS version:
>
> (gdb) bt
> #0 0x80a8fc0 in inb (port=100)
> at ../../../../../../programs/Xserver/hw/xfree86/common/compiler.h:1135
> #1 0x80a8dff in xf86SetKbdRepeat (rad=32 ' ') at lnx_io.c:197
> #2 0x808fb91 in xf86InitKBD (init=1) at xf86Io.c:246
> #3 0x808fd8c in xf86KbdProc (pKeyboard=0x8372518, what=0) at xf86Io.c:338
> #4 0x80cc2eb in InitAndStartDevices () at devices.c:169
> #5 0x80e3728 in main (argc=1, argv=0xbccc, envp=0xbcd4) at main.c:403
>
> It hangs in this loop:
>
> while ((inb(0x64) & 2) == 2); /* wait */
>
> I guess that line is trying to read something from an AT keyboard, but
> I don't have one... 4.0.3 did not have this problem.
>
> Just putting a 'return' before the infinite loop fixes the problem,
> and I can also use my USB keyboard. But I don't believe this is a good
> fix... :)
>
> What's a good solution?
>
> Regards,
>
> Pontus
>
(This issue may have been discussed before, but I
wasn't able to find it in a quick scan of my archives.
My apologies if I duplicate any information previously
discussed.)
What type of hardware are you using?
This problem can arise if you are talking to a USB
keyboard on a system without a PS/2 controller.
The attached patch should address this. Unfortunately,
it's not based on the latest snapshot of that
file, so someone could have already applied a similar
fix, or the line numbers could be slightly off.
It essentially adds a countdown to the 'while' loops in
xf86SetKbdRepeat() so that it doesn't spin forever.
Let me know if this works.
-paul
---
--- xc-orig/programs/Xserver/hw/xfree86/os-support/linux/lnx_io.c Fri Aug 3
16:12:41 2001
+++ xc/programs/Xserver/hw/xfree86/os-support/linux/lnx_io.cMon Aug 6 12:08:54
+2001
@@ -34,6 +34,8 @@
#include "xf86Priv.h"
#include "xf86_OSlib.h"
+#define KBC_TIMEOUT 250/* Timeout in ms for sending to keyboard controller */
+
void
xf86SoundKbdBell(int loudness, int pitch, int duration)
{
@@ -140,6 +142,7 @@
#endif
{
int i;
+ int timeout;
int value = 0x7f;/* Maximum delay with slowest rate */
#ifdef __sparc__
@@ -194,9 +197,14 @@
break;
}
- while ((inb(0x64) & 2) == 2); /* wait */
+ timeout = KBC_TIMEOUT;
+ while (((inb(0x64) & 2) == 2) && --timeout)
+ usleep(1000); /* wait */
+
outb(0x60, 0xf3); /* set typematic rate */
- while ((inb(0x64) & 2) == 2); /* wait */
+ while (((inb(0x64) & 2) == 2) && --timeout)
+ usleep(1000); /* wait */
+
usleep(1);
outb(0x60, value);
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Re: Two ways to framebuffer
Hi Peter: > > On Thu, Sep 20, 2001 at 11:16:29PM +0100, Gustavo Carvalho Homem wrote: > > Hello all: > hi. > > > One thing I can't find in the documentation. Are there advantages (or > > disadvantages) of using XF86_FBDev over XFree86 4.1 with fbdev driver ? > I only see an advantage for situations where Xserver is missing for your card > but there is a fbdev driver for kernel. If there is no fbdev driver for the > card for the kernel, it is better to use Xfree's vesa driver, because it is > slightly faster than vesafb+fbdev and also can do DGA. > What I was really trying to ask was some comparison between the old XF86_FBDev 3.3.6 frame buffer XServer and the new XFree 4.1 using fbdev driver X server, i.e., a comparison between two framebuffer based XServers and not between the solutions and specific drivers. The reason is I am looking for the most general XServer, the one that will work with more cards, and from what I see the Vesa 1.2 cards are not supported by any framebuffer server, but the Vesa 2.0 ones should be.But still comes the doubt about wich one to use. (seems to be a rule of thumb to use framebuffer XServers in the installer, see Suse, Mandrake, Red Hat for instance). Without further information I'd say it shoud be logical to choose the newer release but I see all the distros using the old one. > > > Any other relevant comments ? > If there is a driver for your card for Xf86, forget it. We don't want to detect/configure the whole X for the installation neither does the user wan't to see that, and you know that it can sometimes be tricky. > > > Thank you very much > > Gustavo Homem > Bye, > Gustavo 2+2=5 for extremely large values of 2 .-``'. I.S.T - Instituto Superior Tecnico .` .`~ _.-' '._ Licenciatura em Engenharia Fisica e Tecnologica [EMAIL PROTECTED] http://mega.ist.utl.pt/~gach ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]lockup in keyboard initialization when no keyboard is connected
