Embedded Linux
Dont know if anyone has a good tool package in mind that I can use to develop an Embedded Linux. We want a linux that is just enough to run the flash_rom program+usb support and we hope it would be small enough so we can fit it into a Bios chip. Its all for the convenience of updating Bios in the future. As you image, we are closed to support Linuxbios officially in our products. gin
RE: Embedded Linux
If it's just about reading a file from usb and writing it to flash, you might want to have a look at filo. It should be easy to integrate the functionality of flash_rom into it, so you don't need a full blown linux system in flash. That will be a good idea if FILO can launch the flash_rom program but I thought FILO is excepting a bootable image. So I guess the effort will be in How to make FILO run an executable? Don't' know if I miss anything. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: It booted
wonderful news. Can you tell us more about your board? I can put it on the web page if you want. Yes, it's a server board with Intel E7501/ICH3-s. Details as below: http://www.nexcom.com/product/nex/nex7220/ Actually i am going to port it to another board later, which we want to use Linuxbios on. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
It booted
I just want to tell everyone that my board now booted SMP/non-SMP kernel ok. The problem is the setting of IOAPIC in ICH3. Thanks all for helping. gin
SMP linux
Ive been stuck at this problem for a while. My linuxbios boots a NON-SMP kernel ok but not the SMP one. The SMP kernel gets loaded and run. When it tries to access an IDE drive, an IDE command timed out before anything returns(kernel messages). My I/O APIC settings correspond to the MP table. All ISA devices are mapped to the I/O IOAPIC at 0xFEC0. But where do we set this I/O APIC up? How does it know to route the IRQs to the local APIC? Another question is, does linuxbios setup the local APIC as Virtual Wire mode that via I/O APIC? My board is similar to Tyan/s2735, does anyone get this booted with a SMP linux before?? many thinks, gin
RE: SMP linux
All ISA devices are mapped to the I/O IOAPIC at 0xFEC0. But where do we set this I/O APIC up? How does it know to route the IRQs to the local APIC? Linux looks at the mptable. I understand Linux looks at the mptable and looks for IOAPIC entries. I meant where in Linuxbios do we configure the IOAPIC at FEC0 or we don't have to do anything? Linuxbios DOES do something to configure the IOAPICs of P64H2. Thank you, gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: SMP linux
You are kidding. I put junk in the CVS server? Didn't mean to be rude. I just want to be sure. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: problem with booting SMP linux kernel
I've adjusted my MP table to use the correct IRQs. But the IDE interrupt is still lost. There is some message from linux kernel that I didn't understand. Is there something wrong with my timer settings? ..MP-BIOS bug: 8254 timer not connected to IO-APIC ...trying to set up timer (IRQ0) through the 8259A ... . (found pin 0) ... failed. ...trying to set up timer as Virtual Wire IRQ... works. === This is the last line where the kernel stops. = hda:4hda: dma_timer_expiry: dma status == 0x24 = Does anyone have an idea? gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: problem with booting SMP linux kernel
I've adjusted my MP table to use the correct IRQs. But the IDE interrupt is still lost. There is some message from linux kernel that I didn't understand. Is there something wrong with my timer settings? ..MP-BIOS bug: 8254 timer not connected to IO-APIC ...trying to set up timer (IRQ0) through the 8259A ... . (found pin 0) ... failed. ...trying to set up timer as Virtual Wire IRQ... works. === This is the last line where the kernel stops. = hda:4hda: dma_timer_expiry: dma status == 0x24 = Does anyone have an idea? gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: problem with booting SMP linux kernel
bad mp table would be my guess. I think the I/O APICs are incorrectly assigned to buses. From the kernel messages, devices on bus 0 are transformed to high IRQs. Can anyone one confirm that the mptable.c under src/tyan/s2735 is a good one to boot SMP linux kernel? I might need to manually construct the MP table for my board according to a good one. Thanks, gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
mptable.c
Is there an utility help to generate mptable.c? I thought I saw someone saying so. Is it usable? gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
irq_tables
Does linux care the PCI IRQ tables created by LinuxBios? I thought it only look at the MP-table. gin
problem with booting SMP linux kernel
