cvs commit: src/sys/dev/ata atapi-cd.c atapi-cd.h

[Scott Long]

Poul-Henning Kamp wrote:
> In message <473E238D.4030800 at samsco.org>, Scott Long writes:
>> Poul-Henning Kamp wrote:
> 
>>>> Well, there are two questions:  media present (yes/no) and drive capable
>>>> of telling if media present without just trying to read it (yes/no).
>>> This risk reopening the old religious war if drivers should poll
>>> their drives to be up to date on media existence.
>> This has nothing to do with polling, you're just trotting out an 
>> irrelevant argument to divert attention.
> 
> No, I merely pointed out that Nates worries had an obvious and widely
> accepted solution, which unfortunately is religiously unacceptable
> in certain parts of FreeBSD storage subsystems.
> 
> For all I know, the problem sos@ has complained about has a solution
> that just needs somebody to change the relevant few lines of code.
> 

Ok, back on track.  What you want to do is make g_access largely 
irrelevant, and have the _real_ access check done in g_start.  Fine.
Is there any reason at all to make g_access not return success from
now on?  And if so, is there any reason to even keep it around?  This
is what I'm arguing against; you're advocating a hack that works well
enough for you and ignoring the larger architectural issue, which I
think is important.  But anyways.

Scott