From: Dan Williams <dan.j.williams@intel.com>
To: Jan Kara <jack@suse.cz>
Cc: linux-nvdimm <linux-nvdimm@lists.01.org>,
stable <stable@vger.kernel.org>, Jeff Moyer <jmoyer@redhat.com>,
Ingo Molnar <mingo@redhat.com>, Christoph Hellwig <hch@lst.de>,
Al Viro <viro@zeniv.linux.org.uk>,
Thomas Gleixner <tglx@linutronix.de>,
Matthew Wilcox <willy@infradead.org>,
Jeff Smits <jeff.smits@intel.com>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Kees Cook <keescook@chromium.org>
Subject: Re: [PATCH] libnvdimm/pmem: Bypass CONFIG_HARDENED_USERCOPY overhead
Date: Fri, 17 May 2019 08:08:27 -0700 [thread overview]
Message-ID: <CAPcyv4iZZCgcC657ZOysBP9=1ejp3jfFj=VETVBPrgmfg7xUEw@mail.gmail.com> (raw)
In-Reply-To: <20190517084739.GB20550@quack2.suse.cz>
On Fri, May 17, 2019 at 1:47 AM Jan Kara <jack@suse.cz> wrote:
>
> Let's add Kees to CC for usercopy expertise...
>
> On Thu 16-05-19 17:33:38, Dan Williams wrote:
> > Jeff discovered that performance improves from ~375K iops to ~519K iops
> > on a simple psync-write fio workload when moving the location of 'struct
> > page' from the default PMEM location to DRAM. This result is surprising
> > because the expectation is that 'struct page' for dax is only needed for
> > third party references to dax mappings. For example, a dax-mapped buffer
> > passed to another system call for direct-I/O requires 'struct page' for
> > sending the request down the driver stack and pinning the page. There is
> > no usage of 'struct page' for first party access to a file via
> > read(2)/write(2) and friends.
> >
> > However, this "no page needed" expectation is violated by
> > CONFIG_HARDENED_USERCOPY and the check_copy_size() performed in
> > copy_from_iter_full_nocache() and copy_to_iter_mcsafe(). The
> > check_heap_object() helper routine assumes the buffer is backed by a
> > page-allocator DRAM page and applies some checks. Those checks are
> > invalid, dax pages are not from the heap, and redundant,
> > dax_iomap_actor() has already validated that the I/O is within bounds.
>
> So this last paragraph is not obvious to me as check_copy_size() does a lot
> of various checks in CONFIG_HARDENED_USERCOPY case. I agree that some of
> those checks don't make sense for PMEM pages but I'd rather handle that by
> refining check_copy_size() and check_object_size() functions to detect and
> appropriately handle pmem pages rather that generally skip all the checks
> in pmem_copy_from/to_iter(). And yes, every check in such hot path is going
> to cost performance but that's what user asked for with
> CONFIG_HARDENED_USERCOPY... Kees?
As far as I can see it's mostly check_heap_object() that is the
problem, so I'm open to finding a way to just bypass that sub-routine.
However, as far as I can see none of the other block / filesystem user
copy implementations submit to the hardened checks, like
bio_copy_from_iter(), and iov_iter_copy_from_user_atomic() . So,
either those need to grow additional checks, or the hardened copy
implementation is targeting single object copy use cases, not
necessarily block-I/O. Yes, Kees, please advise.
next prev parent reply other threads:[~2019-05-17 15:08 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-17 0:33 [PATCH] libnvdimm/pmem: Bypass CONFIG_HARDENED_USERCOPY overhead Dan Williams
2019-05-17 8:47 ` Jan Kara
2019-05-17 9:06 ` David Laight
2019-05-17 15:53 ` Kees Cook
2019-05-17 16:14 ` David Laight
2019-05-17 16:40 ` Kees Cook
2019-05-17 15:08 ` Dan Williams [this message]
2019-05-17 15:56 ` Kees Cook
2019-05-17 17:28 ` Dan Williams
2019-05-17 19:25 ` Kees Cook
2019-05-19 4:46 ` Dan Williams
2019-05-20 7:52 ` Jan Kara
2019-05-20 15:40 ` 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='CAPcyv4iZZCgcC657ZOysBP9=1ejp3jfFj=VETVBPrgmfg7xUEw@mail.gmail.com' \
--to=dan.j.williams@intel.com \
--cc=hch@lst.de \
--cc=jack@suse.cz \
--cc=jeff.smits@intel.com \
--cc=jmoyer@redhat.com \
--cc=keescook@chromium.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvdimm@lists.01.org \
--cc=mingo@redhat.com \
--cc=stable@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=viro@zeniv.linux.org.uk \
--cc=willy@infradead.org \
/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).