Merge tag 'x86-mm-2020-06-05' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 mm updates from Ingo Molnar:
 "Misc changes:

   - Unexport various PAT primitives

   - Unexport per-CPU tlbstate and uninline TLB helpers"

* tag 'x86-mm-2020-06-05' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
  x86/tlb/uv: Add a forward declaration for struct flush_tlb_info
  x86/cpu: Export native_write_cr4() only when CONFIG_LKTDM=m
  x86/tlb: Restrict access to tlbstate
  xen/privcmd: Remove unneeded asm/tlb.h include
  x86/tlb: Move PCID helpers where they are used
  x86/tlb: Uninline nmi_uaccess_okay()
  x86/tlb: Move cr4_set_bits_and_update_boot() to the usage site
  x86/tlb: Move paravirt_tlb_remove_table() to the usage site
  x86/tlb: Move __flush_tlb_all() out of line
  x86/tlb: Move flush_tlb_others() out of line
  x86/tlb: Move __flush_tlb_one_kernel() out of line
  x86/tlb: Move __flush_tlb_one_user() out of line
  x86/tlb: Move __flush_tlb_global() out of line
  x86/tlb: Move __flush_tlb() out of line
  x86/alternatives: Move temporary_mm helpers into C
  x86/cr4: Sanitize CR4.PCE update
  x86/cpu: Uninline CR4 accessors
  x86/tlb: Uninline __get_current_cr3_fast()
  x86/mm: Use pgprotval_t in protval_4k_2_large() and protval_large_2_4k()
  x86/mm: Unexport __cachemode2pte_tbl
  ...

1 files changed, 378 insertions, 6 deletions

diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index f3fe261e5936..1a3569b43aa5 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -18,6 +18,16 @@
 
 #include "mm_internal.h"
 
+#ifdef CONFIG_PARAVIRT
+# define STATIC_NOPV
+#else
+# define STATIC_NOPV			static
+# define __flush_tlb_local		native_flush_tlb_local
+# define __flush_tlb_global		native_flush_tlb_global
+# define __flush_tlb_one_user(addr)	native_flush_tlb_one_user(addr)
+# define __flush_tlb_others(msk, info)	native_flush_tlb_others(msk, info)
+#endif
+
 /*
  *	TLB flushing, formerly SMP-only
  *		c/o Linus Torvalds.
@@ -39,6 +49,126 @@
 #define LAST_USER_MM_IBPB	0x1UL
 
