x86/amd: Fix L1i and L2 cache sharing information for AMD family 15h processors

commit 32c3233885eb10ac9cb9410f2f8cd64b8df2b2a1 upstream.

For L1 instruction cache and L2 cache the shared CPU information
is wrong. On current AMD family 15h CPUs those caches are shared
between both cores of a compute unit.

This fixes https://bugzilla.kernel.org/show_bug.cgi?id=42607

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Petkov Borislav <Borislav.Petkov@amd.com>
Cc: Dave Jones <davej@redhat.com>
Link: http://lkml.kernel.org/r/20120208195229.GA17523@alberich.amd.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index c105c533ed94575e54b047ab131ff4a9170282b8..fde44284cf21447230d3863fbb838ddea3d102a0 100644 (file)
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -330,8 +330,7 @@ static void __cpuinit amd_calc_l3_indices(struct amd_l3_cache *l3)
        l3->indices = (max(max3(sc0, sc1, sc2), sc3) << 10) - 1;
 }
 
-static void __cpuinit amd_init_l3_cache(struct _cpuid4_info_regs *this_leaf,
-                                       int index)
+static void __cpuinit amd_init_l3_cache(struct _cpuid4_info_regs *this_leaf, int index)
 {
        static struct amd_l3_cache *__cpuinitdata l3_caches;
        int node;
@@ -748,14 +747,16 @@ static DEFINE_PER_CPU(struct _cpuid4_info *, ici_cpuid4_info);
 #define CPUID4_INFO_IDX(x, y)  (&((per_cpu(ici_cpuid4_info, x))[y]))
 
 #ifdef CONFIG_SMP
-static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
+
+static int __cpuinit cache_shared_amd_cpu_map_setup(unsigned int cpu, int index)
 {
-       struct _cpuid4_info     *this_leaf, *sibling_leaf;
-       unsigned long num_threads_sharing;
-       int index_msb, i, sibling;
+       struct _cpuid4_info *this_leaf;
+       int ret, i, sibling;
        struct cpuinfo_x86 *c = &cpu_data(cpu);
 
-       if ((index == 3) && (c->x86_vendor == X86_VENDOR_AMD)) {
+       ret = 0;
+       if (index == 3) {
+               ret = 1;
                for_each_cpu(i, cpu_llc_shared_mask(cpu)) {
                        if (!per_cpu(ici_cpuid4_info, i))
                                continue;
@@ -766,8 +767,35 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
                                set_bit(sibling, this_leaf->shared_cpu_map);
                        }
                }
-               return;
+       } else if ((c->x86 == 0x15) && ((index == 1) || (index == 2))) {
+               ret = 1;
+               for_each_cpu(i, cpu_sibling_mask(cpu)) {
+                       if (!per_cpu(ici_cpuid4_info, i))
+                               continue;
+                       this_leaf = CPUID4_INFO_IDX(i, index);
+                       for_each_cpu(sibling, cpu_sibling_mask(cpu)) {
+                               if (!cpu_online(sibling))
+                                       continue;
+                               set_bit(sibling, this_leaf->shared_cpu_map);
+                       }
+               }
        }
+
+       return ret;
+}
+
+static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
+{
+       struct _cpuid4_info *this_leaf, *sibling_leaf;
+       unsigned long num_threads_sharing;
+       int index_msb, i;
+       struct cpuinfo_x86 *c = &cpu_data(cpu);
+
+       if (c->x86_vendor == X86_VENDOR_AMD) {
+               if (cache_shared_amd_cpu_map_setup(cpu, index))
+                       return;
+       }
+
        this_leaf = CPUID4_INFO_IDX(cpu, index);
        num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing;