updating to mainline 4.14.12

config

@@ -1,6 +1,6 @@
 #
 # Automatically generated file; DO NOT EDIT.
-# Linux/x86_64 4.14.11-jakeday Kernel Configuration
+# Linux/x86_64 4.14.12-jakeday Kernel Configuration
 #
 CONFIG_64BIT=y
 CONFIG_X86_64=y

kernel/Makefile

@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 VERSION = 4
 PATCHLEVEL = 14
-SUBLEVEL = 11
+SUBLEVEL = 12
 EXTRAVERSION =
 NAME = Petit Gorille
 

kernel/arch/x86/entry/entry_64_compat.S

@@ -190,8 +190,13 @@ ENTRY(entry_SYSCALL_compat)
 	/* Interrupts are off on entry. */
 	swapgs
 
-	/* Stash user ESP and switch to the kernel stack. */
+	/* Stash user ESP */
 	movl	%esp, %r8d
+
+	/* Use %rsp as scratch reg. User ESP is stashed in r8 */
+	SWITCH_TO_KERNEL_CR3 scratch_reg=%rsp
+
+	/* Switch to the kernel stack */
 	movq	PER_CPU_VAR(cpu_current_top_of_stack), %rsp
 
 	/* Construct struct pt_regs on stack */
@@ -219,12 +224,6 @@ GLOBAL(entry_SYSCALL_compat_after_hwframe)
 	pushq   $0			/* pt_regs->r14 = 0 */
 	pushq   $0			/* pt_regs->r15 = 0 */
 
-	/*
-	 * We just saved %rdi so it is safe to clobber.  It is not
-	 * preserved during the C calls inside TRACE_IRQS_OFF anyway.
-	 */
-	SWITCH_TO_KERNEL_CR3 scratch_reg=%rdi
-
 	/*
 	 * User mode is traced as though IRQs are on, and SYSENTER
 	 * turned them off.

kernel/arch/x86/include/asm/unwind.h

@@ -56,18 +56,27 @@ void unwind_start(struct unwind_state *state, struct task_struct *task,
 
 #if defined(CONFIG_UNWINDER_ORC) || defined(CONFIG_UNWINDER_FRAME_POINTER)
 /*
- * WARNING: The entire pt_regs may not be safe to dereference.  In some cases,
+ * If 'partial' returns true, only the iret frame registers are valid.
- * only the iret frame registers are accessible.  Use with caution!
  */
-static inline struct pt_regs *unwind_get_entry_regs(struct unwind_state *state)
+static inline struct pt_regs *unwind_get_entry_regs(struct unwind_state *state,
+						    bool *partial)
 {
 	if (unwind_done(state))
 		return NULL;
 
+	if (partial) {
+#ifdef CONFIG_UNWINDER_ORC
+		*partial = !state->full_regs;
+#else
+		*partial = false;
+#endif
+	}
+
 	return state->regs;
 }
 #else
-static inline struct pt_regs *unwind_get_entry_regs(struct unwind_state *state)
+static inline struct pt_regs *unwind_get_entry_regs(struct unwind_state *state,
+						    bool *partial)
 {
 	return NULL;
 }

kernel/arch/x86/kernel/cpu/common.c

@@ -899,7 +899,7 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 
 	setup_force_cpu_cap(X86_FEATURE_ALWAYS);
 
-	/* Assume for now that ALL x86 CPUs are insecure */
+	if (c->x86_vendor != X86_VENDOR_AMD)
 		setup_force_cpu_bug(X86_BUG_CPU_INSECURE);
 
 	fpu__init_system(c);

kernel/arch/x86/kernel/dumpstack.c

@@ -76,11 +76,22 @@ void show_iret_regs(struct pt_regs *regs)
 		regs->sp, regs->flags);
 }
 
