Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next

Alexei Starovoitov says:

====================
pull-request: bpf-next 2020-05-01 (v2)

The following pull-request contains BPF updates for your *net-next* tree.

We've added 61 non-merge commits during the last 6 day(s) which contain
a total of 153 files changed, 6739 insertions(+), 3367 deletions(-).

The main changes are:

1) pulled work.sysctl from vfs tree with sysctl bpf changes.

2) bpf_link observability, from Andrii.

3) BTF-defined map in map, from Andrii.

4) asan fixes for selftests, from Andrii.

5) Allow bpf_map_lookup_elem for SOCKMAP and SOCKHASH, from Jakub.

6) production cloudflare classifier as a selftes, from Lorenz.

7) bpf_ktime_get_*_ns() helper improvements, from Maciej.

8) unprivileged bpftool feature probe, from Quentin.

9) BPF_ENABLE_STATS command, from Song.

10) enable bpf_[gs]etsockopt() helpers for sock_ops progs, from Stanislav.

11) enable a bunch of common helpers for cg-device, sysctl, sockopt progs,
 from Stanislav.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

arch/arm64/kernel/armv8_deprecated.c

@@ -203,7 +203,7 @@ static void __init register_insn_emulation(struct insn_emulation_ops *ops)
 }
 
 static int emulation_proc_handler(struct ctl_table *table, int write,
-				  void __user *buffer, size_t *lenp,
+				  void *buffer, size_t *lenp,
 				  loff_t *ppos)
 {
 	int ret = 0;

arch/arm64/kernel/fpsimd.c

@@ -341,8 +341,7 @@ static unsigned int find_supported_vector_length(unsigned int vl)
 #ifdef CONFIG_SYSCTL
 
 static int sve_proc_do_default_vl(struct ctl_table *table, int write,
-				  void __user *buffer, size_t *lenp,
+				  void *buffer, size_t *lenp, loff_t *ppos)
-				  loff_t *ppos)
 {
 	int ret;
 	int vl = sve_default_vl;

arch/mips/lasat/sysctl.c

@@ -95,16 +95,15 @@ int proc_lasat_ip(struct ctl_table *table, int write,
 		len = 0;
 		p = buffer;
 		while (len < *lenp) {
-			if (get_user(c, p++))
+			c = *p;
-				return -EFAULT;
+			p++;
 			if (c == 0 || c == '\n')
 				break;
 			len++;
 		}
 		if (len >= sizeof(ipbuf)-1)
 			len = sizeof(ipbuf) - 1;
-		if (copy_from_user(ipbuf, buffer, len))
+		memcpy(ipbuf, buffer, len);
-			return -EFAULT;
 		ipbuf[len] = 0;
 		*ppos += *lenp;
 		/* Now see if we can convert it to a valid IP */
@@ -122,11 +121,9 @@ int proc_lasat_ip(struct ctl_table *table, int write,
 		if (len > *lenp)
 			len = *lenp;
 		if (len)
-			if (copy_to_user(buffer, ipbuf, len))
+			memcpy(buffer, ipbuf, len);
-				return -EFAULT;
 		if (len < *lenp) {
-			if (put_user('\n', ((char *) buffer) + len))
+			*((char *)buffer + len) = '\n';
-				return -EFAULT;
 			len++;
 		}
 		*lenp = len;

arch/riscv/net/bpf_jit_comp32.c

@@ -13,8 +13,35 @@
 #include <linux/filter.h>
 #include "bpf_jit.h"
 
+/*
+ * Stack layout during BPF program execution:
+ *
+ *                     high
+ *     RV32 fp =>  +----------+
+ *                 | saved ra |
+ *                 | saved fp | RV32 callee-saved registers
+ *                 |   ...    |
+ *                 +----------+ <= (fp - 4 * NR_SAVED_REGISTERS)
+ *                 |  hi(R6)  |
+ *                 |  lo(R6)  |
+ *                 |  hi(R7)  | JIT scratch space for BPF registers
+ *                 |  lo(R7)  |
+ *                 |   ...    |
+ *  BPF_REG_FP =>  +----------+ <= (fp - 4 * NR_SAVED_REGISTERS
+ *                 |          |        - 4 * BPF_JIT_SCRATCH_REGS)
+ *                 |          |
+ *                 |   ...    | BPF program stack
+ *                 |          |
+ *     RV32 sp =>  +----------+
+ *                 |          |
+ *                 |   ...    | Function call stack
+ *                 |          |
+ *                 +----------+
+ *                     low
+ */
+
 enum {
-	/* Stack layout - these are offsets from (top of stack - 4). */
+	/* Stack layout - these are offsets from top of JIT scratch space. */
 	BPF_R6_HI,
 	BPF_R6_LO,
 	BPF_R7_HI,
@@ -29,7 +56,11 @@ enum {
 	BPF_JIT_SCRATCH_REGS,
 };
 
-#define STACK_OFFSET(k) (-4 - ((k) * 4))
+/* Number of callee-saved registers stored to stack: ra, fp, s1--s7. */
+#define NR_SAVED_REGISTERS	9
+
+/* Offset from fp for BPF registers stored on stack. */
+#define STACK_OFFSET(k)	(-4 - (4 * NR_SAVED_REGISTERS) - (4 * (k)))
 
 #define TMP_REG_1	(MAX_BPF_JIT_REG + 0)
 #define TMP_REG_2	(MAX_BPF_JIT_REG + 1)
@@ -111,11 +142,9 @@ static void emit_imm64(const s8 *rd, s32 imm_hi, s32 imm_lo,
 
 static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx)
 {
-	int stack_adjust = ctx->stack_size, store_offset = stack_adjust - 4;
+	int stack_adjust = ctx->stack_size;
 	const s8 *r0 = bpf2rv32[BPF_REG_0];
 
-	store_offset -= 4 * BPF_JIT_SCRATCH_REGS;
-
 	/* Set return value if not tail call. */
 	if (!is_tail_call) {
 		emit(rv_addi(RV_REG_A0, lo(r0), 0), ctx);
@@ -123,15 +152,15 @@ static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx)
 	}
 
 	/* Restore callee-saved registers. */
-	emit(rv_lw(RV_REG_RA, store_offset - 0, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_RA, stack_adjust - 4, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_FP, store_offset - 4, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_FP, stack_adjust - 8, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S1, store_offset - 8, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S1, stack_adjust - 12, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S2, store_offset - 12, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S2, stack_adjust - 16, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S3, store_offset - 16, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S3, stack_adjust - 20, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S4, store_offset - 20, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S4, stack_adjust - 24, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S5, store_offset - 24, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S5, stack_adjust - 28, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S6, store_offset - 28, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S6, stack_adjust - 32, RV_REG_SP), ctx);
-	emit(rv_lw(RV_REG_S7, store_offset - 32, RV_REG_SP), ctx);
+	emit(rv_lw(RV_REG_S7, stack_adjust - 36, RV_REG_SP), ctx);
 
 	emit(rv_addi(RV_REG_SP, RV_REG_SP, stack_adjust), ctx);
 
@@ -770,12 +799,13 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 	emit_bcc(BPF_JGE, lo(idx_reg), RV_REG_T1, off, ctx);
 
 	/*
-	 * if ((temp_tcc = tcc - 1) < 0)
+	 * temp_tcc = tcc - 1;
+	 * if (tcc < 0)
 	 *   goto out;
 	 */
 	emit(rv_addi(RV_REG_T1, RV_REG_TCC, -1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	emit_bcc(BPF_JSLT, RV_REG_T1, RV_REG_ZERO, off, ctx);
+	emit_bcc(BPF_JSLT, RV_REG_TCC, RV_REG_ZERO, off, ctx);
 
 	/*
 	 * prog = array->ptrs[index];
@@ -1259,17 +1289,20 @@ notsupported:
 
 void bpf_jit_build_prologue(struct rv_jit_context *ctx)
 {
-	/* Make space to save 9 registers: ra, fp, s1--s7. */
-	int stack_adjust = 9 * sizeof(u32), store_offset, bpf_stack_adjust;
 	const s8 *fp = bpf2rv32[BPF_REG_FP];
 	const s8 *r1 = bpf2rv32[BPF_REG_1];
+	int stack_adjust = 0;
+	int bpf_stack_adjust =
+		round_up(ctx->prog->aux->stack_depth, STACK_ALIGN);
 
-	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
+	/* Make space for callee-saved registers. */
+	stack_adjust += NR_SAVED_REGISTERS * sizeof(u32);
+	/* Make space for BPF registers on stack. */
+	stack_adjust += BPF_JIT_SCRATCH_REGS * sizeof(u32);
+	/* Make space for BPF stack. */
 	stack_adjust += bpf_stack_adjust;
-
+	/* Round up for stack alignment. */
-	store_offset = stack_adjust - 4;
+	stack_adjust = round_up(stack_adjust, STACK_ALIGN);
-
-	stack_adjust += 4 * BPF_JIT_SCRATCH_REGS;
 
 	/*
 	 * The first instruction sets the tail-call-counter (TCC) register.
@@ -1280,24 +1313,24 @@ void bpf_jit_build_prologue(struct rv_jit_context *ctx)
 	emit(rv_addi(RV_REG_SP, RV_REG_SP, -stack_adjust), ctx);
 
 	/* Save callee-save registers. */
-	emit(rv_sw(RV_REG_SP, store_offset - 0, RV_REG_RA), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 4, RV_REG_RA), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 4, RV_REG_FP), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 8, RV_REG_FP), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 8, RV_REG_S1), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 12, RV_REG_S1), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 12, RV_REG_S2), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 16, RV_REG_S2), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 16, RV_REG_S3), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 20, RV_REG_S3), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 20, RV_REG_S4), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 24, RV_REG_S4), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 24, RV_REG_S5), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 28, RV_REG_S5), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 28, RV_REG_S6), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 32, RV_REG_S6), ctx);
-	emit(rv_sw(RV_REG_SP, store_offset - 32, RV_REG_S7), ctx);
+	emit(rv_sw(RV_REG_SP, stack_adjust - 36, RV_REG_S7), ctx);
 
 	/* Set fp: used as the base address for stacked BPF registers. */
 	emit(rv_addi(RV_REG_FP, RV_REG_SP, stack_adjust), ctx);
 
-	/* Set up BPF stack pointer. */
+	/* Set up BPF frame pointer. */
 	emit(rv_addi(lo(fp), RV_REG_SP, bpf_stack_adjust), ctx);
 	emit(rv_addi(hi(fp), RV_REG_ZERO, 0), ctx);
 
-	/* Set up context pointer. */
+	/* Set up BPF context pointer. */
 	emit(rv_addi(lo(r1), RV_REG_A0, 0), ctx);
 	emit(rv_addi(hi(r1), RV_REG_ZERO, 0), ctx);
 

arch/s390/appldata/appldata_base.c

@@ -51,10 +51,9 @@ static struct platform_device *appldata_pdev;
  */
 static const char appldata_proc_name[APPLDATA_PROC_NAME_LENGTH] = "appldata";
 static int appldata_timer_handler(struct ctl_table *ctl, int write,
-				  void __user *buffer, size_t *lenp, loff_t *ppos);
+				  void *buffer, size_t *lenp, loff_t *ppos);
 static int appldata_interval_handler(struct ctl_table *ctl, int write,
-					 void __user *buffer,
+				     void *buffer, size_t *lenp, loff_t *ppos);
-					 size_t *lenp, loff_t *ppos);
 
 static struct ctl_table_header *appldata_sysctl_header;
 static struct ctl_table appldata_table[] = {
@@ -217,7 +216,7 @@ static void __appldata_vtimer_setup(int cmd)
  */
 static int
 appldata_timer_handler(struct ctl_table *ctl, int write,
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
+			   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int timer_active = appldata_timer_active;
 	int rc;
@@ -250,7 +249,7 @@ appldata_timer_handler(struct ctl_table *ctl, int write,
  */
 static int
 appldata_interval_handler(struct ctl_table *ctl, int write,
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
+			   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int interval = appldata_interval;
 	int rc;
@@ -280,7 +279,7 @@ appldata_interval_handler(struct ctl_table *ctl, int write,
  */
 static int
 appldata_generic_handler(struct ctl_table *ctl, int write,
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
+			   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct appldata_ops *ops = NULL, *tmp_ops;
 	struct list_head *lh;

arch/s390/kernel/debug.c

@@ -867,7 +867,7 @@ static int debug_active = 1;
  * if debug_active is already off
  */
 static int s390dbf_procactive(struct ctl_table *table, int write,
-			      void __user *buffer, size_t *lenp, loff_t *ppos)
+			      void *buffer, size_t *lenp, loff_t *ppos)
 {
 	if (!write || debug_stoppable || !debug_active)
 		return proc_dointvec(table, write, buffer, lenp, ppos);

arch/s390/kernel/topology.c

@@ -594,7 +594,7 @@ static int __init topology_setup(char *str)
 early_param("topology", topology_setup);
 
 static int topology_ctl_handler(struct ctl_table *ctl, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int enabled = topology_is_enabled();
 	int new_mode;

arch/s390/mm/cmm.c

@@ -245,7 +245,7 @@ static int cmm_skip_blanks(char *cp, char **endp)
 }
 
 static int cmm_pages_handler(struct ctl_table *ctl, int write,
-			     void __user *buffer, size_t *lenp, loff_t *ppos)
+			     void *buffer, size_t *lenp, loff_t *ppos)
 {
 	long nr = cmm_get_pages();
 	struct ctl_table ctl_entry = {
@@ -264,7 +264,7 @@ static int cmm_pages_handler(struct ctl_table *ctl, int write,
 }
 
 static int cmm_timed_pages_handler(struct ctl_table *ctl, int write,
-				   void __user *buffer, size_t *lenp,
+				   void *buffer, size_t *lenp,
 				   loff_t *ppos)
 {
 	long nr = cmm_get_timed_pages();
@@ -284,7 +284,7 @@ static int cmm_timed_pages_handler(struct ctl_table *ctl, int write,
 }
 
 static int cmm_timeout_handler(struct ctl_table *ctl, int write,
-			       void __user *buffer, size_t *lenp, loff_t *ppos)
+			       void *buffer, size_t *lenp, loff_t *ppos)
 {
 	char buf[64], *p;
 	long nr, seconds;
@@ -297,8 +297,7 @@ static int cmm_timeout_handler(struct ctl_table *ctl, int write,
 
 	if (write) {
 		len = min(*lenp, sizeof(buf));
-		if (copy_from_user(buf, buffer, len))
+		memcpy(buf, buffer, len);
-			return -EFAULT;
 		buf[len - 1] = '\0';
 		cmm_skip_blanks(buf, &p);
 		nr = simple_strtoul(p, &p, 0);
@@ -311,8 +310,7 @@ static int cmm_timeout_handler(struct ctl_table *ctl, int write,
 			      cmm_timeout_pages, cmm_timeout_seconds);
 		if (len > *lenp)
 			len = *lenp;
-		if (copy_to_user(buffer, buf, len))
+		memcpy(buffer, buf, len);
-			return -EFAULT;
 		*lenp = len;
 		*ppos += len;
 	}

arch/x86/kernel/itmt.c

@@ -39,8 +39,7 @@ static bool __read_mostly sched_itmt_capable;
 unsigned int __read_mostly sysctl_sched_itmt_enabled;
 
 static int sched_itmt_update_handler(struct ctl_table *table, int write,
-				     void __user *buffer, size_t *lenp,
+				     void *buffer, size_t *lenp, loff_t *ppos)
-				     loff_t *ppos)
 {
 	unsigned int old_sysctl;
 	int ret;

drivers/cdrom/cdrom.c

@@ -3631,7 +3631,7 @@ static void cdrom_update_settings(void)
 }
 
 static int cdrom_sysctl_handler(struct ctl_table *ctl, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret;
 	

drivers/char/random.c

@@ -2057,7 +2057,7 @@ static char sysctl_bootid[16];
  * sysctl system call, as 16 bytes of binary data.
  */
 static int proc_do_uuid(struct ctl_table *table, int write,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
+			void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table fake_table;
 	unsigned char buf[64], tmp_uuid[16], *uuid;

drivers/macintosh/mac_hid.c

@@ -183,8 +183,7 @@ static void mac_hid_stop_emulation(void)
 }
 
 static int mac_hid_toggle_emumouse(struct ctl_table *table, int write,
-				   void __user *buffer, size_t *lenp,
+				   void *buffer, size_t *lenp, loff_t *ppos)
-				   loff_t *ppos)
 {
 	int *valp = table->data;
 	int old_val = *valp;

drivers/media/rc/bpf-lirc.c

@@ -103,6 +103,8 @@ lirc_mode2_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_map_peek_elem_proto;
 	case BPF_FUNC_ktime_get_ns:
 		return &bpf_ktime_get_ns_proto;
+	case BPF_FUNC_ktime_get_boot_ns:
+		return &bpf_ktime_get_boot_ns_proto;
 	case BPF_FUNC_tail_call:
 		return &bpf_tail_call_proto;
 	case BPF_FUNC_get_prandom_u32:

drivers/parport/procfs.c

@@ -34,7 +34,7 @@
 #define PARPORT_MAX_SPINTIME_VALUE 1000
 
 static int do_active_device(struct ctl_table *table, int write,
-		      void __user *result, size_t *lenp, loff_t *ppos)
+		      void *result, size_t *lenp, loff_t *ppos)
 {
 	struct parport *port = (struct parport *)table->extra1;
 	char buffer[256];
@@ -65,13 +65,13 @@ static int do_active_device(struct ctl_table *table, int write,
 		*lenp = len;
 
 	*ppos += len;
-
+	memcpy(result, buffer, len);
-	return copy_to_user(result, buffer, len) ? -EFAULT : 0;
+	return 0;
 }
 
 #ifdef CONFIG_PARPORT_1284
 static int do_autoprobe(struct ctl_table *table, int write,
-			void __user *result, size_t *lenp, loff_t *ppos)
+			void *result, size_t *lenp, loff_t *ppos)
 {
 	struct parport_device_info *info = table->extra2;
 	const char *str;
@@ -108,13 +108,13 @@ static int do_autoprobe(struct ctl_table *table, int write,
 
 	*ppos += len;
 
-	return copy_to_user (result, buffer, len) ? -EFAULT : 0;
+	memcpy(result, buffer, len);
+	return 0;
 }
 #endif /* IEEE1284.3 support. */
 
 static int do_hardware_base_addr(struct ctl_table *table, int write,
-				 void __user *result,
+				 void *result, size_t *lenp, loff_t *ppos)
-				 size_t *lenp, loff_t *ppos)
 {
 	struct parport *port = (struct parport *)table->extra1;
 	char buffer[20];
@@ -136,13 +136,12 @@ static int do_hardware_base_addr(struct ctl_table *table, int write,
 		*lenp = len;
 
 	*ppos += len;
-
+	memcpy(result, buffer, len);
-	return copy_to_user(result, buffer, len) ? -EFAULT : 0;
+	return 0;
 }
 
 static int do_hardware_irq(struct ctl_table *table, int write,
-			   void __user *result,
+			   void *result, size_t *lenp, loff_t *ppos)
-			   size_t *lenp, loff_t *ppos)
 {
 	struct parport *port = (struct parport *)table->extra1;
 	char buffer[20];
@@ -164,13 +163,12 @@ static int do_hardware_irq(struct ctl_table *table, int write,
 		*lenp = len;
 
 	*ppos += len;
-
+	memcpy(result, buffer, len);
-	return copy_to_user(result, buffer, len) ? -EFAULT : 0;
+	return 0;
 }
 
 static int do_hardware_dma(struct ctl_table *table, int write,
-			   void __user *result,
+			   void *result, size_t *lenp, loff_t *ppos)
-			   size_t *lenp, loff_t *ppos)
 {
 	struct parport *port = (struct parport *)table->extra1;
 	char buffer[20];
@@ -192,13 +190,12 @@ static int do_hardware_dma(struct ctl_table *table, int write,
 		*lenp = len;
 
 	*ppos += len;
-
+	memcpy(result, buffer, len);
-	return copy_to_user(result, buffer, len) ? -EFAULT : 0;
+	return 0;
 }
 
 static int do_hardware_modes(struct ctl_table *table, int write,
-			     void __user *result,
+			     void *result, size_t *lenp, loff_t *ppos)
-			     size_t *lenp, loff_t *ppos)
 {
 	struct parport *port = (struct parport *)table->extra1;
 	char buffer[40];
@@ -231,8 +228,8 @@ static int do_hardware_modes(struct ctl_table *table, int write,
 		*lenp = len;
 
 	*ppos += len;
-
+	memcpy(result, buffer, len);
-	return copy_to_user(result, buffer, len) ? -EFAULT : 0;
+	return 0;
 }
 
 #define PARPORT_PORT_DIR(CHILD) { .procname = NULL, .mode = 0555, .child = CHILD }

fs/dcache.c

@@ -165,7 +165,7 @@ static long get_nr_dentry_negative(void)
 	return sum < 0 ? 0 : sum;
 }
 
