Re: Problem with resolution
On Mon, 2008-11-24 at 18:39 +0100, A.Madesani wrote: On Fri, Nov 21, 2008 at 08:03:02PM +, Ferris McCormick wrote: You can change it at the OBP 'ok' prompt; I don't recall the exact command off hand but someone here will know it. The frame buffer handbook shows what resolutions the afb supports --- search for 816-3582-11 at http://docs.sun.com and download the latest. Hi. The problem isn't in OBP setting, but in the fb's setting in linux. mode 1152x900 geometry 1152 900 1152 900 32 timings 0 0 0 0 0 0 0 accel true rgba 8/0,8/8,8/16,0/0 endmode Is that fron your xorg.conf file? Last I knew, the sunffb driver (for ffb and afb frame buffers) ignored mode lines completely. I *think* the only way to set resolution is from OBP. This is the actual setting and evidently isn't correct. I read the man of frame buffer, but i've not understood the sintassi to change fb modes. When i was using an intel processor lilo, I used VGA=ask, but I don't know if I may use the same solution to find the right resolution with a blade, silo and elite card. This is a piece of my silo.conf: ## read-only timeout=10 image=1/vmlinuz.new label=Linux.new initrd=/boot/initrd.img.new root=/dev/sdb5 Is it possible to insert VGA=ask to check possibilities to change resolution? Where it's necessary to put VGA=ask to obtain what I'm finding? I really don't know. Sorry. TIA Antonio -- #+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+# Antonio Madesani [EMAIL PROTECTED] [EMAIL PROTECTED] #+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+# Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Sparc, Userrel, Trustees) signature.asc Description: This is a digitally signed message part
Re: problem with keyboard and mouse
On Fri, 2008-11-21 at 14:42 +0100, A.Madesani wrote: On Fri, Nov 21, 2008 at 02:18:43PM +0100, Frans Pop wrote: A.Madesani wrote: Section InputDevice Identifier Generic Keyboard Driver keyboard Does it help if you replace keyboard by kbd? Hi Frans.Thanks for your message. Doesn't change. Remember that the mouse too doesn't work. The error messag I receive is: ~ cat /var/log/Xorg.0.log | grep WW (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (WW) The directory /usr/lib/X11/fonts/misc does not exist. (WW) The directory /usr/lib/X11/fonts/cyrillic does not exist. (WW) The directory /usr/lib/X11/fonts/100dpi/ does not exist. (WW) The directory /usr/lib/X11/fonts/75dpi/ does not exist. (WW) The directory /usr/lib/X11/fonts/Type1 does not exist. (WW) The directory /usr/lib/X11/fonts/CID does not exist. (WW) The directory /usr/lib/X11/fonts/100dpi does not exist. (WW) The directory /usr/lib/X11/fonts/75dpi does not exist. (WW) Warning, couldn't open module type1 (WW) Warning, couldn't open module kbd (WW) Warning, couldn't open module mouse (WW) /dev/fb0: Forcing no acceleration on Elite3D M3/M6 My experiences with Xorg and Elite framebuffer suggest that if you don't load the microcode for the Elite device, Xorg almost immediately terminates with a Signal 11 (SIGSEGV). I don't know if Debian makes the microcode available or if you have to get it yourself directly, but I am sure Debian provides the microcode loader. However, your problem indicates that you can't load type1, kbd, or mouse modules, all of which should be available. Check the log to see if you loaded any modules please (well, you must have loaded sunffb at least). All of those should be in whatever directory the xorg modules are in. On my system, for example, I see /usr/lib/xorg/modules/input/mouse_drv.so (and /usr/lib/xorg/modules/fonts/libtype1.so and so on). But I am running Gentoo, not Debian, so your modules might be elsewhere, or you might have an incomplete install. I hope this help someone :) Cheers Antonio -- #+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+# Antonio Madesani [EMAIL PROTECTED] [EMAIL PROTECTED] #+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+# Hope this is of some use, Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Sparc, Userrel, Trustees) signature.asc Description: This is a digitally signed message part
Re: problem with keyboard and mouse
On Fri, 2008-11-21 at 19:21 +0100, A.Madesani wrote: On Fri, Nov 21, 2008 at 03:13:14PM +, Ferris McCormick wrote: (WW) Warning, couldn't open module type1 (WW) Warning, couldn't open module kbd (WW) Warning, couldn't open module mouse (WW) /dev/fb0: Forcing no acceleration on Elite3D M3/M6 My experiences with Xorg and Elite framebuffer suggest that if you don't load the microcode for the Elite device, Xorg almost immediately terminates with a Signal 11 (SIGSEGV). I don't know if Debian makes the microcode available or if you have to get it yourself directly, but I am sure Debian provides the microcode loader. However, your problem indicates that you can't load type1, kbd, or mouse modules, all of which should be available. Check the log to see if you loaded any modules please (well, you must have loaded sunffb at least). All