linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Dont use num_processors as index to generate logical cpu# in x86_64
@ 2006-01-31 15:59 Ashok Raj
  0 siblings, 0 replies; only message in thread
From: Ashok Raj @ 2006-01-31 15:59 UTC (permalink / raw)
  To: ak; +Cc: linux-kernel, akpm

Hi Andi

Please consider the following patch. Iam in process of developing ACPI based
cpu hotplug support. Using num_processors as index is not friendly to 
hotplug, so i turned it to use cpu_present_map instead. The code is 
also little bit more cleaner and easy to ready.

Cheers,
ashok

-- 
Cheers,
Ashok Raj
- Open Source Technology Center


Minor cleanup to lend better for physical CPU hotplug.
Earlier way of using num_processors as index doesnt
fit if CPUs come and go. This makes the code little bit better
to read, and helps physical hotplug use the same functions as boot.

Reserving CPU0 for BSP is too late to be done in smp_prepare_boot_cpu().
Since logical assignments from MADT is already done via
setup_arch()->acpi_boot_init()->parse lapic

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
------------------------------------------------------
 arch/x86_64/kernel/mpparse.c |   19 ++++++++-----------
 arch/x86_64/kernel/setup.c   |    6 ++++++
 2 files changed, 14 insertions(+), 11 deletions(-)

Index: linux-2.6.16-rc1-mm2/arch/x86_64/kernel/mpparse.c
===================================================================
--- linux-2.6.16-rc1-mm2.orig/arch/x86_64/kernel/mpparse.c
+++ linux-2.6.16-rc1-mm2/arch/x86_64/kernel/mpparse.c
@@ -106,11 +106,11 @@ static int __init mpf_checksum(unsigned 
 	return sum & 0xFF;
 }
 
-static void __init MP_processor_info (struct mpc_config_processor *m)
+static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 {
 	int cpu;
 	unsigned char ver;
-	static int found_bsp=0;
+	cpumask_t tmp_map;
 
 	if (!(m->mpc_cpuflag & CPU_ENABLED)) {
 		disabled_cpus++;
@@ -133,8 +133,10 @@ static void __init MP_processor_info (st
 		return;
 	}
 
-	cpu = num_processors++;
-	
+	num_processors++;
+	cpus_complement(tmp_map, cpu_present_map);
+	cpu = first_cpu(tmp_map);
+
 #if MAX_APICS < 255	
 	if ((int)m->mpc_apicid > MAX_APICS) {
 		printk(KERN_ERR "Processor #%d INVALID. (Max ID: %d).\n",
@@ -160,12 +162,7 @@ static void __init MP_processor_info (st
  		 * entry is BSP, and so on.
  		 */
 		cpu = 0;
-
- 		bios_cpu_apicid[0] = m->mpc_apicid;
- 		x86_cpu_to_apicid[0] = m->mpc_apicid;
- 		found_bsp = 1;
- 	} else
-		cpu = num_processors - found_bsp;
+ 	}
 	bios_cpu_apicid[cpu] = m->mpc_apicid;
 	x86_cpu_to_apicid[cpu] = m->mpc_apicid;
 
@@ -691,7 +688,7 @@ void __init mp_register_lapic_address (
 }
 
 
-void __init mp_register_lapic (
+void __cpuinit mp_register_lapic (
 	u8			id, 
 	u8			enabled)
 {
Index: linux-2.6.16-rc1-mm2/arch/x86_64/kernel/setup.c
===================================================================
--- linux-2.6.16-rc1-mm2.orig/arch/x86_64/kernel/setup.c
+++ linux-2.6.16-rc1-mm2/arch/x86_64/kernel/setup.c
@@ -704,6 +704,12 @@ void __init setup_arch(char **cmdline_p)
 
 	check_ioapic();
 
+	/*
+	 * set this early, so we dont allocate cpu0
+	 * if MADT list doesnt list BSP first
+	 * mpparse.c/MP_processor_info() allocates logical cpu numbers.
+	 */
+	cpu_set(0, cpu_present_map);
 #ifdef CONFIG_ACPI
 	/*
 	 * Read APIC and some other early information from ACPI tables.

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2006-01-31 16:00 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2006-01-31 15:59 Dont use num_processors as index to generate logical cpu# in x86_64 Ashok Raj

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).