http://blog.gmane.org/gmane.os.netbsd.ports.arm hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.os.netbsd.ports.arm/3377 <pre>I'm thinking of ordering one of these for the heck of it: http://apc.io/about/ Does anyone know of a reason why this wouldn't be supported with NetBSD? It might be too early to know I guess. There are other machines out there that are interesting as well. The Raspberry Pi won't be available for a while so one of these other alternatives might be a better option. Andy </pre> Andy Ruhl 2012-05-25T15:42:21 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3367 <pre>Hi all, I'm using NetBSD-current on a D-Link DNS-325 NAS. It's a Marvell Kirkwood 88F6281, which appears to be supported by the evbarm port according to the kernel source. My problem is a hang when booting. I've made myself a kernel using the sys/arch/evbarm/conf/DNS323 configuration. The DNS-323 is actually a different SoC, but I believe this is discovered by mvsoc_model(), and I can confirm that kirkwood_intr_bootstrap() is called. In other words, I think makeoptions BOARDTYPE="dns323" and options EVBARM_BOARDTYPE=dns323 are not actually used, and therefore that using the DNS323 on a 325 machine is okay. Booting from U-boot: Marvell&gt;&gt; tftpboot 2000000 netbsd.gz.ub; bootm 2000000 [snip] Bytes transferred = 2117011 (204d93 hex) ## Booting image at 02000000 ... Image Name: NetBSD/dns323 6.99.7 Created: 2012-05-21 22:17:58 UTC Image Type: ARM NetBSD Kernel Image (gzip compressed) Data Size: 2116947 Bytes = 2 MB Load Address: 00008000 Entry Point: 00008000 Verifying Checksum ... OK Uncompressing Kernel Image ... OK ## Transferring control to NetBSD stage-2 loader (at address 00008000) ... [silence] Poking around in sys/arch/evbarm/marvell/marvell_machdep.c I found the hang occurs on the first printf(). I presume because consinit() is not yet called. So I borrowed some code to let me write to the serial port directly, and I'm using that for debugging: #define KW_UART0 (0xf1012000) #define KW_THR (0x0) #define KW_LSR (0x14) #define KW_LSR_THRE (0x1&lt;&lt;5) /* Transmit enable */ #define KW_DBG(c) \ do { \ volatile uint8_t *kw_base = (uint8_t *) KW_UART0; \ while ((kw_base[KW_LSR] &amp; KW_LSR_THRE) == 0) \ ; \ kw_base[KW_THR] = (uint8_t)(c); \ } while (0) #define KW_PUTS(s) \ do { \ const char *kw_p; \ for (kw_p = (s); *kw_p; kw_p++) { \ KW_DBG(*kw_p); \ } \ } while (0) Scattering a few calls to that in marvell_machdep.c, for the first call to consinit: ... KW_PUTS("H CONSINIT BEING CALLED\n\r"); consinit(); KW_PUTS("I CONSINIT DONE\n\r"); ... And inside consinit: void consinit(void) { static int consinit_called = 0; if (consinit_called != 0) { KW_PUTS("CI 0 called already\n\r"); return; } consinit_called = 1; ... } Prints the following on boot: ## Transferring control to NetBSD stage-2 loader (at address 00008000) ... H CONSINIT BEING CALLED CI 0 called already I CONSINIT DONE So here I presume consinit_called isn't having its static storage initialised to 0, as I would expect. nm marvell_machdep.o says consinit_called does indeed live in the data segment: 000004a4 T consinit 00000000 d consinit_called.9898 And gcc's marvell_machdep.s corresponds: .data .align 2 .set .LANCHOR0,. + 0 .type consinit_called.9898, %object .size consinit_called.9898, 4 consinit_called.9898: .space 4 So I believe the data segment is not being initialised correctly. Unfortunately this is where I fall short; I have no idea where to look next, or what to do about it. Any suggestions, please? Incidentally, gcc is compiling with -fno-zero-initialized-in-bss which I presume is correct, because (I think) the bss initialisation is done by the kernel during load. Is that right? Thanks, </pre> Kate F 2012-05-22T12:42:49 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3365 <pre>I normally lurk on current-users,. but saw a message from this group and thought I'd bother you for a minute. I've recently started trying to build a new prototype for my "room status" device. It going to be doing a bunch of stuff, and I've recently purchased a Mini2440 as the processor of choice for the actual implementation. I'm having trouble getting started installing/netbooting NetBSD on the thing. Most of the programs that come with this beast are in Chinese, which would be OK if I read Chinese..... I've got the TFTP directory set up and ready to use (I already use it for netbootking diskless workstations and phones) and have (I think) all of the software I need to get it working. If anyone has any experience