[GIT PULL] x86 fixes
From: Ingo Molnar
Date: Wed Dec 06 2017 - 17:36:35 EST
Linus,
Please pull the latest x86-urgent-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-urgent-for-linus
# HEAD: 88edb57d1e0b262e669c5cad36646dcf5a7f37f5 x86/vdso: Change time() prototype to match __vdso_time()
Misc fixes:
- Make CR4 handling irq-safe, which bug vmware guests ran into
- Don't crash on early IRQs in Xen guests
- Don't crash secondary CPU bringup if #UD assisted WARN()ings are triggered
- Make X86_BUG_FXSAVE_LEAK optional on newer AMD CPUs that have the fix
- Fix AMD Fam17h microcode loading
- Fix broadcom_postcore_init() if ACPI is disabled
- Fix resume regression in __restore_processor_context()
- Fix Sparse warnings
- Fix a GCC-8 warning
Thanks,
Ingo
------------------>
Andy Lutomirski (1):
x86/power: Fix some ordering bugs in __restore_processor_context()
Arnd Bergmann (1):
x86/vdso: Change time() prototype to match __vdso_time()
Chunyu Hu (1):
x86/idt: Load idt early in start_secondary
Colin Ian King (1):
x86: Fix Sparse warnings about non-static functions
Juergen Gross (1):
x86/xen: Support early interrupts in xen pv guests
Nadav Amit (2):
x86/tlb: Refactor CR4 setting and shadow write
x86/tlb: Disable interrupts when changing CR4
Rafael J. Wysocki (1):
x86/PCI: Make broadcom_postcore_init() check acpi_disabled
Rudolf Marek (1):
x86/cpufeatures: Make X86_BUG_FXSAVE_LEAK detectable in CPUID on AMD
Tom Lendacky (1):
x86/microcode/AMD: Add support for fam17h microcode loading
arch/x86/entry/vdso/vclock_gettime.c | 2 +-
arch/x86/include/asm/cpufeatures.h | 1 +
arch/x86/include/asm/segment.h | 12 ++++++++++++
arch/x86/include/asm/tlbflush.h | 35 ++++++++++++++++++----------------
arch/x86/kernel/apic/vector.c | 4 ++--
arch/x86/kernel/cpu/amd.c | 7 +++++--
arch/x86/kernel/cpu/microcode/amd.c | 4 ++++
arch/x86/kernel/process.c | 2 +-
arch/x86/kernel/smpboot.c | 2 +-
arch/x86/mm/extable.c | 4 +++-
arch/x86/pci/broadcom_bus.c | 2 +-
arch/x86/platform/uv/uv_nmi.c | 4 ++--
arch/x86/power/cpu.c | 21 ++++++++++++++++----
arch/x86/xen/enlighten_pv.c | 37 +++++++++++++++++++++++-------------
arch/x86/xen/xen-asm_64.S | 14 ++++++++++++++
15 files changed, 107 insertions(+), 44 deletions(-)
diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c
index 11b13c4b43d5..f19856d95c60 100644
--- a/arch/x86/entry/vdso/vclock_gettime.c
+++ b/arch/x86/entry/vdso/vclock_gettime.c
@@ -324,5 +324,5 @@ notrace time_t __vdso_time(time_t *t)
*t = result;
return result;
}
-int time(time_t *t)
+time_t time(time_t *t)
__attribute__((weak, alias("__vdso_time")));
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index c0b0e9e8aa66..800104c8a3ed 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -266,6 +266,7 @@
/* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */
#define X86_FEATURE_CLZERO (13*32+ 0) /* CLZERO instruction */
#define X86_FEATURE_IRPERF (13*32+ 1) /* Instructions Retired Count */
+#define X86_FEATURE_XSAVEERPTR (13*32+ 2) /* Always save/restore FP error pointers */
/* Thermal and Power Management Leaf, CPUID level 0x00000006 (EAX), word 14 */
#define X86_FEATURE_DTHERM (14*32+ 0) /* Digital Thermal Sensor */
diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h
index b20f9d623f9c..8f09012b92e7 100644
--- a/arch/x86/include/asm/segment.h
+++ b/arch/x86/include/asm/segment.h
@@ -236,11 +236,23 @@
*/
#define EARLY_IDT_HANDLER_SIZE 9
+/*
+ * xen_early_idt_handler_array is for Xen pv guests: for each entry in
+ * early_idt_handler_array it contains a prequel in the form of
+ * pop %rcx; pop %r11; jmp early_idt_handler_array[i]; summing up to
+ * max 8 bytes.