-int proc_nr_dentry(struct ctl_table *table, int write, void __user *buffer,
+int proc_nr_dentry(struct ctl_table *table, int write, void *buffer,
 		   size_t *lenp, loff_t *ppos)
 {
 	dentry_stat.nr_dentry = get_nr_dentry();

fs/drop_caches.c

@@ -47,7 +47,7 @@ static void drop_pagecache_sb(struct super_block *sb, void *unused)
 }
 
 int drop_caches_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	int ret;
 

fs/file_table.c

@@ -80,14 +80,14 @@ EXPORT_SYMBOL_GPL(get_max_files);
  */
 #if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS)
 int proc_nr_files(struct ctl_table *table, int write,
-                     void __user *buffer, size_t *lenp, loff_t *ppos)
+                     void *buffer, size_t *lenp, loff_t *ppos)
 {
 	files_stat.nr_files = get_nr_files();
 	return proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
 }
 #else
 int proc_nr_files(struct ctl_table *table, int write,
-                     void __user *buffer, size_t *lenp, loff_t *ppos)
+                     void *buffer, size_t *lenp, loff_t *ppos)
 {
 	return -ENOSYS;
 }

fs/fscache/main.c

@@ -51,8 +51,7 @@ static unsigned fscache_op_max_active = 2;
 static struct ctl_table_header *fscache_sysctl_header;
 
 static int fscache_max_active_sysctl(struct ctl_table *table, int write,
-				     void __user *buffer,
+				     void *buffer, size_t *lenp, loff_t *ppos)
-				     size_t *lenp, loff_t *ppos)
 {
 	struct workqueue_struct **wqp = table->extra1;
 	unsigned int *datap = table->data;

fs/inode.c

@@ -108,7 +108,7 @@ long get_nr_dirty_inodes(void)
  */
 #ifdef CONFIG_SYSCTL
 int proc_nr_inodes(struct ctl_table *table, int write,
-		   void __user *buffer, size_t *lenp, loff_t *ppos)
+		   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	inodes_stat.nr_inodes = get_nr_inodes();
 	inodes_stat.nr_unused = get_nr_inodes_unused();

fs/proc/proc_sysctl.c

@@ -539,13 +539,13 @@ out:
 	return err;
 }
 
-static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf,
+static ssize_t proc_sys_call_handler(struct file *filp, void __user *ubuf,
 		size_t count, loff_t *ppos, int write)
 {
 	struct inode *inode = file_inode(filp);
 	struct ctl_table_header *head = grab_header(inode);
 	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
-	void *new_buf = NULL;
+	void *kbuf;
 	ssize_t error;
 
 	if (IS_ERR(head))
@@ -564,27 +564,38 @@ static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf,
 	if (!table->proc_handler)
 		goto out;
 
-	error = BPF_CGROUP_RUN_PROG_SYSCTL(head, table, write, buf, &count,
+	if (write) {
-					   ppos, &new_buf);
+		kbuf = memdup_user_nul(ubuf, count);
-	if (error)
+		if (IS_ERR(kbuf)) {
-		goto out;
+			error = PTR_ERR(kbuf);
-
+			goto out;
-	/* careful: calling conventions are nasty here */
+		}
-	if (new_buf) {
-		mm_segment_t old_fs;
-
-		old_fs = get_fs();
-		set_fs(KERNEL_DS);
-		error = table->proc_handler(table, write, (void __user *)new_buf,
-					    &count, ppos);
-		set_fs(old_fs);
-		kfree(new_buf);
 	} else {
-		error = table->proc_handler(table, write, buf, &count, ppos);
+		error = -ENOMEM;
+		kbuf = kzalloc(count, GFP_KERNEL);
+		if (!kbuf)
+			goto out;
 	}
 
-	if (!error)
+	error = BPF_CGROUP_RUN_PROG_SYSCTL(head, table, write, &kbuf, &count,
-		error = count;
+					   ppos);
+	if (error)
+		goto out_free_buf;
+
+	/* careful: calling conventions are nasty here */
+	error = table->proc_handler(table, write, kbuf, &count, ppos);
+	if (error)
+		goto out_free_buf;
+
+	if (!write) {
+		error = -EFAULT;
+		if (copy_to_user(ubuf, kbuf, count))
+			goto out_free_buf;
+	}
+
+	error = count;
+out_free_buf:
+	kfree(kbuf);
 out:
 	sysctl_head_finish(head);
 

fs/quota/dquot.c

@@ -2841,7 +2841,7 @@ const struct quotactl_ops dquot_quotactl_sysfile_ops = {
 EXPORT_SYMBOL(dquot_quotactl_sysfile_ops);
 
 static int do_proc_dqstats(struct ctl_table *table, int write,
-		     void __user *buffer, size_t *lenp, loff_t *ppos)
+		     void *buffer, size_t *lenp, loff_t *ppos)
 {
 	unsigned int type = (unsigned long *)table->data - dqstats.stat;
 	s64 value = percpu_counter_sum(&dqstats.counter[type]);

fs/xfs/xfs_sysctl.c

@@ -13,7 +13,7 @@ STATIC int
 xfs_stats_clear_proc_handler(
 	struct ctl_table	*ctl,
 	int			write,
-	void			__user *buffer,
+	void			*buffer,
 	size_t			*lenp,
 	loff_t			*ppos)
 {
@@ -33,7 +33,7 @@ STATIC int
 xfs_panic_mask_proc_handler(
 	struct ctl_table	*ctl,
 	int			write,
-	void			__user *buffer,
+	void			*buffer,
 	size_t			*lenp,
 	loff_t			*ppos)
 {

include/linux/bpf-cgroup.h

@@ -57,8 +57,6 @@ struct bpf_cgroup_link {
 	enum bpf_attach_type type;
 };
 
-extern const struct bpf_link_ops bpf_cgroup_link_lops;
-
 struct bpf_prog_list {
 	struct list_head node;
 	struct bpf_prog *prog;
@@ -100,8 +98,6 @@ int __cgroup_bpf_attach(struct cgroup *cgrp,
 int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 			struct bpf_cgroup_link *link,
 			enum bpf_attach_type type);
-int __cgroup_bpf_replace(struct cgroup *cgrp, struct bpf_cgroup_link *link,
-			 struct bpf_prog *new_prog);
 int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
 		       union bpf_attr __user *uattr);
 
@@ -112,8 +108,6 @@ int cgroup_bpf_attach(struct cgroup *cgrp,
 		      u32 flags);
 int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 		      enum bpf_attach_type type);
-int cgroup_bpf_replace(struct bpf_link *link, struct bpf_prog *old_prog,
-		       struct bpf_prog *new_prog);
 int cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
 		     union bpf_attr __user *uattr);
 
@@ -138,8 +132,7 @@ int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
 
 int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 				   struct ctl_table *table, int write,
-				   void __user *buf, size_t *pcount,
+				   void **buf, size_t *pcount, loff_t *ppos,
-				   loff_t *ppos, void **new_buf,
 				   enum bpf_attach_type type);
 
 int __cgroup_bpf_run_filter_setsockopt(struct sock *sock, int *level,
@@ -302,12 +295,12 @@ int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,
 })
 
 
-#define BPF_CGROUP_RUN_PROG_SYSCTL(head, table, write, buf, count, pos, nbuf)  \
+#define BPF_CGROUP_RUN_PROG_SYSCTL(head, table, write, buf, count, pos)  \
 ({									       \
 	int __ret = 0;							       \
 	if (cgroup_bpf_enabled)						       \
 		__ret = __cgroup_bpf_run_filter_sysctl(head, table, write,     \
-						       buf, count, pos, nbuf,  \
+						       buf, count, pos,        \
 						       BPF_CGROUP_SYSCTL);     \
 	__ret;								       \
 })
@@ -354,7 +347,6 @@ int cgroup_bpf_prog_query(const union bpf_attr *attr,
 #else
 
 struct bpf_prog;
-struct bpf_link;
 struct cgroup_bpf {};
 static inline int cgroup_bpf_inherit(struct cgroup *cgrp) { return 0; }
 static inline void cgroup_bpf_offline(struct cgroup *cgrp) {}
@@ -378,13 +370,6 @@ static inline int cgroup_bpf_link_attach(const union bpf_attr *attr,
 	return -EINVAL;
 }
 
-static inline int cgroup_bpf_replace(struct bpf_link *link,
-				     struct bpf_prog *old_prog,
-				     struct bpf_prog *new_prog)
-{
-	return -EINVAL;
-}
-
 static inline int cgroup_bpf_prog_query(const union bpf_attr *attr,
 					union bpf_attr __user *uattr)
 {
@@ -429,7 +414,7 @@ static inline int bpf_percpu_cgroup_storage_update(struct bpf_map *map,
 #define BPF_CGROUP_RUN_PROG_UDP6_RECVMSG_LOCK(sk, uaddr) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_SOCK_OPS(sock_ops) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type,major,minor,access) ({ 0; })
-#define BPF_CGROUP_RUN_PROG_SYSCTL(head,table,write,buf,count,pos,nbuf) ({ 0; })
+#define BPF_CGROUP_RUN_PROG_SYSCTL(head,table,write,buf,count,pos) ({ 0; })
 #define BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN(optlen) ({ 0; })
 #define BPF_CGROUP_RUN_PROG_GETSOCKOPT(sock, level, optname, optval, \
 				       optlen, max_optlen, retval) ({ retval; })

include/linux/bpf.h

@@ -987,6 +987,7 @@ _out:							\
 
 #ifdef CONFIG_BPF_SYSCALL
 DECLARE_PER_CPU(int, bpf_prog_active);
+extern struct mutex bpf_stats_enabled_mutex;
 
 /*
  * Block execution of BPF programs attached to instrumentation (perf,
@@ -1026,9 +1027,11 @@ extern const struct file_operations bpf_prog_fops;
 	extern const struct bpf_verifier_ops _name ## _verifier_ops;
 #define BPF_MAP_TYPE(_id, _ops) \
 	extern const struct bpf_map_ops _ops;
+#define BPF_LINK_TYPE(_id, _name)
 #include <linux/bpf_types.h>
 #undef BPF_PROG_TYPE
 #undef BPF_MAP_TYPE
+#undef BPF_LINK_TYPE
 
 extern const struct bpf_prog_ops bpf_offload_prog_ops;
 extern const struct bpf_verifier_ops tc_cls_act_analyzer_ops;
@@ -1085,21 +1088,35 @@ int bpf_prog_new_fd(struct bpf_prog *prog);
 
 struct bpf_link {
 	atomic64_t refcnt;
+	u32 id;
+	enum bpf_link_type type;
 	const struct bpf_link_ops *ops;
 	struct bpf_prog *prog;
 	struct work_struct work;
 };
 
+struct bpf_link_primer {
+	struct bpf_link *link;
+	struct file *file;
+	int fd;
+	u32 id;
+};
+
 struct bpf_link_ops {
 	void (*release)(struct bpf_link *link);
 	void (*dealloc)(struct bpf_link *link);
-
+	int (*update_prog)(struct bpf_link *link, struct bpf_prog *new_prog,
+			   struct bpf_prog *old_prog);
+	void (*show_fdinfo)(const struct bpf_link *link, struct seq_file *seq);
+	int (*fill_link_info)(const struct bpf_link *link,
+			      struct bpf_link_info *info);
 };
 
-void bpf_link_init(struct bpf_link *link, const struct bpf_link_ops *ops,
+void bpf_link_init(struct bpf_link *link, enum bpf_link_type type,
-		   struct bpf_prog *prog);
+		   const struct bpf_link_ops *ops, struct bpf_prog *prog);
-void bpf_link_cleanup(struct bpf_link *link, struct file *link_file,
+int bpf_link_prime(struct bpf_link *link, struct bpf_link_primer *primer);
-		      int link_fd);
+int bpf_link_settle(struct bpf_link_primer *primer);
+void bpf_link_cleanup(struct bpf_link_primer *primer);
 void bpf_link_inc(struct bpf_link *link);
 void bpf_link_put(struct bpf_link *link);
 int bpf_link_new_fd(struct bpf_link *link);
@@ -1215,6 +1232,7 @@ int btf_check_type_match(struct bpf_verifier_env *env, struct bpf_prog *prog,
 
 struct bpf_prog *bpf_prog_by_id(u32 id);
 
+const struct bpf_func_proto *bpf_base_func_proto(enum bpf_func_id func_id);
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
 {
@@ -1365,6 +1383,12 @@ static inline struct bpf_prog *bpf_prog_by_id(u32 id)
 {
 	return ERR_PTR(-ENOTSUPP);
 }
+
+static inline const struct bpf_func_proto *
+bpf_base_func_proto(enum bpf_func_id func_id)
+{
+	return NULL;
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 static inline struct bpf_prog *bpf_prog_get_type(u32 ufd,
@@ -1502,6 +1526,7 @@ extern const struct bpf_func_proto bpf_get_smp_processor_id_proto;
 extern const struct bpf_func_proto bpf_get_numa_node_id_proto;
 extern const struct bpf_func_proto bpf_tail_call_proto;
 extern const struct bpf_func_proto bpf_ktime_get_ns_proto;
+extern const struct bpf_func_proto bpf_ktime_get_boot_ns_proto;
 extern const struct bpf_func_proto bpf_get_current_pid_tgid_proto;
 extern const struct bpf_func_proto bpf_get_current_uid_gid_proto;
 extern const struct bpf_func_proto bpf_get_current_comm_proto;
@@ -1523,6 +1548,7 @@ extern const struct bpf_func_proto bpf_strtoul_proto;
 extern const struct bpf_func_proto bpf_tcp_sock_proto;
 extern const struct bpf_func_proto bpf_jiffies64_proto;
 extern const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto;
+extern const struct bpf_func_proto bpf_event_output_data_proto;
 
 const struct bpf_func_proto *bpf_tracing_func_proto(
 	enum bpf_func_id func_id, const struct bpf_prog *prog);
@@ -1530,6 +1556,7 @@ const struct bpf_func_proto *bpf_tracing_func_proto(
 /* Shared helpers among cBPF and eBPF. */
 void bpf_user_rnd_init_once(void);
 u64 bpf_user_rnd_u32(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
+u64 bpf_get_raw_cpu_id(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5);
 
 #if defined(CONFIG_NET)
 bool bpf_sock_common_is_valid_access(int off, int size,

include/linux/bpf_types.h

@@ -118,3 +118,9 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_STACK, stack_map_ops)
 #if defined(CONFIG_BPF_JIT)
 BPF_MAP_TYPE(BPF_MAP_TYPE_STRUCT_OPS, bpf_struct_ops_map_ops)
 #endif
+
+BPF_LINK_TYPE(BPF_LINK_TYPE_RAW_TRACEPOINT, raw_tracepoint)
+BPF_LINK_TYPE(BPF_LINK_TYPE_TRACING, tracing)
+#ifdef CONFIG_CGROUP_BPF
+BPF_LINK_TYPE(BPF_LINK_TYPE_CGROUP, cgroup)
+#endif

include/linux/compaction.h

@@ -86,7 +86,7 @@ static inline unsigned long compact_gap(unsigned int order)
 #ifdef CONFIG_COMPACTION
 extern int sysctl_compact_memory;
 extern int sysctl_compaction_handler(struct ctl_table *table, int write,
-			void __user *buffer, size_t *length, loff_t *ppos);
+			void *buffer, size_t *length, loff_t *ppos);
 extern int sysctl_extfrag_threshold;
 extern int sysctl_compact_unevictable_allowed;
 

include/linux/coredump.h

@@ -22,4 +22,8 @@ extern void do_coredump(const kernel_siginfo_t *siginfo);
 static inline void do_coredump(const kernel_siginfo_t *siginfo) {}
 #endif
 
+extern int core_uses_pid;
+extern char core_pattern[];
+extern unsigned int core_pipe_limit;
+
 #endif /* _LINUX_COREDUMP_H */

include/linux/file.h

@@ -94,4 +94,6 @@ extern void fd_install(unsigned int fd, struct file *file);
 extern void flush_delayed_fput(void);
 extern void __fput_sync(struct file *);
 
+extern unsigned int sysctl_nr_open_min, sysctl_nr_open_max;
+
 #endif /* __LINUX_FILE_H */

include/linux/filter.h

@@ -863,8 +863,6 @@ int bpf_prog_create(struct bpf_prog **pfp, struct sock_fprog_kern *fprog);
 int bpf_prog_create_from_user(struct bpf_prog **pfp, struct sock_fprog *fprog,
 			      bpf_aux_classic_check_t trans, bool save_orig);
 void bpf_prog_destroy(struct bpf_prog *fp);
-const struct bpf_func_proto *
-bpf_base_func_proto(enum bpf_func_id func_id);
 
 int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk);
 int sk_attach_bpf(u32 ufd, struct sock *sk);

include/linux/fs.h

@@ -3536,11 +3536,11 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf,
 
 struct ctl_table;
 int proc_nr_files(struct ctl_table *table, int write,
-		  void __user *buffer, size_t *lenp, loff_t *ppos);
+		  void *buffer, size_t *lenp, loff_t *ppos);
 int proc_nr_dentry(struct ctl_table *table, int write,
-		  void __user *buffer, size_t *lenp, loff_t *ppos);
+		  void *buffer, size_t *lenp, loff_t *ppos);
 int proc_nr_inodes(struct ctl_table *table, int write,
-		   void __user *buffer, size_t *lenp, loff_t *ppos);
+		   void *buffer, size_t *lenp, loff_t *ppos);
 int __init get_filesystem_list(char *buf);
 
 #define __FMODE_EXEC		((__force int) FMODE_EXEC)

include/linux/ftrace.h

@@ -1005,8 +1005,7 @@ extern void disable_trace_on_warning(void);
 extern int __disable_trace_on_warning;
 
 int tracepoint_printk_sysctl(struct ctl_table *table, int write,
-			     void __user *buffer, size_t *lenp,
+			     void *buffer, size_t *lenp, loff_t *ppos);
-			     loff_t *ppos);
 
 #else /* CONFIG_TRACING */
 static inline void  disable_trace_on_warning(void) { }

include/linux/hugetlb.h

@@ -105,14 +105,13 @@ struct hugepage_subpool *hugepage_new_subpool(struct hstate *h, long max_hpages,
 void hugepage_put_subpool(struct hugepage_subpool *spool);
 
 void reset_vma_resv_huge_pages(struct vm_area_struct *vma);
-int hugetlb_sysctl_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
+int hugetlb_sysctl_handler(struct ctl_table *, int, void *, size_t *, loff_t *);
-int hugetlb_overcommit_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
+int hugetlb_overcommit_handler(struct ctl_table *, int, void *, size_t *,
-int hugetlb_treat_movable_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
+		loff_t *);
-
+int hugetlb_treat_movable_handler(struct ctl_table *, int, void *, size_t *,
-#ifdef CONFIG_NUMA
+		loff_t *);
-int hugetlb_mempolicy_sysctl_handler(struct ctl_table *, int,
+int hugetlb_mempolicy_sysctl_handler(struct ctl_table *, int, void *, size_t *,
-					void __user *, size_t *, loff_t *);
+		loff_t *);
-#endif
 
 int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *);
 long follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *,

include/linux/kprobes.h

@@ -312,7 +312,7 @@ DEFINE_INSN_CACHE_OPS(optinsn);
 #ifdef CONFIG_SYSCTL
 extern int sysctl_kprobes_optimization;
 extern int proc_kprobes_optimization_handler(struct ctl_table *table,
-					     int write, void __user *buffer,
+					     int write, void *buffer,
 					     size_t *length, loff_t *ppos);
 #endif
 extern void wait_for_kprobe_optimizer(void);

include/linux/latencytop.h

@@ -38,8 +38,8 @@ account_scheduler_latency(struct task_struct *task, int usecs, int inter)
 
 void clear_tsk_latency_tracing(struct task_struct *p);
 
-extern int sysctl_latencytop(struct ctl_table *table, int write,
+int sysctl_latencytop(struct ctl_table *table, int write, void *buffer,
-			void __user *buffer, size_t *lenp, loff_t *ppos);
+		size_t *lenp, loff_t *ppos);
 
 #else
 

include/linux/mm.h

@@ -201,10 +201,10 @@ extern int sysctl_overcommit_memory;
 extern int sysctl_overcommit_ratio;
 extern unsigned long sysctl_overcommit_kbytes;
 
-extern int overcommit_ratio_handler(struct ctl_table *, int, void __user *,
+int overcommit_ratio_handler(struct ctl_table *, int, void *, size_t *,
-				    size_t *, loff_t *);
+		loff_t *);
-extern int overcommit_kbytes_handler(struct ctl_table *, int, void __user *,
+int overcommit_kbytes_handler(struct ctl_table *, int, void *, size_t *,
-				    size_t *, loff_t *);
+		loff_t *);
 
 #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
 
@@ -2957,8 +2957,8 @@ extern bool process_shares_mm(struct task_struct *p, struct mm_struct *mm);
 
 #ifdef CONFIG_SYSCTL
 extern int sysctl_drop_caches;
-int drop_caches_sysctl_handler(struct ctl_table *, int,
+int drop_caches_sysctl_handler(struct ctl_table *, int, void *, size_t *,
-					void __user *, size_t *, loff_t *);
+		loff_t *);
 #endif
 
 void drop_slab(void);
@@ -3140,5 +3140,7 @@ unsigned long wp_shared_mapping_range(struct address_space *mapping,
 				      pgoff_t first_index, pgoff_t nr);
 #endif
 
+extern int sysctl_nr_trim_pages;
+
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */

include/linux/mmzone.h

@@ -909,24 +909,23 @@ static inline int is_highmem(struct zone *zone)
 
 /* These two functions are used to setup the per zone pages min values */
 struct ctl_table;
-int min_free_kbytes_sysctl_handler(struct ctl_table *, int,
-					void __user *, size_t *, loff_t *);
-int watermark_boost_factor_sysctl_handler(struct ctl_table *, int,
-					void __user *, size_t *, loff_t *);
-int watermark_scale_factor_sysctl_handler(struct ctl_table *, int,
-					void __user *, size_t *, loff_t *);
-extern int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES];
-int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int,
-					void __user *, size_t *, loff_t *);
-int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *, int,
-					void __user *, size_t *, loff_t *);
-int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *, int,
-			void __user *, size_t *, loff_t *);
-int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *, int,
-			void __user *, size_t *, loff_t *);
 
-extern int numa_zonelist_order_handler(struct ctl_table *, int,
+int min_free_kbytes_sysctl_handler(struct ctl_table *, int, void *, size_t *,
-			void __user *, size_t *, loff_t *);
+		loff_t *);
+int watermark_scale_factor_sysctl_handler(struct ctl_table *, int, void *,
+		size_t *, loff_t *);
+extern int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES];
+int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, void *,
+		size_t *, loff_t *);
+int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *, int,
+		void *, size_t *, loff_t *);
+int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *, int,
+		void *, size_t *, loff_t *);
+int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *, int,
+		void *, size_t *, loff_t *);
+int numa_zonelist_order_handler(struct ctl_table *, int,
+		void *, size_t *, loff_t *);
+extern int percpu_pagelist_fraction;
 extern char numa_zonelist_order[];
 #define NUMA_ZONELIST_ORDER_LEN	16
 

include/linux/nmi.h

@@ -202,16 +202,11 @@ static inline void watchdog_update_hrtimer_threshold(u64 period) { }
 #endif
 
 struct ctl_table;
-extern int proc_watchdog(struct ctl_table *, int ,
+int proc_watchdog(struct ctl_table *, int, void *, size_t *, loff_t *);
-			 void __user *, size_t *, loff_t *);
+int proc_nmi_watchdog(struct ctl_table *, int , void *, size_t *, loff_t *);
-extern int proc_nmi_watchdog(struct ctl_table *, int ,
+int proc_soft_watchdog(struct ctl_table *, int , void *, size_t *, loff_t *);
-			     void __user *, size_t *, loff_t *);
+int proc_watchdog_thresh(struct ctl_table *, int , void *, size_t *, loff_t *);
-extern int proc_soft_watchdog(struct ctl_table *, int ,
+int proc_watchdog_cpumask(struct ctl_table *, int, void *, size_t *, loff_t *);
-			      void __user *, size_t *, loff_t *);
-extern int proc_watchdog_thresh(struct ctl_table *, int ,
-				void __user *, size_t *, loff_t *);
-extern int proc_watchdog_cpumask(struct ctl_table *, int,
-				 void __user *, size_t *, loff_t *);
 
 #ifdef CONFIG_HAVE_ACPI_APEI_NMI
 #include <asm/nmi.h>

include/linux/perf_event.h

@@ -1280,15 +1280,12 @@ extern int sysctl_perf_cpu_time_max_percent;
 
 extern void perf_sample_event_took(u64 sample_len_ns);
 
-extern int perf_proc_update_handler(struct ctl_table *table, int write,
+int perf_proc_update_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos);
-		loff_t *ppos);
+int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
-extern int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
+		void *buffer, size_t *lenp, loff_t *ppos);
-		void __user *buffer, size_t *lenp,
-		loff_t *ppos);
-
 int perf_event_max_stack_handler(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp, loff_t *ppos);
+		void *buffer, size_t *lenp, loff_t *ppos);
 
 /* Access to perf_event_open(2) syscall. */
 #define PERF_SECURITY_OPEN		0

include/linux/pid.h

@@ -108,6 +108,9 @@ extern void transfer_pid(struct task_struct *old, struct task_struct *new,
 struct pid_namespace;
 extern struct pid_namespace init_pid_ns;
 
+extern int pid_max;
+extern int pid_max_min, pid_max_max;
+
 /*
  * look up a PID in the hash table. Must be called with the tasklist_lock
  * or rcu_read_lock() held.

include/linux/printk.h

@@ -189,7 +189,7 @@ extern int printk_delay_msec;
 extern int dmesg_restrict;
 
 extern int
-devkmsg_sysctl_set_loglvl(struct ctl_table *table, int write, void __user *buf,
+devkmsg_sysctl_set_loglvl(struct ctl_table *table, int write, void *buf,
 			  size_t *lenp, loff_t *ppos);
 
 extern void wake_up_klogd(void);

include/linux/sched/sysctl.h

@@ -12,9 +12,8 @@ extern unsigned int  sysctl_hung_task_panic;
 extern unsigned long sysctl_hung_task_timeout_secs;
 extern unsigned long sysctl_hung_task_check_interval_secs;
 extern int sysctl_hung_task_warnings;
-extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
+int proc_dohung_task_timeout_secs(struct ctl_table *table, int write,
-					 void __user *buffer,
+		void *buffer, size_t *lenp, loff_t *ppos);
-					 size_t *lenp, loff_t *ppos);
 #else
 /* Avoid need for ifdefs elsewhere in the code */
 enum { sysctl_hung_task_timeout_secs = 0 };
@@ -43,8 +42,7 @@ extern __read_mostly unsigned int sysctl_sched_migration_cost;
 extern __read_mostly unsigned int sysctl_sched_nr_migrate;
 
 int sched_proc_update_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *length,
+		void *buffer, size_t *length, loff_t *ppos);
-		loff_t *ppos);
 #endif
 
 /*
@@ -72,33 +70,21 @@ extern unsigned int sysctl_sched_autogroup_enabled;
 extern int sysctl_sched_rr_timeslice;
 extern int sched_rr_timeslice;
 
-extern int sched_rr_handler(struct ctl_table *table, int write,
+int sched_rr_handler(struct ctl_table *table, int write, void *buffer,
-		void __user *buffer, size_t *lenp,
+		size_t *lenp, loff_t *ppos);
-		loff_t *ppos);
+int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
-
+		size_t *lenp, loff_t *ppos);
-extern int sched_rt_handler(struct ctl_table *table, int write,
+int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos);
-		loff_t *ppos);
+int sysctl_numa_balancing(struct ctl_table *table, int write, void *buffer,
-
+		size_t *lenp, loff_t *ppos);
-#ifdef CONFIG_UCLAMP_TASK
+int sysctl_schedstats(struct ctl_table *table, int write, void *buffer,
-extern int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
+		size_t *lenp, loff_t *ppos);
-				       void __user *buffer, size_t *lenp,
-				       loff_t *ppos);
-#endif
-
-extern int sysctl_numa_balancing(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp,
-				 loff_t *ppos);
-
-extern int sysctl_schedstats(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp,
-				 loff_t *ppos);
 
 #if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL)
 extern unsigned int sysctl_sched_energy_aware;
-extern int sched_energy_aware_handler(struct ctl_table *table, int write,
+int sched_energy_aware_handler(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos);
-				 loff_t *ppos);
 #endif
 
 #endif /* _LINUX_SCHED_SYSCTL_H */

include/linux/security.h

@@ -211,7 +211,7 @@ struct request_sock;
 
 #ifdef CONFIG_MMU
 extern int mmap_min_addr_handler(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp, loff_t *ppos);
+				 void *buffer, size_t *lenp, loff_t *ppos);
 #endif
 
 /* security_inode_init_security callback function to write xattrs */

include/linux/sysctl.h

@@ -44,35 +44,26 @@ struct ctl_dir;
 
 extern const int sysctl_vals[];
 
-typedef int proc_handler (struct ctl_table *ctl, int write,
+typedef int proc_handler(struct ctl_table *ctl, int write, void *buffer,
-			  void __user *buffer, size_t *lenp, loff_t *ppos);
+		size_t *lenp, loff_t *ppos);
 
-extern int proc_dostring(struct ctl_table *, int,
+int proc_dostring(struct ctl_table *, int, void *, size_t *, loff_t *);
-			 void __user *, size_t *, loff_t *);
+int proc_dointvec(struct ctl_table *, int, void *, size_t *, loff_t *);
-extern int proc_dointvec(struct ctl_table *, int,
+int proc_douintvec(struct ctl_table *, int, void *, size_t *, loff_t *);
-			 void __user *, size_t *, loff_t *);
+int proc_dointvec_minmax(struct ctl_table *, int, void *, size_t *, loff_t *);
-extern int proc_douintvec(struct ctl_table *, int,
+int proc_douintvec_minmax(struct ctl_table *table, int write, void *buffer,
-			 void __user *, size_t *, loff_t *);
+		size_t *lenp, loff_t *ppos);
-extern int proc_dointvec_minmax(struct ctl_table *, int,
+int proc_dointvec_jiffies(struct ctl_table *, int, void *, size_t *, loff_t *);
-				void __user *, size_t *, loff_t *);
+int proc_dointvec_userhz_jiffies(struct ctl_table *, int, void *, size_t *,
-extern int proc_douintvec_minmax(struct ctl_table *table, int write,
+		loff_t *);
-				 void __user *buffer, size_t *lenp,
+int proc_dointvec_ms_jiffies(struct ctl_table *, int, void *, size_t *,
-				 loff_t *ppos);
+		loff_t *);
-extern int proc_dointvec_jiffies(struct ctl_table *, int,
+int proc_doulongvec_minmax(struct ctl_table *, int, void *, size_t *, loff_t *);
-				 void __user *, size_t *, loff_t *);
+int proc_doulongvec_ms_jiffies_minmax(struct ctl_table *table, int, void *,
-extern int proc_dointvec_userhz_jiffies(struct ctl_table *, int,
+		size_t *, loff_t *);
-					void __user *, size_t *, loff_t *);
+int proc_do_large_bitmap(struct ctl_table *, int, void *, size_t *, loff_t *);
-extern int proc_dointvec_ms_jiffies(struct ctl_table *, int,
+int proc_do_static_key(struct ctl_table *table, int write, void *buffer,
-				    void __user *, size_t *, loff_t *);
+		size_t *lenp, loff_t *ppos);
-extern int proc_doulongvec_minmax(struct ctl_table *, int,
-				  void __user *, size_t *, loff_t *);
-extern int proc_doulongvec_ms_jiffies_minmax(struct ctl_table *table, int,
-				      void __user *, size_t *, loff_t *);
-extern int proc_do_large_bitmap(struct ctl_table *, int,
-				void __user *, size_t *, loff_t *);
-extern int proc_do_static_key(struct ctl_table *table, int write,
-			      void __user *buffer, size_t *lenp,
-			      loff_t *ppos);
 
 /*
  * Register a set of sysctl names by calling register_sysctl_table
@@ -207,7 +198,15 @@ void unregister_sysctl_table(struct ctl_table_header * table);
 
 extern int sysctl_init(void);
 
+extern int pwrsw_enabled;
+extern int unaligned_enabled;
+extern int unaligned_dump_stack;
+extern int no_unaligned_warning;
+
 extern struct ctl_table sysctl_mount_point[];
+extern struct ctl_table random_table[];
+extern struct ctl_table firmware_config_table[];
+extern struct ctl_table epoll_table[];
 
 #else /* CONFIG_SYSCTL */
 static inline struct ctl_table_header *register_sysctl_table(struct ctl_table * table)
@@ -238,7 +237,7 @@ static inline void setup_sysctl_set(struct ctl_table_set *p,
 
 #endif /* CONFIG_SYSCTL */
 
-int sysctl_max_threads(struct ctl_table *table, int write,
+int sysctl_max_threads(struct ctl_table *table, int write, void *buffer,
-		       void __user *buffer, size_t *lenp, loff_t *ppos);
+		size_t *lenp, loff_t *ppos);
 
 #endif /* _LINUX_SYSCTL_H */

include/linux/timer.h

@@ -201,8 +201,7 @@ struct ctl_table;
 
 extern unsigned int sysctl_timer_migration;
 int timer_migration_handler(struct ctl_table *table, int write,
-			    void __user *buffer, size_t *lenp,
+			    void *buffer, size_t *lenp, loff_t *ppos);
-			    loff_t *ppos);
 #endif
 
 unsigned long __round_jiffies(unsigned long j, int cpu);

include/linux/vmstat.h

@@ -16,8 +16,8 @@ extern int sysctl_stat_interval;
 #define DISABLE_NUMA_STAT   0
 extern int sysctl_vm_numa_stat;
 DECLARE_STATIC_KEY_TRUE(vm_numa_stat_key);
-extern int sysctl_vm_numa_stat_handler(struct ctl_table *table,
+int sysctl_vm_numa_stat_handler(struct ctl_table *table, int write,
-		int write, void __user *buffer, size_t *length, loff_t *ppos);
+		void *buffer, size_t *length, loff_t *ppos);
 #endif
 
 struct reclaim_stat {
@@ -274,8 +274,8 @@ void cpu_vm_stats_fold(int cpu);
 void refresh_zone_stat_thresholds(void);
 
 struct ctl_table;
-int vmstat_refresh(struct ctl_table *, int write,
+int vmstat_refresh(struct ctl_table *, int write, void *buffer, size_t *lenp,
-		   void __user *buffer, size_t *lenp, loff_t *ppos);
+		loff_t *ppos);
 
 void drain_zonestat(struct zone *zone, struct per_cpu_pageset *);
 

include/linux/writeback.h

@@ -362,24 +362,18 @@ extern int vm_highmem_is_dirtyable;
 extern int block_dump;
 extern int laptop_mode;
 
-extern int dirty_background_ratio_handler(struct ctl_table *table, int write,
+int dirty_background_ratio_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos);
-		loff_t *ppos);
+int dirty_background_bytes_handler(struct ctl_table *table, int write,
-extern int dirty_background_bytes_handler(struct ctl_table *table, int write,
+		void *buffer, size_t *lenp, loff_t *ppos);
-		void __user *buffer, size_t *lenp,
+int dirty_ratio_handler(struct ctl_table *table, int write,
-		loff_t *ppos);
+		void *buffer, size_t *lenp, loff_t *ppos);
-extern int dirty_ratio_handler(struct ctl_table *table, int write,
+int dirty_bytes_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos);
-		loff_t *ppos);
-extern int dirty_bytes_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
-		loff_t *ppos);
 int dirtytime_interval_handler(struct ctl_table *table, int write,
-			       void __user *buffer, size_t *lenp, loff_t *ppos);
+		void *buffer, size_t *lenp, loff_t *ppos);
-
+int dirty_writeback_centisecs_handler(struct ctl_table *table, int write,
-struct ctl_table;
+		void *buffer, size_t *lenp, loff_t *ppos);
-int dirty_writeback_centisecs_handler(struct ctl_table *, int,
-				      void __user *, size_t *, loff_t *);
 
 void global_dirty_limits(unsigned long *pbackground, unsigned long *pdirty);
 unsigned long wb_calc_thresh(struct bdi_writeback *wb, unsigned long thresh);

include/uapi/linux/bpf.h

@@ -113,6 +113,9 @@ enum bpf_cmd {
 	BPF_MAP_DELETE_BATCH,
 	BPF_LINK_CREATE,
 	BPF_LINK_UPDATE,
+	BPF_LINK_GET_FD_BY_ID,
+	BPF_LINK_GET_NEXT_ID,
+	BPF_ENABLE_STATS,
 };
 
 enum bpf_map_type {
@@ -220,6 +223,15 @@ enum bpf_attach_type {
 
 #define MAX_BPF_ATTACH_TYPE __MAX_BPF_ATTACH_TYPE
 
+enum bpf_link_type {
+	BPF_LINK_TYPE_UNSPEC = 0,
+	BPF_LINK_TYPE_RAW_TRACEPOINT = 1,
+	BPF_LINK_TYPE_TRACING = 2,
+	BPF_LINK_TYPE_CGROUP = 3,
+
+	MAX_BPF_LINK_TYPE,
+};
+
 /* cgroup-bpf attach flags used in BPF_PROG_ATTACH command
  *
  * NONE(default): No further bpf programs allowed in the subtree.
@@ -379,6 +391,12 @@ enum {
  */
 #define BPF_F_QUERY_EFFECTIVE	(1U << 0)
 
+/* type for BPF_ENABLE_STATS */
+enum bpf_stats_type {
+	/* enabled run_time_ns and run_cnt */
+	BPF_STATS_RUN_TIME = 0,
+};
+
 enum bpf_stack_build_id_status {
 	/* user space need an empty entry to identify end of a trace */
 	BPF_STACK_BUILD_ID_EMPTY = 0,
@@ -523,6 +541,7 @@ union bpf_attr {
 			__u32		prog_id;
 			__u32		map_id;
 			__u32		btf_id;
+			__u32		link_id;
 		};
 		__u32		next_id;
 		__u32		open_flags;
@@ -589,6 +608,10 @@ union bpf_attr {
 		__u32		old_prog_fd;
 	} link_update;
 
+	struct { /* struct used by BPF_ENABLE_STATS command */
+		__u32		type;
+	} enable_stats;
+
 } __attribute__((aligned(8)));
 
 /* The description below is an attempt at providing documentation to eBPF
@@ -652,6 +675,8 @@ union bpf_attr {
  * u64 bpf_ktime_get_ns(void)
  * 	Description
  * 		Return the time elapsed since system boot, in nanoseconds.
+ * 		Does not include time the system was suspended.
+ * 		See: clock_gettime(CLOCK_MONOTONIC)
  * 	Return
  * 		Current *ktime*.
  *
@@ -1562,7 +1587,7 @@ union bpf_attr {
  * 	Return
  * 		0
  *
- * int bpf_setsockopt(struct bpf_sock_ops *bpf_socket, int level, int optname, void *optval, int optlen)
+ * int bpf_setsockopt(void *bpf_socket, int level, int optname, void *optval, int optlen)
  * 	Description
  * 		Emulate a call to **setsockopt()** on the socket associated to
  * 		*bpf_socket*, which must be a full socket. The *level* at
@@ -1570,6 +1595,11 @@ union bpf_attr {
  * 		must be specified, see **setsockopt(2)** for more information.
  * 		The option value of length *optlen* is pointed by *optval*.
  *
+ * 		*bpf_socket* should be one of the following:
+ * 		* **struct bpf_sock_ops** for **BPF_PROG_TYPE_SOCK_OPS**.
+ * 		* **struct bpf_sock_addr** for **BPF_CGROUP_INET4_CONNECT**
+ * 		  and **BPF_CGROUP_INET6_CONNECT**.
+ *
  * 		This helper actually implements a subset of **setsockopt()**.
  * 		It supports the following *level*\ s:
  *
@@ -1764,7 +1794,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_getsockopt(struct bpf_sock_ops *bpf_socket, int level, int optname, void *optval, int optlen)
+ * int bpf_getsockopt(void *bpf_socket, int level, int optname, void *optval, int optlen)
  * 	Description
  * 		Emulate a call to **getsockopt()** on the socket associated to
  * 		*bpf_socket*, which must be a full socket. The *level* at
@@ -1773,6 +1803,11 @@ union bpf_attr {
  * 		The retrieved value is stored in the structure pointed by
  * 		*opval* and of length *optlen*.
  *
+ * 		*bpf_socket* should be one of the following:
+ * 		* **struct bpf_sock_ops** for **BPF_PROG_TYPE_SOCK_OPS**.
+ * 		* **struct bpf_sock_addr** for **BPF_CGROUP_INET4_CONNECT**
+ * 		  and **BPF_CGROUP_INET6_CONNECT**.
+ *
  * 		This helper actually implements a subset of **getsockopt()**.
  * 		It supports the following *level*\ s:
  *
@@ -3025,6 +3060,14 @@ union bpf_attr {
  *		* **-EOPNOTSUPP**	Unsupported operation, for example a
  *					call from outside of TC ingress.
  *		* **-ESOCKTNOSUPPORT**	Socket type not supported (reuseport).
+ *
+ * u64 bpf_ktime_get_boot_ns(void)
+ * 	Description
+ * 		Return the time elapsed since system boot, in nanoseconds.
+ * 		Does include the time the system was suspended.
+ * 		See: clock_gettime(CLOCK_BOOTTIME)
+ * 	Return
+ * 		Current *ktime*.
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -3151,7 +3194,8 @@ union bpf_attr {
 	FN(xdp_output),			\
 	FN(get_netns_cookie),		\
 	FN(get_current_ancestor_cgroup_id),	\
-	FN(sk_assign),
+	FN(sk_assign),			\
+	FN(ktime_get_boot_ns),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
@@ -3598,6 +3642,25 @@ struct bpf_btf_info {
 	__u32 id;
 } __attribute__((aligned(8)));
 
+struct bpf_link_info {
+	__u32 type;
+	__u32 id;
+	__u32 prog_id;
+	union {
+		struct {
+			__aligned_u64 tp_name; /* in/out: tp_name buffer ptr */
+			__u32 tp_name_len;     /* in/out: tp_name buffer len */
+		} raw_tracepoint;
+		struct {
+			__u32 attach_type;
+		} tracing;
+		struct {
+			__u64 cgroup_id;
+			__u32 attach_type;
+		} cgroup;
+	};
+} __attribute__((aligned(8)));
+
 /* User bpf_sock_addr struct to access socket fields and sockaddr struct passed
  * by user and intended to be used by socket (e.g. to bind to, depends on
  * attach attach type).

ipc/ipc_sysctl.c

@@ -24,7 +24,7 @@ static void *get_ipc(struct ctl_table *table)
 
 #ifdef CONFIG_PROC_SYSCTL
 static int proc_ipc_dointvec(struct ctl_table *table, int write,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table ipc_table;
 
@@ -35,7 +35,7 @@ static int proc_ipc_dointvec(struct ctl_table *table, int write,
 }
 
 static int proc_ipc_dointvec_minmax(struct ctl_table *table, int write,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table ipc_table;
 
@@ -46,7 +46,7 @@ static int proc_ipc_dointvec_minmax(struct ctl_table *table, int write,
 }
 
 static int proc_ipc_dointvec_minmax_orphans(struct ctl_table *table, int write,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ipc_namespace *ns = current->nsproxy->ipc_ns;
 	int err = proc_ipc_dointvec_minmax(table, write, buffer, lenp, ppos);
@@ -59,7 +59,7 @@ static int proc_ipc_dointvec_minmax_orphans(struct ctl_table *table, int write,
 }
 
 static int proc_ipc_doulongvec_minmax(struct ctl_table *table, int write,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table ipc_table;
 	memcpy(&ipc_table, table, sizeof(ipc_table));
@@ -70,7 +70,7 @@ static int proc_ipc_doulongvec_minmax(struct ctl_table *table, int write,
 }
 
 static int proc_ipc_auto_msgmni(struct ctl_table *table, int write,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table ipc_table;
 	int dummy = 0;

ipc/mq_sysctl.c

@@ -19,7 +19,7 @@ static void *get_mq(struct ctl_table *table)
 }
 
 static int proc_mq_dointvec(struct ctl_table *table, int write,
-			    void __user *buffer, size_t *lenp, loff_t *ppos)
+			    void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table mq_table;
 	memcpy(&mq_table, table, sizeof(mq_table));
@@ -29,7 +29,7 @@ static int proc_mq_dointvec(struct ctl_table *table, int write,
 }
 
 static int proc_mq_dointvec_minmax(struct ctl_table *table, int write,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table mq_table;
 	memcpy(&mq_table, table, sizeof(mq_table));

kernel/bpf/btf.c

@@ -3482,6 +3482,7 @@ extern char __weak __stop_BTF[];
 extern struct btf *btf_vmlinux;
 
 #define BPF_MAP_TYPE(_id, _ops)
+#define BPF_LINK_TYPE(_id, _name)
 static union {
 	struct bpf_ctx_convert {
 #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
@@ -3508,6 +3509,7 @@ static u8 bpf_ctx_convert_map[] = {
 	0, /* avoid empty array */
 };
 #undef BPF_MAP_TYPE
+#undef BPF_LINK_TYPE
 
 static const struct btf_member *
 btf_get_prog_ctx_type(struct bpf_verifier_log *log, struct btf *btf,

kernel/bpf/cgroup.c

@@ -557,8 +557,9 @@ found:
  *
  * Must be called with cgroup_mutex held.
  */
-int __cgroup_bpf_replace(struct cgroup *cgrp, struct bpf_cgroup_link *link,
+static int __cgroup_bpf_replace(struct cgroup *cgrp,
-			 struct bpf_prog *new_prog)
+				struct bpf_cgroup_link *link,
+				struct bpf_prog *new_prog)
 {
 	struct list_head *progs = &cgrp->bpf.progs[link->type];
 	struct bpf_prog *old_prog;
@@ -583,6 +584,30 @@ int __cgroup_bpf_replace(struct cgroup *cgrp, struct bpf_cgroup_link *link,
 	return 0;
 }
 
+static int cgroup_bpf_replace(struct bpf_link *link, struct bpf_prog *new_prog,
+			      struct bpf_prog *old_prog)
+{
+	struct bpf_cgroup_link *cg_link;
+	int ret;
+
+	cg_link = container_of(link, struct bpf_cgroup_link, link);
+
+	mutex_lock(&cgroup_mutex);
+	/* link might have been auto-released by dying cgroup, so fail */
+	if (!cg_link->cgroup) {
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+	if (old_prog && link->prog != old_prog) {
+		ret = -EPERM;
+		goto out_unlock;
+	}
+	ret = __cgroup_bpf_replace(cg_link->cgroup, cg_link, new_prog);
+out_unlock:
+	mutex_unlock(&cgroup_mutex);
+	return ret;
+}
+
 static struct bpf_prog_list *find_detach_entry(struct list_head *progs,
 					       struct bpf_prog *prog,
 					       struct bpf_cgroup_link *link,
@@ -808,17 +833,56 @@ static void bpf_cgroup_link_dealloc(struct bpf_link *link)
 	kfree(cg_link);
 }
 
-const struct bpf_link_ops bpf_cgroup_link_lops = {
+static void bpf_cgroup_link_show_fdinfo(const struct bpf_link *link,
+					struct seq_file *seq)
+{
+	struct bpf_cgroup_link *cg_link =
+		container_of(link, struct bpf_cgroup_link, link);
+	u64 cg_id = 0;
+
+	mutex_lock(&cgroup_mutex);
+	if (cg_link->cgroup)
+		cg_id = cgroup_id(cg_link->cgroup);
+	mutex_unlock(&cgroup_mutex);
+
+	seq_printf(seq,
+		   "cgroup_id:\t%llu\n"
+		   "attach_type:\t%d\n",
+		   cg_id,
+		   cg_link->type);
+}
+
+static int bpf_cgroup_link_fill_link_info(const struct bpf_link *link,
+					  struct bpf_link_info *info)
+{
+	struct bpf_cgroup_link *cg_link =
+		container_of(link, struct bpf_cgroup_link, link);
+	u64 cg_id = 0;
+
+	mutex_lock(&cgroup_mutex);
+	if (cg_link->cgroup)
+		cg_id = cgroup_id(cg_link->cgroup);
+	mutex_unlock(&cgroup_mutex);
+
+	info->cgroup.cgroup_id = cg_id;
+	info->cgroup.attach_type = cg_link->type;
+	return 0;
+}
+
+static const struct bpf_link_ops bpf_cgroup_link_lops = {
 	.release = bpf_cgroup_link_release,
 	.dealloc = bpf_cgroup_link_dealloc,
+	.update_prog = cgroup_bpf_replace,
+	.show_fdinfo = bpf_cgroup_link_show_fdinfo,
+	.fill_link_info = bpf_cgroup_link_fill_link_info,
 };
 
 int cgroup_bpf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
 {
+	struct bpf_link_primer link_primer;
 	struct bpf_cgroup_link *link;
-	struct file *link_file;
 	struct cgroup *cgrp;
-	int err, link_fd;
+	int err;
 
 	if (attr->link_create.flags)
 		return -EINVAL;
@@ -832,26 +896,25 @@ int cgroup_bpf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog)
 		err = -ENOMEM;
 		goto out_put_cgroup;
 	}
-	bpf_link_init(&link->link, &bpf_cgroup_link_lops, prog);
+	bpf_link_init(&link->link, BPF_LINK_TYPE_CGROUP, &bpf_cgroup_link_lops,
+		      prog);
 	link->cgroup = cgrp;
 	link->type = attr->link_create.attach_type;
 
-	link_file = bpf_link_new_file(&link->link, &link_fd);
+	err  = bpf_link_prime(&link->link, &link_primer);
-	if (IS_ERR(link_file)) {
+	if (err) {
 		kfree(link);
-		err = PTR_ERR(link_file);
 		goto out_put_cgroup;
 	}
 
 	err = cgroup_bpf_attach(cgrp, NULL, NULL, link, link->type,
 				BPF_F_ALLOW_MULTI);
 	if (err) {
-		bpf_link_cleanup(&link->link, link_file, link_fd);
+		bpf_link_cleanup(&link_primer);
 		goto out_put_cgroup;
 	}
 
-	fd_install(link_fd, link_file);
+	return bpf_link_settle(&link_primer);
-	return link_fd;
 
 out_put_cgroup:
 	cgroup_put(cgrp);
@@ -1054,36 +1117,21 @@ int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
 
 	return !allow;
 }
-EXPORT_SYMBOL(__cgroup_bpf_check_dev_permission);
 
 static const struct bpf_func_proto *
 cgroup_base_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 {
 	switch (func_id) {
-	case BPF_FUNC_map_lookup_elem:
-		return &bpf_map_lookup_elem_proto;
-	case BPF_FUNC_map_update_elem:
-		return &bpf_map_update_elem_proto;
-	case BPF_FUNC_map_delete_elem:
-		return &bpf_map_delete_elem_proto;
-	case BPF_FUNC_map_push_elem:
-		return &bpf_map_push_elem_proto;
-	case BPF_FUNC_map_pop_elem:
-		return &bpf_map_pop_elem_proto;
-	case BPF_FUNC_map_peek_elem:
-		return &bpf_map_peek_elem_proto;
 	case BPF_FUNC_get_current_uid_gid:
 		return &bpf_get_current_uid_gid_proto;
 	case BPF_FUNC_get_local_storage:
 		return &bpf_get_local_storage_proto;
 	case BPF_FUNC_get_current_cgroup_id:
 		return &bpf_get_current_cgroup_id_proto;
-	case BPF_FUNC_trace_printk:
+	case BPF_FUNC_perf_event_output:
-		if (capable(CAP_SYS_ADMIN))
+		return &bpf_event_output_data_proto;
-			return bpf_get_trace_printk_proto();
-		/* fall through */
 	default:
-		return NULL;
+		return bpf_base_func_proto(func_id);
 	}
 }
 
@@ -1137,16 +1185,13 @@ const struct bpf_verifier_ops cg_dev_verifier_ops = {
  * @head: sysctl table header
  * @table: sysctl table
  * @write: sysctl is being read (= 0) or written (= 1)
- * @buf: pointer to buffer passed by user space
+ * @buf: pointer to buffer (in and out)
  * @pcount: value-result argument: value is size of buffer pointed to by @buf,
  *	result is size of @new_buf if program set new value, initial value
  *	otherwise
  * @ppos: value-result argument: value is position at which read from or write
  *	to sysctl is happening, result is new position if program overrode it,
  *	initial value otherwise
- * @new_buf: pointer to pointer to new buffer that will be allocated if program
- *	overrides new value provided by user space on sysctl write
- *	NOTE: it's caller responsibility to free *new_buf if it was set
  * @type: type of program to be executed
  *
  * Program is run when sysctl is being accessed, either read or written, and
@@ -1157,8 +1202,7 @@ const struct bpf_verifier_ops cg_dev_verifier_ops = {
  */
 int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 				   struct ctl_table *table, int write,
-				   void __user *buf, size_t *pcount,
+				   void **buf, size_t *pcount, loff_t *ppos,
-				   loff_t *ppos, void **new_buf,
 				   enum bpf_attach_type type)
 {
 	struct bpf_sysctl_kern ctx = {
@@ -1173,36 +1217,28 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 		.new_updated = 0,
 	};
 	struct cgroup *cgrp;
+	loff_t pos = 0;
 	int ret;
 
 	ctx.cur_val = kmalloc_track_caller(ctx.cur_len, GFP_KERNEL);
-	if (ctx.cur_val) {
+	if (!ctx.cur_val ||
-		mm_segment_t old_fs;
+	    table->proc_handler(table, 0, ctx.cur_val, &ctx.cur_len, &pos)) {
-		loff_t pos = 0;
-
-		old_fs = get_fs();
-		set_fs(KERNEL_DS);
-		if (table->proc_handler(table, 0, (void __user *)ctx.cur_val,
-					&ctx.cur_len, &pos)) {
-			/* Let BPF program decide how to proceed. */
-			ctx.cur_len = 0;
-		}
-		set_fs(old_fs);
-	} else {
 		/* Let BPF program decide how to proceed. */
 		ctx.cur_len = 0;
 	}
 
-	if (write && buf && *pcount) {
+	if (write && *buf && *pcount) {
 		/* BPF program should be able to override new value with a
 		 * buffer bigger than provided by user.
 		 */
 		ctx.new_val = kmalloc_track_caller(PAGE_SIZE, GFP_KERNEL);
 		ctx.new_len = min_t(size_t, PAGE_SIZE, *pcount);
-		if (!ctx.new_val ||
+		if (ctx.new_val) {
-		    copy_from_user(ctx.new_val, buf, ctx.new_len))
+			memcpy(ctx.new_val, *buf, ctx.new_len);
+		} else {
 			/* Let BPF program decide how to proceed. */
 			ctx.new_len = 0;
+		}
 	}
 
 	rcu_read_lock();
@@ -1213,7 +1249,8 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 	kfree(ctx.cur_val);
 
 	if (ret == 1 && ctx.new_updated) {
-		*new_buf = ctx.new_val;
+		kfree(*buf);
+		*buf = ctx.new_val;
 		*pcount = ctx.new_len;
 	} else {
 		kfree(ctx.new_val);
@@ -1221,7 +1258,6 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 
 	return ret == 1 ? 0 : -EPERM;
 }
-EXPORT_SYMBOL(__cgroup_bpf_run_filter_sysctl);
 
 #ifdef CONFIG_NET
 static bool __cgroup_bpf_prog_array_is_empty(struct cgroup *cgrp,
@@ -1326,7 +1362,6 @@ out:
 		sockopt_free_buf(&ctx);
 	return ret;
 }
-EXPORT_SYMBOL(__cgroup_bpf_run_filter_setsockopt);
 
 int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
 				       int optname, char __user *optval,
@@ -1413,7 +1448,6 @@ out:
 	sockopt_free_buf(&ctx);
 	return ret;
 }
-EXPORT_SYMBOL(__cgroup_bpf_run_filter_getsockopt);
 #endif
 
 static ssize_t sysctl_cpy_dir(const struct ctl_dir *dir, char **bufp,

kernel/bpf/core.c

@@ -2136,6 +2136,11 @@ BPF_CALL_0(bpf_user_rnd_u32)
 	return res;
 }
 
+BPF_CALL_0(bpf_get_raw_cpu_id)
+{
+	return raw_smp_processor_id();
+}
+
 /* Weak definitions of helper functions in case we don't have bpf syscall. */
 const struct bpf_func_proto bpf_map_lookup_elem_proto __weak;
 const struct bpf_func_proto bpf_map_update_elem_proto __weak;
@@ -2151,6 +2156,7 @@ const struct bpf_func_proto bpf_get_prandom_u32_proto __weak;
 const struct bpf_func_proto bpf_get_smp_processor_id_proto __weak;
 const struct bpf_func_proto bpf_get_numa_node_id_proto __weak;
 const struct bpf_func_proto bpf_ktime_get_ns_proto __weak;
+const struct bpf_func_proto bpf_ktime_get_boot_ns_proto __weak;
 
 const struct bpf_func_proto bpf_get_current_pid_tgid_proto __weak;
 const struct bpf_func_proto bpf_get_current_uid_gid_proto __weak;

kernel/bpf/helpers.c

@@ -151,7 +151,19 @@ BPF_CALL_0(bpf_ktime_get_ns)
 
 const struct bpf_func_proto bpf_ktime_get_ns_proto = {
 	.func		= bpf_ktime_get_ns,
-	.gpl_only	= true,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+};
+
+BPF_CALL_0(bpf_ktime_get_boot_ns)
+{
+	/* NMI safe access to clock boottime */
+	return ktime_get_boot_fast_ns();
+}
+
+const struct bpf_func_proto bpf_ktime_get_boot_ns_proto = {
+	.func		= bpf_ktime_get_boot_ns,
+	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
 };
 
@@ -562,3 +574,78 @@ const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto = {
 	.arg3_type      = ARG_PTR_TO_UNINIT_MEM,
 	.arg4_type      = ARG_CONST_SIZE,
 };
+
+static const struct bpf_func_proto bpf_get_raw_smp_processor_id_proto = {
+	.func		= bpf_get_raw_cpu_id,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+};
+
+BPF_CALL_5(bpf_event_output_data, void *, ctx, struct bpf_map *, map,
+	   u64, flags, void *, data, u64, size)
+{
+	if (unlikely(flags & ~(BPF_F_INDEX_MASK)))
+		return -EINVAL;
+
+	return bpf_event_output(map, flags, data, size, NULL, 0, NULL);
+}
+
+const struct bpf_func_proto bpf_event_output_data_proto =  {
+	.func		= bpf_event_output_data,
+	.gpl_only       = true,
+	.ret_type       = RET_INTEGER,
+	.arg1_type      = ARG_PTR_TO_CTX,
+	.arg2_type      = ARG_CONST_MAP_PTR,
+	.arg3_type      = ARG_ANYTHING,
+	.arg4_type      = ARG_PTR_TO_MEM,
+	.arg5_type      = ARG_CONST_SIZE_OR_ZERO,
+};
+
+const struct bpf_func_proto *
+bpf_base_func_proto(enum bpf_func_id func_id)
+{
+	switch (func_id) {
+	case BPF_FUNC_map_lookup_elem:
+		return &bpf_map_lookup_elem_proto;
+	case BPF_FUNC_map_update_elem:
+		return &bpf_map_update_elem_proto;
+	case BPF_FUNC_map_delete_elem:
+		return &bpf_map_delete_elem_proto;
+	case BPF_FUNC_map_push_elem:
+		return &bpf_map_push_elem_proto;
+	case BPF_FUNC_map_pop_elem:
+		return &bpf_map_pop_elem_proto;
+	case BPF_FUNC_map_peek_elem:
+		return &bpf_map_peek_elem_proto;
+	case BPF_FUNC_get_prandom_u32:
+		return &bpf_get_prandom_u32_proto;
+	case BPF_FUNC_get_smp_processor_id:
+		return &bpf_get_raw_smp_processor_id_proto;
+	case BPF_FUNC_get_numa_node_id:
+		return &bpf_get_numa_node_id_proto;
+	case BPF_FUNC_tail_call:
+		return &bpf_tail_call_proto;
+	case BPF_FUNC_ktime_get_ns:
+		return &bpf_ktime_get_ns_proto;
+	case BPF_FUNC_ktime_get_boot_ns:
+		return &bpf_ktime_get_boot_ns_proto;
+	default:
+		break;
+	}
+
+	if (!capable(CAP_SYS_ADMIN))
+		return NULL;
+
+	switch (func_id) {
+	case BPF_FUNC_spin_lock:
+		return &bpf_spin_lock_proto;
+	case BPF_FUNC_spin_unlock:
+		return &bpf_spin_unlock_proto;
+	case BPF_FUNC_trace_printk:
+		return bpf_get_trace_printk_proto();
+	case BPF_FUNC_jiffies64:
+		return &bpf_jiffies64_proto;
+	default:
+		return NULL;
+	}
+}

kernel/bpf/syscall.c

@@ -42,6 +42,8 @@ static DEFINE_IDR(prog_idr);
 static DEFINE_SPINLOCK(prog_idr_lock);
 static DEFINE_IDR(map_idr);
 static DEFINE_SPINLOCK(map_idr_lock);
+static DEFINE_IDR(link_idr);
+static DEFINE_SPINLOCK(link_idr_lock);
 
 int sysctl_unprivileged_bpf_disabled __read_mostly;
 
@@ -49,9 +51,11 @@ static const struct bpf_map_ops * const bpf_map_types[] = {
 #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type)
 #define BPF_MAP_TYPE(_id, _ops) \
 	[_id] = &_ops,
+#define BPF_LINK_TYPE(_id, _name)
 #include <linux/bpf_types.h>
 #undef BPF_PROG_TYPE
 #undef BPF_MAP_TYPE
+#undef BPF_LINK_TYPE
 };
 
 /*
@@ -1546,9 +1550,11 @@ static const struct bpf_prog_ops * const bpf_prog_types[] = {
 #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
 	[_id] = & _name ## _prog_ops,
 #define BPF_MAP_TYPE(_id, _ops)
+#define BPF_LINK_TYPE(_id, _name)
 #include <linux/bpf_types.h>
 #undef BPF_PROG_TYPE
 #undef BPF_MAP_TYPE
+#undef BPF_LINK_TYPE
 };
 
 static int find_prog_type(enum bpf_prog_type type, struct bpf_prog *prog)
@@ -2181,25 +2187,39 @@ static int bpf_obj_get(const union bpf_attr *attr)
 				attr->file_flags);
 }
 
-void bpf_link_init(struct bpf_link *link, const struct bpf_link_ops *ops,
+void bpf_link_init(struct bpf_link *link, enum bpf_link_type type,
-		   struct bpf_prog *prog)
+		   const struct bpf_link_ops *ops, struct bpf_prog *prog)
 {
 	atomic64_set(&link->refcnt, 1);
+	link->type = type;
+	link->id = 0;
 	link->ops = ops;
 	link->prog = prog;
 }
 
+static void bpf_link_free_id(int id)
+{
+	if (!id)
+		return;
+
+	spin_lock_bh(&link_idr_lock);
+	idr_remove(&link_idr, id);
+	spin_unlock_bh(&link_idr_lock);
+}
+
 /* Clean up bpf_link and corresponding anon_inode file and FD. After
  * anon_inode is created, bpf_link can't be just kfree()'d due to deferred
- * anon_inode's release() call. This helper manages marking bpf_link as
+ * anon_inode's release() call. This helper marksbpf_link as
- * defunct, releases anon_inode file and puts reserved FD.
+ * defunct, releases anon_inode file and puts reserved FD. bpf_prog's refcnt
+ * is not decremented, it's the responsibility of a calling code that failed
+ * to complete bpf_link initialization.
  */
-void bpf_link_cleanup(struct bpf_link *link, struct file *link_file,
+void bpf_link_cleanup(struct bpf_link_primer *primer)
-		      int link_fd)
 {
-	link->prog = NULL;
+	primer->link->prog = NULL;
-	fput(link_file);
+	bpf_link_free_id(primer->id);
-	put_unused_fd(link_fd);
+	fput(primer->file);
+	put_unused_fd(primer->fd);
 }
 
 void bpf_link_inc(struct bpf_link *link)
@@ -2210,6 +2230,7 @@ void bpf_link_inc(struct bpf_link *link)
 /* bpf_link_free is guaranteed to be called from process context */
 static void bpf_link_free(struct bpf_link *link)
 {
+	bpf_link_free_id(link->id);
 	if (link->prog) {
 		/* detach BPF program, clean up used resources */
 		link->ops->release(link);
@@ -2251,35 +2272,35 @@ static int bpf_link_release(struct inode *inode, struct file *filp)
 }
 
 #ifdef CONFIG_PROC_FS
-static const struct bpf_link_ops bpf_raw_tp_lops;
+#define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type)
-static const struct bpf_link_ops bpf_tracing_link_lops;
+#define BPF_MAP_TYPE(_id, _ops)
+#define BPF_LINK_TYPE(_id, _name) [_id] = #_name,
+static const char *bpf_link_type_strs[] = {
+	[BPF_LINK_TYPE_UNSPEC] = "<invalid>",
+#include <linux/bpf_types.h>
+};
+#undef BPF_PROG_TYPE
+#undef BPF_MAP_TYPE
+#undef BPF_LINK_TYPE
 
 static void bpf_link_show_fdinfo(struct seq_file *m, struct file *filp)
 {
 	const struct bpf_link *link = filp->private_data;
 	const struct bpf_prog *prog = link->prog;
 	char prog_tag[sizeof(prog->tag) * 2 + 1] = { };
-	const char *link_type;
-
-	if (link->ops == &bpf_raw_tp_lops)
-		link_type = "raw_tracepoint";
-	else if (link->ops == &bpf_tracing_link_lops)
-		link_type = "tracing";
-#ifdef CONFIG_CGROUP_BPF
-	else if (link->ops == &bpf_cgroup_link_lops)
-		link_type = "cgroup";
-#endif
-	else
-		link_type = "unknown";
 
 	bin2hex(prog_tag, prog->tag, sizeof(prog->tag));
 	seq_printf(m,
 		   "link_type:\t%s\n"
+		   "link_id:\t%u\n"
 		   "prog_tag:\t%s\n"
 		   "prog_id:\t%u\n",
-		   link_type,
+		   bpf_link_type_strs[link->type],
+		   link->id,
 		   prog_tag,
 		   prog->aux->id);
+	if (link->ops->show_fdinfo)
+		link->ops->show_fdinfo(link, m);
 }
 #endif
 
@@ -2292,36 +2313,77 @@ static const struct file_operations bpf_link_fops = {
 	.write		= bpf_dummy_write,
 };
 
-int bpf_link_new_fd(struct bpf_link *link)
+static int bpf_link_alloc_id(struct bpf_link *link)
 {
-	return anon_inode_getfd("bpf-link", &bpf_link_fops, link, O_CLOEXEC);
+	int id;
+
+	idr_preload(GFP_KERNEL);
+	spin_lock_bh(&link_idr_lock);
+	id = idr_alloc_cyclic(&link_idr, link, 1, INT_MAX, GFP_ATOMIC);
+	spin_unlock_bh(&link_idr_lock);
+	idr_preload_end();
+
+	return id;
 }
 
-/* Similar to bpf_link_new_fd, create anon_inode for given bpf_link, but
+/* Prepare bpf_link to be exposed to user-space by allocating anon_inode file,
- * instead of immediately installing fd in fdtable, just reserve it and
+ * reserving unused FD and allocating ID from link_idr. This is to be paired
- * return. Caller then need to either install it with fd_install(fd, file) or
+ * with bpf_link_settle() to install FD and ID and expose bpf_link to
- * release with put_unused_fd(fd).
+ * user-space, if bpf_link is successfully attached. If not, bpf_link and
- * This is useful for cases when bpf_link attachment/detachment are
+ * pre-allocated resources are to be freed with bpf_cleanup() call. All the
- * complicated and expensive operations and should be delayed until all the fd
+ * transient state is passed around in struct bpf_link_primer.
- * reservation and anon_inode creation succeeds.
+ * This is preferred way to create and initialize bpf_link, especially when
+ * there are complicated and expensive operations inbetween creating bpf_link
+ * itself and attaching it to BPF hook. By using bpf_link_prime() and
+ * bpf_link_settle() kernel code using bpf_link doesn't have to perform
+ * expensive (and potentially failing) roll back operations in a rare case
+ * that file, FD, or ID can't be allocated.
  */
-struct file *bpf_link_new_file(struct bpf_link *link, int *reserved_fd)
+int bpf_link_prime(struct bpf_link *link, struct bpf_link_primer *primer)
 {
 	struct file *file;
-	int fd;
+	int fd, id;
 
 	fd = get_unused_fd_flags(O_CLOEXEC);
 	if (fd < 0)
-		return ERR_PTR(fd);
+		return fd;
+
+
+	id = bpf_link_alloc_id(link);
+	if (id < 0) {
+		put_unused_fd(fd);
+		return id;
+	}
 
 	file = anon_inode_getfile("bpf_link", &bpf_link_fops, link, O_CLOEXEC);
 	if (IS_ERR(file)) {
+		bpf_link_free_id(id);
 		put_unused_fd(fd);
-		return file;
+		return PTR_ERR(file);
 	}
 
-	*reserved_fd = fd;
+	primer->link = link;
-	return file;
+	primer->file = file;
+	primer->fd = fd;
+	primer->id = id;
+	return 0;
+}
+
+int bpf_link_settle(struct bpf_link_primer *primer)
+{
+	/* make bpf_link fetchable by ID */
+	spin_lock_bh(&link_idr_lock);
+	primer->link->id = primer->id;
+	spin_unlock_bh(&link_idr_lock);
+	/* make bpf_link fetchable by FD */
+	fd_install(primer->fd, primer->file);
+	/* pass through installed FD */
+	return primer->fd;
+}
+
+int bpf_link_new_fd(struct bpf_link *link)
+{
+	return anon_inode_getfd("bpf-link", &bpf_link_fops, link, O_CLOEXEC);
 }
 
 struct bpf_link *bpf_link_get_from_fd(u32 ufd)
@@ -2345,6 +2407,7 @@ struct bpf_link *bpf_link_get_from_fd(u32 ufd)
 
 struct bpf_tracing_link {
 	struct bpf_link link;
+	enum bpf_attach_type attach_type;
 };
 
 static void bpf_tracing_link_release(struct bpf_link *link)
@@ -2360,16 +2423,40 @@ static void bpf_tracing_link_dealloc(struct bpf_link *link)
 	kfree(tr_link);
 }
 
+static void bpf_tracing_link_show_fdinfo(const struct bpf_link *link,
+					 struct seq_file *seq)
+{
+	struct bpf_tracing_link *tr_link =
+		container_of(link, struct bpf_tracing_link, link);
+
+	seq_printf(seq,
+		   "attach_type:\t%d\n",
+		   tr_link->attach_type);
+}
+
+static int bpf_tracing_link_fill_link_info(const struct bpf_link *link,
+					   struct bpf_link_info *info)
+{
+	struct bpf_tracing_link *tr_link =
+		container_of(link, struct bpf_tracing_link, link);
+
+	info->tracing.attach_type = tr_link->attach_type;
+
+	return 0;
+}
+
 static const struct bpf_link_ops bpf_tracing_link_lops = {
 	.release = bpf_tracing_link_release,
 	.dealloc = bpf_tracing_link_dealloc,
+	.show_fdinfo = bpf_tracing_link_show_fdinfo,
+	.fill_link_info = bpf_tracing_link_fill_link_info,
 };
 
 static int bpf_tracing_prog_attach(struct bpf_prog *prog)
 {
+	struct bpf_link_primer link_primer;
 	struct bpf_tracing_link *link;
-	struct file *link_file;
+	int err;
-	int link_fd, err;
 
 	switch (prog->type) {
 	case BPF_PROG_TYPE_TRACING:
@@ -2402,24 +2489,23 @@ static int bpf_tracing_prog_attach(struct bpf_prog *prog)
 		err = -ENOMEM;
 		goto out_put_prog;
 	}
-	bpf_link_init(&link->link, &bpf_tracing_link_lops, prog);
+	bpf_link_init(&link->link, BPF_LINK_TYPE_TRACING,
+		      &bpf_tracing_link_lops, prog);
+	link->attach_type = prog->expected_attach_type;
 
-	link_file = bpf_link_new_file(&link->link, &link_fd);
+	err = bpf_link_prime(&link->link, &link_primer);
-	if (IS_ERR(link_file)) {
+	if (err) {
 		kfree(link);
-		err = PTR_ERR(link_file);
 		goto out_put_prog;
 	}
 
 	err = bpf_trampoline_link_prog(prog);
 	if (err) {
-		bpf_link_cleanup(&link->link, link_file, link_fd);
+		bpf_link_cleanup(&link_primer);
 		goto out_put_prog;
 	}
 
-	fd_install(link_fd, link_file);
+	return bpf_link_settle(&link_primer);
-	return link_fd;
-
 out_put_prog:
 	bpf_prog_put(prog);
 	return err;
@@ -2447,22 +2533,69 @@ static void bpf_raw_tp_link_dealloc(struct bpf_link *link)
 	kfree(raw_tp);
 }
 
-static const struct bpf_link_ops bpf_raw_tp_lops = {
+static void bpf_raw_tp_link_show_fdinfo(const struct bpf_link *link,
+					struct seq_file *seq)
+{
+	struct bpf_raw_tp_link *raw_tp_link =
+		container_of(link, struct bpf_raw_tp_link, link);
+
+	seq_printf(seq,
+		   "tp_name:\t%s\n",
+		   raw_tp_link->btp->tp->name);
+}
+
+static int bpf_raw_tp_link_fill_link_info(const struct bpf_link *link,
+					  struct bpf_link_info *info)
+{
+	struct bpf_raw_tp_link *raw_tp_link =
+		container_of(link, struct bpf_raw_tp_link, link);
+	char __user *ubuf = u64_to_user_ptr(info->raw_tracepoint.tp_name);
+	const char *tp_name = raw_tp_link->btp->tp->name;
+	u32 ulen = info->raw_tracepoint.tp_name_len;
+	size_t tp_len = strlen(tp_name);
+
+	if (ulen && !ubuf)
+		return -EINVAL;
+
+	info->raw_tracepoint.tp_name_len = tp_len + 1;
+
+	if (!ubuf)
+		return 0;
+
+	if (ulen >= tp_len + 1) {
+		if (copy_to_user(ubuf, tp_name, tp_len + 1))
+			return -EFAULT;
+	} else {
+		char zero = '\0';
+
+		if (copy_to_user(ubuf, tp_name, ulen - 1))
+			return -EFAULT;
+		if (put_user(zero, ubuf + ulen - 1))
+			return -EFAULT;
+		return -ENOSPC;
+	}
+
+	return 0;
+}
+
+static const struct bpf_link_ops bpf_raw_tp_link_lops = {
 	.release = bpf_raw_tp_link_release,
 	.dealloc = bpf_raw_tp_link_dealloc,
+	.show_fdinfo = bpf_raw_tp_link_show_fdinfo,
+	.fill_link_info = bpf_raw_tp_link_fill_link_info,
 };
 
 #define BPF_RAW_TRACEPOINT_OPEN_LAST_FIELD raw_tracepoint.prog_fd
 
 static int bpf_raw_tracepoint_open(const union bpf_attr *attr)
 {
+	struct bpf_link_primer link_primer;
 	struct bpf_raw_tp_link *link;
 	struct bpf_raw_event_map *btp;
-	struct file *link_file;
 	struct bpf_prog *prog;
 	const char *tp_name;
 	char buf[128];
-	int link_fd, err;
+	int err;
 
 	if (CHECK_ATTR(BPF_RAW_TRACEPOINT_OPEN))
 		return -EINVAL;
@@ -2515,24 +2648,23 @@ static int bpf_raw_tracepoint_open(const union bpf_attr *attr)
 		err = -ENOMEM;
 		goto out_put_btp;
 	}
-	bpf_link_init(&link->link, &bpf_raw_tp_lops, prog);
+	bpf_link_init(&link->link, BPF_LINK_TYPE_RAW_TRACEPOINT,
+		      &bpf_raw_tp_link_lops, prog);
 	link->btp = btp;
 
-	link_file = bpf_link_new_file(&link->link, &link_fd);
+	err = bpf_link_prime(&link->link, &link_primer);
-	if (IS_ERR(link_file)) {
+	if (err) {
 		kfree(link);
-		err = PTR_ERR(link_file);
 		goto out_put_btp;
 	}
 
 	err = bpf_probe_register(link->btp, prog);
 	if (err) {
-		bpf_link_cleanup(&link->link, link_file, link_fd);
+		bpf_link_cleanup(&link_primer);
 		goto out_put_btp;
 	}
 
-	fd_install(link_fd, link_file);
+	return bpf_link_settle(&link_primer);
-	return link_fd;
 
 out_put_btp:
 	bpf_put_raw_tracepoint(btp);
@@ -3313,6 +3445,42 @@ static int bpf_btf_get_info_by_fd(struct btf *btf,
 	return btf_get_info_by_fd(btf, attr, uattr);
 }
 
+static int bpf_link_get_info_by_fd(struct bpf_link *link,
+				  const union bpf_attr *attr,
+				  union bpf_attr __user *uattr)
+{
+	struct bpf_link_info __user *uinfo = u64_to_user_ptr(attr->info.info);
+	struct bpf_link_info info;
+	u32 info_len = attr->info.info_len;
+	int err;
+
+	err = bpf_check_uarg_tail_zero(uinfo, sizeof(info), info_len);
+	if (err)
+		return err;
+	info_len = min_t(u32, sizeof(info), info_len);
+
+	memset(&info, 0, sizeof(info));
+	if (copy_from_user(&info, uinfo, info_len))
+		return -EFAULT;
+
+	info.type = link->type;
+	info.id = link->id;
+	info.prog_id = link->prog->aux->id;
+
+	if (link->ops->fill_link_info) {
+		err = link->ops->fill_link_info(link, &info);
+		if (err)
+			return err;
+	}
+
+	if (copy_to_user(uinfo, &info, info_len) ||
+	    put_user(info_len, &uattr->info.info_len))
+		return -EFAULT;
+
+	return 0;
+}
+
+
 #define BPF_OBJ_GET_INFO_BY_FD_LAST_FIELD info.info
 
 static int bpf_obj_get_info_by_fd(const union bpf_attr *attr,
@@ -3337,6 +3505,9 @@ static int bpf_obj_get_info_by_fd(const union bpf_attr *attr,
 					     uattr);
 	else if (f.file->f_op == &btf_fops)
 		err = bpf_btf_get_info_by_fd(f.file->private_data, attr, uattr);
+	else if (f.file->f_op == &bpf_link_fops)
+		err = bpf_link_get_info_by_fd(f.file->private_data,
+					      attr, uattr);
 	else
 		err = -EINVAL;
 
@@ -3464,7 +3635,7 @@ static int bpf_task_fd_query(const union bpf_attr *attr,
 	if (file->f_op == &bpf_link_fops) {
 		struct bpf_link *link = file->private_data;
 
-		if (link->ops == &bpf_raw_tp_lops) {
+		if (link->ops == &bpf_raw_tp_link_lops) {
 			struct bpf_raw_tp_link *raw_tp =
 				container_of(link, struct bpf_raw_tp_link, link);
 			struct bpf_raw_event_map *btp = raw_tp->btp;
@@ -3645,13 +3816,10 @@ static int link_update(union bpf_attr *attr)
 		goto out_put_progs;
 	}
 
-#ifdef CONFIG_CGROUP_BPF
+	if (link->ops->update_prog)
-	if (link->ops == &bpf_cgroup_link_lops) {
+		ret = link->ops->update_prog(link, new_prog, old_prog);
-		ret = cgroup_bpf_replace(link, old_prog, new_prog);
+	else
-		goto out_put_progs;
+		ret = EINVAL;
-	}
-#endif
-	ret = -EINVAL;
 
 out_put_progs:
 	if (old_prog)
@@ -3663,6 +3831,102 @@ out_put_link:
 	return ret;
 }
 
+static int bpf_link_inc_not_zero(struct bpf_link *link)
+{
+	return atomic64_fetch_add_unless(&link->refcnt, 1, 0) ? 0 : -ENOENT;
+}
+
+#define BPF_LINK_GET_FD_BY_ID_LAST_FIELD link_id
+
+static int bpf_link_get_fd_by_id(const union bpf_attr *attr)
+{
+	struct bpf_link *link;
+	u32 id = attr->link_id;
+	int fd, err;
+
+	if (CHECK_ATTR(BPF_LINK_GET_FD_BY_ID))
+		return -EINVAL;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	spin_lock_bh(&link_idr_lock);
+	link = idr_find(&link_idr, id);
+	/* before link is "settled", ID is 0, pretend it doesn't exist yet */
+	if (link) {
+		if (link->id)
+			err = bpf_link_inc_not_zero(link);
+		else
+			err = -EAGAIN;
+	} else {
+		err = -ENOENT;
+	}
+	spin_unlock_bh(&link_idr_lock);
+
+	if (err)
+		return err;
+
+	fd = bpf_link_new_fd(link);
+	if (fd < 0)
+		bpf_link_put(link);
+
+	return fd;
+}
+
+DEFINE_MUTEX(bpf_stats_enabled_mutex);
+
+static int bpf_stats_release(struct inode *inode, struct file *file)
+{
+	mutex_lock(&bpf_stats_enabled_mutex);
+	static_key_slow_dec(&bpf_stats_enabled_key.key);
+	mutex_unlock(&bpf_stats_enabled_mutex);
+	return 0;
+}
+
+static const struct file_operations bpf_stats_fops = {
+	.release = bpf_stats_release,
+};
+
+static int bpf_enable_runtime_stats(void)
+{
+	int fd;
+
+	mutex_lock(&bpf_stats_enabled_mutex);
+
+	/* Set a very high limit to avoid overflow */
+	if (static_key_count(&bpf_stats_enabled_key.key) > INT_MAX / 2) {
+		mutex_unlock(&bpf_stats_enabled_mutex);
+		return -EBUSY;
+	}
+
+	fd = anon_inode_getfd("bpf-stats", &bpf_stats_fops, NULL, O_CLOEXEC);
+	if (fd >= 0)
+		static_key_slow_inc(&bpf_stats_enabled_key.key);
+
+	mutex_unlock(&bpf_stats_enabled_mutex);
+	return fd;
+}
+
+#define BPF_ENABLE_STATS_LAST_FIELD enable_stats.type
+
+static int bpf_enable_stats(union bpf_attr *attr)
+{
+
+	if (CHECK_ATTR(BPF_ENABLE_STATS))
+		return -EINVAL;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	switch (attr->enable_stats.type) {
+	case BPF_STATS_RUN_TIME:
+		return bpf_enable_runtime_stats();
+	default:
+		break;
+	}
+	return -EINVAL;
+}
+
 SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, size)
 {
 	union bpf_attr attr;
@@ -3780,6 +4044,16 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
 	case BPF_LINK_UPDATE:
 		err = link_update(&attr);
 		break;
+	case BPF_LINK_GET_FD_BY_ID:
+		err = bpf_link_get_fd_by_id(&attr);
+		break;
+	case BPF_LINK_GET_NEXT_ID:
+		err = bpf_obj_get_next_id(&attr, uattr,
+					  &link_idr, &link_idr_lock);
+		break;
+	case BPF_ENABLE_STATS:
+		err = bpf_enable_stats(&attr);
+		break;
 	default:
 		err = -EINVAL;
 		break;

kernel/bpf/verifier.c

@@ -28,9 +28,11 @@ static const struct bpf_verifier_ops * const bpf_verifier_ops[] = {
 #define BPF_PROG_TYPE(_id, _name, prog_ctx_type, kern_ctx_type) \
 	[_id] = & _name ## _verifier_ops,
 #define BPF_MAP_TYPE(_id, _ops)
+#define BPF_LINK_TYPE(_id, _name)
 #include <linux/bpf_types.h>
 #undef BPF_PROG_TYPE
 #undef BPF_MAP_TYPE
+#undef BPF_LINK_TYPE
 };
 
 /* bpf_check() is a static code analyzer that walks eBPF program
@@ -168,6 +170,8 @@ struct bpf_verifier_stack_elem {
 	int insn_idx;
 	int prev_insn_idx;
 	struct bpf_verifier_stack_elem *next;
+	/* length of verifier log at the time this state was pushed on stack */
+	u32 log_pos;
 };
 
 #define BPF_COMPLEXITY_LIMIT_JMP_SEQ	8192
@@ -283,6 +287,18 @@ void bpf_verifier_vlog(struct bpf_verifier_log *log, const char *fmt,
 		log->ubuf = NULL;
 }
 
+static void bpf_vlog_reset(struct bpf_verifier_log *log, u32 new_pos)
+{
+	char zero = 0;
+
+	if (!bpf_verifier_log_needed(log))
+		return;
+
+	log->len_used = new_pos;
+	if (put_user(zero, log->ubuf + new_pos))
+		log->ubuf = NULL;
+}
+
 /* log_level controls verbosity level of eBPF verifier.
  * bpf_verifier_log_write() is used to dump the verification trace to the log,
  * so the user can figure out what's wrong with the program
@@ -413,11 +429,30 @@ static bool is_release_function(enum bpf_func_id func_id)
 	return func_id == BPF_FUNC_sk_release;
 }
 