setting these things to light, I'd love to hear about it. </pre> Dave Burgess 2012-05-21T15:58:37 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3361 <pre>Hello, I'm looking at modding my Buffalo Linkstation Mini (http://goo.gl/25bec). Before using Linux Debian on it, I wanted to know if NetBSD supports this motherboard. According to there (http://goo.gl/3dFv0), the ARM926EJ would be supported. What is the support status for this board ? If supported, can it boot from internal disks ? And if so, is there some notes available somewhere that explains the installation process ? Thanks, Jo </pre> Joel Carnat 2012-05-07T12:28:56 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3359 <pre> What's the chance of seeing NetBSD on a Raspberry Pi? It has a Broadcom BCM2835 SoC that includes a 700MHz ARM1176JZF-S CPU core and some other nice stuff - see details e.g. at http://www.techspot.com/review/527-raspberry-pi/ - Hubert </pre> Hubert Feyrer 2012-05-05T12:29:24 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3358 <pre>Hi, I have purchased two Gumstix OVERO COMs (Tide and Water) and a Gumstix board (Tobi). I am having trouble booting NetBSD-current on these, has anyone got recent experience of doing this? Below is a console dump whilst booting the OVERO Tide. Nothing is displayed after passing control to the kernel. From looking at mailing list archives, I am expecting the following: NetBSD/evbarm (overo) booting ... I can confirm that code in gumstix_start.S does start to execute. By inserting code to print to the console I can confirm that it runs until it flushes the TLB and sets the TTB. I managed to print the following diagnostic details just before the TLB flush. Interpretation of the register values has been done by hand, most of it makes sense except for the control register suggesting that the MMU is already turned on? Kernel Start: 0x80200000 RAM size: 0x20000000 Boot Arg[0]: 0x9ff27fe0 Boot Arg[1]: 0x00000000 Boot Arg[2]: 0x9ffb482a Boot Arg[3]: 0x9ffb482a Processor ID: 0x411fc083 [31:24] Implementor = 0x41 ARM [23:20] Major revision = 0x1 [19:16] Architecture = 0xF [15:4] Primary part = 0xC08 Cortex-A8 [3:0] Revision = 0x3 Status Reg: 0x600001d3 Z, C, A, I, F, Mode=0x13 Z = Zero C = Carry A = Disable imprecise data aborts I = IRQ interrupts are disabled F = FIQ interrupts are disabled Mode = Supervisor Control Reg: 0x00c5187b V = 0 = Normal exception vectors selected I = 1 = instruction caching enabled Z = 1 = program flow prediction enabled C = 0 = data caching disabled at all levels A = 1 = strict alignment fault checking enabled M = 1 = MMU enabled TTB Reg: 0x9fff0000 Domain Reg: 0xffffffff thanks for any help you can provide Michael. -- Texas Instruments X-Loader 1.5.0 (Aug 29 2011 - 12:52:49) No NAND detected OMAP3530-GP ES3.1 Board revision: 1 Reading boot sector Loading u-boot.bin from mmc U-Boot 2010.12 (Aug 22 2011 - 09:49:35) OMAP3530-GP ES3.1, CPU-OPP2, L3-165MHz, Max CPU Clock 720 mHz Gumstix Overo board + LPDDR/NAND I2C: ready DRAM: 512 MiB NAND: 0 MiB MMC: OMAP SD/MMC: 0 *** Warning - readenv() failed, using default environment In: serial Out: serial Err: serial Board revision: 1 Direct connection on mmc2 No EEPROM on expansion board Die ID #09b20004000000000403951c07012007 Net: smc911x-0 Hit any key to stop autoboot: 0 Overo # mmc init Overo # mmc part 0 Partition Map for MMC device 0 -- Partition Type: DOS Partition Start Sector Num Sectors Type 1 128 130944 c 2 131072 3731456 83 Overo # fatls mmc 0:1 / 24240 mlo 235252 u-boot.bin 2914772 uimage 1586112 netbsd.gz.ub 3378688 netbsd.bin 5 file(s), 0 dir(s) Overo # fatload mmc 0:1 ${loadaddr} netbsd.gz.ub reading netbsd.gz.ub 1586112 bytes read Overo # bootm ${loadaddr} ## Booting kernel from Legacy Image at 82000000 ... Image Name: NetBSD/overo 6.99.4 Image Type: ARM NetBSD Kernel Image (gzip compressed) Data Size: 1586048 Bytes = 1.5 MiB Load Address: 80200000 Entry Point: 80200000 Verifying Checksum ... OK Uncompressing Kernel Image ... OK ## Transferring control to NetBSD stage-2 loader (at address 80200000) ... </pre> Michael Taylor | Omniscient 2012-04-14T07:19:01 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3350 <pre>In last week I summarized "NetBSD/evbarm supported hardware list" for NetBSD presentation at Open Source Conferenece 2012 Ehime. Is there a good place to put this file, like sys/arch/README? Any updates are also appreciated. --- configdateboard --------------------------------------------------------------------------- ADI_BRH2003/01/25ADI Eng. Big