+ */
+#define XEN_EARLY_IDT_HANDLER_SIZE 8
+
#ifndef __ASSEMBLY__
extern const char early_idt_handler_array[NUM_EXCEPTION_VECTORS][EARLY_IDT_HANDLER_SIZE];
extern void early_ignore_irq(void);
+#if defined(CONFIG_X86_64) && defined(CONFIG_XEN_PV)
+extern const char xen_early_idt_handler_array[NUM_EXCEPTION_VECTORS][XEN_EARLY_IDT_HANDLER_SIZE];
+#endif
+
/*
* Load a segment. Fall back on loading the zero segment if something goes
* wrong. This variant assumes that loading zero fully clears the segment.
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 509046cfa5ce..877b5c1a1b12 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -173,40 +173,43 @@ static inline void cr4_init_shadow(void)
this_cpu_write(cpu_tlbstate.cr4, __read_cr4());
}
+static inline void __cr4_set(unsigned long cr4)
+{
+ lockdep_assert_irqs_disabled();
+ this_cpu_write(cpu_tlbstate.cr4, cr4);
+ __write_cr4(cr4);
+}
+
/* Set in this cpu's CR4. */
static inline void cr4_set_bits(unsigned long mask)
{
- unsigned long cr4;
+ unsigned long cr4, flags;
+ local_irq_save(flags);
cr4 = this_cpu_read(cpu_tlbstate.cr4);
- if ((cr4 | mask) != cr4) {
- cr4 |= mask;
- this_cpu_write(cpu_tlbstate.cr4, cr4);
- __write_cr4(cr4);
- }
+ if ((cr4 | mask) != cr4)
+ __cr4_set(cr4 | mask);
+ local_irq_restore(flags);
}
/* Clear in this cpu's CR4. */
static inline void cr4_clear_bits(unsigned long mask)
{
- unsigned long cr4;
+ unsigned long cr4, flags;
+ local_irq_save(flags);
cr4 = this_cpu_read(cpu_tlbstate.cr4);
- if ((cr4 & ~mask) != cr4) {
- cr4 &= ~mask;
- this_cpu_write(cpu_tlbstate.cr4, cr4);
- __write_cr4(cr4);
- }
+ if ((cr4 & ~mask) != cr4)
+ __cr4_set(cr4 & ~mask);
+ local_irq_restore(flags);
}
-static inline void cr4_toggle_bits(unsigned long mask)
+static inline void cr4_toggle_bits_irqsoff(unsigned long mask)
{
unsigned long cr4;
cr4 = this_cpu_read(cpu_tlbstate.cr4);
- cr4 ^= mask;
- this_cpu_write(cpu_tlbstate.cr4, cr4);
- __write_cr4(cr4);
+ __cr4_set(cr4 ^ mask);
}
/* Read the CR4 shadow. */
diff --git a/arch/x86/kernel/apic/vector.c b/arch/x86/kernel/apic/vector.c
index 6a823a25eaff..750449152b04 100644
--- a/arch/x86/kernel/apic/vector.c
+++ b/arch/x86/kernel/apic/vector.c
@@ -542,8 +542,8 @@ static int x86_vector_alloc_irqs(struct irq_domain *domain, unsigned int virq,
}
#ifdef CONFIG_GENERIC_IRQ_DEBUGFS
-void x86_vector_debug_show(struct seq_file *m, struct irq_domain *d,
- struct irq_data *irqd, int ind)
+static void x86_vector_debug_show(struct seq_file *m, struct irq_domain *d,
+ struct irq_data *irqd, int ind)
{
unsigned int cpu, vector, prev_cpu, prev_vector;
struct apic_chip_data *apicd;
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index d58184b7cd44..bcb75dc97d44 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -804,8 +804,11 @@ static void init_amd(struct cpuinfo_x86 *c)
case 0x17: init_amd_zn(c); break;
}
- /* Enable workaround for FXSAVE leak */
- if (c->x86 >= 6)
+ /*
+ * Enable workaround for FXSAVE leak on CPUs
+ * without a XSaveErPtr feature
+ */
+ if ((c->x86 >= 6) && (!cpu_has(c, X86_FEATURE_XSAVEERPTR)))
set_cpu_bug(c, X86_BUG_FXSAVE_LEAK);
cpu_detect_cache_sizes(c);
diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c
index c6daec4bdba5..330b8462d426 100644
--- a/arch/x86/kernel/cpu/microcode/amd.c
+++ b/arch/x86/kernel/cpu/microcode/amd.c
@@ -470,6 +470,7 @@ static unsigned int verify_patch_size(u8 family, u32 patch_size,
#define F14H_MPB_MAX_SIZE 1824
#define F15H_MPB_MAX_SIZE 4096
