Bug#260658: Installation report (amd64)
On Wed, Jul 21, 2004 at 05:17:35PM +0200, Harald Dunkel wrote: > :00:00.0 Host bridge: nVidia Corporation nForce3 Host Bridge (rev a4) > :00:01.0 ISA bridge: nVidia Corporation nForce3 LPC Bridge (rev a6) > :00:01.1 SMBus: nVidia Corporation nForce3 SMBus (rev a4) > :00:02.0 USB Controller: nVidia Corporation nForce3 USB 1.1 (rev a5) > :00:02.1 USB Controller: nVidia Corporation nForce3 USB 1.1 (rev a5) > :00:02.2 USB Controller: nVidia Corporation nForce3 USB 2.0 (rev a2) > :00:05.0 Ethernet controller: nVidia Corporation nForce3 Ethernet (rev > a5) > :00:06.0 Multimedia audio controller: nVidia Corporation nForce3 Audio > (rev a2) > :00:08.0 IDE interface: nVidia Corporation nForce3 IDE (rev a5) > :00:0a.0 PCI bridge: nVidia Corporation nForce3 PCI Bridge (rev a2) > :00:0b.0 PCI bridge: nVidia Corporation nForce3 AGP Bridge (rev a4) > :00:18.0 Host bridge: Advanced Micro Devices [AMD] K8 NorthBridge > :00:18.1 Host bridge: Advanced Micro Devices [AMD] K8 NorthBridge > :00:18.2 Host bridge: Advanced Micro Devices [AMD] K8 NorthBridge > :00:18.3 Host bridge: Advanced Micro Devices [AMD] K8 NorthBridge > :01:06.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host > Controller (rev 80) > :01:07.0 RAID bus controller: Silicon Image, Inc. (formerly CMD > Technology Inc) SiI 3512 [SATALink/SATARaid] Serial ATA Controller (rev 01) > :02:00.0 VGA compatible controller: nVidia Corporation NV28GL [Quadro4 > 980 XGL] (rev a1) > > Configure network HW: > A lot of IDE drivers were listed! No option to select > "sata_sil", though. Discover would have found it. Can you provide the output of lspci -n please ? Thanks for your report. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Wrong kernel after install
On Wed, Aug 04, 2004 at 06:39:41AM -0600, listas wrote: > hello > > I've downloaded the current Sarge installer > http://cdimage.debian.org/pub/cdimage-testing/daily/i386/20040803/sarge-i386-businesscard.iso > > all went fine but the kernel installed > > I've start the instalation with a 2.6 kernel ( linux26 ) > but after the instalation the installer put a 2.6 kernel for AMD k6 > this test machine is a PMMX 233 Hi, thanks for your report. Can you send the output of the /proc/cpuinfo file ? And what mirror did you choose and what suite (testing, sarge, unstable, sid) ? Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Wrong kernel after install
On Wed, Aug 04, 2004 at 09:09:25AM -0600, listas wrote: > Thomas Poindessous wrote: > > >On Wed, Aug 04, 2004 at 06:39:41AM -0600, listas wrote: > > > >>hello > >> > >>I've downloaded the current Sarge installer > >>http://cdimage.debian.org/pub/cdimage-testing/daily/i386/20040803/sarge-i386-businesscard.iso > >> > >>all went fine but the kernel installed > >> > >>I've start the instalation with a 2.6 kernel ( linux26 ) > >>but after the instalation the installer put a 2.6 kernel for AMD k6 > >>this test machine is a PMMX 233 > > > >Hi, > > > >thanks for your report. > > > >Can you send the output of the /proc/cpuinfo file ? > > > > I have to copy by hand this and booted with beta2 sarge install it was a bug in base-installer. There is no more a kernel-image-2.6-k6 package. Thanks for your report. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#265098: rc1 installation on dual processor G4
On Wed, Aug 11, 2004 at 07:04:37PM +0100, Clive Menzies wrote: > > The installer doesn't pick up that this is a dual processor box > but it's no big deal to install the smp kernel. Hi, thanks for your report. Can you send the content of /var/log/debian-installer/syslog and /var/log/debian-installer/hardware-summary ? Thanks in advance. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
sparc64 ultra5 RC1 tftpboot install
Package: installation-reports INSTALL REPORT Debian-installer-version: RC1 sparc64/netboot/boot.img uname -a: Linux ultra5 2.4.26-sparc64 #1 Sun Jun 20 02:39:17 PDT 2004 sparc64 GNU/Linux Date: 2004_08_11 Method: tftpboot, dhcp, ftp install Machine: Sun ultra5 Processor: UltraSparc IIi 270 Mhz Memory: 192 Mo Root Device: /dev/hda 20 Go Root Size/partition table: Disk /dev/hda (Sun disk label): 16 heads, 63 sectors, 38792 cylinders Units = cylinders of 1008 * 512 bytes Device FlagStart EndBlocks Id System /dev/hda1 0 37799 190506961 Boot /dev/hda2 37799 38792500472 82 Linux swap /dev/hda3 0 38792 195511685 Whole disk Output of lspci and lspci -n: :00:00.0 Host bridge: Sun Microsystems Computer Corp. Ultra IIi :00:01.0 PCI bridge: Sun Microsystems Computer Corp. Simba Advanced PCI Bridge (rev 11) :00:01.1 PCI bridge: Sun Microsystems Computer Corp. Simba Advanced PCI Bridge (rev 11) :01:01.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01) :01:01.1 Ethernet controller: Sun Microsystems Computer Corp. Happy Meal (rev 01) :01:02.0 VGA compatible controller: ATI Technologies Inc 3D Rage I/II 215GT [Mach64 GT] (rev 9a) :01:03.0 IDE interface: Silicon Image, Inc. (formerly CMD Technology Inc) PCI0646 (rev 03) :00:00.0 0600: 108e:a000 :00:01.0 0604: 108e:5000 (rev 11) :00:01.1 0604: 108e:5000 (rev 11) :01:01.0 0680: 108e:1000 (rev 01) :01:01.1 0200: 108e:1001 (rev 01) :01:02.0 0300: 1002:4754 (rev 9a) :01:03.0 0101: 1095:0646 (rev 03) Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot worked:[ O] Configure network HW: [ O] Config network: [ O] Detect CD: [ ] Load installer modules: [ O] Detect hard drives: [ O] Partition hard drives: [ O] Create file systems:[ O] Mount partitions: [ O] Install base system:[ O] Install boot loader:[ O] Reboot: [ O] Comments/Problems: Tftpboot. No framebuffer (disabled by default on sparc I think) so languagechooser was a little dirty but readable. I was able to do a french install with a standard us sunkeyboard. Partman didn't propose the automatic partitioning but it did a great job and created the sun disklabel and the partitions. I didn't try to do a X install. Will do another install. Thanks for your work ! -- Thomas Poindessous partman.gz Description: Binary data
Bug#259914: Bug#257260: NO Detect CD with sarge-sparc-netinst.iso on SUN Ultra60
On Thu, Aug 12, 2004 at 09:02:24PM +0300, Anton Zinoviev wrote: > Hi! > > I don't know much about sparcs and I don't know how to deal with this > bug. I think it was related to the busybox modprobe bug. Insmoding a sparc64 module was impossible, so no cdrom detection was possible on sparc64. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#265381: installation report
On Fri, Aug 13, 2004 at 01:29:40AM +0300, Johan Ehnberg wrote: > Package: installation-reports > > Debian-installer-version: > Method: > > required booting with linux26 or expert. The kernel installed was ok, > but for 386. Autodetect my 686 so I don't have install it later (ok I > know, CD space limited). The questions asked were good; not too many, there is no 686 kernel on netinst cd. You can use the businesscard CD which is smaller and which gets deb by network. So it will autodetect and install 686 kernel. I think full CD will also have multiple kernel flavors. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#265477: discover1-data: Wrong detection of aacraid for Dell R.A.C 10280008
Package: discover1-data Severity: normal Hi, today I did a install of RC1 on a DELL 1750 I got a error message about aacraid. pci id 10280008 was affected to aacraid module but in fact, it's the R.A.C for Dell. Here is a patch. Thanks. -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kernel: Linux 2.6.7-1-k7 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous diff -ur discover1-data-1.2004.07.28/pci.lst discover1-data-1.2004.07.28.new/pci.lst --- discover1-data-1.2004.07.28/pci.lst 2004-08-05 17:41:49.0 +0200 +++ discover1-data-1.2004.07.28.new/pci.lst 2004-08-13 12:46:49.0 +0200 @@ -797,7 +797,7 @@ 10280005scsiaacraid PowerEdge Expandable RAID Controller 3/Di 10280006scsiaacraid PowerEdge Expandable RAID Controller 3/Di 10280007unknown racser Remote Access Controller:DRAC III - 10280008scsiaacraid Remote Access Controller + 10280008unknown unknown Remote Access Controller 10280009unknown unknown BMC/SMIC device not present 1028000ascsiaacraid PowerEdge Expandable RAID Controller 3 1028000cunknown racser Remote Access Controller:ERA or ERA/O
Bug#266564: two network cards problem
On Wed, Aug 18, 2004 at 11:42:47AM +0200, Ondra Kudlik wrote: > Package: installation-reports > > One big problem was that as you can see in lspci output, i have two > identical network cards with realtek chipset. Correct driver for > them is 8139too. > > But with installed kernel one card had 8139too module and second one > 8139cp and didn't work. After compiling and installing my own kernel > everything work's fine. Friend of me has same problem with RC1 > installer. I need to test, but I think this is a problem with testing hotplug (and discover maybe). Even with a single realtek card, the two modules are loaded. [EMAIL PROTECTED]:~$ ( lspci ; lspci -n ) | grep 10.0 :00:10.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) :00:10.0 0200: 10ec:8139 (rev 10) [EMAIL PROTECTED]:~$ lsmod | grep ^8139 8139cp 20672 0 8139too26112 0 [EMAIL PROTECTED]:~$ dpkg -l hotplug | grep ^ii ii hotplug0.0.20040329-1 Linux Hotplug Scripts -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#266947: After switching to legacy ATA mode in BIOS, install completes successfully. Of course, not all drive
On Thu, Aug 19, 2004 at 11:48:52PM +0200, Christian Hofstädtler wrote: reassign 266947 discover1-data tags 266947 + patch d-i thanks > After switching to legacy ATA mode in BIOS, install completes > successfully. Of course, not all drives are (would be) accessible. > lspci, lspci -n, uname -a follows. Here is a patch for discover1-data, which adds your e100 network card and all 2.4.27/modules.pcimap e100 network cards which are not detected. > About the kernel version: I thought the detection stuff has been fixed > so Pentium IV machines would not get the i386 kernel, but it seems this > doesn't work on this machine? (cpuinfo follows too.) it does work, if it finds -686 kernel. But there is only -386 kernel on netinst CD, for space reduction. With full CD, businesscard (network) install, you will get -686 kernel. > :00:1f.2 IDE interface: Intel Corp. 82801FB/FW (ICH6/ICH6W) SATA > Controller (rev 03) > :00:1f.3 SMBus: Intel Corp. 82801FB/FBM/FR/FW/FRW (ICH6 Family) > SMBus Controller (rev 03) > :00:1f.2 0101: 8086:2651 (rev 03) > :00:1f.3 0c05: 8086:266a (rev 03) This still needs to be added in discover1-data. -- Thomas Poindessous --- pci.lst.orig2004-08-20 11:19:47.0 +0200 +++ pci.lst 2004-08-20 11:31:27.0 +0200 @@ -5215,6 +5215,14 @@ 80861054ethernete10082801EB (ICH5) PRO/100 VE Ethernet Controller 80861055ethernete10082801EB (ICH5) PRO/100 VM Ethernet Controller 80861059ethernete10082551QM Ethernet Controller + 80861064ethernete100Ethernet Controller + 80861065ethernete100Ethernet Controller + 80861066ethernete100Ethernet Controller + 80861067ethernete100Ethernet Controller + 80861068ethernete100Ethernet Controller + 80861069ethernete100Ethernet Controller + 8086106aethernete100Ethernet Controller + 8086106bethernete100Ethernet Controller 80861075ethernete1000 82547GI Gigabit Ethernet Controller 80861076ethernete1000 82541GI/PI Gigabit Ethernet Controller 80861077ethernete1000 82541GI Gigabit Ethernet Controller
Bug#254071: Fails to load ata_piix module
On Fri, Aug 20, 2004 at 07:55:18AM +0200, Rémi Denis-Courmont wrote: reassign 254071 discover1-data > Hello, > > I had the same problem on a Dell Dimension 4700 with an Intel ICH6 > chipset. > Running by hand: > # modprobe ata_piix > from the installer's console shell before hard drive auto-detection > solved the problem. Hi, can you send the output of lspci ; lspci -n Thanks a lot ! -- Thomas Poindessous