The system failed when booting SMP linux kernel. It complains the IDE lost interrupt. Things were working fine with the non-SMP linux kernel. Has anyone seen it before? Note that the SMP kernel image was stored in the IDE hard disk. So BIOS was able to load the kernel to be executed. After Linux kernel took control, somehow the interrupts was messed up. Here are the last few messages over the console. === ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 hda: lost interrupt hda: lost interrupt hda: lost interrupt hda: host protected area = 1 hda: 160086528 sectors (81964 MB) w/7936KiB Cache, CHS=158816/16/63, UDMA(100) hda: lost interrupt hda: lost interrupt ide-floppy driver 0.99.newide Partition check: hda:4hda: dma_timer_expiry: dma status == 0x24 ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: Linux kernel in Rom
Here is my config file for a kernel version of linuxbios. opteron_phase1_p4_noapic is a kernel that I ran mkelfImage on. Ok. I will test on this config file. I am trying to make my linux kernel image smaller as I want to embed it with the ROM. Is there anyway you suggest to do this instead of using the menuconfig in RedHat linux. I can only take out options provided in the menu but the sometimes the dependency doesn't work that well. I get compile errors later when make bzImage. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
Linux kernel in Rom
What Options should I enable if I want to embed a linux kernel with Linuxbios in Version2? CONFIG_FS_STREAM? What should I use as the Linuxbios payload? What file format should it be? gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
fallback and normal image
The fallback image tries to load the normal image but if that fails it'll take over and continue with the fallback image. So I am guessing the 2 images are the same. Then why 2 exact the same image? In case of damaged flash area? I am asking this because I want to embed a linux kernel in the BIOS rom to run Flash_and_Burn util (For convenience when updating Bios). There probably won't be enough ROM space if I use both images. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
Etherboot and Linuxbios
Now my linuxbios found the ELF file created by Etherboot and tried to jump to it. But it hangs afterward. This is the debug output from linuxbios. It loads 2 seg from the ELF file. But in a objdump on the ELF file. There is only one seg and the loading addr is different. = New segment addr 0x2 size 0x114c8 offset 0xb0 filesize 0x8c4c Loading Segment: addr: 0x3ffd4000 memsz: 0x8000 filesz: 0x8000 Loading Segment: addr: 0x00028000 memsz: 0x94c8 filesz: 0x0c4c Clearing Segment: addr: 0x00028c4c memsz: 0x887c = Program Header: NOTE off0x0074 vaddr 0x0074 paddr 0x0074 align 2**0 filesz 0x003c memsz 0x003c flags rwx LOAD off0x00b0 vaddr 0x0002 paddr 0x0002 align 2**0 filesz 0x8c4c memsz 0x000114c8 flags rwx Sections: Idx Name Size VMA LMA File off Algn = ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
Etherboot and Linuxbios
Etherboot version used: 5.3.11 I am trying to build a Etherboot image to work with Linuxbios but got a compile error. There are couple undefined references to some functions due to the settings in the Config file. I commented out the PCBIOS option, and define the following. -DLINUXBIOS -DCONFIG_TSC_CURRTICKS -DCONSOLE_SERIAL -DCOMCONSOLE=0x3f8 -DCOMPRESERVE -DCONFIG_PCI_DIRECT -DELF_IMAGE -DPXE_IMAGE -DPXE_EXPORT p.s The undefined reference is due to the undefined PCBIOS. ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
flash rom utility
Is there a flash rom utility I can use to flash linuxbios to my EEPROM? Currently I am WICE to do it. But I cant depend on a hardware device to update bios. gin
RE: PXE support
etherboot supports PXE. If you use etherboot as LB payload it should work. Never tried this though since I can't really recognize an advantage over the usual dhcp/tftpboot scenario. It's unlikely that you are able to use any driver like code on the network adapters without 16bit bios. I saw the Config options for PXE boot in Etherboot. Has anyone successfully remote booted an image using PXE through Etherboot? If so, what is the target name I should give to the makefile. I am not sure how to build the file to use as the payload in Etherboot. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
Merry X'mas
Thanks for everyone's help this year. Wish you guys a wonderful holiday. gin ___ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: No booting
I can never get any output from the com port. I have console redirect running in grub, soI no that part is running and good test point. Your com port might not be turned on yet. Check your super IO. See if its the same with what your code is based on.
