LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Dan Williams <dan.j.williams@intel.com>
To: Dave Airlie <airlied@gmail.com>
Cc: Jerome Glisse <jglisse@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Logan Gunthorpe <logang@deltatee.com>,
	Christoph Hellwig <hch@lst.de>, Michal Hocko <mhocko@suse.com>,
	Linux MM <linux-mm@kvack.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 0/5] mm: rework hmm to use devm_memremap_pages
Date: Tue, 29 May 2018 16:33:49 -0700	[thread overview]
Message-ID: <CAPcyv4jTty4k1xXCOWbeRjzv-KjxNH1L4oOkWW1EbJt66jF4_w@mail.gmail.com> (raw)
In-Reply-To: <CAPM=9twm=17t=2=M27ELB=vZWzpqM7GuwCUsC891jJ0t3JM4vg@mail.gmail.com>

On Tue, May 29, 2018 at 4:00 PM, Dave Airlie <airlied@gmail.com> wrote:
> On 30 May 2018 at 08:31, Dan Williams <dan.j.williams@intel.com> wrote:
>> On Tue, May 29, 2018 at 3:22 PM, Dave Airlie <airlied@gmail.com> wrote:
>>>
>>> On 24 May 2018 at 13:18, Dan Williams <dan.j.williams@intel.com> wrote:
>>> > On Wed, May 23, 2018 at 5:10 PM, Jerome Glisse <jglisse@redhat.com> wrote:
>>> >> On Mon, May 21, 2018 at 03:35:14PM -0700, Dan Williams wrote:
>>> >>> Hi Andrew, please consider this series for 4.18.
>>> >>>
>>> >>> For maintainability, as ZONE_DEVICE continues to attract new users,
>>> >>> it is useful to keep all users consolidated on devm_memremap_pages() as
>>> >>> the interface for create "device pages".
>>> >>>
>>> >>> The devm_memremap_pages() implementation was recently reworked to make
>>> >>> it more generic for arbitrary users, like the proposed peer-to-peer
>>> >>> PCI-E enabling. HMM pre-dated this rework and opted to duplicate
>>> >>> devm_memremap_pages() as hmm_devmem_pages_create().
>>> >>>
>>> >>> Rework HMM to be a consumer of devm_memremap_pages() directly and fix up
>>> >>> the licensing on the exports given the deep dependencies on the mm.
>>> >>
>>> >> I am on PTO right now so i won't be able to quickly review it all
>>> >> but forcing GPL export is problematic for me now. I rather have
>>> >> device driver using "sane" common helpers than creating their own
>>> >> crazy thing.
>>> >
>>> > Sane drivers that need this level of deep integration with Linux
>>> > memory management need to be upstream. Otherwise, HMM is an
>>> > unprecedented departure from the norms of Linux kernel development.
>>>
>>> Isn't it the author of code choice what EXPORT_SYMBOL to use? and
>>> isn't the agreement that if something is EXPORT_SYMBOL now, changing
>>> underlying exports isn't considered a good idea. We've seen this before
>>> with the refcount fun,
>>>
>>> See d557d1b58b3546bab2c5bc2d624c5709840e6b10
>>>
>>> Not commenting on the legality or what derived works are considered,
>>> since really the markings are just an indication of the authors opinion,
>>> and at this stage I think are actually meaningless, since we've diverged
>>> considerably from the advice given to Linus back when this started.
>>
>> Yes, and in this case devm_memremap_pages() was originally written by
>> Christoph and I:
>>
>>     41e94a851304 add devm_memremap_pages
>
> So you wrote some code in 2015 (3 years ago) and you've now decided
> to change the EXPORT marker on it? what changed in 3 years, and why
> would changing that marker 3 years later have any effect on your original
> statement that it was an EXPORT_SYMBOL.
>
> Think what EXPORT_SYMBOL vs GPL means, it isn't a bit stick that magically
> makes things into derived works. If something wasn't a derived work for 3 years
> using that API, then it isn't a derived work now 3 years later because you
> changed the marker. Retrospectively changing the markers doesn't really
> make any sense legally or otherwise.

It honestly was an oversight, and as we've gone on to add deeper and
deeper ties into the mm and filesystems [1] I realized this symbol was
mis-labeled.  It would be one thing if this was just some random
kernel leaf / library function, but this capability when turned on
causes the entire kernel to be recompiled as things like the
definition of put_page() changes. It's deeply integrated with how
Linux manages memory.