of those should be in whatever directory the xorg modules are in. On my system, for example, I see /usr/lib/xorg/modules/input/mouse_drv.so (and /usr/lib/xorg/modules/fonts/libtype1.so and so on). Hi. Thank for your mail. I was already using the the Elite microcode. I found the problem: usually I don't use X so when I changed from XFree to xorg I forgot to install xserver-xorg-input-kbd and xserver-xorg-input-mouse . Now it works correcltly. Remain the problem to change the framebuffer setting because the image is a bit taller than my monitor. I hope to find information to change this by google ;) Cheers Antonio You can change it at the OBP 'ok' prompt; I don't recall the exact command off hand but someone here will know it. The frame buffer handbook shows what resolutions the afb supports --- search for 816-3582-11 at http://docs.sun.com and download the latest. -- #+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+# Antonio Madesani [EMAIL PROTECTED] [EMAIL PROTECTED] #+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+#+--+# -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Sparc, Userrel, Trustees) signature.asc Description: This is a digitally signed message part
Re: Fans at full speed on blade 1000 with kernel 2.6
On Wed, 2006-09-06 at 18:20 +0200, Luis Martinez wrote: Hello every one, Is there any kernel module for to control the temperature sensors and thus the fans for the blade 1000. Regards, Luis CONFIG_BBC_I2C=m in your .config file builds the bbc module, and it is supposed to do what you are asking. But last time I tried it, it had a horrible side effect: it took over the system completely for about 2 seconds out of every 5. That was with a 2.6.15 kernel, though, so it might do better with something more current today. If you build the bbc module, of course you can play with it and see how it does today. Hope this helps, Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Devrel, Sparc) signature.asc Description: This is a digitally signed message part
Re: Soprot for expert 3d on xorg 7?
On Wed, 2006-09-06 at 18:22 +0200, Luis Martinez wrote: Hello again, Does anybody know if there is or there would be any support for Expert3d cards (PCI). Regards, Luis Almost certainly not. I don't think any sparc developers have specifications for it; there is no kernel support; a driver would be a major effort with small return. Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Devrel, Sparc) signature.asc Description: This is a digitally signed message part
Re: ultra2 ... stuff
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Mon, 17 Jul 2006, Kevin Diggs wrote: Hi, I finally stumbled around and got X working on this thing. I had to use sunffb as the device, set the mouse device to /dev/sunmouse, and change the protocol to MouseSystems. How do I get the resolution up to 1280x1024? Got a modeline, anyone? It turns out you have to do it through OBP commands; I don't recall the exact format, but you can force a resolution on the output device for the ffb. (There is a frame buffer handbook on the sun web site, and there is a Creator manual. Between them, they will give you what ffb is capable of. You have to match your monitor to something ffb can do, and go from there in my experience.) Mode lines will not help. sunffb ignores them, and uses whatever ffb is using. Sorry I don't have more information at hand, but I hope this helps some. Any pointers on how to upgrade the kernel to 2.6 would be appreciated as well. In my experience, on U2 very carefully. It's not quite stable. kevin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] Sorry I can't give more details. I don't have the information available where I am right now. Regards, Ferris - -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Devrel, Sparc) -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.4-ecc0.1.6 (GNU/Linux) iD8DBQFEvALDQa6M3+I///cRAjF/AKCdeUVL0UO1CaqPqwqQ92g6+V2mTQCfXYZ9 tpkEy+nzWPblvp5hpch9Pgk= =hgxo -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: sparc64 2.6 SMP RED state exception
On Fri, 2006-04-07 at 13:03 +0200, BERTRAND Joël wrote: Blars Blarson a écrit : I've submitted bug 361214 about the frequent sparc SMP crashes with 2.6 kernels. Is anyone running a sparc 2.6 SMP kernel and NOT having the system crash under heavy disk I/O? Please let the bug know if you have any additional information. I use a sparc64 SMP kernel on my U60 (and on my U420) without any trouble. Kernel is official kernel (not from debian tree): Linux bohr 2.6.15.6 #1 SMP PREEMPT Fri Mar 10 23:27:31 CET 2006 sparc64 GNU/Linux I don't have any trouble with sparc64 architecture, only with 2.6 and sparc32 :-( Maybe this bug come from the same location than the bug I have submitted in bugzilla.kernel.org (#6344)... DMA mistake that only occurs with heavy disk I/O... Regards, JKB We have found that some systems are fine with later 2.6 kernels (say, in the 2.6.15 series) and some are not. sparc32 systems are a problem; U1/U2 systems routinely crash under heavy disk I/O, U60 now seem fine, SB1000 is fine, I think U5/U10 are OK, and so on. Regards, -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Devrel, Sparc) signature.asc Description: This is a digitally signed message part
