These headers shadow system headers.
While they are necessary for building on the LM32, it breaks any
code that wants the other WR headers on a different architecture.

By moving these headers, we can pick whether to use them or not.

Cesar Prados authored 11 years ago and Alessandro Rubini committed 11 years ago

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

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

The associated README explains how to use this stuff with w1 eeprom.

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

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

This is needed to build sdb-lib for the target, in case you rely on
the default value for CROSS_COMPILE instead of using the environment
variable.

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

This is a straight and complete copy of sdbfs/lib from the
fpga-config-space project (ohwr.org).  The last sdbfs commit in there
as I pick this is d37e888.

Further commits in sdb-lib within wrpc-sw are expected to exactly
track fpga-config-space (and report the upstream commit), to avoid
maintainance issues.  However, I prefer not to have a submodule since
it makes things difficult for such a small code subset.

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

sdb-lib/access.c: FIXUP --> fpga-config-space

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

sdb-lib:FIXUP -> fpga-config-s-ace

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

SDB-LIB: fix missing ->read (fppa-c-space)

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

SDBLIB: buffer is void

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

Now "meanPathDelay" is called "oneWayDelay", so the new commits
force a one-line change.

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>

When WR PTP Core is set in GrandMaster mode it aligns its nanosecond
counter to 1-PPS and 10MHz coming from external source. When user wants
to set seconds counter to a desired value (with _time set_ command), the
nanosecond counter was also set (zeroed) causing additional (random)
offset to the external source. The commit adds _time setsec_ and _time
setnsec_ for setting only seconds or nanoseconds counter without
affecting the other one. Now, when user wants to set current time on a
GrandMaster Node he has to call _time setsec_ command after _mode gm_

Conflicts:
	wrc_main.c

when Etherbone support is turned off ARP/ICMP/.. (in general all non-PTP)
traffic has to go into external fabric interface of WRPC because it won't be
handled by WRPC software

We don't want to call the real printf, which brings in the whole of
newlib, but it sometimes happen. So let's diagnose the problem
immediately rather than letting users wonder what is wrong with the
size.

Similarly, if building under ppsi, ensure we are not calling libgcc
division functions: any 64-bit division my be done with __div64_32
offered by ppsi/lib (copied from the kernel).

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

This reverts commit 0bcbe429

.

Now that we are able to check for bogus printf calls, don't define the
name any more (ptp-noposix defines it anyways, but let's not do it
when building with ppsi, so the code explicitly calls pp_printf (or
mprintf, which will remain as an alias until we get rid of
ptp-noposix).

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>

Tomasz Wlostowski authored 11 years ago and Grzegorz Daniluk committed 11 years ago

Signed-off-by: Grzegorz Daniluk <grzegorz.daniluk@cern.ch>