Re: Listing of Xmodmap pointer versus button actions
On Mon, Mar 02, 2009 at 12:52:11PM -0800, Alan Coopersmith wrote: > R. G. Newbury wrote: > > I have been trying to find a reference for the expected actions of mouse > > buttons (actually a Logitech trackball), in order to create a useful > > 'pointer = 1 2 3' etc. line. > > > > For example what is 'pointer = 8' *supposed* to do? This action would be > > bound to button 1 if this line were fed to xmodmap. > > It is supposed to report button 8 presses to the X clients - what the clients > do on those is up to the client, toolkit & window manager, and is configurable > in many of them. > > There's a defacto standard that buttons ids 1 2 3 are the main mouse buttons > (primary as 1, either leftmost or rightmost depending on whether you use your > mouse right-handed or left-handed, secondary as 3, and the "middle" one as > 2), > ids 4 & 5 are vertical scroll wheel events, and 6 & 7 are horizontal scroll > wheel events - all others are just more buttons with no specific role > assigned. Buttons 8 and 9 are often interpreted as history navigation buttons (e.g. by Firefox). > Even what the scroll wheel does is up to the application - it can scroll your > text area, zoom in and out on an image, go forward/back in a browser or > perform some other application-specific function. Marius Gedminas -- "All programs evolve until they can send email." -- Richard Letts "Except Microsoft Exchange." -- Art (found on the Snort web site) signature.asc Description: Digital signature ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: Newer libtool breaking libXfont with -Wl, -single_module (Was Re: autoreconf issue with libXfont)
Jeremy Huddleston wrote: > Ok, after a bit more poking, this seems to be a fairly annoying bug > with glibtool. > > Minimizing any chance of conflict, I have a working build using the > libtool provided by configure in the tarball, and I have a broken > build if I replace it with glibtool-1.5.26 after running ./configure. > The delta between the verbose output of the builds is below. For > those not willing to go crosseyed looking for the difference, it's > just a single -Wl,-single_module being added to the linking in the > 'broken' build. This is not my area of expertise... so if someone > could fill me in on why that is winding up in LDFLAGS and why it's > causing up problems, I'd appreciate it... > Hi Jeremy, Sorry, I don't read the Xorg lists that often, I was marking messages as read today when I happened to notice this. Please bring up issues with libtool either with me, or on the bug-libt...@gnu.org lists. I added -Wl,-single_module to libtool on the request of a couple of Apple engineers a few years ago. For the vast majority of circumstances it should make no difference whatsoever, however there exist some areas where it may make a difference (interposing). export LT_MULTI_MODULE=yes; ./configure; make should "fix" this for you. -single_module is the default in 10.5, and -multi_module is marked as "obsolete" in the ld man page, are you seeing these problems on 10.5? Peter -- Peter O'Gorman http://pogma.com ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
installing xf86-video-intel-2.6.0
Running I get the following: # lspci | grep VGA 00:02.0 VGA compatible controller: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller (rev 04) I trying to follow the steps at http://intellinuxgraphics.org/install.html I've dowloaded and extracted xf86-video-intel-2.6.0.tar.bz2 Then I run : # ./configure -- everything goes fine here # make When I run make I get the following error : make[4]: Entering directory `/home/ericomtx/Desktop/xf86-video-intel-2.6.0/src' ../doltcompile gcc -DHAVE_CONFIG_H -I. -I.. -I/usr/include/xorg -I/usr/include/pixman-1 -I/usr/local/include -I/usr/local/include/drm -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -fno-strict-aliasing -I/usr/include/xorg -I/usr/include/pixman-1 -I/usr/local/include -I/usr/local/include/drm -I/usr/include/X11/dri -I../uxa -DI830_XV -DI830_USE_XAA -DI830_USE_EXA -g -O2 -MT i830_lvds.lo -MD -MP -MF .deps/i830_lvds.Tpo -c -o i830_lvds.lo i830_lvds.c i830_lvds.c: In function ‘i830_lvds_get_modes’: i830_lvds.c:827: error: ‘DS_VENDOR’ undeclared (first use in this function) i830_lvds.c:827: error: (Each undeclared identifier is reported only once i830_lvds.c:827: error: for each function it appears in.) make[4]: *** [i830_lvds.lo] Error 1 make[4]: Leaving directory `/home/ericomtx/Desktop/xf86-video-intel-2.6.0/src' make[3]: *** [all-recursive] Error 1 make[3]: Leaving directory `/home/ericomtx/Desktop/xf86-video-intel-2.6.0/src' make[2]: *** [all] Error 2 make[2]: Leaving directory `/home/ericomtx/Desktop/xf86-video-intel-2.6.0/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/ericomtx/Desktop/xf86-video-intel-2.6.0' make: *** [all] Error 2 What am I missing ? ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: [ANNOUNCE] xf86-input-evdev 2.1.99.1
On Mon, Mar 02, 2009 at 09:06:12AM +, Alex Bennee wrote: > 2009/3/2 Peter Hutterer : > > First snapshot for evdev 2.2. > > > > Lots of cleanup, and - most notably - general axis/button support. > > For those running an X server from master, evdev will label axes and buttons > > for you. > > > > shortlog is a bit longer than it actually is, it includes a number of > > patches > > that were cherry-picked onto 2.1 and have since been released. > > > > Add support for button labelling. > > evdev 2.1.99.1 > > Seems to have broken: > > gcc -DHAVE_CONFIG_H -I. -I.. -I../include/ -Wall -Wpointer-arith > -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations > -Wnested-externs -fno-strict-aliasing -Wbad-function-cast > -Wold-style-definition -Wdeclaration-after-statement > -fvisibility=hidden -I/home/alex/src/xorg2/install/include/xorg > -I/home/alex/src/xorg2/install/include/pixman-1 > -I/home/alex/src/xorg2/install/include -g3 -MT evdev.lo -MD -MP -MF > .deps/evdev.Tpo -c evdev.c -fPIC -DPIC -o .libs/evdev.o > evdev.c:1806: error: 'BTN_LABEL_PROP_BTN_0' undeclared here (not in a > function) > evdev.c:1807: error: 'BTN_LABEL_PROP_BTN_1' undeclared here (not in a > function) > evdev.c:1808: error: 'BTN_LABEL_PROP_BTN_2' undeclared here (not in a > function) > evdev.c:1809: error: 'BTN_LABEL_PROP_BTN_3' undeclared here (not in a > function) > evdev.c:1810: error: 'BTN_LABEL_PROP_BTN_4' undeclared here (not in a > function) > evdev.c:1811: error: 'BTN_LABEL_PROP_BTN_5' undeclared here (not in a > function) > evdev.c:1812: error: 'BTN_LABEL_PROP_BTN_6' undeclared here (not in a > function) > evdev.c:1813: error: 'BTN_LABEL_PROP_BTN_7' undeclared here (not in a > function) > evdev.c:1814: error: 'BTN_LABEL_PROP_BTN_8' undeclared here (not in a > function) > evdev.c:1815: error: 'BTN_LABEL_PROP_BTN_9' undeclared here (not in a > function) > oops. forgot to push xserver, a706dd8771f is the matching commit now. Thanks. Cheers, Peter ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: Listing of Xmodmap pointer versus button actions
R. G. Newbury wrote: > I have been trying to find a reference for the expected actions of mouse > buttons (actually a Logitech trackball), in order to create a useful > 'pointer = 1 2 3' etc. line. > > For example what is 'pointer = 8' *supposed* to do? This action would be > bound to button 1 if this line were fed to xmodmap. It is supposed to report button 8 presses to the X clients - what the clients do on those is up to the client, toolkit & window manager, and is configurable in many of them. There's a defacto standard that buttons ids 1 2 3 are the main mouse buttons (primary as 1, either leftmost or rightmost depending on whether you use your mouse right-handed or left-handed, secondary as 3, and the "middle" one as 2), ids 4 & 5 are vertical scroll wheel events, and 6 & 7 are horizontal scroll wheel events - all others are just more buttons with no specific role assigned. Even what the scroll wheel does is up to the application - it can scroll your text area, zoom in and out on an image, go forward/back in a browser or perform some other application-specific function. -- -Alan Coopersmith- alan.coopersm...@sun.com Sun Microsystems, Inc. - X Window System Engineering ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Eeepc901 KMS: LDVS detection problems