Re: Weird things regarding Xfree and Creator 3d (or why don't Imanage to get Xvideo and DRI working)
On Tue, 2005-07-19 at 11:44 +0200, Christian Tusche wrote: DRI has been disabled in current xfree86 for creator3d, IIRC. That's correct. In order to support XAA and the modern font rendering stuff correctly, DRI support had to get turned off on sunffb. I had the same problem with a Creator3D in an Ultra60. Do you know if there is any way to turn on the DRI support again? I was looking through the changelog of the xfree package but found nothing about changes in DRI, and the documentaiton on the Xfree86 site stated it works (is this a debian specific change?). So, is there an option to the sunffb module or some other trick to enable it? The necesary dri driver ffb_dri.so and the kernel modul is installed, at least. Not really. It's disabled in the source for sunffb by #if 0 DRI stuff #endif So, the trick would be to read the source and revert to the original conditionals. You'ld be on your own at that point, though. Perhaps easier would be just to use an older version of sunffb; except for the changes for XAA, it hardly ever (i.e. never) changes otherwise. There also is a second problem with sunffb: When moving a window on the desktop, sometimes it gets overlayed by a color pattern until the window is redrawn. I can't say if this related to acceleration of the sunffb driver. Is there a way to turn off acceleration for testing? -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Sparc, Devrel) signature.asc Description: This is a digitally signed message part
Re: Bug#287249: xserver-xfree86: Crashes on SPARC with Elite3D/m6
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, 26 Mar 2005, David S. Miller wrote: On Sat, 26 Mar 2005 20:32:44 +0100 T.Amm [EMAIL PROTECTED] wrote: Any other install kernel just gives me a MMU-Fast Instruction Miss. So I am actually damned with Solaris - which is not *so* bad - but what's life without Linux? Use the latest Sarge testing boot images, I'm happily running Debian on my 1.5GHZ SB1500 Back to the bug report. The X server already makes a check, but there was some talk the other week about this check crashing Creator3D chips (since we poke a register that only exists on the Elite3D card to check this) so it's possible someone patched the test out or changed it. No. That check in sunffb is not changed. Apparently a reboot cured the original problem; I have heard nothing further on it. I was prepared to modify the check for the reporter and for the reporter only: after all, sunffb has been running fine for years, and this looked like a one-user problem. It seems it was a no-user problem, and sunffb is as it always has been. I haven't looked at the X server sources in a year or so unfortunately, so I'm pretty out of touch with the current state of the tree. And with a million and a half networking and sparc64 kernel things to do, I doubt I'll get back to it any time soon. Only recent changes to sunffb are yours for XAA friends. By the way, you get that abort from xfree/xorg and sunffb and Elite when, as here, you do not have the microcode loaded. If you install and load the afb microcode, that abort should go away. (As, I believe, the bug reporter indicates.) Someday I might even look at why this is the case. Hope this is close to on topic, Regards, Ferris - -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (sparc) -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFCRexUQa6M3+I///cRAinxAJ0ZO1s0iPZWoexR04MU/ZM+XPlxPQCfZSs9 y2r4nznr+nrLgj2UdRCdLKs= =V0en -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sparcbook 3GX: current state of support?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, 10 Nov 2004, Roger Burton West wrote: I've recently acquired a Sparcbook 3GX and I'd like to get it up and running Debian. I've found the message http://lists.debian.org/debian-sparc/2000/03/msg00098.html but can't see anything more recent. Is X support a plausible thing to try for, and should I be going for stable or unstable installers? Roger For what it's worth, 15 months ago or so I was able to install Gentoo Linux onto a Sparcbook 3GX, using a kernel at about revision 2.4.20. X would work using the CG3 driver (if I recall correctly) but I could never figure out how to configure in the pointer (mouse). -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] - -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Sparc) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) iD8DBQFBklz8Qa6M3+I///cRAi1AAJ96WxWerY4cddkNC2adoUOq7y6e9QCfWdBv Z/o1/iQsIM7ts8jH6nWdeVA= =YbWO -END PGP SIGNATURE-
