linux-kernel - Re: [PATCH net-next v9 10/13] mm: page_frag: introduce prepare/probe/commit API

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAKgT0UdH1yD=LSCXFJ=YM_aiA4OomD-2wXykO42bizaWMt_HOA@mail.gmail.com>
Date: Mon, 8 Jul 2024 07:30:30 -0700
From: Alexander Duyck <alexander.duyck@...il.com>
To: Yunsheng Lin <linyunsheng@...wei.com>
Cc: Yunsheng Lin <yunshenglin0825@...il.com>, davem@...emloft.net, kuba@...nel.org, 
	pabeni@...hat.com, netdev@...r.kernel.org, linux-kernel@...r.kernel.org, 
	Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org
Subject: Re: [PATCH net-next v9 10/13] mm: page_frag: introduce
 prepare/probe/commit API

On Mon, Jul 8, 2024 at 3:58 AM Yunsheng Lin <linyunsheng@...wei.com> wrote:
>
> On 2024/7/8 1:12, Alexander Duyck wrote:
>
> ...
>
> > The issue is the dependency mess that has been created with patch 11
> > in the set. Again you are conflating patches which makes this really
> > hard to debug or discuss as I make suggestions on one patch and you
> > claim it breaks things that are really due to issues in another patch.
> > So the issue is you included this header into include/linux/sched.h
> > which is included in linux/mm_types.h. So what happens then is that
> > you have to include page_frag_cache.h *before* you can include the
> > bits from mm_types.h
> >
> > What might make more sense to solve this is to look at just moving the
> > page_frag_cache into mm_types_task.h and then having it replace the
> > page_frag struct there since mm_types.h will pull that in anyway. That
> > way sched.h can avoid having to pull in page_frag_cache.h.
>
> It seems the above didn't work either, as asm-offsets.c does depend on
> mm_types_task.h too.
>
> In file included from ./include/linux/mm.h:16,
>                  from ./include/linux/page_frag_cache.h:10,
>                  from ./include/linux/mm_types_task.h:11,
>                  from ./include/linux/mm_types.h:5,
>                  from ./include/linux/mmzone.h:22,
>                  from ./include/linux/gfp.h:7,
>                  from ./include/linux/slab.h:16,
>                  from ./include/linux/resource_ext.h:11,
>                  from ./include/linux/acpi.h:13,
>                  from ./include/acpi/apei.h:9,
>                  from ./include/acpi/ghes.h:5,
>                  from ./include/linux/arm_sdei.h:8,
>                  from arch/arm64/kernel/asm-offsets.c:10:
> ./include/linux/mmap_lock.h: In function ‘mmap_assert_locked’:
> ./include/linux/mmap_lock.h:65:23: error: invalid use of undefined type ‘const struct mm_struct’
>    65 |  rwsem_assert_held(&mm->mmap_lock);

Do not include page_frag_cache.h in mm_types_task.h. Just move the
struct page_frag_cache there to replace struct page_frag.