armel/armhf arch qualification for buster: call for DSA, Security, toolchain concernsj
Quoting Uwe Kleine-König : If the concerns are mostly about the hardware not being rackable, there is a rackable NAS by Netgear: https://www.netgear.com/business/products/storage/readynas/RN2120.aspx#tab-techspecs This seems to be out of stock and discontinued, unfortunately. Anyway, I'm relatively sure, that I can convince my boss to sponsor/donate both armel and armhf hardware for Debian, if that is of any help. Or arm64 used in "32 bits mode".
Re: Porter roll call for Debian Stretch
* ni...@thykier.net [2016-08-17 22:05]: > 2020), please respond with a signed email containing the following > before Friday, the 9th of September: Can you please specify where to respond to? I don't think dozens of emails to -ports and -devel make any sense. Maybe debian-release with CC debian- or to you personally and you'll collect the info? -- Martin Michlmayr http://www.cyrius.com/
Re: Roll call for porters of architectures in sid and testing (Status update)
Hi, b...@decadent.org.uk said: > > I've also provided a couple of kernel patches in the past. I'm cross > > testing with Gentoo to ensure that bugs I report are Debian-specific > > or ia64-generic. > > > > I'll continue testing/software development activity on ia64 for the > > Jessie cycle, and more generally, until Debian drops ia64. I'm already > > waiting for Wayland on ia64 and other big updates. > > > > So please, keep ia64 in the bandwagon ;-) > > But I don't think ia64 is well-supported even in wheezy. The kernel > doesn't boot on some common machines and no-one seems to be able to fix > it. I'm the most recent person to report that bug. Unfortunately I run a sole ia64 box in a production setting and I only have physical access to it during business hours (and no management card) which makes experimenting with kernels difficult. I would be more than happy to sign up as a porter for ia64 however I would need to obtain more hardware. HP rx1620s can be had on eBay for $100 but they are located in California and shipping + customs to Slovakia is not worth it :-/ If someone is willing to donate hardware, or knows of a cheap source of ia64 hardware in Europe I'm more than happy to host it here, maintain it, provide access to developers, and time permitting test and and help diagnose ia64-specific problems with the base system for the lifetime of the jessie release. I am also a long-time Debian user (>10 years, mostly x86/amd64, alpha in the past, some arm work and ia64 from the time I've had this hardware). I am not a DD/DM but have experience as a package maintainer via sponsorship in the past. Martin signature.asc Description: Digital signature
DSA concerns for jessie architectures
[please consider replacing debian-ports@ldo with the appropriate port specific list when replying.] Comrades! At our recent Essen sprint, DSA went through the release qualification matrix (for wheezy, as there isn't one for jessie, yet) and defined a set of requirements that we consider necessary for us to support a port for the next stable release. We have limited these requirements to whether DSA can support a port well or not, and we wanted to establish these requirements early in the release cycle so that our concerns can be addressed. Our requirements for machines are not new; they are: * reliability - The stable release manager requires that we operate three machines for each port: two buildd machines in different locations and one porter machine. These machines must be reliable (see mips for counterexample). * out of band management - We require the ability to manage the machines independently of their primary network interface: serial console or better, remotely-controllable power. * supportability - We require that the machines be commercialy available (within financial constraints) and that they be supportable through a warranty or post-warranty support or are otherwise easy to replace. * stability - We require that the machine's architecture have an actively-maintained stable kernel in the archive. * environment - We require that packages critical for DSA operations be available: puppet, samhain, syslog-ng, ferm/pf, etc. Historically, we have not been enforcing these requirements strictly and this has caused / continues to cause us significant operational challenges resulting in our inability to render the service levels that should reasonably be expected of us. Therefore, we believe it is important that all debian.org machines meet these requirements. Based on the list of requirements enumerated above, we currentlty are concerned about the following architectures from the perspective of using them as debian.org machines: * armel: no remote management (being worked on); no archive kernel for the machines we use. * armhf: no remote management (being worked on). * hurd: no puppet/ruby broken (for 3 months+); lack of firewall support. * mips: existing machines are either not reliable or too slow to keep up; we suspect that they may not be easily replaceable. * mipsel: the porter machine and some of the buildd machines have an implementation error for one opcode; missing kernel in the archive * sparc: no working nflog (mild concern); no stable kernels in stable (compiling clisp for instance crashes the kernel reliably on smetana). We need to run sparc with oldstable kernels to provide stable machines. That's not an option for long. * s390/x: no stable kernels; not sourceable within our budgets (currently relying on sponsors - this has not been a problem so far). We believe that it is the responsibility of the porter community to either source hardware or provide DSA with proposals regarding the hardware Debian should buy. We encourage the porter community to actively assist DSA with the resolution of the above noted concerns regarding various ports. Thanks, Martin Zobel-Helas Debian System Administration Team -- "It pays to be obvious, especially if you have a reputation for being subtle." Martin Zobel-Helas Debian System Administrator Debian & GNU/Linux Developer Debian Listmaster http://about.me/zobel Debian Webmaster GPG Fingerprint: 6B18 5642 8E41 EC89 3D5D BDBB 53B1 AC6D B11B 627B signature.asc Description: Digital signature
Re: Install Linux on Ultra Sparc 10
On Wed, 2012-12-19 at 22:44 +, Amy Tran wrote: > Hello there, > > > > I tried to install Debian version 6.0.6 sparc on sparc ultra machine > and got an error: > > Console: color dummy device 80x25 > > Console [tty0] enable, bootable disable > > And the system was hang, no keyboard response. > > > > Any idea? What happens if you hit Stop-A? What do you have connected to the machine (monitor, ethernet, serial console, Sun keyboard, etc.)? Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1355958282.30544.71.ca...@white.sevalidation.com
Re: Sun Fire V210 NIC's don't work
On Mon, 2012-11-19 at 20:04 +, Anonymous wrote: > SOLARIS 10 SPARC!!! > > You KNOW you want it! ;-) > > No Linux, no FSF, no problem! I'm going to assume this is meant as a helpful suggestion rather than a troll*. But, this is the Debian SPARC mailing list, about running Debian (mostly GNU+Linux based) on SPARC hardware, so saying Solaris, although it may be the fast route to getting a working system, isn't really that helpful and doesn't solve the problem actually being discussed. Cheers, - Martin *. Anyway, the first thing anyone does with a Solaris box is install the FSF/GNU userland and a bunch of developed-for-Linux applications, I mean, that's how you make it useful, right? -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1353356668.19277.552.ca...@white.sevalidation.com
Re: Running applications remotely on headless SPARC box
On Tue, 2012-09-18 at 09:41 +0100, Kaya Saman wrote: > On 09/18/2012 09:13 AM, Martin wrote: > > On Sun, 2012-09-16 at 22:09 +0100, Kaya Saman wrote: > > [...] > > > > > > I guess that I will need to have the X server enabled on both client and > > server machines however, since my V210 doesn't have a GPU only the > > serial console what are my options? > > Not necessarily but you will need some of the X libraries installed so > > you have xhost and similar. I don't recall which packages are needed > > (they may be included in the suggested packages for openssh) but > > installing xterm was sufficient last time I ran into a problem like > > this. > > Thanks Martin for the suggestions > > I was actually trying to get /usr/bin/startkde working as a remote > desktop equivalent to running XDMCP. I'm not sure I've ever seen window manager / desktop environment run over X forwarding. The normal usecase is running individual applications. > My main emphasis has been to try to integrate remote X11 with Sun > Microsystems (Oracle) VDI solution; which on SPARC platform means Sun > Ray Server Software and Secure Global Desktop. > > I managed to get X11 forwarding up and running which is all fine now on > my Sun Ray. > > > However, I think I will start exploring telnet with X11 forwarding and > perhaps even rsh and rexec for less CPU overhead induced by SSH which > continuously needs to encrypt/decrypt the data stream. SSH is probably a lot less hassel and CPU usage tends to be pretty low. The persistent connection options might help. > I just wish that there was some kind of easy integration for both X and > audio between Solaris 10 and Debian. The good side about Solaris 10 is > integration, the bad side, it's over 10 years old and without running > third party packages from sunfreeware/opencsw/blastwave the applications > for it are pretty limited and compiling things isn't that easy at times > (hence power of Linux for desktop usage). There was something called 'network audio', maybe NAS (network audio server) which was intended for forwarding sound to remote X clients. You also might want to look at nxproxy and nomachine's other software. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1347971357.14962.1429.ca...@white.sevalidation.com
Re: Running applications remotely on headless SPARC box
On Sun, 2012-09-16 at 22:09 +0100, Kaya Saman wrote: > Hi, > > I would like to use one of my V210's as an application server however, > there is an issue while using X11 forwarding. > > I have enabled X11_forwarding under /etc/ssh/sshd_config however, > whenever I try to start an application from my x86 box remotely using: > > ssh -X or ssh -Y > > I get presented with "cannot connect to X server". > > > I guess that I will need to have the X server enabled on both client and > server machines however, since my V210 doesn't have a GPU only the > serial console what are my options? Not necessarily but you will need some of the X libraries installed so you have xhost and similar. I don't recall which packages are needed (they may be included in the suggested packages for openssh) but installing xterm was sufficient last time I ran into a problem like this. > Can anyone suggest anything or help? Try running ssh with -v, -vv, -vvv, etc. as it gives some very useful (and verbose output), also check the logs on the server as some errors are not given to the client for security reasons. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1347956014.14962.1393.ca...@white.sevalidation.com
Re: sparc qualification for Wheezy
Hi, On Wed May 30, 2012 at 22:29:32 +0200, Bernd Zeimetz wrote: > Hi, > > > > On Wed May 16, 2012 at 13:19:48 +0100, Adam D. Barratt wrote: > >> Hi, > >> > >> With the sound of the ever approaching freeze ringing loudly in our ears, > >> we're (somewhat belatedly) looking at finalising the list of release > >> architectures for the Wheezy release. > >> > >> Comments on / additions and corrections to the content of > >> http://release.debian.org/wheezy/arch_qualify.html would be appreciated, > >> as would any other information you think is relevant to helping us > >> determine sparc's status for the release. > > > > with my DSA hat on: > > > > We no longer have an UltraSparc II porterbox, and we are considering > > decommissioning our single remaining UltraSparc II buildd machine. > > > > Maybe it would be a good idea to officially drop US II support from > > wheezy since we won't have hardware to test issues on. > > Might make sense, but on the other hand the USII machines were usually those > where issues did *not* show up. Especially USIII liked to be a problem child > as > the specs were never released by sun. It got better with USIV again... So some > state like 'not officially supported, but should just work' is probably the > best > for USII. if that means, DSA does not need to operate those machines for the project any more, i am fine with that. Cheers, Martin -- Martin Zobel-Helas | Debian System Administrator Debian & GNU/Linux Developer | Debian Listmaster GPG key http://go.debian.net/B11B627B | GPG Fingerprint: 6B18 5642 8E41 EC89 3D5D BDBB 53B1 AC6D B11B 627B -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120530203434.gh29...@ftbfs.de
Re: sparc qualification for Wheezy
Hi, On Wed May 16, 2012 at 13:19:48 +0100, Adam D. Barratt wrote: > Hi, > > With the sound of the ever approaching freeze ringing loudly in our ears, > we're (somewhat belatedly) looking at finalising the list of release > architectures for the Wheezy release. > > Comments on / additions and corrections to the content of > http://release.debian.org/wheezy/arch_qualify.html would be appreciated, > as would any other information you think is relevant to helping us > determine sparc's status for the release. with my DSA hat on: We no longer have an UltraSparc II porterbox, and we are considering decommissioning our single remaining UltraSparc II buildd machine. Maybe it would be a good idea to officially drop US II support from wheezy since we won't have hardware to test issues on. Cheers, Martin -- Martin Zobel-Helas | Debian System Administrator Debian & GNU/Linux Developer | Debian Listmaster GPG key http://go.debian.net/B11B627B | GPG Fingerprint: 6B18 5642 8E41 EC89 3D5D BDBB 53B1 AC6D B11B 627B -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120530201601.gg29...@ftbfs.de
Re: kernel panic
On Fri, 2011-08-26 at 07:34 -0700, Sean Whitney wrote: > I have a ultra1 that has been getting kernel panic's since about 2.6.18. > I've changed out all the hardware (except the HD, and network card) and > the panics have followed. I finally caught it happening so my question > is recommendations on where to post it to get some help. > > Thanks > > Sean > > [ 1406.069019] Kernel panic - not syncing: Aiee, killing interrupt handler! > [ 1406.147371] Call Trace: > [ 1406.176544] [0045e5b0] do_exit+0xb4/0x704 > [ 1406.233827] [00427608] die_if_kernel+0x29c/0x2c4 > [ 1406.298408] [00427f98] bad_trap+0x7c/0xf0 > [ 1406.355701] [004220b0] tl0_resv104+0x30/0xa0 > [ 1406.416225] [100d0a6c] esp_free_lun_tag+0x70/0x74 [esp_scsi] > [ 1406.493311] [100d1e44] esp_cmd_is_done+0x24/0x154 [esp_scsi] > [ 1406.570398] [100d3634] scsi_esp_intr+0x128c/0x1de0 [esp_scsi] Looks like it could be a problem with the SCSI hardware or driver; have you tried getting open boot to run a full check of the hardware? Also you might want to check the cabling as on some older SCSI hardware loose or dodgy cabling can cause strange SCSI faults including driver crashes. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1314479958.4074.730.ca...@white.sevalidation.com
Re: failed sparc/powerpc build of libidn 1.22-1
On Jun 20, 2011, at 4:47 PM, Simon Josefsson wrote: > Martin Alfke writes: > >>> /usr/lib/sparc-linux-gnu/gcc/sparc-linux-gnu/4.4.6/ecj1: error while >>> loading shared libraries: libgcj_bc.so.1: cannot open shared object >> >> Seems the same as bug #630417 >> >> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=630417 > > Thanks! > >> Which version of gcj are you using? >> According to the mentioned bug this is fixed in version 4.6.0-6 > > Libidn depend on unversioned 'gcj', so it will use whatever version is > available on the buildd's. It seems sparc has gcj 4.4.6-6 and powerpc > has 4.6.0-13. What's holding up gcj on sparc? > > For powerpc, the error seems strange if indeed powerpc has a more recent > gcj. The error message suggest it is 4.6.1 though?! > > /usr/lib/powerpc-linux-gnu/gcc/powerpc-linux-gnu/4.6.1/ecj1: error while > loading shared libraries: libgcj_bc.so.1: cannot open shared object file: No > such file or directory > > /Simon Looking further into the bug #630417: "Found in version gcj-4.6/4.6.0-13 Fixed in version gcc-defaults/1.105" But: the upload was made for gcc-defaults/1.105 and gcc-4.6.0-6 -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/0cfc662d-79e5-4483-a40e-50d4ae037...@gmail.com
Re: failed sparc/powerpc build of libidn 1.22-1
On Jun 20, 2011, at 4:05 PM, Simon Josefsson wrote: > Hi, > > The java code in libidn fails to build using GCJ on Sparc and PowerPC. > Complete link to log quoted below, but the relevant part is shown > inline. Any ideas? It seems like the gcj compiler is broken on these > architectures. > > /Simon > > make[6]: Entering directory > `/build/buildd-libidn_1.22-1-sparc-O07rsn/libidn-1.22/java/gnu/inet/encoding' > CLASSPATH=../../../../java:./../../../../java:$CLASSPATH javac -d > ../../../../javaCombiningClass.java Composition.java > DecompositionKeys.java DecompositionMappings.java IDNA.java > IDNAException.java NFKC.java Punycode.java PunycodeException.java > RFC3454.java Stringprep.java StringprepException.java > /usr/lib/sparc-linux-gnu/gcc/sparc-linux-gnu/4.4.6/ecj1: error while loading > shared libraries: libgcj_bc.so.1: cannot open shared object file: No such > file or directory > make[6]: *** [classdist_noinst.stamp] Error 1 Seems the same as bug #630417 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=630417 Which version of gcj are you using? According to the mentioned bug this is fixed in version 4.6.0-6 > > > Debian buildds writes: > >> * Source package: libidn >> * Version: 1.22-1 >> * Architecture: sparc >> * State: failed >> * Suite: sid >> * Builder: lebrun.debian.org >> * Build log: >> https://buildd.debian.org/fetch.cgi?pkg=libidn&arch=sparc&ver=1.22-1&stamp=1308563014&file=log >> >> Please note that these notifications do not necessarily mean bug reports >> in your package but could also be caused by other packages, temporary >> uninstallabilities and arch-specific breakages. A look at the build log >> despite this disclaimer would be appreciated however. > > Debian buildds writes: > >> * Source package: libidn >> * Version: 1.22-1 >> * Architecture: powerpc >> * State: failed >> * Suite: sid >> * Builder: poulenc.debian.org >> * Build log: >> https://buildd.debian.org/fetch.cgi?pkg=libidn&arch=powerpc&ver=1.22-1&stamp=1308562221&file=log >> >> Please note that these notifications do not necessarily mean bug reports >> in your package but could also be caused by other packages, temporary >> uninstallabilities and arch-specific breakages. A look at the build log >> despite this disclaimer would be appreciated however. > > > -- > To UNSUBSCRIBE, email to debian-powerpc-requ...@lists.debian.org > with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org > Archive: http://lists.debian.org/87zklczjgl@latte.josefsson.org > -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/05f9393b-9637-41ee-b364-3f71b2502...@gmail.com
Re: virtualization on Sparc architecture?