Re: XFree86 X server: does loading the dri module still hose sunffb users?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, 26 May 2004, Branden Robinson wrote: There are some issues that may be worth mentioning: local OpenGL clients segfault (though remote ones are fine, so that suggests a mesalib problem) Please file a bug report against xlibmesa-gl. Installation of the xlibmesa-{gl,glu,dri}-dbg packages, and a backtrace from a segfaulting client, would be *very* helpful. It's a combination of mesa xfree problems which are now fixed in latest sources. For an overlong discussion, look at http://bugs.gentoo.org/show_bug.cgi?id=19776 Regards, - -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Developer, Gentoo Linux (Sparc) -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAtNnXQa6M3+I///cRAlYaAKCrQSpEaoakoFPx+OeIp7LPo8P2cwCaAp3c xLun9JIX+80T9Im+mmC8Nf0= =qGIH -END PGP SIGNATURE-
Re: [debian-sparc] ultra 1
On Tue, 30 Sep 2003, Erwann Abalea wrote: Hi David, On Tue, 30 Sep 2003, David S. Miller wrote: On Tue, 30 Sep 2003 16:23:06 +0200 (CEST) Erwann Abalea [EMAIL PROTECTED] wrote: Can it be higher? No. Sunffb will give you an error message and quit if you try 32. -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: Problem getting X Server to tun on Ultra Sparc 10 with Debian 3.0.r1
On Tue, 13 May 2003, kevin curran wrote: Folks I have been trying unsuccessfully for about three weeks (on and off) to get the graphical User interface to work on my Sun UltraSparc 10 box. I presume it has an Elite 3D PCI card inside as the sticker is on the front of the box. My monitor is a Sun 19inch with a CM751U label on the back. Try /bin/dmesg | grep Elite that will tell you if you have an Elite3D video card. Also, the /var/log/XFree86.0.log should have information from SUNFFB on what it has found on your system. I have been running xf86config for weeks now trying all sorts of configurations. I don't really know what my horizontal and vertical refresh rates are for the monitor that I have. I also don't seem to see Elite 3D or CM751U listed in the video card and monitor lists that xf86config presents me with. I can enclose the full output if people like from the server log. The last lines however mention the following: (EE) SUNFFB: Failed to load module 'dri' ... (II) unloading /usr/x11r6/lib/modules (II) unloading /usr/x11r6/lib/modules ... What version of XF86? If it is 4.2.xxx, comment out 'Load dri' from the Modules section in your XF86Config file, if you have it. (This is changed with 4.3.0). Fatal server error No Screens found I guess that the last configuration that I choose must have been for the subffb card. I can't remember. I have tried so many. SUNFFB is the driver for Creator/Elite graphics cards. Once it is loaded, it will figure out if it is talking to a Creator3D or to an Elite3D. I have read elsewhere that I should try running the xserver-mach64. I don't actually know how to do that. It is a command line instruction? It says also after that line in the debian online sparc problems page that 'one should.. use this X server for ATI Rage framebuffers, such as found in many PCI-based UltraSPARC machines'. My question here is 'DOes the Ultrasparc 10 have an ATI rage framebuffer'? Some do. The messages at boot time will tell you what is on yours. In fact, I have an Ultra10 with both a mach64 and with a creator3d. You need your monitor plugged into the one the XF86Config file wants to use. Quick check: If your monitor is plugged into the system using a PC-like HD15 connector, you are not connected to a Creator/Elite card. If you are plugged in with a Sun 13W3 connector, It's a Sun graphics card of some sort. (HD15 = 3 rows of 5 pins; 13W3 looks like nothing else on your system). Everything else I can get working. I installed all the network settings manually but the X server has totally baffled. I have installed about 7 different versions of linux on over 20 PC's in the last 18 months and I have always managed to somehow get the X server to work but my ultrasparc has beaten me. Any help appreciated. It can be done, but the documentation doesn't make it easy. Regards, -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: XFree on Ultra10
On Thu, 1 May 2003, Marc Dilasser wrote: Hello, I'm running Debian Woody on an Ultra10 with a Creator3D card. XFree seems slow and freeze times to times during 20-30 secondes, I don't understand why. Is the afbinit /dev/fb0 /usr/lib/afb.ucode required ? Marc, No, that's for an Elite3D. If you try it with a Creator, it will leave you in a very bad state (ask me how I know). It seems to take one or more complete power on/off cycles to recover. Unfortunately, I can't say anything useful about your real problem, but I am almost certain that trying to load the AFB microcode into an FFB device will do nothing good for you. Marc Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: Problems with X (Try taking out 'Load dri')
) Option XkbSymbols sun/us(sun5) EndSection Section InputDevice Identifier Mouse0 Driver mouse Option Protocol MouseSystems Option Device /dev/sunmouse EndSection Section Monitor Identifier Monitor0 VendorName Sony ModelNameGDM-20E20 Option DPMS on EndSection Section Device Identifier Card0 BoardName Sun|FFB2+ Horizontal Creator 3D (ffb) VendorName Sun Microsystems Driver sunffb EndSection Section Screen Identifier Screen0 Device Card0 Monitor Monitor0 DefaultDepth 24 SubSection Display Depth 1 EndSubSection SubSection Display Depth 4 EndSubSection SubSection Display Depth 8 EndSubSection SubSection Display Depth 15 EndSubSection SubSection Display Depth 16 EndSubSection SubSection Display Depth 24 EndSubSection SubSection Display Depth 32 EndSubSection EndSection Section DRI Group video Mode 0666 EndSection === Hope this helps, Regards -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: Creator3D XF86Config-4
I can send you one tomorrow. I use it on an Ultra2/creator3d system. It's not a flat panel monitor, though, it's Sony GDM20E20, I don't recall Sun's part number. Regards, -- Ferris E. McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (h) 703-368-6723 (w) 703-392-0303 (fax) 703-392-0401 On Thu, 7 Nov 2002, AUSTIN MURPHY wrote: If anyone has a working XF86Config-4 for an Ultra with a Creator3D, I would love to see it. My Sun Flat panel displays a few glitches with the creator which i think are slightly out of range sync signals. The glitch is one or two lines is offset to the left by one pixel. Any hints on tweaking the creator woudl be great too. Thanks, Austin Murphy -- Systems Programmer Rutgers University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: audio issue
According to the service manual, the U10 has 2 EIA standard 3.5mm output jacks and 2 input jacks, with the cable tips assigned to left channel and the ring assigned to right channel. The manual labels them 'headphone, line-out, line-in, and microphone,' but on my U10, they are not marked, and I've never tried them. The electrical specifications are shown on pages C-24 -- C-26 of the Ultra10 Service Manual (Part # 805-7764-12); this manual is available in pdf format for free download on Sun's web site. Unfortunately, I do not remember where on the site it is, but it is there (and, if you have an Ultra10, it is well worth having). With a little work it can be found. No, I don't know why Sun's web site is so hard to find anything in. That's not much help, I'm afraid, but maybe it is of some use. Regards, -- Ferris E. McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (h) 703-368-6723 (w) 703-392-0303 (fax) 703-392-0401 On 20 Aug 2002, Hakan Kuecuekyilmaz wrote: Hello, I am running an Ultra10 debian/unstable with custom 2.4.19 kernel. Now I am trying to record some audio on my box, playing audio is no problem. Since I have no handbook of my Ultra10 I do not know exactly which socket is for what. I guess the sockets are: (view from the back) [] [] for recording? [] [] for playback, which works.. I installed some audio programs, when I try to record sound from line in some programs are crashing, some hanging and other do not record anything. Does anybody knows which socket is for what? Is it possible to record audio with Ultra10? PS: Does anybody run mplayer on an Ultra10 (I have a big problem with sound on mplayer, too..) regards -- Hakan Kuecuekyilmaz, University of Applied Sciences Esslingen, Germany [EMAIL PROTECTED] | [EMAIL PROTECTED] | -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Which Sparc is best?
On Tue, 23 Jul 2002, Craig Morehouse wrote: On Tue, 23 Jul 2002 22:54:48 -0400, Ben Collins wrote: On Tue, Jul 23, 2002 at 06:46:56PM -0400, Craig Morehouse wrote: I'm new to Debian, but am going to be buying 30 workstations for a new operation, and I'd like to use Sun hardware running 3.0 Woody. Question, which Sun box has proven to be REALLY good and solid with Debian? Are the Ultra 2's better than Ultra 5's or 10's, or vice versa? Should I build a bunch os SS20's with 4 cpus, or should I get the latest Blade? For the most tested and stable machines, go with an UltraSPARC. I'd suggest one of two types of systems. 1) The early IDE line. E.g. Ultra 5 or 10. The reason being that the machines are cheap to maintain and offer decent performance. 2) PCI SCSI based systems. E.g. Ultra 2, 30, 60. If you're interested in SMP based systems, the 2 and 60 are the way to go. Debian's primary archive (aka ftp-master, aka auric) is an Ultra60 with dual 450mhz cpu's, 1.5 gigs of ram and a ~240gig raid5 For what it's worth, my experience with Linux (Debian SuSE) on an Ultra10 suggests that if you are going to stress the disk much at all, on such a system, you will want to use one of your expansion slots for a SCSI card. Others will have better information, though. Regards, Ferris Thanks, Ben. We had some early Blades last year, and I really thought that a dual CPU Ultra2 was a better box. Does Debian install nicely on the Ultra 5's and 10's? As you said, there are plenty of 10's around at nice prices. Another question... Are there any Package disadvantages in using a Sparc over an Intel box using Debian? Is the same range of Debian software available for all CPU's? Thanks, Craig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Which Sparc is best? (SCSI card)
On Wed, 24 Jul 2002, Craig Morehouse wrote: On Wed, 24 Jul 2002 08:03:39 -0400, Ferris McCormick wrote: On Tue, 23 Jul 2002, Craig Morehouse wrote: On Tue, 23 Jul 2002 22:54:48 -0400, Ben Collins wrote: ... 1) The early IDE line. E.g. Ultra 5 or 10. The reason being that the machines are cheap to maintain and offer decent performance. 2) PCI SCSI based systems. E.g. Ultra 2, 30, 60. If you're interested in SMP based systems, the 2 and 60 are the way to go. Debian's primary archive (aka ftp-master, aka auric) is an Ultra60 with dual 450mhz cpu's, 1.5 gigs of ram and a ~240gig raid5 For what it's worth, my experience with Linux (Debian SuSE) on an Ultra10 suggests that if you are going to stress the disk much at all, on such a system, you will want to use one of your expansion slots for a SCSI card. Others will have better information, though. Regards, Ferris This makes much sense. I use SCSI on all the important Intel boxes. One question, though, which SCSI cards work best on UltraSparcs and Debian? Thanks. I have an Ultra10 with a card the system identifies as: 02:04.0 SCSI storage controller: Symbios Logic Inc. (formerly NCR) 53c875 (rev 04) Flags: medium devsel, IRQ 6646528 I/O ports at 2000400 Memory at f9ff2000 (32-bit, non-prefetchable) Memory at f9ff4000 (32-bit, non-prefetchable) Expansion ROM at 0004 [disabled] with external SCSI-III disk(s). It uses the sym53c8xx.o scsi driver, which David S. Miller called one of the best scsi drivers in the tree in response to a similar question last April on the sparclinux@vger.kernel.org mailing list. It's worked fine for me for about 9 months now. In the interest of full disclosure: Debian installs fine for me over this driver, but for various compatibility reasons, my primary OS on this system is SuSE 7.3. I don't know why that should matter at the driver level, though, since that should be independent from whose name is on the release(?) Regards, -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [Xpert]Re: Screen blanking/DPMS with Sun Ultra1
On Thu, 20 Dec 2001, Moritz Bunkus wrote: Hi. Thanks for all your work. I will try it after christmas. Right now I'm struggling to find the sources for xserver-xfree86... I'm using woody and I CAN'T find anything!? I'm not that used to debian yet. Where should I look? -- == Ciao, Mosu I went to the www.xfree86.org site and downloaded the current CVS development tree as described at www.xfree86.org/cvs/ in the Accessing Anoncvs via cvs pserver... section. All this takes is a lot of time and disk space. For the build, I just did what the INSTALL-X.org instructions suggested: Edit a configuration file (config/cf/site.def) to define where I wanted the test version installed, and 'Make World'. Although XF86 is very large, the only things I changed are at xc/programs/Xserver/hw/xfree86/drivers/sunffb/. I didn't use Debian for this, so I don't know exactly where Debian's source for XFree86 is. But, it should be one of the largest packages on their source disks, and it should have a name that looks right. I am sending a copy of this reply to the Debian list, so someone who knows can tell you for sure what you already have, or perhaps what might work best for your configuration. Until later, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: [Xpert]Re: Screen blanking/DPMS with Sun Ultra1
(More comments below -- FEM) On Mon, 3 Dec 2001, Moritz Bunkus wrote: Hello everyone. I have a lot of Sun Ultra1 boxes with Sun monitors. Until recently they ran under Solaris. With this setup they monitors went into some sort of power saving mode after some time of inactivity. All those boxes now run Woody with XFree 4.1.0.1. Unfortunately either the graphics hardware (the suncg6 driver is used) or the monitor don't support DPMS (option DPMS is given in the monitor section): the logfile says (WW) SUNCG6(0): Option DPMS is not used (WW) SUNCG6(0): Option StandByTime is not used (WW) SUNCG6(0): Option SuspendTime is not used (WW) SUNCG6(0): Option OffTime is not used Now the boxes all have their monitors on all the time which is definitely not what I want. Is there a way to use power saving with this setup? Or at least some sort of screen blanking and/or a screensaver? Moritz, I don't know if you ever got this resolved or not, nor do I know if you have been following the follow-on discussions or not (helpful to me, not so directly to the CG6 case.) In any case, here is an idea which I have no way of testing, but which parallels David Miller's suggestion regarding the Creator graphics card. The kernel code seems to blank/unblank a display device connected to a CG6 frame buffer seems to use the mask CG6_THC_MISC_VIDEO in the module [cgsixfb.c