Hello, I decided to try XFree86 4.1.0, but it won't start; the X server hangs during keyboard initialization. The following backtrace is from a recent CVS version: (gdb) bt #0 0x80a8fc0 in inb (port=100) at ../../../../../../programs/Xserver/hw/xfree86/common/compiler.h:1135 #1 0x80a8dff in xf86SetKbdRepeat (rad=32 ' ') at lnx_io.c:197 #2 0x808fb91 in xf86InitKBD (init=1) at xf86Io.c:246 #3 0x808fd8c in xf86KbdProc (pKeyboard=0x8372518, what=0) at xf86Io.c:338 #4 0x80cc2eb in InitAndStartDevices () at devices.c:169 #5 0x80e3728 in main (argc=1, argv=0xbccc, envp=0xbcd4) at main.c:403 It hangs in this loop: while ((inb(0x64) & 2) == 2); /* wait */ I guess that line is trying to read something from an AT keyboard, but I don't have one... 4.0.3 did not have this problem. Just putting a 'return' before the infinite loop fixes the problem, and I can also use my USB keyboard. But I don't believe this is a good fix... :) What's a good solution? Regards, Pontus -- Pontus Lidman Email: [EMAIL PROTECTED] Nokia Home Communications Phone: +46 (0)13 4611201 Diskettgatan 11 Fax: +46 (0)13 4611001 583 35 Linköping, Sweden ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
-- Forwarded message -- Date: Fri, 21 Sep 2001 15:07:57 +0200 From: "Michel [iso-8859-1] Dänzer" <[EMAIL PROTECTED]> To: Gustavo Carvalho Homem <[EMAIL PROTECTED]> Cc: [EMAIL PROTECTED] Subject: Re: [Xpert]Two ways to framebuffer Gustavo Carvalho Homem wrote: > One thing I can't find in the documentation. Are there advantages (or > disadvantages) of using XF86_FBDev over XFree86 4.1 with fbdev driver ? Any > other relevant comments ? The big disadvantage is that XF86_FBDev is ooold and lacks a lot of features like shadow framebuffer, RENDER extension, ... The only possible advantages I see are for exotic cases like 1 or 4 bit depth which might be broken in the 4.x fbdev driver (but should be easy to fix) and that maybe XF86_FBDev offers some acceleration for some chipsets, but I'm not even sure there. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]3D on ATI Mobility Radeon-M
Peter Surda wrote: > > On Fri, Sep 21, 2001 at 05:08:29PM +0200, Michel Dänzer wrote: > > Yes, PCI GART is still disabled in source for the Radeon AFAIK. > Hmm I thought pci gart isn't implemented yet for radeon? The code has been there for as long as for the Rage128, it's just still disabled on i386 because there are issues. It's enabled on alpha though. > If it is, could the responsible developers please contact "R C" > <[EMAIL PROTECTED]>, he is developing capturing for radeon AIW and told > me he wrote to dri-devel about this (pci gart for radeon) several months > ago, but got no response, Must have been missed, but he could have found out by searching the archives or reading the code. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
Peter Surda wrote: > On Fri, Sep 21, 2001 at 04:07:11PM +0200, Michel Dänzer wrote: > > When did you last try the fbdev driver? It supports DGA as of 4.1.0 and I > > wonder why it should be slower than the vesa driver as both use the same > > shadow framebuffer. > I stopped using fbdev in January so may be I missed something. > > Shadow is sometimes slower (e.g. on videos), hence I used to run with > disabled shadow. The point stands, both drivers use the same lowlevel code so I'd be surprised if there is a real performance difference. > Anyway, is there really a reason to use vesafb + fbdev instead of vesa? I > don't think so. But of course! Penguin(s) on bootup! ;) Seriously though, fbcon offers a lot of advantages over vgacon. But I won't elaborate as this is OT here. > BTW did you get the patch for R128PutImage+DMA I posted with ICQ? It fixed > the last bug I was aware of. Didn't get it, please mail it to me privately. I'm cleaning up the code and preparing to commit it to DRI CVS. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]3D on ATI Mobility Radeon-M