Read Head i80200 eval board ARMADILLO2102006/02/06Atmark Techno Armadillo-210 ARMADILLO92005/11/13Atmark Techno Armadillo-9 BEAGLEBOARD2008/10/22TI OMAP3530 BeagleBoard (incomplete?) CP31002006/11/08Certance IOP321 CP-3100 DEVKIT80002010/09/08Embest OMAP3530 DevKit8000 eval Kit DNS3232010/10/02D-Link DNS-323 Marvell SoC based NAS GEMINI2008/10/24Cortina Systems SL3516 eval board GUMSTIX2006/10/16PXA255 Gumstix board HDL_G2006/04/16I-O DATA HDL-G Giga LANDISK HPT53252012/03/31HP t5325 Thin Client IGEPV22010/06/16IGEPv2 OMAP3530 eval board IMX31LITE2008/04/27Freescale i.M31 DEV LITE KIT INTEGRATOR2001/10/27ARM Integrator board IQ312442003/05/14Intel IQ31244 reference board IQ803102001/09/05Intel IQ80310 eval board IQ803212002/03/27Intel IQ321 eval board IXDP4252003/04/08Intel IXDP425/IXCDP1100 development platform IXM12002002/07/15Intel IMX1200 eval board KUROBOX_PRO2010/10/02Kuroutoshikou KURO-BOX/PRO LUBBOCK2003/06/18Intel Lubbock DBPXA250 board MARBELL_NAS2010/10/02Generic Marvell SoC based NAS MINI24402012/01/30FrendlyARM Mini2440 S3C2440 SoC board MMNET_GENERIC2011/11/04Propox MMnet1002 board MPCSA_GENERIC2008/07/03MPCSA Atmel AT91RM9200 based board MV21202011/07/20HP Media Vault MV2011 Marvell Orion board NAPPI2002/07/15Netwise APlication Platform Board NETWALKER2010/11/13Sharp NetWalker NSLU22006/02/28Linksys NSLU2 (a.k.a. "Slag") OSK59122007/01/06TI OMAP 5912 OSK board OVERO2010/07/10Gumstix Inc Overo board SHEEVAPLUG2010/10/02Marvell SheevaPlug SMDK24102003/07/31Samsung SMDK2410 S3C2410 eval board SMDK28002002/11/20Samsung SMDK2800 S3C2800 eval board TEAMASA_NPWR2002/02/07Team ASA Npwr IOP310 based server appliance TEAMASA_NPWR_FC2003/12/24Team ASA NPWR-FC i80321 server appliance TISDP24202008/04/27TI OMAP 2420 eval board TISDP24302008/04/27TI OMAP 2430 eval board TOASTER2005/08/14NetBSD/toaster based on TS7200 TS72002004/12/23Technologic Systems TS-7200 board TWINTAIL2005/02/26Genetec corp. "Twintail" PXA255 eval board VIPER2005/06/06Arcom Viper PXA255 ARM board ZAO4252003/05/23NOVATEC NTNP425B "ZAO425" IXP425 eval based --- Izumi Tsutsui </pre> Izumi Tsutsui 2012-03-31T03:04:37 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3346 <pre>Hi, I'm going to enable x11 build for evbarm as in attached patch, if there is no objections. Xorg server with wsfb driver works OK on some of evbarm platforms which have LCD frame buffers. BR bsh Index: external/mit/xorg/server/drivers/Makefile =================================================================== RCS file: /cvsroot/src/external/mit/xorg/server/drivers/Makefile,v retrieving revision 1.60 diff -u -r1.60 Makefile --- external/mit/xorg/server/drivers/Makefile30 Aug 2011 04:22:56 -00001.60 +++ external/mit/xorg/server/drivers/Makefile30 Mar 2012 03:02:00 -0000 &lt; at &gt;&lt; at &gt; -147,6 +147,11 &lt; at &gt;&lt; at &gt; xf86-video-wsfb .endif# ${MACHINE} == "dreamcast" +.if ${MACHINE} == "evbarm" +SUBDIR+= \ +xf86-video-wsfb +.endif # ${MACHINE} == "evbarm" + .if ${MACHINE} == "evbmips" SUBDIR+= \ xf86-video-siliconmotion \ Index: external/mit/xorg/server/xorg-server/Makefile.common =================================================================== RCS file: /cvsroot/src/external/mit/xorg/server/xorg-server/Makefile.common,v retrieving revision 1.23 diff -u -r1.23 Makefile.common --- external/mit/xorg/server/xorg-server/Makefile.common30 Aug 2011 04:22:56 -00001.23 +++ external/mit/xorg/server/xorg-server/Makefile.common30 Mar 2012 03:02:01 -0000 &lt; at &gt;&lt; at &gt; -13,6 +13,7 &lt; at &gt;&lt; at &gt; ${MACHINE} == "cats"|| \ ${MACHINE} == "dreamcast"|| \ ${MACHINE} == "ews4800mips"|| \ + ${MACHINE} == "evbarm"|| \ ${MACHINE} == "evbmips"|| \ ${MACHINE} == "hp300"|| \ ${MACHINE} == "hpcarm"|| \ Index: external/mit/xorg/server/xorg-server/hw/xfree86/xorgos/Makefile =================================================================== RCS file: /cvsroot/src/external/mit/xorg/server/xorg-server/hw/xfree86/xorgos/Makefile,v retrieving revision 1.31 diff -u -r1.31 Makefile --- external/mit/xorg/server/xorg-server/hw/xfree86/xorgos/Makefile30 Aug 2011 04:22:56 -00001.31 +++ external/mit/xorg/server/xorg-server/hw/xfree86/xorgos/Makefile30 Mar 2012 03:02:01 -0000 &lt; at &gt;&lt; at &gt; -69,6 +69,7 &lt; at &gt;&lt; at &gt; .endif .if ${MACHINE} == "cats" || \ + ${MACHINE} == "evbarm" || \ ${MACHINE} == "hpcarm" || \ ${MACHINE} == "shark" || \ ${MACHINE} == "netwinder" || \ </pre> Hiroyuki Bessho 2012-03-30T03:13:36 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3345 <pre>Hi, While trying to