All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ian Jackson <ian.jackson@eu.citrix.com>
To: <xen-devel@lists.xenproject.org>
Cc: Anthony PERARD <anthony.perard@citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>,
	George Dunlap <george.dunlap@citrix.com>, Wei Liu <wl@xen.org>
Subject: [Xen-devel] [PATCH v3 07/10] libxl: event: poller pipe optimisation
Date: Fri, 17 Jan 2020 14:47:23 +0000	[thread overview]
Message-ID: <20200117144726.582-8-ian.jackson@eu.citrix.com> (raw)
In-Reply-To: <20200117144726.582-1-ian.jackson@eu.citrix.com>

Track in userland whether the poller pipe is nonempty.  This saves us
writing many many bytes to the pipe if nothing ever reads them.

This is going to be relevant in a moment, where we are going to create
a situation where this will happen quite a lot.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Reviewed-by: George Dunlap <george.dunlap@citrix.com>
Tested-by: George Dunlap <george.dunlap@citrix.com>
---
 tools/libxl/libxl_event.c    | 3 +++
 tools/libxl/libxl_internal.h | 1 +
 2 files changed, 4 insertions(+)

diff --git a/tools/libxl/libxl_event.c b/tools/libxl/libxl_event.c
index b50d4e5074..3e76fa5af5 100644
--- a/tools/libxl/libxl_event.c
+++ b/tools/libxl/libxl_event.c
@@ -1417,6 +1417,7 @@ static void afterpoll_internal(libxl__egc *egc, libxl__poller *poller,
     }
 
     if (afterpoll_check_fd(poller,fds,nfds, poller->wakeup_pipe[0],POLLIN)) {
+        poller->pipe_nonempty = 0;
         int e = libxl__self_pipe_eatall(poller->wakeup_pipe[0]);
         if (e) LIBXL__EVENT_DISASTER(gc, "read wakeup", e, 0);
     }
@@ -1809,6 +1810,8 @@ void libxl__poller_put(libxl_ctx *ctx, libxl__poller *p)
 
 void libxl__poller_wakeup(libxl__gc *gc, libxl__poller *p)
 {
+    if (p->pipe_nonempty) return;
+    p->pipe_nonempty = 1;
     int e = libxl__self_pipe_wakeup(p->wakeup_pipe[1]);
     if (e) LIBXL__EVENT_DISASTER(gc, "cannot poke watch pipe", e, 0);
 }
diff --git a/tools/libxl/libxl_internal.h b/tools/libxl/libxl_internal.h
index eec4bf767d..0ab324102b 100644
--- a/tools/libxl/libxl_internal.h
+++ b/tools/libxl/libxl_internal.h
@@ -625,6 +625,7 @@ struct libxl__poller {
     int (*fd_rindices)[3]; /* see libxl_event.c:beforepoll_internal */
 
     int wakeup_pipe[2]; /* 0 means no fd allocated */
+    bool pipe_nonempty;
 
     /*
      * We also use the poller to record whether any fds have been
-- 
2.11.0


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

  parent reply	other threads:[~2020-01-17 14:48 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-17 14:47 [Xen-devel] [PATCH v3 00/10] libxl: event: Fix hang for some applications Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 01/10] libxl: event: Rename poller.fds_changed to .fds_deregistered Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 02/10] libxl: event: Rename ctx.pollers_fd_changed to .pollers_active Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 03/10] libxl: event: Introduce CTX_UNLOCK_EGC_FREE Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 04/10] libxl: event: Make LIBXL__EVENT_DISASTER take a gc, not an egc Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 05/10] libxl: event: Make libxl__poller_wakeup " Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 06/10] libxl: event: Fix hang when mixing blocking and eventy calls Ian Jackson
2020-01-17 14:47 ` Ian Jackson [this message]
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 08/10] libxl: event: Break out baton_wake Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 09/10] libxl: event: Fix possible hang with libxl_osevent_beforepoll Ian Jackson
2020-01-17 14:47 ` [Xen-devel] [PATCH v3 10/10] libxl: event: Move poller pipe emptying to the end of afterpoll Ian Jackson
2020-01-27 16:09 ` [Xen-devel] [PATCH v3 00/10] libxl: event: Fix hang for some applications Ian Jackson

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=20200117144726.582-8-ian.jackson@eu.citrix.com \
    --to=ian.jackson@eu.citrix.com \
    --cc=anthony.perard@citrix.com \
    --cc=george.dunlap@citrix.com \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.