From mboxrd@z Thu Jan  1 00:00:00 1970
From: Matthew Wilcox <willy@infradead.org>
Subject: Re: drm pull for v5.3-rc1
Date: Tue, 6 Aug 2019 12:09:38 -0700
Message-ID: <20190806190937.GD30179@bombadil.infradead.org>
References: <CAPM=9tzJQ+26n_Df1eBPG1A=tXf4xNuVEjbG3aZj-aqYQ9nnAg@mail.gmail.com>
 <CAPM=9twvwhm318btWy_WkQxOcpRCzjpok52R8zPQxQrnQ8QzwQ@mail.gmail.com>
 <CAHk-=wjC3VX5hSeGRA1SCLjT+hewPbbG4vSJPFK7iy26z4QAyw@mail.gmail.com>
 <CAHk-=wiD6a189CXj-ugRzCxA9r1+siSCA0eP_eoZ_bk_bLTRMw@mail.gmail.com>
 <48890b55-afc5-ced8-5913-5a755ce6c1ab@shipmail.org>
 <CAHk-=whwcMLwcQZTmWgCnSn=LHpQG+EBbWevJEj5YTKMiE_-oQ@mail.gmail.com>
 <CAHk-=wghASUU7QmoibQK7XS09na7rDRrjSrWPwkGz=qLnGp_Xw@mail.gmail.com>
 <20190806073831.GA26668@infradead.org>
 <CAHk-=wi7L0MDG7DY39Hx6v8jUMSq3ZCE3QTnKKirba_8KAFNyw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Return-path: <linux-kernel-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <CAHk-=wi7L0MDG7DY39Hx6v8jUMSq3ZCE3QTnKKirba_8KAFNyw@mail.gmail.com>
Sender: linux-kernel-owner@vger.kernel.org
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Christoph Hellwig <hch@infradead.org>, Thomas =?iso-8859-1?Q?Hellstr=F6m_=28VMware=29?= <thomas@shipmail.org>, Dave Airlie <airlied@gmail.com>, Thomas Hellstrom <thellstrom@vmware.com>, Daniel Vetter <daniel.vetter@ffwll.ch>, LKML <linux-kernel@vger.kernel.org>, dri-devel <dri-devel@lists.freedesktop.org>, Jerome Glisse <jglisse@redhat.com>, Jason Gunthorpe <jgg@mellanox.com>, Andrew Morton <akpm@linux-foundation.org>, Steven Price <steven.price@arm.com>, Linux-MM <linux-mm@kvack.org>
List-Id: dri-devel@lists.freedesktop.org

On Tue, Aug 06, 2019 at 11:50:42AM -0700, Linus Torvalds wrote:
> In fact, I do note that a lot of the users don't actually use the
> "void *private" argument at all - they just want the walker - and just
> pass in a NULL private pointer. So we have things like this:
> 
> > +       if (walk_page_range(&init_mm, va, va + size, &set_nocache_walk_ops,
> > +                       NULL)) {
> 
> and in a perfect world we'd have arguments with default values so that
> we could skip those entirely for when people just don't need it.
> 
> I'm not a huge fan of C++ because of a lot of the complexity (and some
> really bad decisions), but many of the _syntactic_ things in C++ would
> be nice to use. This one doesn't seem to be one that the gcc people
> have picked up as an extension ;(
> 
> Yes, yes, we could do it with a macro, I guess.
> 
>    #define walk_page_range(mm, start,end, ops, ...) \
>        __walk_page_range(mm, start, end, (NULL , ## __VA_ARGS__))
> 
> but I'm not sure it's worthwhile.

Has anyone looked at turning the interface inside-out?  ie something like:

	struct mm_walk_state state = { .mm = mm, .start = start, .end = end, };

	for_each_page_range(&state, page) {
		... do something with page ...
	}

with appropriate macrology along the lines of:

#define for_each_page_range(state, page)				\
	while ((page = page_range_walk_next(state)))

Then you don't need to package anything up into structs that are shared
between the caller and the iterated function.