Bug#266947: [christian@hofstaedtler.com: Antw: Re: Bug#266947: After switching to legacy ATA mode in BIOS, install completes successfully. Of cours]
for the record. -- Thomas Poindessous --- Begin Message --- > Here is a patch for discover1-data, which adds your e100 network card > and all 2.4.27/modules.pcimap e100 network cards which are not detected. I should have told you more about the NIC. Intel's Windows drivers claim it is an PRO/100 VE Ethernet Connection. The board spec says the controller is called 82562EZ, and the rest of the board chips is ICH6. Therefore, a better name would be (I guess): 82562EZ (ICH6) PRO/100 VE Ethernet Controller -- Christian --- End Message ---
Bug#267055: discover1-data: patch for all 2.4.27 sata modules
Package: discover1-data Severity: normal Hi, there are a lot of bugreports for RC1 about missing sata detection. Here are three patchs which corrects this detection : - first one : sata_mdk.diff : I took pci.lst for Mandrake CVS and I checked sata lines. I verified them with /lib/modules/2.4.27-1-k7/modules.pcimap - second one : pci.lst.diff : sata_mdk.diff + 2 corrections from /lib/modules/2.4.27-1-k7/modules.pcimap - third one : ata_2.6.diff : this one is not really a patch, it's all sata_ and ata_ modules which are in 2.6 kernel and not in 2.4. I don't know how discover1-data handles 2.6 kernel, so I didn't make a aptch for theses ids. Thanks for your work. -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kernel: Linux 2.6.7-1-k7 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous --- /tmp/ata_2.42004-08-20 14:28:00.0 +0200 +++ /tmp/ata_2.62004-08-20 14:27:51.0 +0200 @@ -11,7 +11,20 @@ 10953112 sata_sil 10953114 sata_sil 10953512 sata_sil +10de0036 sata_nv +10de003e sata_nv +10de0054 sata_nv +10de0055 sata_nv +10de008e sata_nv +10de00e3 sata_nv +10de00ee sata_nv 11063149 sata_via 11660240 sata_svw 17257174 sata_vsc +808624d1 ata_piix +808624df ata_piix +808625a3 ata_piix +808625b0 ata_piix +80862651 ata_piix +80862652 ata_piix 80863200 sata_vsc --- pci.lst.orig2004-08-20 11:19:47.0 +0200 +++ pci.lst 2004-08-20 14:16:14.0 +0200 @@ -932,6 +932,8 @@ 10390008unknown unknown SiS85C503/5513 (LPC Bridge) 10390009pmc ignore ACPI 10390018bridge ignore SiS85C503/5513 (LPC Bridge) + 10390180ide sata_sisRAID bus controller 180 SATA/PATA + 10390181ide sata_sisSATA 10390200video Server:XFree86(sis) 5597/5598/6326 VGA 10390204video Server:XF86_SVGA82C204 10390205video Server:XFree86(sis) SG86C205 @@ -1236,6 +1238,7 @@ 105a1275unknown unknown 20275 105a3318ide sata_promisePDC20318 (SATA150 TX4) 105a3319ide sata_promisePDC20319 (SATA150 TX4) + 105a3371ide sata_promisePDC20371 (SATA150 TX2plus) 105a3373ide sata_promisePDC20378 (SATA150 TX) 105a3375ide sata_promisePDC20375 (SATA150 TX2plus) 105a3376ide sata_promisePDC20376 (SATA150 TX2) @@ -1557,7 +1560,7 @@ 1093c831unknown unknown PCI-GPIB bridge 1094 First International Computers [FIC] 1095 CMD Technology Inc - 10950240unknown unknown Adaptec AAR-1210SA SATA HostRAID Controller + 10950240ide sata_silAdaptec AAR-1210SA SATA HostRAID Controller 10950640unknown unknown PCI0640 10950643unknown unknown PCI0643 10950646unknown unknown PCI0646 @@ -1568,7 +1571,7 @@ 10950670usb usb-ohciUSB0670 10950673usb usb-ohciUSB0673 10950680ide siimage PCI0680 Ultra ATA-133 Host Controller - 10953112ide medley Silicon Image Serial ATARaid Controller [ CMD/Sil 3112/3112A ] + 10953112ide sata_silSilicon Image Serial ATARaid Controller [ CMD/Sil 3112/3112A ] 10953114ide sata_silSilicon Image SiI 3114 SATARaid Controller 10953512ide sata_silSilicon Image Serial ATARaid Controller [ CMD/Sil 3512 ] 1096 Alacron @@ -4815,6 +4818,8 @@ 170b0100unknown unknown NSP2000-SSL crypto accelerator 170c YottaYotta Inc. 172a Accelerated Encryption +1725 Vitesse Semiconductor + 17257174ide sata_vscVSC7174 PCI/PCI-X Serial ATA Host Bus Controller 1737 Linksys 17371032ethernetunknown Gigabit Network Adapter 17371064ethernetunknown Gigabit Network Adapter @@ -5421,7 +5426,7 @@ 8086266funknown unknown I/O Controller Hub PATA 80862782unknown unknown Graphics Controller 80863092unknown i2o_block Integrated RAID - 80863200unknown unknown GD31244 PCI-X SATA HBA + 80863200ide sata_vscGD31244 PCI-X SATA HBA 80863340unknown unknown 82855PM Processor to I/O Controller 80863341unknown unknown 82855PM Processor to AGP Controller 80863575bridge agpgart 82830 830 Chipset Host Bridge --- pci.lst.orig2004-08-20 11:19:47.0 +0200 +++ pci.lst 2004-08-20 14:06:18.0 +0200 @@ -932,6 +932,8 @@ 10390008unknown unknown S
Bug#267059: Installation report: RC1
On Fri, Aug 20, 2004 at 08:17:32AM -0500, Michael Satterwhite wrote: > > ***PROBLEM: The new installation routine does not install development tools. > The nature of Linux with the near standard of ./configure make makes the > resultant system very weak. In addition, adding the tools afterward is very > error prone. I've used apt to install gcc, g++, etc and (after 2 days) am > still not able to compile because of a missing cc1plus. Older versions of the > installer made installation of development tools a simple matter of one > selection during installation. Either apt needs to be able to actually > install working compilers or the installer needs to be able to set up the > compilation tools - preferably both should work. The removal of the feature > to install development tools makes the installation of Linux from this > version of the installer all but useless. Please consider putting the ability > to install development tools as part of the base install back in the > intaller. This change is a giant step backward. you just need to install build-essential and the problem of the missing package is maybe about the gcc transition, it will be fixed for the release. And there is no more a development task. Search debian-boot archives for rationale. Thanks for your bug report. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: plans for d-i string freeze
On Sat, Dec 06, 2003 at 09:11:43PM -0500, David Nusinow wrote: > On Fri, Dec 05, 2003 at 11:24:56AM -0500, Joey Hess wrote: > > Christian Perrier wrote: > > Hi Joey and everyone, thanks for the cc. I'm willing to maintain > discover in maintainance mode, but if Progeny plans to focus on > discover2, I don't really see the point in keeping discover1 alive > beyond the sarge release. The fix from Thomas Poindessous isn't really > a fix at all, if you look at it. It seems the real bug is in > linux-kernel-headers, from what I can tell, so once that's fixed an > upload can be made. If anyone at all is interested in maintaining > discover1, I won't stand in their way. Hi ! as I explained in the bts, the recent change in glibc and linux-kernel-headers has changed the way to handle inclusion of kernel headers. userland application ( like discover ) must copy headers they are using, they must not including kernel headers. Further informations can be found in glibc mailing list. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: plans for d-i string freeze
On Sat, Dec 06, 2003 at 09:11:43PM -0500, David Nusinow wrote: > On Fri, Dec 05, 2003 at 11:24:56AM -0500, Joey Hess wrote: > > Christian Perrier wrote: > > Hi Joey and everyone, thanks for the cc. I'm willing to maintain > discover in maintainance mode, but if Progeny plans to focus on > discover2, I don't really see the point in keeping discover1 alive > beyond the sarge release. The fix from Thomas Poindessous isn't really > a fix at all, if you look at it. It seems the real bug is in > linux-kernel-headers, from what I can tell, so once that's fixed an > upload can be made. If anyone at all is interested in maintaining > discover1, I won't stand in their way. Hi, I found this : /usr/share/doc/linux-kernel-headers/README.Debian I hope this is a better explanation than mine. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
update for discover-data
Hi, I made a update for discover-data with hwdata list and data from bug reports. It's available in bug #224428 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=224428 It would be a good idea to have a updated list for beta2. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#225926: choose-mirror: doesn't ask for manual mirror in "normal" mode
Package: choose-mirror Severity: normal businesscard iso from ~manty, 1st Jan. In "normal" mode (default mode), choose-mirror doesn't propose "manual mirror". Since it needs a manual selection, why not propose manual mirror ? Thanks in advance. -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux thomas 2.4.23-1-k7 #1 Mon Dec 1 00:05:09 EST 2003 i686 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: sarge netinst of today: pcmcia problem
On Sun, Jan 04, 2004 at 11:00:16AM +0100, Harald Dunkel wrote: > - discover loaded i82365.o instead of i82092.o. > Maybe this can be configured; I haven't checked. > The workaround was to add i82092 to /etc/modules. please, can you provide output of $ lspci and $ lspci -n thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: sarge netinst of today: pcmcia problem
[ Tried to resent it to debian-boot without the attachments, since it didn't make it the first time ] On Sun, Jan 04, 2004 at 04:42:32PM +0100, Harald Dunkel wrote: > >>- discover loaded i82365.o instead of i82092.o. > >> Maybe this can be configured; I haven't checked. > >> The workaround was to add i82092 to /etc/modules. > 00:03.1 CardBus bridge: Texas Instruments PCI1420 > 00:03.1 Class 0607: 104c:ac51 Hi, can you confirm that if you applied this patch ( ou copy pci.lst to /usr/share/discover/pci.lst ), then you suppress i82092 from /etc/modules, then after a reboot or a rmmod, discover does detect your card correctly ? thanks. --- /usr/share/discover/pci.lst 2003-12-27 14:11:36.0 +0100 +++ pci.lst 2004-01-08 20:48:49.0 +0100 @@ -1023,7 +1023,7 @@ 104cac42bridge i82365 PCI4451 PC card Cardbus Controller 104cac44unknown unknown PCI4510 PC card Cardbus Controller 104cac50bridge i82365 PCI1410 PC card Cardbus Controller - 104cac51bridge i82365 PCI1420 PC card Cardbus Controller + 104cac51bridge i82092 PCI1420 PC card Cardbus Controller 104cac52bridge i82365 PCI1451 PC card Cardbus Controller 104cac53bridge i82365 PCI1421 PC card Cardbus Controller 104cac55unknown unknown PCI1250 PC card Cardbus Controller -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: sparc boot progress -- or at least a different error
On Sun, Jan 11, 2004 at 07:04:42AM -0800, Blars Blarson wrote: > After spending way to much time beeting on it, (and burning a couple > coasters) I've finally gotten the sparc d-i to mount the generated > root filesystem. Unfortuanatly it doesn't get much farther: Hi, I tried to use netboot method, and I can't get my sparc to use initrd. I tried two methods : dhcp,netboot and setenv boot-file : $ setenv boot-file "vga=normal initrd=netboot-initrd.gz ramdisk_size=8192 root=/dev/rd/0 devfs=mount,dall rw" $ boot net:dhcp this loads the di.img (tftp image build from cvs), but I got an error after I pressed three times ENTER when asking for a floopy : from memory : "... can't open rd/0, please use a correct root= option" Second time, I use a silo.conf method, with vmlinux-2.4.21-sparc64 kernel and netboot-initrd.gz ( and same append option ). i noticed that during kernel booting, devfs reported "0x1" as options ( i will re-check later ). How can I use netboot for testing d-i ? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: [debian_inst@yahoo.com: d-i fails on eepro100]
