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

Pull networking updates from David Miller:
 "Highlights:

   1) Fix the iwlwifi regression, from Johannes Berg.

   2) Support BSS coloring and 802.11 encapsulation offloading in
      hardware, from John Crispin.

   3) Fix some potential Spectre issues in qtnfmac, from Sergey
      Matyukevich.

   4) Add TTL decrement action to openvswitch, from Matteo Croce.

   5) Allow paralleization through flow_action setup by not taking the
      RTNL mutex, from Vlad Buslov.

   6) A lot of zero-length array to flexible-array conversions, from
      Gustavo A. R. Silva.

   7) Align XDP statistics names across several drivers for consistency,
      from Lorenzo Bianconi.

   8) Add various pieces of infrastructure for offloading conntrack, and
      make use of it in mlx5 driver, from Paul Blakey.

   9) Allow using listening sockets in BPF sockmap, from Jakub Sitnicki.

  10) Lots of parallelization improvements during configuration changes
      in mlxsw driver, from Ido Schimmel.

  11) Add support to devlink for generic packet traps, which report
      packets dropped during ACL processing. And use them in mlxsw
      driver. From Jiri Pirko.

  12) Support bcmgenet on ACPI, from Jeremy Linton.

  13) Make BPF compatible with RT, from Thomas Gleixnet, Alexei
      Starovoitov, and your's truly.

  14) Support XDP meta-data in virtio_net, from Yuya Kusakabe.

  15) Fix sysfs permissions when network devices change namespaces, from
      Christian Brauner.

  16) Add a flags element to ethtool_ops so that drivers can more simply
      indicate which coalescing parameters they actually support, and
      therefore the generic layer can validate the user's ethtool
      request. Use this in all drivers, from Jakub Kicinski.

  17) Offload FIFO qdisc in mlxsw, from Petr Machata.

  18) Support UDP sockets in sockmap, from Lorenz Bauer.

  19) Fix stretch ACK bugs in several TCP congestion control modules,
      from Pengcheng Yang.

  20) Support virtual functiosn in octeontx2 driver, from Tomasz
      Duszynski.

  21) Add region operations for devlink and use it in ice driver to dump
      NVM contents, from Jacob Keller.

  22) Add support for hw offload of MACSEC, from Antoine Tenart.

  23) Add support for BPF programs that can be attached to LSM hooks,
      from KP Singh.

  24) Support for multiple paths, path managers, and counters in MPTCP.
      From Peter Krystad, Paolo Abeni, Florian Westphal, Davide Caratti,
      and others.

  25) More progress on adding the netlink interface to ethtool, from
      Michal Kubecek"

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2121 commits)
  net: ipv6: rpl_iptunnel: Fix potential memory leak in rpl_do_srh_inline
  cxgb4/chcr: nic-tls stats in ethtool
  net: dsa: fix oops while probing Marvell DSA switches
  net/bpfilter: remove superfluous testing message
  net: macb: Fix handling of fixed-link node
  net: dsa: ksz: Select KSZ protocol tag
  netdevsim: dev: Fix memory leak in nsim_dev_take_snapshot_write
  net: stmmac: add EHL 2.5Gbps PCI info and PCI ID
  net: stmmac: add EHL PSE0 & PSE1 1Gbps PCI info and PCI ID
  net: stmmac: create dwmac-intel.c to contain all Intel platform
  net: dsa: bcm_sf2: Support specifying VLAN tag egress rule
  net: dsa: bcm_sf2: Add support for matching VLAN TCI
  net: dsa: bcm_sf2: Move writing of CFP_DATA(5) into slicing functions
  net: dsa: bcm_sf2: Check earlier for FLOW_EXT and FLOW_MAC_EXT
  net: dsa: bcm_sf2: Disable learning for ASP port
  net: dsa: b53: Deny enslaving port 7 for 7278 into a bridge
  net: dsa: b53: Prevent tagged VLAN on port 7 for 7278
  net: dsa: b53: Restore VLAN entries upon (re)configuration
  net: dsa: bcm_sf2: Fix overflow checks
  hv_netvsc: Remove unnecessary round_up for recv_completion_cnt
  ...

1 files changed, 1153 insertions, 496 deletions

diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 1cc945daa9c8..04c6630cc18f 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -19,6 +19,8 @@
 #include <linux/sort.h>
 #include <linux/perf_event.h>
 #include <linux/ctype.h>
+#include <linux/error-injection.h>
+#include <linux/bpf_lsm.h>
 
 #include "disasm.h"
 