Hello, recently I tried to make kernel modesetting work on my Asus Eeepc 901 laptop (it has an Intel 945GSE display chipset). Whenever KMS is activated in kernel (2.6.29-rc6) the LCD gets black (no output), but VGA output is working OK. If I deactivate KMS the built-in LCD (AUOptronics model) works OK. My software system configuration is: Xorg server 1.6, xorg-video-intel is latest from git (since 2.6.2 does suffer from memory leaking error previously discussed here). Xrandr output with KMS: Screen 0: minimum 320 x 200, current 1680 x 1050, maximum 2048 x 2048 VGA1 connected 1680x1050+0+0 (normal left inverted right x axis y axis) 474mm x 296mm 1680x1050 59.9*+ 60.0 1280x1024 75.0 1152x864 75.0 1024x768 75.1 60.0 832x62474.6 800x60075.0 60.3 56.2 640x48075.0 60.0 720x40070.1 LVDS1 connected 1024x600+0+0 (normal left inverted right x axis y axis) 195mm x 113mm 1024x600 60.0*+ Xrandr output witthout KMS ( external monitor disconected): Screen 0: minimum 320 x 200, current 1024 x 600, maximum 1024 x 1024 VGA disconnected (normal left inverted right x axis y axis) LVDS connected 1024x600+0+0 (normal left inverted right x axis y axis) 195mm x 113mm 1024x600 60.0*+ 800x60085.1 72.2 75.0 60.3 56.2 640x48085.0 72.8 75.0 59.9 720x40085.0 640x40085.1 640x35085.1 I tried to make some changes to LVDS quirks inside of the xorg-video-intel driver, but had no success. I also saw that there were some kernel changes related to intel DRM lately, maybe the problems are in the kernel ... Do you know what else can I try? Daniel ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: Intel driver - Dual head and 2D performance
Hi, It could be you are runnig into bug http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6708580 I wrote a small patch which has been integrated into JDK6u12, which reverts to software-only rendering in the case EXA is used. So, does the problem also happen with JDK6u12? - Clemens 2009/3/2 Luiz Felipe Talvik : > Hi, > > I`m trying to run an java app that runs on full screen on two monitors > and it is constantly drawing on the screen. I`m having trouble with > performance on new drivers(Ubuntu 8.10, 8.04). And on older > drivers(Ubuntu 7.04, CentOS 5) it runs much faster, but I`m having > trouble with the dual monitor(the LVDS monitor uses the virtual > resolution, instead of assuming just half). > The best combination was Xorg 1.3.0 and intel 2.1.1, it ran fairly > decent and the dual monitor config worked correctly. > > What would be the best combination of Xorg, intel and kernel version, > considering Java2D performance on dual monitor? (i`d like to know even > if the solution involves compiling, coding and patching) > > Thanks > ___ > xorg mailing list > xorg@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/xorg > ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: Xlib/GLX multiple displays and threads
Rohit Garg wrote: > Hi, > > I am writing an app that needs to render different stuff to different > windows (not full screen). Windows (the correct X name for that is > display right?) are completely independent and share no data and have > no communication. Naturally, I want to use multiple threads to do it. > Ie, each thread has some local data and renders to it's window, > blissfully unaware of what other thread does in the other window. All > manipulation is via keyboard/mouse input. IE, no toolbars/widgets to > bother with. > > Can Xlib+glx+opengl be used in this manner? It should work in principle. In practice it may not. In the past when I've tried to use Xlib from multiple threads I've run into bugs. Depending on the version of Xlib I've seen X protocol/sequence errors or mutex deadlocks. Using XInitThreads(), XLock/UnlockDisplay(), etc. didn't seem to help much. However, I haven't tried the latest/greatest version of Xlib w/ XCB. Also, while OpenGL supports multiple rendering threads, some drivers probably haven't been exercised in that area. There's a Mesa test (progs/xdemos/glthreads.c) that tests rendering from multiple threads. You might grab that and give it a try on your system. There's a few command-line options that might be of interest. > In this case would I need > to lock a mutex or call XInitThreads()/Xlockdisplay()/xunlockdisplay() > at each instance of a x function call? I really don't want to do that. > > Running multiple processes will alleviate the pain but will make some > other things difficult in other places. If it is not going to be as > simple as an embarrassingly parallel thing, then please let me know. > -Brian ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