On Sun, Jan 11, 2004 at 06:47:47PM +, Martin Michlmayr wrote: > The submitter claims that his laptop only works with the eepro100 > module, not with e100. This has been fixed this then, but since > (afaik) all occurences of eepro100 have been changed to e100 in a > previous version of discover-data, I wonder if anyone actually checked > if e100 supports all cards eepro100 does. Just for information, redhat doesn't use eepro100 neither, and only uses e100 ( from /usr/share/hwdata/pcitable ). -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#227343: discover-data: wrong module for CardBus bridge: Texas Instruments PCI1420
Package: discover-data Version: 1.2002.08.21-2 Severity: normal Tags: patch Hi, as reported by Harald Dunkel in http://lists.debian.org/debian-boot/2004/debian-boot-200401/msg00237.html discover uses the wrong module for CardBus bridge: Texas Instruments PCI1420. Here is a patch. Thanks for your work. -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux thomas 2.4.23-1-k7 #1 Mon Dec 1 00:05:09 EST 2003 i686 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- no debconf information -- Thomas Poindessous --- /usr/share/discover/pci.lst 2003-12-27 14:11:36.0 +0100 +++ pci.lst 2004-01-08 20:48:49.0 +0100 @@ -1023,7 +1023,7 @@ 104cac42bridge i82365 PCI4451 PC card Cardbus Controller 104cac44unknown unknown PCI4510 PC card Cardbus Controller 104cac50bridge i82365 PCI1410 PC card Cardbus Controller - 104cac51bridge i82365 PCI1420 PC card Cardbus Controller + 104cac51bridge i82092 PCI1420 PC card Cardbus Controller 104cac52bridge i82365 PCI1451 PC card Cardbus Controller 104cac53bridge i82365 PCI1421 PC card Cardbus Controller 104cac55unknown unknown PCI1250 PC card Cardbus Controller
Sparc netboot install failed
[ I resent this mail, since previous one didn't pass beacause of the attachments. I can provide syslog and "find /" list on demand ] Hi, today I tried a netbood install on a Sparc Ultra5. I did a make TYPE=netboot in current cvs, and I build a tftpboot.img for a hacked tftpboot.sh ( tilo doesn't work). Then I booted via rarp/tftp. kdb-chooser doesn't work. hw-detect doesn't work very well, it seems that discover is missing. dhcp configuration worked, but net-retriever doesn't work. Sorry for not filing a proper installation report. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#227851: udpkg: [PATCH] wrong architecture for sparc64
Package: udpkg
Severity: normal
Hi, udpkg --print-architecture reports sparc64 for UltraSparc; it needs
to report sparc instead.
here is a small patch which corrects that. It was tested.
Thanks.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux thomas 2.4.24-1-k7 #1 Wed Jan 7 00:47:47 EST 2004 i686
Locale: [EMAIL PROTECTED], [EMAIL PROTECTED]
--
Thomas Poindessous
Index: tools/udpkg/udpkg.c
===
RCS file: /cvsroot/d-i/debian-installer/tools/udpkg/udpkg.c,v
retrieving revision 1.45
diff -u -r1.45 udpkg.c
--- tools/udpkg/udpkg.c 13 Dec 2003 00:17:49 - 1.45
+++ tools/udpkg/udpkg.c 14 Jan 2004 23:54:02 -
@@ -62,6 +62,7 @@
{ "parisc64", "hppa" },
{ "ppc", "powerpc" },
{ "ppc64", "powerpc" },
+ { "sparc64", "sparc" },
};
if (uname(&name) < 0)
Bug#227853: debian-installer: [PATCH] small fix for sparc, installs discover-udeb
Package: debian-installer Version: [PATCH] small fix for sparc, installs discover-udeb Severity: normal Hi, here is a small fix, which adds discover and discover-data in sparc netboot image. Thanks. PS : I deleted kbd-chooser in netboot since it's already in common. -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux thomas 2.4.24-1-k7 #1 Wed Jan 7 00:47:47 EST 2004 i686 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous Index: pkg-lists/netboot/sparc === RCS file: /cvsroot/d-i/debian-installer/build/pkg-lists/netboot/sparc,v retrieving revision 1.8 diff -u -r1.8 sparc --- pkg-lists/netboot/sparc 24 Dec 2003 00:48:13 - 1.8 +++ pkg-lists/netboot/sparc 14 Jan 2004 23:56:55 - @@ -1,4 +1,5 @@ console-keymaps-usb console-keymaps-sun -kbd-chooser dhcp-client-udeb +discover-udeb +discover-data-udeb
Bug#227852: kbd-chooser: [PATCH] wrong name for sparc/sun keyboard
Package: kbd-chooser
Severity: normal
Hi,
kdb-chooser search for sparc directory, but it's now in sun directory.
This patch corrects that. It was tested.
Thanks.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux thomas 2.4.24-1-k7 #1 Wed Jan 7 00:47:47 EST 2004 i686
Locale: [EMAIL PROTECTED], [EMAIL PROTECTED]
--
Thomas Poindessous
Index: tools/kbd-chooser/sparc-kbd.c
===
RCS file: /cvsroot/d-i/debian-installer/tools/kbd-chooser/sparc-kbd.c,v
retrieving revision 1.9
diff -u -r1.9 sparc-kbd.c
--- tools/kbd-chooser/sparc-kbd.c 14 Nov 2003 20:53:16 - 1.9
+++ tools/kbd-chooser/sparc-kbd.c 14 Jan 2004 23:54:26 -
@@ -21,7 +21,7 @@
{
kbd_t *k = xmalloc (sizeof(kbd_t));
- k->name = "sparc"; // This must match the name "sparc" in console-keymaps-sparc
+ k->name = "sun"; // This must match the name "sparc" in console-keymaps-sparc
k->deflt = NULL;
k->fd = -1;
k->present = UNKNOWN;
Bug#227854: Install report for Sparc netboot.
Package: installation-reports INSTALL REPORT Debian-installer-version: sparc netboot build from cvs of 2004/01/14 and with hacked udpkg, kdb-chooser. uname -a: Linux ultra5 2.4.21 #1 Thu Aug 7 20:30:12 EDT 2003 sparc64 GNU/Linux Date: 2004/01/14 23h Method: netboot method with rarpd/tftpd. dhcp network without proxy, but with a apt-proxy as mirror source. Machine: Ultra 5 Processor: ultra-sparc II, 270 Memory: 192 Mo Root Device: IDE Root Size/partition table: (from memory) /dev/hda1 ext2other_linux /dev/hda2 ext2other_linux /dev/hda3 ext2other_linux /dev/hda4 ext2/ 500 Mo /dev/hda5 swap70 Mo. Output of lspci: 00:00.0 Host bridge: Sun Microsystems Computer Corp. Ultra IIi 00:01.0 PCI bridge: Sun Microsystems Computer Corp. Simba Advanced PCI Bridge (rev 11) 00:01.1 PCI bridge: Sun Microsystems Computer Corp. Simba Advanced PCI Bridge (rev 11) 01:01.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01) 01:01.1 Ethernet controller: Sun Microsystems Computer Corp. Happy Meal (rev 01) 01:02.0 VGA compatible controller: ATI Technologies Inc 3D Rage I/II 215GT [Mach64 GT] (rev 9a) 01:03.0 IDE interface: CMD Technology Inc PCI0646 (rev 03) Base System Installation Checklist: Initial boot worked:[O] Configure network HW: [O] Config network: [O] Detect CD: [ ] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [E] Create file systems:[O] Mount partitions: [O] Install base system:[O] Install boot loader:[ ] Reboot: [O] [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Comments/Problems: With patched udpkg and kbd-chooser, almost everything was alright. Since everything was included in kernel, no module was installed. partition hard-drives didn't work, I didn't check if fdisk was installed since my hard-drive was partitioned. something strange, the -smp kernel was installed. and since there is no silo-installer, I need to use my current silo with extra parameter root=/dev/hda4 But it was a very smooth install ! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#227853: debian-installer: [PATCH] small fix for sparc, installs discover-udeb
On Fri, Jan 16, 2004 at 01:11:50PM +0100, Gaudenz Steinlin wrote: > Am Don, den 15.01.2004 schrieb Thomas Poindessous um 10:22: > > Package: debian-installer > > Version: [PATCH] small fix for sparc, installs discover-udeb > > Severity: normal > > > > Hi, here is a small fix, which adds discover and discover-data in sparc > > netboot image. > > Did you check if discover is actually usable on sparc? Yes, and it's usable. Here is a log on my Ultra5 : ultra5:~# /etc/init.d/discover Detecting hardware: ide-scsi sunhme Loading ide-scsi module. Skipping sunhme module; assuming it is compiled into the kernel. ultra5:~# exit > Does sparc use > PCI Buses? Mine yes. > Does it fail gracefully on sparcs without PCI? I don't know. I assume that some bugs would have been reported if it didn't work. I CC-ed debian-sparc for additional information. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#228327: installation-reports: PCMCIA of the yenta_socket variety not detected properly
On Sun, Jan 18, 2004 at 11:32:58AM +0100, Matijs van Zuijlen wrote: > 00:08.0 CardBus bridge: Texas Instruments PCI1225 (rev 01) > 00:08.1 CardBus bridge: Texas Instruments PCI1225 (rev 01) > PCMCIA drivers. It turned out the system tried to use the i82365 > driver, whereas this machine needs yenta_socket. Changing > /etc/default/pcmcia solved that, and after reinserting my network > card, I had a working eth0. Hi, could you provide us the output of lspci -n with this, it will be possible to change the automatic detection of your card. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#228399: base-installer: [PATCH] kernel package name for sparc are wrong
Package: base-installer Severity: normal Hi, the kernel package names are wrong for sparc. current kernel packages names are : kernel-image-2.4.21-sparc32 kernel-image-2.4.21-sparc32-smp kernel-image-2.4.21-sparc64 kernel-image-2.4.21-sparc64-smp Here is a patch to correct that. I didn't test it yet. -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux thomas 2.4.24-1-k7 #1 Wed Jan 7 00:47:47 EST 2004 i686 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#228399: Attach the patch
I forgot to attach my patch. -- Thomas Poindessous Index: postinst === RCS file: /cvsroot/d-i/debian-installer/tools/base-installer/debian/postinst,v retrieving revision 1.60 diff -u -r1.60 postinst --- postinst17 Jan 2004 19:28:03 - 1.60 +++ postinst18 Jan 2004 19:31:24 - @@ -209,16 +209,16 @@ case "$MODEL" in sparc) if [ "$CPUS" -eq 1 ]; then - trykernel=kernel-image-$version-sun4cdm + trykernel=kernel-image-$version-sparc32 else - trykernel=kernel-image-$version-sun4dm-smp + trykernel=kernel-image-$version-sparc32-smp fi ;; sparc64) if [ "$CPUS" -eq 1 ]; then - trykernel=kernel-image-$version-sun4u + trykernel=kernel-image-$version-sparc64 else - trykernel=kernel-image-$version-sun4u-smp + trykernel=kernel-image-$version-sparc64-smp fi ;; esac
Bug#228252: SMP detection doesn't work
Hi, for the moment, smp detection doesn't work as the logic is wrong. For a SMP system, a no-smp kernel reports : Processors: 1 and for a uni-cpu, there is no line at all. So you would install a smp kernel only if the grep is positive. But anyway, current kernel used by d-i doesn't work with this technique, I don't know why. After install, I have installed kernel-image-2.4.24-1-686 and I got a "Processors: 1" line. I will install kernel-image-2.4.24-1-386 to see I get the line with a 386 kernel. And are you interessed in more x86 cpu detection (others than athlon and P4) or you will use archdetect ? PS : for dmesg ring buffer, maybe you could use the info in syslog ? -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#228519: [PATCH] tftpboot.sh for sparc
Package: debian-installer Severity: normal Hi, here is a small fix which corrects netboot image building on sparc. Thanks in advance. -- Thomas Poindessous Index: tftpboot.sh === RCS file: /cvsroot/d-i/debian-installer/build/tftpboot.sh,v retrieving revision 1.6 diff -u -r1.6 tftpboot.sh --- tftpboot.sh 30 Dec 2003 21:02:18 - 1.6 +++ tftpboot.sh 18 Jan 2004 19:44:14 - @@ -44,7 +44,7 @@ done case "$arch" in -arm | i386 | mips | mipsel) +sparc | arm | i386 | mips | mipsel) cp $kernel $tftpimage.tmp ;; *)
Bug#227852: kbd-chooser: New patch
Package: kbd-chooser
Severity: normal
Followup-For: Bug #227852
Hi,
here is a new patch which is more clean.
