On Tue, Sep 21, 2021 at 3:04 PM Andrew Lunn [off-list ref] wrote:

On Tue, Sep 21, 2021 at 02:54:47PM -0700, Saravana Kannan wrote:

quoted

On Tue, Sep 21, 2021 at 2:03 PM Andrew Lunn [off-list ref] wrote:

quoted

quoted

There are cases where the children try to probe too quickly (before
the parent has had time to set up all the resources it's setting up)
and the child defers the probe. Even Andrew had an example of that
with some ethernet driver where the deferred probe is attempted
multiple times wasting time and then it eventually succeeds.

And i prefer an occasional EPROBE_DEFER over a broken Ethernet switch,
which is the current state. I'm happy to see optimisations, but not at
the expense of breaking working stuff.

Right, but in that case, the long term solution should be to make
changes so we don't expect the child to be bound as soon as it's
added. Not disable the optimization. Agree?

Maybe. Lets see how you fix what is currently broken. At the moment, i
don't care too much about the long term solution. The current quick
fix for stable does not seem to be making any progress. So we need the
real fix now, to unbreak what is currently broken, then we can think
about the long term.

Wait, what's the difference between a real fix vs a long term fix? To
me those are the same.

I agree that having DSA be broken till we have the final fix isn't
good. Especially because DSA is complicated and the over eager gen PHY
driver makes it even harder to fix it quickly.

Merging this patch gives an exception to DSA, while we figure out a
good fix. It also makes sure more drivers don't get merged with the
same assumptions (because fw_devlink=on won't give them the
exception).

Greg/Rafael, can we merge this please while we figure out a fix for
DSA/PHY. It's a non-trivial problem to solve because PHYs need some
kind of driver "match rating".

-Saravana