June 2003 - coreboot - mail.coreboot.org

Re: doc ipl image
by roger June 11, 2003

June 11, 2003

ok. docipl /path/to/src/mainboard/asus/cua/ipl.S and Python configure made the Makefiles, however dd fails with a negative number with "docipl" in the config file. So i'm doing this seperately and copying the created dociple to another folder and then re-running the configure/make without the "docipl" option to build the linuxbios.image and linuxkernel.image images. I'm rebuilding the kernel again for the DoC (looks like it might be wise to build mtd/mtdchar/doc2001 modules into the kernel) and am working on a serial console for remote console. So far, i've gotten nothing to boot. Just the same scenario as a box without a bios chip. Although, I could have sworn i had some additional hdd activity on the last attempt. One additional question, if HAVE_FRAMEBUFFER=0, there still is console activity on the monitor right? (-i think i'm interpretting this correctly-). wew! ain't google grand?!? one can actually have a one-way conversion with himself via email!!! On Tue, 2003-06-10 at 22:24, roger wrote: > Just a quick follow-up. I noticed that I had to add the following line > in the config file: > > docipl /pathto/ipl.S > > ok. now for ipl.S (for a 440BX chipset?) -- Roger http://www.eskimo.com/~roger/index.html

4 4

doc ipl image
by roger June 11, 2003

June 11, 2003

ok. wew. searched, read, reboots (did this in a loop for 10x's). i've gotten p5 images built and have been able to recognize and write to the DoC (many thanx to a kernel oops w/ devbios project and it's apperant incompatability with apm). After erasing the DoC, I'm realizing that I need an ipl image burned prior to burning linuxbios.image & linuxkernel.image images to the rom. Ipl is needed to boot linuxbios. Linuxbios is needed to boot linuxkernel. (And so on...) I see that the ipl.asm is provided only by m-sys on a NDA. (btw, using an Intel 440bx chipset here). -- Roger http://www.eskimo.com/~roger/index.html

1 1

romcc progress.
by ebiederman＠lnxi.com June 10, 2003

June 10, 2003

I have just implemented inline assembly and a better allocator in romcc. It will now use sse and mmx registers when you give it the appropriate -mcpu= option. I am not happy with the command line options but fixing them is easy. In addition the quality of the register allocations has improved to the point that it looks quite a bit like it was done by hand. There are limits but there are fewer stupid things happening. The inline assembly is syntactically the same as gcc's. But it is not 100% compatible with gcc. The goal is to be as compatible with gcc as gcc is compatible with gcc between ports. In practice this means that romcc supports a subset of what gcc does on x86, with respect to constraints. On inline assembly versus builtin operations. A compiler has a lot more potential to optimize code in a builtin so I will continue to implement builtins. Support wise inline assembly with parameters is a pain to initially implement but there really is not a support burden after that. And there are some operations where there is not gain making them a builtin. The code is now checked into the freebios2 tree. And I can get back to exercising romcc by writing code with it. Eric