Thanks.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux thomas 2.4.24-1-k7 #1 Wed Jan 7 00:47:47 EST 2004 i686
Locale: [EMAIL PROTECTED], [EMAIL PROTECTED]
--
Thomas Poindessous
Index: sparc-kbd.c
===
RCS file: /cvsroot/d-i/debian-installer/tools/kbd-chooser/sparc-kbd.c,v
retrieving revision 1.9
diff -u -r1.9 sparc-kbd.c
--- sparc-kbd.c 14 Nov 2003 20:53:16 - 1.9
+++ sparc-kbd.c 18 Jan 2004 19:43:05 -
@@ -21,7 +21,7 @@
{
kbd_t *k = xmalloc (sizeof(kbd_t));
- k->name = "sparc"; // This must match the name "sparc" in console-keymaps-sparc
+ k->name = "sun"; // This must match the name "sun" in console-keymaps-sun
k->deflt = NULL;
k->fd = -1;
k->present = UNKNOWN;
Bug#228252: kernel-image-2.4.*-386 needs apic
Hi, I've done some tests, and I think that kernel-image-2.4.*-386 needs apic option ( kernel-image-2.4.18-bf2.4 does it ). Computer : bi-xeon with HT. 2.4.18-386 : No Processors: line 2.4.18-686 : Processors: 2, but only one in /proc/cpuinfo 2.4.24-686-smp-thomas4 : Processors: 4, and 4 in cpuinfo 2.4.24-386-thomas1 ( 386 with APIC ) : Processors: 1, and 1 in cpuinfo. So, if you enable APIC in 386, then if there is a line "Processors:", this is a smp computer. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
problem with hw-detect and ide-scsi ?
Hi, today I tried to install daily build on vmware and cvs build on sparc, and I got stuck in "detect hardware" step. It tried to load ide-scsi which is missing but it didn't ignore it. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Installer builds for non x86 archs. - Was Re: Daily builds may go away
On Wed, Feb 04, 2004 at 07:15:00PM -0500, Ray Knight wrote: > OK, there is a link to an m68k Mac image at this site. > But, there doesn't appear to be an installer for NuBus PPC Macs, no link > to installer images for SGI Indy or Sparc. Is anyone working on > supporting the NuBus PPC Macs? How about SGI Indy or Sparc? I have netboot images for Sparc, but I'm waiting for exim4*. Once exim4* are in unstable, I will publish my images. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#228519: actually this patch wasn't needed
Hi, today, I re-tried to build netboot image from new system, and I found that my patch was a mistake. I'm sorry, it's not needed. Since there is no sparc autobuild yet, I hope that it caused no real harm. since kernel image for sparc are gzip, it needs to gunzip them, so this patch must be undone. I was using a gunzipped kernel. Sorry for that. I attached a new patch to apply against current cvs. Thanks for you work. -- Thomas Poindessous Index: tftpboot.sh === RCS file: /cvsroot/d-i/debian-installer/build/tftpboot.sh,v retrieving revision 1.7 diff -u -r1.7 tftpboot.sh --- tftpboot.sh 19 Jan 2004 16:50:27 - 1.7 +++ tftpboot.sh 9 Feb 2004 00:13:00 - @@ -44,7 +44,7 @@ done case "$arch" in -sparc | arm | i386 | mips | mipsel) +arm | i386 | mips | mipsel) cp $kernel $tftpimage.tmp ;; *)
[PATCH] error with genromfs in new build system
I think that there is a error with genromfs in build/Makefile I was building netboot image for sparc and I got a error. the "-f" flag is missing. In old build system, it was present ( http://cvs.alioth.debian.org/cgi-bin/cvsweb.cgi/debian-installer/build/Makefile?rev=1.265&content-type=text/x-cvsweb-markup&cvsroot=d-i ) I think this patch could solve this problem. I only test the first case. Thanks -- Thomas Poindessous Index: Makefile === RCS file: /cvsroot/d-i/debian-installer/build/Makefile,v retrieving revision 1.269 diff -u -r1.269 Makefile --- Makefile8 Feb 2004 08:15:05 - 1.269 +++ Makefile9 Feb 2004 00:19:37 - @@ -564,7 +564,7 @@ if [ $(INITRD_FS) = ext2 ]; then \ $(genext2fs) $(TEMP)/initrd; \ elif [ $(INITRD_FS) = romfs ]; then \ - genromfs -d $(TREE) $(TEMP)/initrd; \ + genromfs -d $(TREE) -f $(TEMP)/initrd; \ else \ echo "Unsupported filesystem type"; \ exit 1; \ @@ -606,7 +606,7 @@ set -e; if [ $(INITRD_FS) = ext2 ]; then \ $(genext2fs) $@; \ elif [ $(INITRD_FS) = romfs ]; then \ - genromfs -d $(TREE) $@; \ + genromfs -d $(TREE) -f $@; \ else \ echo "Unsupported filesystem type"; \ exit 1; \
Re: Lowering mirror questions priority to medium
On Wed, Feb 11, 2004 at 06:46:58PM +0100, Christian Perrier wrote: > Same rationale for mirror host. As long as we check that default hosts > for each country are rock solid mirrors, I see no real interest in > keeping this question at high priority. > > Does someone object to this? It seems a good idea. The only drawback is that secondary mirror will be never used (for installation and further apt), like ftp2.fr.debian.org for exemple. But it's a minor issue. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
sparc, netboot and linux-kernel-di naming convention
Hi, today, I saw that linux-kernel-di was uploaded for sparc. I tried to do a netboot build. I got these errors : Selecting previously deselected package kernel-image-2.4.24-sparc64-di. (Reading database ... 196 files and directories currently installed.) Unpacking kernel-image-2.4.24-sparc64-di (from .../kernel-image-2.4.24-sparc64-di.udeb) ... dpkg - warning, overriding problem because --force enabled: trying to overwrite `/boot/vmlinuz', which is also in package kernel-image-2.4.24-sparc32-di dpkg - warning, overriding problem because --force enabled: trying to overwrite `/boot/System.map', which is also in package kernel-image-2.4.24-sparc32-di and a little later : # Move the kernel image out of the way. mv -f ./tmp/netboot/tree/boot/vmlinuz-2.4.24-sparc32 ./tmp/netboot/vmlinuz-2.4.24-sparc32; \ mv -f ./tmp/netboot/tree/boot/vmlinuz-2.4.24-sparc64 ./tmp/netboot/vmlinuz-2.4.24-sparc64; mv: cannot stat `./tmp/netboot/tree/boot/vmlinuz-2.4.24-sparc32': No such file or directory mv: cannot stat `./tmp/netboot/tree/boot/vmlinuz-2.4.24-sparc64': No such file or directory With a bit of hacking build/Makefile and config/sparc.cfg, I will be able to build a image (I did it a few days ago), but the current building system is flawed for sparc. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#227852: kbd-chooser: [PATCH] another sparc->sun change
Package: kbd-chooser Severity: normal Followup-For: Bug #227852 Hi, I found another change to do : Rationale for others persons : the console-keymaps package for sparc is console-keymaps-sun. In kbd-chooser.h, there is a option which defines the preferred keymap package for sparc architecture and the name must be "sun" instead of "sparc". The second patch corrects a same error in TODO file, it's not very important. Once this patch is applied, it would be very appreciated if you do a upload, because this bug blocks sparc testing, even if it's not the only one :) Thanks for your work -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.4.24-1-k7 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous Index: kbd-chooser.h === RCS file: /cvsroot/d-i/debian-installer/tools/kbd-chooser/kbd-chooser.h,v retrieving revision 1.14 diff -u -r1.14 kbd-chooser.h --- kbd-chooser.h 21 Jan 2004 14:00:45 - 1.14 +++ kbd-chooser.h 11 Feb 2004 22:56:45 - @@ -54,7 +54,7 @@ extern int check_dir (const char *dirname); #ifdef SPARC_KBD -#define PREFERRED_KBD "sparc" +#define PREFERRED_KBD "sun" #endif #ifndef __m68k__ Index: TODO === RCS file: /cvsroot/d-i/debian-installer/tools/kbd-chooser/TODO,v retrieving revision 1.7 diff -u -r1.7 TODO --- TODO29 Jul 2003 18:21:59 - 1.7 +++ TODO11 Feb 2004 22:56:26 - @@ -5,7 +5,7 @@ - on netinst, all console-keymap-* packages are installed. Try installing only the ones that are required (eg don't pull in - console-keymaps-sparc on an i386 install) + console-keymaps-sun on an i386 install) - Only install console-keymaps-at on arm arch for the appropriate hardware.
Bug#227852: kbd-chooser: [PATCH] the last one, I hope
Package: kbd-chooser Severity: normal Followup-For: Bug #227852 Hi, today, I tried netboot install for sparc, and I found a last bug in kbd-chooser. The word "sparc" in the debconf template must be changed to "sun" to match previous change. Could somebody applied the two last patchs ? Thanks in advance. -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.2-1-k7 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous Index: tools/kbd-chooser/debian/keyboard-sparc.templates === RCS file: /cvsroot/d-i/debian-installer/tools/kbd-chooser/debian/keyboard-sparc.templates,v retrieving revision 1.2 diff -u -r1.2 keyboard-sparc.templates --- tools/kbd-chooser/debian/keyboard-sparc.templates 6 Nov 2003 22:36:45 - 1.2 +++ tools/kbd-chooser/debian/keyboard-sparc.templates 13 Feb 2004 01:06:50 - @@ -1,4 +1,4 @@ -Template: kbd-chooser/kbd/sparc +Template: kbd-chooser/kbd/sun Type: text _Description: Sun keyboard
[PATCH] small patch to fix cdrom build for sparc
Hi, this patch is needed to bix cdrom building for sparc. It's quite ugly, so if someone has a better idea ... tilo needs uncompressed kernel image and vmlinuz* are gzipped. -- Thomas Poindessous Index: sparc/cdrom.cfg === RCS file: /cvsroot/d-i/debian-installer/build/config/sparc/cdrom.cfg,v retrieving revision 1.2 diff -u -r1.2 cdrom.cfg --- sparc/cdrom.cfg 9 Feb 2004 18:31:53 - 1.2 +++ sparc/cdrom.cfg 13 Feb 2004 01:07:42 - @@ -7,6 +7,7 @@ .PHONY: arch_boot arch_boot: + $(foreach img,$(TEMP_KERNEL), mv $(img) $(img).gz ; gunzip $(img).gz ; ) tilo -o $(TEMP_BOOT) $(TEMP_KERNEL) $(TEMP_INITRD) MANIFEST-BOOT = "CDROM image for all Sun Sparcstations"
sparc : some problems and netboot image available
Hi,
today I tried netboot install for sparc64. It did worked, but with several
problems:
- first, make build_netboot provides a image which displays a lot of
lines "bus error" when booting (after umounting /initrd). It's not
useable.