booting slow aftering mouting root
I compare 2 output of linux command pci from both Linuxbios booted linux and a regular bios booted linux. There are no IRQ assignments for lots of devices including IDE controller, SMBUS. Does linuxbios assign IRQs at all in the code? == Linuxbios == Bus 0, device 31, function 1: IDE interface: Intel Corp. 82801CA IDE U100 (rev 2). I/O at 0x3080 [0x308f]. Non-prefetchable 32 bit memory at 0xfe601000 [0xfe6013ff]. Bus 0, device 31, function 3: SMBus: Intel Corp. 82801CA/CAM SMBus (rev 2). I/O at 0x3060 [0x307f]. = regular bios Bus 0, device 31, function 1: IDE interface: Intel Corp. 82801CA IDE U100 (rev 2). IRQ 12. I/O at 0xf000 [0xf00f]. Non-prefetchable 32 bit memory at 0x3fff [0x3fff03ff]. Bus 0, device 31, function 3: SMBus: Intel Corp. 82801CA/CAM SMBus (rev 2). IRQ 5. I/O at 0x500 [0x51f]. ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
irq_tables.c
You can only use getpir if you have a working bios. Otherwise you Will get junk. Oh..i got it. I do have a working bios. Thank God. And I generated the irq_tables.c using getpir. It still doesn't help. One thing I did notice, it starts getting slow after mounting file system. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: booting slow
Seen this when interrupts in irq.c were wrong based on The motherboard. In our case we'd use etherboot to Load the kernel (which doesn't handle interrupts) then Once kernel was loaded we'd see the problem. This Was fixed by fixing the irq table. Thanks for the hint. I am able to load the kernel now. It just executes commands very slowly. One question for the irq_routing table(is that what you mean by irq.c?). The getpir util generates a routing table by looking at the memory config of my developing machine. How's it related to my target machine? Any document I can reference? This link mentioned on irq_tables.c http://www.microsoft.com/hwdev/busbios/PCIIRQ.HTM doesn't work. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: booting slow
It was easy to see when doing A cat /proc/interrupts and watching the offending interrupt configured Wrong. Just increment like mad. This is the result of a cat /proc/interrupts. The interrupt increments of timer and ide0 seem fine. I compared it with a normal bios boot. = CPU0 0:5331065 XT-PIC timer 1: 0 XT-PIC keyboard 2: 0 XT-PIC cascade 4: 0 XT-PIC serial 5: 0 XT-PIC usb-uhci 8: 1 XT-PIC rtc 10: 0 XT-PIC usb-uhci 11: 0 XT-PIC usb-uhci 14: 22880 XT-PIC ide0 NMI: 0 ERR: 0 ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: booting slow
that's the interesting part. What happens is that once init is running and things get a bit more formal, things start to slow down. There are lots of causes we've found for this, but interrupts is typically the big one. The serial interrupt? Shouldn't it be easy like assigning it an irq? What other causes that you've found out? One possibility I can think of is the cache. It was turned off. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
booting slow
I got the keyboard controller problem fixed by enabling Keyboard separately (Not through the PNP devices list). Since now the PNP devices don't get enabled at all. For now it's ok because we need only serial com port and keyboard. But I will love to have the code for SMSC if you have one available. Thanks. Now linux kernel boots fine except for one problem. It is extremely slow after this line Red Hat nash version 3.4.42 starting It displays each character slowly including what I typed from the keyboard(over serial console). It is very slow all the way to linux command prompt. I think it might have something to do with the routing table. I haven't changed much of mptable.c which is specific for another board. I couldn't think of anything else. gin = Red Hat nash version 3.4.42 starting Loading jbd.o module Loading ext3.o module Mounting /proc filesystem Creating block devices Creating root device Mounting root filesystem Freeing unused kernel memory: 132k freed INIT: version 2.84 booting Welcome to RedEntering runlevel: 3 : = : ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: booting slow
This is our proprietary mobo. In case you want a quick glance, http://www.nexcom.com/product/nex/nex7220/ North Bridge: E7501 South Bridge: ICH3 - S SuperIO: SMSC(LPC47B27x) I base on the code for tyan/s2735. If it was the serial interrupt, why didn't it happen earlier? It starts getting slow at certain point in the linux kernel. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
keyboard controller jammed
What is your southbridge (LPC) again ? I guess you LPC bridge does not have scan_static_bus is its device:scan_bus() method. I guess you are correct. So the PNP devices are there but didn't get looped through. But that's in the chip_op of the LPC bridge. It should be fixed in the code?? Ron, I did change the board config file for my target board. Here is my platform: North Bridge: E7501 South Bridge: ICH3 - S SuperIO: SMSC(LPC47B27x) The code I am based on right now is ICH5 and Winbond w83627hf. Some Dev ID of ICH5 and ICH3-s are different. LPC bridge is one of them. Don't know if it matters? Is there code for ICH3-s and SMSC superIO that I can use? Otherwise I will need to modify from the code I have now. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: keyboard controller jammed