1 0

答复: 答复: 答复: new config language.
by YhLu June 10, 2003

June 10, 2003

I changeg the stack size from 0x2000 to 0x10000, and the result is that it retrys more times. I have got 1.5G memory installed and I found the code in ramtest only hard code that to 512M, is cause the problem? LinuxBIOS-1.1.02.0Fallback Mon Jun 9 23:14:16 PDT 2003 starting... Bootstrap cpu setting up coherent ht domain.... done. PCI: 00:01.00 PCI: 00:01.01 PCI: 00:02.00 PCI: 00:02.01 PCI: 00:03.00 PCI: 00:04.00 PCI: 00:04.01 PCI: 00:04.02 PCI: 00:04.03 PCI: 00:04.05 PCI: 00:04.06 PCI: 00:18.00 PCI: 00:18.01 PCI: 00:18.02 PCI: 00:18.03 SMBus controller enabled Ram1 setting up CPU0 northbridge registers done. Ram2 Ram3 dcl: 080c8000 Initializing memory: done Ram4 Ram5 Ram6 dimm: 50 00: 80 08 07 0d 0a 01 48 00 04 60 70 02 82 08 08 01 10: 0e 04 0c 01 02 26 c0 75 70 00 00 48 30 48 2a 40 20: 80 80 45 45 00 00 00 00 00 3c 48 30 2d 55 64 00 30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 10 a2 40: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 50: 00 00 00 00 00 00 00 00 00 00 00 00 00 03 11 00 60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff 90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff dimm: 5080 dimm: 5180 dimm: 5280 dimm: 5380 DRAM fill: 00000000-00001000 00001000 DRAM filled DRAM verify: 00000000-00001000 00001000 DRAM verified Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.1.02.0Fallback Mon Jun 9 23:14:16 PDT 2003 starting... LinuxBIOS-1.1.02.0Fallback Mon Jun 9 23:14:16 PDT 2003 starting... Bootstrap cpu CPU INIT Detected. Copying LinuxBIOS to ram. Jumping to LinuxBIOS. LinuxBIOS-1.1.02.0Fallback Mon Jun 9 23:14:16 PDT 2003 starting... LinuxBIOS-1.1.02.0Fallback Mon Jun 9 23:14:16 PDT 2003 starting... Bootstrap cpu CPU INIT Detected. Copying LinuxBIOS to ram. Jumping to LinuxBIOS. So I would find one AMD to verify the code. -----邮件原件----- 发件人: ebiederman(a)lnxi.com [mailto:ebiederman@lnxi.com] 发送时间: 2003年6月9日 22:36 收件人: YhLu 抄送: ron minnich; linuxbios(a)clustermatic.org 主题: Re: 答复: 答复: new config language. YhLu <YhLu(a)tyan.com> writes: > Ron, > > Is there any MB can boot to execute into LinuxBIOS after copying it to RAM? > > The _RAMBASE WE use for E7501 is 0x8000, and the FreeBIOS2 default for K8 is > 0x4000. How should I set the value to make the LinuxBIOS can boot into > hardwaremain.c. > It seems MB only execute the auto.c and crt0.base and can not jump to the > right address. ??????????????? I think there is a fundamental disconnect here. Except when I miscompile auto.c the AMD reference board boots fine. And in this aspect the code is fundamentally the same as earlier revision of LinuxBIOS. auto.c just gets compiles into auto.inc And I can't think of a board where I have had problems with jumping to hardwaremain. _RAMBASE should not strongly affect anything, it just need to be in a location where the ram initialization code always puts ram. Only with respect to stack size, and the LinuxBIOS tables should it have any issues with various locations in ram. Eric _______________________________________________ Linuxbios mailing list Linuxbios(a)clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios

1 0

答复: 答复: new config language.
by YhLu June 10, 2003

June 10, 2003

Ron, Is there any MB can boot to execute into LinuxBIOS after copying it to RAM? The _RAMBASE WE use for E7501 is 0x8000, and the FreeBIOS2 default for K8 is 0x4000. How should I set the value to make the LinuxBIOS can boot into hardwaremain.c. It seems MB only execute the auto.c and crt0.base and can not jump to the right address. Regards Yinghai Lu -----邮件原件----- 发件人: ron minnich [mailto:rminnich@lanl.gov] 发送时间: 2003年6月9日 22:01 收件人: YhLu 抄送: linuxbios(a)clustermatic.org 主题: Re: 答复: new config language. On Mon, 9 Jun 2003, YhLu wrote: > How about the status of freebios2 support on K8? we're working on it. There are 3 big efforts going on. First, Eric is working on romcc for K8 so that it works well. Suse is trying to get it to boot. I am working on a new config language which I hope is rich enough to capture the needs of freebios2. Where are we? We are all collectively close. But we're not done. ron

