mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Yinghai Lu <yinghai@kernel.org>
To: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Jesse Barnes <jbarnes@virtuousgeek.org>,
	Christoph Lameter <cl@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
	Suresh Siddha <suresh.b.siddha@intel.com>,
	"yinghai@kernel.org" <yinghai@kernel.org>,
	Ingo Molnar <mingo@elte.hu>
Subject: [PATCH 32/37] x86, apic: Use logical flat on intel with <= 8 logical cpus
Date: Fri, 15 Jan 2010 19:07:03 -0800	[thread overview]
Message-ID: <1263611228-6751-33-git-send-email-yinghai@kernel.org> (raw)
In-Reply-To: <1263611228-6751-1-git-send-email-yinghai@kernel.org>

From: Suresh Siddha <suresh.b.siddha@intel.com>

On Intel platforms, we can use logical flat mode if there are <= 8
logical cpu's (irrespective of physical apic id values). This will
enable simplified and efficient IPI and device interrupt routing on
such platforms.

Fix the relevant comments while we are at it.

We can clean up default_setup_apic_routing() by using apic->probe()
but that is a different item.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "yinghai@kernel.org" <yinghai@kernel.org>
LKML-Reference: <1253327399.3948.747.camel@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/apic/apic.c     |   26 ++++++++------------------
 arch/x86/kernel/apic/probe_64.c |   15 +++++++++++----
 2 files changed, 19 insertions(+), 22 deletions(-)

diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index e80f291..aa57c07 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -62,7 +62,7 @@ unsigned int boot_cpu_physical_apicid = -1U;
 /*
  * The highest APIC ID seen during enumeration.
  *
- * This determines the messaging protocol we can use: if all APIC IDs
+ * On AMD, this determines the messaging protocol we can use: if all APIC IDs
  * are in the 0 ... 7 range, then we can use logical addressing which
  * has some performance advantages (better broadcasting).
  *
@@ -1898,24 +1898,14 @@ void __cpuinit generic_processor_info(int apicid, int version)
 		max_physical_apicid = apicid;
 
 #ifdef CONFIG_X86_32
-	/*
-	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
-	 * but we need to work other dependencies like SMP_SUSPEND etc
-	 * before this can be done without some confusion.
-	 * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
-	 *       - Ashok Raj <ashok.raj@intel.com>
-	 */
-	if (max_physical_apicid >= 8) {
-		switch (boot_cpu_data.x86_vendor) {
-		case X86_VENDOR_INTEL:
-			if (!APIC_XAPIC(version)) {
-				def_to_bigsmp = 0;
-				break;
-			}
-			/* If P4 and above fall through */
-		case X86_VENDOR_AMD:
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_INTEL:
+		if (num_processors > 8)
+			def_to_bigsmp = 1;
+		break;
+	case X86_VENDOR_AMD:
+		if (max_physical_apicid >= 8)
 			def_to_bigsmp = 1;
-		}
 	}
 #endif
 
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index 65edc18..c4cbd30 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -64,16 +64,23 @@ void __init default_setup_apic_routing(void)
 			apic = &apic_x2apic_phys;
 		else
 			apic = &apic_x2apic_cluster;
-		printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
 	}
 #endif
 
 	if (apic == &apic_flat) {
-		if (max_physical_apicid >= 8)
-			apic = &apic_physflat;
-		printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_INTEL:
+			if (num_processors > 8)
+				apic = &apic_physflat;
+			break;
+		case X86_VENDOR_AMD:
+			if (max_physical_apicid >= 8)
+				apic = &apic_physflat;
+		}
 	}
 
+	printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
+
 	if (is_vsmp_box()) {
 		/* need to update phys_pkg_id */
 		apic->phys_pkg_id = apicid_phys_pkg_id;
-- 
1.6.4.2