[ANNOUNCE] xproto 7.0.15
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 James Cloos (4): Add dead keysyms for DOUBLE GRAVE and INVERTED BREVE accents Add a dead_belowcomma keysym [keysymdef.h] Add dead_currency symbol. Bump to 7.0.15. Paulo Cesar Pereira de Andrade (1): Janitor: Correct make distcheck and dont distribute autogen.sh Peter Hutterer (1): Add XF86XK_Suspend and XF86XK_Hibernate keysym defs. Sayamindu Dasgupta (1): Add dead vowels in keysymdef.h git tag: xproto-7.0.15 http://xorg.freedesktop.org/archive/individual/proto/xproto-7.0.15.tar.bz2 MD5: 1a2b31430d04340be2e49e8b6445e076 xproto-7.0.15.tar.bz2 SHA1: 891e3f8e2c17128a8fc6930583e42fd3db75839a xproto-7.0.15.tar.bz2 http://xorg.freedesktop.org/archive/individual/proto/xproto-7.0.15.tar.gz MD5: 9186713de3ec8a9fdd6ec3f4c4534fde xproto-7.0.15.tar.gz SHA1: b86ffce7904f7ca3b9d522a51a4051b9e69c6d96 xproto-7.0.15.tar.gz -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.10 (GNU/Linux) iEYEARECAAYFAkmsLeoACgkQmXqfF+19rqbQBwCgpSHHA8XHoZV+Qav9ipr99Xe5 /HsAn2waHaT40QjwR6zc2FogHzL2eHfq =G7B+ -END PGP SIGNATURE- ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Intel driver - Dual head and 2D performance
Hi, I`m trying to run an java app that runs on full screen on two monitors and it is constantly drawing on the screen. I`m having trouble with performance on new drivers(Ubuntu 8.10, 8.04). And on older drivers(Ubuntu 7.04, CentOS 5) it runs much faster, but I`m having trouble with the dual monitor(the LVDS monitor uses the virtual resolution, instead of assuming just half). The best combination was Xorg 1.3.0 and intel 2.1.1, it ran fairly decent and the dual monitor config worked correctly. What would be the best combination of Xorg, intel and kernel version, considering Java2D performance on dual monitor? (i`d like to know even if the solution involves compiling, coding and patching) Thanks ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: int10 using vbios from wrong card
On Sat, 2009-02-21 at 19:44 +1100, Timothy S. Nelson wrote: > On Fri, 20 Feb 2009, Steven J Newbury wrote: > > > While trying to figure out where my VGA->SCART video output has > > disappeared, I noticed my secondary Radeon 9250 is being booted with the > > VBIOS from the primary X800 as can be seen in the attached log. Older > > logs from working versions seem to show the same thing, so that's not > > necessarily caused my immediate problem of no modes on VGA-0... > > Does it look like it's at all related to my final comments on this bug > here? > > https://bugs.freedesktop.org/show_bug.cgi?id=18160 > > Looking at my log, it doesn't mention the BIOS either way, but I'm > using the nv and sis drivers, not radeon, so that may have something to do > with it. > > You may or may not find some of the information attached to that bug > report helpful. > > HTH. > > :) Definitely related... I'll post anything further there. ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Xlib/GLX multiple displays and threads
Hi, I am writing an app that needs to render different stuff to different windows (not full screen). Windows (the correct X name for that is display right?) are completely independent and share no data and have no communication. Naturally, I want to use multiple threads to do it. Ie, each thread has some local data and renders to it's window, blissfully unaware of what other thread does in the other window. All manipulation is via keyboard/mouse input. IE, no toolbars/widgets to bother with. Can Xlib+glx+opengl be used in this manner? In this case would I need to lock a mutex or call XInitThreads()/Xlockdisplay()/xunlockdisplay() at each instance of a x function call? I really don't want to do that. Running multiple processes will alleviate the pain but will make some other things difficult in other places. If it is not going to be as simple as an embarrassingly parallel thing, then please let me know. -- Rohit Garg http://rpg-314.blogspot.com/ Senior Undergraduate Department of Physics Indian Institute of Technology Bombay ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
