From: Matthew Wilcox <willy@infradead.org>
To: Christopher Lameter <cl@linux.com>
Cc: Vlastimil Babka <vbabka@suse.cz>,
Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Pekka Enberg <penberg@kernel.org>,
David Rientjes <rientjes@google.com>,
Ming Lei <ming.lei@redhat.com>,
Dave Chinner <david@fromorbit.com>,
"Darrick J . Wong" <darrick.wong@oracle.com>,
Christoph Hellwig <hch@lst.de>,
linux-xfs@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-block@vger.kernel.org,
James Bottomley <James.Bottomley@hansenpartnership.com>,
linux-btrfs@vger.kernel.org
Subject: Re: [PATCH v2 2/2] mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two)
Date: Wed, 28 Aug 2019 12:46:08 -0700 [thread overview]
Message-ID: <20190828194607.GB6590@bombadil.infradead.org> (raw)
In-Reply-To: <0100016cd98bb2c1-a2af7539-706f-47ba-a68e-5f6a91f2f495-000000@email.amazonses.com>
On Wed, Aug 28, 2019 at 06:45:07PM +0000, Christopher Lameter wrote:
> > Ideally we should provide to mm users what they need without difficult
> > workarounds or own reimplementations, so let's make the kmalloc() alignment to
> > size explicitly guaranteed for power-of-two sizes under all configurations.
>
> The objection remains that this will create exceptions for the general
> notion that all kmalloc caches are aligned to KMALLOC_MINALIGN which may
Hmm? kmalloc caches will be aligned to both KMALLOC_MINALIGN and the
natural alignment of the object.
> be suprising and it limits the optimizations that slab allocators may use
> for optimizing data use. The SLOB allocator was designed in such a way
> that data wastage is limited. The changes here sabotage that goal and show
> that future slab allocators may be similarly constrained with the
> exceptional alignents implemented. Additional debugging features etc etc
> must all support the exceptional alignment requirements.
While I sympathise with the poor programmer who has to write the
fourth implementation of the sl*b interface, it's more for the pain of
picking a new letter than the pain of needing to honour the alignment
of allocations.
There are many places in the kernel which assume alignment. They break
when it's not supplied. I believe we have a better overall system if
the MM developers provide stronger guarantees than the MM consumers have
to work around only weak guarantees.
> > * SLOB has no implicit alignment so this patch adds it explicitly for
> > kmalloc(). The potential downside is increased fragmentation. While
> > pathological allocation scenarios are certainly possible, in my testing,
> > after booting a x86_64 kernel+userspace with virtme, around 16MB memory
> > was consumed by slab pages both before and after the patch, with difference
> > in the noise.
>
> This change to slob will cause a significant additional use of memory. The
> advertised advantage of SLOB is that *minimal* memory will be used since
> it is targeted for embedded systems. Different types of slab objects of
> varying sizes can be allocated in the same memory page to reduce
> allocation overhead.
Did you not read the part where he said the difference was in the noise?
> The result of this patch is just to use more memory to be safe from
> certain pathologies where one subsystem was relying on an alignment that
> was not specified. That is why this approach should not be called
> �natural" but "implicit alignment". The one using the slab cache is not
> aware that the slab allocator provides objects aligned in a special way
> (which is in general not needed. There seems to be a single pathological
> case that needs to be addressed and I thought that was due to some
> brokenness in the hardware?).
It turns out there are lots of places which assume this, including the
pmem driver, the ramdisk driver and a few other similar drivers.
> It is better to ensure that subsystems that require special alignment
> explicitly tell the allocator about this.
But it's not the subsystems which have this limitation which do the
allocation; it's the subsystems who allocate the memory that they then
pass to the subsystems. So you're forcing communication of these limits
up & down the stack.
> I still think implicit exceptions to alignments are a bad idea. Those need
> to be explicity specified and that is possible using kmem_cache_create().
I swear we covered this last time the topic came up, but XFS would need
to create special slab caches for each size between 512 and PAGE_SIZE.
Potentially larger, depending on whether the MM developers are willing to
guarantee that kmalloc(PAGE_SIZE * 2, GFP_KERNEL) will return a PAGE_SIZE
aligned block of memory indefinitely.
next prev parent reply other threads:[~2019-08-28 19:46 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-26 11:16 [PATCH v2 0/2] guarantee natural alignment for kmalloc() Vlastimil Babka
2019-08-26 11:16 ` [PATCH v2 1/2] mm, sl[ou]b: improve memory accounting Vlastimil Babka
2019-08-26 11:16 ` [PATCH v2 2/2] mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two) Vlastimil Babka
2019-08-28 18:45 ` Christopher Lameter
2019-08-28 19:46 ` Matthew Wilcox [this message]
2019-08-28 22:24 ` Dave Chinner
2019-08-29 7:56 ` Vlastimil Babka
2019-08-30 0:29 ` Dave Chinner
2019-08-29 7:39 ` Michal Hocko
2019-08-30 17:41 ` Christopher Lameter
2019-09-01 0:52 ` Matthew Wilcox
2019-09-03 20:13 ` Christopher Lameter
2019-09-03 20:53 ` Matthew Wilcox
2019-09-04 5:19 ` Christoph Hellwig
2019-09-04 6:40 ` Ming Lei
2019-09-04 7:20 ` Vlastimil Babka
2019-09-04 19:31 ` Christopher Lameter
2019-09-23 16:36 ` Vlastimil Babka
2019-09-23 17:17 ` David Sterba
2019-09-23 17:51 ` Darrick J. Wong
2019-09-24 20:47 ` cl
2019-09-24 20:51 ` Matthew Wilcox
2019-09-24 20:55 ` cl
2019-09-26 13:02 ` David Sterba
2019-09-24 21:19 ` Vlastimil Babka
2019-09-24 21:53 ` Dave Chinner
2019-09-24 22:21 ` Darrick J. Wong
2019-09-24 20:52 ` cl
2019-09-24 23:54 ` Andrew Morton
2019-09-25 7:17 ` Vlastimil Babka
2019-09-26 0:16 ` Christopher Lameter
2019-09-26 0:14 ` Christopher Lameter
2019-09-26 7:41 ` Vlastimil Babka
2019-09-28 1:12 ` Christopher Lameter
2019-09-30 13:32 ` Matthew Wilcox
2019-09-23 17:54 ` Roman Gushchin
2019-09-30 8:06 ` Christoph Hellwig
2019-09-30 9:23 ` Michal Hocko
2019-09-30 9:32 ` Kirill A. Shutemov
2019-09-23 18:57 ` Matthew Wilcox
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=20190828194607.GB6590@bombadil.infradead.org \
--to=willy@infradead.org \
--cc=James.Bottomley@hansenpartnership.com \
--cc=akpm@linux-foundation.org \
--cc=cl@linux.com \
--cc=darrick.wong@oracle.com \
--cc=david@fromorbit.com \
--cc=hch@lst.de \
--cc=linux-block@vger.kernel.org \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-xfs@vger.kernel.org \
--cc=ming.lei@redhat.com \
--cc=penberg@kernel.org \
--cc=rientjes@google.com \
--cc=vbabka@suse.cz \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).