-static bool is_acquire_function(enum bpf_func_id func_id)
+static bool may_be_acquire_function(enum bpf_func_id func_id)
 {
 	return func_id == BPF_FUNC_sk_lookup_tcp ||
 		func_id == BPF_FUNC_sk_lookup_udp ||
-		func_id == BPF_FUNC_skc_lookup_tcp;
+		func_id == BPF_FUNC_skc_lookup_tcp ||
+		func_id == BPF_FUNC_map_lookup_elem;
+}
+
+static bool is_acquire_function(enum bpf_func_id func_id,
+				const struct bpf_map *map)
+{
+	enum bpf_map_type map_type = map ? map->map_type : BPF_MAP_TYPE_UNSPEC;
+
+	if (func_id == BPF_FUNC_sk_lookup_tcp ||
+	    func_id == BPF_FUNC_sk_lookup_udp ||
+	    func_id == BPF_FUNC_skc_lookup_tcp)
+		return true;
+
+	if (func_id == BPF_FUNC_map_lookup_elem &&
+	    (map_type == BPF_MAP_TYPE_SOCKMAP ||
+	     map_type == BPF_MAP_TYPE_SOCKHASH))
+		return true;
+
+	return false;
 }
 
 static bool is_ptr_cast_function(enum bpf_func_id func_id)
@@ -846,7 +881,7 @@ static void update_branch_counts(struct bpf_verifier_env *env, struct bpf_verifi
 }
 
 static int pop_stack(struct bpf_verifier_env *env, int *prev_insn_idx,
-		     int *insn_idx)
+		     int *insn_idx, bool pop_log)
 {
 	struct bpf_verifier_state *cur = env->cur_state;
 	struct bpf_verifier_stack_elem *elem, *head = env->head;
@@ -860,6 +895,8 @@ static int pop_stack(struct bpf_verifier_env *env, int *prev_insn_idx,
 		if (err)
 			return err;
 	}
+	if (pop_log)
+		bpf_vlog_reset(&env->log, head->log_pos);
 	if (insn_idx)
 		*insn_idx = head->insn_idx;
 	if (prev_insn_idx)
@@ -887,6 +924,7 @@ static struct bpf_verifier_state *push_stack(struct bpf_verifier_env *env,
 	elem->insn_idx = insn_idx;
 	elem->prev_insn_idx = prev_insn_idx;
 	elem->next = env->head;
+	elem->log_pos = env->log.len_used;
 	env->head = elem;
 	env->stack_size++;
 	err = copy_verifier_state(&elem->st, cur);
@@ -915,7 +953,7 @@ err:
 	free_verifier_state(env->cur_state, true);
 	env->cur_state = NULL;
 	/* pop all elements and return */
-	while (!pop_stack(env, NULL, NULL));
+	while (!pop_stack(env, NULL, NULL, false));
 	return NULL;
 }
 
@@ -3915,7 +3953,8 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 		    func_id != BPF_FUNC_sock_map_update &&
 		    func_id != BPF_FUNC_map_delete_elem &&
 		    func_id != BPF_FUNC_msg_redirect_map &&
-		    func_id != BPF_FUNC_sk_select_reuseport)
+		    func_id != BPF_FUNC_sk_select_reuseport &&
+		    func_id != BPF_FUNC_map_lookup_elem)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_SOCKHASH:
@@ -3923,7 +3962,8 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 		    func_id != BPF_FUNC_sock_hash_update &&
 		    func_id != BPF_FUNC_map_delete_elem &&
 		    func_id != BPF_FUNC_msg_redirect_hash &&
-		    func_id != BPF_FUNC_sk_select_reuseport)
+		    func_id != BPF_FUNC_sk_select_reuseport &&
+		    func_id != BPF_FUNC_map_lookup_elem)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_REUSEPORT_SOCKARRAY:
@@ -4093,7 +4133,7 @@ static bool check_refcount_ok(const struct bpf_func_proto *fn, int func_id)
 	/* A reference acquiring function cannot acquire
 	 * another refcounted ptr.
 	 */
-	if (is_acquire_function(func_id) && count)
+	if (may_be_acquire_function(func_id) && count)
 		return false;
 
 	/* We only support one arg being unreferenced at the moment,
@@ -4604,7 +4644,7 @@ static int check_helper_call(struct bpf_verifier_env *env, int func_id, int insn
 	if (is_ptr_cast_function(func_id)) {
 		/* For release_reference() */
 		regs[BPF_REG_0].ref_obj_id = meta.ref_obj_id;
-	} else if (is_acquire_function(func_id)) {
+	} else if (is_acquire_function(func_id, meta.map_ptr)) {
 		int id = acquire_reference_state(env, insn_idx);
 
 		if (id < 0)
@@ -5609,7 +5649,7 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 {
 	struct bpf_reg_state *regs = cur_regs(env);
 	u8 opcode = BPF_OP(insn->code);
-	bool src_known, dst_known;
+	bool src_known;
 	s64 smin_val, smax_val;
 	u64 umin_val, umax_val;
 	s32 s32_min_val, s32_max_val;
@@ -5631,7 +5671,6 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 
 	if (alu32) {
 		src_known = tnum_subreg_is_const(src_reg.var_off);
-		dst_known = tnum_subreg_is_const(dst_reg->var_off);
 		if ((src_known &&
 		     (s32_min_val != s32_max_val || u32_min_val != u32_max_val)) ||
 		    s32_min_val > s32_max_val || u32_min_val > u32_max_val) {
@@ -5643,7 +5682,6 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 		}
 	} else {
 		src_known = tnum_is_const(src_reg.var_off);
-		dst_known = tnum_is_const(dst_reg->var_off);
 		if ((src_known &&
 		     (smin_val != smax_val || umin_val != umax_val)) ||
 		    smin_val > smax_val || umin_val > umax_val) {
@@ -6515,12 +6553,16 @@ static void mark_ptr_or_null_reg(struct bpf_func_state *state,
 		if (is_null) {
 			reg->type = SCALAR_VALUE;
 		} else if (reg->type == PTR_TO_MAP_VALUE_OR_NULL) {
-			if (reg->map_ptr->inner_map_meta) {
+			const struct bpf_map *map = reg->map_ptr;
+
+			if (map->inner_map_meta) {
 				reg->type = CONST_PTR_TO_MAP;
-				reg->map_ptr = reg->map_ptr->inner_map_meta;
+				reg->map_ptr = map->inner_map_meta;
-			} else if (reg->map_ptr->map_type ==
+			} else if (map->map_type == BPF_MAP_TYPE_XSKMAP) {
-				   BPF_MAP_TYPE_XSKMAP) {
 				reg->type = PTR_TO_XDP_SOCK;
+			} else if (map->map_type == BPF_MAP_TYPE_SOCKMAP ||
+				   map->map_type == BPF_MAP_TYPE_SOCKHASH) {
+				reg->type = PTR_TO_SOCKET;
 			} else {
 				reg->type = PTR_TO_MAP_VALUE;
 			}
@@ -8409,6 +8451,7 @@ static bool reg_type_mismatch(enum bpf_reg_type src, enum bpf_reg_type prev)
 
 static int do_check(struct bpf_verifier_env *env)
 {
+	bool pop_log = !(env->log.level & BPF_LOG_LEVEL2);
 	struct bpf_verifier_state *state = env->cur_state;
 	struct bpf_insn *insns = env->prog->insnsi;
 	struct bpf_reg_state *regs;
@@ -8685,7 +8728,7 @@ static int do_check(struct bpf_verifier_env *env)
 process_bpf_exit:
 				update_branch_counts(env, env->cur_state);
 				err = pop_stack(env, &prev_insn_idx,
-						&env->insn_idx);
+						&env->insn_idx, pop_log);
 				if (err < 0) {
 					if (err != -ENOENT)
 						return err;
@@ -10208,6 +10251,7 @@ static void sanitize_insn_aux_data(struct bpf_verifier_env *env)
 
 static int do_check_common(struct bpf_verifier_env *env, int subprog)
 {
+	bool pop_log = !(env->log.level & BPF_LOG_LEVEL2);
 	struct bpf_verifier_state *state;
 	struct bpf_reg_state *regs;
 	int ret, i;
@@ -10270,7 +10314,9 @@ out:
 		free_verifier_state(env->cur_state, true);
 		env->cur_state = NULL;
 	}
-	while (!pop_stack(env, NULL, NULL));
+	while (!pop_stack(env, NULL, NULL, false));
+	if (!ret && pop_log)
+		bpf_vlog_reset(&env->log, 0);
 	free_states(env);
 	if (ret)
 		/* clean aux data in case subprog was rejected */

kernel/cgroup/cgroup.c

@@ -6508,33 +6508,6 @@ int cgroup_bpf_attach(struct cgroup *cgrp,
 	return ret;
 }
 
-int cgroup_bpf_replace(struct bpf_link *link, struct bpf_prog *old_prog,
-		       struct bpf_prog *new_prog)
-{
-	struct bpf_cgroup_link *cg_link;
-	int ret;
-
-	if (link->ops != &bpf_cgroup_link_lops)
-		return -EINVAL;
-
-	cg_link = container_of(link, struct bpf_cgroup_link, link);
-
-	mutex_lock(&cgroup_mutex);
-	/* link might have been auto-released by dying cgroup, so fail */
-	if (!cg_link->cgroup) {
-		ret = -EINVAL;
-		goto out_unlock;
-	}
-	if (old_prog && link->prog != old_prog) {
-		ret = -EPERM;
-		goto out_unlock;
-	}
-	ret = __cgroup_bpf_replace(cg_link->cgroup, cg_link, new_prog);
-out_unlock:
-	mutex_unlock(&cgroup_mutex);
-	return ret;
-}
-
 int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 		      enum bpf_attach_type type)
 {

kernel/events/callchain.c

@@ -236,7 +236,7 @@ exit_put:
  * sysctl_perf_event_max_contexts_per_stack.
  */
 int perf_event_max_stack_handler(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp, loff_t *ppos)
+				 void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int *value = table->data;
 	int new_value = *value, ret;

kernel/events/core.c

@@ -437,8 +437,7 @@ static void update_perf_cpu_limits(void)
 static bool perf_rotate_context(struct perf_cpu_context *cpuctx);
 
 int perf_proc_update_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int ret;
 	int perf_cpu = sysctl_perf_cpu_time_max_percent;
@@ -462,8 +461,7 @@ int perf_proc_update_handler(struct ctl_table *table, int write,
 int sysctl_perf_cpu_time_max_percent __read_mostly = DEFAULT_CPU_TIME_MAX_PERCENT;
 
 int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos)
-				loff_t *ppos)
 {
 	int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
 

kernel/kprobes.c

@@ -892,7 +892,7 @@ static void unoptimize_all_kprobes(void)
 static DEFINE_MUTEX(kprobe_sysctl_mutex);
 int sysctl_kprobes_optimization;
 int proc_kprobes_optimization_handler(struct ctl_table *table, int write,
-				      void __user *buffer, size_t *length,
+				      void *buffer, size_t *length,
 				      loff_t *ppos)
 {
 	int ret;

kernel/latencytop.c

@@ -269,8 +269,8 @@ static int __init init_lstats_procfs(void)
 	return 0;
 }
 
-int sysctl_latencytop(struct ctl_table *table, int write,
+int sysctl_latencytop(struct ctl_table *table, int write, void *buffer,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
+		size_t *lenp, loff_t *ppos)
 {
 	int err;
 

kernel/pid_namespace.c

@@ -263,7 +263,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns)
 
 #ifdef CONFIG_CHECKPOINT_RESTORE
 static int pid_ns_ctl_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct pid_namespace *pid_ns = task_active_pid_ns(current);
 	struct ctl_table tmp = *table;

kernel/printk/printk.c

@@ -173,7 +173,7 @@ __setup("printk.devkmsg=", control_devkmsg);
 char devkmsg_log_str[DEVKMSG_STR_MAX_SIZE] = "ratelimit";
 
 int devkmsg_sysctl_set_loglvl(struct ctl_table *table, int write,
-			      void __user *buffer, size_t *lenp, loff_t *ppos)
+			      void *buffer, size_t *lenp, loff_t *ppos)
 {
 	char old_str[DEVKMSG_STR_MAX_SIZE];
 	unsigned int old;

kernel/sched/core.c

@@ -1110,8 +1110,7 @@ static void uclamp_update_root_tg(void) { }
 #endif
 
 int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp,
+				void *buffer, size_t *lenp, loff_t *ppos)
-				loff_t *ppos)
 {
 	bool update_root_tg = false;
 	int old_min, old_max;
@@ -2718,7 +2717,7 @@ void set_numabalancing_state(bool enabled)
 
 #ifdef CONFIG_PROC_SYSCTL
 int sysctl_numa_balancing(struct ctl_table *table, int write,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+			  void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table t;
 	int err;
@@ -2792,8 +2791,8 @@ static void __init init_schedstats(void)
 }
 
 #ifdef CONFIG_PROC_SYSCTL
-int sysctl_schedstats(struct ctl_table *table, int write,
+int sysctl_schedstats(struct ctl_table *table, int write, void *buffer,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+		size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table t;
 	int err;

kernel/sched/fair.c

@@ -645,8 +645,7 @@ struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq)
  */
 
 int sched_proc_update_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
 	unsigned int factor = get_update_sysctl_factor();

kernel/sched/rt.c

@@ -2714,9 +2714,8 @@ static void sched_rt_do_global(void)
 	def_rt_bandwidth.rt_period = ns_to_ktime(global_rt_period());
 }
 
-int sched_rt_handler(struct ctl_table *table, int write,
+int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
-		void __user *buffer, size_t *lenp,
+		size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int old_period, old_runtime;
 	static DEFINE_MUTEX(mutex);
@@ -2754,9 +2753,8 @@ undo:
 	return ret;
 }
 
-int sched_rr_handler(struct ctl_table *table, int write,
+int sched_rr_handler(struct ctl_table *table, int write, void *buffer,
-		void __user *buffer, size_t *lenp,
+		size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int ret;
 	static DEFINE_MUTEX(mutex);

kernel/sched/topology.c

@@ -209,7 +209,7 @@ bool sched_energy_update;
 
 #ifdef CONFIG_PROC_SYSCTL
 int sched_energy_aware_handler(struct ctl_table *table, int write,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+		void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret, state;
 

kernel/seccomp.c

@@ -1776,7 +1776,7 @@ static void audit_actions_logged(u32 actions_logged, u32 old_actions_logged,
 }
 
 static int seccomp_actions_logged_handler(struct ctl_table *ro_table, int write,
-					  void __user *buffer, size_t *lenp,
+					  void *buffer, size_t *lenp,
 					  loff_t *ppos)
 {
 	int ret;

kernel/time/timer.c

@@ -249,8 +249,7 @@ void timers_update_nohz(void)
 }
 
 int timer_migration_handler(struct ctl_table *table, int write,
-			    void __user *buffer, size_t *lenp,
+			    void *buffer, size_t *lenp, loff_t *ppos)
-			    loff_t *ppos)
 {
 	int ret;
 

kernel/trace/bpf_trace.c

@@ -797,6 +797,8 @@ bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_map_peek_elem_proto;
 	case BPF_FUNC_ktime_get_ns:
 		return &bpf_ktime_get_ns_proto;
+	case BPF_FUNC_ktime_get_boot_ns:
+		return &bpf_ktime_get_boot_ns_proto;
 	case BPF_FUNC_tail_call:
 		return &bpf_tail_call_proto;
 	case BPF_FUNC_get_current_pid_tgid:

kernel/trace/trace.c

@@ -2661,7 +2661,7 @@ static void output_printk(struct trace_event_buffer *fbuffer)
 }
 
 int tracepoint_printk_sysctl(struct ctl_table *table, int write,
-			     void __user *buffer, size_t *lenp,
+			     void *buffer, size_t *lenp,
 			     loff_t *ppos)
 {
 	int save_tracepoint_printk;

kernel/umh.c

@@ -630,7 +630,7 @@ int call_usermodehelper(const char *path, char **argv, char **envp, int wait)
 EXPORT_SYMBOL(call_usermodehelper);
 
 static int proc_cap_handler(struct ctl_table *table, int write,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+			 void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table t;
 	unsigned long cap_array[_KERNEL_CAPABILITY_U32S];

kernel/utsname_sysctl.c

@@ -30,7 +30,7 @@ static void *get_uts(struct ctl_table *table)
  *	to observe. Should this be in kernel/sys.c ????
  */
 static int proc_do_uts_string(struct ctl_table *table, int write,
-		  void __user *buffer, size_t *lenp, loff_t *ppos)
+		  void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table uts_table;
 	int r;

kernel/watchdog.c

@@ -661,7 +661,7 @@ static void proc_watchdog_update(void)
  * proc_soft_watchdog | soft_watchdog_user_enabled | SOFT_WATCHDOG_ENABLED
  */
 static int proc_watchdog_common(int which, struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int err, old, *param = table->data;
 
@@ -688,7 +688,7 @@ static int proc_watchdog_common(int which, struct ctl_table *table, int write,
  * /proc/sys/kernel/watchdog
  */
 int proc_watchdog(struct ctl_table *table, int write,
-		  void __user *buffer, size_t *lenp, loff_t *ppos)
+		  void *buffer, size_t *lenp, loff_t *ppos)
 {
 	return proc_watchdog_common(NMI_WATCHDOG_ENABLED|SOFT_WATCHDOG_ENABLED,
 				    table, write, buffer, lenp, ppos);
@@ -698,7 +698,7 @@ int proc_watchdog(struct ctl_table *table, int write,
  * /proc/sys/kernel/nmi_watchdog
  */
 int proc_nmi_watchdog(struct ctl_table *table, int write,
-		      void __user *buffer, size_t *lenp, loff_t *ppos)
+		      void *buffer, size_t *lenp, loff_t *ppos)
 {
 	if (!nmi_watchdog_available && write)
 		return -ENOTSUPP;
@@ -710,7 +710,7 @@ int proc_nmi_watchdog(struct ctl_table *table, int write,
  * /proc/sys/kernel/soft_watchdog
  */
 int proc_soft_watchdog(struct ctl_table *table, int write,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
+			void *buffer, size_t *lenp, loff_t *ppos)
 {
 	return proc_watchdog_common(SOFT_WATCHDOG_ENABLED,
 				    table, write, buffer, lenp, ppos);
@@ -720,7 +720,7 @@ int proc_soft_watchdog(struct ctl_table *table, int write,
  * /proc/sys/kernel/watchdog_thresh
  */
 int proc_watchdog_thresh(struct ctl_table *table, int write,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+			 void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int err, old;
 
@@ -743,7 +743,7 @@ int proc_watchdog_thresh(struct ctl_table *table, int write,
  * been brought online, if desired.
  */
 int proc_watchdog_cpumask(struct ctl_table *table, int write,
-			  void __user *buffer, size_t *lenp, loff_t *ppos)
+			  void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int err;
 

mm/compaction.c

@@ -2463,7 +2463,7 @@ int sysctl_compact_memory;
  * /proc/sys/vm/compact_memory
  */
 int sysctl_compaction_handler(struct ctl_table *table, int write,
-			void __user *buffer, size_t *length, loff_t *ppos)
+			void *buffer, size_t *length, loff_t *ppos)
 {
 	if (write)
 		compact_nodes();

mm/hugetlb.c

@@ -3352,7 +3352,7 @@ static unsigned int cpuset_mems_nr(unsigned int *array)
 #ifdef CONFIG_SYSCTL
 static int hugetlb_sysctl_handler_common(bool obey_mempolicy,
 			 struct ctl_table *table, int write,
-			 void __user *buffer, size_t *length, loff_t *ppos)
+			 void *buffer, size_t *length, loff_t *ppos)
 {
 	struct hstate *h = &default_hstate;
 	unsigned long tmp = h->max_huge_pages;
@@ -3375,7 +3375,7 @@ out:
 }
 
 int hugetlb_sysctl_handler(struct ctl_table *table, int write,
-			  void __user *buffer, size_t *length, loff_t *ppos)
+			  void *buffer, size_t *length, loff_t *ppos)
 {
 
 	return hugetlb_sysctl_handler_common(false, table, write,
@@ -3384,7 +3384,7 @@ int hugetlb_sysctl_handler(struct ctl_table *table, int write,
 
 #ifdef CONFIG_NUMA
 int hugetlb_mempolicy_sysctl_handler(struct ctl_table *table, int write,
-			  void __user *buffer, size_t *length, loff_t *ppos)
+			  void *buffer, size_t *length, loff_t *ppos)
 {
 	return hugetlb_sysctl_handler_common(true, table, write,
 							buffer, length, ppos);
@@ -3392,8 +3392,7 @@ int hugetlb_mempolicy_sysctl_handler(struct ctl_table *table, int write,
 #endif /* CONFIG_NUMA */
 
 int hugetlb_overcommit_handler(struct ctl_table *table, int write,
-			void __user *buffer,
+		void *buffer, size_t *length, loff_t *ppos)
-			size_t *length, loff_t *ppos)
 {
 	struct hstate *h = &default_hstate;
 	unsigned long tmp;

mm/page-writeback.c

@@ -512,8 +512,7 @@ bool node_dirty_ok(struct pglist_data *pgdat)
 }
 
 int dirty_background_ratio_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int ret;
 
@@ -524,8 +523,7 @@ int dirty_background_ratio_handler(struct ctl_table *table, int write,
 }
 
 int dirty_background_bytes_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int ret;
 
@@ -535,9 +533,8 @@ int dirty_background_bytes_handler(struct ctl_table *table, int write,
 	return ret;
 }
 
-int dirty_ratio_handler(struct ctl_table *table, int write,
+int dirty_ratio_handler(struct ctl_table *table, int write, void *buffer,
-		void __user *buffer, size_t *lenp,
+		size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	int old_ratio = vm_dirty_ratio;
 	int ret;
@@ -551,8 +548,7 @@ int dirty_ratio_handler(struct ctl_table *table, int write,
 }
 
 int dirty_bytes_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *lenp,
+		void *buffer, size_t *lenp, loff_t *ppos)
-		loff_t *ppos)
 {
 	unsigned long old_bytes = vm_dirty_bytes;
 	int ret;
@@ -1972,7 +1968,7 @@ bool wb_over_bg_thresh(struct bdi_writeback *wb)
  * sysctl handler for /proc/sys/vm/dirty_writeback_centisecs
  */
 int dirty_writeback_centisecs_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	unsigned int old_interval = dirty_writeback_interval;
 	int ret;

mm/page_alloc.c

@@ -5546,21 +5546,11 @@ char numa_zonelist_order[] = "Node";
  * sysctl handler for numa_zonelist_order
  */
 int numa_zonelist_order_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *length,
+		void *buffer, size_t *length, loff_t *ppos)
-		loff_t *ppos)
 {
-	char *str;
+	if (write)
-	int ret;
+		return __parse_numa_zonelist_order(buffer);
-
+	return proc_dostring(table, write, buffer, length, ppos);
-	if (!write)
-		return proc_dostring(table, write, buffer, length, ppos);
-	str = memdup_user_nul(buffer, 16);
-	if (IS_ERR(str))
-		return PTR_ERR(str);
-
-	ret = __parse_numa_zonelist_order(str);
-	kfree(str);
-	return ret;
 }
 
 
@@ -7963,7 +7953,7 @@ core_initcall(init_per_zone_wmark_min)
  *	changes.
  */
 int min_free_kbytes_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	int rc;
 
@@ -7978,20 +7968,8 @@ int min_free_kbytes_sysctl_handler(struct ctl_table *table, int write,
 	return 0;
 }
 
-int watermark_boost_factor_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
-{
-	int rc;
-
-	rc = proc_dointvec_minmax(table, write, buffer, length, ppos);
-	if (rc)
-		return rc;
-
-	return 0;
-}
-
 int watermark_scale_factor_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	int rc;
 
@@ -8021,7 +7999,7 @@ static void setup_min_unmapped_ratio(void)
 
 
 int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	int rc;
 
@@ -8048,7 +8026,7 @@ static void setup_min_slab_ratio(void)
 }
 
 int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	int rc;
 
@@ -8072,7 +8050,7 @@ int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *table, int write,
  * if in function of the boot time zone sizes.
  */
 int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	proc_dointvec_minmax(table, write, buffer, length, ppos);
 	setup_per_zone_lowmem_reserve();
@@ -8094,7 +8072,7 @@ static void __zone_pcp_update(struct zone *zone)
  * pagelist can have before it gets flushed back to buddy allocator.
  */
 int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *table, int write,
-	void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	struct zone *zone;
 	int old_percpu_pagelist_fraction;

mm/util.c

@@ -717,9 +717,8 @@ int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT;
 unsigned long sysctl_user_reserve_kbytes __read_mostly = 1UL << 17; /* 128MB */
 unsigned long sysctl_admin_reserve_kbytes __read_mostly = 1UL << 13; /* 8MB */
 
-int overcommit_ratio_handler(struct ctl_table *table, int write,
+int overcommit_ratio_handler(struct ctl_table *table, int write, void *buffer,
-			     void __user *buffer, size_t *lenp,
+		size_t *lenp, loff_t *ppos)
-			     loff_t *ppos)
 {
 	int ret;
 
@@ -729,9 +728,8 @@ int overcommit_ratio_handler(struct ctl_table *table, int write,
 	return ret;
 }
 
-int overcommit_kbytes_handler(struct ctl_table *table, int write,
+int overcommit_kbytes_handler(struct ctl_table *table, int write, void *buffer,
-			     void __user *buffer, size_t *lenp,
+		size_t *lenp, loff_t *ppos)
-			     loff_t *ppos)
 {
 	int ret;
 

mm/vmstat.c

@@ -76,7 +76,7 @@ static void invalid_numa_statistics(void)
 static DEFINE_MUTEX(vm_numa_stat_lock);
 
 int sysctl_vm_numa_stat_handler(struct ctl_table *table, int write,
-		void __user *buffer, size_t *length, loff_t *ppos)
+		void *buffer, size_t *length, loff_t *ppos)
 {
 	int ret, oldval;
 
@@ -1751,7 +1751,7 @@ static void refresh_vm_stats(struct work_struct *work)
 }
 
 int vmstat_refresh(struct ctl_table *table, int write,
-		   void __user *buffer, size_t *lenp, loff_t *ppos)
+		   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	long val;
 	int err;

net/bridge/br_netfilter_hooks.c

@@ -1027,7 +1027,7 @@ int br_nf_hook_thresh(unsigned int hook, struct net *net,
 #ifdef CONFIG_SYSCTL
 static
 int brnf_sysctl_call_tables(struct ctl_table *ctl, int write,
-			    void __user *buffer, size_t *lenp, loff_t *ppos)
+			    void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret;
 

net/core/filter.c

@@ -256,17 +256,6 @@ BPF_CALL_2(bpf_skb_load_helper_32_no_cache, const struct sk_buff *, skb,
 					  offset);
 }
 
-BPF_CALL_0(bpf_get_raw_cpu_id)
-{
-	return raw_smp_processor_id();
-}
-
-static const struct bpf_func_proto bpf_get_raw_smp_processor_id_proto = {
-	.func		= bpf_get_raw_cpu_id,
-	.gpl_only	= false,
-	.ret_type	= RET_INTEGER,
-};
-
 static u32 convert_skb_access(int skb_field, int dst_reg, int src_reg,
 			      struct bpf_insn *insn_buf)
 {
@@ -4205,36 +4194,19 @@ static const struct bpf_func_proto bpf_get_socket_uid_proto = {
 	.arg1_type      = ARG_PTR_TO_CTX,
 };
 
-BPF_CALL_5(bpf_event_output_data, void *, ctx, struct bpf_map *, map, u64, flags,
+#define SOCKOPT_CC_REINIT (1 << 0)
-	   void *, data, u64, size)
+
+static int _bpf_setsockopt(struct sock *sk, int level, int optname,
+			   char *optval, int optlen, u32 flags)
 {
-	if (unlikely(flags & ~(BPF_F_INDEX_MASK)))
-		return -EINVAL;
-
-	return bpf_event_output(map, flags, data, size, NULL, 0, NULL);
-}
-
-static const struct bpf_func_proto bpf_event_output_data_proto =  {
-	.func		= bpf_event_output_data,
-	.gpl_only       = true,
-	.ret_type       = RET_INTEGER,
-	.arg1_type      = ARG_PTR_TO_CTX,
-	.arg2_type      = ARG_CONST_MAP_PTR,
-	.arg3_type      = ARG_ANYTHING,
-	.arg4_type      = ARG_PTR_TO_MEM,
-	.arg5_type      = ARG_CONST_SIZE_OR_ZERO,
-};
-
-BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock,
-	   int, level, int, optname, char *, optval, int, optlen)
-{
-	struct sock *sk = bpf_sock->sk;
 	int ret = 0;
 	int val;
 
 	if (!sk_fullsock(sk))
 		return -EINVAL;
 
+	sock_owned_by_me(sk);
+
 	if (level == SOL_SOCKET) {
 		if (optlen != sizeof(int))
 			return -EINVAL;
@@ -4329,7 +4301,7 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock,
 		   sk->sk_prot->setsockopt == tcp_setsockopt) {
 		if (optname == TCP_CONGESTION) {
 			char name[TCP_CA_NAME_MAX];
-			bool reinit = bpf_sock->op > BPF_SOCK_OPS_NEEDS_ECN;
+			bool reinit = flags & SOCKOPT_CC_REINIT;
 
 			strncpy(name, optval, min_t(long, optlen,
 						    TCP_CA_NAME_MAX-1));
@@ -4376,24 +4348,14 @@ BPF_CALL_5(bpf_setsockopt, struct bpf_sock_ops_kern *, bpf_sock,
 	return ret;
 }
 
-static const struct bpf_func_proto bpf_setsockopt_proto = {
+static int _bpf_getsockopt(struct sock *sk, int level, int optname,
-	.func		= bpf_setsockopt,
+			   char *optval, int optlen)
-	.gpl_only	= false,
-	.ret_type	= RET_INTEGER,
-	.arg1_type	= ARG_PTR_TO_CTX,
-	.arg2_type	= ARG_ANYTHING,
-	.arg3_type	= ARG_ANYTHING,
-	.arg4_type	= ARG_PTR_TO_MEM,
-	.arg5_type	= ARG_CONST_SIZE,
-};
-
-BPF_CALL_5(bpf_getsockopt, struct bpf_sock_ops_kern *, bpf_sock,
-	   int, level, int, optname, char *, optval, int, optlen)
 {
-	struct sock *sk = bpf_sock->sk;
-
 	if (!sk_fullsock(sk))
 		goto err_clear;
+
+	sock_owned_by_me(sk);
+
 #ifdef CONFIG_INET
 	if (level == SOL_TCP && sk->sk_prot->getsockopt == tcp_getsockopt) {
 		struct inet_connection_sock *icsk;
@@ -4459,8 +4421,71 @@ err_clear:
 	return -EINVAL;
 }
 
-static const struct bpf_func_proto bpf_getsockopt_proto = {
+BPF_CALL_5(bpf_sock_addr_setsockopt, struct bpf_sock_addr_kern *, ctx,
-	.func		= bpf_getsockopt,
+	   int, level, int, optname, char *, optval, int, optlen)
+{
+	u32 flags = 0;
+	return _bpf_setsockopt(ctx->sk, level, optname, optval, optlen,
+			       flags);
+}
+
+static const struct bpf_func_proto bpf_sock_addr_setsockopt_proto = {
+	.func		= bpf_sock_addr_setsockopt,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_ANYTHING,
+	.arg3_type	= ARG_ANYTHING,
+	.arg4_type	= ARG_PTR_TO_MEM,
+	.arg5_type	= ARG_CONST_SIZE,
+};
+
+BPF_CALL_5(bpf_sock_addr_getsockopt, struct bpf_sock_addr_kern *, ctx,
+	   int, level, int, optname, char *, optval, int, optlen)
+{
+	return _bpf_getsockopt(ctx->sk, level, optname, optval, optlen);
+}
+
+static const struct bpf_func_proto bpf_sock_addr_getsockopt_proto = {
+	.func		= bpf_sock_addr_getsockopt,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_ANYTHING,
+	.arg3_type	= ARG_ANYTHING,
+	.arg4_type	= ARG_PTR_TO_UNINIT_MEM,
+	.arg5_type	= ARG_CONST_SIZE,
+};
+
+BPF_CALL_5(bpf_sock_ops_setsockopt, struct bpf_sock_ops_kern *, bpf_sock,
+	   int, level, int, optname, char *, optval, int, optlen)
+{
+	u32 flags = 0;
+	if (bpf_sock->op > BPF_SOCK_OPS_NEEDS_ECN)
+		flags |= SOCKOPT_CC_REINIT;
+	return _bpf_setsockopt(bpf_sock->sk, level, optname, optval, optlen,
+			       flags);
+}
+
+static const struct bpf_func_proto bpf_sock_ops_setsockopt_proto = {
+	.func		= bpf_sock_ops_setsockopt,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_PTR_TO_CTX,
+	.arg2_type	= ARG_ANYTHING,
+	.arg3_type	= ARG_ANYTHING,
+	.arg4_type	= ARG_PTR_TO_MEM,
+	.arg5_type	= ARG_CONST_SIZE,
+};
+
+BPF_CALL_5(bpf_sock_ops_getsockopt, struct bpf_sock_ops_kern *, bpf_sock,
+	   int, level, int, optname, char *, optval, int, optlen)
+{
+	return _bpf_getsockopt(bpf_sock->sk, level, optname, optval, optlen);
+}
+
+static const struct bpf_func_proto bpf_sock_ops_getsockopt_proto = {
+	.func		= bpf_sock_ops_getsockopt,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
 	.arg1_type	= ARG_PTR_TO_CTX,
@@ -5983,52 +6008,7 @@ bool bpf_helper_changes_pkt_data(void *func)
 	return false;
 }
 
-const struct bpf_func_proto *
+const struct bpf_func_proto bpf_event_output_data_proto __weak;
-bpf_base_func_proto(enum bpf_func_id func_id)
-{
-	switch (func_id) {
-	case BPF_FUNC_map_lookup_elem:
-		return &bpf_map_lookup_elem_proto;
-	case BPF_FUNC_map_update_elem:
-		return &bpf_map_update_elem_proto;
-	case BPF_FUNC_map_delete_elem:
-		return &bpf_map_delete_elem_proto;
-	case BPF_FUNC_map_push_elem:
-		return &bpf_map_push_elem_proto;
-	case BPF_FUNC_map_pop_elem:
-		return &bpf_map_pop_elem_proto;
-	case BPF_FUNC_map_peek_elem:
-		return &bpf_map_peek_elem_proto;
-	case BPF_FUNC_get_prandom_u32:
-		return &bpf_get_prandom_u32_proto;
-	case BPF_FUNC_get_smp_processor_id:
-		return &bpf_get_raw_smp_processor_id_proto;
-	case BPF_FUNC_get_numa_node_id:
-		return &bpf_get_numa_node_id_proto;
-	case BPF_FUNC_tail_call:
-		return &bpf_tail_call_proto;
-	case BPF_FUNC_ktime_get_ns:
-		return &bpf_ktime_get_ns_proto;
-	default:
-		break;
-	}
-
-	if (!capable(CAP_SYS_ADMIN))
-		return NULL;
-
-	switch (func_id) {
-	case BPF_FUNC_spin_lock:
-		return &bpf_spin_lock_proto;
-	case BPF_FUNC_spin_unlock:
-		return &bpf_spin_unlock_proto;
-	case BPF_FUNC_trace_printk:
-		return bpf_get_trace_printk_proto();
-	case BPF_FUNC_jiffies64:
-		return &bpf_jiffies64_proto;
-	default:
-		return NULL;
-	}
-}
 
 static const struct bpf_func_proto *
 sock_filter_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
@@ -6119,6 +6099,22 @@ sock_addr_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_sk_storage_get_proto;
 	case BPF_FUNC_sk_storage_delete:
 		return &bpf_sk_storage_delete_proto;
+	case BPF_FUNC_setsockopt:
+		switch (prog->expected_attach_type) {
+		case BPF_CGROUP_INET4_CONNECT:
+		case BPF_CGROUP_INET6_CONNECT:
+			return &bpf_sock_addr_setsockopt_proto;
+		default:
+			return NULL;
+		}
+	case BPF_FUNC_getsockopt:
+		switch (prog->expected_attach_type) {
+		case BPF_CGROUP_INET4_CONNECT:
+		case BPF_CGROUP_INET6_CONNECT:
+			return &bpf_sock_addr_getsockopt_proto;
+		default:
+			return NULL;
+		}
 	default:
 		return bpf_base_func_proto(func_id);
 	}
@@ -6213,6 +6209,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_skb_adjust_room_proto;
 	case BPF_FUNC_skb_change_tail:
 		return &bpf_skb_change_tail_proto;
+	case BPF_FUNC_skb_change_head:
+		return &bpf_skb_change_head_proto;
 	case BPF_FUNC_skb_get_tunnel_key:
 		return &bpf_skb_get_tunnel_key_proto;
 	case BPF_FUNC_skb_set_tunnel_key:
@@ -6335,9 +6333,9 @@ sock_ops_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 {
 	switch (func_id) {
 	case BPF_FUNC_setsockopt:
-		return &bpf_setsockopt_proto;
+		return &bpf_sock_ops_setsockopt_proto;
 	case BPF_FUNC_getsockopt:
-		return &bpf_getsockopt_proto;
+		return &bpf_sock_ops_getsockopt_proto;
 	case BPF_FUNC_sock_ops_cb_flags_set:
 		return &bpf_sock_ops_cb_flags_set_proto;
 	case BPF_FUNC_sock_map_update:
@@ -8786,6 +8784,10 @@ BPF_CALL_4(sk_select_reuseport, struct sk_reuseport_kern *, reuse_kern,
 
 	reuse = rcu_dereference(selected_sk->sk_reuseport_cb);
 	if (!reuse) {
+		/* Lookup in sock_map can return TCP ESTABLISHED sockets. */
+		if (sk_is_refcounted(selected_sk))
+			sock_put(selected_sk);
+
 		/* reuseport_array has only sk with non NULL sk_reuseport_cb.
 		 * The only (!reuse) case here is - the sk has already been
 		 * unhashed (e.g. by close()), so treat it as -ENOENT.

net/core/neighbour.c

@@ -3379,7 +3379,7 @@ EXPORT_SYMBOL(neigh_app_ns);
 static int unres_qlen_max = INT_MAX / SKB_TRUESIZE(ETH_FRAME_LEN);
 
 static int proc_unres_qlen(struct ctl_table *ctl, int write,
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
+			   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int size, ret;
 	struct ctl_table tmp = *ctl;
@@ -3443,8 +3443,8 @@ static void neigh_proc_update(struct ctl_table *ctl, int write)
 }
 
 static int neigh_proc_dointvec_zero_intmax(struct ctl_table *ctl, int write,
-					   void __user *buffer,
+					   void *buffer, size_t *lenp,
-					   size_t *lenp, loff_t *ppos)
+					   loff_t *ppos)
 {
 	struct ctl_table tmp = *ctl;
 	int ret;
@@ -3457,8 +3457,8 @@ static int neigh_proc_dointvec_zero_intmax(struct ctl_table *ctl, int write,
 	return ret;
 }
 
-int neigh_proc_dointvec(struct ctl_table *ctl, int write,
+int neigh_proc_dointvec(struct ctl_table *ctl, int write, void *buffer,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
+			size_t *lenp, loff_t *ppos)
 {
 	int ret = proc_dointvec(ctl, write, buffer, lenp, ppos);
 
@@ -3467,8 +3467,7 @@ int neigh_proc_dointvec(struct ctl_table *ctl, int write,
 }
 EXPORT_SYMBOL(neigh_proc_dointvec);
 
-int neigh_proc_dointvec_jiffies(struct ctl_table *ctl, int write,
+int neigh_proc_dointvec_jiffies(struct ctl_table *ctl, int write, void *buffer,
-				void __user *buffer,
 				size_t *lenp, loff_t *ppos)
 {
 	int ret = proc_dointvec_jiffies(ctl, write, buffer, lenp, ppos);
@@ -3479,8 +3478,8 @@ int neigh_proc_dointvec_jiffies(struct ctl_table *ctl, int write,
 EXPORT_SYMBOL(neigh_proc_dointvec_jiffies);
 
 static int neigh_proc_dointvec_userhz_jiffies(struct ctl_table *ctl, int write,
-					      void __user *buffer,
+					      void *buffer, size_t *lenp,
-					      size_t *lenp, loff_t *ppos)
+					      loff_t *ppos)
 {
 	int ret = proc_dointvec_userhz_jiffies(ctl, write, buffer, lenp, ppos);
 
@@ -3489,8 +3488,7 @@ static int neigh_proc_dointvec_userhz_jiffies(struct ctl_table *ctl, int write,
 }
 
 int neigh_proc_dointvec_ms_jiffies(struct ctl_table *ctl, int write,
-				   void __user *buffer,
+				   void *buffer, size_t *lenp, loff_t *ppos)
-				   size_t *lenp, loff_t *ppos)
 {
 	int ret = proc_dointvec_ms_jiffies(ctl, write, buffer, lenp, ppos);
 
@@ -3500,8 +3498,8 @@ int neigh_proc_dointvec_ms_jiffies(struct ctl_table *ctl, int write,
 EXPORT_SYMBOL(neigh_proc_dointvec_ms_jiffies);
 
 static int neigh_proc_dointvec_unres_qlen(struct ctl_table *ctl, int write,
-					  void __user *buffer,
+					  void *buffer, size_t *lenp,
-					  size_t *lenp, loff_t *ppos)
+					  loff_t *ppos)
 {
 	int ret = proc_unres_qlen(ctl, write, buffer, lenp, ppos);
 
@@ -3510,8 +3508,8 @@ static int neigh_proc_dointvec_unres_qlen(struct ctl_table *ctl, int write,
 }
 
 static int neigh_proc_base_reachable_time(struct ctl_table *ctl, int write,
-					  void __user *buffer,
+					  void *buffer, size_t *lenp,
-					  size_t *lenp, loff_t *ppos)
+					  loff_t *ppos)
 {
 	struct neigh_parms *p = ctl->extra2;
 	int ret;

net/core/sock_map.c

@@ -343,7 +343,14 @@ static struct sock *__sock_map_lookup_elem(struct bpf_map *map, u32 key)
 
 static void *sock_map_lookup(struct bpf_map *map, void *key)
 {
-	return __sock_map_lookup_elem(map, *(u32 *)key);
+	struct sock *sk;
+
+	sk = __sock_map_lookup_elem(map, *(u32 *)key);
+	if (!sk || !sk_fullsock(sk))
+		return NULL;
+	if (sk_is_refcounted(sk) && !refcount_inc_not_zero(&sk->sk_refcnt))
+		return NULL;
+	return sk;
 }
 
 static void *sock_map_lookup_sys(struct bpf_map *map, void *key)
@@ -1051,7 +1058,14 @@ static void *sock_hash_lookup_sys(struct bpf_map *map, void *key)
 
 static void *sock_hash_lookup(struct bpf_map *map, void *key)
 {
-	return __sock_hash_lookup_elem(map, key);
+	struct sock *sk;
+
+	sk = __sock_hash_lookup_elem(map, key);
+	if (!sk || !sk_fullsock(sk))
+		return NULL;
+	if (sk_is_refcounted(sk) && !refcount_inc_not_zero(&sk->sk_refcnt))
+		return NULL;
+	return sk;
 }
 
 static void sock_hash_release_progs(struct bpf_map *map)

net/core/sysctl_net_core.c

@@ -45,7 +45,7 @@ EXPORT_SYMBOL(sysctl_devconf_inherit_init_net);
 
 #ifdef CONFIG_RPS
 static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	unsigned int orig_size, size;
 	int ret, i;
@@ -115,8 +115,7 @@ static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
 static DEFINE_MUTEX(flow_limit_update_mutex);
 
 static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp,
+				 void *buffer, size_t *lenp, loff_t *ppos)
-				 loff_t *ppos)
 {
 	struct sd_flow_limit *cur;
 	struct softnet_data *sd;
@@ -180,10 +179,7 @@ write_unlock:
 		}
 		if (len < *lenp)
 			kbuf[len++] = '\n';
-		if (copy_to_user(buffer, kbuf, len)) {
+		memcpy(buffer, kbuf, len);
-			ret = -EFAULT;
-			goto done;
-		}
 		*lenp = len;
 		*ppos += len;
 	}
@@ -194,8 +190,7 @@ done:
 }
 
 static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
-				       void __user *buffer, size_t *lenp,
+				       void *buffer, size_t *lenp, loff_t *ppos)
-				       loff_t *ppos)
 {
 	unsigned int old, *ptr;
 	int ret;
@@ -217,7 +212,7 @@ static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
 
