Dear coreboot and Flashrom folks,
the idea is to give out flash chips for free to people (LinuxTag in Berlin from May 23rd to 26th) asking them to test Flashrom and try coreboot. Where the flash chips are given away, at the coreboot or Flashrom booth or at the sponsor’s booth does not matter much.
Reason ------
In my experience having to buy flash chips (or a programmer) is the biggest hurdle to development or testing. Although 20 € for shipping and such is not *that* much it seems to be for some people. Additionally since mostly you cannot buy such things in the shop around the corner it is quite time consuming to shop online.
Benefits --------
For coreboot it could mean just telling people to build an image similar to their board, switch the chip, flash the image to the given chip and then test it. So because people coming home having means to try coreboot (more) easily will get them to the project.
For Flashrom people could test if their mainboard supports all operations, especially erase and wrote, so we could better coverage of mainboard support [1]. Of course only for boards supporting the given away chip.
Type of chip ------------
I have not much of a clue, but I guess the flash chip should have DIP8 form factor and support the SPI protocol. The size should be 16 or 32 Mb MBit.
Help is needed here for suggestions.
Quantity and Price ------------------
500 or 1000 chip should be a reasonable amount to get. One developer (in Europe?) (going to most fairs) could store them and if there is another event (in Europe) the developer will not attend then he could mail a certain amount of chips to developers attending that event.
On IRC scientes found an offer for order of 1000+ pieces for an 32Mbit chip for $1.77 for a piece. Therefore getting 1000 chips would cost 2000 $/€.
Money and sponsors ------------------
It would be great if a company or organization would be willing to finance that project. Maybe the developers who work at company/organizations doing coreboot or Flashrom work or deployment could lobby for the support in their company/organization.
Besides good press coverage and improved reputation I write up some arguments for some companies or organizations.
### AMD ###
Although a statement was made to support coreboot and code for chipset support is contributed, board ports are lacking. Making it easy for people to port coreboot to more AMD boards and by doing so also fix bug or clean up the code should be in AMD’s interest.
### Intel ###
The known reasons apply. Now that the coreboot team at Google made the Sandy Bridge port for their Chromebooks [3], it easier than ever to make a port for other Intel boards with that chipset.
### Google ###
Just plain Free Software principle. By having people work with or try out coreboot, code Google uses will be tested and improved. Also Google doing Flashrom development more testing would be beneficial in that area too.
### Microsoft, Ubuntu and other Free Software distributions ###
Having a BIOS replacement which takes less time to initialize the hardware and let the operating system take over, will (indirectly) improve the experience of their users.
### FSF (Free Software Foundation) ###
As coreboot is a high priority project at the FSF they could support such promotions financially.
### BSI (Bundesamt für Sicherheit in der Informationstechnik) ###
Having companies and citizens using coreboot will be lead to more secure systems, since coreboot being free software it can be audited by everyone and updates do not depend on the willingness of the vendor.
It would be awesome if we could get that project going. I am curious how it will turn out.
Thanks,
Paul
[1] http://flashrom.org/Supported_hardware#Supported_mainboards [2] http://flashrom.org/Technology [3] http://blogs.coreboot.org/blog/2012/04/02/google-releases-sandybridge-suppor... [4] http://www.fsf.org/campaigns/priority-projects/
Paul Menzel wrote:
Where the flash chips are given away, at the coreboot or Flashrom booth or at the sponsor’s booth does not matter much.
Note there will not be a flashrom booth at LinuxTag this year. We'll of course try to answer flashrom questions in the coreboot booth as best we can! :)
Carl-Daniel mentioned that he will be at LinuxTag, but busy in another booth, so we can always get in touch with him if there are some special questions.
I have not much of a clue, but I guess the flash chip should have DIP8 form factor and support the SPI protocol. The size should be 16 or 32 Mb MBit.
SPI DIP 16-32 Mbit is a meaningful target. Some ordering codes to choose from:
AMIC A25L032-F AMIC A25L032-UF Macronix MX25L3205DPI-12G Winbond W25Q32BVDAIG Winbond W25Q32BVDAIP
The first one is in stock at Newark/Farnell, but only in smaller quantity. See below.
500 or 1000 chip should be a reasonable amount to get.
That's a lot of flash chips. If the distributor MOQ is 300 then that would also be a very welcome contribution.