@@ -227,8 +229,7 @@ struct bpf_call_arg_meta {
 	bool pkt_access;
 	int regno;
 	int access_size;
-	s64 msize_smax_value;
-	u64 msize_umax_value;
+	u64 msize_max_value;
 	int ref_obj_id;
 	int func_id;
 	u32 btf_id;
@@ -549,6 +550,22 @@ static void print_verifier_state(struct bpf_verifier_env *env,
 					tnum_strn(tn_buf, sizeof(tn_buf), reg->var_off);
 					verbose(env, ",var_off=%s", tn_buf);
 				}
+				if (reg->s32_min_value != reg->smin_value &&
+				    reg->s32_min_value != S32_MIN)
+					verbose(env, ",s32_min_value=%d",
+						(int)(reg->s32_min_value));
+				if (reg->s32_max_value != reg->smax_value &&
+				    reg->s32_max_value != S32_MAX)
+					verbose(env, ",s32_max_value=%d",
+						(int)(reg->s32_max_value));
+				if (reg->u32_min_value != reg->umin_value &&
+				    reg->u32_min_value != U32_MIN)
+					verbose(env, ",u32_min_value=%d",
+						(int)(reg->u32_min_value));
+				if (reg->u32_max_value != reg->umax_value &&
+				    reg->u32_max_value != U32_MAX)
+					verbose(env, ",u32_max_value=%d",
+						(int)(reg->u32_max_value));
 			}
 			verbose(env, ")");
 		}
@@ -923,6 +940,20 @@ static void __mark_reg_known(struct bpf_reg_state *reg, u64 imm)
 	reg->smax_value = (s64)imm;
 	reg->umin_value = imm;
 	reg->umax_value = imm;
+
+	reg->s32_min_value = (s32)imm;
+	reg->s32_max_value = (s32)imm;
+	reg->u32_min_value = (u32)imm;
+	reg->u32_max_value = (u32)imm;
+}
+
+static void __mark_reg32_known(struct bpf_reg_state *reg, u64 imm)
+{
+	reg->var_off = tnum_const_subreg(reg->var_off, imm);
+	reg->s32_min_value = (s32)imm;
+	reg->s32_max_value = (s32)imm;
+	reg->u32_min_value = (u32)imm;
+	reg->u32_max_value = (u32)imm;
 }
 
 /* Mark the 'variable offset' part of a register as zero.  This should be
@@ -977,8 +1008,52 @@ static bool reg_is_init_pkt_pointer(const struct bpf_reg_state *reg,
 	       tnum_equals_const(reg->var_off, 0);
 }
 
-/* Attempts to improve min/max values based on var_off information */
-static void __update_reg_bounds(struct bpf_reg_state *reg)
+/* Reset the min/max bounds of a register */
+static void __mark_reg_unbounded(struct bpf_reg_state *reg)
+{
+	reg->smin_value = S64_MIN;
+	reg->smax_value = S64_MAX;
+	reg->umin_value = 0;
+	reg->umax_value = U64_MAX;
+
+	reg->s32_min_value = S32_MIN;
+	reg->s32_max_value = S32_MAX;
+	reg->u32_min_value = 0;
+	reg->u32_max_value = U32_MAX;
+}
+
+static void __mark_reg64_unbounded(struct bpf_reg_state *reg)
+{
+	reg->smin_value = S64_MIN;
+	reg->smax_value = S64_MAX;
+	reg->umin_value = 0;
+	reg->umax_value = U64_MAX;
+}
+
+static void __mark_reg32_unbounded(struct bpf_reg_state *reg)
+{
+	reg->s32_min_value = S32_MIN;
+	reg->s32_max_value = S32_MAX;
+	reg->u32_min_value = 0;
+	reg->u32_max_value = U32_MAX;
+}
+
+static void __update_reg32_bounds(struct bpf_reg_state *reg)
+{
+	struct tnum var32_off = tnum_subreg(reg->var_off);
+
+	/* min signed is max(sign bit) | min(other bits) */
+	reg->s32_min_value = max_t(s32, reg->s32_min_value,
+			var32_off.value | (var32_off.mask & S32_MIN));
+	/* max signed is min(sign bit) | max(other bits) */
+	reg->s32_max_value = min_t(s32, reg->s32_max_value,
+			var32_off.value | (var32_off.mask & S32_MAX));
+	reg->u32_min_value = max_t(u32, reg->u32_min_value, (u32)var32_off.value);
+	reg->u32_max_value = min(reg->u32_max_value,
+				 (u32)(var32_off.value | var32_off.mask));
+}
+
+static void __update_reg64_bounds(struct bpf_reg_state *reg)
 {
 	/* min signed is max(sign bit) | min(other bits) */
 	reg->smin_value = max_t(s64, reg->smin_value,
@@ -991,8 +1066,48 @@ static void __update_reg_bounds(struct bpf_reg_state *reg)
 			      reg->var_off.value | reg->var_off.mask);
 }
 
+static void __update_reg_bounds(struct bpf_reg_state *reg)
+{
+	__update_reg32_bounds(reg);
+	__update_reg64_bounds(reg);
+}
+
 /* Uses signed min/max values to inform unsigned, and vice-versa */
-static void __reg_deduce_bounds(struct bpf_reg_state *reg)
+static void __reg32_deduce_bounds(struct bpf_reg_state *reg)
+{
+	/* Learn sign from signed bounds.
+	 * If we cannot cross the sign boundary, then signed and unsigned bounds
+	 * are the same, so combine.  This works even in the negative case, e.g.
+	 * -3 s<= x s<= -1 implies 0xf...fd u<= x u<= 0xf...ff.
+	 */
+	if (reg->s32_min_value >= 0 || reg->s32_max_value < 0) {
+		reg->s32_min_value = reg->u32_min_value =
+			max_t(u32, reg->s32_min_value, reg->u32_min_value);
+		reg->s32_max_value = reg->u32_max_value =
+			min_t(u32, reg->s32_max_value, reg->u32_max_value);
+		return;
+	}
+	/* Learn sign from unsigned bounds.  Signed bounds cross the sign
+	 * boundary, so we must be careful.
+	 */
+	if ((s32)reg->u32_max_value >= 0) {
+		/* Positive.  We can't learn anything from the smin, but smax
+		 * is positive, hence safe.
+		 */
+		reg->s32_min_value = reg->u32_min_value;
+		reg->s32_max_value = reg->u32_max_value =
+			min_t(u32, reg->s32_max_value, reg->u32_max_value);
+	} else if ((s32)reg->u32_min_value < 0) {
+		/* Negative.  We can't learn anything from the smax, but smin
+		 * is negative, hence safe.
+		 */
+		reg->s32_min_value = reg->u32_min_value =
+			max_t(u32, reg->s32_min_value, reg->u32_min_value);
+		reg->s32_max_value = reg->u32_max_value;
+	}
+}
+
+static void __reg64_deduce_bounds(struct bpf_reg_state *reg)
 {
 	/* Learn sign from signed bounds.
 	 * If we cannot cross the sign boundary, then signed and unsigned bounds
@@ -1026,32 +1141,106 @@ static void __reg_deduce_bounds(struct bpf_reg_state *reg)
 	}
 }
 