- make build_cdrom, with my patch, provides a image, but it didn't boot,
displaying a error in tilo. I will retry it and make a bugreport. It
may be related to recent rewrite of silo supporting larger kernel.
- now something really weird : I did a make build_cdrom with
pkg-lists/cdrom/{common,sparc} from pkg-lists/netboot/. Then I did a
tftpboot image with tftpboot.sh script. It worked ... something is
really wrong in build_netboot method.
Now problems with install :
- no fdisk : I will do another image tomorrow with fdisk-udeb
- no dns resolving : I can't install from anything else than my
apt-proxy with numerical ip address. I tried to do a wget
ftp://ftp.debian.org/ and I got an error about dns resolution. My dns
server works fine and /etc/resolv.conf is correct. Maybe it's about
the line in config/i386/cdrom.cfg :
EXTRALIBS = /lib/libnss_dns* /lib/libresolv*
- no bootloader : I used my current silo
- SMP kernel installed ?
I put my tftpboot image on this URL :
http://diablero.free.fr/sparc/
This image includes : kbd-chooser_0.41_sparc.udeb with my patch and
base-installer_0.052_sparc.udeb from current CVS.
Happy testing !
--
Thomas Poindessous
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
sparc : work & results
Hi, I found why netboot images were crashing with "bus error". It comes from a wrong unifont.bgf generated by bdftobogl. Bugreport filed against libbogl-dev. debian-boot was added to x-debbugs-cc, but it seems that didn't work. bug #232720 And I also found why name resolution was not working. Someone should add EXTRALIBS line in sparc/cdrom.cfg and sparc/netboot.cfg like i386/cdrom.cfg this line adds libresolv and libnss_dns and after that I got name resolution. [ BTW, I just found that jbailey already found it since friday, I really need to read more carefully next time :) ] Anyway, if you don't apply his config, please add this line. Jeff, if you want to autobuild working netboot image, you should apply this patch. But it maybe breaks cdrom builds ... Once these bugs are resolved, and fdisk-udeb, new kbd-chooser and partionner are added to the archive, we will have a good d-i for sparc ! -- Thomas Poindessous Index: Makefile === RCS file: /cvsroot/d-i/debian-installer/build/Makefile,v retrieving revision 1.281 diff -u -r1.281 Makefile --- Makefile13 Feb 2004 23:08:55 - 1.281 +++ Makefile15 Feb 2004 22:57:13 - @@ -548,7 +548,7 @@ fi cat $(TEMP)/all.utf | LOCPATH=$(LOCALE_PATH) LC_ALL=C.UTF-8 reduce-font /usr/src/unifont.bdf > $(TEMP)/unifont.bdf # bdftobogl is part of package libbogl-dev - bdftobogl -b $(TEMP)/unifont.bdf > [EMAIL PROTECTED] + bdftobogl -b $(TEMP)///unifont.bdf > [EMAIL PROTECTED] mv [EMAIL PROTECTED] $@
Re: Sparc d-i status
On Tue, Feb 24, 2004 at 09:34:06AM -0500, Jeff Bailey wrote: > I also remember (was it Thomas? Blars? hmmm...) mentioning that the > fdisk udeb wasn't getting loaded for some reason. I'm hoping that the > whole partman thing will make this silently go away before we get there. in fact, there is no fdisk-udeb for sparc (bug #228444 ), and there is no support for sparc in partitioner ( bug #228518). we also need a silo-installer -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: netcfg broken? (20040224 image)
On Wed, Feb 25, 2004 at 11:45:34AM +0100, Sven Luther wrote: > netcfg (0.47) unstable; urgency=low > > * Joey Hess > - Use three dots in ellipses for consistency. > - Link libiw statically. This is temporary, until there is a libiw-udeb. > > Joeyh, can you comment on this ? yesterday, i did a change to build/debian/control to add a dependance for libiw27. Please revert this change if you need. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: netcfg broken? (20040224 image)
On Wed, Feb 25, 2004 at 04:03:18PM +0100, Sven Luther wrote: > > > - Link libiw statically. This is temporary, until there is a libiw-udeb. > > > > > > Joeyh, can you comment on this ? > > > > yesterday, i did a change to build/debian/control to add a dependance > > for libiw27. Please revert this change if you need. > > So, you think the static linking mentioned above will not solve the problem ? I mean "please revert my change in build/debian/control if it's not needed". Sorry. I did this change because I was thinking that would solve the build breakage. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: netcfg broken? (20040224 image)
On Thu, Feb 26, 2004 at 10:13:28AM +0100, Sven Luther wrote: > I don't understand what happened here. netcfg was upload with a depends for libiw27, but there is no libiw udeb which can provide that. So build method which uses netcfg in the initrd were broken. I fix that with a add in build/debian/control. But if netcfg wasn't in the initrd, it failed later, so joey rebuild it with statical libiw. > So supposedly, it would need both the new netcfg (which we have) and a > rebuilding of the installer image to fix this ? once netcfg with libiw dynamic linking and libiw-udeb are in the archive, build/debian/control can be revert. Anyway that I understand. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#235107: discover: Run only at install
On Fri, Feb 27, 2004 at 02:56:54PM +0100, Petter Reinholdtsen wrote: > > Please disable discover after installation. It would even better to > > remove the package. > > I find it very convenient that it work flawlessly to move the hard > drive into another host without having to modify anything to get the > new hardware configured as it should. Please keep discover installed > and enabled also after installation. I agree. And even more, once discover2 will replace discover1, it will work flawless with 2.4 and 2.6 kernel, and you can't do that with /etc/modules -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: release status
On Mon, Feb 23, 2004 at 06:05:25PM -0800, Jeff Bailey wrote: > On Mon, Feb 23, 2004 at 08:39:23PM -0500, Joey Hess wrote: > > > - Make sure the debian-installer package builds for your arch in a > >clean chroot, and let me know so I can add your architecture to the > >control file. Currently only i386, powerpc, mips, and ia64 are listed. > > I'll check this for sparc tomorrow. Hi, since I didn't hear any reports about this, I did a check with pbuilder and a sid environment. debian-installer package for sparc will build, but with these changes : demo must be disabled in config/sparc.cfg. I took the right to do it in CVS. I also did a copy of INSTALLATION-HOWTO in build/debian, but I'm sure that I miss something in the build process. If somebody can confirm this result, you may add sparc in build/debian/control. current build process : cd build cp ../doc/INSTALLATION-HOWTO debian/ dpkg-buildpackage -S cd .. sparc32 pbuilder build debian-installer_20040304.dsc -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: release status
On Fri, Mar 05, 2004 at 05:34:09PM -0900, Joey Hess wrote: > Thomas Poindessous wrote: > > demo must be disabled in config/sparc.cfg. I took the right to do it in > > CVS. > > Hmm, that must make the debian-installer-demo package rather useless. yes, I didn't realize that, sorry :( > What's the problem with building the demo? some variables used by pkg-list weren't defined, then no udeb was installed, and later building failed. I corrected that. I redid a test (not in pbuilder) and it does work. Maybe sparc can be added to Architectures line in debian/control ? Thanks ! PS : I did a dpkg -c on produced .deb and it did look good. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
correction of hppa build failure and new changelog version
Hi, I read this http://buildd.debian.org/fetch.php?&pkg=debian-installer&ver=20040303&arch=hppa&stamp=1078663228&file=log&as=raw and to correct this failure, I added a build depends on palo [hppa]. To do this, I was forced to make a new entry in build/debian/changelog. I hope I did the right thing, please correct my mistake otherwise. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#236364: SPARCstation 5 installation report
On Fri, Mar 05, 2004 at 03:13:00PM -0700, Peter Karbaliotis wrote: > This time it actually got to the point of installing a kernel image. > Unfortunately, the installer chose kernel-image-2.4.24-sparc64-smp when > it should have chosen kernel-image-2.4.24-sparc32. I think this is fixed in cvs version of base-installer. Default kernel version wasn't the lastest. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#236973: Successful but troubled installation on mips (Indy R4k)
On Mon, Mar 08, 2004 at 10:38:23PM -0500, Nicholas Breen wrote: > Package: installation-reports > > #236970.) Fortunately, this also turned up a different problem: > the kernel-image-2.4.22-r5k-ip22 package had been installed, which > will not boot on this machine! Installing the r4k-ip22 image > instead and rerunning arcboot allowed the system to boot successfully. Hi, if you re-do a installation, could you try to launch /bin/archdetect and reports the output. Thanks ! PS : if it doesn't output mips/r4k-ip22, could you send /proc/cpuinfo -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237075: Workaround: disable framebuffer
On Tue, Mar 09, 2004 at 06:14:00PM +0100, Reinhard Tartler wrote: > > I managed to track the problem down to bterm, which segfaults. I could > use the installer passing the boot option > "debian-installer/framebuffer=false". > > This bug seems to be related to #232720. now that sparc image are autobuilding, you can use : http://ftp.uk.debian.org/debian/dists/unstable/main/installer-sparc/ it's not daily build, but these should work -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#220616: discover: actually, the problem is Discover doesn't know anything about ALSA
On Thu, Mar 11, 2004 at 11:18:28AM +0200, Martin-Éric Racine wrote: > Package: discover > Version: 1.5-2 > Severity: normal > Followup-For: Bug #220616 > > In my case, I use a 2.6 kernel with ALSA modules. Discover tries to load the OSS > module called es1371, which of > course fails, since what I have is the ALSA modules ens1371. This means that > Discover's database should be able to > use ALSA equivalents to the old OSS modules, if they exist. Boolean OR case: OSS or > ALSA modules for card XXX. kernel 2.6.* have OSS drivers, and discover1 can't provide boolean logic. Discover2 will do that. With discover1, we need to have a single data list, so, with 2.2, 2.4 and 2.6, and alsa,oss, it always will be wrong for something. I think current discover-data is focused on 2.4 support, with oss for sound. > Then again, having Discover detect the sound driver is the wrong approach, since > alsa-utils has its own configurator > and its own sound module loader, which is executed in the second stage (S20alsa in > rc2.d to rc5.d) of the boot process. If you don't want discover to try loading of es1371, just add "skip es1371" in /etc/discover.conf. But the logic is right : * if you want to use oss driver, discover will load it for you (2.4 and 2.6 will work in this case) * if you want to use alsa driver, just add a skip in discover.conf and use alsa module loader. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: ide-cd error in d-i for 2.4.26-sparc64 kernel (sarge-sparc-businesscard / 20040603)
On Fri, Jun 04, 2004 at 03:14:23PM +0200, Wiktor Wandachowicz wrote: > I am unable to load ide-cd.o module for 2.4.26-sparc64 kernel. It's a known problem. busybox can't handle sparc64 module. > I really would like to have a usable debian sarge on the > SunBlade 150 / sparc64 machine, so I could do some tests > for this architecture. Is anybody interested? you can do a netboot install, it works quite well. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#244886: kbd-chooser: can't select keymap on sparc64 netboot 20040411 image
On Thu, Jun 10, 2004 at 05:04:24PM +0100, Alastair McKinstry wrote: > Hi, > > Can you please confirm whether or not you still see this problem? > > I do not have access to a sparc machine and cannot debug > this without help Hi, I just tried it with last netboot image in sid (url attached). When I boot with DEBCONF_PRIORITY=low, I answer : * langage : en (default) * country : united states * keyboard : sun (choices possibles: sun or usb) and then kbd-chooser assumes that I have a sunkeymap keyboard. But it didn't ask me to choose. If I choose United Kingdom for country, it chooses sun5uk* and sun5fr with France as country. I think that with "low" as priority, it should list available keymap. Thanks. PS : if you want me to do some debugging, please ask. I can even provide you a root account via ssh on this box (in day hours (UTC+2)). -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[PATCH] small error in kbd-chooser.c
I discovered a small error in kbd-chooser.c. I'm not sure that kbd-chooser can parse such files as /usr/share/console/lists/keymaps/console-data.keymaps -- Thomas Poindessous Index: kbd-chooser.c === RCS file: /cvs/debian-boot/debian-installer/tools/kbd-chooser/kbd-chooser.c,v retrieving revision 1.27 diff -u -r1.27 kbd-chooser.c --- kbd-chooser.c 15 Jul 2003 08:29:40 - 1.27 +++ kbd-chooser.c 24 Jul 2003 22:26:53 - @@ -446,7 +446,7 @@ if (strncmp (ent->d_name, "console-keymaps-", 16) == 0) strcpy (p, ent->d_name); else - strncpy (p, ent->d_name, strchr (ent->d_name, '.') - p); + strncpy (p, ent->d_name, strchr (ent->d_name, '.') - ent->d_name); maplist_select (maplist_parse_file (fullname)); } }
kbd-chooser proposes a wrong default
Hi, I did a install of sarge from jigdo-1 of 2003/07/18. I chose french language ( from France ), and in the next step, kbd-choose proposed be-latin1 as default choice. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#203135: kbd-chooser: [PATCH] small error and can't handle other file than console-keymaps-*
Package: kbd-chooser
Version: 0.23 (not installed)
Severity: normal
Tags: patch
Hi,
here are 2 patchs. The first one corrects a small error. The second one
makes kbd-chooser handling only console-keymaps-* files. In
maplist_parse_file:374, there is a strlen("console-keymaps-").
