xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: George Dunlap <george.dunlap@eu.citrix.com>
To: xen-devel@lists.xensource.com
Cc: george.dunlap@eu.citrix.com
Subject: [PATCH 2 of 2] scheduler: Implement yield for credit1
Date: Mon, 9 Aug 2010 12:20:51 +0100	[thread overview]
Message-ID: <83e1162f1fdcbb294918.1281352851@gdunlap-desktop> (raw)
In-Reply-To: <patchbomb.1281352849@gdunlap-desktop>

This patch implements 'yield' for credit1.  It does this by attempting
to put yielding vcpu behind a single lower-priority vcpu on the runqueue.
If no lower-priority vcpus are in the queue, it will go at the back (which if
the queue is empty, will also be the front).

Runqueues are sorted every 30ms, so that's the longest this priority inversion
can happen.

For workloads with heavy concurrency hazard, and guest which implement
yield-on-spinlock, this patch significantly increases performance and total
system throughput.

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>

diff -r 1c9288cd2af3 -r 83e1162f1fdc xen/common/sched_credit.c
--- a/xen/common/sched_credit.c	Fri Aug 06 15:41:56 2010 +0100
+++ b/xen/common/sched_credit.c	Fri Aug 06 15:41:57 2010 +0100
@@ -64,7 +64,8 @@
 /*
  * Flags
  */
-#define CSCHED_FLAG_VCPU_PARKED 0x0001  /* VCPU over capped credits */
+#define CSCHED_FLAG_VCPU_PARKED    0x0001  /* VCPU over capped credits */
+#define CSCHED_FLAG_VCPU_YIELD     0x0002  /* VCPU yielding */
 
 
 /*
@@ -108,6 +109,12 @@
 
 
 /*
+ * Boot parameters
+ */
+int sched_credit_default_yield = 0;
+boolean_param("sched_credit_default_yield", sched_credit_default_yield);
+
+/*
  * Physical CPU
  */
 struct csched_pcpu {
@@ -202,6 +209,18 @@
             break;
     }
 
+    /* If the vcpu yielded, try to put it behind one lower-priority
+     * runnable vcpu if we can.  The next runq_sort will bring it forward
+     * within 30ms if the queue too long. */
+    if ( svc->flags & CSCHED_FLAG_VCPU_YIELD
+         && __runq_elem(iter)->pri > CSCHED_PRI_IDLE )
+    {
+        iter=iter->next;
+
+        /* Some sanity checks */
+        BUG_ON(iter == runq);
+    }
+
     list_add_tail(&svc->runq_elem, iter);
 }
 
@@ -748,6 +767,18 @@
     __runq_tickle(cpu, svc);
 }
 
+static void
+csched_vcpu_yield(const struct scheduler *ops, struct vcpu *vc)
+{
+    struct csched_vcpu * const sv = CSCHED_VCPU(vc);
+
+    if ( !sched_credit_default_yield )
+    {
+        /* Let the scheduler know that this vcpu is trying to yield */
+        sv->flags |= CSCHED_FLAG_VCPU_YIELD;
+    }
+}
+
 static int
 csched_dom_cntl(
     const struct scheduler *ops,
@@ -1282,6 +1313,12 @@
     }
 
     /*
+     * Clear YIELD flag before scheduling out
+     */
+    if ( scurr->flags & CSCHED_FLAG_VCPU_YIELD )
+        scurr->flags &= ~(CSCHED_FLAG_VCPU_YIELD);
+
+    /*
      * SMP Load balance:
      *
      * If the next highest priority local runnable VCPU has already eaten
@@ -1509,6 +1546,7 @@
 
     .sleep          = csched_vcpu_sleep,
     .wake           = csched_vcpu_wake,
+    .yield          = csched_vcpu_yield,
 
     .adjust         = csched_dom_cntl,

      parent reply	other threads:[~2010-08-09 11:20 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-09 11:20 [PATCH 0 of 2] Scheduler: Implement yield for credit scheduler George Dunlap
2010-08-09 11:20 ` [PATCH 1 of 2] scheduler: Add a per-scheduler yield callback George Dunlap
2010-08-09 11:20 ` George Dunlap [this message]

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=83e1162f1fdcbb294918.1281352851@gdunlap-desktop \
    --to=george.dunlap@eu.citrix.com \
    --cc=xen-devel@lists.xensource.com \
    /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).