I hope you did not modify static.c? I did. I did change the dev_root chip_op to 0 otherwise it won't execute the default root op. I don't know if the error is related to my superio. The superio on board is not winbond w83627hf, which is the code is based on. So the pnp devices are never enabled. In file /devices/root_device.c, scan_static_bus supposes to walk through the device tree and enable the PNP devices. I put a printk message. I don't think the PNP devices ever get initialized. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: keyboard controller jammed
I hope you did not modify static.c? I did. I did change the dev_root chip_op to 0 otherwise it won't execute the default root op. I don't know if the error is related to my superio. The superio on board is not winbond w83627hf, which is the code is based on. So the pnp devices are never enabled. In file /devices/root_device.c, scan_static_bus supposes to walk through the device tree and enable the PNP devices. I put a printk message. I don't think the PNP devices ever get initialized. gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Linux kernel hangs
Linuxbios writes to the beginning of the Memory for MP tables. Shouldnt those addresses be interrupt vector table? It doesnt look it contains any pointers to an interrupt handler. Who is the table for? Will Linux kernel reference it? Linuxbios successfully loads the payload FILO which brings up Linux kernel. But the kernel hangs while waiting for a timer interrupt. See below. I guess the interrupt table is not correct. == Linux version 2.4.20-8 ([EMAIL PROTECTED]) (gcc version 3.2.2 200 30222 (Red Hat Linux 3.2.2-5)) #1 Thu Mar 13 17:54:28 EST 2003 BIOS-provided physical RAM map: BIOS-e820: 0c88 - 000a (usable) BIOS-e820: 0010 - 4000 (usable) 128MB HIGHMEM available. 896MB LOWMEM available. hm, page reserved twice. On node 0 totalpages: 262144 zone(0): 4096 pages. zone(1): 225280 pages. zone(2): 32768 pages. Kernel command line: ro root=/dev/hda2 console=ttyS0,57600 Initializing CPU#0 Detected 1597.358 MHz processor. Console: colour dummy device 80x25 Calibrating delay loop... The MP table is specific to Tyan/s2735 not my board. I should change it. We use intel E7501/ICHs-3. But s2735 is ICH5. Dont know if it matters. gin
keyboard jammed
Forget about the previous email. Actually I solved the problem of Calibrating delay loop. The problem is that I used the device list in static.c as my onboard devices and it was missing apic cluster. In the board specific config file(under src/.), I noticed there is a tree-like device list. See below. Should I make it exactly the same as my board? Would it stop me from booting into linux kernel? Now I am getting another error. It prints out keyboard controller jammed. : :gin :=== # sample config for tyan/s2735 chip northbridge/intel/e7501 device pci_domain 0 on device pci 0.0 on end device pci 0.1 on end device pci 2.0 on chip southbridge/intel/i82870 device pci 1c.0 on end device pci 1d.0 on end device pci 1e.0 on end device pci 1f.0 on end end end device pci 6.0 on end chip southbridge/intel/i82801er device pci 1d.0 on end device pci 1d.1 on end device pci 1d.2 on end device pci 1d.3 on end device pci 1d.7 on end device pci 1e.0 on end device pci 1f.0 on # device pci 8.0 end chip superio/winbond/w83627hf device pnp 2e.0 on # Floppy io 0x60 = 0x3f0 irq 0x70 = 6 drq 0x74 = 2 end device pnp 2e.1 off # Parallel Port io 0x60 = 0x378 irq 0x70 = 7 end device pnp 2e.2 on # Com1 : : : === Linux version 2.4.20-8 ([EMAIL PROTECTED]) (gcc version 3.2.2 200 30222 (Red Hat Linux 3.2.2-5)) #1 Thu Mar 13 17:54:28 EST 2003 BIOS-provided physical RAM map: BIOS-e820: 0c88 - 000a (usable) BIOS-e820: 0010 - 4000 (usable) 128MB HIGHMEM available. 896MB LOWMEM available. hm, page reserved twice. On node 0 totalpages: 262144 zone(0): 4096 pages. zone(1): 225280 pages. zone(2): 32768 pages. Kernel command line: ro root=/dev/hda2 console=ttyS0,57600 Initializing CPU#0 Detected 1597.322 MHz processor. Console: colour dummy device 80x25 Calibrating delay loop... 3185.04 BogoMIPS Memory: 1027480k/1048576k available (1347k kernel code, 17508k reserved, 999k da ta, 132k init, 131072k highmem) Dentry cache hash table entries: 131072 (order: 8, 1048576 bytes) Inode cache hash table entries: 65536 (order: 7, 524288 bytes) Mount cache hash table entries: 512 (order: 0, 4096 bytes) Buffer-cache hash table entries: 65536 (order: 6, 262144 bytes) Page-cache hash table entries: 262144 (order: 8, 1048576 bytes) CPU: Trace cache: 12K uops, L1 D cache: 8K CPU: L2 cache: 512K Intel machine check architecture supported. Intel machine check reporting enabled on CPU#0. CPU: Intel(R) Xeon(TM) CPU 1.60GHz stepping 07 Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Checking 'hlt' instruction... OK. POSIX conformance testing by UNIFIX mtrr: v1.40 (20010327) Richard Gooch ([EMAIL PROTECTED]) mtrr: detected mtrr type: Intel PCI: Using configuration type 1 PCI: Probing PCI hardware PCI: Ignoring BAR0-3 of IDE controller 00:1f.1 PCI: Unable to handle 64-bit address space for PCI: Unable to handle 64-bit address space for Transparent bridge - Intel Corp. 82801BA/CA/DB PCI Bridge PCI: Using IRQ router PIIX [8086/2480] at 00:1f.0 isapnp: Scanning for PnP cards... isapnp: No Plug Play device found Linux NET4.0 for Linux 2.4 Based upon Swansea University Computer Society NET3.039 Initializing RT netlink socket apm: BIOS not found. Starting kswapd allocated 32 pages and 32 bhs reserved for the highmem bounces VFS: Disk quotas vdquot_6.5.1 pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF). pc_keyb: controller jammed (0xFF).