+static void __reg_deduce_bounds(struct bpf_reg_state *reg)
+{
+	__reg32_deduce_bounds(reg);
+	__reg64_deduce_bounds(reg);
+}
+
 /* Attempts to improve var_off based on unsigned min/max information */
 static void __reg_bound_offset(struct bpf_reg_state *reg)
 {
-	reg->var_off = tnum_intersect(reg->var_off,
-				      tnum_range(reg->umin_value,
-						 reg->umax_value));
+	struct tnum var64_off = tnum_intersect(reg->var_off,
+					       tnum_range(reg->umin_value,
+							  reg->umax_value));
+	struct tnum var32_off = tnum_intersect(tnum_subreg(reg->var_off),
+						tnum_range(reg->u32_min_value,
+							   reg->u32_max_value));
+
+	reg->var_off = tnum_or(tnum_clear_subreg(var64_off), var32_off);
+}
+
+static void __reg_assign_32_into_64(struct bpf_reg_state *reg)
+{
+	reg->umin_value = reg->u32_min_value;
+	reg->umax_value = reg->u32_max_value;
+	/* Attempt to pull 32-bit signed bounds into 64-bit bounds
+	 * but must be positive otherwise set to worse case bounds
+	 * and refine later from tnum.
+	 */
+	if (reg->s32_min_value > 0)
+		reg->smin_value = reg->s32_min_value;
+	else
+		reg->smin_value = 0;
+	if (reg->s32_max_value > 0)
+		reg->smax_value = reg->s32_max_value;
+	else
+		reg->smax_value = U32_MAX;
 }
 
-static void __reg_bound_offset32(struct bpf_reg_state *reg)
+static void __reg_combine_32_into_64(struct bpf_reg_state *reg)
 {
-	u64 mask = 0xffffFFFF;
-	struct tnum range = tnum_range(reg->umin_value & mask,
-				       reg->umax_value & mask);
-	struct tnum lo32 = tnum_cast(reg->var_off, 4);
-	struct tnum hi32 = tnum_lshift(tnum_rshift(reg->var_off, 32), 32);
+	/* special case when 64-bit register has upper 32-bit register
+	 * zeroed. Typically happens after zext or <<32, >>32 sequence
+	 * allowing us to use 32-bit bounds directly,
+	 */
+	if (tnum_equals_const(tnum_clear_subreg(reg->var_off), 0)) {
+		__reg_assign_32_into_64(reg);
+	} else {
+		/* Otherwise the best we can do is push lower 32bit known and
+		 * unknown bits into register (var_off set from jmp logic)
+		 * then learn as much as possible from the 64-bit tnum
+		 * known and unknown bits. The previous smin/smax bounds are
+		 * invalid here because of jmp32 compare so mark them unknown
+		 * so they do not impact tnum bounds calculation.
+		 */
+		__mark_reg64_unbounded(reg);
+		__update_reg_bounds(reg);
+	}
 
-	reg->var_off = tnum_or(hi32, tnum_intersect(lo32, range));
+	/* Intersecting with the old var_off might have improved our bounds
+	 * slightly.  e.g. if umax was 0x7f...f and var_off was (0; 0xf...fc),
+	 * then new var_off is (0; 0x7f...fc) which improves our umax.
+	 */
+	__reg_deduce_bounds(reg);
+	__reg_bound_offset(reg);
+	__update_reg_bounds(reg);
 }
 
-/* Reset the min/max bounds of a register */
-static void __mark_reg_unbounded(struct bpf_reg_state *reg)
+static bool __reg64_bound_s32(s64 a)
 {
-	reg->smin_value = S64_MIN;
-	reg->smax_value = S64_MAX;
-	reg->umin_value = 0;
-	reg->umax_value = U64_MAX;
+	if (a > S32_MIN && a < S32_MAX)
+		return true;
+	return false;
+}
+
+static bool __reg64_bound_u32(u64 a)
+{
+	if (a > U32_MIN && a < U32_MAX)
+		return true;
+	return false;
+}
+
+static void __reg_combine_64_into_32(struct bpf_reg_state *reg)
+{
+	__mark_reg32_unbounded(reg);
+
+	if (__reg64_bound_s32(reg->smin_value))
+		reg->s32_min_value = (s32)reg->smin_value;
+	if (__reg64_bound_s32(reg->smax_value))
+		reg->s32_max_value = (s32)reg->smax_value;
+	if (__reg64_bound_u32(reg->umin_value))
+		reg->u32_min_value = (u32)reg->umin_value;
+	if (__reg64_bound_u32(reg->umax_value))
+		reg->u32_max_value = (u32)reg->umax_value;
+
+	/* Intersecting with the old var_off might have improved our bounds
+	 * slightly.  e.g. if umax was 0x7f...f and var_off was (0; 0xf...fc),
+	 * then new var_off is (0; 0x7f...fc) which improves our umax.
+	 */
+	__reg_deduce_bounds(reg);
+	__reg_bound_offset(reg);
+	__update_reg_bounds(reg);
 }
 
 /* Mark a register as having a completely unknown (scalar) value. */
@@ -2784,6 +2973,12 @@ static int check_tp_buffer_access(struct bpf_verifier_env *env,
 	return 0;
 }
 
+/* BPF architecture zero extends alu32 ops into 64-bit registesr */
+static void zext_32_to_64(struct bpf_reg_state *reg)
+{
+	reg->var_off = tnum_subreg(reg->var_off);
+	__reg_assign_32_into_64(reg);
+}
 
 /* truncate register to smaller size (in bytes)
  * must be called with size < BPF_REG_SIZE
@@ -2806,6 +3001,14 @@ static void coerce_reg_to_size(struct bpf_reg_state *reg, int size)
 	}
 	reg->smin_value = reg->umin_value;
 	reg->smax_value = reg->umax_value;
+
+	/* If size is smaller than 32bit register the 32bit register
+	 * values are also truncated so we push 64-bit bounds into
+	 * 32-bit bounds. Above were truncated < 32-bits already.
+	 */
+	if (size >= 4)
+		return;
+	__reg_combine_64_into_32(reg);
 }
 
 static bool bpf_map_is_rdonly(const struct bpf_map *map)
@@ -3460,13 +3663,17 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno,
 		expected_type = CONST_PTR_TO_MAP;
 		if (type != expected_type)
 			goto err_type;
-	} else if (arg_type == ARG_PTR_TO_CTX) {
+	} else if (arg_type == ARG_PTR_TO_CTX ||
+		   arg_type == ARG_PTR_TO_CTX_OR_NULL) {
 		expected_type = PTR_TO_CTX;
-		if (type != expected_type)
-			goto err_type;
-		err = check_ctx_reg(env, reg, regno);
-		if (err < 0)
-			return err;
+		if (!(register_is_null(reg) &&
+		      arg_type == ARG_PTR_TO_CTX_OR_NULL)) {
+			if (type != expected_type)
+				goto err_type;
+			err = check_ctx_reg(env, reg, regno);
+			if (err < 0)
+				return err;
+		}
 	} else if (arg_type == ARG_PTR_TO_SOCK_COMMON) {
 		expected_type = PTR_TO_SOCK_COMMON;
 		/* Any sk pointer can be ARG_PTR_TO_SOCK_COMMON */
@@ -3576,11 +3783,15 @@ static int check_func_arg(struct bpf_verifier_env *env, u32 regno,
 	} else if (arg_type_is_mem_size(arg_type)) {
 		bool zero_size_allowed = (arg_type == ARG_CONST_SIZE_OR_ZERO);
 
-		/* remember the mem_size which may be used later
-		 * to refine return values.
+		/* This is used to refine r0 return value bounds for helpers
+		 * that enforce this value as an upper bound on return values.
+		 * See do_refine_retval_range() for helpers that can refine
+		 * the return value. C type of helper is u32 so we pull register
+		 * bound from umax_value however, if negative verifier errors
+		 * out. Only upper bounds can be learned because retval is an
+		 * int type and negative retvals are allowed.
 		 */
-		meta->msize_smax_value = reg->smax_value;
-		meta->msize_umax_value = reg->umax_value;
+		meta->msize_max_value = reg->umax_value;
 
 		/* The register is SCALAR_VALUE; the access check
 		 * happens using its boundaries.
@@ -3649,7 +3860,8 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 		if (func_id != BPF_FUNC_perf_event_read &&
 		    func_id != BPF_FUNC_perf_event_output &&
 		    func_id != BPF_FUNC_skb_output &&
-		    func_id != BPF_FUNC_perf_event_read_value)
+		    func_id != BPF_FUNC_perf_event_read_value &&
+		    func_id != BPF_FUNC_xdp_output)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_STACK_TRACE:
@@ -3693,14 +3905,16 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 		if (func_id != BPF_FUNC_sk_redirect_map &&
 		    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_msg_redirect_map &&
+		    func_id != BPF_FUNC_sk_select_reuseport)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_SOCKHASH:
 		if (func_id != BPF_FUNC_sk_redirect_hash &&
 		    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_msg_redirect_hash &&
+		    func_id != BPF_FUNC_sk_select_reuseport)
 			goto error;
 		break;
 	case BPF_MAP_TYPE_REUSEPORT_SOCKARRAY:
@@ -3737,6 +3951,7 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 	case BPF_FUNC_perf_event_output:
 	case BPF_FUNC_perf_event_read_value:
 	case BPF_FUNC_skb_output:
+	case BPF_FUNC_xdp_output:
 		if (map->map_type != BPF_MAP_TYPE_PERF_EVENT_ARRAY)
 			goto error;
 		break;
@@ -3774,7 +3989,9 @@ static int check_map_func_compatibility(struct bpf_verifier_env *env,
 			goto error;
 		break;
 	case BPF_FUNC_sk_select_reuseport:
-		if (map->map_type != BPF_MAP_TYPE_REUSEPORT_SOCKARRAY)
+		if (map->map_type != BPF_MAP_TYPE_REUSEPORT_SOCKARRAY &&
+		    map->map_type != BPF_MAP_TYPE_SOCKMAP &&
+		    map->map_type != BPF_MAP_TYPE_SOCKHASH)
 			goto error;
 		break;
 	case BPF_FUNC_map_peek_elem:
@@ -4117,10 +4334,11 @@ static void do_refine_retval_range(struct bpf_reg_state *regs, int ret_type,
 	     func_id != BPF_FUNC_probe_read_str))
 		return;
 
