(unknown charset) Re: Updated SATA support patches for Intel ICH & Promise cards

David Schultz writes:
| I'm not asking you to do anything extreme like give up
| maintainership of the ata driver; that would be detrimental if

I strongly agree with this.  Soren has tons of ATA knowledge and
work arounds.  That is a valuable asset that we shouldn't lose.

| anything.  But Doug has merged his changes with -CURRENT several
| times now.  I think that with a minimal amount of effort, you
| could work more closely with other people who are willing to work
| on the ATA driver.  Keep in mind that these people are trying to
| improve the ATA driver while causing *less* work for you than if
| you did it yourself; it ought to be easier to merge someone else's
| patch than to write and test the changes yourself.  If there's a
| technical reason why you don't like someone's changes, please be
| upfront about it so they can address your concerns.

I can fairly easily re-format things or stick storage of flags/bit
in different places.  Most of the code I've submitted has been separate
functions that can be linked into place.  The only code changes I did
to the core was to address issues like the dev pointer not being
initialized before I needed and linking reset code to recover
from a bad SATA state before dealth.  The real core of the code
is really:
      -	Detect an issue
      -	Deal with the issue
I've found if the issues are ignored then bad things start to happen.
If the issues are acknowledge then the system is happy.  This really
has nothing to do with the core ATA code but other bits.

I think my patches could be better and tighter integrated with
some guidence.  I can offer some suggestions some might be usefull 
others might not or there could be better ways to do things.  Granted
I'm not going to invest the time merging if they always get rejected.

Doug A.