X-Git-Url: https://git.proxmox.com/?a=blobdiff_plain;f=arch%2Fx86%2Finclude%2Fasm%2Ftlbflush.h;h=4a08dd2ab32ade77dcbfff6a8e8b77f5d5e43764;hb=52c90f2d32bfa7d6eccd66a56c44ace1f78fbadd;hp=5dcc38b16604d9da0f33db9d697e566670f58b11;hpb=6fd166aae78c0ab738d49bda653cbd9e3b1491cf;p=mirror_ubuntu-bionic-kernel.git

diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 5dcc38b16604..4a08dd2ab32a 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -13,16 +13,33 @@
 #include <asm/pti.h>
 #include <asm/processor-flags.h>
 
-static inline u64 inc_mm_tlb_gen(struct mm_struct *mm)
-{
-	/*
-	 * Bump the generation count.  This also serves as a full barrier
-	 * that synchronizes with switch_mm(): callers are required to order
-	 * their read of mm_cpumask after their writes to the paging
-	 * structures.
-	 */
-	return atomic64_inc_return(&mm->context.tlb_gen);
-}
+/*
+ * 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
@@ -41,7 +58,7 @@ static inline u64 inc_mm_tlb_gen(struct mm_struct *mm)
 
 /*
  * ASIDs are zero-based: 0->MAX_AVAIL_ASID are valid.  -1 below to account
- * for them being zero-based.  Another -1 is because ASID 0 is reserved for
+ * 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)
@@ -52,6 +69,9 @@ static inline u64 inc_mm_tlb_gen(struct mm_struct *mm)
  */
 #define TLB_NR_DYN_ASIDS	6
 
+/*
+ * Given @asid, compute kPCID
+ */
 static inline u16 kern_pcid(u16 asid)
 {
 	VM_WARN_ON_ONCE(asid > MAX_ASID_AVAILABLE);
@@ -85,6 +105,18 @@ static inline u16 kern_pcid(u16 asid)
 	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_SWITCH_BIT;
+#endif
+	return ret;
+}
+
 struct pgd_t;
 static inline unsigned long build_cr3(pgd_t *pgd, u16 asid)
 {
@@ -213,40 +245,43 @@ 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. */
@@ -313,15 +348,17 @@ static inline void invalidate_user_asid(u16 asid)
  */
 static inline void __native_flush_tlb(void)
 {
-	invalidate_user_asid(this_cpu_read(cpu_tlbstate.loaded_mm_asid));
 	/*
-	 * If current->mm == NULL then we borrow a mm which may change
-	 * during a task switch and therefore we must not be preempted
-	 * while we write CR3 back:
+	 * 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().
 	 */
-	preempt_disable();
+	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());
-	preempt_enable();
 }
 
 /*
@@ -335,6 +372,8 @@ static inline void __native_flush_tlb_global(void)
 		/*
 		 * 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;
@@ -368,7 +407,14 @@ static inline void __native_flush_tlb_single(unsigned long addr)
 	if (!static_cpu_has(X86_FEATURE_PTI))
 		return;
 
-	invalidate_user_asid(loaded_mm_asid);
+	/*
+	 * 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);
 }
 
 /*
@@ -463,6 +509,17 @@ static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long a)
 void native_flush_tlb_others(const struct cpumask *cpumask,
 			     const struct flush_tlb_info *info);
 
+static inline u64 inc_mm_tlb_gen(struct mm_struct *mm)
+{
+	/*
+	 * Bump the generation count.  This also serves as a full barrier
+	 * that synchronizes with switch_mm(): callers are required to order
+	 * their read of mm_cpumask after their writes to the paging
+	 * structures.
+	 */
+	return atomic64_inc_return(&mm->context.tlb_gen);
+}
+
 static inline void arch_tlbbatch_add_mm(struct arch_tlbflush_unmap_batch *batch,
 					struct mm_struct *mm)
 {