Re: (another) Intel driver change needs testing.
On Mon, 17 May 2010 22:16:54 +0100 Owain Ainsworth zer...@googlemail.com wrote: For those of you new here, or termnally adsent minded, instructions follow: You've got be careful with those termnally adsent folks. ;) With the 845G and a minimal xorg.conf (only NoTrapSignals) I get the expected: (II) intel(0): DPMS enabled (==) intel(0): Intel XvMC decoder disabled (II) intel(0): Set up overlay video (II) intel(0): direct rendering: DRI2 Enabled When using: Option XvMC true It seems to become enabled, but really doesn't do anything. (II) intel(0): DPMS enabled (==) intel(0): Intel XvMC decoder disabled (II) intel(0): Set up overlay video (II) intel(0): direct rendering: DRI2 Enabled Trying to test xvmc with `mplayer -vo xvmc` just results in: Error opening/initializing the selected video_out (-vo) device. Whether enabled or disabled in xorg.conf, (so far) it seems to have no effect on the occasional/intermittent screen corruption bug seen by dcoppa (915GM) and me (845G). -- The OpenBSD Journal - http://www.undeadly.org
Ο Τύπος της Κατοχής
[IMAGE] Am jat\ k\hor ^qhe s' es\r, aut| to l^mula, paqajak~ apamt^ste lar le tgm ]mdeing DIACQAVG. [demime 1.01d removed an attachment of type image/jpeg which had a name of typos-katoxhs_small.jpg]
Re: (another) Intel driver change needs testing.
On Tue, May 18, 2010 at 4:40 PM, J.C. Roberts list-...@designtools.org wrote: Whether enabled or disabled in xorg.conf, (so far) it seems to have no effect on the occasional/intermittent screen corruption bug seen by dcoppa (915GM) and me (845G). The occasional/intermittent screen corruption bug seems to have disappeared for me with latest xenocara -current + the experimental Mesa 7.8 update. cheers, david
$LINENO in trap
My first patch that solves this :
http://cvs.openbsd.org/cgi-bin/query-pr-wrapper?full=yesnumbers=6195
Comments?
Index: history.c
===
RCS file: /cvs/src/bin/ksh/history.c,v
retrieving revision 1.38
diff -u -p -r1.38 history.c
--- history.c 1 May 2010 21:09:23 - 1.38
+++ history.c 18 May 2010 13:42:38 -
@@ -226,7 +226,7 @@ c_fc(char **wp)
Source *sold = source;
int ret;
- ret = command(editor ? editor : ${FCEDIT:-/bin/ed} $_);
+ ret = command(editor ? editor : ${FCEDIT:-/bin/ed} $_, 0);
source = sold;
if (ret)
return ret;
@@ -295,7 +295,7 @@ hist_execute(char *cmd)
*/
/* XXX: source should not get trashed by this.. */
sold = source;
- ret = command(cmd);
+ ret = command(cmd, 0);
source = sold;
return ret;
}
Index: main.c
===
RCS file: /cvs/src/bin/ksh/main.c,v
retrieving revision 1.45
diff -u -p -r1.45 main.c
--- main.c 29 Jan 2009 23:27:26 - 1.45
+++ main.c 18 May 2010 13:42:38 -
@@ -460,13 +460,18 @@ include(const char *name, int argc, char
return i 0xff;/* 0xff to ensure value not -1 */
}
+/*
+ * spawn a command into a shell optionally keeping track of line
+ * number.