On Fri, Sep 21, 2001 at 05:08:29PM +0200, Michel Dänzer wrote: > Yes, PCI GART is still disabled in source for the Radeon AFAIK. Hmm I thought pci gart isn't implemented yet for radeon? If it is, could the responsible developers please contact "R C" <[EMAIL PROTECTED]>, he is developing capturing for radeon AIW and told me he wrote to dri-devel about this (pci gart for radeon) several months ago, but got no response, so he was forced to work on his own kernel module. It would be nice not to split the DMA infrastructure. Bye, Peter Surda (Shurdeek) <[EMAIL PROTECTED]>, ICQ 10236103, +436505122023 -- PGP signature
Re: [Xpert]Trident Cyberblade XP XV support
On Fri, Sep 21, 2001 at 07:09:20PM +0200, Anders Rune Jensen wrote: > Im running a Toshiba Tecra 8200 laptop with a trident cyberblade xp gfx > card. The card works in Xfree but lacks xv and dga. I saw that there > were some unofficial patches and maybe some of them fixed the problem. > Could anyone put them in cvs or tell me how to install them. Im running > debian sid. > I've just committed them to CVS, which some updates for > 384 wide images. Can people update and test and give more feedback asap. Thanks. Alan. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Trident Cyberblade XP XV support
Hi Im running a Toshiba Tecra 8200 laptop with a trident cyberblade xp gfx card. The card works in Xfree but lacks xv and dga. I saw that there were some unofficial patches and maybe some of them fixed the problem. Could anyone put them in cvs or tell me how to install them. Im running debian sid. Med venlig Hilsen: - Anders Rune Jensen icq : 52926571 Markusgade 7 4. TH email : [EMAIL PROTECTED] 9000 Ålborg web : http://www.gnulinux.dk PGP signature
Re: [Xpert]3D on ATI Mobility Radeon-M
Brian Lavender wrote: > > On Fri, Sep 21, 2001 at 02:29:16PM +0200, Michel D?nzer wrote: > > You probably need to build the radeon.o kernel module from XFree86 source > > and use that. > > Do I need the agpgart module as well? Yes, PCI GART is still disabled in source for the Radeon AFAIK. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]3D on ATI Mobility Radeon-M
On Fri, Sep 21, 2001 at 02:29:16PM +0200, Michel D?nzer wrote: > Brian Lavender wrote: > > > > I have a laptop with the ATI Rage Mobility Radeon-M on a Sony VAIO > > PCG-GR170K. I did an lspci and this is the ID of the card. It has 8 > > Megs of RAM > > > > #define PCI_CHIP_RADEON_VE 0x4C59 > > > > I got the latest XFree86 4.1.99 by doing a CVS checkout a couple weeks > > ago. I am running RedHat 7.1. X is running, but it doesn't look like > > the 3D support is enabled. If I try to run to run tuxracer, and it is > > going really slow. Is there something I need to enable in my XF86Config-4 > > file? Here is what I am currently using: > > Looks good. > > You probably need to build the radeon.o kernel module from XFree86 source and > use that. > > If that doesn't help, and neither does looking at the server log, please > follow the DRI User Guide at http://dri.sf.net/ . Do I need the agpgart module as well? -- Brian Lavender http://www.brie.com/brian/ ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
Hi dude. On Fri, Sep 21, 2001 at 04:07:11PM +0200, Michel Dänzer wrote: > When did you last try the fbdev driver? It supports DGA as of 4.1.0 and I > wonder why it should be slower than the vesa driver as both use the same > shadow framebuffer. I stopped using fbdev in January so may be I missed something. Shadow is sometimes slower (e.g. on videos), hence I used to run with disabled shadow. Anyway, is there really a reason to use vesafb + fbdev instead of vesa? I don't think so. The only reason I can think of is some obscure hardware combination in which pure vesa driver doesn't work. BTW did you get the patch for R128PutImage+DMA I posted with ICQ? It fixed the last bug I was aware of. Bye, Peter Surda (Shurdeek) <[EMAIL PROTECTED]>, ICQ 10236103, +436505122023 -- They called their place of existence the "Universe", not the "Great Programmer/Universe". PGP signature
Re: [Xpert]Both CRT&LCD impossible on Dell Latitude CPt/V with ATI Mobility M1?