#define F16H_MPB_MAX_SIZE 3458
+#define F17H_MPB_MAX_SIZE 3200
switch (family) {
case 0x14:
@@ -481,6 +482,9 @@ static unsigned int verify_patch_size(u8 family, u32 patch_size,
case 0x16:
max_size = F16H_MPB_MAX_SIZE;
break;
+ case 0x17:
+ max_size = F17H_MPB_MAX_SIZE;
+ break;
default:
max_size = F1XH_MPB_MAX_SIZE;
break;
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 97fb3e5737f5..bb988a24db92 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -299,7 +299,7 @@ void __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
}
if ((tifp ^ tifn) & _TIF_NOTSC)
- cr4_toggle_bits(X86_CR4_TSD);
+ cr4_toggle_bits_irqsoff(X86_CR4_TSD);
if ((tifp ^ tifn) & _TIF_NOCPUID)
set_cpuid_faulting(!!(tifn & _TIF_NOCPUID));
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 3d01df7d7cf6..05a97d5fe298 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -237,7 +237,7 @@ static void notrace start_secondary(void *unused)
load_cr3(swapper_pg_dir);
__flush_tlb_all();
#endif
-
+ load_current_idt();
cpu_init();
x86_cpuinit.early_percpu_clock_init();
preempt_disable();
diff --git a/arch/x86/mm/extable.c b/arch/x86/mm/extable.c
index 3321b446b66c..88754bfd425f 100644
--- a/arch/x86/mm/extable.c
+++ b/arch/x86/mm/extable.c
@@ -1,6 +1,7 @@
#include <linux/extable.h>
#include <linux/uaccess.h>
#include <linux/sched/debug.h>
+#include <xen/xen.h>
#include <asm/fpu/internal.h>
#include <asm/traps.h>
@@ -212,8 +213,9 @@ void __init early_fixup_exception(struct pt_regs *regs, int trapnr)
* Old CPUs leave the high bits of CS on the stack
* undefined. I'm not sure which CPUs do this, but at least
* the 486 DX works this way.
+ * Xen pv domains are not using the default __KERNEL_CS.
*/
- if (regs->cs != __KERNEL_CS)
+ if (!xen_pv_domain() && regs->cs != __KERNEL_CS)
goto fail;
/*
diff --git a/arch/x86/pci/broadcom_bus.c b/arch/x86/pci/broadcom_bus.c
index bb461cfd01ab..526536c81ddc 100644
--- a/arch/x86/pci/broadcom_bus.c
+++ b/arch/x86/pci/broadcom_bus.c
@@ -97,7 +97,7 @@ static int __init broadcom_postcore_init(void)
* We should get host bridge information from ACPI unless the BIOS
* doesn't support it.
*/
- if (acpi_os_get_root_pointer())
+ if (!acpi_disabled && acpi_os_get_root_pointer())
return 0;
#endif
diff --git a/arch/x86/platform/uv/uv_nmi.c b/arch/x86/platform/uv/uv_nmi.c
index c34bd8233f7c..5f64f30873e2 100644
--- a/arch/x86/platform/uv/uv_nmi.c
+++ b/arch/x86/platform/uv/uv_nmi.c
@@ -905,7 +905,7 @@ static inline void uv_call_kgdb_kdb(int cpu, struct pt_regs *regs, int master)
/*
* UV NMI handler
*/
-int uv_handle_nmi(unsigned int reason, struct pt_regs *regs)
+static int uv_handle_nmi(unsigned int reason, struct pt_regs *regs)
{
struct uv_hub_nmi_s *hub_nmi = uv_hub_nmi;
int cpu = smp_processor_id();
@@ -1013,7 +1013,7 @@ void uv_nmi_init(void)
}
/* Setup HUB NMI info */
-void __init uv_nmi_setup_common(bool hubbed)
+static void __init uv_nmi_setup_common(bool hubbed)
{
int size = sizeof(void *) * (1 << NODES_SHIFT);
int cpu;
diff --git a/arch/x86/power/cpu.c b/arch/x86/power/cpu.c
index 84fcfde53f8f..5191de14f4df 100644
--- a/arch/x86/power/cpu.c
+++ b/arch/x86/power/cpu.c
@@ -226,8 +226,20 @@ static void notrace __restore_processor_state(struct saved_context *ctxt)
load_idt((const struct desc_ptr *)&ctxt->idt_limit);
#endif
+#ifdef CONFIG_X86_64
/*
- * segment registers
+ * We need GSBASE restored before percpu access can work.
+ * percpu access can happen in exception handlers or in complicated
+ * helpers like load_gs_index().