at cg6_blank] thus: tmp = sbus_readl(fb-s.cg6.thc-thc_misc); tmp = ~CG6_THC_MISC_VIDEO; wbus_writel(tmp, fb-s.cg6.thc-thc_misc); This suggests that changing the XF86 cg6 driver module --- cg6_driver.c --- as follows should get you screen blanking at least: The current CG6SaveScreen procedure is an empty procedure returning TRUE. Something along the following lines might do what you want-- static Bool CG6SaveScreen(ScreenPtr pScreen, int mode) { ScrnInfoPtr pScrn = xf86Screens[pScreen-myNum]; /* Like elsewhere*/ Cg6Ptr pCg6; unsigned int tmp; pCg6 = GET_CG6_FROM_SCRN(pScrn); tmp = pCg6 - thc - thc_misc; switch(mode) { case SCREEN_SAVER_ON: case SCREEN_SAVER_CYCLE: tmp = ~CG6_THC_MISC_VIDEO; break; case SCREEN_SAVER_OFF: case SCREEN_SAVER_FORCER: tmp |= CG6_THC_MISC_VIDEO; break; default: return FALSE; } pCg6 - thc - thc_misc = tmp; return TRUE; } If you try that and it works, there are some other THC_MISC masks which might get the display device to turn off. As I mentioned a couple weeks ago, something along these lines will turn it off: Solaris knows how to do it. Sorry to be so long-winded. Hope this is of some use to you (or that someone who actually knows will correct it into something useful). Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: [Xpert]Re: DPMS, Screen blanking with Sun Creator3D (sun4u systems)
On Tue, 11 Dec 2001, Ferris McCormick wrote: On Fri, 7 Dec 2001, Antonio Prioglio wrote: On Thu, 6 Dec 2001, Ferris McCormick wrote: [...] I'd be getting a Creator 2D soon. I'll be happy to test it with a GDM-17E10. Saluti, Antonio Prioglio [EMAIL PROTECTED] Thanks for the interest. I'll send out a patch file in the next day or so. It modifies three source files: ffb.h ffb_driver.c ffb_dac.c and applies to the CVS version 4.1.99.1, of about November 1. However the three files involved are dated 2001/05/04, and I haven't seen anything else posted which would indicate conflicting changes to the driver. As indicated, attached source patch file puts some DPMS support into the Sun Creator/Elite driver for XF86, release 4.1.99(.1?)-CVS. The driver is in the Xfree source tree at (if I type all this right) xc/programs/Xserver/hw/xfree86/drivers/sunffb In that directory, the command patch -b sunffb-4.1.99-DPMSMode.patch should successfully patch the files ffb.h, ffb_driver.c, ffb_dac.c; after that just 'make' in that directory should get you a new sunffb_drv.o (assuming you have already built everything once.) Now, with X inactive, go to wherever your existing drivers are (something like /usr/X11R6/lib/modules/drivers depending on your configuration), save the existing sunffb_drv.o, and bring in the new one. To play with it, make sure that the already-working-with-your- Creator-card XF86Config file has the line Option DPMS on in its Monitor section, and restart X. xset q should now report that you can use DPMS, and you can play with it. Also, if you want, xset s on; xset s blank should now work. (If DPMS Option is not set, the driver will not enable DPMS, but you can change that behavior, of course.) Also, keep in mind Jon Leonard's note I sent you. I have no way of knowing if his remarks apply to your configuration or not. Good luck, Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401 --- ffb.h.orig Thu Dec 6 23:39:31 2001 +++ ffb.h Thu Dec 6 23:53:59 2001 @@ -43,6 +43,14 @@ #include xf86drm.h #include ffb_drishare.h #endif +/* + Bring in the dpms definitions. +*/ +#ifndef DPMS_SERVER +#define DPMS_SERVER +#endif /* DPMS_SERVER */ +#include extensions/dpms.h + /* Various offsets in virtual (ie. mmap()) spaces Linux and Solaris support. */ /* Note: do not mmap FFB_DFB8R_VOFF and following mappings using one mmap together @@ -233,7 +241,8 @@ extern void FFBDacFini(FFBPtr); extern void FFBDacEnterVT(FFBPtr); extern void FFBDacLeaveVT(FFBPtr); - +extern Bool FFBDacSaveScreen(FFBPtr, int); +extern void FFBDacDPMSMode(FFBPtr, int, int); /* Exported WID layer routines. */ extern void FFBWidPoolInit(FFBPtr); --- ffb_driver.c.orig Sat Nov 17 18:37:18 2001 +++ ffb_driver.cTue Dec 11 22:26:18 2001 @@ -60,7 +60,9 @@ static voidFFBFreeScreen(int scrnIndex, int flags); static int FFBValidMode(int scrnIndex, DisplayModePtr mode, Bool verbose, int flags); +/* DPMS Mode hook */ +static void FFBDPMSMode(ScrnInfoPtr pScrn, int DPMSMode, int flags); /* ffb_dga.c */ extern void FFB_InitDGA(ScreenPtr pScreen); @@ -94,13 +97,15 @@ typedef enum { OPTION_SW_CURSOR, OPTION_HW_CURSOR, -OPTION_NOACCEL +OPTION_NOACCEL, +OPTION_DPMS /* Establish a hook for DPMS Mode */ } FFBOpts; static const OptionInfoRec FFBOptions[] = { { OPTION_SW_CURSOR,SWcursor, OPTV_BOOLEAN, {0}, FALSE }, { OPTION_HW_CURSOR,HWcursor, OPTV_BOOLEAN, {0}, FALSE }, { OPTION_NOACCEL, NoAccel, OPTV_BOOLEAN, {0}, FALSE }, +{ OPTION_DPMS, DPMS, OPTV_BOOLEAN, {0}, FALSE }, { -1, NULL, OPTV_NONE, {0}, FALSE } }; @@ -937,6 +942,15 @@ pScreen-CloseScreen = FFBCloseScreen; pScreen-SaveScreen = FFBSaveScreen; +/* + Check finally for DPMS Mode. If so, set up a hook for it. If not, + keep dpms disabled. +*/ + +if (xf86ReturnOptValBool(pFfb-Options, OPTION_DPMS, FALSE)) { + (void) xf86DPMSInit(pScreen, FFBDPMSMode, 0); +} + /* Report any unused options (only for the first generation) */ if (serverGeneration == 1) { xf86ShowUnusedOptions(pScrn-scrnIndex, pScrn-options); @@ -1085,11 +1099,14 @@ /* Mandatory */ static Bool FFBSaveScreen(ScreenPtr pScreen, int mode) -/* this function should blank the screen when unblank is FALSE and - unblank it when unblank is TRUE -- it doesn't actually seem to be - used for much though */ +/* This function blanks the screen when mode=SCREEN_SAVER_ON and + unblanks it when mode=SCREEN_SAVER_OFF. It is used internally in the + FFBScreenInit code `for aesthetic reasons,' and it is used for + blanking if you set xset s on s blank. The work
Re: DPMS, Screen blanking with Sun Creator3D (sun4u systems)
On Fri, 7 Dec 2001, Antonio Prioglio wrote: On Thu, 6 Dec 2001, Ferris McCormick wrote: [..] else to complain about. The short answer is that with David Miller's hint, I have been able to put both screen blanking (xset s on s blank) and DPMS support (xset dpms p q r) into the sunffb_drv.o driver for the Creator3D, FFB2+/Vertical. I'll keep playing with it, and if it seems to keep on working, I'll make it available to anyone who wants it. I hesitate to suggest something like this as a change for the driver in general because this driver has to handle various Creator and Elite cards, and I am able to play with exactly one possible configuration, which is Creator3D, FFB2+/Vertical, GDM-20E20 monitor. I'd be getting a Creator 2D soon. I'll be happy to test it with a GDM-17E10. Saluti, Antonio Prioglio [EMAIL PROTECTED] Thanks for the interest. I'll send out a patch file in the next day or so. It modifies three source files: ffb.h ffb_driver.c ffb_dac.c and applies to the CVS version 4.1.99.1, of about November 1. However the three files involved are dated 2001/05/04, and I haven't seen anything else posted which would indicate conflicting changes to the driver. Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401
Re: DPMS, Screen blanking with Sun Creator3D (sun4u systems)
On Mon, 3 Dec 2001, David S. Miller wrote: (... snip ... Lots of stuff) You won't find creator headers in anything Sun publishes. What doesn't work is any sort of DPMS power management, even though Solaris knows how (so it is possible). It just kills the VSYNC signal, we do that in the Creator3D kernel FB driver. It's DAC fiddling: tmp = read_dac(0x6000); tmp = ~0x1; write_dac(0x6000, tmp); Just set that bit when you want it back. I'm pretty sure I even document these registers in the ffb_dac.h header file in the xfree86 sources. David, Yes, it is documented in ffb_dac.h, thanks a lot. That said, this solves my problem, and I'll start looking for something else to complain about. The short answer is that with David Miller's hint, I have been able to put both screen blanking (xset s on s blank) and DPMS support (xset dpms p q r) into the sunffb_drv.o driver for the Creator3D, FFB2+/Vertical. I'll keep playing with it, and if it seems to keep on working, I'll make it available to anyone who wants it. I hesitate to suggest something like this as a change for the driver in general because this driver has to handle various Creator and Elite cards, and I am able to play with exactly one possible configuration, which is Creator3D, FFB2+/Vertical, GDM-20E20 monitor. * For those interested and for anyone who can set me right if what I am trying is a bad thing, here is how I interpreted Miller's advice: 1. In the driver, ffb_driver.c, add DPMS to the option list, and if it there, do something like xf86DPMSInit(pScreen, FFBDPMSMode, 0); /* in FFBScreenInit */ where you have a signature static void FFBDPMSMode(ScrnInfoPtr pScrn, int M, int f) FFBDPMSMode just does FFBDacDPMSMode(GET_FFB_FROM_SCRN(pScrn), M, f); 2. In ffb_dac.c, you add something like void FFBDPmsMode(FFBPtr pFfb, int DPMSMode, int flags) { int tmp; ffb_dacPtr dac = pFfb - dac; tmp = DACCFG_READ(dac, FFBDAC_CFG_TGEN); /* Get timing reg. */ And, say, if you want to go into suspend mode, what finally seems to work for me looks like: case DPMSModeSuspend: tmp |= FFBDAC_CFG_TGEN_VSD; /* VSYNC Disable */ tmp = ~FFBDAC_CFG_TGEN_HSD; /* HSYNC Enable */ tmp = ~FFBDAC_CFG_TGEN_VIDE; /* Kill the video */ break; (Turning the monitor on looks like:) case DPMSModeOn: tmp = ~(FFBDAC_CFG_TGEN_VSD | FFBDAC_CFG_TGEN_HSD); /* Un-disable Vsync, Hsync */ tmp |= FFBDAC_CFG_TGEN_VIDE; /* Turn video on */ break; And, finally, change the mode with DACFFG_WRITE(dac, FFBDAC_CFG_TGEN, tmp); In fact, what I sketched out above is about 1/3 the total changes I have needed so far. As I said, I can exercise this in only one specific configuration, but it's my interpretation of Miller's suggestion. Regards, Ferris -- Ferris McCormick (P44646, MI) [EMAIL PROTECTED] Phone: (703) 392-0303 Fax: (703) 392-0401