RE: jump to boot code
Yes, when I tested the filo with Grub. I could see the messages over the serial console. gin -Original Message- From: Ronald G. Minnich [mailto:[EMAIL PROTECTED] Sent: Thursday, December 02, 2004 11:29 PM To: Gin Cc: 'LinuxBIOS' Subject: Re: jump to boot code the problem is that your error messages are kind of ok. Did you build a filo with serial port only console? ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: jump to boot code
This is the header of my payload Filo.elf. it does have the starting address. = start address 0x001092e4 Program Header: LOAD off0x00a0 vaddr 0x0010 paddr 0x0010 align 2**5 filesz 0xd048 memsz 0x000270d0 flags rwx LOAD off0xd100 vaddr 0x001270e0 paddr 0x001270e0 align 2**5 filesz 0x0048 memsz 0x0048 flags rw- NOTE off0x00c0 vaddr 0x00100020 paddr 0x00100020 align 2**5 filesz 0x0088 memsz 0x0088 flags r-- == -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Greg Watson Sent: Thursday, December 02, 2004 10:57 PM To: Gin Cc: 'LinuxBIOS' Subject: Re: jump to boot code If you do an 'objdump -f' on the executable, what does it say is the start address? 0x1092e4 seems a bit strange. Greg On Dec 1, 2004, at 11:17 PM, Gin wrote: I tested my payload(FILO.elf) with linux loader Grub. It has no problem at all. So something must go wrong when linuxbios jumps to the payload. FILO doesn't seem to run at all. No debug message over the console. Dont know if there is anyone familiar with ELF. This is the message over the console at the end. It seems that there are 2 segments in the ELF. In the end, linuxbios jumps to an entry that looks like the in the middle of the first segment. Does it look right? Found ELF candiate at offset 0 (cleaned up) New segment addr 0x10 size 0x270f0 offset 0xa0 filesize 0xd068 (cleaned up) New segment addr 0x127100 size 0x48 offset 0xd120 filesize 0x48 Dropping non PT_LOAD segment Loading Segment: addr: 0x0010 memsz: 0x000270f0 filesz: 0x00 00d068 Loading Segment: addr: 0x00127100 memsz: 0x0048 filesz: 0x00 48 Jumping to boot code at 0x1092e4 entry = 0x001092e4 lb_start = 0x4000 lb_size = 0x00024000 adjust = 0xfe5d8400 buffer = 0xfe5b8400 elf_boot_notes = 0x00015680 adjusted_boot_notes = 0xfe5eda80 ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
jump to boot code
This is the message I got in the end. Linuxbios tries to load the ELF image. It seems that there are 2 segments in the ELF. In the end, linuxbios jumps to an entry that looks like the in the middle of the first segment. Does it look right? The FILO.elf doesnt seem to load succeffully. Gin Found ELF candiate at offset 0 (cleaned up) New segment addr 0x10 size 0x270f0 offset 0xa0 filesize 0xd068 (cleaned up) New segment addr 0x127100 size 0x48 offset 0xd120 filesize 0x48 Dropping non PT_LOAD segment Loading Segment: addr: 0x0010 memsz: 0x000270f0 filesz: 0x00 00d068 Loading Segment: addr: 0x00127100 memsz: 0x0048 filesz: 0x00 48 Jumping to boot code at 0x1092e4 entry = 0x001092e4 lb_start = 0x4000 lb_size = 0x00024000 adjust = 0xfe5d8400 buffer = 0xfe5b8400 elf_boot_notes = 0x00015680 adjusted_boot_notes = 0xfe5eda80
irq routing table
If I have a wrong checksum for my irq routing table, would that stops me from booting successfully? The question should be, does the irq_routing table matter at all? Linuxbios runs through till it jumps to the entry of payload. I am just trying to figure out why. Gin
jump to boot code
I tested my payload(FILO.elf) with linux loader Grub. It has no problem at all. So something must go wrong when linuxbios jumps to the payload. FILO doesn't seem to run at all. No debug message over the console. Dont know if there is anyone familiar with ELF. This is the message over the console at the end. It seems that there are 2 segments in the ELF. In the end, linuxbios jumps to an entry that looks like the in the middle of the first segment. Does it look right? Found ELF candiate at offset 0 (cleaned up) New segment addr 0x10 size 0x270f0 offset 0xa0 filesize 0xd068 (cleaned up) New segment addr 0x127100 size 0x48 offset 0xd120 filesize 0x48 Dropping non PT_LOAD segment Loading Segment: addr: 0x0010 memsz: 0x000270f0 filesz: 0x00 00d068 Loading Segment: addr: 0x00127100 memsz: 0x0048 filesz: 0x00 48 Jumping to boot code at 0x1092e4 entry = 0x001092e4 lb_start = 0x4000 lb_size = 0x00024000 adjust = 0xfe5d8400 buffer = 0xfe5b8400 elf_boot_notes = 0x00015680 adjusted_boot_notes = 0xfe5eda80
elfboot
I know its probably
been a while since you look into the elfboot code.