On Sat, 2011-05-28 at 17:53 -0500, mmejiav wrote: > 2011/5/28 mmejiav : > > 2011/5/27 Martin : > >> On Fri, 2011-05-27 at 09:32 -0500, mmejiav wrote: > >>> hi > >>> I have a Sun Blade 100 with Solaris 10. > >>> Does Debian for Sparc architecture have a tool for making > >>> virtualization on this? Something like openvz… I search information > >>> about this topic on google but I didn’t find anything > >>> Thanks in advance > >> > >> Linux VServer ( http://linux-vserver.org/ ) has been ported to SPARC and > >> is an "OS level virtualisation" system similar in concept to OpenVZ. I > >> ran a system that used it for several years with few problems. At one > >> point Debian shipped VServer enabled kernels for SPARC. > >> > >> Alternatively you may be able to use the cgroup tools to set up > >> something similar using stock kernels. > >> > >> HTH > >> > > > > hi Martin and thanks for your answer > > > > Rigth now i'm installing debian squeeze on the Blade 100 machine (sparc). > > In the wiki debian (http://wiki.debian.org/SystemVirtualization) I > > found more information about vserver, but the page says > > > > LinuxVserver - Mostly used on servers. Creates multiple isolated > > containers. Will be deprecated in squeeze. > > > > Deprecated on squeeze? then for sparc whats the next virtualization > > tool to be used on Debian systems? > > > > -- > > hi > I installed vserver > > apt-get install linux-image-vserver-sparc64 util-vserver > vserver-debiantools linux-headers-2.6-vserver-sparc64 > > But after the installation the system shows: > > Linux-VServer capability not detected in kernel. You'll have to reboot the machine to pick up the new kernel. > Mmmm ok. > I reboot the machine and run > > $ uname -ar > Linux gandalf 2.6.32-5-vserver-sparc64 #1 SMP Thu May 19 08:03:14 UTC > 2011 sparc64 GNU/Linux > > The question: is the system ready for the vservers or not? If you're running the correct kernel then it should work. What do the vserver tools say? You might find you get a better response to vserver specific questions on the vserver list / IRC channel. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1306875686.10039.102.ca...@white.sevalidation.com
Re: virtualization on Sparc architecture?
On Sat, 2011-05-28 at 17:17 -0500, mmejiav wrote: > 2011/5/27 Martin : > > On Fri, 2011-05-27 at 09:32 -0500, mmejiav wrote: > >> hi > >> I have a Sun Blade 100 with Solaris 10. > >> Does Debian for Sparc architecture have a tool for making > >> virtualization on this? Something like openvz… I search information > >> about this topic on google but I didn’t find anything > >> Thanks in advance > > > > Linux VServer ( http://linux-vserver.org/ ) has been ported to SPARC and > > is an "OS level virtualisation" system similar in concept to OpenVZ. I > > ran a system that used it for several years with few problems. At one > > point Debian shipped VServer enabled kernels for SPARC. > > > > Alternatively you may be able to use the cgroup tools to set up > > something similar using stock kernels. > > > > HTH > > > > hi Martin and thanks for your answer > > Rigth now i'm installing debian squeeze on the Blade 100 machine (sparc). > In the wiki debian (http://wiki.debian.org/SystemVirtualization) I > found more information about vserver, but the page says > > LinuxVserver - Mostly used on servers. Creates multiple isolated > containers. Will be deprecated in squeeze. > > Deprecated on squeeze? then for sparc whats the next virtualization > tool to be used on Debian systems? Most of the people who use vserver in commercial systems either patch their own kernel or use the Bristol wireless Debian packages. As development tracks the more recent kernel a Debian/Stable package of vserver isn't too useful, hence the depreciation. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1306875514.10039.99.ca...@white.sevalidation.com
Re: virtualization on Sparc architecture?
On Fri, 2011-05-27 at 09:32 -0500, mmejiav wrote: > hi > I have a Sun Blade 100 with Solaris 10. > Does Debian for Sparc architecture have a tool for making > virtualization on this? Something like openvz… I search information > about this topic on google but I didn’t find anything > Thanks in advance Linux VServer ( http://linux-vserver.org/ ) has been ported to SPARC and is an "OS level virtualisation" system similar in concept to OpenVZ. I ran a system that used it for several years with few problems. At one point Debian shipped VServer enabled kernels for SPARC. Alternatively you may be able to use the cgroup tools to set up something similar using stock kernels. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1306507120.562.333.ca...@white.sevalidation.com
Re: How to tell the compiler NOT to assume cpu provides atomic instructions
On Wed, 2011-03-16 at 23:33 +, brian m. carlson wrote: > *Please* do not CC me. My M-F-T is set appropriately; please respect > it. My apologies; it appears my mail client does not pick it up correctly. > On Wed, Mar 16, 2011 at 11:03:17PM +0000, Martin wrote: > > 1. SPARC V8 only has very weak atomic ops (IIRC it's only test and set) > > which aren't sufficient to implement all of the GCC atomic built-ins > > without some extra work. This is why there will be a need to provide > > your own implementation and a linking problem if you don't. > > > > 2. SPARC V9 has CAS which allows most of the GCC atomic ops to be > > emulated but is 64 bit, which may not be what you want. > > Debian only supports v9 processors, but the sparc port is 32-bit. By > default, GCC uses -mcpu=ultrasparc and is configured for sparc-linux-gnu > (which is 32-bit). There is no need to specify any flags to GCC, since > it does the right thing by default. All UltraSPARC operations are > supported, but the code generated is 32-bit. > > By default I think GCC still generates V8 code on Debian, setting > > -mcpu=v8+ should do what you want. Furthermore I might suggest using > > Hans Boehm's libatomic-ops: > > blackhole no % gcc -mcpu=v8+ foo.c > foo.c:1: error: bad value (v8+) for -mcpu= switch > > I believe Debian's GCC does this by default already, with or without > -mv8plus: > > blackhole ok % file a.out > a.out: ELF 32-bit MSB executable, SPARC32PLUS, V8+ Required, version 1 > (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not > stripped > My apologies for posting outdated information. I shall consider myself corrected on this issue. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1300319119.4094.1462.ca...@white.sevalidation.com
Re: How to tell the compiler NOT to assume cpu provides atomic instructions
On Tue, 2011-03-08 at 02:03 +, brian m. carlson wrote: > On Mon, Mar 07, 2011 at 07:54:13PM -0500, RR wrote: > > checking whether the C compiler (gcc -pthread -I../..//include > > -I/usr/include/libxml2 -pipe -Wall -Wstrict-prototypes -Wmissing-prototypes > > -Wmissing-declarations -g3 -mtune=ultrasparc -mcpu=v8 -fomit-frame-pointer > > -O6) works ... yes > > checking whether the C compiler (gcc -pthread -I../..//include > > -I/usr/include/libxml2 -pipe -Wall -Wstrict-prototypes -Wmissing-prototypes > > -Wmissing-declarations -g3 -mtune=ultrasparc -mcpu=v8 -fomit-frame-pointer > > -O6) is a cross-compiler... no > > Okay. I've pinpointed the problem. If atomic intrinsics are used but > not supported on the target architecture (variant), GCC will emit a link > error for that function. > > Your problem is -mcpu=v8. v8 processors (AFAICT) do not support the > intrinsics necessary. Remove that option (or whatever is causing that > option) and it should work. It may be that for some reason configure > thinks that 32-bit code should use -mcpu=v8, but that isn't the case. > Debian's default configuration (without -mcpu and -mtune) should work > just fine for you. > FWIW... I've dug out my notes on atomic operations on SPARC. 1. SPARC V8 only has very weak atomic ops (IIRC it's only test and set) which aren't sufficient to implement all of the GCC atomic built-ins without some extra work. This is why there will be a need to provide your own implementation and a linking problem if you don't. 2. SPARC V9 has CAS which allows most of the GCC atomic ops to be emulated but is 64 bit, which may not be what you want. 3. SPARC V8+ is the instruction set of the V9 processors running in 32 bit mode. Thus it includes CAS and is 32 bit. By default I think GCC still generates V8 code on Debian, setting -mcpu=v8+ should do what you want. Furthermore I might suggest using Hans Boehm's libatomic-ops: http://www.hpl.hp.com/research/linux/atomic_ops/ http://packages.debian.org/search?keywords=libatomic-ops As it gets around a number of the unpleasent issues with memory barriers. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1300316597.4094.1451.ca...@white.sevalidation.com
Re: GCC-4.5 as the default for (at least some) architectures
On 2 March 2011 02:34, Matthias Klose wrote: > armel (although optimized for a different processor) Hi For which processor (/architecture) is it optimized, and do you mean optimized-for, or only-runs-on? I ask in case this would mean dumping all the armv4t systems that are using Debian armel. M -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/aanlktiktn0zoa_hzgciwhkzbup7_ji6pbeji+p4c7...@mail.gmail.com
Re: Installing Debian from NFS
On Wed, Feb 02, 2011 at 01:49:47PM -0500, RR wrote: > Hello, > > this may be a very obvious one of those things that I should just "Google" > but given my last experience, and unsuccessful Googling, I figured maybe I > should ask the community as to what the CORRECT way to install Debian on a > Sun machine such that it uses the ISO/Distribution on my local NFS server as > opposed to just picking up the boot.img and then have the rest of it be > downloaded from the Internet. I have a LOT of SUN machines to install and > it'd be cool if I can just install them off the distro on the nfs server. > > The way my boot server is setup is, that it has tftp, /etc/ethers and > /etc/hosts files all setup. You'll also need either bootparamd or dhcpd set up. > I have the boot.img from Debian Lenny for sparc > in /tftpboot and also have the hex and decimal/AF_INET format of the > intended IP Address soft-linked to this boot.img in the /tftpboot directory > i.e. 0A01030B -> boot.img. I've go this as C0A8000F.SUN4M -> sparcroot/boot/tftpboot.img The debug log of the daemons above will tell you exactly whan name the SUN machine is looking for. > I also have the entire debian-DVD-lenny.iso > available and is mounted on /mnt/iso by way of lofiadm so that /mnt/iso has > the contents of the ISO and not the bundled iso image. > > Now, how do I tell the installer to go pick it up from there? I have done a > bit of research and all I see everywhere is mentions of some file called > root.tar.gz and debian-installer and what not...but I can't figure out where > do I get that from. > > I have tried: > ok> boot net Should be enough to get thing started. I hope it prints out that it's trying to in touch with a bootp server. Martin > and > > ok> boot net - install nfsroot=10.1.3.1:/mnt/iso > > and > > ok> boot net install > > Any one wish to comment? > > Thanks so much > \RR -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110217231243.GB24381@palantir8
Re: Debian on SPARC T2+?
On Mon, 2010-11-15 at 16:37 +0100, Adrian Zaugg wrote: > Dear List > > Is it possible to boot Debian on a SPARC T2+ System? > > Regards, Adrian. > > PS: Please CC, I'm not on the list. Given there have been successful reports of people running Debian on T1's, it should be possible. The problem is more likely to be hardware that doesn't have drivers or doesn't have drivers in the current default kernel build. Using the hardware virtualisation support may offer a way round this if it is a problem. Why not give it a go and report back. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1289858187.3927.486.ca...@white.sevalidation.com
Re: sungem NIC stops working
On Wed, 2010-10-27 at 18:02 +0100, Bruce Cran wrote: > Hi, > > We have several Netra T1's running debian stable, one of which is > a gateway for the network. We've been finding that about once a > week or more it stops responding to network traffic - we've been > rebooting it, but I'd like to work out why it's happening. The only > error that gets logged is: > > [409455.008758] eth0: RX MAC fifo overflow smac[00010400] > > That message gets logged during normal operation too, but it's the last > message in the buffer when the NIC stops working. Is this a known > problem, or how could we go about finding out what's going wrong? The first thing to do would be submit a bug report, giving as many details about the kernel, hardware and operational environment (particularly network throughput) as possible. Then it might be worth feeding the error message into google 'RX MAC fifo overflow' seems to find quite a few things. Beyond that, the code that seems to be causing the warning is: http://lxr.linux.no/#linux+v2.6.36/drivers/net/sungem.c#L482 So it might be worth diff-ing the kernel source you are running against the latest and an older kernel to see what (if anything) has changed. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1288215500.3738.295.ca...@white.sevalidation.com
Re: kernel hang
On Tue, 2010-10-26 at 18:39 +0200, Árpád Magosányi wrote: > Yet another hang. > I am on a sun netra T1. I cannot yet give you details, as the machine > is hung some 30kms away. When you have details it might be worth submitting a bug report if you haven't done so already. > What I have noticed that /proc related processes (like 'w', 'ps') > hang, but not everything /proc related, for example I could login > through ssh and sudo without problem. > Strace revealed that 'ps ax' hangs when trying to > read /proc//cmdline (the read() have never returned), where > is the process id of an apache. > Kernel logs did not reveal anything related. > (I tried to shut it down, but it also hung, and I killed processes in > the wrong order.) That sounds like the kind of thing that should be reasonably easy to trace. Do you have a test case that triggers this bug? If so then there is a reasonable chance it can be fixed, if not then you'll likely have a harder time. > I am struggling with mysterious hangs on my various suns (ultra2, > netra T1, ...) since years. Have you submitted bug reports? Is it always the same problem? If you have serial access then you should be able to get OpenBoot to dump the register values and the program counter should give you some hint to the location of the problem. > Shall I forget sparc altogether? > Change to Ora^H^HpenSolaris? Without knowing more information about what the actual problems are I'm not sure that's really answerable. That said I might be tempted to point out that Ultra 2's and Netra T1's are likely to be around 10 years old, if not older and were probably designed for an operational life of 3 to 5 years, so your problem might just be hardware aging. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1288134449.3738.115.ca...@white.sevalidation.com
Re: bash performance
On Tue, 2010-10-26 at 22:26 +0300, Tarko Tikan wrote: > hey, > > > Just for comparison purposes, the same command gives me the following on a > > much older Sunfire V100 (Ultrasparc IIe @ 500mhz) running Debian, and > > another of the same machine running Solaris 9: > > 8.5sec almost backs up my result of 11sec. It still seems a little slow. Have you tried compiling a version of bash tuned for that machine? > At 1.2GHz per core (sure, it looks like 32 cores to OS but you should still > be able to use 1 of 4 cores at full speed), Depends on the workload. For example, for some of the work I do, 100% CPU usage translates to 15-20% of cycles actually doing something useful and the remaining 80-85% of the time the processor is stalled waiting on memory or branching. In this case you can use all four cores at 'full speed' simultaneously. As with all performance related things - it depends on what you are trying to do and what about that you want to measure. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1288133713.3738.106.ca...@white.sevalidation.com
Re: bash performance
On Mon, 2010-10-25 at 23:31 +0300, Tarko Tikan wrote: > hey, > > This is a fresh squeeze install on T5220 (same for the laptop I'm comparing > with). Whole machine seems strangely slow but one thing I can point out for > sure is bash. I've been debugging this issue for a week now (even comparing > strace -ttt time outputs with random intel machine). Does this seem strange > to anyone else?: > > randomx86laptop$ time (VAR=0; for i in `seq 0 9`; do let VAR=VAR+i; done) > real0m1.235s > user0m1.212s > sys 0m0.036s > > niagara2$ time (VAR=0; for i in `seq 0 9`; do let VAR=VAR+i; done) > real0m11.193s > user0m10.777s > sys 0m0.388s > > I first noticed this with bash_completion - every new bash instance takes 2-3 > seconds to start. > > Anyone have an idea whats going on? Can you give us a bit more info about the laptop? IIRC a T5220 runs an UltraSPARC T2 at about 1Ghz. These are designed for throughput, not latency. The T2 is out of order but only 2 way dispatch (IIRC), compared to a 3.4Ghz Intel chip with 4 way dispatch and full out of order instruction completion. Thus in terms of single threaded performance, I'm not exactly surprised an x86 laptop is 'faster'. A factor of 10 seems a bit steep but 3-4 maybe 5 times I could believe. However if you look at performance of code that is tightly branching, memory latency/bandwidth sensitive or thread interaction critical, you should see scaling up to .. well ... it depends on the app but I've seen near linear speed up, up to 32 threads on T1s. What other benchmarks have you tried? Also have you tried with any other OS (i.e. Open Solaris) on the same hardware; this should rule out anything Debian / Linux specific. Likewise have you tried other machines? Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1288041064.3738.25.ca...@white.sevalidation.com
Re: OpenVZ upgrade path?
