LKML Archive on lore.kernel.org
 help / color / Atom feed
From: Roman Penyaev <rpenyaev@suse.de>
To: Peter Zijlstra <peterz@infradead.org>
Cc: azat@libevent.org, akpm@linux-foundation.org,
	viro@zeniv.linux.org.uk, torvalds@linux-foundation.org,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 06/13] epoll: introduce helpers for adding/removing events to uring
Date: Fri, 31 May 2019 16:21:30 +0200
Message-ID: <98e74ceeefdffc9b50fb33e597d270f7@suse.de> (raw)
In-Reply-To: <20190531125636.GZ2606@hirez.programming.kicks-ass.net>

On 2019-05-31 14:56, Peter Zijlstra wrote:
> On Fri, May 31, 2019 at 01:15:21PM +0200, Roman Penyaev wrote:
>> On 2019-05-31 11:56, Peter Zijlstra wrote:
>> > On Thu, May 16, 2019 at 10:58:03AM +0200, Roman Penyaev wrote:
>> > > +static inline bool ep_add_event_to_uring(struct epitem *epi,
>> > > __poll_t pollflags)
>> > > +{
>> > > +	struct eventpoll *ep = epi->ep;
>> > > +	struct epoll_uitem *uitem;
>> > > +	bool added = false;
>> > > +
>> > > +	if (WARN_ON(!pollflags))
>> > > +		return false;
>> > > +
>> > > +	uitem = &ep->user_header->items[epi->bit];
>> > > +	/*
>> > > +	 * Can be represented as:
>> > > +	 *
>> > > +	 *    was_ready = uitem->ready_events;
>> > > +	 *    uitem->ready_events &= ~EPOLLREMOVED;
>> > > +	 *    uitem->ready_events |= pollflags;
>> > > +	 *    if (!was_ready) {
>> > > +	 *         // create index entry
>> > > +	 *    }
>> > > +	 *
>> > > +	 * See the big comment inside ep_remove_user_item(), why it is
>> > > +	 * important to mask EPOLLREMOVED.
>> > > +	 */
>> > > +	if (!atomic_or_with_mask(&uitem->ready_events,
>> > > +				 pollflags, EPOLLREMOVED)) {
>> > > +		unsigned int i, *item_idx, index_mask;
>> > > +
>> > > +		/*
>> > > +		 * Item was not ready before, thus we have to insert
>> > > +		 * new index to the ring.
>> > > +		 */
>> > > +
>> > > +		index_mask = ep_max_index_nr(ep) - 1;
>> > > +		i = __atomic_fetch_add(&ep->user_header->tail, 1,
>> > > +				       __ATOMIC_ACQUIRE);
>> > > +		item_idx = &ep->user_index[i & index_mask];
>> > > +
>> > > +		/* Signal with a bit, which is > 0 */
>> > > +		*item_idx = epi->bit + 1;
>> >
>> > Did you just increment the user visible tail pointer before you filled
>> > the data? That is, can the concurrent userspace observe the increment
>> > before you put credible data in its place?
>> 
>> No, the "data" is the "ready_events" mask, which was updated before,
>> using cmpxchg, atomic_or_with_mask() call.  All I need is to put an
>> index of just updated item to the uring.
>> 
>> Userspace, in its turn, gets the index from the ring and then checks
>> the mask.
> 
> But where do you write the index into the shared memory? That index
> should be written before you publish the new tail.

The ep_add_event_to_uring() is lockless, thus I can't increase tail 
after,
I need to reserve the index slot, where to write to.  I can use shadow 
tail,
which is not seen by userspace, but I have to guarantee that tail is 
updated
with shadow tail *after* all callers of ep_add_event_to_uring() are 
left.
That is possible, please see the code below, but it adds more 
complexity:

(code was tested on user side, thus has c11 atomics)

static inline void add_event__kernel(struct ring *ring, unsigned bit)
{
         unsigned i, cntr, commit_cntr, *item_idx, tail, old;

         i = __atomic_fetch_add(&ring->cntr, 1, __ATOMIC_ACQUIRE);
         item_idx = &ring->user_itemsindex[i % ring->nr];

         /* Update data */
         *item_idx = bit;

         commit_cntr = __atomic_add_fetch(&ring->commit_cntr, 1, 
__ATOMIC_RELEASE);

         tail = ring->user_header->tail;
         rmb();
         do {
                 cntr = ring->cntr;
                 if (cntr != commit_cntr)
                         /* Someone else will advance tail */
                         break;

                 old = tail;

