[PATCH 2/3] x86, AMD: Cleanup init_amd
From: Borislav Petkov
Date: Tue Jun 24 2014 - 07:26:03 EST
From: Borislav Petkov <bp@xxxxxxx>
Distribute family-specific code to corresponding functions.
Also,
* move the direct mapping splitting around the TSEG SMM area to
bsp_init_amd().
* kill ancient comment about what we should do for K5.
* merge amd_k7_smp_check() into its only caller init_amd_k7 and drop
cpu_has_mp macro.
Signed-off-by: Borislav Petkov <bp@xxxxxxx>
---
arch/x86/kernel/cpu/amd.c | 339 ++++++++++++++++++++++------------------------
1 file changed, 161 insertions(+), 178 deletions(-)
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 8714a78414bf..bc360d3df60e 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -8,6 +8,7 @@
#include <asm/processor.h>
#include <asm/apic.h>
#include <asm/cpu.h>
+#include <asm/smp.h>
#include <asm/pci-direct.h>
#ifdef CONFIG_X86_64
@@ -50,7 +51,6 @@ static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
return wrmsr_safe_regs(gprs);
}
-#ifdef CONFIG_X86_32
/*
* B step AMD K6 before B 9730xxxx have hardware bugs that can cause
* misexecution of code under Linux. Owners of such processors should
@@ -70,6 +70,7 @@ __asm__(".globl vide\n\t.align 4\nvide: ret");
static void init_amd_k5(struct cpuinfo_x86 *c)
{
+#ifdef CONFIG_X86_32
/*
* General Systems BIOSen alias the cpu frequency registers
* of the Elan at 0x000df000. Unfortuantly, one of the Linux
@@ -83,11 +84,12 @@ static void init_amd_k5(struct cpuinfo_x86 *c)
if (inl(CBAR) & CBAR_ENB)
outl(0 | CBAR_KEY, CBAR);
}
+#endif
}
-
static void init_amd_k6(struct cpuinfo_x86 *c)
{
+#ifdef CONFIG_X86_32
u32 l, h;
int mbytes = get_num_physpages() >> (20-PAGE_SHIFT);
@@ -176,10 +178,44 @@ static void init_amd_k6(struct cpuinfo_x86 *c)
/* placeholder for any needed mods */
return;
}
+#endif
}
-static void amd_k7_smp_check(struct cpuinfo_x86 *c)
+static void init_amd_k7(struct cpuinfo_x86 *c)
{
+#ifdef CONFIG_X86_32
+ u32 l, h;
+
+ /*
+ * Bit 15 of Athlon specific MSR 15, needs to be 0
+ * to enable SSE on Palomino/Morgan/Barton CPU's.
+ * If the BIOS didn't enable it already, enable it here.
+ */
+ if (c->x86_model >= 6 && c->x86_model <= 10) {
+ if (!cpu_has(c, X86_FEATURE_XMM)) {
+ printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
+ msr_clear_bit(MSR_K7_HWCR, 15);
+ set_cpu_cap(c, X86_FEATURE_XMM);
+ }
+ }
+
+ /*
+ * It's been determined by AMD that Athlons since model 8 stepping 1
+ * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
+ * As per AMD technical note 27212 0.2
+ */
+ if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
+ rdmsr(MSR_K7_CLK_CTL, l, h);
+ if ((l & 0xfff00000) != 0x20000000) {
+ printk(KERN_INFO
+ "CPU: CLK_CTL MSR was %x. Reprogramming to %x\n",
+ l, ((l & 0x000fffff)|0x20000000));
+ wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
+ }
+ }
+
+ set_cpu_cap(c, X86_FEATURE_K7);
+
/* calling is from identify_secondary_cpu() ? */
if (!c->cpu_index)
return;
@@ -207,7 +243,7 @@ static void amd_k7_smp_check(struct cpuinfo_x86 *c)
if (((c->x86_model == 6) && (c->x86_mask >= 2)) ||
((c->x86_model == 7) && (c->x86_mask >= 1)) ||
(c->x86_model > 7))
- if (cpu_has_mp)
+ if (cpu_has(c, X86_FEATURE_MP))
return;
/* If we get here, not a certified SMP capable AMD system. */
@@ -219,45 +255,8 @@ static void amd_k7_smp_check(struct cpuinfo_x86 *c)
WARN_ONCE(1, "WARNING: This combination of AMD"
" processors is not suitable for SMP.\n");
add_taint(TAINT_CPU_OUT_OF_SPEC, LOCKDEP_NOW_UNRELIABLE);
-}
-
-static void init_amd_k7(struct cpuinfo_x86 *c)
-{
- u32 l, h;
-
- /*
- * Bit 15 of Athlon specific MSR 15, needs to be 0
- * to enable SSE on Palomino/Morgan/Barton CPU's.
- * If the BIOS didn't enable it already, enable it here.
- */
- if (c->x86_model >= 6 && c->x86_model <= 10) {
- if (!cpu_has(c, X86_FEATURE_XMM)) {
- printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
- msr_clear_bit(MSR_K7_HWCR, 15);
- set_cpu_cap(c, X86_FEATURE_XMM);
- }
- }
-
- /*
- * It's been determined by AMD that Athlons since model 8 stepping 1
- * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
- * As per AMD technical note 27212 0.2
- */
- if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
- rdmsr(MSR_K7_CLK_CTL, l, h);
- if ((l & 0xfff00000) != 0x20000000) {
- printk(KERN_INFO
- "CPU: CLK_CTL MSR was %x. Reprogramming to %x\n",
- l, ((l & 0x000fffff)|0x20000000));
- wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
- }
- }
-
- set_cpu_cap(c, X86_FEATURE_K7);
-
- amd_k7_smp_check(c);
-}
#endif
+}
#ifdef CONFIG_NUMA
/*
@@ -446,6 +445,26 @@ static void early_init_amd_mc(struct cpuinfo_x86 *c)
static void bsp_init_amd(struct cpuinfo_x86 *c)
{
+
+#ifdef CONFIG_X86_64
+ if (c->x86 >= 0xf) {
+ unsigned long long tseg;
+
+ /*
+ * Split up direct mapping around the TSEG SMM area.
+ * Don't do it for gbpages because there seems very little
+ * benefit in doing so.
+ */
+ if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
+ unsigned long pfn = tseg >> PAGE_SHIFT;
+
+ printk(KERN_DEBUG "tseg: %010llx\n", tseg);
+ if (pfn_range_is_mapped(pfn, pfn + 1))
+ set_memory_4k((unsigned long)__va(tseg), 1);
+ }
+ }
+#endif
+
if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
if (c->x86 > 0x10 ||
@@ -515,10 +534,100 @@ static const int amd_erratum_383[];
static const int amd_erratum_400[];
static bool cpu_has_amd_erratum(struct cpuinfo_x86 *cpu, const int *erratum);
+static void init_amd_k8(struct cpuinfo_x86 *c)
+{
+ u32 level;
+ u64 value;
+
+ /* On C+ stepping K8 rep microcode works well for copy/memset */
+ level = cpuid_eax(1);
+ if ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
+ set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+
+ /*
+ * Some BIOSes incorrectly force this feature, but only K8 revision D
+ * (model = 0x14) and later actually support it.
+ * (AMD Erratum #110, docId: 25759).
+ */
+ if (c->x86_model < 0x14 && cpu_has(c, X86_FEATURE_LAHF_LM)) {
+ clear_cpu_cap(c, X86_FEATURE_LAHF_LM);
+ if (!rdmsrl_amd_safe(0xc001100d, &value)) {
+ value &= ~BIT_64(32);
+ wrmsrl_amd_safe(0xc001100d, value);
+ }
+ }
+
+ if (!c->x86_model_id[0])
+ strcpy(c->x86_model_id, "Hammer");
+}
+
+static void init_amd_gh(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_64
+ /* do this for boot cpu */
+ if (c == &boot_cpu_data)
+ check_enable_amd_mmconf_dmi();
+
+ fam10h_check_enable_mmcfg();
+#endif
+
+ /*
+ * Disable GART TLB Walk Errors on Fam10h. We do this here because this
+ * is always needed when GART is enabled, even in a kernel which has no
+ * MCE support built in. BIOS should disable GartTlbWlk Errors already.
