Silicon Errata OMAP-L137

I have a question regarding silicon revisions and usage notes listed in the Silicon Errata sheets for the OMAP-L137. Is it correct to assume that later silicon revisions have fixed any problems listed for previous silicon revisions unless they have been noted as still being present. Or is it correct to assume that all problems listed for any silicon revision are still present or not proved to be fixed in later revisions. I am asking this question because I see what appears to me to be 2 conflicting statements in the latest silicon errata for the OMAP-L137. These statements are as follows:

Section 2.1 document Literature Number: SPRZ291B states the following

These usage notes will be incorporated into future documentation updates for the device (such as the device-specific data sheet), and the behaviors they describe will not be altered in future silicon revisions.

Is the document trying to say that this description of behavior will be reported in future documentation as it appears here and will not be changed. Or is it saying that it will be reported in future documentation and it will not be fixed in any future silicon revesions. I know that sounds silly but to me thats what it looks like it is saying.

I say this because I came across another statement in the same document that leads me to believe that future silicon revisions do address problems noted in earlier revisions however if the problem was not fixed it will be noted again in the usage notes of the later revisions. The statement is as follows

Section 3.1 document Literature Number: SPRZ291B states the following

Section 2.1, Usage Notes for Silicon Revision 2.0.

So in a nutshell my question is this. If I have a silicon revision 2.0 part do I only have to be concerned with the usage notes for that silicon revision (IE any problems that are still present will be listed again in revision 2.0 notes) or are all of the problems found in the earlier revisions still present and what is listed under revision 2.0 are problems that might have come up in this revision or have been there all along together with any previously noted problems.

I am specifically intrested in the problems noted for the SPI interface reported for earlier silicon revisions. Have these been fixed in revision 2.0 or are they still there and will be there forever? The design that I am planning on using this part on will use the SPI boot option, I can hardly afford to build a design around a boot method that may be unreliable.