build: bump to use c++17

Bump build to use c++17.

Change-Id: I0ba2efeba2e88f89a18a8d5786253e5a48334ef0
Signed-off-by: Patrick Venture <venture@google.com>

build: set language to C++

Set the language in the build to C++.

Change-Id: I2c91eb95672f2dd0f1c6f21dea47770f900c9393
Signed-off-by: Patrick Venture <venture@google.com>

Add MAINTAINERS file

Change-Id: Icd41a920752ff6f564469b910d03dea8c0bfd6ac
Signed-off-by: Andrew Jeffery <andrew@aj.id.au>

README: Reword cleaning the repo

Reworded cleaning the repo. The old wording did not make sense.

Change-Id: I0e8809a0ebfb68c63707c8a7b9a1c5669eb16c0a
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>

Spelling fixes

Spelling errors found using github.com/lucasdemarchi/codespell
A tool to fix common misspellings.
This tool is licensed under GNU General Public License, version 2.

Change-Id: I0c07f

Spelling fixes

Spelling errors found using github.com/lucasdemarchi/codespell
A tool to fix common misspellings.
This tool is licensed under GNU General Public License, version 2.

Change-Id: I0c07fd50d855db50b8b8891c17f6806dc39e6599
Signed-off-by: Gunnar Mills <gmills@us.ibm.com>

show more ...

Implemented phosphor-event tests to gtest

This is the implemented portion of the tests
over to gtest. As I have done previously I
used test.C as a guide to complete the
implementation of the tests.

Implemented phosphor-event tests to gtest

This is the implemented portion of the tests
over to gtest. As I have done previously I
used test.C as a guide to complete the
implementation of the tests. One thing worth
noting is that in the main method I had to
modify the final few lines of code so that once
the tests finished running, we would call setup()
which clears the events created during this
test. This is necessary so that if you were
to run the tests a second time you wouldn't
continue off from the previous event number
you were at and instead you will now be starting
fresh.

Resolved openbmc/openbmc#561

Change-Id: Ic192ac7753c7a9da541e8ed6efb272f43486ef95
Signed-off-by: Josh King <jdking@us.ibm.com>

show more ...

Broke down our unit tests into gtest functions

I analyzed test.C to see which of the tests we
had created and were running. I took the broken
down tests and went over to the utest.cpp to start
the

Broke down our unit tests into gtest functions

I analyzed test.C to see which of the tests we
had created and were running. I took the broken
down tests and went over to the utest.cpp to start
the functions and listed them out. These functions
still need to be implemented and that is my next
step. I will use test.C to complete the implementation.
This commit was merly only for the breaking down of
the functions.

Change-Id: Ib3a75b3e2d4a7fba71b4ea337e552a4505100f5f
Signed-off-by: Josh King <jdking@us.ibm.com>

show more ...

Created gtest Framework - added necessary files

In order to convert phosphor-event tests to gtest
I needed to start by creating the framework for
the tests. I created a new sub directory called
tes

Created gtest Framework - added necessary files

In order to convert phosphor-event tests to gtest
I needed to start by creating the framework for
the tests. I created a new sub directory called
test in which I created the cpp file utest where I will
be implementing the unit tests. As of right now
I have a sample test in place to verify that everything
is linked properly and will successfully pass its
test. Sample.cpp and Sample.h are needed for the
example test case created and ran inside utest.cpp.
I had to modify both the Makefile.am and configure.ac
to accomidate my new test directory so it would be
included in the make process.

Change-Id: I163584007f52b763c564de516effe35ae6055eb3
Signed-off-by: Josh King <jdking@us.ibm.com>

show more ...

Convert build process to autotools

Replaced the use of a manual Makefile with the use of autotools to
automatically verify and generate the necessary build files. Follow the
steps outlined within th

Convert build process to autotools

Replaced the use of a manual Makefile with the use of autotools to
automatically verify and generate the necessary build files. Follow the
steps outlined within the README.md file to build the package.

Change-Id: I75a0cc5029a63f448866cf38f3edc6d9be363c7e
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>

show more ...

Makefile: CPPFLAGS -> CXXFLAGS

CPPFLAGS means "C Pre-Processor" and not "C++".

Change-Id: Idd57a27cb60aa50645176bfe30da3b318dcd3e5f
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>

Ensure all tools are building with -Wall and -Werror

Added -Wall & -Werror to CFLAGS. Added -Werror to CPPFLAGS.
Also corrected resulting compiler errors.

Change-Id: I1777c3c122edd4b6382a565caf2967

Ensure all tools are building with -Wall and -Werror

Added -Wall & -Werror to CFLAGS. Added -Werror to CPPFLAGS.
Also corrected resulting compiler errors.

