Merge branch 'linux-linaro-lsk' into linux-linaro-lsk-android

Makefile

@@ -1,6 +1,6 @@
 VERSION = 3
 PATCHLEVEL = 10
-SUBLEVEL = 30
+SUBLEVEL = 32
 EXTRAVERSION =
 NAME = TOSSUG Baby Fish
 

arch/arm64/include/asm/cacheflush.h

@@ -116,6 +116,7 @@ extern void flush_dcache_page(struct page *);
 static inline void __flush_icache_all(void)
 {
 	asm("ic	ialluis");
+	dsb();
 }
 
 #define flush_dcache_mmap_lock(mapping) \

arch/arm64/include/asm/topology.h

@@ -26,7 +26,6 @@ extern struct cputopo_arm cpu_topology[NR_CPUS];
 void init_cpu_topology(void);
 void store_cpu_topology(unsigned int cpuid);
 const struct cpumask *cpu_coregroup_mask(int cpu);
-int cluster_to_logical_mask(unsigned int socket_id, cpumask_t *cluster_mask);
 
 #ifdef CONFIG_DISABLE_CPU_SCHED_DOMAIN_BALANCE
 /* Common values for CPUs */
@@ -63,8 +62,6 @@ int cluster_to_logical_mask(unsigned int socket_id, cpumask_t *cluster_mask);
 
 static inline void init_cpu_topology(void) { }
 static inline void store_cpu_topology(unsigned int cpuid) { }
-static inline int cluster_to_logical_mask(unsigned int socket_id,
-	cpumask_t *cluster_mask) { return -EINVAL; }
 
 #endif
 

arch/arm64/kernel/topology.c

@@ -2,9 +2,9 @@
  * arch/arm64/kernel/topology.c
  *
  * Copyright (C) 2011,2013 Linaro Limited.
- * Written by: Vincent Guittot
  *
- * based on arch/sh/kernel/topology.c
+ * Based on the arm32 version written by Vincent Guittot in turn based on
+ * arch/sh/kernel/topology.c
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -13,7 +13,6 @@
 
 #include <linux/cpu.h>
 #include <linux/cpumask.h>
-#include <linux/export.h>
 #include <linux/init.h>
 #include <linux/percpu.h>
 #include <linux/node.h>
@@ -22,9 +21,10 @@
 #include <linux/sched.h>
 #include <linux/slab.h>
 
+#include <asm/topology.h>
 #include <asm/cputype.h>
 #include <asm/smp_plat.h>
-#include <asm/topology.h>
+
 
 /*
  * cpu power scale management
@@ -115,7 +115,7 @@ static void __init parse_core(struct device_node *core, int core_id)
 		if (t) {
 			leaf = false;
 			cpu = get_cpu_for_node(t);
-			if (cpu) {
+			if (cpu >= 0) {
 				pr_info("CPU%d: socket %d core %d thread %d\n",
 					cpu, cluster_id, core_id, i);
 				cpu_topology[cpu].socket_id = cluster_id;
@@ -146,7 +146,7 @@ static void __init parse_core(struct device_node *core, int core_id)
 	}
 }
 
-static void __init parse_cluster(struct device_node *cluster)
+static void __init parse_cluster(struct device_node *cluster, int depth)
 {
 	char name[10];
 	bool leaf = true;
@@ -165,7 +165,7 @@ static void __init parse_cluster(struct device_node *cluster)
 		snprintf(name, sizeof(name), "cluster%d", i);
 		c = of_get_child_by_name(cluster, name);
 		if (c) {
-			parse_cluster(c);
+			parse_cluster(c, depth + 1);
 			leaf = false;
 		}
 		i++;
@@ -179,6 +179,10 @@ static void __init parse_cluster(struct device_node *cluster)
 		if (c) {
 			has_cores = true;
 
+			if (depth == 0)
+				pr_err("%s: cpu-map children should be clusters\n",
+				       c->full_name);
+
 			if (leaf)
 				parse_core(c, core_id++);
 			else
@@ -228,7 +232,7 @@ static void __init parse_dt_topology(void)
 	cn = of_find_node_by_name(cn, "cpu-map");
 	if (!cn)
 		return;
-	parse_cluster(cn);
+	parse_cluster(cn, 0);
 
 	for_each_possible_cpu(cpu) {
 		const u32 *rate;
@@ -354,9 +358,9 @@ void store_cpu_topology(unsigned int cpuid)
 {
 	struct cputopo_arm *cpuid_topo = &cpu_topology[cpuid];
 
-	/* DT should have been parsed by the time we get here */
+	/* Something should have picked a topology by the time we get here */
 	if (cpuid_topo->core_id == -1)
-		pr_info("CPU%u: No topology information configured\n", cpuid);
+		pr_warn("CPU%u: No topology information configured\n", cpuid);
 	else
 		update_siblings_masks(cpuid);
 