+ * If it doesn't, we do it here as suggested by the BKDG.
+ *
+ * Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=33012
+ */
+ msr_set_bit(MSR_AMD64_MCx_MASK(4), 10);
+
+ /*
+ * On family 10h BIOS may not have properly enabled WC+ support, causing
+ * it to be converted to CD memtype. This may result in performance
+ * degradation for certain nested-paging guests. Prevent this conversion
+ * by clearing bit 24 in MSR_AMD64_BU_CFG2.
+ *
+ * NOTE: we want to use the _safe accessors so as not to #GP kvm
+ * guests on older kvm hosts.
+ */
+ msr_clear_bit(MSR_AMD64_BU_CFG2, 24);
+
+ if (cpu_has_amd_erratum(c, amd_erratum_383))
+ set_cpu_bug(c, X86_BUG_AMD_TLB_MMATCH);
+}
+
+static void init_amd_bd(struct cpuinfo_x86 *c)
+{
+ u64 value;
+
+ /* re-enable TopologyExtensions if switched off by BIOS */
+ if ((c->x86_model >= 0x10) && (c->x86_model <= 0x1f) &&
+ !cpu_has(c, X86_FEATURE_TOPOEXT)) {
+
+ if (msr_set_bit(0xc0011005, 54) > 0) {
+ rdmsrl(0xc0011005, value);
+ if (value & BIT_64(54)) {
+ set_cpu_cap(c, X86_FEATURE_TOPOEXT);
+ pr_info(FW_INFO "CPU: Re-enabling disabled Topology Extensions Support.\n");
+ }
+ }
+ }
+
+ /*
+ * The way access filter has a performance penalty on some workloads.
+ * Disable it on the affected CPUs.
+ */
+ if ((c->x86_model >= 0x02) && (c->x86_model < 0x20)) {
+ if (!rdmsrl_safe(0xc0011021, &value) && !(value & 0x1E)) {
+ value |= 0x1E;
+ wrmsrl_safe(0xc0011021, value);
+ }
+ }
+}
+
static void init_amd(struct cpuinfo_x86 *c)
{
u32 dummy;
- unsigned long long value;
#ifdef CONFIG_SMP
/*
@@ -540,100 +649,29 @@ static void init_amd(struct cpuinfo_x86 *c)
*/
clear_cpu_cap(c, 0*32+31);
-#ifdef CONFIG_X86_64
- /* On C+ stepping K8 rep microcode works well for copy/memset */
- if (c->x86 == 0xf) {
- u32 level;
-
- level = cpuid_eax(1);
- if ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
- set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-
- /*
- * Some BIOSes incorrectly force this feature, but only K8
- * revision D (model = 0x14) and later actually support it.
- * (AMD Erratum #110, docId: 25759).
- */
- if (c->x86_model < 0x14 && cpu_has(c, X86_FEATURE_LAHF_LM)) {
- clear_cpu_cap(c, X86_FEATURE_LAHF_LM);
- if (!rdmsrl_amd_safe(0xc001100d, &value)) {
- value &= ~(1ULL << 32);
- wrmsrl_amd_safe(0xc001100d, value);
- }
- }
-
- }
if (c->x86 >= 0x10)
set_cpu_cap(c, X86_FEATURE_REP_GOOD);
/* get apicid instead of initial apic id from cpuid */
c->apicid = hard_smp_processor_id();
-#else
-
- /*
- * FIXME: We should handle the K5 here. Set up the write
- * range and also turn on MSR 83 bits 4 and 31 (write alloc,
- * no bus pipeline)
- */
-
- switch (c->x86) {
- case 4:
- init_amd_k5(c);
- break;
- case 5:
- init_amd_k6(c);
- break;
- case 6: /* An Athlon/Duron */
- init_amd_k7(c);
- break;
- }
/* K6s reports MCEs but don't actually have all the MSRs */
if (c->x86 < 6)
clear_cpu_cap(c, X86_FEATURE_MCE);
-#endif
+
+ switch (c->x86) {
+ case 4: init_amd_k5(c); break;
+ case 5: init_amd_k6(c); break;
+ case 6: init_amd_k7(c); break;
+ case 0xf: init_amd_k8(c); break;
+ case 0x10: init_amd_gh(c); break;
+ case 0x15: init_amd_bd(c); break;
+ }
/* Enable workaround for FXSAVE leak */
if (c->x86 >= 6)
set_cpu_bug(c, X86_BUG_FXSAVE_LEAK);
- if (!c->x86_model_id[0]) {
- switch (c->x86) {
- case 0xf:
- /* Should distinguish Models here, but this is only
- a fallback anyways. */
- strcpy(c->x86_model_id, "Hammer");
- break;
- }
- }
-
- /* re-enable TopologyExtensions if switched off by BIOS */
- if ((c->x86 == 0x15) &&
- (c->x86_model >= 0x10) && (c->x86_model <= 0x1f) &&
- !cpu_has(c, X86_FEATURE_TOPOEXT)) {
-
- if (msr_set_bit(0xc0011005, 54) > 0) {
- rdmsrl(0xc0011005, value);
- if (value & BIT_64(54)) {
- set_cpu_cap(c, X86_FEATURE_TOPOEXT);
- pr_info(FW_INFO "CPU: Re-enabling disabled Topology Extensions Support.\n");
- }
- }
- }
-
- /*
- * The way access filter has a performance penalty on some workloads.