 /*
+ * The x86 feature is called PCID (Process Context IDentifier). It is similar
+ * to what is traditionally called ASID on the RISC processors.
+ *
+ * We don't use the traditional ASID implementation, where each process/mm gets
+ * its own ASID and flush/restart when we run out of ASID space.
+ *
+ * Instead we have a small per-cpu array of ASIDs and cache the last few mm's
+ * that came by on this CPU, allowing cheaper switch_mm between processes on
+ * this CPU.
+ *
+ * We end up with different spaces for different things. To avoid confusion we
+ * use different names for each of them:
+ *
+ * ASID  - [0, TLB_NR_DYN_ASIDS-1]
+ *         the canonical identifier for an mm
+ *
+ * kPCID - [1, TLB_NR_DYN_ASIDS]
+ *         the value we write into the PCID part of CR3; corresponds to the
+ *         ASID+1, because PCID 0 is special.
+ *
+ * uPCID - [2048 + 1, 2048 + TLB_NR_DYN_ASIDS]
+ *         for KPTI each mm has two address spaces and thus needs two
+ *         PCID values, but we can still do with a single ASID denomination
+ *         for each mm. Corresponds to kPCID + 2048.
+ *
+ */
+
+/* There are 12 bits of space for ASIDS in CR3 */
+#define CR3_HW_ASID_BITS		12
+
+/*
+ * When enabled, PAGE_TABLE_ISOLATION consumes a single bit for
+ * user/kernel switches
+ */
+#ifdef CONFIG_PAGE_TABLE_ISOLATION
+# define PTI_CONSUMED_PCID_BITS	1
+#else
+# define PTI_CONSUMED_PCID_BITS	0
+#endif
+
+#define CR3_AVAIL_PCID_BITS (X86_CR3_PCID_BITS - PTI_CONSUMED_PCID_BITS)
+
+/*
+ * ASIDs are zero-based: 0->MAX_AVAIL_ASID are valid.  -1 below to account
+ * for them being zero-based.  Another -1 is because PCID 0 is reserved for
+ * use by non-PCID-aware users.
+ */
+#define MAX_ASID_AVAILABLE ((1 << CR3_AVAIL_PCID_BITS) - 2)
+
+/*
+ * Given @asid, compute kPCID
+ */
+static inline u16 kern_pcid(u16 asid)
+{
+	VM_WARN_ON_ONCE(asid > MAX_ASID_AVAILABLE);
+
+#ifdef CONFIG_PAGE_TABLE_ISOLATION
+	/*
+	 * Make sure that the dynamic ASID space does not confict with the
+	 * bit we are using to switch between user and kernel ASIDs.
+	 */
+	BUILD_BUG_ON(TLB_NR_DYN_ASIDS >= (1 << X86_CR3_PTI_PCID_USER_BIT));
+
+	/*
+	 * The ASID being passed in here should have respected the
+	 * MAX_ASID_AVAILABLE and thus never have the switch bit set.
+	 */
+	VM_WARN_ON_ONCE(asid & (1 << X86_CR3_PTI_PCID_USER_BIT));
+#endif
+	/*
+	 * The dynamically-assigned ASIDs that get passed in are small
+	 * (<TLB_NR_DYN_ASIDS).  They never have the high switch bit set,
+	 * so do not bother to clear it.
+	 *
+	 * If PCID is on, ASID-aware code paths put the ASID+1 into the
+	 * PCID bits.  This serves two purposes.  It prevents a nasty
+	 * situation in which PCID-unaware code saves CR3, loads some other
+	 * value (with PCID == 0), and then restores CR3, thus corrupting
+	 * the TLB for ASID 0 if the saved ASID was nonzero.  It also means
+	 * that any bugs involving loading a PCID-enabled CR3 with
+	 * CR4.PCIDE off will trigger deterministically.
+	 */
+	return asid + 1;
+}
+
+/*
+ * Given @asid, compute uPCID
+ */
+static inline u16 user_pcid(u16 asid)
+{
+	u16 ret = kern_pcid(asid);
+#ifdef CONFIG_PAGE_TABLE_ISOLATION
+	ret |= 1 << X86_CR3_PTI_PCID_USER_BIT;
+#endif
+	return ret;
+}
+
+static inline unsigned long build_cr3(pgd_t *pgd, u16 asid)
+{
+	if (static_cpu_has(X86_FEATURE_PCID)) {
+		return __sme_pa(pgd) | kern_pcid(asid);
+	} else {
+		VM_WARN_ON_ONCE(asid != 0);
+		return __sme_pa(pgd);
+	}
+}
+
+static inline unsigned long build_cr3_noflush(pgd_t *pgd, u16 asid)
+{
+	VM_WARN_ON_ONCE(asid > MAX_ASID_AVAILABLE);
+	/*
+	 * Use boot_cpu_has() instead of this_cpu_has() as this function
+	 * might be called during early boot. This should work even after
+	 * boot because all CPU's the have same capabilities:
+	 */
+	VM_WARN_ON_ONCE(!boot_cpu_has(X86_FEATURE_PCID));
+	return __sme_pa(pgd) | kern_pcid(asid) | CR3_NOFLUSH;
+}
+
+/*
  * We get here when we do something requiring a TLB invalidation
  * but could not go invalidate all of the contexts.  We do the
  * necessary invalidation by clearing out the 'ctx_id' which
@@ -110,6 +240,32 @@ static void choose_new_asid(struct mm_struct *next, u64 next_tlb_gen,
 	*need_flush = true;
 }
 