understand the arm interrupt controller driver code, got into some confusions. Older interrupt controller drivers which didnt use the pic interface(arch/arm/pic), calls "enable_interrupts" in their attach functions. eg. arm/arm/omap/omap_intr.c. This looks very straightforward since interrupts are expected to be enabled, when the autoconf gets completed and also after the interrupt controller is initialized. But at the same time, the newer interrupt controller drivers which uses the pic interface doesnt call enable_interrupts in their attach function. Couldnt really figure out how and where the processor interrupt is getting enabled(by clearing the I bit in CPSR). Appreciate your help in understanding this. </pre> Linu Cherian 2012-03-29T05:01:53 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3344 <pre>Hi, I'm one of the luddites that still uses a Shark as a home servlet - although with a slightly modern disk (60 GB 2.5", from the time those where relatively new) and maxed out memory (96 MB). As I mostly used it logged in from a laptop, I thought that the relatively slowness I felt in the last couple of weeks was created by my large mailboxes. However, as the two other frequent users complained (and the problem seems to have increased in the last few weeks), I investigated, and found that I had a lot of Ierrs - after a reboot, about 120% of the input packets (see netstat -i). Strange, I thought. Should the new GB 8-port compact switch have developed a thermal problem? Duplex mismatch? Cable broken? Cable exchange didn't help. ifconfig cs0 down up , and explicitly trying mediaopts, didn't help. Reverting to a 100 Mbit/s switch and to a 10 Mbit/s half-duplex hub I had around didn't help. Besides, I could ping the router through the same switch fine. Powering switch and Shark off and on again didn't help. Hmmmm, Software problem? Reverting to the old kernel I had around - 5.0.2 - didn't help. I started to consider swapping the big memory and the disk into the other Shark, when I remembered that I had swapped all Shark power supplies at work years ago, because a couple of the Sharks didn't boot any longer. The freshly booted machine is now at 30 Ierrs (constant for the last 10 hours) and 65000 packets and feels as fast as it used to be - remotely used or local. So, if you have strange errors - consider testing with a known good power supply - it might be a dried filtering capacitor. Especially those with machines, like the Shark, that use external, fanless power supplies. -is </pre> Ignatios Souvatzis 2012-02-27T07:09:05 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3339 <pre>Hello, I have in a previous post announced the Mini2440 port. It is functional, and supports most of the devices on the board, but needs further testing. Therefore I would like to encourage people with a Mini2440 at hand to try to out this port. Please note that neither NAND nor NOR flash memories are supported at this point. Currently, U-Boot is required in order to load the NetBSD bootmini2440-bootloader. NetBSD can currently be installed either to an SD-card, or the Mini2440 can run as a diskless node. Network Install to SD-card: In order to install the NetBSD MINI2440 on an SD-card follow the steps below. These instructions assume that DHCP is used to configure IPv4 addresses, and that a TFTP server is running at 10.0.0.1. All binary files needed can either be build directly from the CVS repository or obtained from the daily snapshot directory at nyftp.netbsd.org (use newest with evbarm in pub/NetBSD-daily/HEAD). * The dhcp-server must be configured with a proper “next-server” option, as in the following example /etc/dhcpd.conf: host mini2440 { hardware ethernet 08:08:11:18:12:27; fixed-address 10.0.0.20; next-server 10.0.0.1; } * TFTP must be setup, such that the bootmini2440-bootloader and netbsd-MINI2440_INSTALL kernel are downloadable: a. Start TFTP service by uncommenting “'#tftp" line in /etc/inetd.conf, and restart inetd invoking '/etc/rc.d/inetd restart' b. Populate the TFTP-directory (usually /tftpboot) with two files: 1. bootmini2440 found at evbarm/installation/ directory of ISO image. 