multiseat automating script(s) for dual headed multiple VGAS
Hi All. I ave been(still) searching for multiseat automating script(s) that work with dual headed multiple VGAS. Kindly if anyone has and is will to share with me, I will appreciate greatly. Thanks. M. PS: Am using five(5) dual headed VGAs per machine. VGAs => nVidia-GeForce FX 5500. => ATI-Radeon Mobility 9200. Distros: Fedora 6,7,8,9 or 10. Ubuntu 7.10,8.04 or 8.10 OpenSuse 11.0 or 11.1 Mandriva 2009. ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
multiseat automating script(s) for dual headed multiple VGAS
Hi All. I ave been(still) searching for multiseat automating script(s) that work with dual headed multiple VGAS. Kindly if anyone has and is will to share with me, I will appreciate greatly. Thanks. M. PS: Am using five(5) dual headed VGAs per machine. VGAs => nVidia-GeForce FX 5500. => ATI-Radeon Mobility 9200. Distros: Fedora 6,7,8,9 or 10. Ubuntu 7.10,8.04 or 8.10 OpenSuse 11.0 or 11.1 Mandriva 2009. ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: visual and PictFormat related question
Alexei Babich wrote: > > xdpyinfo shows, that my X-server supports two visuals: 24 and 32 bits > (visuals IDs is: 0x21, 0x39); I aim to apply 'PictStandardARGB32' > PictFormat to the windows. Visual having ID 0x39 is suitable for this > type of PictFormat. But default visual of the root window is 0x21, so > when I try to create windows with 0x39 visual, the root window as a > parent, I always get a BadMatch error. 9 times out of 10 when I get that error, I find that I forgot to create a new colormap with the new visual. You might also want to check that you aren't trying to CopyFromParent anything else related to depth (border pixmap, etc). Yeah, I know, forcing the app to create a "colormap" for a TrueColor visual is silly, but the protocol requires a Match error if you don't. Peter Harris -- Open Text Connectivity Solutions Group Peter Harrishttp://www.opentext.com/connectivity Research and DevelopmentPhone: +1 905 762 6001 phar...@opentext.comToll Free: 1 877 359 4866 ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: hp 2133, xorg and broken VTs
Leonie Herzberg wrote: > hi! i've just got a hp 2133 laptop computer with a via chrome > graphics card. now i tried to get gnuffy to run on it (which is some > arch linux fork you probably don't know) > > i tried both the vesa and the openchrome video driver which both had > the "nice" effect of making my VTs unreadable after X has been > started. which means, if i use the vesa driver, i get giant dancing > pixel clouds instead of letters, and, if i use the openchrome driver, > i get green squares over the whole display (no matter if X is still > running, or if I zapped or exited) that only reboot can recover. > > X itself shows a rather usual picture (except that the fluxbox task > bar seems to be missing if i use the openchrome driver, i have not > yet figured out why). i basically used the xorg.conf i found on the > gentoo wiki article about the hp2133 which is why i can't understand > why it doesn't work. i adjusted some options that obviously didn't > work on my system (threw errors). and, as you probably already > assume, i have approximately no idea about what i am doing to my poor > xorg. i would be grateful for any help :) > Please update to latest openchrome trunk, the VT switch issue should be fixed. You shouldn't need the modeline nor to force the panel and the software cursor on in you're xorg conf. Regards, Xavier ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: [ANNOUNCE] xf86-input-evdev 2.1.99.1
On Mon, 2009-03-02 at 16:42 +1000, Peter Hutterer wrote: > First snapshot for evdev 2.2. > > Lots of cleanup, and - most notably - general axis/button support. > For those running an X server from master, evdev will label axes and buttons > for you. Theoretically. ;) It seems that the BTN_LABEL_PROP_* definitions haven't been committed to wherever it is they are supposed to be, evdev.c:1806: error: ‘BTN_LABEL_PROP_BTN_0’ undeclared here (not in a function) evdev.c:1807: error: ‘BTN_LABEL_PROP_BTN_1’ undeclared here (not in a function) [it goes on] I'm running pretty much everything from git. Moreover, grep wasn't able to find those symbols anywhere in my xorg tree. Signs point to a missing commit. Just figured I should probably let you know. Thanks a ton for the work, - Ben ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Listing of Xmodmap pointer versus button actions
I have been trying to find a reference for the expected actions of mouse buttons (actually a Logitech trackball), in order to create a useful 'pointer = 1 2 3' etc. line. For example what is 'pointer = 8' *supposed* to do? This action would be bound to button 1 if this line were fed to xmodmap. I have been completely unable to find any assistance inthe man pages or online. Can someone help with a pointer? Geoff Tux says: "Be regular. Eat cron flakes." ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: [ANNOUNCE] xf86-video-intel 2.6.2
Jacek Luczak pisze: > Eric Anholt pisze: >> On Wed, 2009-02-25 at 19:13 +0100, Brice Goglin wrote: >>> Eric Anholt wrote: On Wed, 2009-02-25 at 14:10 +0100, Jacek Luczak wrote: > this release is totally unusable while running in UXA. System eats lot of > memory, including swapping. Is this that, reported earlier, ,,memory > leak''? As > a result X are really slow and lot of lockups occur (everything freeze > for a few > seconds). At the end I can't even switch to text console, but system > reacts on > power button and it goes down successfully. With EXA there's no such > issue: > total system memory usage around 40%, no lockups. > > If it's not know issue I will try to bisect it down. > > -Jacek > > -- > Details: > 1. X.Org X Server 1.5.99.903 (1.6.0 RC 3) > 2. Linux Kernel 2.6.29-rc6 Can you give me exact steps to reproduce this leak? >>> Same problem here with libdrm 2.4.5, intel 2.6.2, Xserver 1.6-rc2, >>> Mesa 7.3 on >>> Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated >>> Graphics Controller [8086:27a2] (rev 03) >>> >>> Nothing specific needed to reproduce the problem, just use X :) >>> As soon as I start a big application such as firefox ou thunderbird, >>> my 1GB RAM is entirely used after a couple seconds and the system >>> becomes vey slow. >>> >>> Only downgrade intel to 2.6.1 => 200MB used, no problem. >> I think the problem here was the DRI2 tiling fix, which was great for >> the 915-class 3D performance regression but bad for 915-class 2D. I've >> pushed a fix to master that should help. If it does, I'll try to get a >> 2.6.3 out soon. >> >> commit 5bfd73cd31ba197a62f549cdbad1a1270b571027 >> Author: Eric Anholt >> Date: Fri Feb 27 19:09:49 2009 -0800 >> >> Only allocate pixmaps aligned for tiling when requested by DRI2 >> GetBuffers. >> >> This saves massive quantities of memory on pre-965 since the DRI2 tiling >> enable caused the minimum size of any pixmap to be 1MB. >> > > This fixes memory leak, but ... another problem occurs. Just launch glxgears > (in > UXA) and leave it for a few minutes. Glxgears stops (I mean gears stops:D) and > screen is broken. I will try to grab screen. > > -Jacek > I've got more info. As I've wrote, screen locks and ugly stuff shows up [1]. I was trying to grab screen with XV, as a result properly looking screen was grabbed (I mean without those strange artifacts). When I'm switching to text mode (with VESA fb) and back to X screen is ,,refreshed'' and gears are spinning (all values which were not displayed previously are now visible) and no artifacts on the screen. When I leave it for a while gears stops again. This issue exists in UXA (btw: glxgears shows ~400FPS) and not in EXA (glxgears shows 56FPS and one can see ,,frame-by-frame'' display effect). My hardware is: VGA compatible controller: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller (rev 03) Kernel: vanilla Linux Kernel 2.6.29-rc6 Xorg: xorg-server 1.6 with all latest libs, etc. installed. -Jacek [1] Screen: http://pin.if.uz.zgora.pl/~difrost/download/02032009.jpg ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: how to display a XImage with 16bits depth in a drawable with a depth of 24bits ?
