>I'm not sure I like this "usermode" parameter that selects between
>memcpy() and copy_to_user().  For one thing, it means that it is
>impossible to fix the sparse warnings without defeating the sparse
>address space testing: you have to add a casts to and from __user
>pointers all over the place.
>
>Also, it seems that there's only one caller that's ever going to use
>the usermode parameter, namely user_mad.c.

This seems true.  I didn't test the usermode piece, so didn't think about
the casting issues.

>Unfortunately I don't see a good way to avoid duplicating the (simple)
>logic in ib_coalesce_recv_mad().

I worked on trying to keep this as simple as possible.  (It started out
quite a bit more complex...)  Looking at it, it's probably not a big deal to
duplicate the logic.  I'll remove the usermode parameter.

- Sean

_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to