2. netbsd-MINI2440_INSTALL found at evbarm/installation/instkernel/ directory as netbsd-MINI2440_INSTALL.gz. gunzip it and copy it to the TFTP-directory as netbsd-MINI2440_INSTALL Make sure to have READ permission for files, or TFTP daemon will not be able to serve them to the MINI2440-board. * From the U-Boot prompt, request an address via DHCP with autoloading disabled: MINI2440 # setenv autoload n MINI2440 # dhcp dm9000 i/o: 0x20000300, id: 0x90000a46 DM9000: running in 16 bit mode MAC: 08:08:11:18:12:27 BOOTP broadcast 1 BOOTP broadcast 2 DHCP client bound to address 10.0.0.20 MINI2440 # ‘printenv’ can be used to see the settings obtained from the DHCP request: autoload=n gatewayip=10.0.0.1 netmask=255.255.255.0 rootpath=/export/mini2440 ipaddr=10.0.0.20 serverip=10.0.0.1 dnsip=10.0.0.1 * Load "bootmini2440" at address 0x30a00000 over tftp by utilizing U-Boots tftp capabilities: U-Boot&gt; tftp 30a00000 bootmini2440 * Launch the bootmini2440-bootloader and ask it to fetch the install image over the network: U-Boot&gt; go 30a00000 tftp:netbsd-MINI2440_INSTALL * The installer should now launch. * When the distribution-selection screen shows, select "d: Custom installation". Select "e: Kernel (MINI2440)" and continue installation. * Setup the disk-partitioning as desired and continue installation. - Please note, that SD-cards do not behave as regular harddrive. In order to get the best performance, care should be taken that partitions start on 8192-byte boundaries and end at them. Also, it is adviced that the block size of UFS is set to 8192 (other values might be more optimal). For more details see http://www.olpcnews.com/forum/index.php?topic=4993.0. - In order to avoid loading bootmini2440 over the network, or store it in NAND, it might be a good idea to have a FAT-partition on the SD-card which can be accessed by U-Boot. * Choose "HTTP" as installation source, configuring the dme0 manually if needed. * A local server can be used as source,or the releng server can be used:: Host: nyftp.netbsd.org Base directory: pub/NetBSD-daily/HEAD/201202011750Z/ * Once installation completes, bootmini2440 is again loaded over the network: U-Boot&gt; tftp 30a00000 bootmini2440 * The NetBSD system is then booted from the SD-card: U-Boot&gt; go 30a00000 ld0a:netbsd Setup as Diskless Node: It is possible to run MINI2440 as an NFS diskless node. In that case a root-directory for the MINI2440 is create on a server, and exported via NFS. In this example /export/mini2440 is exported, and will be mounted as root file system by the MINI2440. * First, /export/mini2440 needs to be exported, this is done by having the following line in /etc/exports: /export/mini2440 -maproot=0 * Restart the NFS-server: # /etc/rc.d/mountd restart * Two files, base.tgz and etc.tgz, are to be inflated at the NFS exported directory. These files are found at evbarm/binary/sets/ directory of ISO image. *. Run MAKEDEV script to populate dev/ directory: # cd /export/mini2440/dev # sh ./MAKEDEV all *. Gunzip netbsd-MINI2440.gz, found in evbarm/binary/kernel/, and copy it as "netbsd" to the NFS exported directory. * From the U-Boot prompt, request an address via DHCP with autoloading disabled: MINI2440 # setenv autoload n MINI2440 # dhcp * run TFTP to load bootmini2440 and starts NetBSD kernel MINI2440 # tftp 30a00000 bootmini2440 dm9000 i/o: 0x20000300, id: 0x90000a46 DM9000: running in 16 bit mode MAC: 08:08:11:18:12:27 TFTP from server 192.168.24.24; our IP address is 192.168.24.69 Filename 'bootmini2440'. Load address: 0x30a00000 Loading: T ######### done Bytes transferred = 117628 (1cb7c hex) MINI2440 # go 30a00000 nfs:netbsd bootmini2440 loads NetBSD kernel with the NFS protocol, and the Mini2440 runs as NFS diskless node. For more information please visit http://xpg.dk/nb-mini2440-ht/ Kind regards, Paul </pre> Paul Fleischer 2012-02-12T11:40:27 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3333 <pre> I'm running a recent -current evbarm on my Marvell Orion box (HP MV2120). All pthreaded processes segfault due to __lwp_getprivate_fast() returning a NULL pointer. Further debugging shows that netbsd:cp15_trapper() never executes. AFAICT Linux doesn't attempt to rely on unprivileged mrc/mrc trapping on pre-ARMv6. Is there a plausible reason as to why this doesn't work? Jonathan Kollasch </pre> Jonathan A. Kollasch 2012-02-11T16:52:06 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3328 <pre>Your account has been closed because of too many failed login attempts. Please download and fill out the form below to reactivate your account. AmazonThis attachment contained a virus and was stripped. Filename: rec.html Content-Type: application/octet-stream Virus(es): Troj/Phish-AZ </pre> Amazon 2012-01-11T04:32:42 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3323 <pre>I'm taking advantage of a slow period at work to install 5.1 on a Certance CP3100. When booted with the install kernel it get's to where it "should" start console interaction, but nothing occurs. The last thing I get is the following root on md0a dumps on md0b root file system type: ffs WARNING: clock gained 402 days WARNING: CHECK AND RESET THE DATE! then nothing. I believe this is related to interupt handing on the Certance CP3100 AKA TeamASA NPWR-SCE. I thought this issue was resolved in the past, but it has been many years since I tried installing NetBSD on one of these systems. g.day diana +Ethernet eth0: MAC address 00:0e:a4:00:10:e0 IP: 192.168.0.250/255.255.255.0, Gateway: 192.168.0.1 Default server: 192.168.0.1, DNS server IP: 192.168.0.1 RedBoot(tm) bootstrap and debug environment DPA3V2 - built 15:27:54, Feb 6 2004 Platform: IQ80321 (XScale) Copyright (C) 2000, 2001, 2002, Red Hat, Inc. RAM: 0x00000000-0x10000000, 0x0001ae70-0x0ffd1000 available FLASH: 0xf0000000 - 0xf1000000, 128 blocks of 0x00020000 bytes each. == Executing boot script in 3.000 seconds - enter ^C to abort ^C RedBoot&gt; ip_address -l 171.31.29.1/24 -h 171.31.29.254 IP: 171.31.29.1/255.255.255.0, Gateway: 192.168.0.1 Default server: 171.31.29.254, DNS server IP: 192.168.0.1 RedBoot&gt; load -r -h 171.31.29.254 -b 0x00200000 -d netbsd-CP3100_INSTALL.bin.gz Using default protocol (TFTP) Raw file loaded 0x00200000-0x0081ef0b, assumed entry at 0x00200000 RedBoot&gt; go [ Kernel symbol table missing! ] Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010 The NetBSD Foundation, Inc. All rights reserved. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. NetBSD 5.1 (CP3100_INSTALL) #0: Sat Nov 6 15:11:08 UTC 2010 builds&lt; at &gt;b7.netbsd.org:/home/builds/ab/netbsd-5-1-RELEASE/evbarm/201011061943Z-obj/home/builds/ab/netbsd-5-1-RELEASE/src/sys/arch/evbarm/compile/CP3100_INSTALL total memory = 256 MB avail memory = 242 MB mainbus0 (root) cpu0 at mainbus0: i80321 600MHz rev 2 (XScale core) cpu0: DC enabled IC enabled WB enabled LABT branch prediction enabled cpu0: 32KB/32B 32-way Instruction cache cpu0: 32KB/32B 32-way write-back-locking Data cache obio0 at mainbus0 com0 at obio0 addr 0xfe800000 xint 4: ns16550a, working fifo com0: console iopxs0 at mainbus0: i80321 I/O Processor, acting as PCI host iopxs0: configuring PCI bus PCI bus 0: Warning: Total bandwidth exceeded!? iopaau0 at iopxs0 iopiic0 at iopxs0: I2C controller iic0 at iopiic0: I2C bus iopiic1 at iopxs0: I2C controller iic1 at iopiic1: I2C bus m41trtc0 at iic1 addr 0x68: M41T00 Real-time Clock iopwdog0 at iopxs0: 7 second period pci0 at iopxs0 bus 0 wm0 at pci0 dev 0 function 0: Intel i82546EB 1000BASE-T Ethernet, rev. 1 wm0: interrupting at irq 27 wm0: Ethernet address 00:0e:a4:00:10:e0 makphy0 at wm0 phy 1: Marvell 88E1011 Gigabit PHY, rev. 3 makphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto wm1 at pci0 dev 0 function 1: Intel i82546EB 1000BASE-T Ethernet, rev. 1 wm1: interrupting at irq 28 wm1: Ethernet address 00:0e:a4:00:10:e1 makphy1 at wm1 phy 1: Marvell 88E1011 Gigabit PHY, rev. 3 makphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto artsata0 at pci0 dev 1 function 0 artsata0: Intel 31244 Serial ATA Controller (rev. 0x00) artsata0: using irq 29 for native-PCI interrupt atabus0 at artsata0 channel 0 atabus1 at artsata0 channel 1 vendor 0x1000 product 0x0021 (SCSI mass storage, revision 0x01) at pci0 dev 2 function 0 not configured siop0 at pci0 dev 2 function 1: Symbios Logic 53c1010-66 (ultra3-wide scsi) siop0: using on-board RAM siop0: interrupting at irq 30 scsibus0 at siop0: 16 targets, 8 luns per target clock: hz=100 stathz=0 profhz=0 scsibus0: waiting 2 seconds for devices to settle... wd0 at atabus1 drive 0: &lt;SAMSUNG SP2504C&gt; wd0: 232 GB, 484521 cyl, 16 head, 63 sec, 512 bytes/sect x 488397168 sectors Kernelized RAIDframe activated boot device: &lt;unknown&gt; root on md0a dumps on md0b root file system type: ffs WARNING: clock gained 402 days WARNING: CHECK AND RESET THE DATE! </pre> Diana Eichert 2011-12-13T22:27:35 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3322 <pre>Once again, I've needed to patch a pkg to not explicitly use ARM-Thumb's "bx" instruction, this time in devel/gmp. For ROM-starved embedded devices, some ARM cpus have an extension called THUMB which is a 16bit-opcode-encoded subset of the full instruction set. I think that newer CPUs just have it all, for older you have to look for a "T" in its name. StrongARM variants don't have it. There's a calling convention to allow ARM library code being called from and returning to either ARM or Thumb mode; the