Change-Id: I1777c3c122edd4b6382a565caf2967bb9b8afd14
Signed-off-by: Matthew Barth <msbarth@us.ibm.com>

show more ...

phosphor-event: Add dbus method acceptBMCMessage

Add dbus method to accept event log from BMC.
This patch partially resolves issue: openbmc/skeleton#64

Change-Id: Icd19e02a1497a1139f7ba76226a3c16be

phosphor-event: Add dbus method acceptBMCMessage

Add dbus method to accept event log from BMC.
This patch partially resolves issue: openbmc/skeleton#64

Change-Id: Icd19e02a1497a1139f7ba76226a3c16bef6e85f0
Signed-off-by: Yi Li <adamliyi@msn.com>

show more ...

event log limit not considering preexisting logs

Turns out the code that checks for event log size limitations forgot
to take in to account the total size of logs that already existed. So
it would

event log limit not considering preexisting logs

Turns out the code that checks for event log size limitations forgot
to take in to account the total size of logs that already existed. So
it would let you create the max size of new logs rather then new+old.

I fixed this by initializing the variable when the process starts. I
also added a test case that catches the problem.

show more ...

Create a possible limit to the number of logs recorded

https://github.com/openbmc/openbmc/issues/268

There was a technique that created over 1000 event logs. This taxed
the REST interface a lot.

Create a possible limit to the number of logs recorded

https://github.com/openbmc/openbmc/issues/268

There was a technique that created over 1000 event logs. This taxed
the REST interface a lot. It caused connections to time out. I am
adding a way to limit the number of physical logs irregardless of the
size. Like the size limiter, logs are not deleted automatically.

That may actually be a debate (save new logs, delete old logs) vs
(delete new logs, save old logs). I welcome the suggestions.

show more ...

Limit logging

1. Change open/tellg/close to a single stat.

Signed-off-by: Nan Li <bjlinan@cn.ibm.com>

Add support for limiting the size of the event log DB

embedded systems only have so much space. This feature enables the event deamon to
limit the number of bytes used for logs via an input paramet

Add support for limiting the size of the event log DB

embedded systems only have so much space. This feature enables the event deamon to
limit the number of bytes used for logs via an input parameter

Limit the DB to 1k...
/usr/sbin/phosphor-eventd -s 1000

Without the parameter the db size is unlimited.

When the db hits the limit, the event is NOT logged to the db. It is however
added (without debug data) to the syslog still AND an error message is posted
in the syslog...

event logger reached maximum capacity, event not logged

Interfaces exist already to clear the entire db and individual events.

show more ...

Add association interface capabilities for event logs

When a new event log is created it may create an interface
with a name of org.openbmc.association. If there are associations
between the log an

Add association interface capabilities for event logs

When a new event log is created it may create an interface
with a name of org.openbmc.association. If there are associations
between the log and other entites it will have a single property
named "association" which will be an a(sss).
That will be enough information for the rest server to present
it.

Note if the log does not have FRU call outs no association interface will exist

Lets say you want to run the acceptTestMessage method. It
will create an association between the event log and 2 dimms.

So
/org/openbmc/inventory/system/chassis/motherboard/dimm2
/org/openbmc/inventory/system/chassis/motherboard/dimm3

/org/openbmc/records/events/1/fru
should return
data : {/org/openbmc/inventory/system/chassis/motherboard/dimm2 , /org/openbmc/inventory/system/chassis/motherboard/dimm3}

You should also be able to go to the dimm and get the association in reverse

/org/openbmc/inventory/system/chassis/motherboard/dimm2/event
should return
data : {/org/openbmc/records/events/1/}

show more ...

Allow the make check to work on a docker image

The docker image would fail to run the make check because the event directory did
not exist. I've resolved that by always making the directory as part

Allow the make check to work on a docker image

The docker image would fail to run the make check because the event directory did
not exist. I've resolved that by always making the directory as part of the app
kickoff

show more ...

Add flush after writing

It was discovered that if you saved the file and then rebooted the BMC
right after using a violent technique such as AC pulls the logs never
made it to disk. Adding a flush

Add flush after writing

It was discovered that if you saved the file and then rebooted the BMC
right after using a violent technique such as AC pulls the logs never
made it to disk. Adding a flush will resolve that.

show more ...

build: Print the gcc in the container, not the host

Signed-off-by: Joel Stanley <joel@jms.id.au>

build: Don't run upgrade

We don't need to do this every time. Pulling the fresh image from docker
should be enough.

Signed-off-by: Joel Stanley <joel@jms.id.au>

build: Run make check

Signed-off-by: Joel Stanley <joel@jms.id.au>

Add gitignore

Signed-off-by: Joel Stanley <joel@jms.id.au>

Add a script to build inside Docker

This allows phosphor-event to be on a machine with Docker installed
that does not have all the dependencies installed.

A docker container is built and used for t

Add a script to build inside Docker

This allows phosphor-event to be on a machine with Docker installed
that does not have all the dependencies installed.

A docker container is built and used for the build process, with the
binary built in-place.

Signed-off-by: Joel Stanley <joel@jms.id.au>

show more ...

Convert to known version of assert

I made my own ASSERT complete with cool text. I was encouraged
to use existing C tooling. So here you go