+/*
+ * Given an ASID, flush the corresponding user ASID.  We can delay this
+ * until the next time we switch to it.
+ *
+ * See SWITCH_TO_USER_CR3.
+ */
+static inline void invalidate_user_asid(u16 asid)
+{
+	/* There is no user ASID if address space separation is off */
+	if (!IS_ENABLED(CONFIG_PAGE_TABLE_ISOLATION))
+		return;
+
+	/*
+	 * We only have a single ASID if PCID is off and the CR3
+	 * write will have flushed it.
+	 */
+	if (!cpu_feature_enabled(X86_FEATURE_PCID))
+		return;
+
+	if (!static_cpu_has(X86_FEATURE_PTI))
+		return;
+
+	__set_bit(kern_pcid(asid),
+		  (unsigned long *)this_cpu_ptr(&cpu_tlbstate.user_pcid_flush_mask));
+}
+
 static void load_new_mm_cr3(pgd_t *pgdir, u16 new_asid, bool need_flush)
 {
 	unsigned long new_mm_cr3;
@@ -244,6 +400,26 @@ static void cond_ibpb(struct task_struct *next)
 	}
 }
 
+#ifdef CONFIG_PERF_EVENTS
+static inline void cr4_update_pce_mm(struct mm_struct *mm)
+{
+	if (static_branch_unlikely(&rdpmc_always_available_key) ||
+	    (!static_branch_unlikely(&rdpmc_never_available_key) &&
+	     atomic_read(&mm->context.perf_rdpmc_allowed)))
+		cr4_set_bits_irqsoff(X86_CR4_PCE);
+	else
+		cr4_clear_bits_irqsoff(X86_CR4_PCE);
+}
+
+void cr4_update_pce(void *ignored)
+{
+	cr4_update_pce_mm(this_cpu_read(cpu_tlbstate.loaded_mm));
+}
+
+#else
+static inline void cr4_update_pce_mm(struct mm_struct *mm) { }
+#endif
+
 void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 			struct task_struct *tsk)
 {
@@ -403,7 +579,7 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 	this_cpu_write(cpu_tlbstate.loaded_mm_asid, new_asid);
 
 	if (next != real_prev) {
-		load_mm_cr4_irqsoff(next);
+		cr4_update_pce_mm(next);
 		switch_ldt(real_prev, next);
 	}
 }
@@ -580,7 +756,7 @@ static void flush_tlb_func_common(const struct flush_tlb_info *f,
 		unsigned long addr = f->start;
 
 		while (addr < f->end) {
-			__flush_tlb_one_user(addr);
+			flush_tlb_one_user(addr);
 			addr += 1UL << f->stride_shift;
 		}
 		if (local)
@@ -588,7 +764,7 @@ static void flush_tlb_func_common(const struct flush_tlb_info *f,
 		trace_tlb_flush(reason, nr_invalidate);
 	} else {
 		/* Full flush. */
-		local_flush_tlb();
+		flush_tlb_local();
 		if (local)
 			count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ALL);
 		trace_tlb_flush(reason, TLB_FLUSH_ALL);
@@ -623,8 +799,8 @@ static bool tlb_is_not_lazy(int cpu, void *data)
 	return !per_cpu(cpu_tlbstate.is_lazy, cpu);
 }
 
-void native_flush_tlb_others(const struct cpumask *cpumask,
-			     const struct flush_tlb_info *info)
+STATIC_NOPV void native_flush_tlb_others(const struct cpumask *cpumask,
+					 const struct flush_tlb_info *info)
 {
 	count_vm_tlb_event(NR_TLB_REMOTE_FLUSH);
 	if (info->end == TLB_FLUSH_ALL)
@@ -674,6 +850,12 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
 				(void *)info, 1, cpumask);
 }
 