thanks for your work.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux thomas 2.4.21-3-k7 #1 Sun Jul 20 19:23:36 EST 2003 i686
Locale: LANG=C, LC_CTYPE=C
--
Thomas Poindessous
Index: kbd-chooser.c
===
RCS file: /cvs/debian-boot/debian-installer/tools/kbd-chooser/kbd-chooser.c,v
retrieving revision 1.27
diff -u -r1.27 kbd-chooser.c
--- kbd-chooser.c 15 Jul 2003 08:29:40 - 1.27
+++ kbd-chooser.c 27 Jul 2003 22:16:48 -
@@ -446,7 +446,7 @@
if (strncmp (ent->d_name, "console-keymaps-", 16) == 0)
strcpy (p, ent->d_name);
else
- strncpy (p, ent->d_name, strchr (ent->d_name, '.') -
p);
+ strncpy (p, ent->d_name, strchr (ent->d_name, '.') -
ent->d_name);
maplist_select (maplist_parse_file (fullname));
}
}
Index: kbd-chooser.c
===
RCS file: /cvs/debian-boot/debian-installer/tools/kbd-chooser/kbd-chooser.c,v
retrieving revision 1.27
diff -u -r1.27 kbd-chooser.c
--- kbd-chooser.c 15 Jul 2003 08:29:40 - 1.27
+++ kbd-chooser.c 27 Jul 2003 22:17:39 -
@@ -443,11 +443,10 @@
/* two types of name allowed (for the moment; )
* legacy 'console-keymaps-* names and *.keymaps names
*/
- if (strncmp (ent->d_name, "console-keymaps-", 16) == 0)
+ if (strncmp (ent->d_name, "console-keymaps-", 16) == 0) {
strcpy (p, ent->d_name);
- else
- strncpy (p, ent->d_name, strchr (ent->d_name, '.') -
p);
- maplist_select (maplist_parse_file (fullname));
+ maplist_select (maplist_parse_file (fullname));
+ }
}
}
Bug#203260: kbd-chooser preselects be-latin1 after choosing french language ( fr_FR@euro )
Package: kbd-chooser Version: 0.24 (not installed) Severity: normal Hi, I did a install with current cvs. I chose french language ( [EMAIL PROTECTED] set for debian-installer/locale ). But kbd-chooser preselected be-latin1 ( Belgian ). -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux thomas 2.4.21-3-k7 #1 Sun Jul 20 19:23:36 EST 2003 i686 Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
main-menu and kbd-chooser from CVS unbuildable
Hi, main-menu and kbd-chooser are unbuildable because of a wrong changelog. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[PATCH] updated pkg-lists for console-keymaps-at transition
Hi,
since console-keymaps-at now replaces console-keymaps-ps2, pkg-lists
files need to be updated.
Here is a patch.
Thanks.
--
Thomas Poindessous
Index: cdrom/alpha
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/cdrom/alpha,v
retrieving revision 1.5
diff -u -r1.5 alpha
--- cdrom/alpha 21 May 2003 09:43:34 - 1.5
+++ cdrom/alpha 7 Aug 2003 12:50:02 -
@@ -4,5 +4,5 @@
usb-modules-${kernel:Version}-udeb
input-modules-${kernel:Version}-udeb
console-keymaps-usb
-console-keymaps-ps2
+console-keymaps-at
kbd-chooser
Index: cdrom/arm
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/cdrom/arm,v
retrieving revision 1.2
diff -u -r1.2 arm
--- cdrom/arm 5 Apr 2003 08:16:18 - 1.2
+++ cdrom/arm 7 Aug 2003 12:50:02 -
@@ -1,4 +1,4 @@
console-keymaps-acorn
-console-keymaps-ps2
+console-keymaps-at
console-keymaps-usb
kbd-chooser
Index: cdrom/hppa
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/cdrom/hppa,v
retrieving revision 1.2
diff -u -r1.2 hppa
--- cdrom/hppa 5 Apr 2003 08:16:18 - 1.2
+++ cdrom/hppa 7 Aug 2003 12:50:02 -
@@ -1,4 +1,4 @@
console-keymaps-usb
-console-keymaps-ps2
+console-keymaps-at
brltty-udeb
kbd-chooser
Index: cdrom/i386
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/cdrom/i386,v
retrieving revision 1.13
diff -u -r1.13 i386
--- cdrom/i386 16 Jul 2003 20:17:09 - 1.13
+++ cdrom/i386 7 Aug 2003 12:50:02 -
@@ -1,7 +1,7 @@
isa-pnp-modules-${kernel:Version}-udeb
socket-modules-${kernel:Version}-udeb
console-keymaps-usb
-console-keymaps-ps2
+console-keymaps-at
floppy-modules-${kernel:Version}-udeb
cdrom-modules-${kernel:Version}-udeb
ide-modules-${kernel:Version}-udeb
Index: cdrom/ia64
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/cdrom/ia64,v
retrieving revision 1.3
diff -u -r1.3 ia64
--- cdrom/ia64 21 May 2003 22:19:04 - 1.3
+++ cdrom/ia64 7 Aug 2003 12:50:02 -
@@ -1,6 +1,6 @@
efi-reader
kbd-chooser
-console-keymaps-ps2
+console-keymaps-at
console-keymaps-usb
usb-modules-${kernel:Version}-udeb
input-modules-${kernel:Version}-udeb
Index: cdrom/powerpc
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/cdrom/powerpc,v
retrieving revision 1.5
diff -u -r1.5 powerpc
--- cdrom/powerpc 28 Jun 2003 05:20:14 - 1.5
+++ cdrom/powerpc 7 Aug 2003 12:50:02 -
@@ -3,6 +3,6 @@
#usb-modules-${kernel:Version}-udeb
#input-modules-${kernel:Version}-udeb
console-keymaps-amiga
-console-keymaps-ps2
+console-keymaps-at
console-keymaps-usb
kbd-chooser
Index: floppy/powerpc
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/floppy/powerpc,v
retrieving revision 1.1.1.1
diff -u -r1.1.1.1 powerpc
--- floppy/powerpc 14 Jul 2003 09:33:20 - 1.1.1.1
+++ floppy/powerpc 7 Aug 2003 12:50:02 -
@@ -6,6 +6,6 @@
#usb-modules-${kernel:Version}-udeb
#input-modules-${kernel:Version}-udeb
console-keymaps-amiga
-console-keymaps-ps2
+console-keymaps-at
console-keymaps-usb
kbd-chooser
Index: net_drivers/i386
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/net_drivers/i386,v
retrieving revision 1.2
diff -u -r1.2 i386
--- net_drivers/i38625 Jul 2003 13:00:34 - 1.2
+++ net_drivers/i3867 Aug 2003 12:50:02 -
@@ -10,7 +10,7 @@
# Don't install these; save space for pcmcia modules
# ipv6-modules-${kernel:Version}-udeb
# irda-modules-${kernel:Version}-udeb
-console-keymaps-ps2
+console-keymaps-at
console-keymaps-usb
# Comment out pcmcia stuff until it is available in the archive
# pcmcia-cs-udeb
Index: netboot/alpha
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/netboot/alpha,v
retrieving revision 1.3
diff -u -r1.3 alpha
--- netboot/alpha 21 May 2003 22:25:10 - 1.3
+++ netboot/alpha 7 Aug 2003 12:50:02 -
@@ -4,6 +4,6 @@
usb-modules-${kernel:Version}-udeb
input-modules-${kernel:Version}-udeb
console-keymaps-usb
-console-keymaps-ps2
+console-keymaps-at
kbd-chooser
netcfg-dhcp
Index: netboot/i386
===
RCS file: /cvs/debian-boot/debian-installer/build/pkg-lists/netboot/i386,v
retrieving revision 1.12
diff -u -r1.12 i386
--- netboot/i38625 Jul 2003 13:02:15 - 1.12
+++ netboot/i3867 Aug 2003 12:50:02 -
@@ -1,4 +1,4 @@
-console-keymaps-ps2
+console-keyma
Re: kbd-chooser segfault
On Mon, Aug 11, 2003 at 09:28:28PM -0400, Noah L. Meyerhans wrote: > Hi. I'm in the process of my first attempt at using d-i in quite some > time, and I'm seeing a segfault in kbd-chooser. The output on the > display indicates that the return code is 35584 and that the segfault > was in the postinst script. Beyond that, I don't see anything notable > in /var/log/syslog or stderr.log. > > I am using the Aug. 11 netinst iso from > http://gluck.debian.org/cdimage/testing/netinst/i386/ on a fairly > un-interesting Athlon system equipped with a fairly standard > PS2-attached Logitech 104 key keyboard. what keymap did you choose ? for example, fr-latin1 doesn't work. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#214441: debian-installer: should suggest a sane default kernel
On Mon, Oct 06, 2003 at 10:33:10PM +0200, Goswin von Brederlow wrote: > Andre Lehovich <[EMAIL PROTECTED]> writes: > > Type of pu is easy. Number of cpus is tricky. Unless you use a SMP > enabled kernel you don't see the extra cpus in /proc/cpuinfo. This question was asked several months ago. On a SMP machine with a no-smp kernel, you can do a grep on dmesg to verify it's a SMP machine. It works on x86, maybe not on another arch. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#214441: debian-installer: should suggest a sane default kernel
On Tue, Oct 07, 2003 at 08:49:08AM -0700, Matt Kraai wrote: > On Tue, Oct 07, 2003 at 12:41:14PM +0200, Thomas Poindessous wrote: > > This question was asked several months ago. On a SMP machine with a > > no-smp kernel, you can do a grep on dmesg to verify it's a SMP machine. > > It works on x86, maybe not on another arch. > > That's the first I'd heard that. Do you know what we should > search for? I found the reference : http://lists.debian.org/debian-devel/2002/debian-devel-200211/msg02903.html I will test it during this week. Please remember me if I don't. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#214441: debian-installer: should suggest a sane default kernel
On Tue, Oct 07, 2003 at 09:16:32AM -0700, Matt Kraai wrote: > > I found the reference : > > > > http://lists.debian.org/debian-devel/2002/debian-devel-200211/msg02903.html > > > > I will test it during this week. Please remember me if I don't. > > That says it works with an SMP kernel but not with a non-SMP > kernel. I did some tests. On a smp system ( 2xPIII ) : - bf24 from woody netinstall : dmesg | grep Processors doesn't work. - mandrake 9.1 with uniproc kernel : dmesg | grep Processors returns "Processors: 2" and /proc/cpuinfo indicates only one processor. Now, since mandrake's kernel is heavily patched, maybe it's a feature from a patch. The config-2.4.21-0.13mdk is in this rpm file : ftp://speakeasy.rpmfind.net/linux/Mandrake/9.1/i586/Mandrake/RPMS/kernel-2.4.21.0.13mdk-1-1mdk.i586.rpm I will try to test with the latest kernel from sid. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#214441: debian-installer: should suggest a sane default kernel
On Wed, Oct 08, 2003 at 12:08:51AM +0200, Thomas Poindessous wrote: > On Tue, Oct 07, 2003 at 09:16:32AM -0700, Matt Kraai wrote: > I did some tests. > > On a smp system ( 2xPIII ) : I re-did some test on the same system : I installed debian woody, and with bf24, it works, as you can see in the script attached. -- Thomas Poindessous Script started on Wed Oct 8 03:01:15 2003 bi-p3:~# uname -a Linux bi-p3 2.4.18-bf2.4 #1 Son Apr 14 09:53:28 CEST 2002 i686 GNU/Linux bi-p3:~# dmesg | grep Processors Processors: 2 bi-p3:~# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 8 model name : Pentium III (Coppermine) stepping: 6 cpu MHz : 701.613 cache size : 256 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse bogomips: 1399.19 bi-p3:~# exit Script done on Wed Oct 8 03:01:35 2003
Bug#237529: kernel should be chosen according to the installed CPU
On Fri, Mar 12, 2004 at 12:09:02AM +0100, Florian Effenberger wrote: > Package: debian-installer > Version: 20040310 > > The kernel used after installation should be chosen according to the > installed CPU. it tries to do that, but the detection logic isn't finished. It's related to bug #214441 and to bug #228252 (this one should be closed I think, I will test it tomorrow), and bug #229062, bug #229570. Anyway, it's not that easy to find the correct kernel. I don't thing it's a good idea to parse model name, it's may be a better idea to use cpu family and model. If anyone knows the logic between cpuinfo and kernel packages, please tell :) Otherwise, I will be apply to receive a lot of /proc/cpuinfo to determine this logic ( please tell me which kernel works for your cpuinfo). Thanks -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237531: debconf should be configured with base-config
On Fri, Mar 12, 2004 at 12:14:26AM +0100, Florian Effenberger wrote: > Package: base-config > Version: 2.14 > > debconf should be configured (dpkg-reconfigure debconf) with base-config to > choose the question priority. Hi, I think that this question is asked if you choose a lower priority before booting. This bug and a lot of your previous bug reports are "features" since the willing of debian-installer is to ask the less possible questions. If you want all these questions, you should boot in expert mode. But it's only my thoughts, maybe it's a real bug -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sparc64 Netbooting
On Fri, Mar 12, 2004 at 10:25:30AM +, [EMAIL PROTECTED] wrote: > NB: Please CC replies, I'm not on the list > > Anyone had any success with booting the sparc64 daily netboot images? > I've tried the images over the last few days and they all fail at the same > point. > > This is trying to boot on an Ultra60 > > If you need any other info let me know. same here with a ultra5. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sparc64 Netbooting
On Fri, Mar 12, 2004 at 10:25:30AM +, [EMAIL PROTECTED] wrote: > NB: Please CC replies, I'm not on the list > > Warning: unable to open an initial console. Ok, I found why. In initrd, /dev is not populated, and so there is no /dev/console when initrd is mounted. I don't know why, maybe it's related to devfs. Anyone has a idea ? PS : i did a mknod console c 5 1 in the initrd image and then my image was working. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Sparc64 Netbooting
On Fri, Mar 12, 2004 at 01:39:42PM +, [EMAIL PROTECTED] wrote: > > I've poked around some more. > I can at least get the system up by booting with the options > devfs=mount init=/bin/sh > > As far as I can tell it's trying to write past the end of the ram disk. > I'd like to try and let it have a much bigger ram disk, > but I'm currently lacking TFM and brain cells to make it happen. > > Tips accepted :) you need a longer options line :) with ramdisk_size but, it was working some times ago, it's not normal that it doesn't work anymore ! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237735: fails if ext3 is built into the kernel
On Sat, Mar 13, 2004 at 01:15:21AM -0500, Joey Hess wrote: > Package: partman-ext3 > Severity: normal > Tags: d-i > > On sparc, ext3 is built into the kernel. So kernelmodules_ext3 fails to > load it, does not find it in /proc/modules, and I think this makes it > not be offered as a choice. It should be simple to fix; just look in > /proc/filesystems too. The other filesystem modules probably need > similar changes. I can confirm that partman doesn't propose neither reiserfs even when it's available in /proc/filesystems -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237779: /etc/hosts is wrong
On Sat, Mar 13, 2004 at 02:16:19PM +0100, Florian Effenberger wrote: > Package: debian-installer > Version: 20040310 > > /etc/hosts is wrong. At least with DHCP setup, I have the following: > > 127.0.0.1myhostname (without domain!)localhost > > But it should be: > > 127.0.0.1localhost > 192.168.0.1 (or similar)myhostname.mydomainmyhostname but I don't think dhcp rewrite /etc/hosts, so next time you got a different ip, /etc/hosts will be wrong. I think the logic is the same for the domain. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237735: fails if ext3 is built into the kernel
On Sat, Mar 13, 2004 at 10:14:43AM -0500, Joey Hess wrote: > Thomas Poindessous wrote: > > On Sat, Mar 13, 2004 at 01:15:21AM -0500, Joey Hess wrote: > > > Package: partman-ext3 > > > Severity: normal > > > Tags: d-i > > > > > > On sparc, ext3 is built into the kernel. So kernelmodules_ext3 fails to > > > load it, does not find it in /proc/modules, and I think this makes it > > > not be offered as a choice. It should be simple to fix; just look in > > > /proc/filesystems too. The other filesystem modules probably need > > > similar changes. > > > > I can confirm that partman doesn't propose neither reiserfs even when > > it's available in /proc/filesystems > > So reiserfs is also built in on sparc? in sparc64 kernel yes. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237779: /etc/hosts is wrong
On Sat, Mar 13, 2004 at 03:34:54PM +0100, Florian Effenberger wrote: > How about a possibility of being asked what to write in /etc/hosts? > This relates to some other bug I have filed. > You have no choice to NOT use DHCP when a DHCP server is reachable. you have this choice, just boot with DEBCONF_PRIORITY=low parameter. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#237888: installation-reports: sparc64 netboot, it works !
Package: installation-reports INSTALL REPORT Debian-installer-version: 20040310 boot.img (sparc64 netboot) uname -a: Linux test 2.4.24-sparc64 #1 Fri Jan 30 18:32:22 EST 2004 sparc64 GNU/Linux Date: 2004-03-13 Method: netboot : boot net:client-ip=10.0.0.25,boot.img devfs=mount Machine: Ultra5 Processor: UltraSparc IIi 270 Mhz Memory: 192 Mb Root Device: /dev/hda1 Root Size/partition table: /dev/hda1 / /dev/hda2 swap /dev/hda3 Whole disk Output of lspci: 00:00.0 Host bridge: Sun Microsystems Computer Corp. Ultra IIi 00:01.0 PCI bridge: Sun Microsystems Computer Corp. Simba Advanced PCI Bridge (rev 11) 00:01.1 PCI bridge: Sun Microsystems Computer Corp. Simba Advanced PCI Bridge (rev 11) 01:01.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01) 01:01.1 Ethernet controller: Sun Microsystems Computer Corp. Happy Meal (rev 01) 01:02.0 VGA compatible controller: ATI Technologies Inc 3D Rage I/II 215GT [Mach64 GT] (rev 9a) 01:03.0 IDE interface: CMD Technology Inc PCI0646 (rev 03) Base System Installation Checklist: Initial boot worked:[O] Configure network HW: [O] Config network: [O] Detect CD: [ ] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [O/E] Create file systems:[O/E] Mount partitions: [O] Install base system:[O] Install boot loader:[O/E] Reboot: [O] [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Comments/Problems: I did several installs and I found several problems ... but I did a install without problems ! First, partman doesn't know how to do a automatic partitioning on the whole disk. I got this error : "can't have the end before the start" And if there is no partitions, it can't even create one. But if there are already some partitions, it can delete 2 and create 1 with the space of the two. partman only proposes ext2, but this bug is fixed I think. And last problem, silo-installer can't work with a reiserfs /. It's normal, but the error message must be more explicit. But sparc64 netboot works (at least for me) ! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238525: Sparc netboot fails on Ultra5
On Wed, Mar 17, 2004 at 01:52:14PM +, Philip Armstrong wrote: > Package: installation-reports > > The tftp transfer goes fine & the kernel boots, mounts the initrd this > time (unlike the CD install I tried earlier today), but then fails > with the following errors at the end of the log: > > VFS: Mounted root (ext2 filesystem) > Warning: unable to open initial console. > Unimplemented SPARC system call 188 > Unimplemented SPARC system call 188 > kernel panic: attempted to kill init! you need to boot with some args : for example : boot net:dhcp devfs=mount -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238527: installation-reports
On Wed, Mar 17, 2004 at 03:05:55PM +0100, Pascal Brisset wrote:
> The network is configured correctly. I choose my local debian mirror
> ("Enter information manually") and I make a mistake. It fails. I
> "Go Back" and I get a new menu ("http" or "ftp") and I choose "http".
> Then a new list of mirrors is proposed where I cannot enter the name of
> my local mirror. From this state, I fail to retrieve the menu where I
> could enter the name of my local mirror.
you should have "enter the information manually" on the top of the
mirror list.
--
Thomas Poindessous
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238525: Sparc netboot fails on Ultra5
On Wed, Mar 17, 2004 at 05:29:52PM +, Philip Armstrong wrote: > > boot net:dhcp devfs=mount > > Ah ha. Thanks. > > Why isn't devfs mounting on by default if it won't boot without it? I don't know. Maybe because there is no way to store kernel parameters in tftp image ? it's just a supposition. But, yes, I will be very happy if there was no need to pass this param :) -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238525: Sparc netboot fails on Ultra5
On Wed, Mar 17, 2004 at 05:29:52PM +, Philip Armstrong wrote: > Ah ha. Thanks. > > Why isn't devfs mounting on by default if it won't boot without it? from the changelog of d-i : * Joey Hess - Always include /dev/console, even for images that will use devfs. This means that netboot users will not have to remember the devfs=mount. So, everything is (now) fine :) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#237075: Bug not related to Sparc64... Maybe related to Creator?