>> HMM started off by duplicating devm_memremap_pages() which is fixed up
>> by this series:
>
> Just looking in my current tree hmm_devmem_pages_create and
> devm_memremap_pages don't look like duplicates, they might have
> code but they definitely aren't one for one copies. I'm not sure you can
> just say Jerome copied that code in, you've now refactored the code
> so HMM can use it and are changing the symbol exports underneath it,

The initial patches for HMM used devm_memremap_pages() directly, and
during review I asked for the exact same arrangement as implemented
here, i.e. for the dev_pagemap structure to be a sub-structure of the
HMM data [2]. At some point along the way we lost that review
feedback. It was not until Christoph and Logan recently reworked
devm_memermap_pages() that I realized that HMM had unnecessarily
diverged.

> Again if Christoph believes all uses of this are a derived work he didn't
> indicate it 3 years ago, but neither does the mark make any legal difference
> in this case, since everything in the kernel is GPL, and if you
> consider something
> a derived work or not is well into legal land.
>
> I'd rather anyways the original author of HMM wishes were respected
> on his code, or at least you wait until he gets back from holidays before
> pushing to merge this.

To be clear this only affects the usage of the ZONE_DEVICE facility,
I'm not touching the other pieces of HMM that are original to HMM. I
didn't realize Jerome was on vacation when I sent the patches, and I
think it is otherwise healthy to have this discussion in the meantime.

[1]: https://lists.01.org/pipermail/linux-nvdimm/2018-May/015905.html
[2]: http://lkml.iu.edu/hypermail/linux/kernel/1701.2/00812.html

  reply	other threads:[~2018-05-29 23:33 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-05-21 22:35 Dan Williams
2018-05-21 22:35 ` [PATCH 1/5] mm, devm_memremap_pages: mark devm_memremap_pages() EXPORT_SYMBOL_GPL Dan Williams
2018-05-22  6:29   ` Christoph Hellwig
2018-05-21 22:35 ` [PATCH 2/5] mm, devm_memremap_pages: handle errors allocating final devres action Dan Williams
2018-05-21 23:10   ` Andrew Morton
2018-05-22  0:07     ` Dan Williams
2018-05-22 16:42       ` Logan Gunthorpe
2018-05-22 16:56         ` Dan Williams
2018-05-22 17:03           ` Logan Gunthorpe
2018-05-22 17:25             ` Dan Williams
2018-05-22 17:36               ` Logan Gunthorpe
2018-05-22  6:30   ` Christoph Hellwig
2018-05-21 22:35 ` [PATCH 3/5] mm, hmm: use devm semantics for hmm_devmem_{add, remove} Dan Williams
2018-05-22  6:30   ` Christoph Hellwig
2018-05-21 22:35 ` [PATCH 4/5] mm, hmm: replace hmm_devmem_pages_create() with devm_memremap_pages() Dan Williams
2018-05-22  6:31   ` Christoph Hellwig
2018-05-22 17:13   ` Logan Gunthorpe
2018-05-22 21:38     ` Dan Williams
2018-05-21 22:35 ` [PATCH 5/5] mm, hmm: mark hmm_devmem_{add, add_resource} EXPORT_SYMBOL_GPL Dan Williams
2018-05-22  6:32   ` Christoph Hellwig
2018-05-22 21:31     ` Andrew Morton
2018-06-05 18:24       ` Jerome Glisse
2018-05-24  0:10 ` [PATCH 0/5] mm: rework hmm to use devm_memremap_pages Jerome Glisse
2018-05-24  3:18   ` Dan Williams
2018-05-24  6:35     ` Christoph Hellwig
2018-05-29 22:22     ` Dave Airlie
2018-05-29 22:31       ` Dan Williams
2018-05-29 23:00         ` Dave Airlie
2018-05-29 23:33           ` Dan Williams [this message]
2018-06-05 18:48             ` Jerome Glisse
2018-06-05 22:19               ` Dave Airlie
2018-06-05 23:06                 ` Dan Williams
2018-06-06  0:08                   ` Jerome Glisse
2018-06-06  1:33                     ` Dan Williams
2018-06-06  7:14                       ` Christoph Hellwig
2018-06-07 14:16                       ` Jerome Glisse
2018-06-07 18:39                         ` Dan Williams

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAPcyv4jTty4k1xXCOWbeRjzv-KjxNH1L4oOkWW1EbJt66jF4_w@mail.gmail.com \
    --to=dan.j.williams@intel.com \
    --cc=airlied@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=hch@lst.de \
    --cc=jglisse@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=logang@deltatee.com \
    --cc=mhocko@suse.com \
    --subject='Re: [PATCH 0/5] mm: rework hmm to use devm_memremap_pages' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).