Glynn Clements a écrit : > hd wrote: > > >> I my app, I receive a raw image in 16 bits depth. >> My defaut visual have a depth of 24 bits (and all other Visual have 24 >> or 32bits depth on my X server). >> >> To display the bitmap I try to create an XImage with XCreateImage() and >> display it with XPutImage(). >> But X displays a "bad match" error >> >> If I try to display a "raw image" with the same depth (24bits) as the >> visual (24bits), my app works. >> >> Question : XImage APIs (XCreateImage() / XPutImage() / ...) can convert >> 'on the fly' the depth of the bitmap to match the visual depth >> or must I convert myself the raw bitmap to match the visual depth ? >> (convert each pixel from 16bits to 24 bits) >> > > You must convert yourself, or use a higher-level interface. The core X > protocol doesn't do format conversions. > > If you want the conversion to be done in hardware, look at OpenGL or > XRender. It's implementation-dependent whether either of these > supports a specific format (other than 24-bit RGB), so you still need > to provide a software conversion. > thanks a lot for your help. regards ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: [ANNOUNCE] xf86-video-intel 2.6.2
Eric Anholt pisze: > On Wed, 2009-02-25 at 19:13 +0100, Brice Goglin wrote: >> Eric Anholt wrote: >>> On Wed, 2009-02-25 at 14:10 +0100, Jacek Luczak wrote: this release is totally unusable while running in UXA. System eats lot of memory, including swapping. Is this that, reported earlier, ,,memory leak''? As a result X are really slow and lot of lockups occur (everything freeze for a few seconds). At the end I can't even switch to text console, but system reacts on power button and it goes down successfully. With EXA there's no such issue: total system memory usage around 40%, no lockups. If it's not know issue I will try to bisect it down. -Jacek -- Details: 1. X.Org X Server 1.5.99.903 (1.6.0 RC 3) 2. Linux Kernel 2.6.29-rc6 >>> Can you give me exact steps to reproduce this leak? >> Same problem here with libdrm 2.4.5, intel 2.6.2, Xserver 1.6-rc2, >> Mesa 7.3 on >> Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated >> Graphics Controller [8086:27a2] (rev 03) >> >> Nothing specific needed to reproduce the problem, just use X :) >> As soon as I start a big application such as firefox ou thunderbird, >> my 1GB RAM is entirely used after a couple seconds and the system >> becomes vey slow. >> >> Only downgrade intel to 2.6.1 => 200MB used, no problem. > > I think the problem here was the DRI2 tiling fix, which was great for > the 915-class 3D performance regression but bad for 915-class 2D. I've > pushed a fix to master that should help. If it does, I'll try to get a > 2.6.3 out soon. > > commit 5bfd73cd31ba197a62f549cdbad1a1270b571027 > Author: Eric Anholt > Date: Fri Feb 27 19:09:49 2009 -0800 > > Only allocate pixmaps aligned for tiling when requested by DRI2 > GetBuffers. > > This saves massive quantities of memory on pre-965 since the DRI2 tiling > enable caused the minimum size of any pixmap to be 1MB. > This fixes memory leak, but ... another problem occurs. Just launch glxgears (in UXA) and leave it for a few minutes. Glxgears stops (I mean gears stops:D) and screen is broken. I will try to grab screen. -Jacek ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: [ANNOUNCE] xf86-video-intel 2.6.2
On Sunday 01 March 2009 13:40:16 Vasily Khoruzhick wrote: > Thanks, fix works for me (no more memory leakage) :) Btw, it seems that it's a fix of catalyst, not of the real leak: after ~24h of usage 398mb of swap is used. lsof | grep "drm mm object" | wc -l shows 700, is it OK? signature.asc Description: This is a digitally signed message part. ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: Severe memory leaks make X.org unuseable