-static void show_regs_safe(struct stack_info *info, struct pt_regs *regs)
+static void show_regs_if_on_stack(struct stack_info *info, struct pt_regs *regs,
+				  bool partial)
 {
-	if (on_stack(info, regs, sizeof(*regs)))
+	/*
+	 * These on_stack() checks aren't strictly necessary: the unwind code
+	 * has already validated the 'regs' pointer.  The checks are done for
+	 * ordering reasons: if the registers are on the next stack, we don't
+	 * want to print them out yet.  Otherwise they'll be shown as part of
+	 * the wrong stack.  Later, when show_trace_log_lvl() switches to the
+	 * next stack, this function will be called again with the same regs so
+	 * they can be printed in the right context.
+	 */
+	if (!partial && on_stack(info, regs, sizeof(*regs))) {
 		__show_regs(regs, 0);
-	else if (on_stack(info, (void *)regs + IRET_FRAME_OFFSET,
+
+	} else if (partial && on_stack(info, (void *)regs + IRET_FRAME_OFFSET,
 				       IRET_FRAME_SIZE)) {
 		/*
 		 * When an interrupt or exception occurs in entry code, the
@@ -98,11 +109,13 @@ void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	struct stack_info stack_info = {0};
 	unsigned long visit_mask = 0;
 	int graph_idx = 0;
+	bool partial;
 
 	printk("%sCall Trace:\n", log_lvl);
 
 	unwind_start(&state, task, regs, stack);
 	stack = stack ? : get_stack_pointer(task, regs);
+	regs = unwind_get_entry_regs(&state, &partial);
 
 	/*
 	 * Iterate through the stacks, starting with the current stack pointer.
@@ -120,7 +133,7 @@ void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	 * - hardirq stack
 	 * - entry stack
 	 */
-	for (regs = NULL; stack; stack = PTR_ALIGN(stack_info.next_sp, sizeof(long))) {
+	for ( ; stack; stack = PTR_ALIGN(stack_info.next_sp, sizeof(long))) {
 		const char *stack_name;
 
 		if (get_stack_info(stack, task, &stack_info, &visit_mask)) {
@@ -140,7 +153,7 @@ void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 			printk("%s <%s>\n", log_lvl, stack_name);
 
 		if (regs)
-			show_regs_safe(&stack_info, regs);
+			show_regs_if_on_stack(&stack_info, regs, partial);
 
 		/*
 		 * Scan the stack, printing any text addresses we find.  At the
@@ -164,7 +177,7 @@ void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 
 			/*
 			 * Don't print regs->ip again if it was already printed
-			 * by show_regs_safe() below.
+			 * by show_regs_if_on_stack().
 			 */
 			if (regs && stack == &regs->ip)
 				goto next;
@@ -199,9 +212,9 @@ next:
 			unwind_next_frame(&state);
 
 			/* if the frame has entry regs, print them */
-			regs = unwind_get_entry_regs(&state);
+			regs = unwind_get_entry_regs(&state, &partial);
 			if (regs)
-				show_regs_safe(&stack_info, regs);
+				show_regs_if_on_stack(&stack_info, regs, partial);
 		}
 
 		if (stack_name)

kernel/arch/x86/kernel/process.c

@@ -47,7 +47,7 @@
  * section. Since TSS's are completely CPU-local, we want them
  * on exact cacheline boundaries, to eliminate cacheline ping-pong.
  */
-__visible DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, cpu_tss_rw) = {
+__visible DEFINE_PER_CPU_PAGE_ALIGNED(struct tss_struct, cpu_tss_rw) = {
 	.x86_tss = {
 		/*
 		 * .sp0 is only used when entering ring 0 from a lower

kernel/arch/x86/kernel/stacktrace.c

@@ -98,7 +98,7 @@ static int __save_stack_trace_reliable(struct stack_trace *trace,
 	for (unwind_start(&state, task, NULL, NULL); !unwind_done(&state);
 	     unwind_next_frame(&state)) {
 