+void flush_tlb_others(const struct cpumask *cpumask,
+		      const struct flush_tlb_info *info)
+{
+	__flush_tlb_others(cpumask, info);
+}
+
 /*
  * See Documentation/x86/tlb.rst for details.  We choose 33
  * because it is large enough to cover the vast majority (at
@@ -784,7 +966,7 @@ static void do_kernel_range_flush(void *info)
 
 	/* flush range by one by one 'invlpg' */
 	for (addr = f->start; addr < f->end; addr += PAGE_SIZE)
-		__flush_tlb_one_kernel(addr);
+		flush_tlb_one_kernel(addr);
 }
 
 void flush_tlb_kernel_range(unsigned long start, unsigned long end)
@@ -807,6 +989,164 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 }
 
 /*
+ * This can be used from process context to figure out what the value of
+ * CR3 is without needing to do a (slow) __read_cr3().
+ *
+ * It's intended to be used for code like KVM that sneakily changes CR3
+ * and needs to restore it.  It needs to be used very carefully.
+ */
+unsigned long __get_current_cr3_fast(void)
+{
+	unsigned long cr3 = build_cr3(this_cpu_read(cpu_tlbstate.loaded_mm)->pgd,
+		this_cpu_read(cpu_tlbstate.loaded_mm_asid));
+
+	/* For now, be very restrictive about when this can be called. */
+	VM_WARN_ON(in_nmi() || preemptible());
+
+	VM_BUG_ON(cr3 != __read_cr3());
+	return cr3;
+}
+EXPORT_SYMBOL_GPL(__get_current_cr3_fast);
+
+/*
+ * Flush one page in the kernel mapping
+ */
+void flush_tlb_one_kernel(unsigned long addr)
+{
+	count_vm_tlb_event(NR_TLB_LOCAL_FLUSH_ONE);
+
+	/*
+	 * If PTI is off, then __flush_tlb_one_user() is just INVLPG or its
+	 * paravirt equivalent.  Even with PCID, this is sufficient: we only
+	 * use PCID if we also use global PTEs for the kernel mapping, and
+	 * INVLPG flushes global translations across all address spaces.
+	 *
+	 * If PTI is on, then the kernel is mapped with non-global PTEs, and
+	 * __flush_tlb_one_user() will flush the given address for the current
+	 * kernel address space and for its usermode counterpart, but it does
+	 * not flush it for other address spaces.
+	 */
+	flush_tlb_one_user(addr);
+
+	if (!static_cpu_has(X86_FEATURE_PTI))
+		return;
+
+	/*
+	 * See above.  We need to propagate the flush to all other address
+	 * spaces.  In principle, we only need to propagate it to kernelmode
+	 * address spaces, but the extra bookkeeping we would need is not
+	 * worth it.
+	 */
+	this_cpu_write(cpu_tlbstate.invalidate_other, true);
+}
+
+/*
+ * Flush one page in the user mapping
+ */
+STATIC_NOPV void native_flush_tlb_one_user(unsigned long addr)
+{
+	u32 loaded_mm_asid = this_cpu_read(cpu_tlbstate.loaded_mm_asid);
+
+	asm volatile("invlpg (%0)" ::"r" (addr) : "memory");
+
+	if (!static_cpu_has(X86_FEATURE_PTI))
+		return;
+
+	/*
+	 * Some platforms #GP if we call invpcid(type=1/2) before CR4.PCIDE=1.
+	 * Just use invalidate_user_asid() in case we are called early.
+	 */
+	if (!this_cpu_has(X86_FEATURE_INVPCID_SINGLE))
+		invalidate_user_asid(loaded_mm_asid);
+	else
+		invpcid_flush_one(user_pcid(loaded_mm_asid), addr);
+}
+
+void flush_tlb_one_user(unsigned long addr)
+{
+	__flush_tlb_one_user(addr);
+}
+
+/*
+ * Flush everything
+ */
+STATIC_NOPV void native_flush_tlb_global(void)
+{
+	unsigned long cr4, flags;
+
+	if (static_cpu_has(X86_FEATURE_INVPCID)) {
+		/*
+		 * Using INVPCID is considerably faster than a pair of writes
+		 * to CR4 sandwiched inside an IRQ flag save/restore.
+		 *
+		 * Note, this works with CR4.PCIDE=0 or 1.
+		 */
+		invpcid_flush_all();
+		return;
+	}
+
+	/*
+	 * Read-modify-write to CR4 - protect it from preemption and
+	 * from interrupts. (Use the raw variant because this code can
+	 * be called from deep inside debugging code.)
+	 */
+	raw_local_irq_save(flags);
+
+	cr4 = this_cpu_read(cpu_tlbstate.cr4);
+	/* toggle PGE */
+	native_write_cr4(cr4 ^ X86_CR4_PGE);
+	/* write old PGE again and flush TLBs */
+	native_write_cr4(cr4);
+
+	raw_local_irq_restore(flags);
+}
+
+/*
+ * Flush the entire current user mapping
+ */
+STATIC_NOPV void native_flush_tlb_local(void)
+{
+	/*
+	 * Preemption or interrupts must be disabled to protect the access
+	 * to the per CPU variable and to prevent being preempted between
+	 * read_cr3() and write_cr3().
+	 */
+	WARN_ON_ONCE(preemptible());
+
+	invalidate_user_asid(this_cpu_read(cpu_tlbstate.loaded_mm_asid));
+
+	/* If current->mm == NULL then the read_cr3() "borrows" an mm */
+	native_write_cr3(__native_read_cr3());
+}
+
+void flush_tlb_local(void)
+{
+	__flush_tlb_local();
+}
+
+/*
+ * Flush everything
+ */
+void __flush_tlb_all(void)
+{
+	/*
+	 * This is to catch users with enabled preemption and the PGE feature
+	 * and don't trigger the warning in __native_flush_tlb().
+	 */
+	VM_WARN_ON_ONCE(preemptible());
+
+	if (boot_cpu_has(X86_FEATURE_PGE)) {
+		__flush_tlb_global();
+	} else {
+		/*
+		 * !PGE -> !PCID (setup_pcid()), thus every flush is total.
+		 */
+		flush_tlb_local();
+	}
+}
+EXPORT_SYMBOL_GPL(__flush_tlb_all);
+
+/*
  * arch_tlbbatch_flush() performs a full TLB flush regardless of the active mm.
  * This means that the 'struct flush_tlb_info' that describes which mappings to
  * flush is actually fixed. We therefore set a single fixed struct and use it in
@@ -837,6 +1177,38 @@ void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch)
 	put_cpu();
 }
 