+ */
+ wrmsrl(MSR_GS_BASE, ctxt->gs_base);
+#endif
+
+ fix_processor_context();
+
+ /*
+ * Restore segment registers. This happens after restoring the GDT
+ * and LDT, which happen in fix_processor_context().
*/
#ifdef CONFIG_X86_32
loadsegment(es, ctxt->es);
@@ -248,13 +260,14 @@ static void notrace __restore_processor_state(struct saved_context *ctxt)
load_gs_index(ctxt->gs);
asm volatile ("movw %0, %%ss" :: "r" (ctxt->ss));
+ /*
+ * Restore FSBASE and user GSBASE after reloading the respective
+ * segment selectors.
+ */
wrmsrl(MSR_FS_BASE, ctxt->fs_base);
- wrmsrl(MSR_GS_BASE, ctxt->gs_base);
wrmsrl(MSR_KERNEL_GS_BASE, ctxt->gs_kernel_base);
#endif
- fix_processor_context();
-
do_fpu_end();
tsc_verify_tsc_adjust(true);
x86_platform.restore_sched_clock_state();
diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c
index 5b2b3f3f6531..f2414c6c5e7c 100644
--- a/arch/x86/xen/enlighten_pv.c
+++ b/arch/x86/xen/enlighten_pv.c
@@ -622,7 +622,7 @@ static struct trap_array_entry trap_array[] = {
{ simd_coprocessor_error, xen_simd_coprocessor_error, false },
};
-static bool get_trap_addr(void **addr, unsigned int ist)
+static bool __ref get_trap_addr(void **addr, unsigned int ist)
{
unsigned int nr;
bool ist_okay = false;
@@ -644,6 +644,14 @@ static bool get_trap_addr(void **addr, unsigned int ist)
}
}
+ if (nr == ARRAY_SIZE(trap_array) &&
+ *addr >= (void *)early_idt_handler_array[0] &&
+ *addr < (void *)early_idt_handler_array[NUM_EXCEPTION_VECTORS]) {
+ nr = (*addr - (void *)early_idt_handler_array[0]) /
+ EARLY_IDT_HANDLER_SIZE;
+ *addr = (void *)xen_early_idt_handler_array[nr];
+ }
+
if (WARN_ON(ist != 0 && !ist_okay))
return false;
@@ -1262,6 +1270,21 @@ asmlinkage __visible void __init xen_start_kernel(void)
xen_setup_gdt(0);
xen_init_irq_ops();
+
+ /* Let's presume PV guests always boot on vCPU with id 0. */
+ per_cpu(xen_vcpu_id, 0) = 0;
+
+ /*
+ * Setup xen_vcpu early because idt_setup_early_handler needs it for
+ * local_irq_disable(), irqs_disabled().
+ *
+ * Don't do the full vcpu_info placement stuff until we have
+ * the cpu_possible_mask and a non-dummy shared_info.
+ */
+ xen_vcpu_info_reset(0);
+
+ idt_setup_early_handler();
+
xen_init_capabilities();
#ifdef CONFIG_X86_LOCAL_APIC
@@ -1295,18 +1318,6 @@ asmlinkage __visible void __init xen_start_kernel(void)
*/
acpi_numa = -1;
#endif
- /* Let's presume PV guests always boot on vCPU with id 0. */
- per_cpu(xen_vcpu_id, 0) = 0;
-
- /*
- * Setup xen_vcpu early because start_kernel needs it for
- * local_irq_disable(), irqs_disabled().
- *
- * Don't do the full vcpu_info placement stuff until we have
- * the cpu_possible_mask and a non-dummy shared_info.
- */
- xen_vcpu_info_reset(0);
-
WARN_ON(xen_cpuhp_setup(xen_cpu_up_prepare_pv, xen_cpu_dead_pv));
local_irq_disable();
diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
index 8a10c9a9e2b5..417b339e5c8e 100644
--- a/arch/x86/xen/xen-asm_64.S
+++ b/arch/x86/xen/xen-asm_64.S
@@ -15,6 +15,7 @@
#include <xen/interface/xen.h>
+#include <linux/init.h>
#include <linux/linkage.h>
.macro xen_pv_trap name
@@ -54,6 +55,19 @@ xen_pv_trap entry_INT80_compat
#endif
xen_pv_trap hypervisor_callback
+ __INIT
+ENTRY(xen_early_idt_handler_array)
+ i = 0
+ .rept NUM_EXCEPTION_VECTORS
+ pop %rcx
+ pop %r11
+ jmp early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE
+ i = i + 1
+ .fill xen_early_idt_handler_array + i*XEN_EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
+ .endr
+END(xen_early_idt_handler_array)
+ __FINIT
+
hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32
/*
* Xen64 iret frame: