diff options
author | Tejun Heo <tj@kernel.org> | 2011-01-23 13:37:39 (GMT) |
---|---|---|
committer | Ingo Molnar <mingo@elte.hu> | 2011-01-28 13:54:09 (GMT) |
commit | bbc9e2f452d9c4b166d1f9a78d941d80173312fe (patch) | |
tree | d75d41187b296235f833e942ed8c1dd938a7bae4 /arch/x86/kernel/cpu | |
parent | 89e5dc218e084e13a3996db6693b01478912f4ee (diff) | |
download | linux-bbc9e2f452d9c4b166d1f9a78d941d80173312fe.tar.xz |
x86: Unify cpu/apicid <-> NUMA node mapping between 32 and 64bit
The mapping between cpu/apicid and node is done via
apicid_to_node[] on 64bit and apicid_2_node[] +
apic->x86_32_numa_cpu_node() on 32bit. This difference makes it
difficult to further unify 32 and 64bit NUMA handling.
This patch unifies it by replacing both apicid_to_node[] and
apicid_2_node[] with __apicid_to_node[] array, which is accessed
by two accessors - set_apicid_to_node() and numa_cpu_node(). On
64bit, numa_cpu_node() always consults __apicid_to_node[]
directly while 32bit goes through apic->numa_cpu_node() method
to allow apic implementations to override it.
srat_detect_node() for amd cpus contains workaround for broken
NUMA configuration which assumes relationship between APIC ID,
HT node ID and NUMA topology. Leave it to access
__apicid_to_node[] directly as mapping through CPU might result
in undesirable behavior change. The comment is reformatted and
updated to note the ugliness.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Pekka Enberg <penberg@kernel.org>
Cc: eric.dumazet@gmail.com
Cc: yinghai@kernel.org
Cc: brgerst@gmail.com
Cc: gorcunov@gmail.com
Cc: shaohui.zheng@intel.com
Cc: rientjes@google.com
LKML-Reference: <1295789862-25482-14-git-send-email-tj@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: David Rientjes <rientjes@google.com>
Diffstat (limited to 'arch/x86/kernel/cpu')
-rw-r--r-- | arch/x86/kernel/cpu/amd.c | 47 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/intel.c | 3 |
2 files changed, 31 insertions, 19 deletions
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 7c7bedb..3cce8f2 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -234,17 +234,21 @@ static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c) #endif #if defined(CONFIG_NUMA) && defined(CONFIG_X86_64) +/* + * To workaround broken NUMA config. Read the comment in + * srat_detect_node(). + */ static int __cpuinit nearby_node(int apicid) { int i, node; for (i = apicid - 1; i >= 0; i--) { - node = apicid_to_node[i]; + node = __apicid_to_node[i]; if (node != NUMA_NO_NODE && node_online(node)) return node; } for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) { - node = apicid_to_node[i]; + node = __apicid_to_node[i]; if (node != NUMA_NO_NODE && node_online(node)) return node; } @@ -339,26 +343,35 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) int node; unsigned apicid = c->apicid; - node = per_cpu(cpu_llc_id, cpu); + node = numa_cpu_node(cpu); + if (node == NUMA_NO_NODE) + node = per_cpu(cpu_llc_id, cpu); - if (apicid_to_node[apicid] != NUMA_NO_NODE) - node = apicid_to_node[apicid]; if (!node_online(node)) { - /* Two possibilities here: - - The CPU is missing memory and no node was created. - In that case try picking one from a nearby CPU - - The APIC IDs differ from the HyperTransport node IDs - which the K8 northbridge parsing fills in. - Assume they are all increased by a constant offset, - but in the same order as the HT nodeids. - If that doesn't result in a usable node fall back to the - path for the previous case. */ - + /* + * Two possibilities here: + * + * - The CPU is missing memory and no node was created. In + * that case try picking one from a nearby CPU. + * + * - The APIC IDs differ from the HyperTransport node IDs + * which the K8 northbridge parsing fills in. Assume + * they are all increased by a constant offset, but in + * the same order as the HT nodeids. If that doesn't + * result in a usable node fall back to the path for the + * previous case. + * + * This workaround operates directly on the mapping between + * APIC ID and NUMA node, assuming certain relationship + * between APIC ID, HT node ID and NUMA topology. As going + * through CPU mapping may alter the outcome, directly + * access __apicid_to_node[]. + */ int ht_nodeid = c->initial_apicid; if (ht_nodeid >= 0 && - apicid_to_node[ht_nodeid] != NUMA_NO_NODE) - node = apicid_to_node[ht_nodeid]; + __apicid_to_node[ht_nodeid] != NUMA_NO_NODE) + node = __apicid_to_node[ht_nodeid]; /* Pick a nearby node */ if (!node_online(node)) node = nearby_node(apicid); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index d16c2c5..6052004 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -279,11 +279,10 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) #if defined(CONFIG_NUMA) && defined(CONFIG_X86_64) unsigned node; int cpu = smp_processor_id(); - int apicid = cpu_has_apic ? hard_smp_processor_id() : c->apicid; /* Don't do the funky fallback heuristics the AMD version employs for now. */ - node = apicid_to_node[apicid]; + node = numa_cpu_node(cpu); if (node == NUMA_NO_NODE || !node_online(node)) { /* reuse the value from init_cpu_to_node() */ node = cpu_to_node(cpu); |