On Tue, 2010-10-19 at 02:27 -0400, Ivan Jager wrote: > Hi, > > I'm not sure if this is the best place to ask this, but I am > wondering what the best plan is to upgrade my openvz system. > > A short summary of how I got here follows: I installed etch when > that was the stable release, and openvz was supported. When lenny > was released, the release notes claimed better openvz support, > but in fact openvz support had been silently dropped. So, I still > feel somewhat betrayed, and still have lenny system running on an > etch kernel with openvz on it. With the upcoming release of > squeeze, etch will go from being oldstable to being ancient and > completely unsupported, so I don't want to continue running an > etch kernel. > > So, I suppose my options are: > 1. Take up maintaining openvz sparc support. (I wwould consider > this if there are other users who would also contribute, but if > I'm the only one it seems like a waste of time. I did have to > patch the patches so I have a tiny bit of experience.) > 2. Switch to vservers and hope not to get dropped again. > 3. Switch to FreeBSD jails and be sad that I'm not running > Debian but happy I have ZFS. > 4. Switch to OpenSlolaris containers and be even more sad I'm not > running Debian. > 5. Start a Debian/kFreeBSD/sparc port... (Seems like it would be > more work than 1, although it may have more users too.) > 6. Use some other virtualization thing I haven't heard of on > Debian. (lxc seems like it should be platform independent, but > seems to be mysteriously missing a sparc package) > > So, any advice on which way to go? What are other people using? > Should I be asking or CCing somewhere else? I have used vserver on SPARC in a production system. It worked and the developers seemed keen on supporting x86 architecture. Note that many vserver users use the project's packages for kernel and tools as they track vserver development in a more timely fashion. Thus having vserver kernels included in the main Debian repositories is less of an issue than you'd imagine. Longer term, if you go with a technology that is in squeeze then you should have at least a handful of years before you are in a similar situation - even in the worst case. When the future has been discussed on the vserver mailing list there was a suggestion that once cgroups were capable of everything required, vserver would slowly become a set of userspace tools. There is no definate timeline for this but the amount of code required in the kernel patches does seem to be decreasing. I would have thought vserver or lxc would have be the best effort vs future security trade off. Good luck and let us know what you pick. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1287491182.11418.103.ca...@white.sevalidation.com
Re: Help creating drivers for Tadpole SPARCLE
On Tue, 2009-12-22 at 21:33 +, Immolo wrote: > > Brian Szymanski wrote: > > If I were you I'd OpenBSD a shot - it has pretty good sparc support, > > and if the installer detects the keyboard, you've got some > > documentation (in the form of a driver) that you can look at. A few > > years back I had a similar machine (tadpole sparc laptop, don't > > remember which variety tho) which linux and NetBSD failed to > > understand keypresses from, but OpenBSD did just fine. > > > > Good luck... > Thanks OpenBSD does work with the keyboard but is there away to use this > so I can get the keyboard to work in Linux? You would likely have to re-write the driver for Linux. A direct port may not be possible due to BSD/GPL licencing issues. Start by reading the source of the driver and brushing up on Linux kernel development. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Help creating drivers for Tadpole SPARCLE
On Mon, 2009-12-21 at 21:29 +, Immolo wrote: > Just recived contact and they are unwilling to give any documentation > other then that what is at > > ftp://ftp.tadpole.com/support/SPARCLE/ It will be significantly more difficult to write drivers for an undocumented device. Your first step should be to gather as much information as possible on it. sunhelp.org might be a good place to start as the rescue list has a lot of folks who are very knowledgeable about Sun kit. Have you identified via OpenBoot, proc, etc. what bus the keyboard is attached to and what the keyboard controllers are? Have you / can you open up the machine and find out what chips / parts are used? Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Help creating drivers for Tadpole SPARCLE
On Mon, 2009-12-21 at 20:32 +, Immolo wrote: > Hello, > > I own a Tadpole SPARCLE and would really love to get Linux to run on > this as lets be honest Solaris is awful but this device seems to use a > special keyboard interface. > > Would anyone be willing to spend some time to help me get it working? > > To clear things up Debian Lenny will boot and use an USB keyboard just > not the laptop keyboard and this is the same in Solaris until the driver > is installed. Do you have documentation for the keyboard controller? I know at one point Sun had someone in charge of answering and helping with documentation requests. If you don't have, then contacting them might be a good place to start. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Bug#549407: ivtools 1.2.6-1 FTBFS on sparc
On Sat, Oct 03, 2009 at 09:17:52AM +0200, Andreas Barth wrote: > Package: ivtools > Version: 1.2.6-1 > Severity: serious > > Hi, > > this package FTBFS on sparc: > > make[2]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6' > depending for LINUX in > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src > make[3]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src' > depending for LINUX in > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common > make[4]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common' > depending for LINUX in > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX > make[5]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX' > g++ -M -w -DMAKEDEPEND -Dcplusplus_2_1 -Wno-deprecated > -I/build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX/.. > -I/build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX/../.. > -I/build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src > -I/build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/include > -I/build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/include/ivstd > -I/usr/include -I/usr/local/include -UHAVE_ACE > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX/../*.c > >Makefile.depend > g++: > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX/../*.c: > No such file or directory > g++: no input files > make[5]: *** [depend] Error 1 > make[5]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX' > make[4]: *** [depend] Error 2 > make[3]: *** [depend] Error 2 > make[4]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common' > make[3]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src' > make[2]: *** [depend] Error 2 > make[2]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6' > > make subdirs > make[2]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6' > making all for LINUX in > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src > make[3]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src' > making all for LINUX in > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common > make[4]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common' > making all for LINUX in > /build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX > make[5]: Entering directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX' > make[5]: *** No rule to make target > `/build/buildd-ivtools_1.2.6-1-1-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX/../math.c', > needed by `math.o'. Stop. > make[5]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common/LINUX' > make[4]: *** [all] Error 2 > make[4]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src/IV-common' > make[3]: *** [all] Error 2 > make[3]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6/src' > make[2]: Leaving directory > `/build/buildd-ivtools_1.2.6-1-sparc-mqkrRP/ivtools-1.2.6' Hi, debian-sparc people, This is a strange FTBFS. I uploaded a NMU fixing the 64bit build (#545834), but seems that this problem is not a temporary buildd problem but is still present. More funny, build previously worked for powerpc, but now a bug similar to this one appeared (nothing was touched in that area), so this particular FTBFS currently happens only in sparc and in powerpc. I am thinking about a race condition, but want to look more. I'd like to try building the package in smetana and see what happens, but libace-dev is missing from its sid dchroot. What is the procedure for asking libace-dev be installed in sid chroot at smetana.debian.org? Cheers, -- Agustin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Mysql 4GB mem limit and SPARC64
On Fri, 2009-07-31 at 09:35 +0200, Davide Ferrari wrote: > Hi > > I've installed Debian SPARC64 port on a Sunfire T5220 (so, a Niagara2) with > 32GB of RAM but unfortunately mysql is, as every userland program in Debain > SPARC64, a 32bit application. > This is a great limitation cause you cannot make mysql use more than 4GB of > RAM and with really big InnoDB tablespace this is a real PITA cause you > cannot > get all the queries cached in RAM and performances drop, so we're just > wasting > RAM cause this should be a mysql box. > > Any idea on how to solve this? Install 64 bit versions of the database and any necessary libraries. Some may be packaged in Debian (there is a 64 bit libc package for example), the rest you will have ot build from source. > Does recompiling work? Should do. If DB hasn't been built / run on a 64 bit platform before you could run into issues but if it's reasonably standards compliant and coded in a sensible way it shouldn't be a big deal. > Even gcc is a 32bit > application so I'm not pretty sure. You might want to look at -m64 and the SPARC options: http://gcc.gnu.org/onlinedocs/gcc-4.3.2/gcc/SPARC-Options.html#SPARC-Options Given the strong similarities between 32 and 64 bit SPARC architectures and the tendancy for people to run mixed systems it makes sense to have GCC able to generate both. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Systems which can run on a SS2.
On Sun, 2009-07-26 at 14:38 -0700, Peter Crawford wrote: > Mon, 25 May 2009 11:29:27 +0200 Joël BERTRAND wrote, > "Etch worked fine." > > Whereas > http://www.debian.org/releases/stable/sparc/ch02s01.html.en#sparc-cpus > states > "sun4, sun4c, sun4d, sun4m >... > The last Debian release to support sparc32 was Etch, but even then only > for sun4m systems. Support for the other 32-bits subarchitectures had > already been discontinued after earlier releases." > > The Web page is wrong? I believe the web page is the official position. The differences between 32 bit SPARC based machines are not that great (multiply and divide added in V8 / sun4m, plus (IIRC) one or two of the synchronisation operations). Thus it is possible that older machines could be run using etch or mostly etch. Plus, without more context I couldn't be entirely sure which machines Joel is refering to. > Mon, 25 May 2009 11:29:27 +0200 Joël BERTRAND wrote, > " ... good results with lenny and 2.6 kernel (only UP and sparcv8 [SS20] ..." > > Would UP be Ultra Panther or Ultra Plus? Not that I > have either; just for interest. I suspect it is uni-processor as SS20's could be fitted with 2 (or with later modules, 4) processors, althought support for this has been iffy for some time. > " ... sun4[cdm] stations, you have to use NetBSD 4.0.1 (or 5.0 in UP)." > > In http://www.netbsd.org/ports/sparc/ > sun4c is listed as supported. Absence of reference > to a release should mean that 5.0 works. I am under the impression that sun4d and sun4m are effectively supersets of sun4c, thus 4c support should be sufficient. > Mon, 25 May 2009 16:31:12 +0100 Martin wrote, > " ... and remember having an insecure box on the net doesn't just affect > you; it potentially affects everyone." > > You've convinced me. I'll learn to install NetBSD. > Again, without context, I'm not sure if you're quoting me or what I was talking about but I guess it's a question of what you want to do with the machines, an older version of Debian may be sufficient, or maybe NetBSD. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Tadpole Display Settings
On Wed, 2009-07-22 at 09:42 -0700, Joseph Simantov wrote:
> When trying to boot Debian (both 4 and 5), it doesn't go beyond the OpenBoot.
>
> I tried Aurora / Corona / Fedora, and the display jams up when the first box
> is supposed to be displayed ("Media test")
>
> BTW, being aware this is not the right mailing list, would you by any chance
> know how I can update the OpenBoot using the Tadpole upgrade patch? (you can
> reply offline)
You may have more luck asking the Sun Rescue list:
http://www.sunhelp.org/mailman/listinfo/rescue
if it is a general hardware question.
The general proceedure is normally either from Solaris using the
provided tools or netbooting the firmware updater as you'd netboot a
kernel. Your best bet is to check the documentation that comes with it.