-	ret_reg->smax_value = meta->msize_smax_value;
-	ret_reg->umax_value = meta->msize_umax_value;
+	ret_reg->smax_value = meta->msize_max_value;
+	ret_reg->s32_max_value = meta->msize_max_value;
 	__reg_deduce_bounds(ret_reg);
 	__reg_bound_offset(ret_reg);
+	__update_reg_bounds(ret_reg);
 }
 
 static int
@@ -4427,7 +4645,17 @@ static bool signed_add_overflows(s64 a, s64 b)
 	return res < a;
 }
 
-static bool signed_sub_overflows(s64 a, s64 b)
+static bool signed_add32_overflows(s64 a, s64 b)
+{
+	/* Do the add in u32, where overflow is well-defined */
+	s32 res = (s32)((u32)a + (u32)b);
+
+	if (b < 0)
+		return res > a;
+	return res < a;
+}
+
+static bool signed_sub_overflows(s32 a, s32 b)
 {
 	/* Do the sub in u64, where overflow is well-defined */
 	s64 res = (s64)((u64)a - (u64)b);
@@ -4437,6 +4665,16 @@ static bool signed_sub_overflows(s64 a, s64 b)
 	return res > a;
 }
 
+static bool signed_sub32_overflows(s32 a, s32 b)
+{
+	/* Do the sub in u64, where overflow is well-defined */
+	s32 res = (s32)((u32)a - (u32)b);
+
+	if (b < 0)
+		return res < a;
+	return res > a;
+}
+
 static bool check_reg_sane_offset(struct bpf_verifier_env *env,
 				  const struct bpf_reg_state *reg,
 				  enum bpf_reg_type type)
@@ -4673,6 +4911,9 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env,
 	    !check_reg_sane_offset(env, ptr_reg, ptr_reg->type))
 		return -EINVAL;
 
+	/* pointer types do not carry 32-bit bounds at the moment. */
+	__mark_reg32_unbounded(dst_reg);
+
 	switch (opcode) {
 	case BPF_ADD:
 		ret = sanitize_ptr_alu(env, insn, ptr_reg, dst_reg, smin_val < 0);
@@ -4836,6 +5077,518 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env,
 	return 0;
 }
 