@@ -534,4 +538,17 @@ void __init init_cpu_topology(void)
 	smp_wmb();
 
 	parse_dt_topology();
+
+	/*
+	 * Assign all remaining CPUs to a cluster so the scheduler
+	 * doesn't get confused.
+	 */
+	for_each_possible_cpu(cpu) {
+		struct cputopo_arm *cpu_topo = &cpu_topology[cpu];
+
+		if (cpu_topo->socket_id == -1) {
+			cpu_topo->socket_id = INT_MAX;
+			cpu_topo->core_id = cpu;
+		}
+	}
 }

arch/arm64/kernel/vdso.c

@@ -235,6 +235,8 @@ void update_vsyscall(struct timekeeper *tk)
 	vdso_data->use_syscall			= use_syscall;
 	vdso_data->xtime_coarse_sec		= xtime_coarse.tv_sec;
 	vdso_data->xtime_coarse_nsec		= xtime_coarse.tv_nsec;
+	vdso_data->wtm_clock_sec		= tk->wall_to_monotonic.tv_sec;
+	vdso_data->wtm_clock_nsec		= tk->wall_to_monotonic.tv_nsec;
 
 	if (!use_syscall) {
 		vdso_data->cs_cycle_last	= tk->clock->cycle_last;
@@ -242,8 +244,6 @@ void update_vsyscall(struct timekeeper *tk)
 		vdso_data->xtime_clock_nsec	= tk->xtime_nsec;
 		vdso_data->cs_mult		= tk->mult;
 		vdso_data->cs_shift		= tk->shift;
-		vdso_data->wtm_clock_sec	= tk->wall_to_monotonic.tv_sec;
-		vdso_data->wtm_clock_nsec	= tk->wall_to_monotonic.tv_nsec;
 	}
 
 	smp_wmb();

arch/arm64/kernel/vdso/Makefile

@@ -48,7 +48,7 @@ $(obj-vdso): %.o: %.S
 
 # Actual build commands
 quiet_cmd_vdsold = VDSOL $@
-      cmd_vdsold = $(CC) $(c_flags) -Wl,-T $^ -o $@
+      cmd_vdsold = $(CC) $(c_flags) -Wl,-n -Wl,-T $^ -o $@
 quiet_cmd_vdsoas = VDSOA $@
       cmd_vdsoas = $(CC) $(a_flags) -c -o $@ $<
 

arch/arm64/kernel/vdso/gettimeofday.S

@@ -103,6 +103,8 @@ ENTRY(__kernel_clock_gettime)
 	bl	__do_get_tspec
 	seqcnt_check w9, 1b
 
+	mov	x30, x2
+
 	cmp	w0, #CLOCK_MONOTONIC
 	b.ne	6f
 
@@ -118,6 +120,9 @@ ENTRY(__kernel_clock_gettime)
 	ccmp	w0, #CLOCK_MONOTONIC_COARSE, #0x4, ne
 	b.ne	8f
 
+	/* xtime_coarse_nsec is already right-shifted */
+	mov	x12, #0
+
 	/* Get coarse timespec. */
 	adr	vdso_data, _vdso_data
 3:	seqcnt_acquire
@@ -156,7 +161,7 @@ ENTRY(__kernel_clock_gettime)
 	lsr	x11, x11, x12
 	stp	x10, x11, [x1, #TSPEC_TV_SEC]
 	mov	x0, xzr
-	ret	x2
+	ret
 7:
 	mov	x30, x2
 8:	/* Syscall fallback. */

arch/arm64/mm/mmu.c

@@ -203,10 +203,18 @@ static void __init alloc_init_pmd(pud_t *pud, unsigned long addr,
 	do {
 		next = pmd_addr_end(addr, end);
 		/* try section mapping first */
-		if (((addr | next | phys) & ~SECTION_MASK) == 0)
+		if (((addr | next | phys) & ~SECTION_MASK) == 0) {
+			pmd_t old_pmd =*pmd;
 			set_pmd(pmd, __pmd(phys | prot_sect_kernel));
-		else
+			/*
+			 * Check for previous table entries created during
+			 * boot (__create_page_tables) and flush them.
+			 */
+			if (!pmd_none(old_pmd))
+				flush_tlb_all();
+		} else {
 			alloc_init_pte(pmd, addr, next, __phys_to_pfn(phys));
+		}
 		phys += next - addr;
 	} while (pmd++, addr = next, addr != end);
 }

arch/s390/crypto/aes_s390.c

@@ -25,6 +25,7 @@
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/init.h>
+#include <linux/spinlock.h>
 #include "crypt_s390.h"
 
 #define AES_KEYLEN_128		1
@@ -32,6 +33,7 @@
 #define AES_KEYLEN_256		4
 
 static u8 *ctrblk;
+static DEFINE_SPINLOCK(ctrblk_lock);
 static char keylen_flag;
 
 struct s390_aes_ctx {
@@ -756,43 +758,67 @@ static int ctr_aes_set_key(struct crypto_tfm *tfm, const u8 *in_key,
 	return aes_set_key(tfm, in_key, key_len);
 }
 
+static unsigned int __ctrblk_init(u8 *ctrptr, unsigned int nbytes)
+{
+	unsigned int i, n;
+
+	/* only use complete blocks, max. PAGE_SIZE */
+	n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : nbytes & ~(AES_BLOCK_SIZE - 1);
+	for (i = AES_BLOCK_SIZE; i < n; i += AES_BLOCK_SIZE) {
+		memcpy(ctrptr + i, ctrptr + i - AES_BLOCK_SIZE,
+		       AES_BLOCK_SIZE);
+		crypto_inc(ctrptr + i, AES_BLOCK_SIZE);
+	}
+	return n;
+}
+
 static int ctr_aes_crypt(struct blkcipher_desc *desc, long func,
 			 struct s390_aes_ctx *sctx, struct blkcipher_walk *walk)
 {
 	int ret = blkcipher_walk_virt_block(desc, walk, AES_BLOCK_SIZE);
-	unsigned int i, n, nbytes;
-	u8 buf[AES_BLOCK_SIZE];
-	u8 *out, *in;
+	unsigned int n, nbytes;
+	u8 buf[AES_BLOCK_SIZE], ctrbuf[AES_BLOCK_SIZE];
+	u8 *out, *in, *ctrptr = ctrbuf;
 
 	if (!walk->nbytes)
 		return ret;
 
-	memcpy(ctrblk, walk->iv, AES_BLOCK_SIZE);
+	if (spin_trylock(&ctrblk_lock))
+		ctrptr = ctrblk;
+
+	memcpy(ctrptr, walk->iv, AES_BLOCK_SIZE);
 	while ((nbytes = walk->nbytes) >= AES_BLOCK_SIZE) {
 		out = walk->dst.virt.addr;
 		in = walk->src.virt.addr;
 		while (nbytes >= AES_BLOCK_SIZE) {
-			/* only use complete blocks, max. PAGE_SIZE */
-			n = (nbytes > PAGE_SIZE) ? PAGE_SIZE :
-						 nbytes & ~(AES_BLOCK_SIZE - 1);
-			for (i = AES_BLOCK_SIZE; i < n; i += AES_BLOCK_SIZE) {
-				memcpy(ctrblk + i, ctrblk + i - AES_BLOCK_SIZE,
-				       AES_BLOCK_SIZE);
-				crypto_inc(ctrblk + i, AES_BLOCK_SIZE);
-			}
-			ret = crypt_s390_kmctr(func, sctx->key, out, in, n, ctrblk);
-			if (ret < 0 || ret != n)
+			if (ctrptr == ctrblk)
+				n = __ctrblk_init(ctrptr, nbytes);
+			else
+				n = AES_BLOCK_SIZE;
+			ret = crypt_s390_kmctr(func, sctx->key, out, in,
+					       n, ctrptr);
+			if (ret < 0 || ret != n) {
+				if (ctrptr == ctrblk)
+					spin_unlock(&ctrblk_lock);
 				return -EIO;
+			}
 			if (n > AES_BLOCK_SIZE)
-				memcpy(ctrblk, ctrblk + n - AES_BLOCK_SIZE,
+				memcpy(ctrptr, ctrptr + n - AES_BLOCK_SIZE,
 				       AES_BLOCK_SIZE);
-			crypto_inc(ctrblk, AES_BLOCK_SIZE);
+			crypto_inc(ctrptr, AES_BLOCK_SIZE);
 			out += n;
 			in += n;
 			nbytes -= n;
 		}
 		ret = blkcipher_walk_done(desc, walk, nbytes);
 	}
+	if (ctrptr == ctrblk) {
+		if (nbytes)
+			memcpy(ctrbuf, ctrptr, AES_BLOCK_SIZE);
+		else
+			memcpy(walk->iv, ctrptr, AES_BLOCK_SIZE);
+		spin_unlock(&ctrblk_lock);
+	}
 	/*
 	 * final block may be < AES_BLOCK_SIZE, copy only nbytes
 	 */
@@ -800,14 +826,15 @@ static int ctr_aes_crypt(struct blkcipher_desc *desc, long func,
 		out = walk->dst.virt.addr;
 		in = walk->src.virt.addr;
 		ret = crypt_s390_kmctr(func, sctx->key, buf, in,
-				       AES_BLOCK_SIZE, ctrblk);
+				       AES_BLOCK_SIZE, ctrbuf);
 		if (ret < 0 || ret != AES_BLOCK_SIZE)
 			return -EIO;
 		memcpy(out, buf, nbytes);
-		crypto_inc(ctrblk, AES_BLOCK_SIZE);
+		crypto_inc(ctrbuf, AES_BLOCK_SIZE);
 		ret = blkcipher_walk_done(desc, walk, 0);
+		memcpy(walk->iv, ctrbuf, AES_BLOCK_SIZE);
 	}
-	memcpy(walk->iv, ctrblk, AES_BLOCK_SIZE);
+
 	return ret;
 }
 

arch/s390/crypto/des_s390.c

@@ -25,6 +25,7 @@
 #define DES3_KEY_SIZE	(3 * DES_KEY_SIZE)
 