In the elfboot.c
When it tries to build the
ELF segment list, it checks if the segment address is valid by walking through the
table of valid memory ranges. What is the reason to do this? Below is the code, the if statement seems only guarantee that the new
addresses INTERSECT with the valid memory range, not fully contained.
for(i = 0; i mem_entries; i++) {
uint64_t mstart, mend;
uint32_t mtype;
mtype = mem-map[i].type;
mstart = mem-map[i].start;
mend = mstart + mem-map[i].size;
if ((mtype == LB_MEM_RAM)
(start mend) (end mstart))
{
break;
}
}
jmp_to_elf_entry
I've got it to run to the point that it jumps to the ELF entry. My payload is a FILO. What should I see after it jumps to the ELF entry? I guess the FILO program will start to run, correct? I have not got my VGA up, so I can't see if it really boots the ELF entry successfully. Maybe that's what I should do. Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
compile vga rom
I have an onboard VGA chip that I want to use. Is there a way to compile the onboard VGA rom with the linuxbios? Or would it be easier to use a VGA card? Gin
boot linux from an ide
I use FILO as the payload to boot from a Hard Disk, which has a installed linux on it. I noticed if I enabled the ide stream. It tries to read the ELF header from my IDE drive. But I thought the FILO payload should be the one to load the image from the IDE drive. When does it come in and play the role? Anyone can explain the flow of booting if I try to load the INSTALLED linux from a IDE drive. Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
boot from an ide
I use FILO as the payload to boot from a Hard Disk, which has a installed linux on it. I noticed if I enabled the ide stream. It tries to read the ELF header from my IDE drive. But I thought the FILO payload should be the one to load the image from the IDE drive. When does it come in and play the role? Anyone can explain the flow of booting if I try to load the INSTALLED linux from a IDE drive. Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
could not find a bounce buffer
Yes, I think I need to set CONFIG_ROM_STREAM=1 in order to have it boot from my FILO payload which then will go load the image in a IDE drive. Thanks. I got an error when it ties to get_bounce_buffer in elfboot, Don't know if you have the same problem. I dumped out the mem entries info. Looks like the 2 mem entries are all LB_MEM_TABLE(ram config tables to be kept in). So there is no LB_MEM_RAM, (where everyone could use) to allocate the bounce buffer from. Any idea? === Wrote the mp table end at: 0020 - 01f4 Wrote linuxbios table at: 0500 - 0be4 checksum 18e1 lb_size=0x2 mem_entries=2 mem entry1: type=0x10, start=0, size=0xc4c mem entry2: type=0x10, start=0xf, size=0x400 Could not find a bounce buffer... Cannot Load ELF Image -=== -Original Message- From: Greg Watson [mailto:[EMAIL PROTECTED] Sent: Thursday, November 25, 2004 11:43 PM To: Gin Cc: [EMAIL PROTECTED] Subject: Re: boot linux from an ide If FILO is a payload in flash, then you need to set CONFIG_ROM_STREAM=1 and CONFIG_ROM_STREAM_START to the address you want to start looking for the payload ELF header (usually just after the LinuxBIOS image). CONFIG_IDE_STREAM is used to load a payload (such as FILO) from an attached IDE device. This is useful if the flash is not large enough to hold LinuxBIOS+payload, but is probably not what you are looking for. Sorry for the confusion. Greg On Nov 25, 2004, at 5:40 AM, Gin wrote: I use FILO as the payload to boot from a Hard Disk, which has a installed linux on it. I noticed if I enabled the ide stream. It tries to read the ELF header from my IDE drive. But I thought the FILO payload should be the one to load the image from the IDE drive. When does it come in and play the role? Anyone can explain the flow of booting if I try to load the INSTALLED linux from a IDE drive. Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: boot from ide
Thanks for the file. But would it work if I specify the ROM_IMAGE_SIZE larger than 0x1? -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ronald G. Minnich Sent: Thursday, November 18, 2004 11:11 PM To: Gin Cc: 'LinuxBIOS' Subject: Re: boot from ide Here is my filo-using config # Sample config file for the Arima HDAMA target hdama.filo mainboard arima/hdama option DEFAULT_CONSOLE_LOGLEVEL=5 option MAXIMUM_CONSOLE_LOGLEVEL=5 romimage normal option USE_FALLBACK_IMAGE=0 option ROM_IMAGE_SIZE=0x1 option LINUXBIOS_EXTRA_VERSION=.0Normal payload /opt/filo/normal/filo.elf end romimage fallback option USE_FALLBACK_IMAGE=1 option ROM_IMAGE_SIZE=0x1 option LINUXBIOS_EXTRA_VERSION=.0Fallback payload /opt/filo/fallback/filo.elf end buildrom ./linuxbios.rom ROM_SIZE normal fallback ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
ROM_IMAGE_SIZE
