RE: [PATCH v4 3/7] drivers:hv: Use new vmbus_mmio_free() from client drivers.

[Jake Oshins]

> -----Original Message-----
> From: Bjorn Helgaas [mailto:helgaas@xxxxxxxxxx]
> Sent: Tuesday, April 5, 2016 11:00 AM
> To: Jake Oshins <jakeo@xxxxxxxxxxxxx>
> Cc: linux-pci@xxxxxxxxxxxxxxx; gregkh@xxxxxxxxxxxxxxxxxxx; KY Srinivasan
> <kys@xxxxxxxxxxxxx>; linux-kernel@xxxxxxxxxxxxxxx;
> devel@xxxxxxxxxxxxxxxxxxxxxx; olaf@xxxxxxxxx; apw@xxxxxxxxxxxxx;
> vkuznets@xxxxxxxxxx; Haiyang Zhang <haiyangz@xxxxxxxxxxxxx>; Hadden
> Hoppert <haddenh@xxxxxxxxxxxxx>
> Subject: Re: [PATCH v4 3/7] drivers:hv: Use new vmbus_mmio_free() from
> client drivers.
> 
> Hi Jake,
> 
> On Fri, Apr 01, 2016 at 05:47:43PM -0700, Jake Oshins wrote:
> > This patch modifies all the callers of vmbus_mmio_allocate()
> > to call vmbus_mmio_free() instead of release_mem_region().
> 
> This changelog merely restates the C code.  Presumably there's some
> important difference between release_mem_region() and
> vmbus_mmio_free(), and we need a hint about what that is.
> 
> Oh, I see, there actually is no difference *yet*, but it's coming.
> I'd combine this with patch 2.  Then the patch is obviously correct
> all by itself, and the changelog for patch 2 makes clear what's
> happening.
> 
> In changelogs, don't bother with "this patch does" or "this function
> is introduced."  The context is obvious because the changelog is part
> of the commit.  Write imperative sentences, e.g., "Call
> vmbus_mmio_free() instead of release_mem_region()."
> 

Will do.  Thanks.

-- Jake Oshins