HTH
Cheers,
- Matin
--
To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: HPPA and Squeeze
On Thu, 2009-06-25 at 10:10 +0200, Matthias Klose wrote: > Mike Hommey schrieb: > > On Thu, Jun 25, 2009 at 01:32:09AM +0200, Matthias Klose wrote: > >> Luk Claes schrieb: > >>> Matthias Klose wrote: > >>>> Grant Grundler schrieb: > >>>>> On Fri, Jun 12, 2009 at 08:49:26AM +0200, Luk Claes wrote: > >>>>>> Grant Grundler wrote: > >>>>>>> On Tue, Jun 02, 2009 at 03:07:35PM +0100, Neil McGovern wrote: > >>>>>>> http://lists.debian.org/debian-release/2009/04/msg00303.html > >>>>>> Note that it's wrong to assume we will come with the answers. > >>>>> I was expecting a summary of specific issues from an organization > >>>>> that claims to operate transperently. The hand waving is easy. But > >>>>> doesn't resolve problems and doesn't meet my expectation of an "open" > >>>>> organization that I've donated money, time, and materials to. > >>>> +1. dropping hppa as a release architecture was not communicated by the > >>>> release > >>>> team at all. I did spend some time to get gcj / default-jdk working on > >>>> hppa, > >>>> and some money (buying a new disk for a hppa machine) to help this port. > >>>> The > >>>> time and the money could have spent better, if d-r would have better > >>>> communicated about their intent. > >>> There are issues with the hppa port where the release team considered > >>> dropping it since 2005 communicated to the porter list... > >>> > >>>> hppa is not in a good shape, but there are other architectures which are > >>>> not > >>>> better (sparc, mips*) from a toolchain point of view. what about these? > >>> I'm not aware of current toolchain issues on sparc and the issues on > >>> mips* still seem to be manageable, no? > >> sparc-biarch defaulting to 32bit isn't supported by upstream; there are > >> requests > >> to move to v9 optimization by default, which requires some work in the > >> compiler. > >> I don't plan to update this for upcoming GCC versions, and there's no > >> interest > >> by upstream to help with this kind of setup. You can't buy v8 software for > >> years > >> now, but afaik all our machines run 64bit kernels. Maybe it's time to > >> acknowledge this, remove sparc from the list of release architectures and > >> go on > >> with sparc64? > > > > Isn't sparc64 a full 64 bits port ? sparc is unfortunately not amd64, > > where the pros of the added registers overcome the cons of bigger > > pointers. In other words, 64 bits code is slower, fatter and more memory > > hungry than 32 bits code on sparc. > > which of the previous statements did you check? I don't every recall seeing any general purpose comparisons on this; if anyone has links to one I'd be most interested. I have recollections of seeing it claimed in some of the Sun manuals and has been accepeted 'cannon' for the entire time I've been using SPARC machines.On specific programs I've tested I've seen a 10-15% slow down moving from 32 to 64 bit code. The key difference is pointer size and the effect this has on caching effects; thus the slow down varies program to program and it is possible that the programs I'm interested in are at the upper end of the range. > E.g. speed comparing the current > 32bit v8 with 64bit ultrasparc code? I think V9 vs. V8+ would be a fairer comparison, plus there is a separate question over the default -mtune options to use. > and even then I wouldn't care that much if it becomes maintainable. (I guess I don't fully understand the context of this but ...) while there is GCC support for V8 / V8+ having the userland 32 / kernel 64 split shouldn't be a big deal, should it? HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Reducing involvement with sparc port
On Thu, 2009-06-25 at 22:45 +0100, Jurij Smakov wrote: > Hi, > > After giving it some thought, I've decided to significantly reduce my > involvement with the sparc port, mainly due to lack of motivation and > other responsibilities I'm in the process of assuming. Last two Debian > kernels (2.6.29 and 2.6.30) do not even boot on my sparc box, and I have > no interest whatsoever in tracking the reason for that down. If someone > out there is still interested in actively keeping sparc port afloat, I > would suggest to try and sort these kernel issues out first. Another > couple of problems I've made pretty much no progress on are disabling > CONFIG_PROM_CONSOLE (bug 525958) and adding support for loading VIO > devices to udev (bug 526621). The former should improve the performance > on Niagara quite a bit, the latter should make the LDOMs usable on sparc > without manual tweaking. > > Good luck. Thank you for all of your hard work over the years; I've appreciated it and I suspect many others have as well. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
RE: Debian on a Sparcstation 2
On Mon, 2009-05-25 at 06:30 -0700, Peter Crawford wrote: > At Mon, 25 May 2009 15:14:00 +0200 > joel.bertr...@systella.fr wrote, > > Etch for both. SS2 cannot run with Sarge. > > Thanks. > > > I know, but as sun4, sun4c and sun4m support have been dropped, you > > have to install another OS or drop your SS2 :-( > > OK thanks. Will try to put Etch on the external > drive and keep NetBSD in mind. What is likely > to fail if Etch is left on the SS2 for years? > The first problem you are likely to have is that you won't recieve security updates. If you're willing to backport / monitor fixes, accept that it is insecure* or lock down the machine to a state that there aren't many viable attack vectors then this *may* not be a problem. Apart from that nothing should 'fail' as such; but you will find that it gets progressively harder to build / install up to date software as more and more of the dependancies have to be built as the system ones are not available or not up to date. If you're OK with this then there is no reason that you can't keep using the system until the hardware finally gives up the ghost. Make sure it's cooled adequately and power cycled as infrequenty as possible and you should be able to run it for years to come. Cheers, - Martin *. and remember having an insecure box on the net doesn't just affect you; it potentially affects everyone. -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Debian on a Sparcstation 2
On Sat, 2009-05-23 at 21:34 -0700, Peter Crawford wrote: > An old Sparcstation 2 has been running Woody since > it was installed ca. 2002 by diskette -> network. > Recently I've added an external 18 GB drive and > am interested to install a newer Debian to it. Can > anyone comment _pro_ or _con_ about installing etch > or lenny or squeeze on a Sparc 2? > > Efforts described below have failed. If a system > upgrade is warranted, can anyone suggest how to > accomplish it. You may want to check when support for SPARC V7 / sun4c machines was dropped. For releases where SPARC V8 / sun4m was supported, the differences may be minimal and recompiling a few key binaries may be sufficient. I believe that Debian is moving to supporting only SPARC V9 based machines so for very recent code you may have problems. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: sparc/debian/linux procedures
[ Apologies for the cross post ]
On Mon, 2009-05-04 at 15:26 -0400, Brian Thompson wrote:
> All,
> It's my understanding that there are a team of people who are focused
> on the sparc kernel itself (which is used by Debian as well as some of
> the other distributions - Aurora, etc).
This is a subset of the kernel developers. David Miller is (I believe)
the key man.
> It's also my understanding that there are a team of people who are
> focused on making sure that the sparc port of Debian works properly
> as a complete Debian OS distribution for sparc.
It's more of a loose affiliation, but yes, these are some of the Debian
developers on the debian-sparc list.
> In addition, I understand that there's also a team of people who are
> focused on making sure that the Debian distribution as whole
> (non-arch specific) functions properly and that changes on one port
> don't end up inadvertently causing problems for other Debian ports.
Again, more a loose affiliation - this is essentially the work of the
Debian developers. A small number of developers have responsibility for
over all integration (i.e. the release team, buildd maintainers, etc.)
but most work is done on a package by package basis with a small number
of folks working on each (often one or two).
> Likewise I understand that there's a team of people who are focused
> on making sure that the linux kernel as a whole functions properly and
> that changes specific to one arch don't end up inadvertently causing
> problems for other linux kernel archs.
This is, in general the Linux kernel developers; although, again, their
responsibilities and organisational structure vary.
> My question is - when I find things that worked in Ubuntu sparc
> but not on Debian, what is the proper procedure for resolving the
> issue? Is there a checklist or flowchart anywhere public that should
> be followed when issues are found?
>
> I'm guessing the first step is probably to determine whether it's a
> kernel issue or an issue external to the kernel so that a bug report can
> be filed with the correct team (while also checking to see if the issue
> has already been reported), but again that's just a guess.
A general procedure might be:
1. Identify which package(s) are causing the problem.
2. Attempt to identify what conditions / factors / circumstances trigger
the issue. All the normal rules about writing bug reports apply.
3. File a bug report against the relevant Debian package.
4. Assist the package maintainer with any follow up queries.
If you have time and access to a version of the package that does work,
it might be helpful to track down which differences are causing the
problem, and if possible, submit a patch. Certainly, including a
reference / pointer to the nearest version of Ubuntu package that works
would be helpful.
If the bug turns out to be something that is not specific to Debian and
is a more general problem then the packages maintainer may forward it
("upstream") to the main developers for that package.
> Any guidance would be greatly appreciated.
Does the above help? I'm far from an expert on this; I'm just an
end-user, but the above procedure has worked for me.
HTH
Cheers,
- Martin
--
To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Regarding older Sparc hardware (32-bit)
On Mon, 2009-05-04 at 19:52 +, brian m. carlson wrote: > On Mon, May 04, 2009 at 01:37:46PM -0500, Dan Oglesby wrote: > > Does Debian need/want more testers for older Sun systems like the IPC, > > SPARCStation 1/2/5/20? I have a handful of older systems and a small > > pile of SBus cards, all in working order, and didn't know if setting > > these systems up for testing would be useful or not. > > As far as I'm aware, Debian does not support 32-bit SPARC system since > lenny. Apparently, there is nobody working on the upstream kernel for > 32-bit SPARC machines. It's also my understanding that jettisoning the > 32-bit systems allows Debian to compile for SPARC v9 systems, which can > provide significant performance benefits. > > So, probably not. IIRC support was dropped because no one wanted to maintain them. At one point there was an effort to provide packages for them; this may still exist. If not, if the original poster wanted to set these machines up and work on supporting them again I'm sure there would be greatful users. I suspect it wouldn't be too much effort, esp. for the SPARC V8 based machines. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: X in lenny: 'cannot run in ramebuffer mode'
you inserted the line BusID "SBUS:/SUNW,f...@1e,o" into your xorg.conf - where did you get that, exactly? I want to try the same (and see if it does me any good), but I'm not sure my card has the same busID. I guess, a bug report xserver-xorg-video-all would make sense, would you write one? (since you can probably sdescribe the problem more precisely than me) regards, Martin > This looks like the same error I was getting on my Ultra 10 w/ Creator 3D > card under a fresh Lenny/5.0.1 install. Under 5.0.0 it get an error of > not > having the sunffb module available. I have a thread going on the list > about > it as well. > > Jon. > > On Sun, May 3, 2009 at 5:09 AM, Martin Lemmen wrote: > > > Hi all, > > recently I made a fresh install of lenny on my SPARC. I didn't keep my > old > > xorg.conf. > > Since then, when trying tostart X, I get the error message: > > "Fatal server error: > > cannot run in framebuffer mode. Please specifyb busIDs for all > framebuffer > > devices" > > > > lspci doesn't list anything that sounds like a grafics card. > > The (to my knowledge) relevant part of xorg.conf reads: > > "Section "Device": > > Identifier: "Configured Video device" > > Driver: "sunffb" > > Option "Useffbdev" "true" > > > > I guess I need to build the specific driver for my card (some sun > grafics > > card) into the kernel, but I don't knoew where to find it. Any pointers? > > > > Best regards, Martin. > > > > -- > > Pt! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit > allen: > > http://www.gmx.net/de/go/multimessenger01 > > > > > > -- > > To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org > > with a subject of "unsubscribe". Trouble? Contact > > listmas...@lists.debian.org > > > > -- Pt! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger01 -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
X in lenny: 'cannot run in ramebuffer mode'
Hi all, recently I made a fresh install of lenny on my SPARC. I didn't keep my old xorg.conf. Since then, when trying tostart X, I get the error message: "Fatal server error: cannot run in framebuffer mode. Please specifyb busIDs for all framebuffer devices" lspci doesn't list anything that sounds like a grafics card. The (to my knowledge) relevant part of xorg.conf reads: "Section "Device": Identifier: "Configured Video device" Driver: "sunffb" Option "Useffbdev" "true" I guess I need to build the specific driver for my card (some sun grafics card) into the kernel, but I don't knoew where to find it. Any pointers? Best regards, Martin. -- Pt! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger01 -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: FTBFS on sparc: __sync_test_and_add_4
On Thu, 2009-04-30 at 00:55 +1000, Felipe Sateler wrote:
> El 29/04/09 22:28 Martin escribió:
> > On Wed, 2009-04-29 at 14:11 +1000, Felipe Sateler wrote:
> > > Hi sparc porters. I'm writing to ask for your assistance on a FTBFS on
> > > my package csound on sparc. The build log is here[1]. The failure is
> > >
> > > the following:
> > > > libCsoundAC.so.5.2: undefined reference to `__sync_fetch_and_add_4'
> > >
> > > Csound uses __sync_lock_test_and_set for spinlocks,
> >
> >
> > Given that pthread_spinlock is in one of the POSIX extensions, is there
> > a good reason you are building your own synchronisation primatives?
> >
>
> Hmm, not so unhelpful, given that csound already uses pthreads... maybe this
> function just escaped upstreams attention. Wouldn't it be slower to use
> (shared) library calls instead of gcc builtins?
Correctness is more important than speed. Synchronisation primatives
are notoriously tricky to implement correctly and are prone to obscure,
subtle race conditions. Thus it is best to use the system (libc) ones.
In rare cases where these are a key bottleneck, changing your algorithm
is likely to make a bigger difference to performance and scalability
than changing the implementation of the synchronisation primatives.
> > > and tests for
> > > their existence at build time to use them.
> >
> > This is the following test?
> >
> > Checking for __sync_lock_test_and_set((int32_t *)0, 0) in C library m...
> > yes
> > found sync lock
> Yes. It just builds a .c file calling that function from main and compiles it.
OK, as long as this is the only primative you need, you should be OK.
If you want to use others then you will need to check for them.
Alternatively libatomic-ops would be a good direction to go in, if you
absolutely *must* have your own synchronisation primatives.
> > > Sparc's gcc apparently
> > > provides said function, since the test succeeds, but I'm getting the
> > > above failure (note that __sync_fetch_and_add_4 is nowhere mentioned
> > > in the csound sources).
> >
> > May it come from a supporting library? Which synchronisation primatives do
> > you use?
>
> fel...@pcfelipe:csound% rgrep __sync *
> H/csound.h:while (__sync_lock_test_and_set(spinlock, 1) == 1) {\
> H/csound.h:__sync_lock_release(spinlock); \
This looks like a custom spinlock. I'd (as a general point) advise
against this and suggest using POSIX spinlocks.
> SConstruct: syncLockTestAndSetFound =
> configure.CheckLibWithHeader('m', 'stdint.h', 'C',
> '__sync_lock_test_and_set((int32_t
> *)0, 0);')
> fel...@pcfelipe:csound%
>
> >
> > > What is a possible cause for this? I don't have access to sparc
> > > machines so I'm kind of unsure what to do here. Note that csound uses
> > > -Wl,--as-needed for most of its libraries.
> > >
> > > Please CC me on replies, I'm not suscribed.
> >
> > OK, __sync_* aren't part of the C library as such, they are GCC
> > built-ins that give a (psuedo) machine independant wrapper around atomic
> > instructions. Unfortunately which atomic ops machines implement varies
> > greatly. The GCC interface is kind of written assuming the union of x86
> > and Alpha, in many cases unavailable atomic ops can be emulated.
> >
> > SPARC, certainly early revisions, are very short on atomic operations.
> > I'll have to check an architecture manual but IIRC V7 only has test and
> > set and atomic swap, V9 / V8+ is needed to get compare and swap. I
> > believe the current default is for Debian to target V8. This would be
> > sufficient to pass a check for test_and_set, but not enough to give
> > fetch_and_add, which you'd need compare and swap to emulated.
>
> But I'm not using fetch_and_add... not directly at least.
Yes, which makes me suspect it might be being used indirectly by one of
the libraries that you use.
> > I'd try to isolate which section of code you needs these instructions,
> > see if there is a more portable way of writing that section of code (if
> > it is a library it may be necessary to bug report this back)
>
> How can I go around to do that? I don't have sparc access,
I /believe/ there are (or at leats were at one point) SPARC machines
that were accessible to all Debian developers. Alsa none of my
Debian/SPARC boxes are in a usable state at the moment.
> and my amd64 binaries have no trace of any __sync functions.
IIRC if the target architecture has suitable assembler, it uses that,
references
Re: FTBFS on sparc: __sync_test_and_add_4
On Thu, 2009-04-30 at 00:40 +1000, Felipe Sateler wrote: > El 29/04/09 23:02 Julien Cristau escribió: > > On Wed, Apr 29, 2009 at 14:11:11 +1000, Felipe Sateler wrote: > > > Hi sparc porters. I'm writing to ask for your assistance on a FTBFS on > > > my package csound on sparc. The build log is here[1]. The failure is > > > > > > the following: > > > > libCsoundAC.so.5.2: undefined reference to `__sync_fetch_and_add_4' > > > > > > Csound uses __sync_lock_test_and_set for spinlocks, and tests for > > > their existence at build time to use them. Sparc's gcc apparently > > > provides said function, since the test succeeds, but I'm getting the > > > above failure (note that __sync_fetch_and_add_4 is nowhere mentioned > > > in the csound sources). > > > > Hi, > > > > any reason you're not using libatomic-ops? > > This is the first time I heard about libatomic-ops. Is it cross-platform? I > mean cross-platform as in OS, not arch, which it clearly is. > It appears to be. It is currently distributed as part of Boehm's GC ( http://bdwgc.sf.net/ ), which takes portability quite seriously. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: FTBFS on sparc: __sync_test_and_add_4
On Wed, 2009-04-29 at 14:11 +1000, Felipe Sateler wrote: > Hi sparc porters. I'm writing to ask for your assistance on a FTBFS on > my package csound on sparc. The build log is here[1]. The failure is > the following: > > > libCsoundAC.so.5.2: undefined reference to `__sync_fetch_and_add_4' > > Csound uses __sync_lock_test_and_set for spinlocks, Given that pthread_spinlock is in one of the POSIX extensions, is there a good reason you are building your own synchronisation primatives? > and tests for > their existence at build time to use them. This is the following test? Checking for __sync_lock_test_and_set((int32_t *)0, 0) in C library m... yes found sync lock > Sparc's gcc apparently > provides said function, since the test succeeds, but I'm getting the > above failure (note that __sync_fetch_and_add_4 is nowhere mentioned > in the csound sources). May it come from a supporting library? Which synchronisation primatives do you use? > What is a possible cause for this? I don't have access to sparc > machines so I'm kind of unsure what to do here. Note that csound uses > -Wl,--as-needed for most of its libraries. > > Please CC me on replies, I'm not suscribed. OK, __sync_* aren't part of the C library as such, they are GCC built-ins that give a (psuedo) machine independant wrapper around atomic instructions. Unfortunately which atomic ops machines implement varies greatly. The GCC interface is kind of written assuming the union of x86 and Alpha, in many cases unavailable atomic ops can be emulated. SPARC, certainly early revisions, are very short on atomic operations. I'll have to check an architecture manual but IIRC V7 only has test and set and atomic swap, V9 / V8+ is needed to get compare and swap. I believe the current default is for Debian to target V8. This would be sufficient to pass a check for test_and_set, but not enough to give fetch_and_add, which you'd need compare and swap to emulated. I'd try to isolate which section of code you needs these instructions, see if there is a more portable way of writing that section of code (if it is a library it may be necessary to bug report this back) and failing all else you may want to try setting the architecture to V8+, but be aware this may have portability issues, plus I'm not entirely sure what the current policy on shipping V8+ packages is; it may be necessary to produce two versions, one V8 and one V8+ HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: UltraSparc IIIi kernel and virtualisation
On Mon, 2009-03-16 at 16:42 +0100, Josip Rodin wrote: > On Mon, Mar 16, 2009 at 03:43:41PM +0100, Sander Marechal wrote: > > Question: If openvz is possible under sparc64, why doesn't Debian > > support this? It's supported on other archs. > > > > Even odder, why does Debian supply SMP and non-SMP kernels for sparc64 > > but only a non-SMP kernel for sparc64-vserver and no > > vserver-sparc64-smp? As I understand it now, using the vserver kernel > > supplied for sparc64 would mean that 3 of my 4 CPUs would do nothing > > since it's not an SMP kernel. > > You'd have to check with the debian-kernel mailing list, but odds are that > simply nobody ever asked? :) Check the BTS too. > It used to exist, I used to use it, it used to work - migrated to a new machine about 2-3 years ago and haven't needed it since. The vserver dev teams were very helpful in getting things working and do seem to care about portability. Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Instalado el codio
* Alan Victor <14puer...@gmail.com> [2009-03-06 09:20]: > I finally got Hurd running. QEMU option is very quick and simple. Please use debian-h...@lists.debian.org and not debian-ports. -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Algun usuario en espa ñol de HURD?
* Juan Carlos Canaza Ayarachi [2009-03-04 22:13]: > I am a programmer and want to participate in the project hurd. > I count as 2 hours a day. The correct mailing list to use is debian-h...@lists.debian.org -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Breaking X.Org on sparc (was: Bug#514418: [FIX]: ultra45 boot failing...)
On 2009-02-08 17:33, Jurij Smakov wrote: > Something like this: Added. -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Breaking X.Org on sparc (was: Bug#514418: [FIX]: ultra45 boot failing...)
On 2009-02-08 00:07, Luk Claes wrote: > I think it's best to delay that to r1. Can someone please provide a text > for the release notes to describe the problem, TIA? Hi all: Please send me a paragraph of text ASAP. (Or file a bug with the proposed text against "release-notes".) -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: apt gpg problem dist-upgrade sarge > etch on SS20
On Mon, 2009-01-19 at 22:16 +, gavin duley wrote: > On 17 Jan 2009, at 22:37, Chris Andrew wrote: > > > Hi, all. > > > > I'm trying to do a dist-upgrade from sarge to etch on my SS20. Sadly, > > I am getting a gpg error when apt and other apt utilities try to > > configure. Does anyone know how I can overcome this? > > I'm not a Debian developer, but: I've had a similar problem with my > Sun Ultra 5. Oddly, it seems to be specific to certain mirrors. > Changing to using another mirror often seems to fix the problem. > > It could be just that changing the mirror and doing apt-get update > gets rid of problems with some local cached files? I'm not sure -- as > I said, I'm not a Debian developer and only have a hazy idea of how > Debian archives work. IIRC this is because some of the repository signing keys have expired. If you check back throught the archives you will find a load of posts about this from when it first became a problem. The answer is to manually install the package with the new GPG keys (having first verified, through other means, that they are correct, etc. etc.). As to why this isn't a problem with some mirrors - I guess it depends on if the mirror is signed or not. Not signed, not a problem. Which are signed, of the top of my head I don't know, I guess it may have something to do with how they where created (i.e. if they are Debian mirrors or 3rd party mirrors). HTH Cheers, - Martin -- To UNSUBSCRIBE, email to debian-sparc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Urgent respond
I have a new email address!You can now email me at: mar_tinden...@ymail.com Sir/Madam, Ive business to discuss with you, please contact me, for more details Dent - Martin Dent
Re: Debian on Sun box any advantage?
On Wed, 2008-10-22 at 13:18 -0700, John Bevins wrote: > Hello Everyone, > > I don't know much about Sun boxes. What advantage, if any, is there to > running Debian on a Sun box as opposed to a brand name x86 server? As the question is phrased I think it is unanswerable. It depends what Sun hardware you are comparing and what x86 hardware you are comparing it with. At the higher end, Sun kit is available in 'wider' parallel configurations (i.e. the recently released 256 way UltraSPARC T2+ based boxen), you will also likley find more server orientated features on Sun boxen, for example ECC RAM, more redundancy, network management hardware, etc. But these are general trends. For a long while Linux on SPARC hardware was (to my impression, please do correct me if I am wrong) mainly run by Linux users who had acquired Sun hardware second hand and wanted to make use of it. I am not aware of many cases of people purchasing new Sun hardware just to run Linux (althought I could be mistaken). This is beginning to change as I believe Linux is now officially supported by Sun on UltraSPARC T based hardware as a guest OS. If you are looking to buy new hardware, then get the manufacturers to give you the sales spheel about why you should buy from them. If you have spare SPARC based hardware, are looking to put it to use and don't want to run Solaris*, then Debian/SPARC is a good solution. HTH Cheers, - Martin *. You may find hardware support, particularly for higher end and more obscure hardware, is better on Solaris, however, there is more the a choice of OS than just drivers. YMMV. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: free(): invalid nextd size?
Hi, On Mon Sep 22, 2008 at 02:41:06 -0500, Steven Robbins wrote: > Hi, > > The Sparc Buildd failed to build ITK [1], aparently because the build > takes too long (see below). Or does the first line indicate something more > nefarious? > > Would the buildd owner please re-try ITK? > > Thanks, > -Steve > > Excerpt from [1]: > > *** glibc detected *** /usr/lib/gcc/sparc-linux-gnu/4.3.1/cc1plus: free(): > invalid next size (fast): 0x0098d7a0 *** > make[3]: *** > [Wrapping/CSwig/Algorithms/CMakeFiles/_ITKAlgorithmsPython.dir/wrap_itkHistogramMatchingImageFilterPython.o] > Terminated > make[2]: *** > [Wrapping/CSwig/Algorithms/CMakeFiles/_ITKAlgorithmsPython.dir/all] Terminated > make[1]: *** [all] Terminated > make: *** [debian/stamp-makefile-build] Terminated > Build killed with signal 15 after 150 minutes of inactivity > > > [1] > http://buildd.debian.org/fetch.cgi?&pkg=insighttoolkit&ver=3.8.0-1&arch=sparc&stamp=1220273077&file=log it could be a problem with the processort (being Ultra III). I will requeue on spontini, which is a Ultra II Greetings Martin -- Martin Zobel-Helas <[EMAIL PROTECTED]> | Debian System Administrator Debian & GNU/Linux Developer | Debian Listmaster Public key http://zobel.ftbfs.de/5d64f870.asc - KeyID: 5D64 F870 GPG Fingerprint: 5DB3 1301 375A A50F 07E7 302F 493E FB8E 5D64 F870 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sparc Live CD
On Tue, 2008-06-03 at 20:51 -0400, Eric Nichols wrote: > I would like to build a live CD for my Ultra 2 that will act as a static web > server. Basically everything would work in memory. The perfect secure > server. For easy of maintenance I would have thought it would be worth exporting (read only) some of the filesystem from another machine. That way updating the website / reconfiguring the web server can be done by simply editing a file - you don't need to burn a new CD. Also you may wish to log to a network filesystem or set up syslog to log to another machine. If you have a look at the LSB you will see which parts of the filesystem have to be writable and which can be mounted read only. > However I have having difficulty trying to figure out how to build one using > debian tools. dfsbuild looked like the obvious choice but it does not support > sparc. I talked with the maintainer and the reason that sparc is not support > is that is he simply does not have the hardware but was willing to accept > patches to the project. > > I honestly am just an end-user in this endeavor but am willing to test > anything. > Can anyone assist? Given that the install CDs are (basically) minimal live CDs, thus is probably just a case of building things, little to no actual 'porting' is likely required. You're probably best off picking a script / program / system that can either build an image from a running system or from debian packages. I believe a number of these exist, a few years ago I the knoppix one was quite highly rated IIRC. Beyond that - dive in and give it a go. You are more likely to be able to find material / documentation / assistance for any detailed problems you encounter, then you are to be able to find general help. One possibility would be to first build the image you want for x86 (documenting in detail what you did) and then try the same thing for SPARC. This should help separate problems with the tools from portability issues. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Netra X1 install problem/oddity
I remember seeing something like this as well. Seems like a bug in partman. The only way I got around it is to partition the disk using fdisk. Getting fdisk on the machine was the tricky part, it was not on my installation media. I put the executable on another server, and copied that over from the install shell. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as MERT/UNIX is today. - Martin Habets --- On Sat, May 31, 2008 at 04:27:57PM -0700, Gary W. Smith wrote: > Hello, > > I've been working to get Debian current installed on a Netra X1 from a > netboot over LOM. I first encountered the problem with the tulip driver > (requiring me to shell out, unload dmfe/tulip and then load tulip). As > is well on the network front. The problem is after it downloads all of > the install from ftp, it comes up and states that it cannot find any > disks and asks me to load a driver. > > So I shell again and I can see /dev/hda in /proc/partitions (20gb). I > run partman from the shell and it sees no drives. Thinking that it's > because of the former Sun partition, I dd if=/dev/zero of=/dev/had bs=1M > count=1 to clear any partitions. I then run partman and it now sees a > 1MB disk drive. > > Any ideas on why this might be happening and how I can work around it. > I have several of these machines that I'd like to build out in the next > couple days for a project, but they just aren't cooperating. > > Gary -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Can Etch be installed on an SS20?
If memory serves me right cdrom is for an IDE CDROM. You'll need 'sr_mod' for SCSI CDROMs. Apart from lsmod, run dmesg to see what devices esp detected. Martin On Thu, May 29, 2008 at 07:52:48PM +0100, Chris Andrew wrote: > Hi, guys. > > Just trying to install Etch on an SS20. The CD boots, and I select my > keyboard etc, but when it comes to detect the CD properly, it can't be > found. Can anyone remember the solution to this? lsmod indicates > that the esp module and cdrom modules are loaded. I think in the past > the problem has been esp, but I'm sure I got further than this. > > Can anyone help? > > Cheers, > > Chris. > > 2008/4/23 Michelle Konzack <[EMAIL PROTECTED]>: > > Am 2008-04-22 20:35:13, schrieb Chris Andrew: > >> Hi, everybody. > >> > >> I know the Lenny will not support 32-bit Sparc Stations. Has anyone > >> successfully booted Etch from a CD and done an install? > >> > >> Cheers, > >> > >> Chris. > > > > I have eight SS10/512 (Dual-CPU) and one SS20 and the last Release was > > Sarge and failed with Etch. Some Kernel related problems... > > > > But if you are happy with an older Kernel, you can roll your own one > > (the one from Sarge) and use it under Etch which should work nicely. > > > > Be carefull, if you upgrade from Sarge to Etch, since you will run into > > trouble, if your Kernel NEED modules... > > > > I had to compile all needed modules into the kernel. > > > > Thanks, Greetings and nice Day > >Michelle Konzack > >Systemadministrator > >24V Electronic Engineer > >Tamay Dogan Network > >Debian GNU/Linux Consultant > > > > > > -- > > Linux-User #280138 with the Linux Counter, http://counter.li.org/ > > # Debian GNU/Linux Consultant # > > Michelle Konzack Apt. 917 ICQ #328449886 > > +49/177/935194750, rue de Soultz MSN LinuxMichi > > +33/6/61925193 67100 Strasbourg/France IRC #Debian (irc.icq.com) > > > > > > -- > Reasons why you may want to try GNU/Linux: > > http://www.getgnulinux.org/ > > A great GNU/Linux distro: > > http://wiki.gnewsense.org/ > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- Martin --- 30 years from now GNU/Linux will be as redundant a term as MERT/UNIX is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#477741: libauthen-dechpwd-perl_2.002-3(sparc/unstable): FTBFS, test failed
Hi, On Thu Apr 24, 2008 at 17:27:39 -0700, Ivan Kohler wrote: > severity 477741 important > tags 477741 help > thanks > > FTBFS appears to be specific to sparc architecture. > > I tried to log onto sperger.debian.org and at least look further into > the problem, but neither debhelper nor libmodule-build-perl are > installed. I'll email debian-admin@ and ask. [sid] sperger:~# apt-get build-dep libauthen-dechpwd-perl Reading package lists... Done Building dependency tree Reading state information... Done 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. -- Martin Zobel-Helas <[EMAIL PROTECTED]> | Debian Release Team Member Debian & GNU/Linux Developer | Debian Listmaster Public key http://zobel.ftbfs.de/5d64f870.asc - KeyID: 5D64 F870 GPG Fingerprint: 5DB3 1301 375A A50F 07E7 302F 493E FB8E 5D64 F870 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sun Leo (ZX)
On Wed, Mar 12, 2008 at 02:41:25PM +, Mark Morgan Lloyd wrote:
> Martin Habets wrote:
> >Does the kernel detect the card when booting? You should see something
> >like the following when booting:
> >/[EMAIL PROTECTED],0/SUNW,[EMAIL PROTECTED],0: leo at x:x
> >Also you should have leo ranges in /proc/iomem and /proc/ioport.
>
> Ahhh... no, I don't see that in dmsg or anything in those proc files. I can
> see the driver enabled in .config and I think that I'm running the
> appropriate kernel but it's obvious that I need to go and check.
