On 10/6/06, Jason Mao <[EMAIL PROTECTED]> wrote:
Hi, Samurai
Well, software may be open, but how could hardware be open
in the same way as software?
You must be trolling.
The furor of the last couple of days (and the last few months/years of
background work) is all about open hardware. Open hardware means not
needing magical blobs in the OS to run. Open hardware means making
register documentation available to those who wish to write drivers.
Open hardware means having complete and accurate documentation.
That rules out NICs that need to have a blob in the driver, rather
than just poking stuff into the chip's registers and leaving the
firmware to figure it out. That rules out video cards that are
minimially functional VESA devices, but need undocumented magic to do
hardware acceleration. That rules out RAID controllers that don't
allow you to read a couple of bytes to query array status, or send a
couple of bytes to start a rebuild. None of that needs to be
proprietary...
Now if you're not satisfied with hardware being black boxes that seem
to do the right thing when you poke registers the right way, look at
the various projects hosted by OpenCores[1] or the LEON[2] GPLed
SPARCv8 clone. Of course, you still need to trust your FPGA...
[1] http://www.opencores.org/browse.cgi/by_category
[2]
http://www.gaisler.com/cms4_5_3/index.php?option=com_content&task=view&id=13&Itemid=53
--
GDB has a 'break' feature; why doesn't it have 'fix' too?
