This is a trivial patch, swapping two blocks and adding a few
lines, so to allow "position=" and "maxsize=" to be used for
subdirectories.

This allows to reserve space for a new SDBFS image to be added
at a later time.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This commit doesn't change actual output for gensdbfs, but it prepares
for recursive scanning of directories. This is a separate commit to ease
my own debugging.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

External projects copying this directory will most likely have
their own "include" directory, but not always "include/linux".
This, use -I../include to they can place sdb.h in there without
modifying their copy of this Makefile.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

The interrupt discussion has been moved to the projects' wiki.
Besides that, some typos, external references and nothing else.
And today is not yesterday: new date is there.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This is mainly due to feedback by Matthieu Cattin and Wesley Terpstra
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This adds a section that explains why we don't describe interrupts.
Most of the section comes from a message by Wesley Terpstra,

  http://lists.ohwr.org/sympa/arc/fpga-config-space/2013-03/msg00000.html

(you may need to re-enter the same url in the browser after clicking on
the "I'm not a spammer" button).

I'm not personally able to make the discussion shorter, because I think
it's very informative.  The message has been slightly reworded to make
it less informal and more suitable to be an "official" rationale, and
I added some glue lines at the beginning and the end.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

With the new forced page breaks (previous commit) two pages had just a
single line in them. So this shrinks a little the text to fix the
problem.  I don't want to force page breaks only in some of the
subsections, for consistency.

Note: I tried the argument "3" to page breaks, to have them strongly
"suggeested" but not "mandated" (argument 4 or default if no argument),
but the result was like without page breaks (i.e., bad).
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This is useful so the image and the table of fields are always on the
same page.  The default effect with float figures and table was
difficult to use as a reference while writing code.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This adds the "repo_url" and "synthesis" records, that are already
used in one of your projects.

Thanks to Matthieu Cattin for his help in designing the structures
and actually implementing them.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

By moving the integration description after device and bridge,
we describe the records in the numerical order, which is also
logical because the integration is only an informative record.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This is a merge of all previous work done on sdbfs using a different
repository (which I used to push to gitorious).

Besides the kernel stuff, which is still in flush, userspace and lib
are actually useful to sdb users in general, so this is a better place
to host it.

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

According to the kernel headers you have, compiling the kernel part
may fail. So compile userspace first, in case you need gensdbfs alone.
Thanks to Matthieu Cattin for noting this.
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

This a major thing, that changes a lot. However, now directories work
properly and the code is generally cleaner. This is what I get,
for example, for the wr-nic release gateware on a SPEC:

  spusa.root# find /mnt
  /mnt
  /mnt/WB4-Crossbar-GSI
  /mnt/WR-CORE
  /mnt/WR-CORE/WB4-Crossbar-GSI
  /mnt/WR-CORE/WB4-BlockRAM
  /mnt/WR-CORE/WB4-Bridge-GSI
  /mnt/WR-CORE/WB4-Bridge-GSI/WB4-Crossbar-GSI
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-Mini-NIC
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-Endpoint
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-Soft-PLL
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-PPS-Generator
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-Periph-Syscon
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-Periph-UART
  /mnt/WR-CORE/WB4-Bridge-GSI/WR-Periph-1Wire
  /mnt/WR-CORE/WB4-Bridge-GSI/WB4-Streaming-DMA_0
  /mnt/WR-NIC
  /mnt/WB-VIC-Int.Control
  /mnt/WR-TXTSU
  /mnt/WR-DIO-Core
  /mnt/WR-DIO-Core/WB4-Crossbar-GSI
  /mnt/WR-DIO-Core/WR-1Wire-master
  /mnt/WR-DIO-Core/WB-I2C-Master
  /mnt/WR-DIO-Core/WB-GPIO-Port
  /mnt/WR-DIO-Core/WR-DIO-Registers
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>