 static u8 *ctrblk;
+static DEFINE_SPINLOCK(ctrblk_lock);
 
 struct s390_des_ctx {
 	u8 iv[DES_BLOCK_SIZE];
@@ -105,29 +106,35 @@ static int ecb_desall_crypt(struct blkcipher_desc *desc, long func,
 }
 
 static int cbc_desall_crypt(struct blkcipher_desc *desc, long func,
-			    u8 *iv, struct blkcipher_walk *walk)
+			    struct blkcipher_walk *walk)
 {
+	struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm);
 	int ret = blkcipher_walk_virt(desc, walk);
 	unsigned int nbytes = walk->nbytes;
+	struct {
+		u8 iv[DES_BLOCK_SIZE];
+		u8 key[DES3_KEY_SIZE];
+	} param;
 
 	if (!nbytes)
 		goto out;
 
-	memcpy(iv, walk->iv, DES_BLOCK_SIZE);
+	memcpy(param.iv, walk->iv, DES_BLOCK_SIZE);
+	memcpy(param.key, ctx->key, DES3_KEY_SIZE);
 	do {
 		/* only use complete blocks */
 		unsigned int n = nbytes & ~(DES_BLOCK_SIZE - 1);
 		u8 *out = walk->dst.virt.addr;
 		u8 *in = walk->src.virt.addr;
 
-		ret = crypt_s390_kmc(func, iv, out, in, n);
+		ret = crypt_s390_kmc(func, &param, out, in, n);
 		if (ret < 0 || ret != n)
 			return -EIO;
 
 		nbytes &= DES_BLOCK_SIZE - 1;
 		ret = blkcipher_walk_done(desc, walk, nbytes);
 	} while ((nbytes = walk->nbytes));
-	memcpy(walk->iv, iv, DES_BLOCK_SIZE);
+	memcpy(walk->iv, param.iv, DES_BLOCK_SIZE);
 
 out:
 	return ret;
@@ -179,22 +186,20 @@ static int cbc_des_encrypt(struct blkcipher_desc *desc,
 			   struct scatterlist *dst, struct scatterlist *src,
 			   unsigned int nbytes)
 {
-	struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm);
 	struct blkcipher_walk walk;
 
 	blkcipher_walk_init(&walk, dst, src, nbytes);
-	return cbc_desall_crypt(desc, KMC_DEA_ENCRYPT, ctx->iv, &walk);
+	return cbc_desall_crypt(desc, KMC_DEA_ENCRYPT, &walk);
 }
 
 static int cbc_des_decrypt(struct blkcipher_desc *desc,
 			   struct scatterlist *dst, struct scatterlist *src,
 			   unsigned int nbytes)
 {
-	struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm);
 	struct blkcipher_walk walk;
 
 	blkcipher_walk_init(&walk, dst, src, nbytes);
-	return cbc_desall_crypt(desc, KMC_DEA_DECRYPT, ctx->iv, &walk);
+	return cbc_desall_crypt(desc, KMC_DEA_DECRYPT, &walk);
 }
 
 static struct crypto_alg cbc_des_alg = {
@@ -327,22 +332,20 @@ static int cbc_des3_encrypt(struct blkcipher_desc *desc,
 			    struct scatterlist *dst, struct scatterlist *src,
 			    unsigned int nbytes)
 {
-	struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm);
 	struct blkcipher_walk walk;
 
 	blkcipher_walk_init(&walk, dst, src, nbytes);
-	return cbc_desall_crypt(desc, KMC_TDEA_192_ENCRYPT, ctx->iv, &walk);
+	return cbc_desall_crypt(desc, KMC_TDEA_192_ENCRYPT, &walk);
 }
 
 static int cbc_des3_decrypt(struct blkcipher_desc *desc,
 			    struct scatterlist *dst, struct scatterlist *src,
 			    unsigned int nbytes)
 {
-	struct s390_des_ctx *ctx = crypto_blkcipher_ctx(desc->tfm);
 	struct blkcipher_walk walk;
 
 	blkcipher_walk_init(&walk, dst, src, nbytes);
-	return cbc_desall_crypt(desc, KMC_TDEA_192_DECRYPT, ctx->iv, &walk);
+	return cbc_desall_crypt(desc, KMC_TDEA_192_DECRYPT, &walk);
 }
 
 static struct crypto_alg cbc_des3_alg = {
@@ -366,54 +369,80 @@ static struct crypto_alg cbc_des3_alg = {
 	}
 };
 