Not sure why this is happening. When I specify the ROM_IMAGE_SIZE to be larger than 64k. The file linuxbios.strip is larger than linuxbios, which is the file before the objcopy. This doesn't happen to the rom_image_size that is 64k. Here is the file size info. [EMAIL PROTECTED] fallback]# ls -l linuxbios* -rwxr-xr-x1 root root 126425 Nov 19 11:28 linuxbios -rw-r--r--1 root root 155892 Nov 19 11:28 linuxbios.a -rw-r--r--1 root root39445 Nov 19 11:28 linuxbios.map -rwxr-xr-x1 root root 149179 Nov 19 11:28 linuxbios_ram -rwxr-xr-x1 root root77008 Nov 19 11:28 linuxbios_ram.bin -rw-r--r--1 root root13331 Nov 19 11:28 linuxbios_ram.map -rw-r--r--1 root root36353 Nov 19 11:28 linuxbios_ram.nrv2b -rw-r--r--1 root root 114373 Nov 19 11:28 linuxbios_ram.o -rw-r--r--1 root root36353 Nov 19 11:28 linuxbios_ram.rom -rw-r--r--1 root root 262144 Nov 19 09:58 linuxbios.rom -rwxr-xr-x1 root root 131087 Nov 19 11:28 linuxbios.strip -rw-r--r--1 root root 4872 Nov 17 19:19 linuxbios_table.o Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
boot from ide
Hello, I've got my linuxbios run to the point it tries to load the image from an IDE drive. I use FILO as the payload. 1. What do I have to add in the Config.lb file to make it use filo? I know CONFIG_FS_STREAM should be set to true. What else? 2. I remembered linuxbios supports ROM image size larger than 64K now. But when got compiler errors when specifies it as 128K. Do I miss anything? Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
Re: boot from ide
Thanks, I got an error while specifying the fall back rom image size of larger than 64K. See below: 1. When ROM_IMAGE_SIZE = 0x2000 ./buildrom linuxbios.strip linuxbios.rom /opt/filo/fallback/filo.elf 0x2 0x4 linuxbios image is 131087 bytes; only 131072 allowed Linuxbios input file larger than allowed size! : Success make: *** [linuxbios.rom] Error 2 2.When ROM_IMAGE_SIZE = 0x3000 ./buildrom linuxbios.strip linuxbios.rom /opt/filo/fallback/filo.elf 0x3 0x4 linuxbios image is 196623 bytes; only 196608 allowed Linuxbios input file larger than allowed size! : Success make: *** [linuxbios.rom] Error 2 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Greg Watson Sent: Thursday, November 18, 2004 10:21 AM To: Gin Cc: 'LinuxBIOS' Subject: Re: boot from ide Actually, don't set CONFIG_FS_STREAM if you're using Filo as a payload. Use the CONFIG_IDE_STREAM instead. Greg On Nov 17, 2004, at 7:05 PM, Gin wrote: Hello, I've got my linuxbios run to the point it tries to load the image from an IDE drive. I use FILO as the payload. 1. What do I have to add in the Config.lb file to make it use filo? I know CONFIG_FS_STREAM should be set to true. What else? 2. I remembered linuxbios supports ROM image size larger than 64K now. But when got compiler errors when specifies it as 128K. Do I miss anything? Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: Tyan/s2735
Thanks Steven, Do you think you will have time testing this on V2 tree? I've been working on that code base and keep having problems with it. Gin -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steven James Sent: Monday, November 15, 2004 4:31 AM To: Gin Cc: [EMAIL PROTECTED] Subject: Re: Tyan/s2735 Greetings, I have successfully used the v1 code on that board. Things have been in flux since then, and I haven't had time to test a recent CVS checkout or the v2 tree. G'day, sjames On Thu, 11 Nov 2004, Gin wrote: Steven, Thanks for the info. I've got the ram up by filling the some values in North Bridge(I reference a regular bios). I have an essential question about the code base for Tyan/s2735. Has it ever successfully built and run on that board? I have to make sure the code I am currently building our board on is good. Thanks -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steven James Sent: Friday, November 05, 2004 10:40 PM To: Gin Lin Cc: [EMAIL PROTECTED] Subject: Re: ram init Greetings, It's been a while since I've looked at it, but when I last looked, it did sometimes need power cycled to restart. I worked around that by making the kernel reboot by writing 0x0e to port 0xcf9. That causes the southbridge to do a 3 second power off. G'day, sjames On Fri, 5 Nov 2004, Gin Lin wrote: Does anyone know if the raminit.c under intel/E7501 is stable enough? I am having a problem with the system reseting when it jumps to the first address in the ram. Could the procedures of writing NorthBrige registers have problems? thanks, Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios | | ||| by Linux Labs International, Inc. Steven James, CTO 55 Marietta Street Suite 1830 Atlanta, Ga 30303 866 824 9737 support ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios | | ||| by Linux Labs International, Inc. Steven James, CTO 55 Marietta Street Suite 1830 Atlanta, Ga 30303 866 824 9737 support ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
ram_check
I have a question with the ram. Dont know if anyone has heard the same problem before. I ran the ram_check procedure and it reports that there are always 4 bytes out of every 64 bytes that reads zero. I think thats why I can run the code in the ram but it gets reset in the in different places. Gin
Tyan/s2735