+static void scalar32_min_max_add(struct bpf_reg_state *dst_reg,
+				 struct bpf_reg_state *src_reg)
+{
+	s32 smin_val = src_reg->s32_min_value;
+	s32 smax_val = src_reg->s32_max_value;
+	u32 umin_val = src_reg->u32_min_value;
+	u32 umax_val = src_reg->u32_max_value;
+
+	if (signed_add32_overflows(dst_reg->s32_min_value, smin_val) ||
+	    signed_add32_overflows(dst_reg->s32_max_value, smax_val)) {
+		dst_reg->s32_min_value = S32_MIN;
+		dst_reg->s32_max_value = S32_MAX;
+	} else {
+		dst_reg->s32_min_value += smin_val;
+		dst_reg->s32_max_value += smax_val;
+	}
+	if (dst_reg->u32_min_value + umin_val < umin_val ||
+	    dst_reg->u32_max_value + umax_val < umax_val) {
+		dst_reg->u32_min_value = 0;
+		dst_reg->u32_max_value = U32_MAX;
+	} else {
+		dst_reg->u32_min_value += umin_val;
+		dst_reg->u32_max_value += umax_val;
+	}
+}
+
+static void scalar_min_max_add(struct bpf_reg_state *dst_reg,
+			       struct bpf_reg_state *src_reg)
+{
+	s64 smin_val = src_reg->smin_value;
+	s64 smax_val = src_reg->smax_value;
+	u64 umin_val = src_reg->umin_value;
+	u64 umax_val = src_reg->umax_value;
+
+	if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
+	    signed_add_overflows(dst_reg->smax_value, smax_val)) {
+		dst_reg->smin_value = S64_MIN;
+		dst_reg->smax_value = S64_MAX;
+	} else {
+		dst_reg->smin_value += smin_val;
+		dst_reg->smax_value += smax_val;
+	}
+	if (dst_reg->umin_value + umin_val < umin_val ||
+	    dst_reg->umax_value + umax_val < umax_val) {
+		dst_reg->umin_value = 0;
+		dst_reg->umax_value = U64_MAX;
+	} else {
+		dst_reg->umin_value += umin_val;
+		dst_reg->umax_value += umax_val;
+	}
+}
+
+static void scalar32_min_max_sub(struct bpf_reg_state *dst_reg,
+				 struct bpf_reg_state *src_reg)
+{
+	s32 smin_val = src_reg->s32_min_value;
+	s32 smax_val = src_reg->s32_max_value;
+	u32 umin_val = src_reg->u32_min_value;
+	u32 umax_val = src_reg->u32_max_value;
+
+	if (signed_sub32_overflows(dst_reg->s32_min_value, smax_val) ||
+	    signed_sub32_overflows(dst_reg->s32_max_value, smin_val)) {
+		/* Overflow possible, we know nothing */
+		dst_reg->s32_min_value = S32_MIN;
+		dst_reg->s32_max_value = S32_MAX;
+	} else {
+		dst_reg->s32_min_value -= smax_val;
+		dst_reg->s32_max_value -= smin_val;
+	}
+	if (dst_reg->u32_min_value < umax_val) {
+		/* Overflow possible, we know nothing */
+		dst_reg->u32_min_value = 0;
+		dst_reg->u32_max_value = U32_MAX;
+	} else {
+		/* Cannot overflow (as long as bounds are consistent) */
+		dst_reg->u32_min_value -= umax_val;
+		dst_reg->u32_max_value -= umin_val;
+	}
+}
+
+static void scalar_min_max_sub(struct bpf_reg_state *dst_reg,
+			       struct bpf_reg_state *src_reg)
+{
+	s64 smin_val = src_reg->smin_value;
+	s64 smax_val = src_reg->smax_value;
+	u64 umin_val = src_reg->umin_value;
+	u64 umax_val = src_reg->umax_value;
+
+	if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
+	    signed_sub_overflows(dst_reg->smax_value, smin_val)) {
+		/* Overflow possible, we know nothing */
+		dst_reg->smin_value = S64_MIN;
+		dst_reg->smax_value = S64_MAX;
+	} else {
+		dst_reg->smin_value -= smax_val;
+		dst_reg->smax_value -= smin_val;
+	}
+	if (dst_reg->umin_value < umax_val) {
+		/* Overflow possible, we know nothing */
+		dst_reg->umin_value = 0;
+		dst_reg->umax_value = U64_MAX;
+	} else {
+		/* Cannot overflow (as long as bounds are consistent) */
+		dst_reg->umin_value -= umax_val;
+		dst_reg->umax_value -= umin_val;
+	}
+}
+
+static void scalar32_min_max_mul(struct bpf_reg_state *dst_reg,
+				 struct bpf_reg_state *src_reg)
+{
+	s32 smin_val = src_reg->s32_min_value;
+	u32 umin_val = src_reg->u32_min_value;
+	u32 umax_val = src_reg->u32_max_value;
+
+	if (smin_val < 0 || dst_reg->s32_min_value < 0) {
+		/* Ain't nobody got time to multiply that sign */
+		__mark_reg32_unbounded(dst_reg);
+		return;
+	}
+	/* Both values are positive, so we can work with unsigned and
+	 * copy the result to signed (unless it exceeds S32_MAX).
+	 */
+	if (umax_val > U16_MAX || dst_reg->u32_max_value > U16_MAX) {
+		/* Potential overflow, we know nothing */
+		__mark_reg32_unbounded(dst_reg);
+		return;
+	}
+	dst_reg->u32_min_value *= umin_val;
+	dst_reg->u32_max_value *= umax_val;
+	if (dst_reg->u32_max_value > S32_MAX) {
+		/* Overflow possible, we know nothing */
+		dst_reg->s32_min_value = S32_MIN;
+		dst_reg->s32_max_value = S32_MAX;
+	} else {
+		dst_reg->s32_min_value = dst_reg->u32_min_value;
+		dst_reg->s32_max_value = dst_reg->u32_max_value;
+	}
+}
+
+static void scalar_min_max_mul(struct bpf_reg_state *dst_reg,
+			       struct bpf_reg_state *src_reg)
+{
+	s64 smin_val = src_reg->smin_value;
+	u64 umin_val = src_reg->umin_value;
+	u64 umax_val = src_reg->umax_value;
+
+	if (smin_val < 0 || dst_reg->smin_value < 0) {
+		/* Ain't nobody got time to multiply that sign */
+		__mark_reg64_unbounded(dst_reg);
+		return;
+	}
+	/* Both values are positive, so we can work with unsigned and
+	 * copy the result to signed (unless it exceeds S64_MAX).
+	 */
+	if (umax_val > U32_MAX || dst_reg->umax_value > U32_MAX) {
+		/* Potential overflow, we know nothing */
+		__mark_reg64_unbounded(dst_reg);
+		return;
+	}
+	dst_reg->umin_value *= umin_val;
+	dst_reg->umax_value *= umax_val;
+	if (dst_reg->umax_value > S64_MAX) {
+		/* Overflow possible, we know nothing */
+		dst_reg->smin_value = S64_MIN;
+		dst_reg->smax_value = S64_MAX;
+	} else {
+		dst_reg->smin_value = dst_reg->umin_value;
+		dst_reg->smax_value = dst_reg->umax_value;
+	}
+}
+
+static void scalar32_min_max_and(struct bpf_reg_state *dst_reg,
+				 struct bpf_reg_state *src_reg)
+{
+	bool src_known = tnum_subreg_is_const(src_reg->var_off);
+	bool dst_known = tnum_subreg_is_const(dst_reg->var_off);
+	struct tnum var32_off = tnum_subreg(dst_reg->var_off);
+	s32 smin_val = src_reg->s32_min_value;
+	u32 umax_val = src_reg->u32_max_value;
+
+	/* Assuming scalar64_min_max_and will be called so its safe
+	 * to skip updating register for known 32-bit case.
+	 */
+	if (src_known && dst_known)
+		return;
+
+	/* We get our minimum from the var_off, since that's inherently
+	 * bitwise.  Our maximum is the minimum of the operands' maxima.
+	 */
+	dst_reg->u32_min_value = var32_off.value;
+	dst_reg->u32_max_value = min(dst_reg->u32_max_value, umax_val);
+	if (dst_reg->s32_min_value < 0 || smin_val < 0) {
+		/* Lose signed bounds when ANDing negative numbers,
+		 * ain't nobody got time for that.
+		 */
+		dst_reg->s32_min_value = S32_MIN;
+		dst_reg->s32_max_value = S32_MAX;
+	} else {
+		/* ANDing two positives gives a positive, so safe to
+		 * cast result into s64.
+		 */
+		dst_reg->s32_min_value = dst_reg->u32_min_value;
+		dst_reg->s32_max_value = dst_reg->u32_max_value;
+	}
+
+}
+
+static void scalar_min_max_and(struct bpf_reg_state *dst_reg,
+			       struct bpf_reg_state *src_reg)
+{
+	bool src_known = tnum_is_const(src_reg->var_off);
+	bool dst_known = tnum_is_const(dst_reg->var_off);
+	s64 smin_val = src_reg->smin_value;
+	u64 umax_val = src_reg->umax_value;
+
+	if (src_known && dst_known) {
+		__mark_reg_known(dst_reg, dst_reg->var_off.value &
+					  src_reg->var_off.value);
+		return;
+	}
+
+	/* We get our minimum from the var_off, since that's inherently
+	 * bitwise.  Our maximum is the minimum of the operands' maxima.
+	 */
+	dst_reg->umin_value = dst_reg->var_off.value;
+	dst_reg->umax_value = min(dst_reg->umax_value, umax_val);
+	if (dst_reg->smin_value < 0 || smin_val < 0) {
+		/* Lose signed bounds when ANDing negative numbers,
+		 * ain't nobody got time for that.
+		 */
+		dst_reg->smin_value = S64_MIN;
+		dst_reg->smax_value = S64_MAX;
+	} else {
+		/* ANDing two positives gives a positive, so safe to
+		 * cast result into s64.
+		 */
+		dst_reg->smin_value = dst_reg->umin_value;
+		dst_reg->smax_value = dst_reg->umax_value;
+	}
+	/* We may learn something more from the var_off */
+	__update_reg_bounds(dst_reg);
+}
+
+static void scalar32_min_max_or(struct bpf_reg_state *dst_reg,
+				struct bpf_reg_state *src_reg)
+{
+	bool src_known = tnum_subreg_is_const(src_reg->var_off);
+	bool dst_known = tnum_subreg_is_const(dst_reg->var_off);
+	struct tnum var32_off = tnum_subreg(dst_reg->var_off);
+	s32 smin_val = src_reg->smin_value;
+	u32 umin_val = src_reg->umin_value;
+
+	/* Assuming scalar64_min_max_or will be called so it is safe
+	 * to skip updating register for known case.
+	 */
+	if (src_known && dst_known)
+		return;
+
+	/* We get our maximum from the var_off, and our minimum is the
+	 * maximum of the operands' minima
+	 */
+	dst_reg->u32_min_value = max(dst_reg->u32_min_value, umin_val);
+	dst_reg->u32_max_value = var32_off.value | var32_off.mask;
+	if (dst_reg->s32_min_value < 0 || smin_val < 0) {
+		/* Lose signed bounds when ORing negative numbers,
+		 * ain't nobody got time for that.
+		 */
+		dst_reg->s32_min_value = S32_MIN;
+		dst_reg->s32_max_value = S32_MAX;
+	} else {
+		/* ORing two positives gives a positive, so safe to
+		 * cast result into s64.
+		 */
+		dst_reg->s32_min_value = dst_reg->umin_value;
+		dst_reg->s32_max_value = dst_reg->umax_value;
+	}
+}
+
+static void scalar_min_max_or(struct bpf_reg_state *dst_reg,
+			      struct bpf_reg_state *src_reg)
+{
+	bool src_known = tnum_is_const(src_reg->var_off);
+	bool dst_known = tnum_is_const(dst_reg->var_off);
+	s64 smin_val = src_reg->smin_value;
+	u64 umin_val = src_reg->umin_value;
+
+	if (src_known && dst_known) {
+		__mark_reg_known(dst_reg, dst_reg->var_off.value |
+					  src_reg->var_off.value);
+		return;
+	}
+
+	/* We get our maximum from the var_off, and our minimum is the
+	 * maximum of the operands' minima
+	 */
+	dst_reg->umin_value = max(dst_reg->umin_value, umin_val);
+	dst_reg->umax_value = dst_reg->var_off.value | dst_reg->var_off.mask;
+	if (dst_reg->smin_value < 0 || smin_val < 0) {
+		/* Lose signed bounds when ORing negative numbers,
+		 * ain't nobody got time for that.
+		 */
+		dst_reg->smin_value = S64_MIN;
+		dst_reg->smax_value = S64_MAX;
+	} else {
+		/* ORing two positives gives a positive, so safe to
+		 * cast result into s64.
+		 */
+		dst_reg->smin_value = dst_reg->umin_value;
+		dst_reg->smax_value = dst_reg->umax_value;
+	}
+	/* We may learn something more from the var_off */
+	__update_reg_bounds(dst_reg);
+}
+
+static void __scalar32_min_max_lsh(struct bpf_reg_state *dst_reg,
+				   u64 umin_val, u64 umax_val)
+{
+	/* We lose all sign bit information (except what we can pick