> I have looked at the source code for X.org and the savage driver. I cannot > really quite understand it at all. Is there a guide to describe and document > the X.org internals in detail? There are documents for EXA and XAA. The majority of each driver is shared code so if you've got a chipset specific leak as it sounds then you can should only need to look at the driver specific code. If its platform related you may need to dig into os/hw but again this is all isolated from the core code. Fortunately most drivers do very little private memory allocation so it shouldn't take long to check all the cases. Also worth running the same test on another OS to pin down the trigger ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
Re: [ANNOUNCE] xf86-input-evdev 2.1.99.1
2009/3/2 Peter Hutterer : > First snapshot for evdev 2.2. > > Lots of cleanup, and - most notably - general axis/button support. > For those running an X server from master, evdev will label axes and buttons > for you. > > shortlog is a bit longer than it actually is, it includes a number of patches > that were cherry-picked onto 2.1 and have since been released. > > Add support for button labelling. > evdev 2.1.99.1 Seems to have broken: gcc -DHAVE_CONFIG_H -I. -I.. -I../include/ -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -fno-strict-aliasing -Wbad-function-cast -Wold-style-definition -Wdeclaration-after-statement -fvisibility=hidden -I/home/alex/src/xorg2/install/include/xorg -I/home/alex/src/xorg2/install/include/pixman-1 -I/home/alex/src/xorg2/install/include -g3 -MT evdev.lo -MD -MP -MF .deps/evdev.Tpo -c evdev.c -fPIC -DPIC -o .libs/evdev.o evdev.c:1806: error: 'BTN_LABEL_PROP_BTN_0' undeclared here (not in a function) evdev.c:1807: error: 'BTN_LABEL_PROP_BTN_1' undeclared here (not in a function) evdev.c:1808: error: 'BTN_LABEL_PROP_BTN_2' undeclared here (not in a function) evdev.c:1809: error: 'BTN_LABEL_PROP_BTN_3' undeclared here (not in a function) evdev.c:1810: error: 'BTN_LABEL_PROP_BTN_4' undeclared here (not in a function) evdev.c:1811: error: 'BTN_LABEL_PROP_BTN_5' undeclared here (not in a function) evdev.c:1812: error: 'BTN_LABEL_PROP_BTN_6' undeclared here (not in a function) evdev.c:1813: error: 'BTN_LABEL_PROP_BTN_7' undeclared here (not in a function) evdev.c:1814: error: 'BTN_LABEL_PROP_BTN_8' undeclared here (not in a function) evdev.c:1815: error: 'BTN_LABEL_PROP_BTN_9' undeclared here (not in a function) -- Alex, homepage: http://www.bennee.com/~alex/ CV: http://www.bennee.com/~alex/cv.php ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
evdev commit 051cb8b528f6eddddccee44842bcb2152ed0e418
This commit seems to break building of evdev: gcc -DHAVE_CONFIG_H -I. -I.. -I../include/ -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -fno-strict-aliasing -Wbad-function-cast -Wold-style-definition -Wdeclaration-after-statement -fvisibility=hidden -I/home/alex/src/xorg2/install/include/xorg -I/home/alex/src/xorg2/install/include/pixman-1 -I/home/alex/src/xorg2/install/include -g3 -MT evdev.lo -MD -MP -MF .deps/evdev.Tpo -c evdev.c -fPIC -DPIC -o .libs/evdev.o evdev.c:1806: error: 'BTN_LABEL_PROP_BTN_0' undeclared here (not in a function) evdev.c:1807: error: 'BTN_LABEL_PROP_BTN_1' undeclared here (not in a function) evdev.c:1808: error: 'BTN_LABEL_PROP_BTN_2' undeclared here (not in a function) evdev.c:1809: error: 'BTN_LABEL_PROP_BTN_3' undeclared here (not in a function) evdev.c:1810: error: 'BTN_LABEL_PROP_BTN_4' undeclared here (not in a function) evdev.c:1811: error: 'BTN_LABEL_PROP_BTN_5' undeclared here (not in a function) evdev.c:1812: error: 'BTN_LABEL_PROP_BTN_6' undeclared here (not in a function) evdev.c:1813: error: 'BTN_LABEL_PROP_BTN_7' undeclared here (not in a function) evdev.c:1814: error: 'BTN_LABEL_PROP_BTN_8' undeclared here (not in a function) evdev.c:1815: error: 'BTN_LABEL_PROP_BTN_9' undeclared here (not in a function) Is there a missing dependancy somewhere else in the source? -- Alex, homepage: http://www.bennee.com/~alex/ CV: http://www.bennee.com/~alex/cv.php ___ xorg mailing list xorg@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/xorg