Looking in drivers/video/leo.c leo_init() I see
if (fb_get_options("leofb", NULL))
which leads me to suggest booting with option 'video=leofb:on'.
> Am I correct in interpreting what you've said as meaning that X's "No
> devices detected" actually means "I'm not seeing the appropriate device
> support in the kernel"?
That's one option. It could also be the wrong device was specified in the
Xconfig.
> >Seems this driver creates a normal fb device. Make sure you have /dev/fb0
> >and/or /dev/fb0autodetect.
> >You could try the fbdev X driver:
> > Driver "fbdev"
> >without specifying a BusID.
> >
> >This is just to see if it detects a device. If it does the issue is with
> >the
> >leo X driver. If it does not my guess it more toward a kernel issue.
>
> Noted. Unfortunately the disc attached to the machine has started failing
> and is rapidly getting worse so that's something that's got to be tackled
> before I can progress this much further.
Yes, obviously. But if Tom's right (and he usually is) the fbdev trick
won't work.
Good luck,
Martin
> I'll be back, but I might be some time...
>
> >Martin
> >
> >On Sun, Mar 09, 2008 at 07:17:49PM +, Mark Morgan Lloyd wrote:
> >>Jurij Smakov wrote:
> >>>On Sun, Mar 09, 2008 at 04:01:40PM +, Mark Morgan Lloyd wrote:
> >>>[...]
> >>>>(II) SUNLEO: driver for Leo (ZX)
> >>>>(EE) No devices detected.
> >>>>
> >>>>Fatal server error:
> >>>>no screens found
> >>>>
> >>>>I've tried several combinations of BusID etc. without success.
> >>>Can PROM/prtconf see this device? I guess that would give you the
> >>>correct location, if any.
> >>>
> >>>Best regards,
> >>Yes, print-devs (?) shows it as /[EMAIL PROTECTED],0/SUNW,[EMAIL
> >>PROTECTED],0 and I can see it
> >>in the /sys tree.
> >>
> >>lssbus doesn't work on this system but I've never tried running it on 2.6
> >>before- I need to check whether there's an updated version.
> >>
> >>I don't believe there's a hardware problem since it displays the Sun logo
> >>with a Lion sitting on it when video first comes up, I do note however
> >>that it's a rather magenta Lion and that console text is very slow.
> >>
> >>This is a sacrificial development machine and I'm quite happy to
> >>experiment on it. Which is fortunate since I tried installing Lenny
> >>earlier and am now getting "Fast data access MMU miss" [sigh].
>
> For the record, that last was a red herring- I was saying "?" to OpenPROM
> rather than "help".
>
> --
> Mark Morgan Lloyd
> markMLl .AT. telemetry.co .DOT. uk
>
> [Opinions above are the author's, not those of his employers or colleagues]
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact
> [EMAIL PROTECTED]
--
Martin
---
30 years from now GNU/Linux will be as redundant a term as MERT/UNIX is
today. - Martin Habets
---
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sun Leo (ZX)
Does the kernel detect the card when booting? You should see something like the following when booting: /[EMAIL PROTECTED],0/SUNW,[EMAIL PROTECTED],0: leo at x:x Also you should have leo ranges in /proc/iomem and /proc/ioport. Seems this driver creates a normal fb device. Make sure you have /dev/fb0 and/or /dev/fb0autodetect. You could try the fbdev X driver: Driver "fbdev" without specifying a BusID. This is just to see if it detects a device. If it does the issue is with the leo X driver. If it does not my guess it more toward a kernel issue. Martin On Sun, Mar 09, 2008 at 07:17:49PM +, Mark Morgan Lloyd wrote: > Jurij Smakov wrote: > >On Sun, Mar 09, 2008 at 04:01:40PM +, Mark Morgan Lloyd wrote: > >[...] > >>(II) SUNLEO: driver for Leo (ZX) > >>(EE) No devices detected. > >> > >>Fatal server error: > >>no screens found > >> > >>I've tried several combinations of BusID etc. without success. > > > >Can PROM/prtconf see this device? I guess that would give you the > >correct location, if any. > > > >Best regards, > > Yes, print-devs (?) shows it as /[EMAIL PROTECTED],0/SUNW,[EMAIL PROTECTED],0 > and I can see it in > the /sys tree. > > lssbus doesn't work on this system but I've never tried running it on 2.6 > before- I need to check whether there's an updated version. > > I don't believe there's a hardware problem since it displays the Sun logo > with a Lion sitting on it when video first comes up, I do note however that > it's a rather magenta Lion and that console text is very slow. > > This is a sacrificial development machine and I'm quite happy to experiment > on it. Which is fortunate since I tried installing Lenny earlier and am now > getting "Fast data access MMU miss" [sigh]. > > -- > Mark Morgan Lloyd > markMLl .AT. telemetry.co .DOT. uk > > [Opinions above are the author's, not those of his employers or colleagues] > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact > [EMAIL PROTECTED] -- Martin --- 30 years from now GNU/Linux will be as redundant a term as MERT/UNIX is today. - Martin Habets --- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [Ann] SparcV8 depot
Hi Joël, I've got an SS20 with dual hypersparc. Kernels post 2.6.21 boot okay, run in general, but a 'make -j 3' build of the kernel aborts a couple of times with a bus error or similar vague errors. I'd be interested in a V8 repository. Martin On Sat, Feb 16, 2008 at 10:34:03AM +0100, BERTRAND Joël wrote: > Hi folks, > > I have built a 2.6.24.2 linux kernel on my SS20 (dual SM71) and it > is stable in SMP configuration with HIGHMEM. I haven't tested with > HyperSPARC. Has anyone any information about HyperSPARC support ? I > shall test, but if someone has tried before me... ;-) > > I'm building deb packages to upgrade this workstation to > debian/testing (with of course only V8 support not V8+). I don't know > if > someone can be interested, but I think we can open a depot for sparc-V8 > package. I think I shall open an anonymous ftp server next week, but I > won't do anything if I'm alone. > > Regards, > > JKB -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: data overrun detected in Data-in phase
On Thu, 2008-01-31 at 18:27 +0100, BERTRAND Joël wrote: > Hello, > > I'm using a Adaptec 29160 in a U60/SMP workstation. For 2.6.23.11, I > see these messages in log file : > > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x12. > (scsi2:A:4:0): Have seen Data Phase. Length = 24576. NumSGs = 1. > sg[0] - Addr 0x0c087e840 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x42. > (scsi2:A:4:0): Have seen Data Phase. Length = 8192. NumSGs = 1. > sg[0] - Addr 0x0c087e820 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x13. > (scsi2:A:4:0): Have seen Data Phase. Length = 8192. NumSGs = 1. > sg[0] - Addr 0x0c087e800 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x41. > (scsi2:A:4:0): Have seen Data Phase. Length = 24576. NumSGs = 1. > sg[0] - Addr 0x0c087e100 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x7a. > (scsi2:A:4:0): Have seen Data Phase. Length = 8192. NumSGs = 1. > sg[0] - Addr 0x0c087e7c0 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0xb. > (scsi2:A:4:0): Have seen Data Phase. Length = 8192. NumSGs = 1. > sg[0] - Addr 0x0c087ede0 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x54. > (scsi2:A:4:0): Have seen Data Phase. Length = 8192. NumSGs = 1. > sg[0] - Addr 0x0c087e180 : Length 32 > (scsi2:A:4:0): data overrun detected in Data-in phase. Tag == 0x30. > (scsi2:A:4:0): Have seen Data Phase. Length = 8192. NumSGs = 1. > sg[0] - Addr 0x0c087e160 : Length 32 > > I don't understand this message. Disk seems to be good. This device is > a part of a 7-disk raid6 array. I don't find any information about this > message. Any idea ? It's probably not much help but problems with cabling and termination, particularly on buses with lots of disks can create lots of seemingly obscure errors. Another option would be to temporarily remove the disk with ID 4; which should tell you if it's a disk specific problem. HTH Cheers, - Martin
Re: Netboot E250 -- tftp problem
On Wed, 2008-01-23 at 09:38 +0800, Hero_xbd!.RRR wrote: > > This seems vaugely familiar. I have a vauge memory of having problems > > getting a Sun machine to netboot and finding out it was a cabling > > problem. Have you tried swaping the Sun machine for another boxen and > > pulling the netboot file from the server? This should tell you if it's > > a cabling / switch problem. > > > I have successfully pulled the files through TFTP to my PC in dormitory > (TFTP server is at lab). Does it infer there is no problem here? That depends on the architecture of the network. If it's something like this: PC in dorm <--> ... some network ... <--> switch <--> server ^ | | V E250 (Where arrows are ethernet cables) then you will have tested all but the cable that goes from the switch to the E250. If your topology is different then what you have tested will be different. HTH Cheers, - Martin PS If you get really stuck netbooting, you could pull the disks, dump a minimal filesystem on them and then boot from disk. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Netboot E250 -- tftp problem
On Tue, 2008-01-22 at 21:31 +0800, Hero_xbd!.RRR wrote: > I am using fail2ban[1] on the TFTP server, which functions on iptables > and is supposed to modify only "ftp ftps ftp-data ftps-data ssh"ports. > Except it, I did not do anything with firewall. Apologies if I'm telling you something you already know but that would mean your TFTP server isn't protected. > > Oh, maybe I should use a shiny minimum debian mechine to act as the TFTP > server. That will eliminate various potential blocks. > > Is it possible the NICs and Net Switch produce the problem? Does the > working rarp session eliminate all the possibility? This seems vaugely familiar. I have a vauge memory of having problems getting a Sun machine to netboot and finding out it was a cabling problem. Have you tried swaping the Sun machine for another boxen and pulling the netboot file from the server? This should tell you if it's a cabling / switch problem. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Interesting apt problem during update
On Mon, Oct 29, 2007 at 10:51:40PM +0100, Ulrich Teichert wrote: > Hi, > > >Unfortunately this didn't solve my problem. I tried many difference > >options from the apt man pages, but kept getting a libc6 (I think) > >error. This is provided by apt, which can't be upgraded. I couldn't > >install a 2.6 kernel without this. Run 'apt-get check' to see what is broken. If you feel up to it you can try an 'apt-get -f' as well, depending on what check reports. > Be happy. My box is unbootable right now: > > SPARCstation 20 MP (2 X 390Z55), No Keyboard > ROM Rev. 2.15, 192 MB memory installed, Serial #3528342. > Ethernet address 8:0:20:21:68:97, Host ID: 7235d696. > > Rebooting with command: > Boot device: /iommu/sbus/[EMAIL PROTECTED],40/[EMAIL > PROTECTED],80/[EMAIL PROTECTED],0:a File and args: > SProgram terminated > Type help for more information > <#0> ok dis > 7 Memory Address not Aligned > > Looks like a broken silo to me. Ran into the same issue 2 weeks ago. I thought I'd just been hacking away too much. :) Got around this by net-booting a kernel and running silo -f. Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Any progress?
I don't see how the users are getting lost. Almost all emails clearly describe on what kind of machine the issue is seen. That immediately makes it clear wether it's sparc 32 or 64. For those interested in sparc32, a usefull mailing list to subscribe to is [EMAIL PROTECTED] With all respect for the good intentions, we don't need another ML. Martin On Mon, Oct 08, 2007 at 08:26:20PM +0100, Chris Andrew wrote: > Julien, > > You may be right, but it seems hat sparc (32) users are understandably > getting lost, because the interest is in sparc64. If nobody is > interested (or are unable) to work on 32-bit stuff, then maybe a new > list is what is needed. Hence the point about Debian having a > separate sparc (32) port. > > I'm reluctant to say more, as the sparc developers will tell me that I > can say this as much as I want, but it won't change the fact that > nobody is able to support 32-bit. > > Cheers, > > Chris. > > On 08/10/2007, Julien Cristau <[EMAIL PROTECTED]> wrote: > > On Mon, Oct 8, 2007 at 19:26:49 +0100, Chris Andrew wrote: > > > > > Hi, Julien. > > > > > > People have mentioned sparc (32) becoming a separate port on Debian, > > > so that we can rally the troops and make a concerted effort. This > > > idea seems not to be of any interest, so I think that it is good that > > > this list has been created. We need to get a greater readership, > > > though. Not sure how. > > > > > Right, and I'm saying that using a new list with no subscribers isn't > > the greatest idea if you want to get a broader readership. IMHO, YMMV, > > etc. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SunRays
On Wed, 2007-09-26 at 16:32 -0500, Alejandro Paredes wrote: > Hello everybody. > > I've been trying to find out if it is possible to use Debian in a Sparc as > the manager of several SunRays, but no luck so far. > Has anyone done it? If so could you give me some pointers on how to do it. The earlier Sun thin clients basically used remote X and so could eaily be presuaded to run without Solaris. The SunRays are a bit different. As I understand it, they are closer to screen scraping approaches such as VNC. Thus they talk a custom protocol that includes graphics, key strokes, etc. Although the server app for this can be run on Linux (apparently) I am not aware of the protocol being documented nor there being any non commercial implementations. At one point the licences for the server app were expensive, I have no idea if this is still the case, they may be free (as in beer) for non commercial use. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: non-SMP Kernel on Sparc machines using dual-cpu boards
On Sun, 2007-09-23 at 02:48 +0200, Bernd Zeimetz wrote: > On which machines does the kernel boot properly? > Remember that UltraSPARC II processors have a different architecture > and UltraSPARC IIi and IIIi also are different! Please forgive me if I am stating the obvious but (as I understand it) the main difference between the II and IIi (and respectively the III and IIIi) is that the i series are for single processor machines and have a lot of the cache control and memory consistancy hardware removed as it is not needed. Thus running an UltraSPARC III and an UltraSPARC IIIi with non SMP kernels are fundamentally different propositions. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sun Ultra10
If someone is interested in a Sun Ultra10 in the Netherlands, please contact [EMAIL PROTECTED] * Casper van der Bilt <[EMAIL PROTECTED]> [2007-09-20 23:07]: > I have a number of spare Ultra 10 machines that I would like to > donate to open source projects. At least 1 is going to NLnet Labs > (www.nlnetlabs.nl) but there are several more. > > If there are debian sparc developer in the Netherlands you know of > that can use an Ultra 10 you can bring them in contact with me. Any > other OpenSource projects are also good. > > Best regards, > > Casper van der Bilt -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Black screen on sunfire v240
Try booting with the -p option. -- Martin --- 30 years from now GNU/Linux will be as redundant a term as MERT/UNIX is today. - Martin Habets --- On Wed, Sep 12, 2007 at 05:23:39PM -0700, Vidblain Amaro wrote: > I}v tried all the things that you told me, the fb=true, etc, but i don't get > nothing different, my screen is in black when °booting linux ... ° i only > have one graphic card, can somebody help me with this problem, i need set > debian in the sunfire v240 black and the sunfire don't do nothing. > > Or somebody can explain me > > Can somebody help me! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: audioctl
Debian is using ALSA, you use alsamixer or amixer. Martin On Mon, Sep 03, 2007 at 10:24:33AM +0200, BERTRAND Joël wrote: > Hello, > > How can I select sound output between internal speaker and line? > With Debian 3.1, there was audioctl, but this program is not provided > by > etch or lenny ? > > Thanks, > > JKB -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Creation of Sparc (32-bit) port.
On Thu, 2007-08-30 at 18:43 +, Markus Dahms wrote: > Hi all, > > Adrian von Bidder wrote: > > Chris Andrew: > >> I know that discussion is on-going regarding the future of 32-bit > >> sparc. > > AFAICT the discussion is not ongoing. It just repeats itself every few > > weeks, and nothing changes: There is no kernel support for sparc32 > > anymore in Linux. > > excuse me, but what exactly is the problem with the sparc32 kernel? I > (cross-)compiled a current git version (2.6.23-rc4+e) and stress tested > it a little on my SparcStation 5. I got the load average up to 16+, got > the oom-killer to kill some processes (no wonder when allocation 8 x 256M > on a box with 128M physical + 170M swap), but the machine still works as > expected (ok, uptime is now only 5:38h, but it's a start). > Userland is Debian Etch. Judging by what I remember of the past discussion (check the archives), the problems are more with SMP boxen and the hyperSPARC processors. That said, IIRC, there was a problem with pipes even on UP machines. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sparc (32) kernel development.
On Tue, 2007-08-21 at 22:36 +0100, Chris Andrew wrote: > Hi, all. > > If I was interested in contributing to maintaining the kernel for > sparc (32), where would be the best place to go, to get involved? Any > good mailing lists for kernel newbies, specific to sparc (32)? [EMAIL PROTECTED] http://www.ultralinux.org/lists.html is the main SPARC kernel mailing list, you probably also want to keep an eye on what's going on on the LKML [EMAIL PROTECTED] http://kernel.org/ although it is fairly high traffic and you may want to read a summary instead: http://www.kernel-traffic.org/ http://lwn.net/Kernel/ http://www.kerneltrap.org/ there is also a dedicated kernel newbies list / site: http://kernelnewbies.org/ (also see http://kernelnewbies.org/KernelBooks for some of the books available) http://www.sparc.org/ has the architecture specifications, which may be of use. http://lxr.linux.no/source/ Is the output of LXR, which I find a very approachable interface to browsing kernel code, although I know some prefer cscope(1). Beyond that I think it's a case of finding a bug, fixing it and repeating. Personally I find the best way of learning a new piece of software is to pick a task and start hacking. HTH and good luck. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Creation of Sparc (32-bit) port.
On Tue, 2007-08-21 at 23:22 +0100, Chris Andrew wrote: > All, > > It seems that people are very keen to see 32 bit support die. I wouldn't have said that. When it was first announced that a new maintainer for the 32 bit kernel was needed, or failing that sparc32 support would have to be dropped, there were lots of people who said they felt sad by the prospect of loosing support. Don't interpret silence as active hostility, perhaps no one knows. > Who do > I have to contact to get a separate 32 bit sparc port set-up? As > suggested before, we can give the new port 6 months, and see what > happens? At the moment, 32 bit future is not looking good, at least > let's go out trying! Why not just start? As long as the main userland pacakges are still being built for V8 rather than V9 you can use them so all you need to do is produce the few sparc32 specific packages (i.e. the kernel). Simply set up a Debian repository ( http://www.debian.org/doc/manuals/repository-howto/repository-howto ) and start creating packages. Then post the URL to the list and get people to test them out. Once you've got a set of packages that increase the functionality of the sparc32 port I would have thought that getting people to test and to help debug would be much easier and once you have users acquiring a name and space on official Debian servers shouldn't be difficult. Just go for it. Cheers, - Martin PS FWIW, I haven't used sparc32 in quite a while but I would be sad to see it go, as I posted when the topic first came up, but as I am not in a position to contribute any time to maintaining it, I don't feel I can blame the porters for not wishing to spend their free time doing so. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Creation of Sparc (32-bit) port.
On Tue, 2007-08-21 at 17:41 -0400, Clint Adams wrote: > On Tue, Aug 21, 2007 at 12:39:16AM +0100, Martin wrote: > > As far as I know nothing has happened for precisely the same reason that > > sparc32 had to be dropped - there isn't anyone to work on the kernel. > > Without that then a separate port isn't going to get too far, but if > > there was some one / some people who were willing to work on it, there > > wouldn't need to be a separate port... > > Actually, there would be unless you add additional CPU emulation to > Linux. Sorry - I don't follow. Presumably you are making reference to emulating the few instructions that are in SPARC V9 but not in SPARC V8. In my opinion that's a fairly minor issue. Last I looked the sparc binaries were being built for V8 and the libraries in which it would make an appreciable difference have V9 versions available. Essentially /if/ there was someone who was willing to handle kernel development for sparc32, things could (more or less) continue as they have been. At least, that's my impression. > Chris is suggesting the only viable option. I don't think so, see above. The one port / two kernel approach has been working well, an extra port seems unnecessary. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Creation of Sparc (32-bit) port.
On Mon, 2007-08-20 at 22:16 +0100, Chris Andrew wrote: > Hi, all. > > I know that discussion is on-going regarding the future of 32-bit > sparc. > > I heard it suggested that a separate port could be set-up for 32 -bit > sparc, permitting maintainers and supporters to pool resources and > work-out whether 32-bit support is possible. This would also give > sparc64 the chance to be optimised, without worrying about 32-bit. > > Does anybody know whether any plans have been made for this to happen? As far as I know nothing has happened for precisely the same reason that sparc32 had to be dropped - there isn't anyone to work on the kernel. Without that then a separate port isn't going to get too far, but if there was some one / some people who were willing to work on it, there wouldn't need to be a separate port... Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: pure 64-bit SPARC tool chain
On Tue, 2007-08-07 at 11:32 +0200, Sunil Amitkumar Janki wrote: > Does anyone have a pure 64-bit SPARC tool chain or know > where one can be downloaded? I have a Sun Enterprise 250 > server containing 2 US II processors and it will only boot > 64-bit kernels. > > To that end I need a SPARC64 compiler but so far I have not > succeeded building that. At the moment I have Splack > (Slackware for SPARC) running but I will install Debian on > a spare hard disk. > > Even though it's not really beneficial to run a pure 64-bit > environment on SPARC I would still like to have a basic > 64-bit tool chain to experiment with. Is gcc -m64 sufficient for your needs? You may also need libc6-dev-sparc64 and (if available) any other libraries you wish to use, compiled as 64 bit binaries. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Webalizer weirdness
On Mon, 2007-07-30 at 20:16 +0100, andrew holway wrote: > Hi, > > I've been asking in IRC and forums for some time but no one has been > able to help. > > Has anyone any ideas why this? > http://www.moonet.co.uk/webalizer/usage_200707.html#TOPSITES > > log is here http://www.moonet.co.uk/logs/access.log > and config is here http://www.moonet.co.uk/webalizer/webalizer.conf > webalizer is Webalizer V2.01-10 (Linux 2.6.18-028stab031.dev-ovz031) English > floating point conversion error? > > Linux daisy.moognu.co.uk 2.6.18-028stab031.dev-ovz031 #2 SMP Wed May > 30 10:45:50 BST 2007 sparc64 GNU/Linux Were all of the input data files generated on sparc64? I have seen strange things happen when parsing floating point numbers written on x86 machines (as ASCII). I always assumed it was something to do with the x86 support for 80 bit long doubles (and the sometimes iffy support for 64 bit doubles) verses the SPARC support of 64 bit and (nominally) 128 bit doubles. HTH, Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: purse strings
On Wed, 2007-07-25 at 14:58 +0100, andrew holway wrote: > Hi, > > If a large organisation is thinking about purchasing a HPC machine who > would make the final decision. Who would you be pointing your > marketing towards? > > Do you think this differs from academia to the commercial sector? At the risk of sounding unfriendly; this really isn't on topic for this list. You might get a better response from something like debian-consultants or one of the other more business orientated lists. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: linking SPARC applications
On Sat, 2007-07-21 at 22:38 +1000, Jim Watson wrote: > Martin wrote: > > On Sat, 2007-07-21 at 16:53 +1000, Jim Watson wrote: > > > >> Hi, > >> > >> I have been building SPARC applications - I mean the shared libraries > >> report SPARC in response to the file command. > >> > >> Now I noticed these are linked to various standard libs in /lib/v9 which > >> report SPARC32PLUS, instead of linking to the required SPARC libs in /lib > >> > >> Is this something that has to be configured when building or is it > >> something happens when running ldd? (I only have sun4u here) > >> > > > > IIRC the output of ldd is something like > > > > libraryNameInProgram => /where/it/maps/to/on/this/system > > > > if the first is incorrect, you have to change you build system / > > environment. If the second is incorrect you have to change how ld is > > set up on your system. > > > > Thus it would sound, to me, like your system was behaving correctly and > > the binaries should work on sun4m. > > > Yes, but I like them to run on sparc (32) systems too. sun4m are the only 32 bit sparc systems supported by Debian at the moment as far as I know. sun4 have been unsupported for a long time, sun4c where dropped during the development of sarge IIRC and sun4d has kernel issues and hasn't been supported for a while (and when they do, they work much as sun4m). > Previously they always linked to /lib/blah. > Now something has changed so they now link to /lib/v9/blah > But I didn't change my build system so I guess it is something changed > in debian. > Thats the bit I don't understand - what has changed? If the problem definately is with what ld maps the libraries to then it is likely to be a problem / change in ld. I would guess that someone added a hack to the binary or the config file so that on sun4u it attempts to load /lib/v9/ before /lib/. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: linking SPARC applications
On Sat, 2007-07-21 at 16:53 +1000, Jim Watson wrote: > Hi, > > I have been building SPARC applications - I mean the shared libraries > report SPARC in response to the file command. > > Now I noticed these are linked to various standard libs in /lib/v9 which > report SPARC32PLUS, instead of linking to the required SPARC libs in /lib > > Is this something that has to be configured when building or is it > something happens when running ldd? (I only have sun4u here) IIRC the output of ldd is something like libraryNameInProgram => /where/it/maps/to/on/this/system if the first is incorrect, you have to change you build system / environment. If the second is incorrect you have to change how ld is set up on your system. Thus it would sound, to me, like your system was behaving correctly and the binaries should work on sun4m. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Debian on SS20.
On Thu, 2007-07-19 at 20:40 +0100, Chris Andrew wrote: > All, > > Well, I thought I'd got Debian running on my SS20. Unfortunately when > I install Sarge, at the end of the install, when Debian tries the > internet for updates, it for some reason wants to remove my SMP kernel > (I have two processors), and as far as I can see, I am not left with > another kernel. > > In addition to this, I get 3 apt related error messages, because the > packages cannot be verified due to a lack of signature. If you want help with this you will likely have to say which three packages these are. > Finally, I tried to do a dist-upgrade to try to remedy this, and I got > the same problems, but also, it doesn't seem like a 32 bit sparc smp > kernel exists. See the archives for the history / status of sparc32/SMP. In short, it's problematic. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Can't use all of my disk in my SS20.
On Wed, Jul 18, 2007 at 04:13:56PM +0100, Chris Andrew wrote: > Hi, all. > > I have succesfully installed Etch on my SS20. I have 2 harddrives, both > 2.1Gb. > > The first HDD will only let me use 200 megs of it. I have tried all sorts > of formatting, but it just won't let me touch most of the drive. When I do > fdisk /dev/sdb, it just shows the two partitions that I have created, plus > the Sun "whole disk" partition. I still can't do anything with almost 2 > gigs of HDD. Can anyone tell me how I can get around this? I ran into this when I used parted to partition the disk. Could not complete the install as a result. Got around it by using good ol' fdisk in stead. Mind you, it only happened on 1 machine out of 5. I think all the others had bigger disks. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Can't use all of my disk in my SS20.
On Wed, 2007-07-18 at 16:52 +0100, Chris Andrew wrote: > Martin, > > With regard to the Etch install, I installed Sarge, which was fine. i > then did a dist-upgrade to Etch. > > Previously, I had been getting "File does not appear to be > executable", on my first reboot. I had a bit of a long-shot and > installed debian on my 2nd HDD. My two disks were labelled something > like disk 0 and 1, when I picked where to install debian. I was > making the mistake of thinking that the default boot was disk 0. This should be configurable in OpenBoot. > I was wrong. When I installed on the 2nd disk, everything worked > just fine. Hence my install of Etch, which is just finishing. Cool. > With regard to the second part of your message, copying the disk > geometry etc, yo totally lost me, I'm afraid. It's a lot easier than that. /If/ the problem is with the disk label on the second disk, then something like: 1. Copy the data off the "200Mb" disk cp -arv /mnt/200Mb /mnt/somewhere-safe/ 2. Zero the start of the disk (thus overwriting the disk label) (remember to unmount first though...) dd if=/dev/zero of=/dev/200Mb count=100 3. Use fdisk to build a new disk label and make sure the values are correct for the disk (real ones should be given in dmesg when the disk is recognised, or maybe in /proc/, or probably printed on the disk) fdisk /dev/200Mb > Hope this helps the archive. Hopefully it will help if someone has the same problem and searches the archives for a solution. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Can't use all of my disk in my SS20.
On Wed, 2007-07-18 at 16:13 +0100, Chris Andrew wrote: > Hi, all. > > I have succesfully installed Etch on my SS20. I have 2 harddrives, > both 2.1Gb. Given the previous problems, it might be nice (for the archives) to explain how you did it. > The first HDD will only let me use 200 megs of it. I have tried all > sorts of formatting, but it just won't let me touch most of the drive. > When I do fdisk /dev/sdb, it just shows the two partitions that I have > created, plus the Sun "whole disk" partition. I still can't do > anything with almost 2 gigs of HDD. Can anyone tell me how I can get > around this? Is this a problem with disk labels? Some partition tools require builiding a Sun disk label which includes the number of heads, cylinders, etc. Some of this information is then used by Linux. Copy and data to the other disk, zero the start of the disk and then use something like fdisk which (IIRC) will create a Sun disk label. HTH Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [ss20]cdrom
On Tue, 2007-07-17 at 23:44 +0200, Frans Pop wrote: > On Tuesday 17 July 2007 23:35, Ludovic Courtès wrote: > > Frans Pop <[EMAIL PROTECTED]> writes: > > > The daily built images of the installer for sparc [1] have kernel > > > 2.6.20. I don't know if the rewrite happened before or after that > > > release, but it is somewhat more recent than the Etch kernel. > > > > The rewrite dates back to April [0] while 2.6.20 was released two > > months earlier [1]. Apparently, the new ESP driver first appeared in > > 2.6.22 [2]. > > In that case there is unfortunately extremely little chance that this will > become available in Debian. The only option seems to be a self-compiled > kernel. Unless I'm mistaken, wouldn't booting using an earlier version of Debian and then upgrading or netbooting also be options for at least the install? Cheers, - Martin
Re: Fwd: "File does not appear executable" on SS20.
On Wed, Jul 04, 2007 at 04:56:28PM +0100, Chris Andrew wrote: > Same again, I'm afraid. A year ago, I had Sarge running on an SS20, so I > know it can work. Something is obviously not right. The other strange > thing is that I have two 2.1 Gb drives, and whatever I try to do with the > 2nd drive, it only lets me use about 250 megs. I have removed all the > partitions on it, and tried several times. I've come upon this issue as well. Parted completely misinterpreted the disk geometry. The only way I got around it is to use good ol' fdisk. The only tricky part is getting is during the install: I put it on an ftp server and wget it. I used etch when I hit this. Did try sarge as well, did not get it to install. Martin > On 04/07/07, Matthias Kreis <[EMAIL PROTECTED]> wrote: > > > >Hello Chris > > > > > >disk:a means the first partition. You can tray to boot from disk:c. From > >this partition debian is booting on my sun servers. > > > >Just use: boot disk:c > > > > > >Regards Matthias > > > > > >> I'll have a look, but as I say, Solaris was booting fine, on power-up, > >> and nothing has changed. > >> > >> printenv reads: > >> > >> "boot-device disk:a disk" in the second column it says "disk net" > >> > >> Does this help? > >> > >> Thank, > >> > >> Chris. > >> > >> -- Forwarded message -- > >> From: *ayo jegede* <[EMAIL PROTECTED] > >> <mailto:[EMAIL PROTECTED]> > > >> Date: 04-Jul-2007 12:42 > >> Subject: Re: "File does not appear executable" on SS20. > >> To: Chris Andrew <[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> > >> > >> You may need to set your OBP for your boot device. > >> > >> Run printenv at the boot prom and check which device is set for > >> boot-device > >> > >> Regards > >> Ayo > >> -- > >> Sent with Instant Email from T-Mobile > >> > >> -Original Message- > >> From: "Chris Andrew" < [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> > >> Date: Wed, 4 Jul 2007 09:00:39 > >> To:debian-sparc >> <mailto:debian-sparc@lists.debian.org>> > >> Subject: "File does not appear executable" on SS20. > >> > >> Hi, all. > >> > >> I have been trying to install Sarge on a SPARCstation20. The install > >> goes fine, but when I reboot into my new installation, I get the OBP > >> message "File does not appear executable". > >> > >> BTW, I haven't tried Etch because I haven't got any spare CD's to burn > >> it to. > >> > >> Can anyone help? > >> > >> Thanks, > >> > >> Chris. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RSC console redirection on a Fire 280R
On Sun, Jun 24, 2007 at 10:24:20PM +0200, Josip Rodin wrote: > I wish I knew the explanation for that -p switch :) I can't find it > documented anywhere... Basically all printk's are also written using a PROM write. This means you see output before the linux console has registered, or if something goes wrong with the console. Note that once your console registers okay, you'll see all lines twice. Yeah, this is undocumented. It's pretty much the only usefull switch. -- Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: RSC console redirection on a Fire 280R
Looks okay to me. Try booting linux215sc with '-p', hopefully you'll see where it goes wrong. Martin On Sat, Jun 23, 2007 at 09:33:13PM +0200, Josip Rodin wrote: > Hi, > > Has anyone got a working console redirection set up with a Sun Fire 280R's > RSC card and a Linux kernel 2.6.x? I do a 'bootmode -u normal' followed > by a 'reset', and that temporarily resets the PROM variables output-device > and input-device so that it goes to rsc-console. (The keyboard and monitor > remain plugged in physically at the time.) > > This gets me the PROM input/output nicely, then I boot into SILO, but then > problems arise. With a 2.4.x kernel, I get the console output, but no input. > With 2.6.x I get nothing after the screen is cleared for the first time by > the kernel. > > The relevant silo.conf bits are: > > append="video=atyfb:off" > [...] > > # this is the 2.4.x image that output to RSC (but doesn't take input from it) > image=/boot/vmlinuz.old > label=linux.oldsc > append="video=atyfb:off console=ttyS0,9600n8" > > # these two don't output anything to the RSC > image=/boot/vmlinuz-2.6.21.5 > label=linux215 > initrd=/boot/initrd.img-2.6.21.5 > image=/boot/vmlinuz-2.6.21.5 > label=linux215sc > initrd=/boot/initrd.img-2.6.21.5 > append="video=atyfb:off console=ttyS0,9600n8" > > I also tried removing all of framebuffer and boot logo from the 2.6 kernel, > but nothing changed. > > What am I doing wrong? > > -- > 2. That which causes joy or happiness. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: benchmarks
On Wed, 2007-06-13 at 00:38 +0100, andrew holway wrote: > Hi, > > We've been doing some benchmarks, please make a visit to the two pages below. > > http://www.moopix.co.uk/bench.php - sunfire V210 2 gig ram dual 1.3 > ultrasparc IIIi (debian) > http://www.moonet.co.uk/bench.php - celeron 1.6 512 gig ram (ubuntu) > > You can see that the sparc seems to be performing very badly. Would > this be an expected result or have I made a mess of it. That entirely depends on what you've done and how you've done it. As the pages contain no methodology it's impossible to give anything other than generalities. Doing benchmarking properly is difficult; I tend to start by reading the section of Hennessey and Patterson relevant to what I want to measure. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [sun]debian : setup
On Tue, 2007-06-12 at 22:17 +0200, Eric Rapilly wrote: > Hi there; For my Debian on sun, I need some informations, because I'm > not familiar with debian; You might find that a general purpose Debian mailing list or guide is of more use for this kind of question. Debian is remarkably uniform across architectures, so what works on x86 will likely work on SPARC (with the obvious exceptions). > 1 ° how to setup a fixed IP address ? > 2° how to indicate the IP address of the gateway, which I use to go on > internet See interfaces(5), the man page for /etc/network/interfaces. Or use $YOURFAVOURITENETWORKGUI. Cheers, - Martin
Re: [sun]debian
On Wed, 2007-06-06 at 11:23 +0200, Eric Rapilly wrote: > I have some questions, may be you wille be able to answer me. I will > receive soon a sun ultra 60; > But there is few memory. I want to know which kind of RAM is installed > on the motherboard: is it ECC SDRAM or NVRAM? SDRAM and NVRAM are very different kinds of memory. NVRAM is a very small amount of non volatile (i.e. doesn't get wiped when the system powers down) memory used to save OpenBoot (like a BIOS but properly designed and well thought out) settings. It is completely different and separate from the system RAM. Note that the Ultra 60 takes 200-pin Sun RAM; not PC compatable RAM (owners of Ultra 5's and Ultra 10's claim mixed success in using PC RAM; like PCI USB cards, it can be made to work but YMMV). See: http://www.sunstuff.org/hardware/ > What is the video connector : sub -D 15 or 13w3 ? Depends on what graphics card it has installed. Could be either. > where could I find the latest Debian ETCH for sparc ? In a similar place to Debian for other architectures: http://www.debian.org/releases/stable/sparc/ch01s04.html.en http://www.debian.org/releases/stable/sparc/ch02s02.html.en http://www.debian.org/releases/stable/sparc/ch04.html.en note that downloading all of the CDs is unlikely to be necessary. If you have a reasonably fast net connection, just download the net install CD and use that. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [hardware-donations] Sun Enterprise 4500 and Netra T4, USA
* Josip Rodin <[EMAIL PROTECTED]> [2007-06-04 23:03]: > > but it shouldn't be too hard to find hosting in California. Maybe you > > want to check whether this box would be useful and whether someone can > > host it? > We don't have any sort of a database of hosting offers, do we? I don't think so. -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [hardware-donations] Sun Enterprise 4500 and Netra T4, USA
* Josip Rodin <[EMAIL PROTECTED]> [2007-06-04 15:52]: > An E4500 is more or less a shipping nightmare. Do they have its rack mount > kit? Which CPUs are in it, and how many of them? Another person in England > recently told me about a similar possible donation, and they also have no > idea how to ship it... ... > A Netra T4 is a 4U server which should be pretty modern, and likely worth > having. What are the specifications for that one? I'm not sure, but I'll put you in contact with the donor privately so you can check. > If it's sufficiently modern and beefed up, it might be worth > shipping it to a place with a fat pipe, but I've no idea if such a > thing is being practised. I could host it over here in Europe, if we > decided that kind of a trip was worth it. I doubt shipping to Europe is worth it, but it shouldn't be too hard to find hosting in California. Maybe you want to check whether this box would be useful and whether someone can host it? -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[hardware-donations] Sun Enterprise 4500 and Netra T4, USA
Is anyone interested in the following machines for Debian related work? The machines are currently located in Carlsbad, CA, near Lego Land. > We recently freed up some old Sun hardware that we're willing to donate > for the cost of shipping. Available are: > > - Sun Enterprise 4500 > - Sun Netra T4 -- Martin Michlmayr http://www.cyrius.com/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Intention to drop sparc32 support for Lenny
Frans, Thanks for that pointer. After trawling through some of the mailing lists and searching for bug reports I found Robert Reif's reply indicating this is related to drm and the fact that sparc does not support cmpxcgh at the moment. Still not sure how to exactly reproduce this, but I'll try to kick off a build with drm in it and see what happens. It's a shame, Debian is normally so good at bug management, and now I'm searching for one I cannot find it... If anyone has any more info on this I'd welcome it. Thanks, Martin On Mon, May 21, 2007 at 10:09:49PM +0200, Frans Pop wrote: > On Monday 21 May 2007 21:49, Martin Habets wrote: > > FYI, 2.6.21 is rock solid on my SS20 here. Do you consider it broken > > just because of some cdrom issues? Or is there more? > > I don't have more details than this: > http://lists.debian.org/debian-kernel/2007/05/msg00305.html -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Intention to drop sparc32 support for Lenny
On Mon, May 21, 2007 at 12:00:24PM +0200, Frans Pop wrote: > The sparc32 port has been struggling for some time. Last month Jurij > Smakov, currently the most active Debian Sparc porter, raised the question > if support sparc32 should be dropped for Lenny [1]. I agree with this intent, as I wrote at the time. > The main reason is the fact that sparc32 support is no longer being > maintained upstream for the kernel [2]. A result of that is that the > 2.6.21 kernel is currently broken, which forces the issue. FYI, 2.6.21 is rock solid on my SS20 here. Do you consider it broken just because of some cdrom issues? Or is there more? Even though David does not actively support sparc32, he does push our patches upstream and creates patches himself (and I for one am very thankfull for that). > Another reason is that dropping sparc32 support will allow optimization > for sparc64 which will result in improved support for the modern sparc > systems from Sun and Hitachi. Other distributions have already made this > choice. > > Unless a group of people is willing to commit to providing the needed > upstream kernel and toolchain support for sparc32, the Debian project > will be forced to drop sparc32 for Lenny. Doing this early in the release > cycle has the advantage of allowing sufficient time to work on optimizing > support for sparc64. > > Given the current problems with sparc32 in the 2.6.21 kernel, the final > decision on this will be made before the end of this month. Please include your specific problem(s) here. Cheers, Martin Habets -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Thank you for the Sun Fire v240 support!
On Mon, 2007-04-30 at 11:02 -0700, Matt Weatherford wrote: > Greetings, > > On a similar note, we are considering getting a Sun Fire T1000 system. I > have read a lot about the press with > Ubuntu/Canonical + Sun - but was wondering if anyone could comment about > those vague "optimizations to Linux" made for the T1000, When talking to a Sun rep. they said the next firmware version for the T1 based machines would allow Linux to be run as a guest OS; using their virtualisation tech (and Solaris as a host). I don't know of any annoucement of which distro's would be supported; the rep I spoke to didn't seem to know. > if they have "made > it" in to Etch, Kernel support for the T1 was added in 2.6.17 > and if I should stick with a sun fire V240 or if the T1000 > is really going to be a significantly better performing machine (as a web/db > server) under debian Etch sparc64 SMP If the work load is bottlenecking on memory access, is integer only and sufficiently parallel then it should go faster. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [sun]debian
On Sun, 2007-04-29 at 17:51 +0200, Eric Rapilly wrote: > Hi, ladies and gentlemen . I could'nt install Debian on my SUn SPARC > ULTRA 1, because the crom has benn written with K3B under Fedora core 5, > i.e with the ISO 9660 format, which unknown by the SUN OS ver 5.5.1. If you're looking to install Debian then you shouldn't need to go near Sun OS. Just get OpenBoot to boot from CD and install from there. > Thus, I believe the only thing to do, is to transfer Debian from fedora > to the sun, It is possible to install Debian from a running UNIX but I'd advise against it. I've seen some very smart people pull this trick and successfully install over the system they were running but I really wouldn't recommend trying it. Installing from CD is buch easier. > but for this, I have to install an net between these two > computerFor this, I nedd your help, because the "ifconfig" man page on > the SUN OS is very confusing for me . The IP address (inet Address) is > correct , and in the same net and subnet as the one of the fedora core > machine. But nevertheless, I can't ping neither "nmap" the sun machine > from my fedora machine; I believe in missed an option, like "plumb" > which appaers to be used to associate hm0 device (i;e eth0) to the inet > address; > How to use this option ? and how to enter a gateway ?? many thanks Networking questions about Sun OS are a little off topic here; plus IMHO that's the wrong direction is you just want to do a basic install. Have a read of the rather spiffing install guide - it's pretty straight forward. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Second video card in a SPARC Ultra 5
On Fri, 2007-04-27 at 06:03 +, Jordan Bettis wrote: > Hi everyone. > > Now, I use dual-head so I figured I would install a video card in one > of the PCI slots so I could attach my second monitor. I chose an ATI > Rage 128 that I had, figuring that while it doesn't have a Sun > framebuffer bios, that shouldn't prevent it from working with X. FWIW... I ran an UltraAXi board (similar to an Ultra 5) with dual head using a pair of sun ATi Mach 64 cards with OpenBoot firmware. I tried but don't recall ever suceeding using an ATI Mach 64 card with a PC BIOS. Although 'in theory' it should be workable, it actually trashed the OpebBoot set up on the motherboad; requiring a hefty reset. I got slightly further using old Matrox Millenium cards (which had a toggle switch so they didn't need intialisation from a BIOS); multiple of these were recognised but IIRC I didn't quite get the framebuffer driver for them in the kernel working. Anyway, YMMV but using non Sun graphics cards is tricky; it should work but there seem to be lots of little oddities that need to be ironed out; requiring a knowledge of the hardware, kernel and X. Good luck. Cheers, - Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Dropping sparc32 for lenny
On Wed, 2007-04-25 at 09:01 +0200, BERTRAND Joël wrote: > I don't aggree. Some new hardware are created today with Leon and some > other sparc v8 processors (in fpga or other technology). If sparc32 port > is dropped, no one usable system will be able to use these systems. > > 1/ Sparc32 (with sun4m) support was dropped with solaris 10 and solaris > 9 is not stable with more than two procs; > 2/ NetBSD 3.1 is not stable when it runs on SMP configuration; > 3/ NetBSD 4.x is not usable (on SS20 with CG14); > 4/ OpenBSD cannot run on any SMP station; > 5/ Linux sparc32 is not dropped, some bugs remain in sparc32 SMP kernel, > but is better than all other systems on sparc32. > > I use several SS20 (with raid, CG14, SMP, X, windowmaker...) with only > one trouble: a bug in read_pipe()... And I have worked on another > project that uses Leon. Berkeley University work today on a calculator > that shall use several Leon processors... Sparc32 is not died even Sun > does not sell Sparc32 anymore. > > Thus dropped sparc32 arch will be a big strategic error for Debian. I don't dispute the facts you've given, and personally, I'd like to see Debian ported as widely as possible* but ... if I may play devil's advocate for a moment: 1. There are still sparc32 systems in production and in use; but what proportion of them are 1. publically available (as in COTS rather than Verliog like Leon or the UltraSPARC T1), 2. run Linux, 3. run Debian? If they care about having a Debian/sparc32 port then where are they? And why aren't they offering some support in keeping it running? For developers with experience of custom designed embeded systems, helping maintain kernel code shouldn't be too difficult. 2. This unlikely to make a major difference until a year after lenny is released, when support will be fully stopped. Given most users seem to be using old Sun hardware; what proportion of that will still be in use in 2/3 years time? I guess it comes down to a question of how much work it is to maintain verses how much the Debian projects gains by the extra user base / being the only option of free OS for these platforms. I note that according to popcon : http://popcon.debian.org/ sparc is one of the larger ports. I guess with access to the raw data, it should be possible to extrapolate / guestimate the number of sparc32 systems in use. Cheers, - Martin *. but as I can't directly contribute to making this happen I don't think anyone should pay attention to what I think on the matter. I'd like it if they did but I understand completely if they say "you want it; you go do it".
