netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: therbert@google.com (Tom Herbert)
To: davem@davemloft.net, netdev@vger.kernel.org
Subject: RFC [PATCH net-2.6 2/6] net: Changes to x86_64 for softirq scheduling
Date: Wed,  5 Mar 2008 12:51:16 -0800 (PST)	[thread overview]
Message-ID: <20080305205116.5B86741278F@localhost> (raw)

This patches includes changes to x86_64 to support scheduling of softirq's between CPUs.

Signed-off-by: Tom Herbert <therbert@google.com>

---

diff -uprN -X /tmp/donts/si_2 net-2.6/arch/x86/kernel/smp_64.c net-2.6.patch/arch/x86/kernel/smp_64.c
--- net-2.6/arch/x86/kernel/smp_64.c	2008-03-05 09:02:22.768649000 -0800
+++ net-2.6.patch/arch/x86/kernel/smp_64.c	2008-03-05 09:25:33.403752000 -0800
@@ -296,6 +296,15 @@ void smp_send_reschedule(int cpu)
 }
 
 /*
+ * this function sends a 'reschedule' IPI to multiple
+ * CPUs in the mask.
+ */
+void smp_send_reschedule_mask(cpumask_t mask)
+{
+	send_IPI_mask(mask, RESCHEDULE_VECTOR);
+}
+
+/*
  * Structure and data for smp_call_function(). This is designed to minimise
  * static memory requirements. It also looks cleaner.
  */
@@ -494,11 +503,20 @@ void smp_send_stop(void)
  * Reschedule call back. Nothing to do,
  * all the work is done automatically when
  * we return from the interrupt.
+ * Reschedule call back.  Process rescheduling is done
+ * automatically on return from interrupt.  Also, 'or' into
+ * local_softirq_pending any softirq scheduling requests sent
+ * from other CPUs, these softirq's will execute in irq_exit.
  */
 asmlinkage void smp_reschedule_interrupt(void)
 {
 	ack_APIC_irq();
+
+	exit_idle();
+	irq_enter();
+	or_alt_softirqs_pending_irqoff();
 	add_pda(irq_resched_count, 1);
+	irq_exit();
 }
 
 asmlinkage void smp_call_function_interrupt(void)
diff -uprN -X /tmp/donts/si_2 net-2.6/include/asm-x86/smp_64.h net-2.6.patch/include/asm-x86/smp_64.h
--- net-2.6/include/asm-x86/smp_64.h	2008-03-05 09:03:19.005115000 -0800
+++ net-2.6.patch/include/asm-x86/smp_64.h	2008-03-05 09:27:32.532262000 -0800
@@ -74,6 +74,9 @@ static inline int num_booting_cpus(void)
 }
 
 extern void smp_send_reschedule(int cpu);
+extern void smp_send_reschedule_mask(cpumask_t mask);
+#define       ARCH_HAS_SEND_RESCHEDULE_MASK
+
 
 #else /* CONFIG_SMP */
 

                 reply	other threads:[~2008-03-05 20:51 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20080305205116.5B86741278F@localhost \
    --to=therbert@google.com \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.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).