LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Andy Lutomirski <email@example.com>
To: Matthew Wilcox <firstname.lastname@example.org>
Cc: Eric Dumazet <email@example.com>,
Christoph Hellwig <firstname.lastname@example.org>,
Eric Dumazet <email@example.com>,
"David S . Miller" <firstname.lastname@example.org>,
Andy Lutomirski <email@example.com>,
Soheil Hassas Yeganeh <firstname.lastname@example.org>
Subject: Re: [PATCH net-next 1/2] tcp: add TCP_ZEROCOPY_RECEIVE support for zerocopy receive
Date: Wed, 25 Apr 2018 09:22:01 -0700 [thread overview]
Message-ID: <CALCETrWaekirEe+rKiPB-Zim6ZHKL-n7nfk9wrsHra_FtrS=DA@mail.gmail.com> (raw)
On Wed, Apr 25, 2018 at 9:04 AM, Matthew Wilcox <email@example.com> wrote:
> On Wed, Apr 25, 2018 at 06:01:02AM -0700, Eric Dumazet wrote:
>> On 04/24/2018 11:28 PM, Christoph Hellwig wrote:
>> > On Tue, Apr 24, 2018 at 10:27:21PM -0700, Eric Dumazet wrote:
>> >> When adding tcp mmap() implementation, I forgot that socket lock
>> >> had to be taken before current->mm->mmap_sem. syzbot eventually caught
>> >> the bug.
>> >> Since we can not lock the socket in tcp mmap() handler we have to
>> >> split the operation in two phases.
>> >> 1) mmap() on a tcp socket simply reserves VMA space, and nothing else.
>> >> This operation does not involve any TCP locking.
>> >> 2) setsockopt(fd, IPPROTO_TCP, TCP_ZEROCOPY_RECEIVE, ...) implements
>> >> the transfert of pages from skbs to one VMA.
>> >> This operation only uses down_read(¤t->mm->mmap_sem) after
>> >> holding TCP lock, thus solving the lockdep issue.
>> >> This new implementation was suggested by Andy Lutomirski with great details.
>> > Thanks, this looks much more sensible to me.
>> Thanks Christoph
>> Note the high cost of zap_page_range(), needed to avoid -EBUSY being returned
>> from vm_insert_page() the second time TCP_ZEROCOPY_RECEIVE is used on one VMA.
>> Ideally a vm_replace_page() would avoid this cost ?
> If you don't zap the page range, any of the CPUs in the system where
> any thread in this task have ever run may have a TLB entry pointing to
> this page ... if the page is being recycled into the page allocator,
> then that page might end up as a slab page or page table or page cache
> while the other CPU still have access to it.
Indeed. This is one of the reasons that Linus has generally been
quite vocal that he doesn't like MMU-based zerocopy schemes.
> You could hang onto the page until you've built up a sufficiently large
> batch, then bulk-invalidate all of the TLB entries, but we start to get
> into weirdnesses on different CPU architectures.
The existing mmu_gather code should already handle this at least
moderately well. If it's not, then it should be fixed.
On x86, there is no operation to flush a range of addresses. You can
flush one address or you can flush all of them. If you flush one page
at a time, then you might never recover the performance of a plain old
memcpy(). If you flush all of them, then you're hurting the
performance of everything else in the task.
In general, I suspect that the zerocopy receive mechanism will only
really be a win in single-threaded applications that consume large
amounts of receive bandwidth on a single TCP socket using lots of
memory and don't do all that much else.
next prev parent reply other threads:[~2018-04-25 16:22 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-25 5:27 [PATCH net-next 0/2] tcp: mmap: rework " Eric Dumazet
2018-04-25 5:27 ` [PATCH net-next 1/2] tcp: add TCP_ZEROCOPY_RECEIVE support for " Eric Dumazet
2018-04-25 6:28 ` Christoph Hellwig
2018-04-25 13:01 ` Eric Dumazet
2018-04-25 15:24 ` Christoph Hellwig
2018-04-25 16:04 ` Matthew Wilcox
2018-04-25 16:20 ` Eric Dumazet
2018-04-25 16:30 ` Matthew Wilcox
2018-04-25 16:22 ` Andy Lutomirski [this message]
2018-04-25 16:35 ` Eric Dumazet
2018-04-25 16:44 ` Eric Dumazet
2018-04-25 13:42 ` Eric Dumazet
2018-04-25 5:27 ` [PATCH net-next 2/2] selftests: net: tcp_mmap must use TCP_ZEROCOPY_RECEIVE Eric Dumazet
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:
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--subject='Re: [PATCH net-next 1/2] tcp: add TCP_ZEROCOPY_RECEIVE support for zerocopy receive' \
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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).