 #ifdef CONFIG_NET_SCHED
 static int set_default_qdisc(struct ctl_table *table, int write,
-			     void __user *buffer, size_t *lenp, loff_t *ppos)
+			     void *buffer, size_t *lenp, loff_t *ppos)
 {
 	char id[IFNAMSIZ];
 	struct ctl_table tbl = {
@@ -236,7 +231,7 @@ static int set_default_qdisc(struct ctl_table *table, int write,
 #endif
 
 static int proc_do_dev_weight(struct ctl_table *table, int write,
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
+			   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret;
 
@@ -251,7 +246,7 @@ static int proc_do_dev_weight(struct ctl_table *table, int write,
 }
 
 static int proc_do_rss_key(struct ctl_table *table, int write,
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
+			   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct ctl_table fake_table;
 	char buf[NETDEV_RSS_KEY_LEN * 3];
@@ -264,7 +259,7 @@ static int proc_do_rss_key(struct ctl_table *table, int write,
 
 #ifdef CONFIG_BPF_JIT
 static int proc_dointvec_minmax_bpf_enable(struct ctl_table *table, int write,
-					   void __user *buffer, size_t *lenp,
+					   void *buffer, size_t *lenp,
 					   loff_t *ppos)
 {
 	int ret, jit_enable = *(int *)table->data;
@@ -291,8 +286,7 @@ static int proc_dointvec_minmax_bpf_enable(struct ctl_table *table, int write,
 # ifdef CONFIG_HAVE_EBPF_JIT
 static int
 proc_dointvec_minmax_bpf_restricted(struct ctl_table *table, int write,
-				    void __user *buffer, size_t *lenp,
+				    void *buffer, size_t *lenp, loff_t *ppos)
-				    loff_t *ppos)
 {
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -303,8 +297,7 @@ proc_dointvec_minmax_bpf_restricted(struct ctl_table *table, int write,
 
 static int
 proc_dolongvec_minmax_bpf_restricted(struct ctl_table *table, int write,
-				     void __user *buffer, size_t *lenp,
+				     void *buffer, size_t *lenp, loff_t *ppos)
-				     loff_t *ppos)
 {
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;

net/decnet/dn_dev.c

@@ -160,8 +160,8 @@ static int max_t3[] = { 8191 }; /* Must fit in 16 bits when multiplied by BCT3MU
 static int min_priority[1];
 static int max_priority[] = { 127 }; /* From DECnet spec */
 
-static int dn_forwarding_proc(struct ctl_table *, int,
+static int dn_forwarding_proc(struct ctl_table *, int, void *, size_t *,
-			void __user *, size_t *, loff_t *);
+		loff_t *);
 static struct dn_dev_sysctl_table {
 	struct ctl_table_header *sysctl_header;
 	struct ctl_table dn_dev_vars[5];
@@ -245,8 +245,7 @@ static void dn_dev_sysctl_unregister(struct dn_dev_parms *parms)
 }
 
 static int dn_forwarding_proc(struct ctl_table *table, int write,
-				void __user *buffer,
+		void *buffer, size_t *lenp, loff_t *ppos)
-				size_t *lenp, loff_t *ppos)
 {
 #ifdef CONFIG_DECNET_ROUTER
 	struct net_device *dev = table->extra1;

net/decnet/sysctl_net_decnet.c

@@ -134,8 +134,7 @@ static int parse_addr(__le16 *addr, char *str)
 }
 
 static int dn_node_address_handler(struct ctl_table *table, int write,
-				void __user *buffer,
+		void *buffer, size_t *lenp, loff_t *ppos)
-				size_t *lenp, loff_t *ppos)
 {
 	char addr[DN_ASCBUF_LEN];
 	size_t len;
@@ -148,10 +147,7 @@ static int dn_node_address_handler(struct ctl_table *table, int write,
 
 	if (write) {
 		len = (*lenp < DN_ASCBUF_LEN) ? *lenp : (DN_ASCBUF_LEN-1);
-
+		memcpy(addr, buffer, len);
-		if (copy_from_user(addr, buffer, len))
-			return -EFAULT;
-
 		addr[len] = 0;
 		strip_it(addr);
 
@@ -173,11 +169,9 @@ static int dn_node_address_handler(struct ctl_table *table, int write,
 	len = strlen(addr);
 	addr[len++] = '\n';
 
-	if (len > *lenp) len = *lenp;
+	if (len > *lenp)
-
+		len = *lenp;
-	if (copy_to_user(buffer, addr, len))
+	memcpy(buffer, addr, len);
-		return -EFAULT;
-
 	*lenp = len;
 	*ppos += len;
 
@@ -185,8 +179,7 @@ static int dn_node_address_handler(struct ctl_table *table, int write,
 }
 
 static int dn_def_dev_handler(struct ctl_table *table, int write,
-				void __user *buffer,
+		void *buffer, size_t *lenp, loff_t *ppos)
-				size_t *lenp, loff_t *ppos)
 {
 	size_t len;
 	struct net_device *dev;
@@ -201,9 +194,7 @@ static int dn_def_dev_handler(struct ctl_table *table, int write,
 		if (*lenp > 16)
 			return -E2BIG;
 
-		if (copy_from_user(devname, buffer, *lenp))
+		memcpy(devname, buffer, *lenp);
-			return -EFAULT;
-
 		devname[*lenp] = 0;
 		strip_it(devname);
 
@@ -238,9 +229,7 @@ static int dn_def_dev_handler(struct ctl_table *table, int write,
 
 	if (len > *lenp) len = *lenp;
 
-	if (copy_to_user(buffer, devname, len))
+	memcpy(buffer, devname, len);
-		return -EFAULT;
-
 	*lenp = len;
 	*ppos += len;
 

net/ipv4/devinet.c

@@ -2366,8 +2366,7 @@ static int devinet_conf_ifindex(struct net *net, struct ipv4_devconf *cnf)
 }
 
 static int devinet_conf_proc(struct ctl_table *ctl, int write,
-			     void __user *buffer,
+			     void *buffer, size_t *lenp, loff_t *ppos)
-			     size_t *lenp, loff_t *ppos)
 {
 	int old_value = *(int *)ctl->data;
 	int ret = proc_dointvec(ctl, write, buffer, lenp, ppos);
@@ -2419,8 +2418,7 @@ static int devinet_conf_proc(struct ctl_table *ctl, int write,
 }
 
 static int devinet_sysctl_forward(struct ctl_table *ctl, int write,
-				  void __user *buffer,
+				  void *buffer, size_t *lenp, loff_t *ppos)
-				  size_t *lenp, loff_t *ppos)
 {
 	int *valp = ctl->data;
 	int val = *valp;
@@ -2463,8 +2461,7 @@ static int devinet_sysctl_forward(struct ctl_table *ctl, int write,
 }
 
 static int ipv4_doint_and_flush(struct ctl_table *ctl, int write,
-				void __user *buffer,
+				void *buffer, size_t *lenp, loff_t *ppos)
-				size_t *lenp, loff_t *ppos)
 {
 	int *valp = ctl->data;
 	int val = *valp;

net/ipv4/route.c

@@ -3336,8 +3336,7 @@ static int ip_rt_gc_elasticity __read_mostly	= 8;
 static int ip_min_valid_pmtu __read_mostly	= IPV4_MIN_MTU;
 
 static int ipv4_sysctl_rtcache_flush(struct ctl_table *__ctl, int write,
-					void __user *buffer,
+		void *buffer, size_t *lenp, loff_t *ppos)
-					size_t *lenp, loff_t *ppos)
 {
 	struct net *net = (struct net *)__ctl->extra1;
 

net/ipv4/sysctl_net_ipv4.c

@@ -71,8 +71,7 @@ static void set_local_port_range(struct net *net, int range[2])
 
 /* Validate changes from /proc interface. */
 static int ipv4_local_port_range(struct ctl_table *table, int write,
-				 void __user *buffer,
+				 void *buffer, size_t *lenp, loff_t *ppos)
-				 size_t *lenp, loff_t *ppos)
 {
 	struct net *net =
 		container_of(table->data, struct net, ipv4.ip_local_ports.range);
@@ -107,7 +106,7 @@ static int ipv4_local_port_range(struct ctl_table *table, int write,
 
 /* Validate changes from /proc interface. */
 static int ipv4_privileged_ports(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct net *net = container_of(table->data, struct net,
 	    ipv4.sysctl_ip_prot_sock);
@@ -168,8 +167,7 @@ static void set_ping_group_range(struct ctl_table *table, kgid_t low, kgid_t hig
 
 /* Validate changes from /proc interface. */
 static int ipv4_ping_group_range(struct ctl_table *table, int write,
-				 void __user *buffer,
+				 void *buffer, size_t *lenp, loff_t *ppos)
-				 size_t *lenp, loff_t *ppos)
 {
 	struct user_namespace *user_ns = current_user_ns();
 	int ret;
@@ -204,8 +202,7 @@ static int ipv4_ping_group_range(struct ctl_table *table, int write,
 }
 
 static int ipv4_fwd_update_priority(struct ctl_table *table, int write,
-				    void __user *buffer,
+				    void *buffer, size_t *lenp, loff_t *ppos)
-				    size_t *lenp, loff_t *ppos)
 {
 	struct net *net;
 	int ret;
@@ -221,7 +218,7 @@ static int ipv4_fwd_update_priority(struct ctl_table *table, int write,
 }
 
 static int proc_tcp_congestion_control(struct ctl_table *ctl, int write,
-				       void __user *buffer, size_t *lenp, loff_t *ppos)
+				       void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct net *net = container_of(ctl->data, struct net,
 				       ipv4.tcp_congestion_control);
@@ -241,9 +238,8 @@ static int proc_tcp_congestion_control(struct ctl_table *ctl, int write,
 }
 
 static int proc_tcp_available_congestion_control(struct ctl_table *ctl,
-						 int write,
+						 int write, void *buffer,
-						 void __user *buffer, size_t *lenp,
+						 size_t *lenp, loff_t *ppos)
-						 loff_t *ppos)
 {
 	struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX, };
 	int ret;
@@ -258,9 +254,8 @@ static int proc_tcp_available_congestion_control(struct ctl_table *ctl,
 }
 
 static int proc_allowed_congestion_control(struct ctl_table *ctl,
-					   int write,
+					   int write, void *buffer,
-					   void __user *buffer, size_t *lenp,
+					   size_t *lenp, loff_t *ppos)
-					   loff_t *ppos)
 {
 	struct ctl_table tbl = { .maxlen = TCP_CA_BUF_MAX };
 	int ret;
@@ -296,8 +291,7 @@ static int sscanf_key(char *buf, __le32 *key)
 }
 
 static int proc_tcp_fastopen_key(struct ctl_table *table, int write,
-				 void __user *buffer, size_t *lenp,
+				 void *buffer, size_t *lenp, loff_t *ppos)
-				 loff_t *ppos)
 {
 	struct net *net = container_of(table->data, struct net,
 	    ipv4.sysctl_tcp_fastopen);
@@ -399,7 +393,7 @@ static void proc_configure_early_demux(int enabled, int protocol)
 }
 
 static int proc_tcp_early_demux(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret = 0;
 
@@ -415,7 +409,7 @@ static int proc_tcp_early_demux(struct ctl_table *table, int write,
 }
 
 static int proc_udp_early_demux(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret = 0;
 
@@ -431,8 +425,7 @@ static int proc_udp_early_demux(struct ctl_table *table, int write,
 }
 
 static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table,
-					     int write,
+					     int write, void *buffer,
-					     void __user *buffer,
 					     size_t *lenp, loff_t *ppos)
 {
 	struct net *net = container_of(table->data, struct net,
@@ -447,8 +440,7 @@ static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table,
 }
 
 static int proc_tcp_available_ulp(struct ctl_table *ctl,
-				  int write,
+				  int write, void *buffer, size_t *lenp,
-				  void __user *buffer, size_t *lenp,
 				  loff_t *ppos)
 {
 	struct ctl_table tbl = { .maxlen = TCP_ULP_BUF_MAX, };
@@ -466,7 +458,7 @@ static int proc_tcp_available_ulp(struct ctl_table *ctl,
 
 #ifdef CONFIG_IP_ROUTE_MULTIPATH
 static int proc_fib_multipath_hash_policy(struct ctl_table *table, int write,
-					  void __user *buffer, size_t *lenp,
+					  void *buffer, size_t *lenp,
 					  loff_t *ppos)
 {
 	struct net *net = container_of(table->data, struct net,

net/ipv6/addrconf.c

@@ -6095,9 +6095,8 @@ static void ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 
 #ifdef CONFIG_SYSCTL
 
-static
+static int addrconf_sysctl_forward(struct ctl_table *ctl, int write,
-int addrconf_sysctl_forward(struct ctl_table *ctl, int write,
+		void *buffer, size_t *lenp, loff_t *ppos)
-			   void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	int *valp = ctl->data;
 	int val = *valp;
@@ -6121,9 +6120,8 @@ int addrconf_sysctl_forward(struct ctl_table *ctl, int write,
 	return ret;
 }
 
-static
+static int addrconf_sysctl_mtu(struct ctl_table *ctl, int write,
-int addrconf_sysctl_mtu(struct ctl_table *ctl, int write,
+		void *buffer, size_t *lenp, loff_t *ppos)
-			void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct inet6_dev *idev = ctl->extra1;
 	int min_mtu = IPV6_MIN_MTU;
@@ -6193,9 +6191,8 @@ static int addrconf_disable_ipv6(struct ctl_table *table, int *p, int newf)
 	return 0;
 }
 
-static
+static int addrconf_sysctl_disable(struct ctl_table *ctl, int write,
-int addrconf_sysctl_disable(struct ctl_table *ctl, int write,
+		void *buffer, size_t *lenp, loff_t *ppos)
-			    void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	int *valp = ctl->data;
 	int val = *valp;
@@ -6219,9 +6216,8 @@ int addrconf_sysctl_disable(struct ctl_table *ctl, int write,
 	return ret;
 }
 
-static
+static int addrconf_sysctl_proxy_ndp(struct ctl_table *ctl, int write,
-int addrconf_sysctl_proxy_ndp(struct ctl_table *ctl, int write,
+		void *buffer, size_t *lenp, loff_t *ppos)
-			      void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	int *valp = ctl->data;
 	int ret;
@@ -6262,7 +6258,7 @@ int addrconf_sysctl_proxy_ndp(struct ctl_table *ctl, int write,
 }
 
 static int addrconf_sysctl_addr_gen_mode(struct ctl_table *ctl, int write,
-					 void __user *buffer, size_t *lenp,
+					 void *buffer, size_t *lenp,
 					 loff_t *ppos)
 {
 	int ret = 0;
@@ -6324,7 +6320,7 @@ out:
 }
 
 static int addrconf_sysctl_stable_secret(struct ctl_table *ctl, int write,
-					 void __user *buffer, size_t *lenp,
+					 void *buffer, size_t *lenp,
 					 loff_t *ppos)
 {
 	int err;
@@ -6391,8 +6387,7 @@ out:
 
 static
 int addrconf_sysctl_ignore_routes_with_linkdown(struct ctl_table *ctl,
-						int write,
+						int write, void *buffer,
-						void __user *buffer,
 						size_t *lenp,
 						loff_t *ppos)
 {
@@ -6492,10 +6487,8 @@ int addrconf_disable_policy(struct ctl_table *ctl, int *valp, int val)
 	return 0;
 }
 
-static
+static int addrconf_sysctl_disable_policy(struct ctl_table *ctl, int write,
-int addrconf_sysctl_disable_policy(struct ctl_table *ctl, int write,
+				   void *buffer, size_t *lenp, loff_t *ppos)
-				   void __user *buffer, size_t *lenp,
-				   loff_t *ppos)
 {
 	int *valp = ctl->data;
 	int val = *valp;

net/ipv6/ndisc.c

@@ -1835,7 +1835,8 @@ static void ndisc_warn_deprecated_sysctl(struct ctl_table *ctl,
 	}
 }
 
-int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *ppos)
+int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, int write, void *buffer,
+		size_t *lenp, loff_t *ppos)
 {
 	struct net_device *dev = ctl->extra1;
 	struct inet6_dev *idev;

net/ipv6/route.c

@@ -6092,9 +6092,8 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v)
 
 #ifdef CONFIG_SYSCTL
 
-static
+static int ipv6_sysctl_rtcache_flush(struct ctl_table *ctl, int write,
-int ipv6_sysctl_rtcache_flush(struct ctl_table *ctl, int write,
+			      void *buffer, size_t *lenp, loff_t *ppos)
-			      void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct net *net;
 	int delay;

net/ipv6/sysctl_net_ipv6.c

@@ -26,8 +26,7 @@ static int auto_flowlabels_min;
 static int auto_flowlabels_max = IP6_AUTO_FLOW_LABEL_MAX;
 
 static int proc_rt6_multipath_hash_policy(struct ctl_table *table, int write,
-					  void __user *buffer, size_t *lenp,
+					  void *buffer, size_t *lenp, loff_t *ppos)
-					  loff_t *ppos)
 {
 	struct net *net;
 	int ret;

net/mpls/af_mpls.c

@@ -1362,8 +1362,7 @@ done:
 	(&((struct mpls_dev *)0)->field)
 
 static int mpls_conf_proc(struct ctl_table *ctl, int write,
-			  void __user *buffer,
+			  void *buffer, size_t *lenp, loff_t *ppos)
-			  size_t *lenp, loff_t *ppos)
 {
 	int oval = *(int *)ctl->data;
 	int ret = proc_dointvec(ctl, write, buffer, lenp, ppos);
@@ -2594,7 +2593,7 @@ nolabels:
 }
 
 static int mpls_platform_labels(struct ctl_table *table, int write,
-				void __user *buffer, size_t *lenp, loff_t *ppos)
+				void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct net *net = table->data;
 	int platform_labels = net->mpls.platform_labels;

net/netfilter/ipvs/ip_vs_ctl.c

@@ -1736,7 +1736,7 @@ static int three = 3;
 
 static int
 proc_do_defense_mode(struct ctl_table *table, int write,
-		     void __user *buffer, size_t *lenp, loff_t *ppos)
+		     void *buffer, size_t *lenp, loff_t *ppos)
 {
 	struct netns_ipvs *ipvs = table->extra2;
 	int *valp = table->data;
@@ -1763,7 +1763,7 @@ proc_do_defense_mode(struct ctl_table *table, int write,
 
 static int
 proc_do_sync_threshold(struct ctl_table *table, int write,
-		       void __user *buffer, size_t *lenp, loff_t *ppos)
+		       void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int *valp = table->data;
 	int val[2];
@@ -1788,7 +1788,7 @@ proc_do_sync_threshold(struct ctl_table *table, int write,
 
 static int
 proc_do_sync_ports(struct ctl_table *table, int write,
-		   void __user *buffer, size_t *lenp, loff_t *ppos)
+		   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int *valp = table->data;
 	int val = *valp;

net/netfilter/nf_conntrack_standalone.c

@@ -519,7 +519,7 @@ static unsigned int nf_conntrack_htable_size_user __read_mostly;
 
 static int
 nf_conntrack_hash_sysctl(struct ctl_table *table, int write,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+			 void *buffer, size_t *lenp, loff_t *ppos)
 {
 	int ret;
 

net/netfilter/nf_log.c

@@ -414,7 +414,7 @@ static struct ctl_table nf_log_sysctl_ftable[] = {
 };
 
 static int nf_log_proc_dostring(struct ctl_table *table, int write,
-			 void __user *buffer, size_t *lenp, loff_t *ppos)
+			 void *buffer, size_t *lenp, loff_t *ppos)
 {
 	const struct nf_logger *logger;
 	char buf[NFLOGGER_NAME_LEN];

net/phonet/sysctl.c

@@ -49,8 +49,7 @@ void phonet_get_local_port_range(int *min, int *max)
 }
 
 static int proc_local_port_range(struct ctl_table *table, int write,
-				void __user *buffer,
+				 void *buffer, size_t *lenp, loff_t *ppos)
-				size_t *lenp, loff_t *ppos)
 {
 	int ret;
 	int range[2] = {local_port_range[0], local_port_range[1]};

net/rds/tcp.c

@@ -62,8 +62,7 @@ static atomic_t rds_tcp_unloading = ATOMIC_INIT(0);
 static struct kmem_cache *rds_tcp_conn_slab;
 
 static int rds_tcp_skbuf_handler(struct ctl_table *ctl, int write,
-				 void __user *buffer, size_t *lenp,
+				 void *buffer, size_t *lenp, loff_t *fpos);
-				 loff_t *fpos);
 
 static int rds_tcp_min_sndbuf = SOCK_MIN_SNDBUF;
 static int rds_tcp_min_rcvbuf = SOCK_MIN_RCVBUF;
@@ -676,8 +675,7 @@ static void rds_tcp_sysctl_reset(struct net *net)
 }
 
 static int rds_tcp_skbuf_handler(struct ctl_table *ctl, int write,
-				 void __user *buffer, size_t *lenp,
+				 void *buffer, size_t *lenp, loff_t *fpos)
-				 loff_t *fpos)
 {
 	struct net *net = current->nsproxy->net_ns;
 	int err;