MIME-Version: 1.0
References: <20230130130739.563628-1-arnd@kernel.org> <Y9fITnjnIuDz8NYw@dhcp22.suse.cz>
 <CAG_fn=UC-zPKfXvhnXO-Hb7Wp3+BJyT4WvotATigigb63N0tdA@mail.gmail.com> <Y9kwTXhAQiEWz0IJ@dhcp22.suse.cz>
In-Reply-To: <Y9kwTXhAQiEWz0IJ@dhcp22.suse.cz>
From:   Alexander Potapenko <glider@google.com>
Date:   Tue, 31 Jan 2023 16:17:18 +0100
Message-ID: <CAG_fn=UiLM+5uKR+SP_=XVfqxKATi26uByZnV5b=3KNSiV3eHw@mail.gmail.com>
Subject: Re: [PATCH] mm: extend max struct page size for kmsan
To:     Michal Hocko <mhocko@suse.com>
Cc:     Arnd Bergmann <arnd@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Alexander Duyck <alexander.h.duyck@linux.intel.com>,
        Pavel Tatashin <pavel.tatashin@microsoft.com>,
        Arnd Bergmann <arnd@arndb.de>,
        "Matthew Wilcox (Oracle)" <willy@infradead.org>,
        David Hildenbrand <david@redhat.com>,
        "Liam R. Howlett" <Liam.Howlett@oracle.com>,
        John Hubbard <jhubbard@nvidia.com>,
        Naoya Horiguchi <naoya.horiguchi@nec.com>,
        Hugh Dickins <hughd@google.com>,
        Suren Baghdasaryan <surenb@google.com>,
        Alex Sierra <alex.sierra@amd.com>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Jan 31, 2023 at 4:14 PM Michal Hocko <mhocko@suse.com> wrote:
>
> On Mon 30-01-23 18:59:45, Alexander Potapenko wrote:
> > On Mon, Jan 30, 2023 at 2:38 PM Michal Hocko <mhocko@suse.com> wrote:
> > >
> > > On Mon 30-01-23 14:07:26, Arnd Bergmann wrote:
> > > > From: Arnd Bergmann <arnd@arndb.de>
> > > >
> > > > After x86 has enabled support for KMSAN, it has become possible
> > > > to have larger 'struct page' than was expected when commit
> > > > 5470dea49f53 ("mm: use mm_zero_struct_page from SPARC on all 64b
> > > > architectures") was merged:
> > > >
> > > > include/linux/mm.h:156:10: warning: no case matching constant switch condition '96'
> > > >         switch (sizeof(struct page)) {
> > > >
> > > > Extend the maximum accordingly.
> > > >
> > > > Fixes: 5470dea49f53 ("mm: use mm_zero_struct_page from SPARC on all 64b architectures")
> > > > Fixes: 4ca8cc8d1bbe ("x86: kmsan: enable KMSAN builds for x86")
> > > > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > >
> > > Acked-by: Michal Hocko <mhocko@suse.com>
> > >
> > > I haven't really followed KMSAN development but I would have expected
> > > that it would, like other debugging tools, add its metadata to page_ext
> > > rather than page directly.
> >
> > Thanks for the comment!
> > I was considering page_ext at some point, but managed to convince
> > myself it didn't suit the purpose well enough.
> >
> > Right now KMSAN allocates its metadata at boot time, when tearing down memblock.
> > At that point only a handful of memory ranges exist, and it is pretty
> > easy to carve out some unused pages for the metadata for those ranges,
> > then divide the rest evenly and return 1/3 to the system, spending 2/3
> > to keep the metadata for the returned pages.
> > I tried allocating the memory lazily (at page_alloc(), for example),
> > and it turned out to be very tricky because of fragmentation: for an
> > allocation of a given order, one needs shadow and origin allocations
> > of the same order [1], and alloc_pages() simply started with ripping
> > apart the biggest chunk of memory available.
>
> page_ext allocation happens quite early as well. There shouldn't be any
> real fragmentation that early during the boot.
>
> > IIRC if we choose to allocate metadata via page_ext, the memory will
> > be already too fragmented to easily handle it, because it will only
> > happen once alloc_pages() is available.
> > We also can't get rid of the shadow/origin pointers in struct page_ext
> > (storing two 4K-sized arrays in that struct would defeat all the
> > possible alignments), so we won't save any memory by switching to
> > page_ext.
>
> With page_ext you would allow to compile the feature in disabled by
> default and allow to boot time enable it.

This makes little sense to do, because KMSAN requires heavy
compile-time instrumentation to work. One cannot simply enable/disable
it at boot time anyway.