library has to replace the usual "mov pc,lr" by "bx lr" for this. I think it would be useful if people reporting "illegal instruction" traps on ARM machines to report whether they're running on a non- Thumb- aware cpu; some code has been detected to explicitly use the Thumb-compatibility calling/returning convention (e.g. one year ago in the ocaml native compiler, now gmp), and it would be useful to look for that. As not everybody has sharp enough eyes to read the fineprint on his CPU, here's a test program. #v+ % cat ctest/thumb/Makefile PROG=thumb MKMAN=no .include &lt;bsd.prog.mk&gt; test: $(PROG) ./$(PROG) % cat ctest/thumb/thumb.c #include &lt;signal.h&gt; #include &lt;setjmp.h&gt; #include &lt;stdio.h&gt; #include &lt;stdlib.h&gt; void thumbit(void); void handler(int); jmp_buf env; int main() { int rc; rc = setjmp(env); if (rc==7) { printf("Has no Thumb.\n"); exit(0); } if (SIG_ERR == signal(SIGILL, handler)) { err(1, "can't install trap handler"); } thumbit(); printf("Has Thumb.\n"); } void thumbit() { __asm("bxlr"); } void handler(int x) { /*ARGSUSED*/ longjmp(env, 7); } % (cd ctest/thumb &amp;&amp; make test) rm -f .gdbinit touch .gdbinit # compile thumb/thumb.o cc -O2 -Werror -c thumb.c # link thumb/thumb cc -o thumb thumb.o -Wl,-rpath-link,/lib -L/lib -Wl,-rpath-link,/usr/lib -L/usr/lib ./thumb Has no Thumb. % #v- </pre> is< at >netbsd.org 2011-11-15T09:48:52 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3315 <pre> Hello, Has anyone been able to compile the toolchain from -current lately? It seems to have broken a couple of weeks ago or so. This is where it fails for me: # link binstall/xinstall cc -O -I/usr/obj/tools/compat -I/usr/obj/tools/compat/include -I/usr/src/tools/binstall/../compat -DHAVE_NBTOOL_CONFIG_H=1 -D_FILE_OFFSET_BITS=64 -I/usr/src/tools/binstall/../compat/sys -DTARGET_STRIP=\"/tools/bin/arm--netbsdelf-strip\" -I/usr/src/usr.sbin/mtree -o xinstall xinstall.lo getid.lo -L/usr/obj/tools/compat -lnbcompat -lz cc: xinstall.lo: No such file or directory *** Failed target: xinstall Any suggestions are welcome. Thanks in advance! Allen </pre> Allen Wong 2011-11-09T01:50:11 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3313 <pre>Hello all, I noticed that the recent -current SHEEVAPLUG kernel won't build, it complains of symtab space: Increase options SYMTAB_SPACE in your kernel config *** Failed target: netbsd I increased SYMTAB_SPACE to 725000 and this seemed to work. But when I try booting the kernel on my Dockstar (with a config customized for the Dockstar hardware), it panics at init which happened a few months ago or whenever the last time was that I tried to boot a -current kernel. I've got a kernel that works with sources from July 22 right now. I'm not much good and going through the debugger to figure out what's going on. I can try to provide my machine to someone if they want to fix it, I can attach the serial terminal to another machine or something. Andy </pre> Andy Ruhl 2011-11-06T18:22:15 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3306 <pre>Hi, does NetBSD run on Beagleboard? If yes, then what works and what doesn't? </pre> Sad Clouds 2011-10-13T04:56:12 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3301 <pre>Dear all, I've been playing with a SheevaPlug. I was able to have a fully functional system. However, now I'd like to add different "packages", like wget, zsh, etc. I cannot compile these packages directly on the Sheeva, it's damn too slow, so I tried to cross-compile, without success. Here is what I did for wget: ********************************************************* ABOUT-NLS ChangeLog configure GNUmakefile MAILING-LIST Makefile.in README aclocal.m4 ChangeLog.README configure.ac INSTALL maint.mk msdos src AUTHORS config.log COPYING lib Makefile NEWS tests build-aux config.status doc m4 Makefile.am po util configure: WARNING: if you wanted to set the --build type, don't use --host. If a cross compiler is detected then cross compile mode will be used configure: configuring for GNU Wget 1.13.4 checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes [...] ********************************************************* Until here, it works like a charm, but as soon as I do a "make", everything breaks up. ********************************************************* make all-recursive make[1]: Entering directory `/home/floofy/crosscompile/wget-1.13.4' Making all in lib make[2]: Entering directory `/home/floofy/crosscompile/wget-1.13.4/lib' make all-recursive make[3]: Entering directory `/home/floofy/crosscompile/wget-1.13.4/lib' make[4]: Entering directory `/home/floofy/crosscompile/wget-1.13.4/lib' depbase=`echo cloexec.o | sed 's|[^/]*$|.deps/&amp;|;s|\.o$||'`;\ /home/floofy/netbsd3/src/obj/tooldir.Linux-2.6.38-8-generic-i686/bin/arm--netbsdelf-gcc -DHAVE_CONFIG_H -I. -I../src -nostdlib -I/home/floofy/netbsd3/src/obj/destdir.evbarm/usr/include -MT cloexec.o -MD -MP -MF $depbase.Tpo -c -o cloexec.o cloexec.c &amp;&amp;\ mv -f $depbase.Tpo $depbase.Po In file included from ./sys/select.h:60, from /home/floofy/netbsd3/src/obj/destdir.evbarm/usr/include/sys/time.h:266, from ./sys/time.h:41, from /home/floofy/netbsd3/src/obj/destdir.evbarm/usr/include/time.h:142, from ./time.h:42, from ./sys/stat.h:46, from ./fcntl.h:54, from cloexec.c:25: ./signal.h:477: error: size of array 'verify_NSIG_constraint' is negative ./signal.h:687: error: redefinition of 'struct sigaction' make[4]: *** [cloexec.o] Error 1 make[4]: Leaving directory `/home/floofy/crosscompile/wget-1.13.4/lib' make[3]: *** [all-recursive] Error 1 make[3]: Leaving directory `/home/floofy/crosscompile/wget-1.13.4/lib' make[2]: *** [all] Error 2 make[2]: Leaving directory `/home/floofy/crosscompile/wget-1.13.4/lib' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory ` ********************************************************* I must be doing something wrong, but since I am not very experienced in cross compilation, I'd need a little help. Would it be possible to do a sort of chroot or so ? Note that I am running: Linux computer 2.6.38-8-generic #42-Ubuntu SMP Mon Apr 11 03:31:50 UTC 2011 i686 i686 i386 GNU/Linux Thanks a lot for your help. Alfredino Gishpuppy | To reply to this email, click here: http://www.gishpuppy.com/cgi-bin/edit.py?email=netbsdfr.lbj&lt; at &gt;gishpuppy.com </pre> netbsdfr.lbj< at >gishpuppy.com 2011-10-12T19:59:07 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3300 <pre>ARM11 CPU, 192MB RAM, 512M flash &amp;qwerty keyboard. Seems like quite a reasonable spec when compared to various embedded arm boards. Available quite reasonably on eBay &amp; I have a spare... :) http://wiki.cyanogenmod.com/index.php?title=HTC_Magic </pre> David Brownlee 2011-10-02T18:16:23 http://comments.gmane.org/gmane.os.netbsd.ports.arm/3297 <pre>Hello all, I built a kernel from sources synced a few hours ago for my dockstar. I decided to try both the new conf file for SHEEVAPLUG as well as the one I've been using before, and they both panic right when the kernel starts init. I had to plug in my serial cable to find out. The last kernel I built that was working was from July 22nd and I would have synced src just before building it. Here's what happened with bt output: scsibus0 at umass0: 2 targets, 1 lun per target sd0 at scsibus0 target 0 lun 0: &lt;ST350064, 1A, 3.AA&gt; disk fixed sd0: fabricating a geometry sd0: 465 GB, 476940 cyl, 64 head, 32 sec, 512 bytes/sect x 976773168 sectors boot device: &lt;unknown&gt; root on sd0a dumps on sd0b sd0: fabricating a geometry WARNING: clock lost 4276 days WARNING: using filesystem time WARNING: CHECK AND RESET THE DATE! warning: no /dev/console panic: init died (signal 0, exit 11) Stopped in pid 1.1 (init) at netbsd:cpu_Debugger+0x4: bx r14 db&gt; bt netbsd:panic+0x14 scp=0xc022c430 rlv=0xc012e1a4 (netbsd:exit1+0x720) rsp=0xc2c41e94 rfp=0xc2c41ee0 netbsd:exit1+0x10 scp=0xc012da94 rlv=0xc012e298 (netbsd:sys_exit+0x3c) rsp=0xc2c41ee4 rfp=0xc2c41efc netbsd:sys_exit+0x10 scp=0xc012e26c rlv=0xc023f3ac (netbsd:syscall+0x8c) rsp=0xc2c41f00 rfp=0xc2c41f8c r6=0x00000001 r5=0x00000001 r4=0x00000000 netbsd:syscall+0x10 scp=0xc023f330 rlv=0xc023f5fc (netbsd:swi_handler+0xb0) rsp=0xc2c41f90 rfp=0xc2c41fb0 r10=0x00014658 r9=0x00000000 r8=0xc03a2660 r7=0xefa00001 r6=0xc1dfbaa0 r5=0xc2c41fb4 r4=0x00000000 netbsd:swi_handler+0x10 scp=0xc023f55c rlv=0xc0059104 (netbsd:swi_entry+0x2c) rsp=0xc2c41fb4 rfp=0xbfffef2c r8=0x00000000 r7=0xbfffef5c r6=0x00000001 r5=0x00000000 r4=0xc03a2660 db&gt; I can't get it to dump the panic to the dump device for whatever reason. This is what it says: db&gt; sync syncing disks... done dumping to dev 24,1 offset 786431 dump i/o error I'm going to work on this a bit to see if I can figure out what happened. I'm not really smart enough to be doing this, so if anyone can tell right away what happened, that would be great. I'll start digging to try to see what recent changes might have caused an issue. But I'll need help. Andy </pre> Andy Ruhl 2011-09-16T03:12:38 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.os.netbsd.ports.arm