+static unsigned int __ctrblk_init(u8 *ctrptr, unsigned int nbytes)
+{
+	unsigned int i, n;
+
+	/* align to block size, max. PAGE_SIZE */
+	n = (nbytes > PAGE_SIZE) ? PAGE_SIZE : nbytes & ~(DES_BLOCK_SIZE - 1);
+	for (i = DES_BLOCK_SIZE; i < n; i += DES_BLOCK_SIZE) {
+		memcpy(ctrptr + i, ctrptr + i - DES_BLOCK_SIZE, DES_BLOCK_SIZE);
+		crypto_inc(ctrptr + i, DES_BLOCK_SIZE);
+	}
+	return n;
+}
+
 static int ctr_desall_crypt(struct blkcipher_desc *desc, long func,
-			    struct s390_des_ctx *ctx, struct blkcipher_walk *walk)
+			    struct s390_des_ctx *ctx,
+			    struct blkcipher_walk *walk)
 {
 	int ret = blkcipher_walk_virt_block(desc, walk, DES_BLOCK_SIZE);
-	unsigned int i, n, nbytes;
-	u8 buf[DES_BLOCK_SIZE];
-	u8 *out, *in;
+	unsigned int n, nbytes;
+	u8 buf[DES_BLOCK_SIZE], ctrbuf[DES_BLOCK_SIZE];
+	u8 *out, *in, *ctrptr = ctrbuf;
 
-	memcpy(ctrblk, walk->iv, DES_BLOCK_SIZE);
+	if (!walk->nbytes)
+		return ret;
+
+	if (spin_trylock(&ctrblk_lock))
+		ctrptr = ctrblk;
+
+	memcpy(ctrptr, walk->iv, DES_BLOCK_SIZE);
 	while ((nbytes = walk->nbytes) >= DES_BLOCK_SIZE) {
 		out = walk->dst.virt.addr;
 		in = walk->src.virt.addr;
 		while (nbytes >= DES_BLOCK_SIZE) {
-			/* align to block size, max. PAGE_SIZE */
-			n = (nbytes > PAGE_SIZE) ? PAGE_SIZE :
-				nbytes & ~(DES_BLOCK_SIZE - 1);
-			for (i = DES_BLOCK_SIZE; i < n; i += DES_BLOCK_SIZE) {
-				memcpy(ctrblk + i, ctrblk + i - DES_BLOCK_SIZE,
-				       DES_BLOCK_SIZE);
-				crypto_inc(ctrblk + i, DES_BLOCK_SIZE);
-			}
-			ret = crypt_s390_kmctr(func, ctx->key, out, in, n, ctrblk);
-			if (ret < 0 || ret != n)
+			if (ctrptr == ctrblk)
+				n = __ctrblk_init(ctrptr, nbytes);
+			else
+				n = DES_BLOCK_SIZE;
+			ret = crypt_s390_kmctr(func, ctx->key, out, in,
+					       n, ctrptr);
+			if (ret < 0 || ret != n) {
+				if (ctrptr == ctrblk)
+					spin_unlock(&ctrblk_lock);
 				return -EIO;
+			}
 			if (n > DES_BLOCK_SIZE)
-				memcpy(ctrblk, ctrblk + n - DES_BLOCK_SIZE,
+				memcpy(ctrptr, ctrptr + n - DES_BLOCK_SIZE,
 				       DES_BLOCK_SIZE);
-			crypto_inc(ctrblk, DES_BLOCK_SIZE);
+			crypto_inc(ctrptr, DES_BLOCK_SIZE);
 			out += n;
 			in += n;
 			nbytes -= n;
 		}
 		ret = blkcipher_walk_done(desc, walk, nbytes);
 	}
-
+	if (ctrptr == ctrblk) {
+		if (nbytes)
+			memcpy(ctrbuf, ctrptr, DES_BLOCK_SIZE);
+		else
+			memcpy(walk->iv, ctrptr, DES_BLOCK_SIZE);
+		spin_unlock(&ctrblk_lock);
+	}
 	/* final block may be < DES_BLOCK_SIZE, copy only nbytes */
 	if (nbytes) {
 		out = walk->dst.virt.addr;
 		in = walk->src.virt.addr;
 		ret = crypt_s390_kmctr(func, ctx->key, buf, in,
-				       DES_BLOCK_SIZE, ctrblk);
+				       DES_BLOCK_SIZE, ctrbuf);
 		if (ret < 0 || ret != DES_BLOCK_SIZE)
 			return -EIO;
 		memcpy(out, buf, nbytes);
-		crypto_inc(ctrblk, DES_BLOCK_SIZE);
+		crypto_inc(ctrbuf, DES_BLOCK_SIZE);
 		ret = blkcipher_walk_done(desc, walk, 0);
+		memcpy(walk->iv, ctrbuf, DES_BLOCK_SIZE);
 	}
-	memcpy(walk->iv, ctrblk, DES_BLOCK_SIZE);
 	return ret;
 }
 