I recommend buying from a higher tier sales channel than Newark/Farnell/Mouser/Digi-Key because of cost and because this order is likely to reach MOQ. I'm happy to help find the most appropriate sales channel for anyone interested in contributing to the project in this way.
Besides good press coverage and improved reputation
It would of course also buy prominent exposure on printed material accompanying the flash chips.
//Peter
I'm supprised you found DIP8 chips in those sizes. I tried to buy them and could not find them, granted I was looking at single quantity items. Instead, I've orderd some SOIC8 parts and some adapters for them. With those Items in hand, I started designing the 'dual-flash' module that has passed this list a few times before.
I am currently trying to incorporate Peter's suggestions into my design which should be done in a week or so.
I think your idea is true and great, but bundled with the PCB i'm designing would be even beter and should cost about the same, as SOIC parts are cheaper then DIP8 parts it seems.
Just think of it, handing out a little adapter, where the user can insert his own DIP8 flash and with the change of a switch/jumper test coreboot! Just replace the jumper and their trusted old bios returns.
Just my two cents :)
Oliver
On 17-04-12 09:55, Paul Menzel wrote:
Dear coreboot and Flashrom folks,
the idea is to give out flash chips for free to people (LinuxTag in Berlin from May 23rd to 26th) asking them to test Flashrom and try coreboot. Where the flash chips are given away, at the coreboot or Flashrom booth or at the sponsor's booth does not matter much.
Reason
In my experience having to buy flash chips (or a programmer) is the biggest hurdle to development or testing. Although 20 EUR for shipping and such is not *that* much it seems to be for some people. Additionally since mostly you cannot buy such things in the shop around the corner it is quite time consuming to shop online.
Benefits
For coreboot it could mean just telling people to build an image similar to their board, switch the chip, flash the image to the given chip and then test it. So because people coming home having means to try coreboot (more) easily will get them to the project.
For Flashrom people could test if their mainboard supports all operations, especially erase and wrote, so we could better coverage of mainboard support [1]. Of course only for boards supporting the given away chip.
Type of chip
I have not much of a clue, but I guess the flash chip should have DIP8 form factor and support the SPI protocol. The size should be 16 or 32 Mb MBit.
Help is needed here for suggestions.
Quantity and Price
500 or 1000 chip should be a reasonable amount to get. One developer (in Europe?) (going to most fairs) could store them and if there is another event (in Europe) the developer will not attend then he could mail a certain amount of chips to developers attending that event.
On IRC scientes found an offer for order of 1000+ pieces for an 32Mbit chip for $1.77 for a piece. Therefore getting 1000 chips would cost 2000 $/EUR.
Money and sponsors
It would be great if a company or organization would be willing to finance that project. Maybe the developers who work at company/organizations doing coreboot or Flashrom work or deployment could lobby for the support in their company/organization.
Besides good press coverage and improved reputation I write up some arguments for some companies or organizations.
### AMD ###
Although a statement was made to support coreboot and code for chipset support is contributed, board ports are lacking. Making it easy for people to port coreboot to more AMD boards and by doing so also fix bug or clean up the code should be in AMD's interest.
### Intel ###
The known reasons apply. Now that the coreboot team at Google made the Sandy Bridge port for their Chromebooks [3], it easier than ever to make a port for other Intel boards with that chipset.
### Google ###
Just plain Free Software principle. By having people work with or try out coreboot, code Google uses will be tested and improved. Also Google doing Flashrom development more testing would be beneficial in that area too.
### Microsoft, Ubuntu and other Free Software distributions ###
Having a BIOS replacement which takes less time to initialize the hardware and let the operating system take over, will (indirectly) improve the experience of their users.
### FSF (Free Software Foundation) ###
As coreboot is a high priority project at the FSF they could support such promotions financially.
### BSI (Bundesamt für Sicherheit in der Informationstechnik) ###
Having companies and citizens using coreboot will be lead to more secure systems, since coreboot being free software it can be audited by everyone and updates do not depend on the willingness of the vendor.
It would be awesome if we could get that project going. I am curious how it will turn out.
Thanks,
Paul
[1] http://flashrom.org/Supported_hardware#Supported_mainboards [2] http://flashrom.org/Technology [3] http://blogs.coreboot.org/blog/2012/04/02/google-releases-sandybridge-suppor... [4] http://www.fsf.org/campaigns/priority-projects/