+ */
int
-command(const char *comm)
+command(const char *comm, int line)
{
Source *s;
s = pushs(SSTRING, ATEMP);
s-start = s-str = comm;
+ s-line = line;
return shell(s, false);
}
Index: proto.h
===
RCS file: /cvs/src/bin/ksh/proto.h,v
retrieving revision 1.32
diff -u -p -r1.32 proto.h
--- proto.h 29 Jan 2009 23:27:26 - 1.32
+++ proto.h 18 May 2010 13:42:38 -
@@ -159,7 +159,7 @@ voidmbset(char *);
void mpset(char *);
/* main.c */
intinclude(const char *, int, char **, int);
-intcommand(const char *);
+intcommand(const char *, int);
intshell(Source *volatile, int volatile);
void unwind(int) __attribute__((__noreturn__));
void newenv(int);
Index: trap.c
===
RCS file: /cvs/src/bin/ksh/trap.c,v
retrieving revision 1.22
diff -u -p -r1.22 trap.c
--- trap.c 30 Mar 2005 17:16:37 - 1.22
+++ trap.c 18 May 2010 13:42:38 -
@@ -235,7 +235,7 @@ runtrap(Trap *p)
/* Note: trapstr is fully parsed before anything is executed, thus
* no problem with afree(p-trap) in settrap() while still in use.
*/
- command(trapstr);
+ command(trapstr, current_lineno);
exstat = oexstat;
if (i == SIGEXIT_ || i == SIGERR_) {
if (p-flags TF_CHANGED)
--
Manuel Giraud
Re: $LINENO in trap
Comments? ...snip... After a quick read, it looks good to me, but I've not built it and tested it because I don't have access to my openbsd system at the moment.
ULTIMAS VACANTES! Charla gratuita de marketing digital
Publicidad: TALLER GRATUITO PARA EMPRESAS NUEVAS TENDENCIAS DE MARKETING EN INTERNET, LA WEB 2.0 - ULTIMAS VACANTES - CIERRE DE INSCRIPCION: MIERCOLES 19/05/2010 13:30 Hs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sepa csmo convertir usuarios en clientes Participe de un evento para empresas que quieren conocer csmo pueden vender mas a travis de Internet. Solo 15 empresas participaran gratuitamente de este evento exclusivo organizado por Grupo BGL. Benefmciese conociendo las zltimas tendencias en marketing digital, con profesionales que asesoran a empresas en la vida real, generandoles mas y mejores negocios. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TEMARIO ⢠Conozca las estrategias, medios y herramientas que hacen vender mas en Internet a miles de empresas. ⢠Sepa cuales son y csmo utilizar los recursos para que su empresa aumente sus ventas gracias al marketing y la publicidad Online. ⢠Como convertir su Sitio Web en un verdadero canal de ventas? ⢠La web 2.0 no esperemos que los usuarios nos encuentren, salgamos en su bzsqueda. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DESTINADO a: Responsables de marketing, publicidad y ventas y/o directivos de empresas Fecha: Jueves 20 de Mayo de 2010 Horario: de 9 a 11 hs. Lugar: Salguero 2128 (e/ Mitre y Matheu) Ptdo. San Martmn, Bs. As. Organiza: Grupo BGL Marketing Digital www.grupobgl.com CIERRE DE INSCRIPCION: MIERCOLES 19/05/2010 13:30 Hs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IMPORTANTE Reserve hasta 2 vacantes por empresa. Si esta interesado en participar rogamos nos confirme su asistencia completando el formulario. Este email fue enviado a tech@openbsd.org, Si no desea recibir mas newsletters haga click aqum Nombre:Grupo BGL Razsn Social: BGL INTERNATIONAL BUSINESS S.R.L. Direccisn: Castro 2873 San Martin C.P: 1650 Enviado en nombre de BGL INTERNATIONAL BUSINESS S.R.L por Grupo BGL [IMAGE]
Re: (another) Intel driver change needs testing.
On Tue, 18 May 2010 16:52:11 +0200 David Coppa dco...@gmail.com wrote: On Tue, May 18, 2010 at 4:40 PM, J.C. Roberts list-...@designtools.org wrote: Whether enabled or disabled in xorg.conf, (so far) it seems to have no effect on the occasional/intermittent screen corruption bug seen by dcoppa (915GM) and me (845G). The occasional/intermittent screen corruption bug seems to have disappeared for me with latest xenocara -current + the experimental Mesa 7.8 update. That's good news! I don't have the experimental mesa patch, but if you want me to test it, mail it to me off list. I'm still trying to track down what kernel drm change between April 19 and April 24 (inclusive) is responsible for regular xv failing to work any more with the i845G. When I did the initial test of the new intel driver, it worked on the 19th, but when I rebuilt the (GEM) kernel on the 24th, xv stopped working. jcr -- The OpenBSD Journal - http://www.undeadly.org
Re: (another) Intel driver change needs testing.
