MIME-Version: 1.0
References: <20230203-get_kernel_pages-v2-0-f1dc4af273f1@intel.com>
In-Reply-To: <20230203-get_kernel_pages-v2-0-f1dc4af273f1@intel.com>
From:   Sumit Garg <sumit.garg@linaro.org>
Date:   Mon, 6 Feb 2023 11:52:43 +0530
Message-ID: <CAFA6WYM5tzs3T1tDgspA=R=VxfWrknm2YPHMCZihx405pF-ftA@mail.gmail.com>
Subject: Re: [PATCH v2 0/4] Remove get_kernel_pages()
To:     Ira Weiny <ira.weiny@intel.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Christoph Hellwig <hch@lst.de>, linux-kernel@vger.kernel.org,
        op-tee@lists.trustedfirmware.org, linux-mm@kvack.org,
        Jens Wiklander <jens.wiklander@linaro.org>,
        "Fabio M. De Francesco" <fmdefrancesco@gmail.com>,
        John Hubbard <jhubbard@nvidia.com>,
        Matthew Wilcox <willy@infradead.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Mel Gorman <mgorman@suse.de>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sat, 4 Feb 2023 at 09:36, Ira Weiny <ira.weiny@intel.com> wrote:
>
> Sumit,
>
> I did not see a follow up on this series per your last email.[1]  I'd like to
> move forward with getting rid of kmap_to_page().  So Hopefully this can land
> and you can build on this rather than the other way around?

Apologies Ira for keeping you waiting. Actually I was fully involved
with other high priority work with my upstream review backlog
increasing. So I wasn't able to devote time to this work. Sure I will
rebase my work on top of your changes.

-Sumit

>
> All,
>
> Al Viro found[2] that kmap_to_page() is broken.  But not only is it broken, it
> presents confusion over how highmem should be used because kmap() and friends
> should not be used for 'long term' mappings.
>
> get_kernel_pages() is a caller of kmap_to_page().  It only has one caller
> [shm_get_kernel_pages()] which does not need the functionality.
>
> Alter shm_get_kernel_pages() to no longer call get_kernel_pages() and remove
> get_kernel_pages().  Along the way it was noted that shm_get_kernel_pages()
> does not have any need to support vmalloc'ed addresses either.  Remove that
> functionality to clean up the logic.
>
> This series also fixes is_kmap_addr() and uses it to ensure no kmap addresses
> slip in later.
>
> [1] https://lore.kernel.org/all/CAFA6WYMqEVDVW-ifoh-V9ni1zntYdes8adQKf2XXAUpqdaW53w@mail.gmail.com/
> [2] https://lore.kernel.org/lkml/YzSSl1ItVlARDvG3@ZenIV
>
> To: Sumit Garg <sumit.garg@linaro.org>
> To: Andrew Morton <akpm@linux-foundation.org>
> Cc: "Al Viro" <viro@zeniv.linux.org.uk>
> Cc: "Christoph Hellwig" <hch@lst.de>
> Cc: linux-kernel@vger.kernel.org
> Cc: op-tee@lists.trustedfirmware.org
> Cc: linux-mm@kvack.org
> Cc: Jens Wiklander <jens.wiklander@linaro.org>
> Cc: "Fabio M. De Francesco" <fmdefrancesco@gmail.com>
> Signed-off-by: Ira Weiny <ira.weiny@intel.com>
>
> ---
> Changes in v2:
> - Al Viro: Avoid allocating the kiov.
> - Sumit: Update cover letter to clarify the motivation behind removing
>   get_kernel_pages()
> - Link to v1: https://lore.kernel.org/r/20221002002326.946620-1-ira.weiny@intel.com
>
> ---
> Ira Weiny (4):
>       highmem: Enhance is_kmap_addr() to check kmap_local_page() mappings
>       tee: Remove vmalloc page support
>       tee: Remove call to get_kernel_pages()
>       mm: Remove get_kernel_pages()
>
>  drivers/tee/tee_shm.c            | 37 ++++++++++---------------------------
>  include/linux/highmem-internal.h |  5 ++++-
>  include/linux/mm.h               |  2 --
>  mm/swap.c                        | 30 ------------------------------
>  4 files changed, 14 insertions(+), 60 deletions(-)
> ---
> base-commit: 0136d86b78522bbd5755f8194c97a987f0586ba5
> change-id: 20230203-get_kernel_pages-199342cfba79
>
> Best regards,
> --
> Ira Weiny <ira.weiny@intel.com>