Clean up ES startup object table

[skliper]

In reviewing #159 it was determined that a better approach to changing the type of the "StackPtr" field would be to do some more general cleanup here and remove the unused field.

[skliper]

Imported from trac issue 139. Created by jphickey on 2016-03-01T11:12:34, last modified: 2019-03-05T14:57:55

[skliper]

Trac comment by jphickey on 2016-03-01 11:20:10:

Commit [changeset:05afcd28] addresses this, branch trac-139-cleanup-objtable

In doing this change I reviewed the other existing fields and determined that the ObjectPtr, ObjectId, StackPtr, TcbPtr fields are also unreferenced. These were also removed.

The only thing not entirely clear is with respect to the ObjectId vs. ObjectKey fields. The existing code had been writing the OSAL-assigned ID back into the ObjectKey field, not ObjectId as the description/comment would suggest.

It is possible we want to rename ObjectKey to ObjectId to better match the comment.

[skliper]

Trac comment by sduran on 2016-03-14 17:00:24:

Seems reasonable to remove these unused fields, but they were put there at some point for a reason - definitely double check this change with Alan Cudmore.

[skliper]

Trac comment by stashakk on 2016-03-15 08:19:09:

Nice clean up makes the file more explicit and clear

[skliper]

Trac comment by acudmore on 2016-03-15 13:31:58:

I think it is OK to remove these fields.
Many of these fields were carried over from heritage FSW, and they were probably included at the start of cFE development.
They have not been needed for any of our current operating systems or platforms and they will probably not be needed in the future.

Also, since this code is in generic cFE code, it is not supposed to be changed or customized for each OS or mission.

My original thought for including the table at all was to allow for patching in flight to add a task or startup function call if needed. ( this is still possible, the unused fields don't really matter in this case )

[skliper]

Trac comment by jphickey on 2016-03-22 17:41:00:

Follow-on change in [changeset:aef7917] which removes "ObjectKey"

I investigated the possibility of qualifying this structure as const since no software is writing to the structure any more. However, I discovered that the unit test, specifically es_UT.c starting on line 496, test case titled "Bad function pointer", is overwriting a field in this table.

To avoid growing this change therefore this does not modify the const-ness of the structure, this will have to be done as a separate ticket.

The startup procedure is being re-evaluated in #173 and this would be the logical place to also address the read-only status of this table.

[skliper]

Trac comment by jphickey on 2016-03-22 18:14:14:

This has added into the ic-2016-03-22 merge

[skliper]

Trac comment by glimes on 2016-03-30 15:10:01:

development now includes ic-2016-03-22 merge.

[skliper]

Trac comment by jhageman on 2019-03-05 14:57:55:

Milestone renamed