mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: joachim.deguara@amd.com, linux-kernel@vger.kernel.org
Subject: [PATCH for review] [37/48] x86_64: make k8topology multi-core aware
Date: Thu, 19 Jul 2007 15:49:07 +0200 (CEST)	[thread overview]
Message-ID: <20070719134907.E40C514E6E@wotan.suse.de> (raw)
In-Reply-To: <20070719348.540885000@suse.de>


From: "Joachim Deguara" <joachim.deguara@amd.com>

This makes k8topology multicore aware instead of limited to signle- and 
dual-core CPUs.  It uses the CPUID to be more future proof.

Signed-off-by: Joachim Deguara <joachim.deguara@amd.com>
Signed-off-by: Andi Kleen <ak@suse.de>

--
---
 arch/x86_64/mm/k8topology.c |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

Index: linux/arch/x86_64/mm/k8topology.c
===================================================================
--- linux.orig/arch/x86_64/mm/k8topology.c
+++ linux/arch/x86_64/mm/k8topology.c
@@ -48,12 +48,12 @@ int __init k8_scan_nodes(unsigned long s
 { 
 	unsigned long prevbase;
 	struct bootnode nodes[8];
-	int nodeid, i, nb; 
+	int nodeid, i, j, nb;
 	unsigned char nodeids[8];
 	int found = 0;
 	u32 reg;
 	unsigned numnodes;
-	unsigned dualcore = 0;
+	unsigned num_cores;
 
 	if (!early_pci_allowed())
 		return -1;
@@ -64,6 +64,9 @@ int __init k8_scan_nodes(unsigned long s
 
 	printk(KERN_INFO "Scanning NUMA topology in Northbridge %d\n", nb); 
 
+	num_cores = (cpuid_ecx(0x80000008) & 0xff) + 1;
+	printk(KERN_INFO "CPU has %d num_cores\n", num_cores);
+
 	reg = read_pci_config(0, nb, 0, 0x60); 
 	numnodes = ((reg >> 4) & 0xF) + 1;
 	if (numnodes <= 1)
@@ -77,8 +80,6 @@ int __init k8_scan_nodes(unsigned long s
 		unsigned long base,limit; 
 		u32 nodeid;
 		
-		/* Undefined before E stepping, but hopefully 0 */
-		dualcore |= ((read_pci_config(0, nb, 3, 0xe8) >> 12) & 3) == 1;
 		base = read_pci_config(0, nb, 1, 0x40 + i*8);
 		limit = read_pci_config(0, nb, 1, 0x44 + i*8);
 
@@ -174,8 +175,8 @@ int __init k8_scan_nodes(unsigned long s
 	for (i = 0; i < 8; i++) {
 		if (nodes[i].start != nodes[i].end) { 
 			nodeid = nodeids[i];
-			apicid_to_node[nodeid << dualcore] = i;
-			apicid_to_node[(nodeid << dualcore) + dualcore] = i;
+			for (j = 0; j < num_cores; j++)
+				apicid_to_node[(nodeid * num_cores) + j] = i;
 			setup_node_bootmem(i, nodes[i].start, nodes[i].end); 
 		} 
 	}