+/*
+ * Blindly accessing user memory from NMI context can be dangerous
+ * if we're in the middle of switching the current user task or
+ * switching the loaded mm.  It can also be dangerous if we
+ * interrupted some kernel code that was temporarily using a
+ * different mm.
+ */
+bool nmi_uaccess_okay(void)
+{
+	struct mm_struct *loaded_mm = this_cpu_read(cpu_tlbstate.loaded_mm);
+	struct mm_struct *current_mm = current->mm;
+
+	VM_WARN_ON_ONCE(!loaded_mm);
+
+	/*
+	 * The condition we want to check is
+	 * current_mm->pgd == __va(read_cr3_pa()).  This may be slow, though,
+	 * if we're running in a VM with shadow paging, and nmi_uaccess_okay()
+	 * is supposed to be reasonably fast.
+	 *
+	 * Instead, we check the almost equivalent but somewhat conservative
+	 * condition below, and we rely on the fact that switch_mm_irqs_off()
+	 * sets loaded_mm to LOADED_MM_SWITCHING before writing to CR3.
+	 */
+	if (loaded_mm != current_mm)
+		return false;
+
+	VM_WARN_ON_ONCE(current_mm->pgd != __va(read_cr3_pa()));
+
+	return true;
+}
+
 static ssize_t tlbflush_read_file(struct file *file, char __user *user_buf,
 			     size_t count, loff_t *ppos)
 {