Steven, Thanks for the info. I've got the ram up by filling the some values in North Bridge(I reference a regular bios). I have an essential question about the code base for Tyan/s2735. Has it ever successfully built and run on that board? I have to make sure the code I am currently building our board on is good. Thanks -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steven James Sent: Friday, November 05, 2004 10:40 PM To: Gin Lin Cc: [EMAIL PROTECTED] Subject: Re: ram init Greetings, It's been a while since I've looked at it, but when I last looked, it did sometimes need power cycled to restart. I worked around that by making the kernel reboot by writing 0x0e to port 0xcf9. That causes the southbridge to do a 3 second power off. G'day, sjames On Fri, 5 Nov 2004, Gin Lin wrote: Does anyone know if the raminit.c under intel/E7501 is stable enough? I am having a problem with the system reseting when it jumps to the first address in the ram. Could the procedures of writing NorthBrige registers have problems? thanks, Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios | | ||| by Linux Labs International, Inc. Steven James, CTO 55 Marietta Street Suite 1830 Atlanta, Ga 30303 866 824 9737 support ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
ram init
Does anyone know if the raminit.c under intel/E7501 is stable enough? I am having a problem with the system reseting when it jumps to the first address in the ram. Could the procedures of writing NorthBrige registers have problems? thanks, Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
PXE support
Hello, It seems like linuxbios supports PXE. Can anyone tell me how it works if you have experience with it? Thanks Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
fallback and normal image
I could build an image now. Thanks all. I am curious about how fallback image and normal image are used. It looks like in the normal image, it always jumps to the Protected_start. I don't think it should be the start when the system boots up since it will be in real-mode. So which image should I use? Regards, Gin - ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
RE: linux kernel patch
Ron, Thanks for the email. I downloaded the freebios2 version. It looks much cleaner. Good work. Currently we are looking for a proper mainboard among our products to run linuxbios on. I don't know the specific chipset but we tend to use Intel solution on our products. Is there any suggestion for us to pick a chipset/cpu model? We hope to build a more stable machine at the first time. I happened to have a old board(440GX). As I can't wait for they to pick a mainboard, I decide to try linuxbios on this one first. Can I build the linuxbios image from the new source- freebios2? Would that work as the included directories might have been changed in the new source? For the new source(freebios2) 1. Do we need a Config file for each mainboard to run? Does it work(the building process)like the old way or it has been changed? 2. Is any document of how to build linuxbios on the new source? Regards, Gin -Original Message- From: Ronald G. Minnich [mailto:[EMAIL PROTECTED] Sent: Thursday, October 14, 2004 11:32 PM To: Gin Cc: [EMAIL PROTECTED] Subject: Re: linux kernel patch we don't patch kernels any more. Just ignore anything about kernel patches. What's the hardware in question ?What chipset? We're happy to help. ron ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
linux kernel patch
Hello Ron, I dont know if you are still working on this project since its 2004 now. I just finished a presentation of Linuxbios at work. We are highly interested in getting linuxbios running on our systems(we design motherboards). I have a question about the way you do the kernel patch. If the patch is to initialize the uninitialized hardware in order to run linux. Then why didnt we add it into the C code. Let it become part of linuxbios rom image? Furthermore, why do you have different patches for different linux kernel versions? Sorry if the questions are too basic. We just started trying to get it built. Thanks, Gin
RE: linux kernel patch
Hello David, Thanks for your quick response. Do you know where I can get the latest linuxbios source code(freebios2)? Call me crazy, but I couldn't seem able to find it through SourceForge. http://sourceforge.net/ thanks, Gin -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of David Hendricks Sent: Thursday, October 14, 2004 4:08 PM To: [EMAIL PROTECTED] Subject: Re: linux kernel patch We appreciate your interest in the project! As I recall, the patches were just something for some SiS chipset support in older LinuxBIOS versions (freebios1). Since then, LinuxBIOS has advanced to include more generic chipset support that does not require kernel patches. The current LinuxBIOS (freebios2) will work without patches to your kernel. On Thu, 14 Oct 2004 15:34:13 +0800 Gin [EMAIL PROTECTED] wrote: Hello Ron, I don't' know if you are still working on this project since it's 2004 now. I just finished a presentation of Linuxbios at work. We are highly interested in getting linuxbios running on our systems(we design motherboards). I have a question about the way you do the kernel patch. If the patch is to initialize the uninitialized hardware in order to run linux. Then why didn't we add it into the C code. Let it become part of linuxbios rom image? Furthermore, why do you have different patches for different linux kernel versions? Sorry if the questions are too basic. We just started trying to get it built. Thanks, Gin ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios ___ Linuxbios mailing list [EMAIL PROTECTED] http://www.clustermatic.org/mailman/listinfo/linuxbios