On Tue, 18 May 2010 07:40:23 -0700 J.C. Roberts list-...@designtools.org wrote: It seems to become enabled, but really doesn't do anything. (II) intel(0): DPMS enabled (==) intel(0): Intel XvMC decoder disabled copy-n-paste snafu. should read enabled -- The OpenBSD Journal - http://www.undeadly.org
Re: move klist to process
On Mon, 17 May 2010 21:08:05 -0400 (EDT) Ted Unangst ted.unan...@gmail.com wrote: This moves the knote list to struct process for better process semantics with rthreads. It should have little noticeable effect otherwise, but I've been wrong about simpler diffs before It has no noticeable effect on i386, but similarly I've been wrong about noticeable before. jcr
je je 1985 ne pe 4 mi
http://HICKS7725.makesboyscrazy.com/list.html
Re: (another) Intel driver change needs testing.
On Tue, May 18, 2010 at 11:01:29AM -0600, Ted Roby wrote: On Tue, May 18, 2010 at 8:40 AM, J.C. Roberts list-...@designtools.orgwrote: On Mon, 17 May 2010 22:16:54 +0100 Owain Ainsworth zer...@googlemail.com wrote: For those of you new here, or termnally adsent minded, instructions follow: You've got be careful with those termnally adsent folks. ;) absinthe-minded? Hard to get decent Absinthe in the UK, i'm afraid. Whether enabled or disabled in xorg.conf, (so far) it seems to have no effect on the occasional/intermittent screen corruption bug seen by dcoppa (915GM) and me (845G). I have a corruption artifact that affects my pidgin IM window, and pop-up windows like my mozilla-firefox URL history in the address bar. It did NOT exist in a April 28 snapshot build with GEM kernel and intel-current.tgz drivers. I never experienced this artifact until the GEM commit on May 10th. There were three, could you confirm which one you mean? Also, can you please retest with current, i commited a diff today that may help. -0- -- f u cn rd ths, itn tyg h myxbl cd.
Re: (another) Intel driver change needs testing.
On Tue, May 18, 2010 at 07:40:23AM -0700, J.C. Roberts wrote: On Mon, 17 May 2010 22:16:54 +0100 Owain Ainsworth zer...@googlemail.com wrote: For those of you new here, or termnally adsent minded, instructions follow: You've got be careful with those termnally adsent folks. ;) Feh, typo ;). With the 845G and a minimal xorg.conf (only NoTrapSignals) I get the expected: (II) intel(0): DPMS enabled (==) intel(0): Intel XvMC decoder disabled (II) intel(0): Set up overlay video (II) intel(0): direct rendering: DRI2 Enabled When using: Option XvMC true It seems to become enabled, but really doesn't do anything. (II) intel(0): DPMS enabled (==) intel(0): Intel XvMC decoder disabled (II) intel(0): Set up overlay video (II) intel(0): direct rendering: DRI2 Enabled All that actually does is say that the option is turned on (defaults to yes for 965+, no for the rest). an extra message from the decoding code comes up in the case where it actually does something (915+ only, i'm afraid). It is a rather stupid message, I agree. Trying to test xvmc with `mplayer -vo xvmc` just results in: Error opening/initializing the selected video_out (-vo) device. mplayer in ports doesn't support xvmc. I really ought to check what does. Then again, it may not be hard to enable. Thanks for testing (to everyone who has done so so far). -0- -- When the speaker and he to whom he is speaks do not understand, that is metaphysics. -- Voltaire
Re: (another) Intel driver change needs testing.
