Re: ahci, SActive flag, and the HD activity LED

On 8/3/05, Jens Axboe [off-list ref] wrote:

On Wed, Aug 03 2005, Martin Wilck wrote:

quoted

Have you (or has anybody else) also seen the wrong behavior of the
activity LED?

No, but I have observed that SActive never gets cleared by the device
for non-NCQ commands (which is probably which gets you the stuck LED on
some systems?), which to me is another indication that we should not be
setting the tag bits for those commands.

The drives won't send a SetBits FIS when not using NCQ, as it can
really confuse some host adapters that don't understand NCQ.  I'd
imagine you're correct that the driver shouldn't be setting the bit in
the first place.

--eric