arch/s390/kernel/head64.S

@@ -59,7 +59,7 @@ ENTRY(startup_continue)
 	.quad	0			# cr12: tracing off
 	.quad	0			# cr13: home space segment table
 	.quad	0xc0000000		# cr14: machine check handling off
-	.quad	0			# cr15: linkage stack operations
+	.quad	.Llinkage_stack		# cr15: linkage stack operations
 .Lpcmsk:.quad	0x0000000180000000
 .L4malign:.quad 0xffffffffffc00000
 .Lscan2g:.quad	0x80000000 + 0x20000 - 8	# 2GB + 128K - 8
@@ -67,12 +67,15 @@ ENTRY(startup_continue)
 .Lparmaddr:
 	.quad	PARMAREA
 	.align	64
-.Lduct: .long	0,0,0,0,.Lduald,0,0,0
+.Lduct: .long	0,.Laste,.Laste,0,.Lduald,0,0,0
 	.long	0,0,0,0,0,0,0,0
+.Laste:	.quad	0,0xffffffffffffffff,0,0,0,0,0,0
 	.align	128
 .Lduald:.rept	8
 	.long	0x80000000,0,0,0	# invalid access-list entries
 	.endr
+.Llinkage_stack:
+	.long	0,0,0x89000000,0,0,0,0x8a000000,0
 
 ENTRY(_ehead)
 

arch/s390/mm/page-states.c

@@ -12,6 +12,8 @@
 #include <linux/mm.h>
 #include <linux/gfp.h>
 #include <linux/init.h>
+#include <asm/setup.h>
+#include <asm/ipl.h>
 
 #define ESSA_SET_STABLE		1
 #define ESSA_SET_UNUSED		2
@@ -41,6 +43,14 @@ void __init cmma_init(void)
 
 	if (!cmma_flag)
 		return;
+	/*
+	 * Disable CMM for dump, otherwise  the tprot based memory
+	 * detection can fail because of unstable pages.
+	 */
+	if (OLDMEM_BASE || ipl_info.type == IPL_TYPE_FCP_DUMP) {
+		cmma_flag = 0;
+		return;
+	}
 	asm volatile(
 		"       .insn rrf,0xb9ab0000,%1,%1,0,0\n"
 		"0:     la      %0,0\n"

arch/x86/include/asm/xen/page.h

@@ -79,30 +79,38 @@ static inline int phys_to_machine_mapping_valid(unsigned long pfn)
 	return get_phys_to_machine(pfn) != INVALID_P2M_ENTRY;
 }
 
-static inline unsigned long mfn_to_pfn(unsigned long mfn)
+static inline unsigned long mfn_to_pfn_no_overrides(unsigned long mfn)
 {
 	unsigned long pfn;
-	int ret = 0;
+	int ret;
 
 	if (xen_feature(XENFEAT_auto_translated_physmap))
 		return mfn;
 
-	if (unlikely(mfn >= machine_to_phys_nr)) {
-		pfn = ~0;
-		goto try_override;
-	}
-	pfn = 0;
+	if (unlikely(mfn >= machine_to_phys_nr))
+		return ~0;
+
 	/*
 	 * The array access can fail (e.g., device space beyond end of RAM).
 	 * In such cases it doesn't matter what we return (we return garbage),
 	 * but we must handle the fault without crashing!
 	 */
 	ret = __get_user(pfn, &machine_to_phys_mapping[mfn]);
-try_override:
-	/* ret might be < 0 if there are no entries in the m2p for mfn */
 	if (ret < 0)
-		pfn = ~0;
-	else if (get_phys_to_machine(pfn) != mfn)
+		return ~0;
+
+	return pfn;
+}
+
+static inline unsigned long mfn_to_pfn(unsigned long mfn)
+{
+	unsigned long pfn;
+
+	if (xen_feature(XENFEAT_auto_translated_physmap))
+		return mfn;
+
+	pfn = mfn_to_pfn_no_overrides(mfn);
+	if (get_phys_to_machine(pfn) != mfn) {
 		/*
 		 * If this appears to be a foreign mfn (because the pfn
 		 * doesn't map back to the mfn), then check the local override
@@ -111,6 +119,7 @@ try_override:
 		 * m2p_find_override_pfn returns ~0 if it doesn't find anything.
 		 */
 		pfn = m2p_find_override_pfn(mfn, ~0);
+	}
 
 	/* 
 	 * pfn is ~0 if there are no entries in the m2p for mfn or if the

arch/x86/kernel/cpu/common.c

@@ -284,8 +284,13 @@ static __always_inline void setup_smap(struct cpuinfo_x86 *c)
 	raw_local_save_flags(eflags);
 	BUG_ON(eflags & X86_EFLAGS_AC);
 
-	if (cpu_has(c, X86_FEATURE_SMAP))
+	if (cpu_has(c, X86_FEATURE_SMAP)) {
+#ifdef CONFIG_X86_SMAP
 		set_in_cr4(X86_CR4_SMAP);
+#else
+		clear_in_cr4(X86_CR4_SMAP);
+#endif
+	}
 }
 
 /*

arch/x86/kernel/cpu/intel.c

@@ -628,7 +628,7 @@ static void __cpuinit intel_tlb_flushall_shift_set(struct cpuinfo_x86 *c)
 		tlb_flushall_shift = 5;
 		break;
 	case 0x63a: /* Ivybridge */
-		tlb_flushall_shift = 1;
+		tlb_flushall_shift = 2;
 		break;
 	default:
 		tlb_flushall_shift = 6;

arch/x86/kernel/ftrace.c

@@ -77,8 +77,7 @@ within(unsigned long addr, unsigned long start, unsigned long end)
 	return addr >= start && addr < end;
 }
 