On Mon, 17 May 2010 22:16:54 +0100 Owain Ainsworth zer...@googlemail.com wrote: The diff found at http://xenocara.org/xvmc.diff could do with some testing. This is some stuff that I didn't backport back to 2.9.1 when I did the intial intel driver backport pile. This contains a huge cleanup of the XVMC code (and enabled it on 965+ by default, it is still there by an option on lower). Otherwise there are a few small changes, but the vast majority of it is xvmc. With this the intel driver renderer is equal to just after intel 2.11 upstream. Please test and report to myself and matthieu@, good feedback or bad, when i've got this and one more thing out of the way I plan to work on radeon dri2. For those of you new here, or termnally adsent minded, instructions follow: make sure your xenocara tree is up to date. $ cd /usr/xenocara/driver/xf86-video-intel $ patch -E /path/to/xvmc.diff $ make -f Makefile.bsd-wrapper obj $ make -f Makefile.bsd-wrapper build restart X and robert is your mother's brother. Cheers, -0- -- Abstainer, n.: A weak person who yields to the temptation of denying himself a pleasure. -- Ambrose Bierce, The Devil's Dictionary Installed latest snapshot (from May 12, judging by the kernel), updated /usr/xenocara, applied patch, built the driver. Everything seems to be okay. After I enabled XVMC in xorg.conf, it says in Xorg.log that XVMC is enabled. Checked out the latest source of mplayer and built it with --enable-xvmc, this is what it says when I try to play a video with -vo xvmc: s...@d400:0:tmp$ /usr/local/mplayer-svn/bin/mplayer -vo xvmc video.mp4 MPlayer SVN-r31179-3.3.5 (C) 2000-2010 MPlayer Team Playing video.mp4. libavformat file format detected. Compiler did not align stack variables. Libavcodec has been miscompiled and may be very slow or crash. This is not a bug in libavcodec, but in the compiler. You may try recompiling using gcc = 4.2. Do not report crashes to FFmpeg developers. [lavf] stream 0: audio (aac), -aid 0, -alang und [lavf] stream 1: video (h264), -vid 0 VIDEO: [H264] 480x360 24bpp 25.000 fps 501.0 kbps (61.2 kbyte/s) Clip info: major_brand: mp42 minor_version: 0 compatible_brands: isomavc1mp42 vo_xvmc: X-Video extension 2.2 vo_xvmc: No X-Video MotionCompensation Extension on :0.0 Error opening/initializing the selected video_out (-vo) device. == Opening audio decoder: [faad] AAC (MPEG2/4 Advanced Audio Coding) AUDIO: 44100 Hz, 2 ch, s16le, 125.6 kbit/8.90% (ratio: 15698-176400) Selected audio codec: [faad] afm: faad (FAAD AAC (MPEG-2/MPEG-4 Audio)) == [AO OSS] audio_setup: Can't open audio device /dev/sound: Device busy ao2: 44100 Hz 2 chans s16le [0x9] [AO SUN] Can't open audio device /dev/audio, Device busy - nosound. DVB card number must be between 1 and 4 AO: [null] 44100Hz 2ch s16le (2 bytes per sample) Video: no video Starting playback... Exiting... (Quit) But from your mail to J. C. Roberts I gather it's supposed to be like this. Also, after your previous request to test GEM (like when we needed to recompile a kernel with an option), from some later snapshot onwards, I started having two problems: - Screen corruption, easiest way to get it was to scroll a webpage fast in Firefox, when this screen corruption happened -- errors would appear on the console from drm With this current software combination -- I don't see this problem anymore. - Second problem is that if I put something like `bgs image.jpg` in my .xinitrc, then when I run X, the background is never changed and remains to be that default X b/w pattern and the screen/window repainting gets completely messed up: the windows are simply not drawn. I run xterm, the cursor changes to text-cursor when I hover the mouse over the area where the window should be. I can start programs from this xterm, but the window is simply invisible. Same for any other program windows. My .xinitrc: bgs .fluxbox/backgrounds/space_005.jpg exec cwm If I comment out the bgs line, and run it manually after starting X, everything's fine. Also, I think this might be connected to this issue, when I type startx, when the mode is set and the screen is filled with X's default pattern, for approximately 0.4~0.6 seconds, in the lower right edge of the screen I see a black rectangle, about 1/3 of the screen's width and of ~4 pixels height, after that it disappears and I see the X pattern in this area as I normally should have. # dmesg# OpenBSD 4.7-current (GENERIC) #643: Wed May 12 02:00:04 MDT 2010 t...@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: Intel(R) Pentium(R) M processor
Re: (another) Intel driver change needs testing.