-		regs = unwind_get_entry_regs(&state);
+		regs = unwind_get_entry_regs(&state, NULL);
 		if (regs) {
 			/*
 			 * Kernel mode registers on the stack indicate an

kernel/arch/x86/mm/pti.c

@@ -367,7 +367,8 @@ static void __init pti_setup_espfix64(void)
 static void __init pti_clone_entry_text(void)
 {
 	pti_clone_pmds((unsigned long) __entry_text_start,
-			(unsigned long) __irqentry_text_end, _PAGE_RW);
+			(unsigned long) __irqentry_text_end,
+		       _PAGE_RW | _PAGE_GLOBAL);
 }
 
 /*

kernel/drivers/rtc/rtc-m41t80.c

@@ -154,6 +154,8 @@ struct m41t80_data {
 	struct rtc_device *rtc;
 #ifdef CONFIG_COMMON_CLK
 	struct clk_hw sqw;
+	unsigned long freq;
+	unsigned int sqwe;
 #endif
 };
 
@@ -443,43 +445,40 @@ static SIMPLE_DEV_PM_OPS(m41t80_pm, m41t80_suspend, m41t80_resume);
 #ifdef CONFIG_COMMON_CLK
 #define sqw_to_m41t80_data(_hw) container_of(_hw, struct m41t80_data, sqw)
 
-static unsigned long m41t80_sqw_recalc_rate(struct clk_hw *hw,
+static unsigned long m41t80_decode_freq(int setting)
-					    unsigned long parent_rate)
+{
+	return (setting == 0) ? 0 : (setting == 1) ? M41T80_SQW_MAX_FREQ :
+		M41T80_SQW_MAX_FREQ >> setting;
+}
+
+static unsigned long m41t80_get_freq(struct m41t80_data *m41t80)
 {
-	struct m41t80_data *m41t80 = sqw_to_m41t80_data(hw);
 	struct i2c_client *client = m41t80->client;
 	int reg_sqw = (m41t80->features & M41T80_FEATURE_SQ_ALT) ?
 		M41T80_REG_WDAY : M41T80_REG_SQW;
 	int ret = i2c_smbus_read_byte_data(client, reg_sqw);
-	unsigned long val = M41T80_SQW_MAX_FREQ;
 
 	if (ret < 0)
 		return 0;
+	return m41t80_decode_freq(ret >> 4);
+}
 
-	ret >>= 4;
+static unsigned long m41t80_sqw_recalc_rate(struct clk_hw *hw,
-	if (ret == 0)
+					    unsigned long parent_rate)
-		val = 0;
+{
-	else if (ret > 1)
+	return sqw_to_m41t80_data(hw)->freq;
-		val = val / (1 << ret);
-
-	return val;
 }
 
 static long m41t80_sqw_round_rate(struct clk_hw *hw, unsigned long rate,
 				  unsigned long *prate)
 {
-	int i, freq = M41T80_SQW_MAX_FREQ;
+	if (rate >= M41T80_SQW_MAX_FREQ)
-
+		return M41T80_SQW_MAX_FREQ;
-	if (freq <= rate)
+	if (rate >= M41T80_SQW_MAX_FREQ / 4)
-		return freq;
+		return M41T80_SQW_MAX_FREQ / 4;
-
+	if (!rate)
-	for (i = 2; i <= ilog2(M41T80_SQW_MAX_FREQ); i++) {
-		freq /= 1 << i;
-		if (freq <= rate)
-			return freq;
-	}
-
 		return 0;
+	return 1 << ilog2(rate);
 }
 
 static int m41t80_sqw_set_rate(struct clk_hw *hw, unsigned long rate,
@@ -491,17 +490,12 @@ static int m41t80_sqw_set_rate(struct clk_hw *hw, unsigned long rate,
 		M41T80_REG_WDAY : M41T80_REG_SQW;
 	int reg, ret, val = 0;
 
-	if (rate) {
+	if (rate >= M41T80_SQW_MAX_FREQ)
-		if (!is_power_of_2(rate))
-			return -EINVAL;
-		val = ilog2(rate);
-		if (val == ilog2(M41T80_SQW_MAX_FREQ))
 		val = 1;
-		else if (val < (ilog2(M41T80_SQW_MAX_FREQ) - 1))
+	else if (rate >= M41T80_SQW_MAX_FREQ / 4)
-			val = ilog2(M41T80_SQW_MAX_FREQ) - val;
+		val = 2;
-		else
+	else if (rate)
-			return -EINVAL;
+		val = 15 - ilog2(rate);
-	}
 
 	reg = i2c_smbus_read_byte_data(client, reg_sqw);
 	if (reg < 0)
@@ -510,10 +504,9 @@ static int m41t80_sqw_set_rate(struct clk_hw *hw, unsigned long rate,
 	reg = (reg & 0x0f) | (val << 4);
 
 	ret = i2c_smbus_write_byte_data(client, reg_sqw, reg);
-	if (ret < 0)
+	if (!ret)
+		m41t80->freq = m41t80_decode_freq(val);
 	return ret;
-
-	return -EINVAL;
 }
 
 static int m41t80_sqw_control(struct clk_hw *hw, bool enable)
@@ -530,7 +523,10 @@ static int m41t80_sqw_control(struct clk_hw *hw, bool enable)
 	else
 		ret &= ~M41T80_ALMON_SQWE;
 
-	return i2c_smbus_write_byte_data(client, M41T80_REG_ALARM_MON, ret);
+	ret = i2c_smbus_write_byte_data(client, M41T80_REG_ALARM_MON, ret);
+	if (!ret)
+		m41t80->sqwe = enable;
+	return ret;
 }
 
 static int m41t80_sqw_prepare(struct clk_hw *hw)
@@ -545,14 +541,7 @@ static void m41t80_sqw_unprepare(struct clk_hw *hw)
 
 static int m41t80_sqw_is_prepared(struct clk_hw *hw)
 {
-	struct m41t80_data *m41t80 = sqw_to_m41t80_data(hw);
+	return sqw_to_m41t80_data(hw)->sqwe;
-	struct i2c_client *client = m41t80->client;
-	int ret = i2c_smbus_read_byte_data(client, M41T80_REG_ALARM_MON);
-
-	if (ret < 0)
-		return ret;
-
-	return !!(ret & M41T80_ALMON_SQWE);
 }
 
 static const struct clk_ops m41t80_sqw_ops = {
@@ -587,6 +576,7 @@ static struct clk *m41t80_sqw_register_clk(struct m41t80_data *m41t80)
 	init.parent_names = NULL;
 	init.num_parents = 0;
 	m41t80->sqw.init = &init;
+	m41t80->freq = m41t80_get_freq(m41t80);
 
 	/* optional override of the clockname */
 	of_property_read_string(node, "clock-output-names", &init.name);

kernel/fs/exec.c

@@ -1350,9 +1350,14 @@ void setup_new_exec(struct linux_binprm * bprm)
 
 	current->sas_ss_sp = current->sas_ss_size = 0;
 
-	/* Figure out dumpability. */
+	/*
+	 * Figure out dumpability. Note that this checking only of current
+	 * is wrong, but userspace depends on it. This should be testing
+	 * bprm->secureexec instead.
+	 */
 	if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP ||
-	    bprm->secureexec)
+	    !(uid_eq(current_euid(), current_uid()) &&
+	      gid_eq(current_egid(), current_gid())))
 		set_dumpable(current->mm, suid_dumpable);
 	else
 		set_dumpable(current->mm, SUID_DUMP_USER);

kernel/net/xfrm/xfrm_policy.c

@@ -1362,14 +1362,18 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl,
 	struct net *net = xp_net(policy);
 	int nx;
 	int i, error;
+	xfrm_address_t *daddr = xfrm_flowi_daddr(fl, family);
+	xfrm_address_t *saddr = xfrm_flowi_saddr(fl, family);
 	xfrm_address_t tmp;
 
 	for (nx = 0, i = 0; i < policy->xfrm_nr; i++) {
 		struct xfrm_state *x;
-		xfrm_address_t *local;
+		xfrm_address_t *remote = daddr;
-		xfrm_address_t *remote;
+		xfrm_address_t *local  = saddr;
 		struct xfrm_tmpl *tmpl = &policy->xfrm_vec[i];
 
+		if (tmpl->mode == XFRM_MODE_TUNNEL ||
+		    tmpl->mode == XFRM_MODE_BEET) {
 			remote = &tmpl->id.daddr;
 			local = &tmpl->saddr;
 			if (xfrm_addr_any(local, tmpl->encap_family)) {
@@ -1380,11 +1384,14 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl,
 					goto fail;
 				local = &tmp;
 			}
+		}
 
 		x = xfrm_state_find(remote, local, fl, tmpl, policy, &error, family);
 
 		if (x && x->km.state == XFRM_STATE_VALID) {
 			xfrm[nx++] = x;
+			daddr = remote;
+			saddr = local;
 			continue;
 		}
 		if (x) {

kernel/security/commoncap.c

@@ -348,21 +348,18 @@ static __u32 sansflags(__u32 m)
 	return m & ~VFS_CAP_FLAGS_EFFECTIVE;
 }
 
-static bool is_v2header(size_t size, __le32 magic)
+static bool is_v2header(size_t size, const struct vfs_cap_data *cap)
 {
-	__u32 m = le32_to_cpu(magic);
 	if (size != XATTR_CAPS_SZ_2)
 		return false;
-	return sansflags(m) == VFS_CAP_REVISION_2;
+	return sansflags(le32_to_cpu(cap->magic_etc)) == VFS_CAP_REVISION_2;
 }
 
-static bool is_v3header(size_t size, __le32 magic)
+static bool is_v3header(size_t size, const struct vfs_cap_data *cap)
 {
-	__u32 m = le32_to_cpu(magic);
-
 	if (size != XATTR_CAPS_SZ_3)
 		return false;
-	return sansflags(m) == VFS_CAP_REVISION_3;
+	return sansflags(le32_to_cpu(cap->magic_etc)) == VFS_CAP_REVISION_3;
 }
 
 /*
@@ -405,7 +402,7 @@ int cap_inode_getsecurity(struct inode *inode, const char *name, void **buffer,
 
 	fs_ns = inode->i_sb->s_user_ns;
 	cap = (struct vfs_cap_data *) tmpbuf;
-	if (is_v2header((size_t) ret, cap->magic_etc)) {
+	if (is_v2header((size_t) ret, cap)) {
 		/* If this is sizeof(vfs_cap_data) then we're ok with the
 		 * on-disk value, so return that.  */
 		if (alloc)
@@ -413,7 +410,7 @@ int cap_inode_getsecurity(struct inode *inode, const char *name, void **buffer,
 		else
 			kfree(tmpbuf);
 		return ret;
-	} else if (!is_v3header((size_t) ret, cap->magic_etc)) {
+	} else if (!is_v3header((size_t) ret, cap)) {
 		kfree(tmpbuf);
 		return -EINVAL;
 	}
@@ -470,9 +467,9 @@ static kuid_t rootid_from_xattr(const void *value, size_t size,
 	return make_kuid(task_ns, rootid);
 }
 
-static bool validheader(size_t size, __le32 magic)
+static bool validheader(size_t size, const struct vfs_cap_data *cap)
 {
-	return is_v2header(size, magic) || is_v3header(size, magic);
+	return is_v2header(size, cap) || is_v3header(size, cap);
 }
 
 /*
@@ -495,7 +492,7 @@ int cap_convert_nscap(struct dentry *dentry, void **ivalue, size_t size)
 
 	if (!*ivalue)
 		return -EINVAL;
-	if (!validheader(size, cap->magic_etc))
+	if (!validheader(size, cap))
 		return -EINVAL;
 	if (!capable_wrt_inode_uidgid(inode, CAP_SETFCAP))
 		return -EPERM;