4 files changed, 82 insertions, 13 deletions

diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 35f1498e9832..6e3e8a124903 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -190,6 +190,53 @@ static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
 	return pmd_k;
 }
 
+/*
+ *   Handle a fault on the vmalloc or module mapping area
+ *
+ *   This is needed because there is a race condition between the time
+ *   when the vmalloc mapping code updates the PMD to the point in time
+ *   where it synchronizes this update with the other page-tables in the
+ *   system.
+ *
+ *   In this race window another thread/CPU can map an area on the same
+ *   PMD, finds it already present and does not synchronize it with the
+ *   rest of the system yet. As a result v[mz]alloc might return areas
+ *   which are not mapped in every page-table in the system, causing an
+ *   unhandled page-fault when they are accessed.
+ */
+static noinline int vmalloc_fault(unsigned long address)
+{
+	unsigned long pgd_paddr;
+	pmd_t *pmd_k;
+	pte_t *pte_k;
+
+	/* Make sure we are in vmalloc area: */
+	if (!(address >= VMALLOC_START && address < VMALLOC_END))
+		return -1;
+
+	/*
+	 * Synchronize this task's top level page-table
+	 * with the 'reference' page table.
+	 *
+	 * Do _not_ use "current" here. We might be inside
+	 * an interrupt in the middle of a task switch..
+	 */
+	pgd_paddr = read_cr3_pa();
+	pmd_k = vmalloc_sync_one(__va(pgd_paddr), address);
+	if (!pmd_k)
+		return -1;
+
+	if (pmd_large(*pmd_k))
+		return 0;
+
+	pte_k = pte_offset_kernel(pmd_k, address);
+	if (!pte_present(*pte_k))
+		return -1;
+
+	return 0;
+}
+NOKPROBE_SYMBOL(vmalloc_fault);
+
 void arch_sync_kernel_mappings(unsigned long start, unsigned long end)
 {
 	unsigned long addr;
@@ -1110,6 +1157,37 @@ do_kern_addr_fault(struct pt_regs *regs, unsigned long hw_error_code,
 	 */
 	WARN_ON_ONCE(hw_error_code & X86_PF_PK);
 
+#ifdef CONFIG_X86_32
+	/*
+	 * We can fault-in kernel-space virtual memory on-demand. The
+	 * 'reference' page table is init_mm.pgd.
+	 *
+	 * NOTE! We MUST NOT take any locks for this case. We may
+	 * be in an interrupt or a critical region, and should
+	 * only copy the information from the master page table,
+	 * nothing more.
+	 *
+	 * Before doing this on-demand faulting, ensure that the
+	 * fault is not any of the following:
+	 * 1. A fault on a PTE with a reserved bit set.
+	 * 2. A fault caused by a user-mode access.  (Do not demand-
+	 *    fault kernel memory due to user-mode accesses).
+	 * 3. A fault caused by a page-level protection violation.
+	 *    (A demand fault would be on a non-present page which
+	 *     would have X86_PF_PROT==0).
+	 *
+	 * This is only needed to close a race condition on x86-32 in
+	 * the vmalloc mapping/unmapping code. See the comment above
+	 * vmalloc_fault() for details. On x86-64 the race does not
+	 * exist as the vmalloc mappings don't need to be synchronized
+	 * there.
+	 */
+	if (!(hw_error_code & (X86_PF_RSVD | X86_PF_USER | X86_PF_PROT))) {
+		if (vmalloc_fault(address) >= 0)
+			return;
+	}
+#endif
+
 	/* Was the fault spurious, caused by lazy TLB invalidation? */
 	if (spurious_kernel_fault(hw_error_code, address))
 		return;
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 84d85dbd1dad..9e5ccc56f8e0 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -574,7 +574,7 @@ static bool memremap_should_map_decrypted(resource_size_t phys_addr,
 		/* For SEV, these areas are encrypted */
 		if (sev_active())
 			break;
-		/* Fallthrough */
+		fallthrough;
 
 	case E820_TYPE_PRAM:
 		return true;
diff --git a/arch/x86/mm/numa_emulation.c b/arch/x86/mm/numa_emulation.c
index c5174b4e318b..683cd12f4793 100644
--- a/arch/x86/mm/numa_emulation.c
+++ b/arch/x86/mm/numa_emulation.c
@@ -321,7 +321,7 @@ static int __init split_nodes_size_interleave(struct numa_meminfo *ei,
 					      u64 addr, u64 max_addr, u64 size)
 {
 	return split_nodes_size_interleave_uniform(ei, pi, addr, max_addr, size,
-			0, NULL, NUMA_NO_NODE);
+			0, NULL, 0);
 }
 
 static int __init setup_emu2phys_nid(int *dfl_phys_nid)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 1a3569b43aa5..0951b47e64c1 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -555,21 +555,12 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
 		this_cpu_write(cpu_tlbstate.ctxs[new_asid].tlb_gen, next_tlb_gen);
 		load_new_mm_cr3(next->pgd, new_asid, true);
 
-		/*
-		 * NB: This gets called via leave_mm() in the idle path
-		 * where RCU functions differently.  Tracing normally
-		 * uses RCU, so we need to use the _rcuidle variant.
-		 *
-		 * (There is no good reason for this.  The idle code should
-		 *  be rearranged to call this before rcu_idle_enter().)
-		 */
-		trace_tlb_flush_rcuidle(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
+		trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL);
 	} else {
 		/* The new ASID is already up to date. */
 		load_new_mm_cr3(next->pgd, new_asid, false);
 
-		/* See above wrt _rcuidle. */
-		trace_tlb_flush_rcuidle(TLB_FLUSH_ON_TASK_SWITCH, 0);
+		trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, 0);
 	}
 
 	/* Make sure we write CR3 before loaded_mm. */