On Tue, May 18, 2010 at 3:04 PM, Owain Ainsworth zer...@googlemail.comwrote: absinthe-minded? Hard to get decent Absinthe in the UK, i'm afraid. Sorry. There were three, could you confirm which one you mean? (II) Loading extension GLX (II) LoadModule: intel (II) Loading /usr/X11R6/lib/modules/ drivers//intel_drv.so (II) Module intel: vendor=X.Org Foundation compiled for 1.6.5, module version = 2.9.1 Module class: X.Org Video Driver ABI class: X.Org Video Driver, version 5.0 Also, can you please retest with current, i commited a diff today that may help. If you mean sys/dev/pci/drm/i915_drv.c already tested when I saw it. No change. If you mean the additions and updates to Xenocara, I was about to.
Re: (another) Intel driver change needs testing.
On Tue, May 18, 2010 at 5:25 PM, Owain Ainsworth zer...@googlemail.comwrote: I was referring to which of the three drm commits on the 10th. Oh. They were all applied. Unfortunately, I built everything late that night and early on the 11th. So, I never determined which patch caused it. I was watching with great hopefulness from the time you attached pineview to the time I saw this: Log message: unifdef INTELDRM_GEM.
Re: (another) Intel driver change needs testing.
On Tue, May 18, 2010 at 05:42:09PM -0600, Ted Roby wrote: On Tue, May 18, 2010 at 5:25 PM, Owain Ainsworth zer...@googlemail.comwrote: I was referring to which of the three drm commits on the 10th. Oh. They were all applied. Unfortunately, I built everything late that night and early on the 11th. So, I never determined which patch caused it. I was watching with great hopefulness from the time you attached pineview to the time I saw this: Log message: unifdef INTELDRM_GEM. Can you please use cvs to bisect which commit caused the problem? (it is only three commits, will not take long). It may help me fix it to know what caused it. -0- -- Life is like a bowl of soup with hairs floating on it. You have to eat it nevertheless. -- Flaubert
Hace de tu cumple una fiesta
EsteViernes 21- Sabado 22 - Domingo 23 y Lunes 24 de Mayo en la Divina San Telmo B o?= Canilla libre de cerveza +o?=Pizzas libres (5 variedades artesanales) o?= Empanadas caseras o?= Ademas te regalamos una Remera, Consumiciones sin cargo para vos y Listas Free para todos tus amigos o?= Si cumplis ao?=os en Mayoo?= Junioo?=o Julioo?= podes obtener hasta 2 cenas sin cargo o?= Solo respondiendo este mail con tu numero de telefono.o?= B Para obtener todos estos obsequios y muchos mas LLamanoso?= a los siguientes telefonoso?= 1138071413o?= oo?= 43311473o?= de 11hso?= a 19hs o?= Tambien te podes comunicar por email: barraycenali...@mail.ru o?= Atencion las consultas por email o?=son validas solo si tieneno?= tu numero de telefono para contactarte personalmente o?= Te gustariao?=NO PAGAR ALQUILER DE SALONo?= para realizar festejos de casamientos, bautismos, agasajos,eventos, graduados, despedidas, fiestas universitariaso?= y cumpleao?=os de 15? B LLamanoso?= a los siguientes telefonoso?= 1138071413o?= oo?= 43311473o?= de 11hso?= a 19hs o email: barraycenali...@mail.ru o?= PROMOCION VALIDA PARA GRAN BUENOS AIRES Y CIUDAD DE BUENOS AIRES