3 2

答复: new config language.
by YhLu June 10, 2003

June 10, 2003

Ron, How about the status of freebios2 support on K8? Today I use to the code and try on Tyan s2880 found two problem 1. I use RH as host compiler system. payload make statement in make.base with perl doesn't work, it produce strange size images near One mega byte. 2. I change that back to dd, and make one 512K romimage, But the code only can execute to "jumping to LinuxBIOS". And then retry several times. I use debug card and see the post code is 01--> 08--> 11 ->01...>11 -> 12. end. Do you know the progress of Suse LinuxBIOS support on opteron? Regards Yinghai Lu -----邮件原件----- 发件人: ron minnich [mailto:rminnich@lanl.gov] 发送时间: 2003年6月8日 20:41 收件人: linuxbios(a)clustermatic.org 主题: new config language. I'm opening up this discussion, which was formerly just between the folks actively doing freebios2. We're getting closer to having things working and I am happy to hear any comments. linuxbios.a now builds for the K8 with the new config tool. Not all of linuxbios, just linuxbios.a, but that's quite a way to working. Things are stricter now. We've been able to operate before with code like this: #if SOME_VARIABLE_THAT_WAS_UNDEFINED == 0 stuff #endif because in the GNU C preprocessor, #if (a == 0) and #ifndef a are equivalent if a is undefined. I think that is a very poor way to operate: it can fool people, and in a BIOS you want more checking. Now that you can use an 'if' operator in the config language, setting up some things is a lot easier: For example, this: expr ROM_SECTION_SIZE =(USE_FALLBACK_IMAGE*65536)+(USE_NORMAL_IMAGE*(ROM_SIZE - 65536)) expr ROM_SECTION_OFFSET=(USE_FALLBACK_IMAGE*(ROM_SIZE-65536))+(USE_NORMAL_IMAGE *0) now becomes this: default FALLBACK_SIZE=65536 if USE_FALLBACK_IMAGE option ROM_SECTION_SIZE = FALLBACK_SIZE option ROM_SECTION_OFFSET= (ROM_SIZE - FALLBACK_SIZE) end if USE_NORMAL_IMAGE option ROM_SECTION_SIZE = (ROM_SIZE - FALLBACK_SIZE) option ROM_SECTION_OFFSET= 0 end Much better. However, if you have not defined USE_FALLBACK_IMAGE or USE_NORMAL_IMAGE, then you're going to get an error, which in the past was not the case. So we'll have to learn to be a bit more careful about defining things before they are used, and not using things that were never defined. This is all proving to be easy, but people will need to take more care with the Config files (which, in deference to Windows and MacOS, are now named Config.lb). ron _______________________________________________ Linuxbios mailing list Linuxbios(a)clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios

3 2

RE: ADLO/Bochs and PCI config write
by hcyun＠etri.re.kr June 9, 2003

June 9, 2003

> On Mon, 9 Jun 2003, Richard Smith wrote: > > > Any ideas? It seems unlikely that the bochs stuff would be > that fragile so I must be missing something. > > It is :/ > > turn off some deubgging stuff. it helps keep code smaller. Can you explain why turning off debugging macros can help?. I have problems when I enabled debug macros. It fail to run bochs bios. But If I turn off all debug macros, it boot normally. > > I suspect it is b/c of those legacy bios entry points > embedded into the > bios code. they have fixed address. Perhaps it is some > problem with BCC. > > -- > Adam Sulmicki > http://www.eax.com The Supreme Headquarters of the 32 bit registers > > _______________________________________________ > Linuxbios mailing list > Linuxbios(a)clustermatic.org > http://www.clustermatic.org/mailman/listinfo/linuxbios >

3 2

new config language.
by ron minnich June 9, 2003

June 9, 2003

I'm opening up this discussion, which was formerly just between the folks actively doing freebios2. We're getting closer to having things working and I am happy to hear any comments. linuxbios.a now builds for the K8 with the new config tool. Not all of linuxbios, just linuxbios.a, but that's quite a way to working. Things are stricter now. We've been able to operate before with code like this: #if SOME_VARIABLE_THAT_WAS_UNDEFINED == 0 stuff #endif because in the GNU C preprocessor, #if (a == 0) and #ifndef a are equivalent if a is undefined. I think that is a very poor way to operate: it can fool people, and in a BIOS you want more checking. Now that you can use an 'if' operator in the config language, setting up some things is a lot easier: For example, this: expr ROM_SECTION_SIZE =(USE_FALLBACK_IMAGE*65536)+(USE_NORMAL_IMAGE*(ROM_SIZE - 65536)) expr ROM_SECTION_OFFSET=(USE_FALLBACK_IMAGE*(ROM_SIZE-65536))+(USE_NORMAL_IMAGE *0) now becomes this: default FALLBACK_SIZE=65536 if USE_FALLBACK_IMAGE option ROM_SECTION_SIZE = FALLBACK_SIZE option ROM_SECTION_OFFSET= (ROM_SIZE - FALLBACK_SIZE) end if USE_NORMAL_IMAGE option ROM_SECTION_SIZE = (ROM_SIZE - FALLBACK_SIZE) option ROM_SECTION_OFFSET= 0 end Much better. However, if you have not defined USE_FALLBACK_IMAGE or USE_NORMAL_IMAGE, then you're going to get an error, which in the past was not the case. So we'll have to learn to be a bit more careful about defining things before they are used, and not using things that were never defined. This is all proving to be easy, but people will need to take more care with the Config files (which, in deference to Windows and MacOS, are now named Config.lb). ron

3 3

After make, where is file "docipl"?
by roger June 8, 2003

June 8, 2003

ok. completed the make/compile for the smartcore-p5/ but yet there's no file called "docipl". I would assume by the prefix that this is meant for the DOC.. for which i do have. I'm following the Sis documentation as a guideline. >From what I'm seeing in the util/mtd/burn_mtd script, I only need to create the images (with dd) and then dd the images to the DiskOnChip (on this 440BX board). -- Roger http://www.eskimo.com/~roger/index.html

1 0

Re: tusl2-c with kernel
by Yap Ee Huey June 8, 2003

June 8, 2003

Hi, I used RAMTEST=1 and memtest (from memtest86.com) as payload : USE_ELF_BOOT=1 payload memtest (default memtest is elf , correct ?) RAMTEST output : Testing SDRAM : 00000000-0009ffff SDRAM fill: 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 0009ffff SDRAM verify: 00000000 00010000 00020000 00030000 00040000 00050000 00060000 00070000 00080000 00090000 0009ffff Done. MEMTEST loading : Welcome to elfboot, the open sourced starter. January 2002, Eric Biederman. Version 1.2 37:init_bytes() - zkernel_start:0xfff00000 zkernel_mask:0x0000ffff Found ELF candiate at offset 0 header_offset is 0 Try to load at offset 0x0 New segment addr 0x10000 size 0x112b8 offset 0x1000 filesize 0x112b8 (cleaned up) New segment addr 0x10000 size 0x112b8 offset 0x1000 filesize 0x112b8 Loading Segment: addr: 0x000000000ff6e5a8 memsz: 0x00000000000112b8 filesz: 0x00000000000112b8 Jumping to boot code at 0x10000 entry = 0x00010000 lb_start = 0x00004000 lb_size = 0x0004ed2c adjust = 0x0ffad2d4 buffer = 0x0ff625a8 elf_boot_notes = 0x0000cb40 adjusted_boot_notes = 0x0ffb9e14 LinuxBIOS-1.0.0 Mon Jun 9 16:13:25 EDT 2003 starting... (REBOOTED) Looks like my elf boot always failed. Any debug direction ? Thanks. regards, eehuey >From: ron minnich <rminnich(a)lanl.gov> >To: Yap Ee Huey <yapeehuey(a)hotmail.com> >CC: linuxbios(a)clustermatic.org >Subject: Re: tusl2-c with kernel >Date: Fri, 6 Jun 2003 07:40:24 -0600 (MDT) > >Have you just tried memtest first? > >I am wondering if you do in fact have a dram problem. > >ron > _________________________________________________________________ Using a handphone prepaid card? Reload your credit online! http://www.msn.com.my/reloadredir/default.asp

2 1