On Thu, May 12, 2016 at 10:39:00AM +0200, Oswald Buddenhagen wrote: > On Wed, May 11, 2016 at 09:25:41PM -0400, Damien Riegel wrote: > > # HG changeset patch > > # User Damien Riegel <[email protected]> > > # Date 1462391862 14400 > > # Wed May 04 15:57:42 2016 -0400 > > # Node ID eb3c1cd64366352abc9f182d7269ac24298533d0 > > # Parent d18cd04e3f5a07ea6c3d0c0b624c917a0024e037 > > create a dedicated structure for mx operations > > > > This commit introduces a dedicated structure for mailbox operations. The > > point is to avoid to clobber the context structure with additional > > > s/clobber/clutter/ > > > callbacks, and to allow each kind of mailboxes to define its own > > structure. > > > you may also want to mention that this structure can be populated > statically. > > regarding the other patches: > the series is over-fragmented. 6, 7, and 8 definitely should be > squashed, as the first two patches make no sense whatsoever without the > third.
I am used to kernel development, so I like to go from one state to another through small and atomic changes. To me, it doesn't matter if a feature takes multiple commits to get implemented as long as the code is in a working state at every step. Keeping changes as small as possible makes blame and bisect way more friendly. Anyway, this is just what I am used to do. If Kevin wants larger commits or to squash them when he picks them, that's fine by me. > the refactorings of the various drivers make sense separately, but > ideally you'd pull them ahead of 1. > the mx_ops assigning parts of 8 arguably belong into 1, but i didn't > look close enough whether that's feasible. It is a remainer of the first version, which used probe callbacks to find the right mx_ops to use. Now that it is out, it is probably possible to do as you're suggesting with a bit of reordering and some work. Kevin: just let me know if you want me to do that, or if it is fine as is. -- Damien