- * Disable it on the affected CPUs.
- */
- if ((c->x86 == 0x15) &&
- (c->x86_model >= 0x02) && (c->x86_model < 0x20)) {
-
- if (!rdmsrl_safe(0xc0011021, &value) && !(value & 0x1E)) {
- value |= 0x1E;
- wrmsrl_safe(0xc0011021, value);
- }
- }
-
cpu_detect_cache_sizes(c);
/* Multi core CPU? */
@@ -656,33 +694,6 @@ static void init_amd(struct cpuinfo_x86 *c)
set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
}
-#ifdef CONFIG_X86_64
- if (c->x86 == 0x10) {
- /* do this for boot cpu */
- if (c == &boot_cpu_data)
- check_enable_amd_mmconf_dmi();
-
- fam10h_check_enable_mmcfg();
- }
-
- if (c == &boot_cpu_data && c->x86 >= 0xf) {
- unsigned long long tseg;
-
- /*
- * Split up direct mapping around the TSEG SMM area.
- * Don't do it for gbpages because there seems very little
- * benefit in doing so.
- */
- if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
- unsigned long pfn = tseg >> PAGE_SHIFT;
-
- printk(KERN_DEBUG "tseg: %010llx\n", tseg);
- if (pfn_range_is_mapped(pfn, pfn + 1))
- set_memory_4k((unsigned long)__va(tseg), 1);
- }
- }
-#endif
-
/*
* Family 0x12 and above processors have APIC timer
* running in deep C states.
@@ -690,34 +701,6 @@ static void init_amd(struct cpuinfo_x86 *c)
if (c->x86 > 0x11)
set_cpu_cap(c, X86_FEATURE_ARAT);
- if (c->x86 == 0x10) {
- /*
- * Disable GART TLB Walk Errors on Fam10h. We do this here
- * because this is always needed when GART is enabled, even in a
- * kernel which has no MCE support built in.
- * BIOS should disable GartTlbWlk Errors already. If
- * it doesn't, do it here as suggested by the BKDG.
- *
- * Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=33012
- */
- msr_set_bit(MSR_AMD64_MCx_MASK(4), 10);
-
- /*
- * On family 10h BIOS may not have properly enabled WC+ support,
- * causing it to be converted to CD memtype. This may result in
- * performance degradation for certain nested-paging guests.
- * Prevent this conversion by clearing bit 24 in
- * MSR_AMD64_BU_CFG2.
- *
- * NOTE: we want to use the _safe accessors so as not to #GP kvm
- * guests on older kvm hosts.
- */
- msr_clear_bit(MSR_AMD64_BU_CFG2, 24);
-
- if (cpu_has_amd_erratum(c, amd_erratum_383))
- set_cpu_bug(c, X86_BUG_AMD_TLB_MMATCH);
- }
-
if (cpu_has_amd_erratum(c, amd_erratum_400))
set_cpu_bug(c, X86_BUG_AMD_APIC_C1E);
--
2.0.0
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/