Nach meinem kleinen Exkurs in die Gefilden der Bios- und Flash-Brenner bin ich nun verständlicherweise etwas sensibilisiert wenns ums Thema Backup geht. Prompt wird mal mit Knoppix gebootet und das ganze Geraffel ins Netz weggeschoben. Am Besten so:
backup
dd if=/dev/hda bs=64k | gzip -c | split -b 600m - image.gz
verify
cat image.gz* | gzip -d | cmp - /dev/hda
restore
cat image.gz* | gzip -d | dd of=/dev/hda bs=64k
"hEAdr00m" hatte das ma auf gulli gepostet, stand wohl mal in der c't.
Montag, 29. Juni 2009
Freitag, 26. Juni 2009
Board is alive again!!!
Yes, we did it! Great success! My mainboard is back from the dead. With the cable we build and some serious compiling we managed to flash the asus p5b flash/bios through the spi_j.
When i had the cable build and spipgm detected the chip the first time, i was almost sure that that was it. Check out my first video with some flashing that looked good but only filled the flash/bios with lots of FF values. Then i learned that one needs to recompile the spipgm executable to work with the special write mode that is needed for the p5b chip. I downloaded one already precompiled executable from http://richard-burke.dyndns.org/wordpress/2009/02/107/#comments. But still no success as i again tried the /p option of spipgm to flash. This seems not to be working for p5b as the new write method only gets invoked when flashing with the /s option. This is what i did and finally, the bios contained sensmaking material. A first md5sum showed the bios contained exactly the flashed data. Then, a test, it worked, great success :) Thanks for all the great help and keep up the good work!
When i had the cable build and spipgm detected the chip the first time, i was almost sure that that was it. Check out my first video with some flashing that looked good but only filled the flash/bios with lots of FF values. Then i learned that one needs to recompile the spipgm executable to work with the special write mode that is needed for the p5b chip. I downloaded one already precompiled executable from http://richard-burke.dyndns.org/wordpress/2009/02/107/#comments. But still no success as i again tried the /p option of spipgm to flash. This seems not to be working for p5b as the new write method only gets invoked when flashing with the /s option. This is what i did and finally, the bios contained sensmaking material. A first md5sum showed the bios contained exactly the flashed data. Then, a test, it worked, great success :) Thanks for all the great help and keep up the good work!
Mittwoch, 24. Juni 2009
soldering result
Here are some pictures of the finished cable we managed to create:
#1 shows the attachment of the spare atx power supply to get +3,3v from somewhere.
#2 shows the parallel port plug we used to attach the wiring to.
#3 shows the whole cable that "worked" in terms of the spipgm tool was able to detect the flash type.
still, the board is dead :(
#1 shows the attachment of the spare atx power supply to get +3,3v from somewhere.
#2 shows the parallel port plug we used to attach the wiring to.
#3 shows the whole cable that "worked" in terms of the spipgm tool was able to detect the flash type.
still, the board is dead :(
Asus P5B SPI flash challenges
Well, following the web's advice on how to revive my disabled Asus P5B mainboard after a gone-very-wrong bios flash attempt out of windows vista 64 i am making progress with recovery. Thanks to mondotech i at least have a rough idea on what to do.
So i finally managed to have a friend create a working cable for me as i am not the soldering guy. Seeing the spipgm tool recognizing the attached spi did get me excited, but unfortunately the board stays dead after multiple "successfull" flash attempts. So out of pure curiosity i used the spipgm tool to download the bios file that is in the chip and write it into a file. Guess what, it does differ from the one i used to flash the chip. I failed back to the P5B.ROM provided on the cd that came with the board, but still, no luck. Downloading the file with spipgm from the board i get a "corrupt" file that differs from the one delivered on the CD:
I used the tool "Fairdell HexCmp" which is avaliable as a 14 day trial over at the fairdell webpage. So how can i overcome this issue? Does it mean my board is seriously damaged? Or is it simple bad timing / bad wiring / user error during the flash process? We'll see.
So i finally managed to have a friend create a working cable for me as i am not the soldering guy. Seeing the spipgm tool recognizing the attached spi did get me excited, but unfortunately the board stays dead after multiple "successfull" flash attempts. So out of pure curiosity i used the spipgm tool to download the bios file that is in the chip and write it into a file. Guess what, it does differ from the one i used to flash the chip. I failed back to the P5B.ROM provided on the cd that came with the board, but still, no luck. Downloading the file with spipgm from the board i get a "corrupt" file that differs from the one delivered on the CD:
I used the tool "Fairdell HexCmp" which is avaliable as a 14 day trial over at the fairdell webpage. So how can i overcome this issue? Does it mean my board is seriously damaged? Or is it simple bad timing / bad wiring / user error during the flash process? We'll see.
Abonnieren
Posts (Atom)