-static int
-do_ftrace_mod_code(unsigned long ip, const void *new_code)
+static unsigned long text_ip_addr(unsigned long ip)
 {
 	/*
 	 * On x86_64, kernel text mappings are mapped read-only with
@@ -91,7 +90,7 @@ do_ftrace_mod_code(unsigned long ip, const void *new_code)
 	if (within(ip, (unsigned long)_text, (unsigned long)_etext))
 		ip = (unsigned long)__va(__pa_symbol(ip));
 
-	return probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE);
+	return ip;
 }
 
 static const unsigned char *ftrace_nop_replace(void)
@@ -123,8 +122,10 @@ ftrace_modify_code_direct(unsigned long ip, unsigned const char *old_code,
 	if (memcmp(replaced, old_code, MCOUNT_INSN_SIZE) != 0)
 		return -EINVAL;
 
+	ip = text_ip_addr(ip);
+
 	/* replace the text with the new text */
-	if (do_ftrace_mod_code(ip, new_code))
+	if (probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE))
 		return -EPERM;
 
 	sync_core();
@@ -221,37 +222,51 @@ int ftrace_modify_call(struct dyn_ftrace *rec, unsigned long old_addr,
 	return -EINVAL;
 }
 
-int ftrace_update_ftrace_func(ftrace_func_t func)
+static unsigned long ftrace_update_func;
+
+static int update_ftrace_func(unsigned long ip, void *new)
 {
-	unsigned long ip = (unsigned long)(&ftrace_call);
-	unsigned char old[MCOUNT_INSN_SIZE], *new;
+	unsigned char old[MCOUNT_INSN_SIZE];
 	int ret;
 
-	memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE);
-	new = ftrace_call_replace(ip, (unsigned long)func);
+	memcpy(old, (void *)ip, MCOUNT_INSN_SIZE);
+
+	ftrace_update_func = ip;
+	/* Make sure the breakpoints see the ftrace_update_func update */
+	smp_wmb();
 
 	/* See comment above by declaration of modifying_ftrace_code */
 	atomic_inc(&modifying_ftrace_code);
 
 	ret = ftrace_modify_code(ip, old, new);
 
+	atomic_dec(&modifying_ftrace_code);
+
+	return ret;
+}
+
+int ftrace_update_ftrace_func(ftrace_func_t func)
+{
+	unsigned long ip = (unsigned long)(&ftrace_call);
+	unsigned char *new;
+	int ret;
+
+	new = ftrace_call_replace(ip, (unsigned long)func);
+	ret = update_ftrace_func(ip, new);
+
 	/* Also update the regs callback function */
 	if (!ret) {
 		ip = (unsigned long)(&ftrace_regs_call);
-		memcpy(old, &ftrace_regs_call, MCOUNT_INSN_SIZE);
 		new = ftrace_call_replace(ip, (unsigned long)func);
-		ret = ftrace_modify_code(ip, old, new);
+		ret = update_ftrace_func(ip, new);
 	}
 
-	atomic_dec(&modifying_ftrace_code);
-
 	return ret;
 }
 
 static int is_ftrace_caller(unsigned long ip)
 {
-	if (ip == (unsigned long)(&ftrace_call) ||
-		ip == (unsigned long)(&ftrace_regs_call))
+	if (ip == ftrace_update_func)
 		return 1;
 
 	return 0;
@@ -677,45 +692,41 @@ int __init ftrace_dyn_arch_init(void *data)
 #ifdef CONFIG_DYNAMIC_FTRACE
 extern void ftrace_graph_call(void);
 
-static int ftrace_mod_jmp(unsigned long ip,
-			  int old_offset, int new_offset)
+static unsigned char *ftrace_jmp_replace(unsigned long ip, unsigned long addr)
 {
-	unsigned char code[MCOUNT_INSN_SIZE];
+	static union ftrace_code_union calc;
 
-	if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE))
-		return -EFAULT;
+	/* Jmp not a call (ignore the .e8) */
+	calc.e8		= 0xe9;
+	calc.offset	= ftrace_calc_offset(ip + MCOUNT_INSN_SIZE, addr);
 
-	if (code[0] != 0xe9 || old_offset != *(int *)(&code[1]))
-		return -EINVAL;
+	/*
+	 * ftrace external locks synchronize the access to the static variable.
+	 */
+	return calc.code;
+}
 
-	*(int *)(&code[1]) = new_offset;
+static int ftrace_mod_jmp(unsigned long ip, void *func)
+{
+	unsigned char *new;
 
-	if (do_ftrace_mod_code(ip, &code))
-		return -EPERM;
+	new = ftrace_jmp_replace(ip, (unsigned long)func);
 