         } while ((tail = 
__sync_val_compare_and_swap(&ring->user_header->tail, old, cntr)) != 
old);
}

Another way (current solution) is to spin on userspace side in order to 
get
index > 0 (valid index is always > 0), i.e.:

	item_idx_ptr = &index[idx & indeces_mask];

	/*
	 * Spin here till we see valid index
	 */
	while (!(idx = __atomic_load_n(item_idx_ptr, __ATOMIC_ACQUIRE)))
		;



So of course tail can be updated after, like you mentioned, but then I 
have
to introduce locks.  I want to keep it lockless on hot event path.

--
Roman



  reply index

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-16  8:57 [PATCH v3 00/13] epoll: support pollable epoll from userspace Roman Penyaev
2019-05-16  8:57 ` [PATCH v3 01/13] epoll: move private helpers from a header to the source Roman Penyaev
2019-05-16  8:57 ` [PATCH v3 02/13] epoll: introduce user structures for polling from userspace Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 03/13] epoll: allocate user header and user events ring " Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 04/13] epoll: some sanity flags checks for epoll syscalls " Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 05/13] epoll: offload polling to a work in case of epfd polled " Roman Penyaev
2019-05-21  7:51   ` Eric Wong
2019-05-22 12:54     ` Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 06/13] epoll: introduce helpers for adding/removing events to uring Roman Penyaev
2019-05-31  9:55   ` Peter Zijlstra
2019-05-31 11:24     ` Roman Penyaev
2019-05-31 13:11       ` Peter Zijlstra
2019-05-31  9:56   ` Peter Zijlstra
2019-05-31 11:15     ` Roman Penyaev
2019-05-31 12:53       ` Peter Zijlstra
2019-05-31 14:28         ` Roman Penyaev
2019-05-31 16:53           ` Peter Zijlstra
2019-05-31 12:56       ` Peter Zijlstra
2019-05-31 14:21         ` Roman Penyaev [this message]
2019-05-31 16:51           ` Peter Zijlstra
2019-05-31 18:58             ` Roman Penyaev
2019-06-03  9:09               ` Peter Zijlstra
2019-06-03 10:02                 ` Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 07/13] epoll: call ep_add_event_to_uring() from ep_poll_callback() Roman Penyaev
2019-05-31  9:56   ` Peter Zijlstra
2019-05-31 11:22     ` Roman Penyaev
2019-05-31 13:05       ` Peter Zijlstra
2019-05-31 15:05         ` Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 08/13] epoll: support polling from userspace for ep_insert() Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 09/13] epoll: support polling from userspace for ep_remove() Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 10/13] epoll: support polling from userspace for ep_modify() Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 11/13] epoll: support polling from userspace for ep_poll() Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 12/13] epoll: support mapping for epfd when polled from userspace Roman Penyaev
2019-05-16  8:58 ` [PATCH v3 13/13] epoll: implement epoll_create2() syscall Roman Penyaev
2019-05-16 10:03   ` Arnd Bergmann
2019-05-16 10:20     ` Roman Penyaev
2019-05-16 10:57       ` Arnd Bergmann
2019-05-22  2:33       ` Andrew Morton
2019-05-22  9:11         ` Roman Penyaev
2019-05-22 11:14         ` Arnd Bergmann
2019-05-22 18:36           ` Andrew Morton
2019-05-31  9:55 ` [PATCH v3 00/13] epoll: support pollable epoll from userspace Peter Zijlstra
2019-05-31 14:48 ` Jens Axboe
2019-05-31 16:02   ` Roman Penyaev
2019-05-31 16:54     ` Jens Axboe
2019-05-31 19:45       ` Roman Penyaev
2019-05-31 21:09         ` Jens Axboe
2019-06-05  6:17           ` Roman Penyaev
2019-05-31 16:33 ` Peter Zijlstra
2019-05-31 18:50   ` Roman Penyaev

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=98e74ceeefdffc9b50fb33e597d270f7@suse.de \
    --to=rpenyaev@suse.de \
    --cc=akpm@linux-foundation.org \
    --cc=azat@libevent.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterz@infradead.org \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    /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

LKML Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git
	git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git
	git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git
	git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git
	git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git
	git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git
	git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git
	git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git
	git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \
		linux-kernel@vger.kernel.org
	public-inbox-index lkml

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git