This is a short-term fix to actual user needs. We'll soon
have a kernel module to track WR time, but most likely people
won't use that module until the next release.

Meanwhile, the normal "wr_date get" is enhanced to report both UTC and
TAI.

This wr_date is a simple plug-in replacement for installed switches.

Example, after setting a wrong host time on purpose:
(the example above includes extra blank lines to separate the output
from the various commands)

   wrs-192.168.16.242#wr_date get; date; wr_date get tohost; date

   1416482191.851373280 TAI
   2014-11-20 11:16:31.851373280 TAI
   2014-11-20 11:15:56.851373280 UTC

   Mon Nov 10 09:08:14 UTC 2014

   1416482191.895071872 TAI
   2014-11-20 11:16:31.895071872 TAI
   2014-11-20 11:15:56.895071872 UTC

   Thu Nov 20 11:15:56 UTC 2014

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

Adam Wujek authored 10 years ago and Alessandro Rubini committed 10 years ago

Also, fix Makefiles accordingly.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

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

The header is not private to the library, wr_phytool uses it.

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

The header is not private to the library: it is used by elsewhere.

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>

termio setup is restored at exit; si save it initially even if we are
not going to modify it.

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

For unknown reasons, setting stdin to nonblocking mode has global
effects, and thus every later command will get EAGAIN instead
of a blocking read:

   vi: can't read user input

   wrs-192.168.16.242#cat
   cat: read error: Resource temporarily unavailable

But wr_mon doesn't need non-blocking mode, because it uses poll on stdin.
So instead of fixing, remove the source of the bug.

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

If the usb cable is disconnected, any write to /dev/ttyGS0 goes to a
buffer until it fills. Then, any further write will block. This means
the messages towards the user ("uncompressing filesystem... done" etc)
caused the update procedure to stop.

This fixes the problem by using non-blocking write for the messages:
if nobody`s reading the messages are just discarded by the kernel.

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

Without this, the upgrade procedure from v4.0 failed to preserve
the mac address. This fixes the bug which is explained here below.

What happened is that the new filesystem (v4.1) returns UNKNOWN as
fpga type, because no hwinfo is there.  This UNKNOWN should be turned
into the default type LX240T, so the fpga is programmed and everything
proceeds.

However, "UNKNOWN" is intermixed with the (expected) error message,
so no LX240T is selected and loading fpga fails.

Later, "ifconifg wr0" fails to retrieve the mac address, because no
WR drivers are running, and the generation of hwinfo fails as well.
As a result, after the second reboot (see user manual) there MAC
address for WR interfaces was not preserved.

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

This upgrades user-space sdb code the new upstream commit.

This is the result of sdb-read on the switch, for the fpga area:

  wrs-192.168.16.242#sdb-read -l -m 0x100000@0x10000000 -e 0x70000 /dev/mem
  0000000000000651:e6a542c9 @ 00000000-0007ffff WB4-Crossbar-GSI
  0000000000000651:eef0b198 @ 00000000-0001ffff WB4-Bridge-GSI
  0000000000000651:e6a542c9 @ 00000000-0001ffff   WB4-Crossbar-GSI
  000000000000ce42:66cfeb52 @ 00000000-0000ffff   WB4-BlockRAM
  000000000000ce42:e2d13d04 @ 00010000-000100ff   WR-Periph-UART
  000000000000ce42:65158dc0 @ 00010100-000101ff   WR-Soft-PLL
  000000000000ce42:e503947e @ 00010200-0001021f   WB-SPI.Control
  000000000000ce42:441c5143 @ 00010300-000103ff   WB-GPIO-Port
  000000000000ce42:57494266 @ 00010400-000104ff   WB Simple Timer
  000000000000ce42:de0d8ced @ 00010500-000105ff   WR-PPS-Generator
  000000000000ce42:ba07b9d3 @ 00020000-0002ffff WRSW NIC
  0000000000000651:eef0b198 @ 00030000-00037fff WB4...

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

We are now strong with ppsi, remove any reference to 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>

All these files come from fpga-config-space/sdbfs: current top of master
(commit 2584f0053cda6db6390354711687c5cca5ea6aef)

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

Inconsistent naming has always been a problem for me. Since other
developers agree, we chose to rename all stuff to be wrs_ or wr_
for switch-specific stuff and generic WR functionality.  Ok, the
split is not very clear-cut, I agree.

This commit only fixes the tools and their users:
     wrs_version
     wrs_vlan
     wrs_pstats

It doesn't touch wrsw_hal and wrsw_rtud (nor it adds any prefix to
non-prefixed tools like load-lm32 and so on).

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

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

We need to reset the CPU outside of the whole loop. This moves stuff
around a lot to do that. Otherwise the variables are set after the CPU
already started, and it may be too late (it is too late in our
specific case).

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

This allos to write and read variable while loading rt_cpu.elf.
You can read a variable at run time, by not loading the elf file.
To do this use -n to "not" load the file:

   load-lm32 -n rt_cpu.elf pstate=

We still need the file to read the symbol table.  So, <var>= reads
the variable, and <write>=<value> changes it. You can pass several
such actions on the same command line. Please note that you can't
change variables in the BSS section, which is zeroed at run time.

All symbols are trated as 32-bit integer values, so this is not really
generic enough for everything:

  wrs-192.168.16.9# load-lm32 -n rt_cpu.elf memset=
  memset = 134253659 (0x08008c5b)

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

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

This removes the hack of forking a child, because I'll need to do more
than just passing a file name. The other tool using the same hack
(load-virtex) is unchanged as I don't need to work on it.

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

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

You can "diff -w" or "show -w" and see there is no difference. I
didn't shorten long lines nor anything "major", I just removed
trailing spaces, that are highlighted in my editor, and fixed
inconsistencies in indentation within functions.

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 name was inconsistent with anything else, and we must simplify
and get consistent.

A symlink with the old name is provided.

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

I'm confused by TAI/UTC: let's wr_date make it clear that it is
reporting TAI.  Nobody parses this output so far: the web interface
uses it, but it prints back the whole output.

So this change is safe.

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

Every other place calls the file wr_date.conf.

userspace/tools/wr_date.c will open the config file (it does not,
currently) but the name is different.

So we fix the executable and leave the rest as it is, since it works:
documentation matches S70wr_date and use of .config for build-time settings.

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

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

This is needed if for example it is called in a pipe that is then closed.

meanwhile, remove trailing white space.

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