public inbox for linux-pm@vger.kernel.org
 help / color / mirror / Atom feed
From: Johannes Berg <johannes@sipsolutions.net>
To: linuxppc-dev@ozlabs.org
Cc: linux-pm@lists.osdl.org
Subject: [PATCH 02/12] powermac: support G5 CPU hotplug
Date: Wed, 07 Feb 2007 13:45:38 +0100	[thread overview]
Message-ID: <20070207124610.392302000@sipsolutions.net> (raw)
In-Reply-To: 20070207124536.963531000@sipsolutions.net

[-- Attachment #1: powermac-g5-cpu-offline.patch --]
[-- Type: text/plain, Size: 3807 bytes --]

This patch allows "hotplugging" of CPUs on G5 machines. CPUs that are
disabled are put into an idle loop with interrupts hard-disabled, to wake
them up again we kick them just like when bringing them up.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>

---
Except for the in-irq count hack I'm happy with this. I still haven't found
where the in-hard-irq count is set to 1 in the down path during suspend or
resume and other platforms do similar things so I'm inclined to leave this.

--- linux-2.6-git.orig/arch/powerpc/platforms/powermac/smp.c	2007-02-07 02:55:05.657884289 +0100
+++ linux-2.6-git/arch/powerpc/platforms/powermac/smp.c	2007-02-07 02:55:27.352884289 +0100
@@ -898,7 +898,7 @@ void smp_core99_cpu_die(unsigned int cpu
 	cpu_dead[cpu] = 0;
 }
 
-#endif
+#endif /* CONFIG_HOTPLUG_CPU && CONFIG_PP32 */
 
 /* Core99 Macs (dual G4s and G5s) */
 struct smp_ops_t core99_smp_ops = {
@@ -908,8 +908,16 @@ struct smp_ops_t core99_smp_ops = {
 	.setup_cpu	= smp_core99_setup_cpu,
 	.give_timebase	= smp_core99_give_timebase,
 	.take_timebase	= smp_core99_take_timebase,
-#if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32)
+#if defined(CONFIG_HOTPLUG_CPU)
+# if defined(CONFIG_PPC32)
 	.cpu_disable	= smp_core99_cpu_disable,
 	.cpu_die	= smp_core99_cpu_die,
+# endif
+# if defined(CONFIG_PPC64)
+	.cpu_disable	= generic_cpu_disable,
+	.cpu_die	= generic_cpu_die,
+	/* intentionally do *NOT* assign cpu_enable,
+	 * the generic code will use kick_cpu then! */
+# endif
 #endif
 };
--- linux-2.6-git.orig/arch/powerpc/platforms/powermac/setup.c	2007-02-07 02:55:23.435884289 +0100
+++ linux-2.6-git/arch/powerpc/platforms/powermac/setup.c	2007-02-07 02:55:27.388884289 +0100
@@ -490,6 +490,9 @@ static int pmac_late_init(void)
 #ifdef CONFIG_SOFTWARE_SUSPEND
 	pm_set_ops(&pmac_pm_ops);
 #endif /* CONFIG_SOFTWARE_SUSPEND */
+	/* this is udbg (which is __init) and we can later use it during
+	 * cpu hotplug (in smp_core99_kick_cpu) */
+	ppc_md.progress = NULL;
 	return 0;
 }
 
@@ -716,6 +719,39 @@ static int pmac_pci_probe_mode(struct pc
 		return PCI_PROBE_NORMAL;
 	return PCI_PROBE_DEVTREE;
 }
+
+#ifdef CONFIG_HOTPLUG_CPU
+/* access per cpu vars from generic smp.c */
+DECLARE_PER_CPU(int, cpu_state);
+
+static void pmac_cpu_die(void)
+{
+	/* turn off as much as possible, we'll be
+	 * kicked out as this will only be invoked
+	 * on core99 platforms for now ... */
+	hard_irq_disable();
+
+	printk(KERN_INFO "CPU#%d offline\n", smp_processor_id());
+	__get_cpu_var(cpu_state) = CPU_DEAD;
+	smp_wmb();
+
+	/* during the path that leads here preemption is disabled,
+	 * reenable it now so that when coming up preempt count is
+	 * zero correctly */
+	preempt_enable();
+
+	while (1) {
+		ppc64_runlatch_off();
+		if (ppc_md.power_save) {
+			ppc_md.power_save();
+		} else {
+			HMT_low();
+			HMT_very_low();
+		}
+	}
+}
+#endif
+
 #endif
 
 static void __init pmac_init_irq(void)
@@ -769,6 +805,6 @@ define_machine(powermac) {
 	.phys_mem_access_prot	= pci_phys_mem_access_prot,
 #endif
 #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC64)
-	.cpu_die		= generic_mach_cpu_die,
+	.cpu_die		= pmac_cpu_die,
 #endif
 };
--- linux-2.6-git.orig/arch/powerpc/kernel/smp.c	2007-02-07 02:55:06.111884289 +0100
+++ linux-2.6-git/arch/powerpc/kernel/smp.c	2007-02-07 02:55:27.439884289 +0100
@@ -558,6 +558,11 @@ int __devinit start_secondary(void *unus
 
 	local_irq_enable();
 