On Fri, Mar 19, 2004 at 12:16:51PM +0100, Reinhard Tartler wrote: > On Tue, Mar 16, 2004 at 03:50:27PM -0800, Blars Blarson wrote: > > Well, due to the fact that I get the debian-installer menu when booting > with the option "framebuffer=false" I think that installing via serial > console works. But that doesn't resolve the Problem that bterm is broken > on sparc64, i.e. that displaying UTF-8 characters is not possible. I think this problem is resolved. You were using unofficial image of d-i made by Jeff Bailey, and it seems that his build system wasn't using the last version of libbogl-dev. > This Problem should at least be documented bevore d-i ist released. Could you please retry with a official build ? Thanks PS : you can use image from http://ftp.debian.org/debian/dists/unstable/main/installer-sparc/ -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238857: Installation Report on Sparc32
On Fri, Mar 19, 2004 at 01:52:03PM +0100, Peter wrote: > Package: installation-reports > > I burned the sparc/beta3/sarge-sparc-businesscard.iso > and booted from it, but after booting the kernel it hangs with a > kernelpanic and the message > VFS: Unable to mount root fs on 00:00 > Press L1-A to return to the boot prom. > > After trying again with the argument linux root=/dev/sda1 it proceeds to > boot into my existing Woody install, but not into the Debian installer. Could you test it with these arguments : devfs=mount root=/dev/rd/0 and maybe add these also : ramdisk_size=8192 thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#237075: Bug not related to Sparc64... Maybe related to Creator?
On Fri, Mar 19, 2004 at 03:18:47PM +0100, Reinhard Tartler wrote: > This image still has the problem with "unable to open initial console" > so the option "devfs=mount" was needed. Not so in Jeff Bailey's build. > > bterm keeps on segfaulting with this image too. then you indeed found a bug in bterm Could test this : booting with "devfs=mount init=/bin/sh" options you will get a shell and then bterm -f unifont.bgf and maybe also bterm -f unifont.bgf -l C.UTF-8 you should get an error message. > > PS : you can use image from > > http://ftp.debian.org/debian/dists/unstable/main/installer-sparc/ > > These are not the daily builds, are they? I used Jeff's because they > seemed to be newer. They also boot without that "devfs=mount" option. Jeff's images are newer because they are builded from daily cvs. Images in installer-sparc/ are builded when a new version of debian-installer (the debian package) is released. I was thinking that your bug was coming from a old version of libbog-dev on Jeff's system. But I was obviously wrong. I hope you will find the error ! Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Bug#237075: Bug not related to Sparc64... Maybe related to Creator?
On Fri, Mar 19, 2004 at 05:35:19PM +0100, Reinhard Tartler wrote: > On Fri, Mar 19, 2004 at 05:20:15PM +0100, Thomas Poindessous wrote: > > nope. The screen in the upper half gets messed up, long thin horizontal > blue lines, and nothings happens. > > As soon as I press any key i get a segmentation fault. > > I tested this with Jeff's image but I don't think that this result will > differ from beta3. > > Any other ideas I could try? yes, can you test bterm with the unifont.bgf in your console on a running debian sparc system ? If it stills segfaults, it will be far more easier to debug it :) If you don't have the unifont.bgf (from cvs's building), I can send it to you. Many thanks ! -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#239154: sparc ultra 2 beta 3 failure
On Sat, Mar 20, 2004 at 08:01:33PM -0800, Blars Blarson wrote: > Same system as previous succesfull install except keyboard and monitor > rather than serial console. > > Puts red garbage on upper half of screen than loops displaying > "Segmentation fault" stairsteping on the screen. You may have the same problem that bug #237075 I was trying to help Reinhard Tartler to debug bterm with cvs's unifont.bgf, but if you have the time, you are far more competent than me :), and you have the hardware. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#239281: Another implementation for selecting right kernel for x86 cpu
On Sun, Mar 21, 2004 at 08:39:22PM -0800, Matt Kraai wrote: > On Mon, Mar 22, 2004 at 01:34:34AM +0100, Frans Pop wrote: > > I've just fixed this in the repository. I was working on another implementation. I just finished it. It uses /proc/cpuinfo, but it uses vendor_id and cpu family instead of model name. It has the advantages of being more precise and more generic. It supports a lot of Intel and Amd model, and it will be easy to add other models. I tested it with a lot of cpuinfo available from internet and I used also the db from Gentoo poll. I can provide more explanations if you want. I think this patch will close some bugs on base-installer (every bugs which says that the wrong kernel was selected for intel and amd cpu). The only bug from my implementation is that there is no more -586tsc for 2.6.x kernels. Thanks for reviewing this patch and for applying it if it fits. -- Thomas Poindessous Index: postinst === --- postinst(revision 11701) +++ postinst(working copy) @@ -223,28 +223,41 @@ i386*) version=2.4 MODEL=`grep 'model name' /proc/cpuinfo | cut -d: -f2` + VENDOR=`grep '^vendor_id' /proc/cpuinfo | cut -d: -f2` + FAMILY=`grep '^cpu family' /proc/cpuinfo | cut -d: -f2` if dmesg | grep -q ^Processors:; then SMP=yes else SMP=no fi - case "$MODEL" in - " Intel(R) Pentium(R) [4M]"*|" Pentium II "*|" Pentium III "*) - if test $SMP = no; then - trykernel=kernel-image-$version-686 - else - trykernel=kernel-image-$version-686-smp - fi - ;; - " AMD Athlon(tm) "*|" AMD Duron(tm) "*) - if test $SMP = no; then + case "$VENDOR" in + " AuthenticAMD"*) + if test "$FAMILY" = " 6"; then + if test $SMP = no; then trykernel=kernel-image-$version-k7 - else + else trykernel=kernel-image-$version-k7-smp + fi + elif test "$FAMILY" = " 5"; then + trykernel=kernel-image-$version-k6 fi ;; - " AMD-K6(tm) "*) - trykernel=kernel-image-$version-k6 + " GenuineIntel"*) + if test "$FAMILY" = " 6"; then + if test $SMP = no; then + trykernel=kernel-image-$version-686 + else + trykernel=kernel-image-$version-686-smp + fi + elif test "$FAMILY" = " 15"; then + if test $SMP = no; then + trykernel=kernel-image-$version-686 + else + trykernel=kernel-image-$version-686-smp + fi + elif test "$FAMILY" = " 5"; then + trykernel=kernel-image-$version-586tsc + fi ;; *) trykernel=kernel-image-$version-386
Bug#239414: /etc/init.d/discover not removed when package is removed
On Mon, Mar 22, 2004 at 05:26:04PM +0100, Martin Leopold wrote: > Package: discover > Version: 1.5-1.1 > Severity: normal > > Trying to get rid of discover (it messed up my alsa driver) I learned > that removing discover doesn't stop it from running - this can't be > right. While I can se the sense in not removing configuation files it > doen't make sense to leave the program running after I've removed it. discover is run only one time in a normal boot operation. It detects what modules you need and loads them. If you remove discover, you will remove the program. But if you don't reboot, modules loaded by discover aren't unloaded, and that's normal. So if discover was loading oss module instead of your alsa modules, once you remove it, please reboot. If your problem is something, please pardon me and please give extra explanations. Thanks ! -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#239414: [leopold@diku.dk: Re: Bug#239414: /etc/init.d/discover not removed when package is removed]
I think I was only to get this private response. -- Thomas Poindessous --- Begin Message --- Thomas Poindessous wrote: Trying to get rid of discover (it messed up my alsa driver) I learned that removing discover doesn't stop it from running - this can't be right. While I can se the sense in not removing configuation files it doen't make sense to leave the program running after I've removed it. discover is run only one time in a normal boot operation. It detects what modules you need and loads them. If you remove discover, you will remove the program. But if you don't reboot, modules loaded by discover aren't unloaded, and that's normal. So if discover was loading oss module instead of your alsa modules, once you remove it, please reboot. If your problem is something, please pardon me and please give extra explanations. The sequence of events: 1. apt-get remove discover 2. reboot 3. Upon booting discover is still run and istalls OSS drivers that I don't want 4. dpkg --purge discover removes everything 5. Reboot 6. Discover is not run and the alsa scripts loads my drivers --- End Message ---
Bug#239943: installation-reports
On Thu, Mar 25, 2004 at 12:33:54AM +0100, Michel van der Klei wrote: > Processor: Dual P3 800 Mhz > > There was a wrong kernel installed (K7) and grub could not read the root partition. hi, can you provide /proc/cpuinfo for this box ? Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#239943: [reproduced]
On Thu, Mar 25, 2004 at 09:13:26PM +0100, Michel van der Klei wrote: > On Thu, 2004-03-25 at 17:52, Michel van der Klei wrote: > > On an old P3 500 Compaq machine i performed an installation of Sarge > first. The installation was awsome. No problems found. > > After that i formated the partitions and tried to perform an install of > Woody. All worked out partioning, base-install no problems found. > > When i received the message "Installing the kernel" i switched to tty3 > to see what kernel there was going to be installed, it was the k7 > kernel. Grub installed fine. Are you trying to install woody (stable) with sarge (testing) debian-installer ? I don't think this operation is supported. And for the wrong kernel selection, a new selecting mode is written, but it's not yet released and in unstable. Thanks. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#239943: [reproduced]
On Thu, Mar 25, 2004 at 11:01:34PM +0100, Michel van der Klei wrote: > Hi Thomas, > > > I don't think this operation is supported. > > That operation is supported. With the businesscard cd you have the > possibillity to choose between stable, testing and unstable. Yes, I know this, but I think this is only for the futur, when sarge will be the "stable" release. So when release manager decides to release sarge as stable, tools doesn't need to be changed. But I can be wrong. -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#240372: SPARCstation 5 installation report
On Sat, Mar 27, 2004 at 03:10:18PM -0800, Joshua Kwan wrote: > On Fri, 26 Mar 2004 17:03:00 -0700, Peter Karbaliotis wrote: > > WARNING: Your /etc/fstab does not contain the fsck passno field. I will > > kludge around things for you, but you should fix your /etc/fstab file as > > soon as you can. > > > > And then the boot hangs because the root filesystem is read-only > > Could you attach the fstab file from the machine if that's possible? I don't know if it's related but current unstable installation won't work after first reboot, because of a bug in kernel and mount -o remount. It was discuted in debian-sparc http://lists.debian.org/debian-sparc/2004/debian-sparc-200403/msg00323.html http://lists.debian.org/debian-sparc/2004/debian-sparc-200403/msg00307.html -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#241073: hppa sarge CD netinst fails on 715/100 (segfault)
On Tue, Mar 30, 2004 at 10:58:17AM -0500, Greg Wooledge wrote: > Setting up filesystem, please wait .. > umount: /initrd: Invalid argument > Segmentation fault > Segmentation fault > Segmentation fault This cames from bterm which has a problem on your system. We have other reports on sparc which have the same probleme. It seems to be a problem with framebuffer (not sure). > The suggestion to use 'DEBCONF_PRIORITY=low' was made in IRC, but I don't > know precisely how to *do* that. I tried booting the CD with > 'boot scsi.2.0 DEBCONF_PRIORITY=low' but it made no visible difference. can you try with disabling framebuffer : debian-installer/framebuffer=false If it doesn't help, it will need some debugging on bterm ... -- Thomas Poindessous -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
newt boxes displaying problems (translation problem)
I think I have found why in a french install, some boxes were displaying over others. I think it's because some french translations are too long and boxes are larger than screen width. I'm not sure, I will test tonight with shorter translations, but it seems logical. I have tested with other translations and there is not problems. If it is that, can you contact french translator please ? The translations with problems are : #: main_menu.c:166 main_menu.c:208 main_menu.c:478 msgid "Initialize and Activate a Swap Partition" #: main_menu.c:167 main_menu.c:209 main_menu.c:479 msgid "Activate a Previously-Initialized Swap Partition" #: main_menu.c:163 msgid "Preload modules from a floppy" But I think there are others ... The good solution would be finding the max possible length ... -- Thomas Poindessous [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