-	return 0;
+	return update_ftrace_func(ip, new);
 }
 
 int ftrace_enable_ftrace_graph_caller(void)
 {
 	unsigned long ip = (unsigned long)(&ftrace_graph_call);
-	int old_offset, new_offset;
 
-	old_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE);
-	new_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE);
-
-	return ftrace_mod_jmp(ip, old_offset, new_offset);
+	return ftrace_mod_jmp(ip, &ftrace_graph_caller);
 }
 
 int ftrace_disable_ftrace_graph_caller(void)
 {
 	unsigned long ip = (unsigned long)(&ftrace_graph_call);
-	int old_offset, new_offset;
 
-	old_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE);
-	new_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE);
-
-	return ftrace_mod_jmp(ip, old_offset, new_offset);
+	return ftrace_mod_jmp(ip, &ftrace_stub);
 }
 
 #endif /* !CONFIG_DYNAMIC_FTRACE */

arch/x86/mm/fault.c

@@ -989,6 +989,12 @@ static int fault_in_kernel_space(unsigned long address)
 
 static inline bool smap_violation(int error_code, struct pt_regs *regs)
 {
+	if (!IS_ENABLED(CONFIG_X86_SMAP))
+		return false;
+
+	if (!static_cpu_has(X86_FEATURE_SMAP))
+		return false;
+
 	if (error_code & PF_USER)
 		return false;
 
@@ -1091,11 +1097,9 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	if (unlikely(error_code & PF_RSVD))
 		pgtable_bad(regs, error_code, address);
 
-	if (static_cpu_has(X86_FEATURE_SMAP)) {
-		if (unlikely(smap_violation(error_code, regs))) {
-			bad_area_nosemaphore(regs, error_code, address);
-			return;
-		}
+	if (unlikely(smap_violation(error_code, regs))) {
+		bad_area_nosemaphore(regs, error_code, address);
+		return;
 	}
 
 	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);

arch/x86/xen/p2m.c

@@ -878,7 +878,6 @@ int m2p_add_override(unsigned long mfn, struct page *page,
 	unsigned long uninitialized_var(address);
 	unsigned level;
 	pte_t *ptep = NULL;
-	int ret = 0;
 
 	pfn = page_to_pfn(page);
 	if (!PageHighMem(page)) {
@@ -925,8 +924,8 @@ int m2p_add_override(unsigned long mfn, struct page *page,
 	 * frontend pages while they are being shared with the backend,
 	 * because mfn_to_pfn (that ends up being called by GUPF) will
 	 * return the backend pfn rather than the frontend pfn. */
-	ret = __get_user(pfn, &machine_to_phys_mapping[mfn]);
-	if (ret == 0 && get_phys_to_machine(pfn) == mfn)
+	pfn = mfn_to_pfn_no_overrides(mfn);
+	if (get_phys_to_machine(pfn) == mfn)
 		set_phys_to_machine(pfn, FOREIGN_FRAME(mfn));
 
 	return 0;
@@ -941,7 +940,6 @@ int m2p_remove_override(struct page *page,
 	unsigned long uninitialized_var(address);
 	unsigned level;
 	pte_t *ptep = NULL;
-	int ret = 0;
 
 	pfn = page_to_pfn(page);
 	mfn = get_phys_to_machine(pfn);
@@ -1019,8 +1017,8 @@ int m2p_remove_override(struct page *page,
 	 * the original pfn causes mfn_to_pfn(mfn) to return the frontend
 	 * pfn again. */
 	mfn &= ~FOREIGN_FRAME_BIT;
-	ret = __get_user(pfn, &machine_to_phys_mapping[mfn]);
-	if (ret == 0 && get_phys_to_machine(pfn) == FOREIGN_FRAME(mfn) &&
+	pfn = mfn_to_pfn_no_overrides(mfn);
+	if (get_phys_to_machine(pfn) == FOREIGN_FRAME(mfn) &&
 			m2p_find_override(mfn) == NULL)
 		set_phys_to_machine(pfn, mfn);
 

arch/x86/xen/smp.c

@@ -245,6 +245,15 @@ static void __init xen_smp_prepare_boot_cpu(void)
 	   old memory can be recycled */
 	make_lowmem_page_readwrite(xen_initial_gdt);
 
+#ifdef CONFIG_X86_32
+	/*
+	 * Xen starts us with XEN_FLAT_RING1_DS, but linux code
+	 * expects __USER_DS
+	 */
+	loadsegment(ds, __USER_DS);
+	loadsegment(es, __USER_DS);
+#endif
+
 	xen_filter_cpu_maps();
 	xen_setup_vcpu_info_placement();
 }

block/blk-lib.c

@@ -121,6 +121,14 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 
 		atomic_inc(&bb.done);
 		submit_bio(type, bio);
+
+		/*
+		 * We can loop for a long time in here, if someone does
+		 * full device discards (like mkfs). Be nice and allow
+		 * us to schedule out to avoid softlocking if preempt
+		 * is disabled.
+		 */
+		cond_resched();
 	}
 	blk_finish_plug(&plug);