Jeff Garzik wrote:

Very strange.  I cannot find this patch at all in my email folders.

Can someone resend it to me?

The original is at
http://seclists.org/lists/linux-kernel/2005/May/0447.html, as indicated
by the footnote in my original message.

That said, it's identical in content to the one I've just sent (attached
as a text/plain MIME attachment.)

Worried now!

If this patch is needed, something VERY VERY WRONG is going on.  This
patch indicates that the queueing state machine has been violated, and
something is trying to IGNORE the command synchronization :(

Further, you cannot always assume that msleep() is valid in that
context.  It should be the caller that waits (libata suspend code), not
ata_do_simple_cmd() itself.

Does anyone have a link to James Bottomley's proposed patch?  That one
seemed to do what was necessary -- send a SYNCHRONIZE_CACHE command then
turn it over to the LLD for further suspend.

I don't think James sent any patch in the original thread with Jens'
patch. But that sounds somewhat cleaner than msleep().

At any rate, the current patch (with or without Jens' new addition)
works for me on Thinkpad T43. So the net result is better anyway, IMO.

quoted

Ok. Can we have this in -mm for a few days just to shake out anything
interesting, and then merge it into mainline?

Once we get a decent patch, I can merge it into my libata-dev.git
repository, which is automatically propagated to -mm.

Is the kludginess of Jens' addition the only problem you have with the
current patch? (If it is, then fixing this is out of the range of my
current ability...)

Thanks for responding.

-- 
Joshua Kwan