  parent reply	other threads:[~2007-07-19 14:06 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-07-19 13:48 [PATCH for review] [0/48] Second batch of x86 patches for .23 Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [1/48] i386: pgd_{c,d}tor() static Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [2/48] i386: fix section mismatch warning in intel_cacheinfo Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [3/48] i386: do not restore reserved memory after hibernation Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [4/48] i386: DMI_MATCH patch in reboot.c for SFF Dell OptiPlex 745 - fixes hang on reboot Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [5/48] i386: HPET, check if the counter works Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [6/48] x86: trim memory not covered by WB MTRRs Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [7/48] i386: divorce CONFIG_X86_PAE from CONFIG_HIGHMEM64G Andi Kleen
2007-07-19 14:46   ` Dave Jones
2007-07-19 17:06     ` Andi Kleen
2007-07-19 14:52   ` Christoph Hellwig
2007-07-20  0:45     ` William Lee Irwin III
2007-07-19 13:48 ` [PATCH for review] [8/48] i386: Remove unneeded test of 'task' in dump_trace() Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [9/48] i386: move the kernel to 16MB for NUMA-Q Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [10/48] x86_64: Move functions declarations to header file Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [11/48] x86_64: During VM oom condition, kill all threads in process group Andi Kleen
2007-07-19 14:04   ` Christoph Hellwig
2007-07-19 14:14     ` Geert Uytterhoeven
2007-07-19 15:03       ` Will Schmidt
2007-07-23 18:09       ` [PATCH respin, was PATCH for review] " Will Schmidt
2007-07-23 21:16         ` Andrew Morton
2007-07-24 14:28           ` Will Schmidt
2007-07-24 14:31             ` Christoph Hellwig
2007-07-31  9:31         ` Pavel Machek
2007-07-31 14:55           ` Will Schmidt
2007-07-19 13:48 ` [PATCH for review] [12/48] x86_64: use the global PIT lock Andi Kleen
2007-07-19 15:22   ` Dmitry Torokhov
2007-07-19 17:29     ` Andi Kleen
2007-07-19 19:23       ` Dmitry Torokhov
2007-07-19 19:52         ` Andi Kleen
2007-07-20  4:24           ` Dmitry Torokhov
2007-07-20  8:25             ` Andi Kleen
2007-07-20 12:50               ` Dmitry Torokhov
2007-07-19 13:48 ` [PATCH for review] [13/48] x86_64: fix typo in acpi_pm.c Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [14/48] x86_64: lower printk severity Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [15/48] x86_64: fix wrong comment regarding set_fixmap() Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [16/48] x86_64: Geode HW Random Number Generator depends on X86_32 Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [17/48] x86_64: change _map_single to static in pci_gart.c etc Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [18/48] x86_64: flush_tlb_kernel_range() warning fix Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [19/48] i386: add cpu_relax() to cmos_lock() Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [20/48] i386: replace hard-coded constant with appropriate macro from kernel.h Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [21/48] x86_64: disable the GART in shutdown Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [22/48] x86_64: fix e820_hole_size based on address ranges Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [23/48] x86_64: disable srat when numa emulation succeeds Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [24/48] x86_64: move iommu declaration from proto to iommu.h Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [25/48] i386: remove volatile in apic.c Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [26/48] i386: hpet assumes boot cpu is 0 Andi Kleen
2007-07-19 16:23   ` Jeremy Fitzhardinge
2007-07-19 13:48 ` [PATCH for review] [27/48] i386: move PIT function declarations and constants to correct header file Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [28/48] i386: fix iounmap's use of vm_struct's size field Andi Kleen
2007-07-19 13:48 ` [PATCH for review] [29/48] x86_64: arch/x86_64/kernel/aperture.c lower printk severity Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [30/48] x86_64: arch/x86_64/kernel/e820.c " Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [31/48] i386: basic infrastructure support for AMD geode-class machines Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [32/48] i386: insert HPET firmware resource after PCI enumeration has completed Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [33/48] i386: remove old IRQ balancing debug cruft Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [34/48] i386: Update alignment when 4K stacks are used Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [35/48] x86_64: remove __smp_alt* sections Andi Kleen
2007-07-19 13:49 ` [PATCH for review] [36/48] x86_64: k8topology add family 10h and 11h PCI IDs Andi Kleen
2007-07-19 13:49 ` Andi Kleen [this message]
2007-07-19 13:49 ` [PATCH for review] [38/48] x86_64: Put allocated ELF notes in read-only data segment Andi Kleen

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=20070719134907.E40C514E6E@wotan.suse.de \
    --to=ak@suse.de \
    --cc=joachim.deguara@amd.com \
    --cc=linux-kernel@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

all inboxes | Powered by JetHome®