  parent reply	other threads:[~2010-01-16  3:10 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-01-16  3:06 [PATCH -v4 0/37] x86: not use bootmem for x86 Yinghai Lu
2010-01-16  3:06 ` [PATCH 01/37] x86: move range related operation to one file Yinghai Lu
2010-01-19 16:54   ` Christoph Lameter
2010-01-20 19:37     ` Yinghai Lu
2010-01-20 19:51       ` Christoph Lameter
2010-01-20 19:59         ` Yinghai Lu
2010-01-16  3:06 ` [PATCH 02/37] x86: check range in update range Yinghai Lu
2010-01-19 16:56   ` Christoph Lameter
2010-01-20 19:39     ` Yinghai Lu
2010-01-20 19:50       ` Christoph Lameter
2010-01-20 19:57         ` Yinghai Lu
2010-01-16  3:06 ` [PATCH 03/37] x86/pci: use u64 instead of size_t in amd_bus.c Yinghai Lu
2010-01-16  3:06 ` [PATCH 04/37] x86/pci: add cap_resource Yinghai Lu
2010-01-16  3:06 ` [PATCH 05/37] x86/pci: enable pci root res read out for 32bit too Yinghai Lu
2010-01-16  3:06 ` [PATCH 06/37] x86: call early_res_to_bootmem one time Yinghai Lu
2010-01-16  3:06 ` [PATCH 07/37] x86: introduce max_early_res and early_res_count Yinghai Lu
2010-01-16  3:06 ` [PATCH 08/37] x86: dynamic increase early_res array size Yinghai Lu
2010-01-16  3:06 ` [PATCH 09/37] x86: print bootmem free before pci_iommu_alloc and free_all_bootmem -v2 Yinghai Lu
2010-01-16  3:06 ` [PATCH 10/37] x86: make early_node_mem get mem > 4g if possible Yinghai Lu
2010-01-16  3:06 ` [PATCH 11/37] x86: only call dma32_reserve_bootmem 64bit !CONFIG_NUMA Yinghai Lu
2010-01-16  3:06 ` [PATCH 12/37] x86: make 64 bit use early_res instead of bootmem before slab Yinghai Lu
2010-01-16  3:06 ` [PATCH 13/37] sparsemem: put usemap for one node together Yinghai Lu
2010-01-16  3:06 ` [PATCH 14/37] sparsemem: put mem map " Yinghai Lu
2010-01-19 17:13   ` Christoph Lameter
2010-01-16  3:06 ` [PATCH 15/37] x86: change range end to start+size Yinghai Lu
2010-01-16  3:06 ` [PATCH 16/37] x86: move bios page reserve early to head32/64.c Yinghai Lu
2010-01-16  3:06 ` [PATCH 17/37] x86: seperate early_res related code from e820.c Yinghai Lu
2010-01-16  3:06 ` [PATCH 18/37] x86: add find_early_area_size Yinghai Lu
2010-01-16  3:06 ` [PATCH 19/37] x86: move back find_e820_area to e820.c Yinghai Lu
2010-01-16  3:06 ` [PATCH 20/37] early_res: enhance check_and_double_early_res Yinghai Lu
2010-01-16  3:06 ` [PATCH 21/37] x86: make 32bit support NO_BOOTMEM Yinghai Lu
2010-01-16  3:06 ` [PATCH 22/37] move round_up/down to kernel.h Yinghai Lu
2010-01-19 17:57   ` Christoph Lameter
2010-01-20 20:01     ` Yinghai Lu
2010-01-20 20:28     ` Yinghai Lu
2010-01-20 20:52       ` Christoph Lameter
2010-01-20 21:02         ` Yinghai Lu
2010-01-26  0:40       ` Andrew Morton
2010-01-26  0:58         ` H. Peter Anvin
2010-01-26  1:26           ` Andrew Morton
2010-01-16  3:06 ` [PATCH 23/37] x86: add find_fw_memmap_area Yinghai Lu
2010-01-16  3:06 ` [PATCH 24/37] core: move early_res Yinghai Lu
2010-01-16  3:06 ` [PATCH 25/37] ram_buffer_extend_print Yinghai Lu
2010-01-19 17:59   ` Christoph Lameter
2010-01-16  3:06 ` [PATCH 26/37] x86: remove bios data range from e820 Yinghai Lu
2010-01-16  3:06 ` [PATCH 27/37] irq: remove not need bootmem code Yinghai Lu
2010-01-16  3:06 ` [PATCH 28/37] radix: move radix init early Yinghai Lu
2010-01-16  3:07 ` [PATCH 29/37] sparseirq: change irq_desc_ptrs to static Yinghai Lu
2010-01-19 18:01   ` Christoph Lameter
2010-01-20 19:49     ` Yinghai Lu
2010-01-16  3:07 ` [PATCH 30/37] sparseirq: use radix_tree instead of ptrs array Yinghai Lu
2010-01-16  3:07 ` [PATCH 31/37] x86: remove arch_probe_nr_irqs Yinghai Lu
2010-01-16  3:07 ` Yinghai Lu [this message]
2010-01-16  3:07 ` [PATCH 33/37] use nr_cpus= to set nr_cpu_ids early Yinghai Lu
2010-01-19 18:03   ` Christoph Lameter
2010-01-20 19:54     ` Yinghai Lu
2010-01-16  3:07 ` [PATCH 34/37] x86: using logical flat for amd cpu too Yinghai Lu
2010-01-16  3:07 ` [PATCH 35/37] x86: according to nr_cpu_ids to decide if need to leave logical flat Yinghai Lu
2010-01-16  3:07 ` [PATCH 36/37] x86: make 32bit apic flat to physflat switch like 64bit Yinghai Lu
2010-01-16  3:07 ` [PATCH 37/37] x86: use num_processors for possible cpus Yinghai Lu

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=1263611228-6751-33-git-send-email-yinghai@kernel.org \
    --to=yinghai@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=cl@linux-foundation.org \
    --cc=hpa@zytor.com \
    --cc=jbarnes@virtuousgeek.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=suresh.b.siddha@intel.com \
    --cc=tglx@linutronix.de \
    /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

Powered by JetHome