I would like to add my name to the list of people who have suceeded with the EPIA-MII LinuxBIOS.
The VGA BIOS offset given by Nick Barker was no good for my board (Model: EPIA-Mii12000 LVDS), I had to use skip=786432 as suggested by Jun.
I am having problems with the kernel recognising the CF device as hde1 though. I am able to manually mount hde1 from a diagnostics shell within the initrd after running cardmgr, so it is available, but once I exit the shell the kernel is saying it can't find hde1.
I'm not sure if Nick used devfs with his setup, but that's not available for my 2.6.14 kernel. I have used mknod to create the device, and also tried using udevstart but still have the same problem. I wonder if it's something to do with me having all the pcmcia stuff compiled directly into the kernel instead of as loadable modules? Unfortunately I'm quite rusty with Linux!
Anyway, I suppose this is a bit off topic. The LinuxBIOS seems to work great, thank you, and big thanks to Nick for all your work on the EPIA M2.
Oh, by the way, I used Award's AWDFLASH to flash linuxbios.bin. Using the undocumented /F option. Nick's HOWTO talked about a flash_and_burn utility which wasn't in the tree, and the flashrom tool just wouldn't compile for me for some reason. AWDFLASH worked, except for some strange quirks. I can't be certain if it was due to bad floppies on my part, but for some reason I felt I was having better success overwriting a good bios, rather than re-trying on a failed flash. The machine would go loopy sometimes. It all works now anyway.
cheers, Carl
this is the second report of flashrom not compiling. Can you send in your error messages, and uname -a, and what distro this is?
ron
* Ronald G Minnich rminnich@lanl.gov [051201 04:59]:
this is the second report of flashrom not compiling. Can you send in your error messages, and uname -a, and what distro this is?
This is very likely to be caused by th -Werror c flag.
Different gcc versions are pretty quick with warning about code.
Another reason might be that pciutils-devel is not installed.
I've added a note about this in the flashrom readme.
The error messages would indeed be very useful.
Stefan
Here is the uname: Linux thinkpad 2.6.12-gentoo-r6 #5 SMP Tue Nov 29 20:39:38 GMT 2005 i686 Intel(R) Pentium(R) M processor 1.70GHz GenuineIntel GNU/Linux
The error is something to do with the include of pci.h. I apologise if it's due to my system being broken. Here's the output of make:
thinkpad flashrom # make gcc -O2 -g -Wall -Werror -c -o flash_enable.o flash_enable.c flash_enable.c:3:17: pci.h: No such file or directory flash_enable.c:6: warning: `struct pci_dev' declared inside parameter list flash_enable.c:6: warning: its scope is only this definition or declaration, which is probably not what you want flash_enable.c:60: warning: `struct pci_dev' declared inside parameter list flash_enable.c: In function `enable_flash_e7500': flash_enable.c:68: warning: implicit declaration of function `pci_read_byte' flash_enable.c:75: warning: implicit declaration of function `pci_write_byte' flash_enable.c: At top level: flash_enable.c:85: warning: `struct pci_dev' declared inside parameter list flash_enable.c:110: warning: `struct pci_dev' declared inside parameter list flash_enable.c:148: warning: `struct pci_dev' declared inside parameter list flash_enable.c:164: warning: `struct pci_dev' declared inside parameter list flash_enable.c:180: warning: `struct pci_dev' declared inside parameter list flash_enable.c:196: warning: `struct pci_dev' declared inside parameter list flash_enable.c:219: warning: `struct pci_dev' declared inside parameter list flash_enable.c:253: warning: `struct pci_dev' declared inside parameter list flash_enable.c:290: warning: `struct pci_dev' declared inside parameter list flash_enable.c:294: warning: initialization from incompatible pointer type flash_enable.c:295: warning: initialization from incompatible pointer type flash_enable.c:296: warning: initialization from incompatible pointer type flash_enable.c:297: warning: initialization from incompatible pointer type flash_enable.c:298: warning: initialization from incompatible pointer type flash_enable.c:299: warning: initialization from incompatible pointer type flash_enable.c:300: warning: initialization from incompatible pointer type flash_enable.c:301: warning: initialization from incompatible pointer type flash_enable.c:302: warning: initialization from incompatible pointer type flash_enable.c:303: warning: initialization from incompatible pointer type flash_enable.c:304: warning: initialization from incompatible pointer type flash_enable.c:305: warning: initialization from incompatible pointer type flash_enable.c: In function `enable_flash_write': flash_enable.c:315: warning: implicit declaration of function `pci_alloc' flash_enable.c:315: warning: assignment makes pointer from integer without a cast flash_enable.c:317: warning: implicit declaration of function `pci_init' flash_enable.c:318: warning: implicit declaration of function `pci_scan_bus' flash_enable.c:323: error: storage size of `f' isn't known flash_enable.c:326: warning: implicit declaration of function `pci_filter_init' flash_enable.c:329: error: dereferencing pointer to incomplete type flash_enable.c:329: error: dereferencing pointer to incomplete type flash_enable.c:330: warning: implicit declaration of function `pci_filter_match' flash_enable.c:323: warning: unused variable `f' flash_enable.c:339: warning: passing arg 1 of pointer to function from incompatible pointer type make: *** [flash_enable.o] Error 1
regards, Carl ----- Original Message ----- From: "Ronald G Minnich" rminnich@lanl.gov To: "Carl Farrington" linuxbios@compsup.demon.co.uk Cc: linuxbios@openbios.org Sent: Thursday, December 01, 2005 3:59 AM Subject: Re: [LinuxBIOS] EPIA M-II (CF Boot)
this is the second report of flashrom not compiling. Can you send in your error messages, and uname -a, and what distro this is?
ron
-- LinuxBIOS mailing list LinuxBIOS@openbios.org http://www.openbios.org/mailman/listinfo/linuxbios
* Carl Farrington linuxbios@compsup.demon.co.uk [051201 12:12]:
Here is the uname: Linux thinkpad 2.6.12-gentoo-r6 #5 SMP Tue Nov 29 20:39:38 GMT 2005 i686 Intel(R) Pentium(R) M processor 1.70GHz GenuineIntel GNU/Linux
The error is something to do with the include of pci.h. I apologise if it's due to my system being broken. Here's the output of make:
No, your system is not broken, it just lacks libpci.a and it's headers. On Redhat and SUSE the packages you need to install are called pciutils and pciutils-devel. I'd assume they have the same or a similar name on Gentoo.
Regards, Stefan
On Thu, Dec 01, 2005 at 12:55:31PM +0100, Stefan Reinauer wrote:
No, your system is not broken, it just lacks libpci.a and it's headers. On Redhat and SUSE the packages you need to install are called pciutils and pciutils-devel. I'd assume they have the same or a similar name on Gentoo.
They do. emerge picutils and it'll work.
//Peter
Thats great, thank you. I went hunting around my system for pci.h and found it in the kernel sources, and couldn't figure out why it wasn't working!
----- Original Message ----- From: "Peter Stuge" stuge-linuxbios@cdy.org To: linuxbios@openbios.org Sent: Friday, December 02, 2005 9:53 AM Subject: Re: [LinuxBIOS] EPIA M-II (CF Boot)
On Thu, Dec 01, 2005 at 12:55:31PM +0100, Stefan Reinauer wrote:
No, your system is not broken, it just lacks libpci.a and it's headers. On Redhat and SUSE the packages you need to install are called pciutils and pciutils-devel. I'd assume they have the same or a similar name on Gentoo.
They do. emerge picutils and it'll work.
//Peter
Carl Farrington wrote:
Here is the uname: Linux thinkpad 2.6.12-gentoo-r6 #5 SMP Tue Nov 29 20:39:38 GMT 2005 i686 Intel(R) Pentium(R) M processor 1.70GHz GenuineIntel GNU/Linux
The error is something to do with the include of pci.h. I apologise if it's due to my system being broken. Here's the output of make:
thinkpad flashrom # make gcc -O2 -g -Wall -Werror -c -o flash_enable.o flash_enable.c flash_enable.c:3:17: pci.h: No such file or directory
you're missing an include -- you need pci-devel.
ron
* Ronald G Minnich rminnich@lanl.gov [051201 04:59]:
this is the second report of flashrom not compiling. Can you send in your error messages, and uname -a, and what distro this is?
I have written the following small patch for the flashrom utility Makefile. It will check for pciutils installation before compiling anything and print an explicit error message if pciutils is not found.
This is much easier to parse than the half screen gcc errors.
Issue 40.
Commit?
Stefan
Stefan Reinauer wrote:
- Ronald G Minnich rminnich@lanl.gov [051201 04:59]:
this is the second report of flashrom not compiling. Can you send in your error messages, and uname -a, and what distro this is?
I have written the following small patch for the flashrom utility Makefile. It will check for pciutils installation before compiling anything and print an explicit error message if pciutils is not found.
This is much easier to parse than the half screen gcc errors.
Issue 40.
Commit?
yes please
I will get back on issue tracker I hope today and add a signed-off-by
ron
Gooood! but you can run X? This is the key. And I can not yet.
--- Okajima, Jun. Tokyo, Japan.
I would like to add my name to the list of people who have suceeded with the EPIA-MII LinuxBIOS.
The VGA BIOS offset given by Nick Barker was no good for my board (Model: EPIA-Mii12000 LVDS), I had to use skip=786432 as suggested by Jun.
I am having problems with the kernel recognising the CF device as hde1 though. I am able to manually mount hde1 from a diagnostics shell within the initrd after running cardmgr, so it is available, but once I exit the shell the kernel is saying it can't find hde1.
I'm not sure if Nick used devfs with his setup, but that's not available for my 2.6.14 kernel. I have used mknod to create the device, and also tried using udevstart but still have the same problem. I wonder if it's something to do with me having all the pcmcia stuff compiled directly into the kernel instead of as loadable modules? Unfortunately I'm quite rusty with Linux!
Anyway, I suppose this is a bit off topic. The LinuxBIOS seems to work great, thank you, and big thanks to Nick for all your work on the EPIA M2.
Oh, by the way, I used Award's AWDFLASH to flash linuxbios.bin. Using the undocumented /F option. Nick's HOWTO talked about a flash_and_burn utility which wasn't in the tree, and the flashrom tool just wouldn't compile for me for some reason. AWDFLASH worked, except for some strange quirks. I can't be certain if it was due to bad floppies on my part, but for some reason I felt I was having better success overwriting a good bios, rather than re-trying on a failed flash. The machine would go loopy sometimes. It all works now anyway.
cheers, Carl
-- LinuxBIOS mailing list LinuxBIOS@openbios.org http://www.openbios.org/mailman/listinfo/linuxbios
I can't get past initrd yet! VFS: Unable to mount root hde1 or unknown-block (0,0). It seems that even though I'm starting up cardmgr in the initrd, which sucessfully finds hde and hde1, upon finishing the initrd (closing the shell or end of linuxrc), the kernel doesn't find hde1. Could this be to do with my 2.6 kernel not having devfs. I have fired up udevstart though, although I'm not 100% sure what I'm doing, the device nodes are being created.
Or could it be that I need to upgrade to pcmciautils - the new pcmcia-stuff, as the kernel is warning me about use of depreciated ioctls with cardmgr.
ho hum. I haven't bothered playing on it tonight, as I've been staying up 'til ~4:30am on this lately and that's bad for my work!
cheers! Carl ----- Original Message ----- From: "Jun OKAJIMA" okajima@digitalinfra.co.jp To: "Carl Farrington" linuxbios@compsup.demon.co.uk Cc: linuxbios@openbios.org Sent: Friday, December 02, 2005 1:33 AM Subject: Re: [LinuxBIOS] EPIA M-II (CF Boot)
Gooood! but you can run X? This is the key. And I can not yet.
--- Okajima, Jun. Tokyo, Japan.
I would like to add my name to the list of people who have suceeded with the EPIA-MII LinuxBIOS.
The VGA BIOS offset given by Nick Barker was no good for my board (Model: EPIA-Mii12000 LVDS), I had to use skip=786432 as suggested by Jun.
I am having problems with the kernel recognising the CF device as hde1 though. I am able to manually mount hde1 from a diagnostics shell within the initrd after running cardmgr, so it is available, but once I exit the shell the kernel is saying it can't find hde1.
I'm not sure if Nick used devfs with his setup, but that's not available for my 2.6.14 kernel. I have used mknod to create the device, and also tried using udevstart but still have the same problem. I wonder if it's something to do with me having all the pcmcia stuff compiled directly into the kernel instead of as loadable modules? Unfortunately I'm quite rusty with Linux!
Anyway, I suppose this is a bit off topic. The LinuxBIOS seems to work great, thank you, and big thanks to Nick for all your work on the EPIA M2.
Oh, by the way, I used Award's AWDFLASH to flash linuxbios.bin. Using the undocumented /F option. Nick's HOWTO talked about a flash_and_burn utility which wasn't in the tree, and the flashrom tool just wouldn't compile for me for some reason. AWDFLASH worked, except for some strange quirks. I can't be certain if it was due to bad floppies on my part, but for some reason I felt I was having better success overwriting a good bios, rather than re-trying on a failed flash. The machine would go loopy sometimes. It all works now anyway.
cheers, Carl
-- LinuxBIOS mailing list LinuxBIOS@openbios.org http://www.openbios.org/mailman/listinfo/linuxbios
How about checking these stuff? -- Dev node is created -- Root dev is properly assigned, path and (major, minor). But, what you have to do now is gettting sleep.
--- Okajima, Jun. Tokyo, Japan.
I can't get past initrd yet! VFS: Unable to mount root hde1 or unknown-block (0,0). It seems that even though I'm starting up cardmgr in the initrd, which sucessfully finds hde and hde1, upon finishing the initrd (closing the shell or end of linuxrc), the kernel doesn't find hde1. Could this be to do with my 2.6 kernel not having devfs. I have fired up udevstart though, although I'm not 100% sure what I'm doing, the device nodes are being created.
Or could it be that I need to upgrade to pcmciautils - the new pcmcia-stuff, as the kernel is warning me about use of depreciated ioctls with cardmgr.
ho hum. I haven't bothered playing on it tonight, as I've been staying up 'til ~4:30am on this lately and that's bad for my work!
cheers! Carl ----- Original Message ----- From: "Jun OKAJIMA" okajima@digitalinfra.co.jp To: "Carl Farrington" linuxbios@compsup.demon.co.uk Cc: linuxbios@openbios.org Sent: Friday, December 02, 2005 1:33 AM Subject: Re: [LinuxBIOS] EPIA M-II (CF Boot)
Gooood! but you can run X? This is the key. And I can not yet.
--- Okajima, Jun. Tokyo, Japan.
I would like to add my name to the list of people who have suceeded with the EPIA-MII LinuxBIOS.
The VGA BIOS offset given by Nick Barker was no good for my board (Model: EPIA-Mii12000 LVDS), I had to use skip=786432 as suggested by Jun.
I am having problems with the kernel recognising the CF device as hde1 though. I am able to manually mount hde1 from a diagnostics shell within the initrd after running cardmgr, so it is available, but once I exit the shell the kernel is saying it can't find hde1.
I'm not sure if Nick used devfs with his setup, but that's not available for my 2.6.14 kernel. I have used mknod to create the device, and also tried using udevstart but still have the same problem. I wonder if it's something to do with me having all the pcmcia stuff compiled directly into the kernel instead of as loadable modules? Unfortunately I'm quite rusty with Linux!
Anyway, I suppose this is a bit off topic. The LinuxBIOS seems to work great, thank you, and big thanks to Nick for all your work on the EPIA M2.
Oh, by the way, I used Award's AWDFLASH to flash linuxbios.bin. Using the undocumented /F option. Nick's HOWTO talked about a flash_and_burn utility which wasn't in the tree, and the flashrom tool just wouldn't compile for me for some reason. AWDFLASH worked, except for some strange quirks. I can't be certain if it was due to bad floppies on my part, but for some reason I felt I was having better success overwriting a good bios, rather than re-trying on a failed flash. The machine would go loopy sometimes. It all works now anyway.
cheers, Carl
-- LinuxBIOS mailing list LinuxBIOS@openbios.org http://www.openbios.org/mailman/listinfo/linuxbios
It seems that even though I'm starting up cardmgr in the initrd, which sucessfully finds hde and hde1, upon finishing the initrd (closing the shell or end of linuxrc), the kernel doesn't find hde1. Could this be to do with my 2.6 kernel not having devfs. I have fired up udevstart though, although I'm not 100% sure what I'm doing, the device nodes are being created.
Where did you get your initrd from? Custom built or script generated? Functional root filesystems can be a bit tricky to get right.
-- Richard A. Smith