+	/* it seems to be possible that we died from within
+	 * an interrupt ... reset the in-irq counts */
+	task_thread_info(paca[cpu].__current)->preempt_count &=
+		~(SOFTIRQ_MASK|HARDIRQ_MASK);
+
 	cpu_idle();
 	return 0;
 }

--

  parent reply	other threads:[~2007-02-07 12:45 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-02-07 12:45 [PATCH 00/12] powerpc/powermac power management patches Johannes Berg
2007-02-07 12:45 ` [PATCH 01/12] powerpc: MPIC sys_device & suspend/resume Johannes Berg
2007-02-07 12:45 ` Johannes Berg [this message]
2007-02-08 13:14   ` [PATCH 02/12] powermac: support G5 CPU hotplug Johannes Berg
2007-02-08 13:28     ` Johannes Berg
2007-02-08 15:38       ` Johannes Berg
2007-02-08 21:59       ` Paul Mackerras
2007-02-08 22:33         ` Benjamin Herrenschmidt
2007-02-09 15:39         ` Johannes Berg
2007-02-08 22:24     ` Benjamin Herrenschmidt
2007-02-08 22:24     ` Benjamin Herrenschmidt
2007-02-09 15:36       ` Johannes Berg
2007-02-12 15:15         ` Johannes Berg
2007-02-15  3:44           ` Paul Mackerras
2007-02-15 15:08             ` Johannes Berg
2007-02-15 19:43             ` Johannes Berg
2007-02-14 14:45     ` Pavel Machek
2007-02-14 19:11       ` Johannes Berg
2007-02-14 21:27       ` Benjamin Herrenschmidt
2007-02-07 12:45 ` [PATCH 03/12] powerpc: dart iommu suspend Johannes Berg
2007-02-07 12:45 ` [PATCH 04/12] powerpc: mark pages that dont exist as Nosave Johannes Berg
2007-02-07 12:45 ` [PATCH 05/12] power management: no valid states w/o pm_ops + docs Johannes Berg
2007-02-13 12:12   ` Pavel Machek
2007-02-13 16:08     ` Johannes Berg
2007-02-13 16:14       ` Pavel Machek
2007-02-07 12:45 ` [PATCH 06/12] powerpc: fix suspend states again Johannes Berg
2007-02-07 12:45 ` [PATCH 07/12] powermac: suspend to disk on G5 Johannes Berg
2007-02-07 12:45 ` [PATCH 08/12] powermac: fix G5-cpufreq for cpu on/offline Johannes Berg
2007-02-07 12:45 ` [PATCH 09/12] powerpc: remove bogus comment about page_is_ram Johannes Berg
2007-02-07 12:45 ` [PATCH 10/12] powerpc: remove unneeded exports in mem.c Johannes Berg
2007-02-08  4:25   ` Paul Mackerras
2007-02-08 13:10     ` Johannes Berg
2007-02-08 13:27       ` [PATCH 10/12] powerpc: remove unneeded page_is_ram export Johannes Berg
2007-02-07 12:45 ` [PATCH 11/12] powermac: disallow pmu sleep notifiers from aborting sleep Johannes Berg
2007-02-07 22:52   ` Andreas Schwab
2007-02-08 13:08     ` Johannes Berg
2007-02-08 13:16   ` Johannes Berg
2007-02-07 12:45 ` [PATCH 12/12] powermac: proper sleep management Johannes Berg
2007-02-08 13:16 ` [PATCH 13/12] mpic: add affinity callback for IPI "chip" Johannes Berg
2007-02-08 21:14   ` Benjamin Herrenschmidt
2007-02-09 15:41     ` Johannes Berg
2007-02-12 15:20     ` [PATCH 13/12] mpic: set IPIs to be per-CPU Johannes Berg

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=20070207124610.392302000@sipsolutions.net \
    --to=johannes@sipsolutions.net \
    --cc=linux-pm@lists.osdl.org \
    --cc=linuxppc-dev@ozlabs.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