I don't know if this has anything to do with anything, but on the Dell Latitudes we have (C600's? I think)--they have ATI Rage 128 Mobility cards--you can use the 'Option "Display" "Mirror"' to do something like that. Of course, that doesn't mean it'll work for you--as a matter of fact, I think there's actually different syntax for a similar option in the documentation on the XFree86 site (which didn't work for me). Anyway, hope you get it fixed. --Daniel On Thu, Sep 20, 2001 at 08:01:17PM +0200, Eskil Brun wrote: > > I have been trying for quite some time now to make XFree86 4.1.0 (and other > versions) display both on a CRT and the LCD of a Dell Latitude CPt/V with a > Mach64 chipset. But so far I have only managed to get it to display on either > one but not both at the same time. > > The Xserver (XFree86 4.1.0) tells me that the chip in question is a: > ATI 3D Rage Mobility > Chip type 4C4D "LM", version 4, foundry TSMC, class 0, revision 0x01 > > After having searched on google, google groups, Dell Tech support and reading > the XFree86 xpert archive I have still not come up with a solution. > > It seems that it is difficult programming this chipset to do the "right thing" > within X. Something to do with controlling both the LCD and the CRT. M$ > Windows on the other hand seems to have a simpler model and therefore have > implemented a working solution. (I might have gotten this wrong). > > Using the same timing on both the external CRT and the LCD would be just fine. > > Have this already been implemented in XFree85 4.1.0? Or is it never going to > be there because of difficulties? > > > Dell seem to be ignoring the problem for Linux. Here's an excerpt from the > user manual: > > «NOTE: True Simul-mode (the ability to simultaneously display > video of the same content and resolution on both the computer > display screen and the external display) is not supported if > the system is booted to MS-DOS®, using a full-screen MS-DOS > session under a Microsoft Windows operating system, or the > ATI Mobility M1 video driver is not loaded. This limitation > may also be noted in operating systems other than Microsoft > Windows.» > > Not having simultaneous display is very frustrating for the owner. She now has > to lecture with her back to her audience. All help with this is much ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
Peter Surda wrote: > > On Thu, Sep 20, 2001 at 11:16:29PM +0100, Gustavo Carvalho Homem wrote: > > > One thing I can't find in the documentation. Are there advantages (or > > disadvantages) of using XF86_FBDev over XFree86 4.1 with fbdev driver ? > I only see an advantage for situations where Xserver is missing for your > card but there is a fbdev driver for kernel. If there is no fbdev driver for > the card for the kernel, it is better to use Xfree's vesa driver, because it > is slightly faster than vesafb+fbdev and also can do DGA. When did you last try the fbdev driver? It supports DGA as of 4.1.0 and I wonder why it should be slower than the vesa driver as both use the same shadow framebuffer. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
On Thu, Sep 20, 2001 at 11:16:29PM +0100, Gustavo Carvalho Homem wrote: > Hello all: hi. > One thing I can't find in the documentation. Are there advantages (or > disadvantages) of using XF86_FBDev over XFree86 4.1 with fbdev driver ? I only see an advantage for situations where Xserver is missing for your card but there is a fbdev driver for kernel. If there is no fbdev driver for the card for the kernel, it is better to use Xfree's vesa driver, because it is slightly faster than vesafb+fbdev and also can do DGA. In the past using vesa was necessary to use TVOut on ATI cards, however I was able to get r128's tvout working and am on my way to get rid of the small bugs and support mach64 and radeon. > Any other relevant comments ? If there is a driver for your card for Xf86, forget it. > Thank you very much > Gustavo Homem Bye, Peter Surda (Shurdeek) <[EMAIL PROTECTED]>, ICQ 10236103, +436505122023 -- The best things in life are free, but the expensive ones are still worth a look. PGP signature
[Xpert]Xv card list and Vesa 2.0 compliance
Just out of curiosity, did the older millennium I (2064w) and millennium II's (2164w) have Xv hardware on them as well, or did that not get added till later? Thanks, Alex -- On Thu, 20 Sep 2001, Bob wrote: > I was under the impression that the Matrox G100 that I'm struggling with > also supports Xv? If not, that will save me the continued anguish of > trying to get it to work :) The driver doesn't support it. Back when I wrote the Xv support for the mga driver, I was under the impression that the hardware couldn't support it. Mark. __ Terrorist Attacks on U.S. - How can you help? Donate cash, emergency relief information http://dailynews.yahoo.com/fc/US/Emergency_Information/ ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Two ways to framebuffer
Gustavo Carvalho Homem wrote: > One thing I can't find in the documentation. Are there advantages (or > disadvantages) of using XF86_FBDev over XFree86 4.1 with fbdev driver ? Any > other relevant comments ? The big disadvantage is that XF86_FBDev is ooold and lacks a lot of features like shadow framebuffer, RENDER extension, ... The only possible advantages I see are for exotic cases like 1 or 4 bit depth which might be broken in the 4.x fbdev driver (but should be easy to fix) and that maybe XF86_FBDev offers some acceleration for some chipsets, but I'm not even sure there. -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Trident / XVideo / restore from apm
On Fri, Sep 21, 2001 at 12:16:01PM +0100, Alan Hourihane wrote: > Can you check the archives. I'd sent a patch about this a few days ago. > Apply it, and let me know. yes. it works with this patch. -- fabrice ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]3D on ATI Mobility Radeon-M
Brian Lavender wrote: > > I have a laptop with the ATI Rage Mobility Radeon-M on a Sony VAIO > PCG-GR170K. I did an lspci and this is the ID of the card. It has 8 > Megs of RAM > > #define PCI_CHIP_RADEON_VE 0x4C59 > > I got the latest XFree86 4.1.99 by doing a CVS checkout a couple weeks > ago. I am running RedHat 7.1. X is running, but it doesn't look like > the 3D support is enabled. If I try to run to run tuxracer, and it is > going really slow. Is there something I need to enable in my XF86Config-4 > file? Here is what I am currently using: Looks good. You probably need to build the radeon.o kernel module from XFree86 source and use that. If that doesn't help, and neither does looking at the server log, please follow the DRI User Guide at http://dri.sf.net/ . -- Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer XFree86 and DRI project member / CS student, Free Software enthusiast ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Trident / XVideo / restore from apm
On Fri, Sep 21, 2001 at 12:20:37PM +0200, Fabrice Bellet wrote: > On Fri, Sep 21, 2001 at 11:49:57AM +0200, Fabrice Bellet wrote: > > +#if 0 > > OUTW_3C4(SSetup); > > OUTW_3C4(SKey); > > OUTW_3C4(SPKey); > > +#endif > > more precisely, pb occurs because these registers can > be written when uninitialized (all values to zero). > Can you check the archives. I'd sent a patch about this a few days ago. Apply it, and let me know. Alan. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]dri/drm from Xfree 4.1.0 for kernel 2.2.19?
On Fri, Sep 21, 2001 at 12:05:42PM +0200, Sven Garbade wrote: > Hi all, > > does dri/drm from Xfree 4.1.0 is also avaiable for the 2.2.19 kernel > series, and if yes, where can I get the corrct kernel modules? > No. 2.2.x support has been withdrawn. Alan. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Trident / XVideo / restore from apm
On Fri, Sep 21, 2001 at 11:49:57AM +0200, Fabrice Bellet wrote: > +#if 0 > OUTW_3C4(SSetup); > OUTW_3C4(SKey); > OUTW_3C4(SPKey); > +#endif more precisely, pb occurs because these registers can be written when uninitialized (all values to zero). -- fabrice ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]dri/drm from Xfree 4.1.0 for kernel 2.2.19?
Hi all, does dri/drm from Xfree 4.1.0 is also avaiable for the 2.2.19 kernel series, and if yes, where can I get the corrct kernel modules? Thanks, Sven ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Trident / XVideo / restore from apm
Hello, I have a cyber 9397DVD chipset : 01:00.0 VGA compatible controller: Trident Microsystems Cyber 9397DVD (rev f3) (prog-if 00 [VGA]) Subsystem: IBM: Unknown device 00ea Flags: 66Mhz, medium devsel, IRQ 11 Memory at 7080 (32-bit, non-prefetchable) [size=8M] Memory at 7100 (32-bit, non-prefetchable) [size=128K] Memory at 7000 (32-bit, non-prefetchable) [size=8M] Expansion ROM at [disabled] [size=64K] Capabilities: [80] AGP version 1.0 Capabilities: [90] Power Management version 1 I have a problem with recent trident driver and the XVideo extension. When coming back from APM suspend, the XVideo data of my apps are not restored, and the XVideo area remains blue. It appears after a quick search that this patch solves this problem: Index: trident_dac.c === RCS file: /cvs/xc/programs/Xserver/hw/xfree86/drivers/trident/trident_dac.c,v retrieving revision 1.45.2.1 diff -u -r1.45.2.1 trident_dac.c --- trident_dac.c 2001/05/23 20:33:41 1.45.2.1 +++ trident_dac.c 2001/09/21 02:25:21 @@ -648,9 +648,11 @@ OUTW_3CE(MiscIntContReg); OUTW_3CE(MiscExtFunc); OUTW_3x4(Offset); +#if 0 OUTW_3C4(SSetup); OUTW_3C4(SKey); OUTW_3C4(SPKey); +#endif OUTW_3x4(PreEndControl); OUTW_3x4(PreEndFetch); if (pTrident->Chipset >= PROVIDIA9685) OUTW_3x4(Enhancement0); -- fabrice ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]S3 ViRGE problems with Xfree86 v4.02
I tried compiling and installing X on my Toshiba Satellite 490CDT laptop .. it compiles and installs. I have a S3 ViRGE /MX card with 2MB on the laptop which is identified by xf86config. When I to start with any screen resolution > 320 X 200 the machine hangs. After spending hours trying to trace the cause, I discovered the S3 device driver would hang trying to update the MTRRs for Write Combining. Tha machine hangs even with all forms of acceleration disabled. The only fix I could use is to change the driver to VESA. Has anyone else had similar problems ? Any fixes ?? -mohan -- __ Mohan Rajagopalan [EMAIL PROTECTED] (520) 621 2858 Graduate Research Assistant Dept. of Computer Science 817 E Drachman St, #817 University of Arizona Tucson, AZ 85719 __ Yesterday I knew a little less than today ... ... and tommorow a little more. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Both CRT&LCD impossible on Dell Latitude CPt/V with ATI Mobility M1?
OS <[EMAIL PROTECTED]> said: > Hello, > My understanding is this chipset (I have an Inspiron) was better > supported in XF 3.3.6 !! The 'reasoning' appears to be that this > chipset is now too old to bother with !! This goes for 3D > acceleration as well which seemed to be fully supported in 3.3.6 !! > All I can think is to try good old 3.3.6. > > Owen Thank you for this information, Owen. I did try XFree86 3.3.6. Two versions, one from XFree86 and one patched to better support this chipset. Both these servers performed fine on either the CRT or LCD. But alas, not on both. If you know any options in the XFree86 Config file that I might have forgotten and that makes the X server display simultaneously on the CRT and LCD I would very much like to try them out. Eskil... :-) ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Re: Trident & XVideo
On Fri, Sep 21, 2001 at 02:17:09AM -0500, Chad Kitching wrote: > With those changes, I can switch VTs or resolutions within X again (yay!), > however, on my laptop, the Xv image is still offset too far out of the > window (at either 640x480 or 800x600), and the problem with videos > >384pixels wide still happens (left side appears on right and vice versa). > It would appear that I still need to add OUTW(0x3C4, 0x0497) to > TRIDENTDisplayVideo to make it display properly on my Cyber9525, and modify > the timings again to prevent the video from being offset in the window. > Chad, I'll be making a fix to sort the >384 problem out. What timings do you need to modify to prevent the window being offset. > Well, it definately fixed some stuff, although I am a little curious about > the video still being offset. I hope this isn't going to end up being > specific to each laptop using an LCD Trident chipset. > It's certainly looking that way. :-< Alan. ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