+	 * up from var_off)
+	 */
+	dst_reg->s32_min_value = S32_MIN;
+	dst_reg->s32_max_value = S32_MAX;
+	/* If we might shift our top bit out, then we know nothing */
+	if (umax_val > 31 || dst_reg->u32_max_value > 1ULL << (31 - umax_val)) {
+		dst_reg->u32_min_value = 0;
+		dst_reg->u32_max_value = U32_MAX;
+	} else {
+		dst_reg->u32_min_value <<= umin_val;
+		dst_reg->u32_max_value <<= umax_val;
+	}
+}
+
+static void scalar32_min_max_lsh(struct bpf_reg_state *dst_reg,
+				 struct bpf_reg_state *src_reg)
+{
+	u32 umax_val = src_reg->u32_max_value;
+	u32 umin_val = src_reg->u32_min_value;
+	/* u32 alu operation will zext upper bits */
+	struct tnum subreg = tnum_subreg(dst_reg->var_off);
+
+	__scalar32_min_max_lsh(dst_reg, umin_val, umax_val);
+	dst_reg->var_off = tnum_subreg(tnum_lshift(subreg, umin_val));
+	/* Not required but being careful mark reg64 bounds as unknown so
+	 * that we are forced to pick them up from tnum and zext later and
+	 * if some path skips this step we are still safe.
+	 */
+	__mark_reg64_unbounded(dst_reg);
+	__update_reg32_bounds(dst_reg);
+}
+
+static void __scalar64_min_max_lsh(struct bpf_reg_state *dst_reg,
+				   u64 umin_val, u64 umax_val)
+{
+	/* Special case <<32 because it is a common compiler pattern to sign
+	 * extend subreg by doing <<32 s>>32. In this case if 32bit bounds are
+	 * positive we know this shift will also be positive so we can track
+	 * bounds correctly. Otherwise we lose all sign bit information except
+	 * what we can pick up from var_off. Perhaps we can generalize this
+	 * later to shifts of any length.
+	 */
+	if (umin_val == 32 && umax_val == 32 && dst_reg->s32_max_value >= 0)
+		dst_reg->smax_value = (s64)dst_reg->s32_max_value << 32;
+	else
+		dst_reg->smax_value = S64_MAX;
+
+	if (umin_val == 32 && umax_val == 32 && dst_reg->s32_min_value >= 0)
+		dst_reg->smin_value = (s64)dst_reg->s32_min_value << 32;
+	else
+		dst_reg->smin_value = S64_MIN;
+
+	/* If we might shift our top bit out, then we know nothing */
+	if (dst_reg->umax_value > 1ULL << (63 - umax_val)) {
+		dst_reg->umin_value = 0;
+		dst_reg->umax_value = U64_MAX;
+	} else {
+		dst_reg->umin_value <<= umin_val;
+		dst_reg->umax_value <<= umax_val;
+	}
+}
+
+static void scalar_min_max_lsh(struct bpf_reg_state *dst_reg,
+			       struct bpf_reg_state *src_reg)
+{
+	u64 umax_val = src_reg->umax_value;
+	u64 umin_val = src_reg->umin_value;
+
+	/* scalar64 calc uses 32bit unshifted bounds so must be called first */
+	__scalar64_min_max_lsh(dst_reg, umin_val, umax_val);
+	__scalar32_min_max_lsh(dst_reg, umin_val, umax_val);
+
+	dst_reg->var_off = tnum_lshift(dst_reg->var_off, umin_val);
+	/* We may learn something more from the var_off */
+	__update_reg_bounds(dst_reg);
+}
+
+static void scalar32_min_max_rsh(struct bpf_reg_state *dst_reg,
+				 struct bpf_reg_state *src_reg)
+{
+	struct tnum subreg = tnum_subreg(dst_reg->var_off);
+	u32 umax_val = src_reg->u32_max_value;
+	u32 umin_val = src_reg->u32_min_value;
+
+	/* BPF_RSH is an unsigned shift.  If the value in dst_reg might
+	 * be negative, then either:
+	 * 1) src_reg might be zero, so the sign bit of the result is
+	 *    unknown, so we lose our signed bounds
+	 * 2) it's known negative, thus the unsigned bounds capture the
+	 *    signed bounds
+	 * 3) the signed bounds cross zero, so they tell us nothing
+	 *    about the result
+	 * If the value in dst_reg is known nonnegative, then again the
+	 * unsigned bounts capture the signed bounds.
+	 * Thus, in all cases it suffices to blow away our signed bounds
+	 * and rely on inferring new ones from the unsigned bounds and
+	 * var_off of the result.
+	 */
+	dst_reg->s32_min_value = S32_MIN;
+	dst_reg->s32_max_value = S32_MAX;
+
+	dst_reg->var_off = tnum_rshift(subreg, umin_val);
+	dst_reg->u32_min_value >>= umax_val;
+	dst_reg->u32_max_value >>= umin_val;
+
+	__mark_reg64_unbounded(dst_reg);
+	__update_reg32_bounds(dst_reg);
+}
+
+static void scalar_min_max_rsh(struct bpf_reg_state *dst_reg,
+			       struct bpf_reg_state *src_reg)
+{
+	u64 umax_val = src_reg->umax_value;
+	u64 umin_val = src_reg->umin_value;
+
+	/* BPF_RSH is an unsigned shift.  If the value in dst_reg might
+	 * be negative, then either:
+	 * 1) src_reg might be zero, so the sign bit of the result is
+	 *    unknown, so we lose our signed bounds
+	 * 2) it's known negative, thus the unsigned bounds capture the
+	 *    signed bounds
+	 * 3) the signed bounds cross zero, so they tell us nothing
+	 *    about the result
+	 * If the value in dst_reg is known nonnegative, then again the
+	 * unsigned bounts capture the signed bounds.
+	 * Thus, in all cases it suffices to blow away our signed bounds
+	 * and rely on inferring new ones from the unsigned bounds and
+	 * var_off of the result.
+	 */
+	dst_reg->smin_value = S64_MIN;
+	dst_reg->smax_value = S64_MAX;
+	dst_reg->var_off = tnum_rshift(dst_reg->var_off, umin_val);
+	dst_reg->umin_value >>= umax_val;
+	dst_reg->umax_value >>= umin_val;
+
+	/* Its not easy to operate on alu32 bounds here because it depends
+	 * on bits being shifted in. Take easy way out and mark unbounded
+	 * so we can recalculate later from tnum.
+	 */
+	__mark_reg32_unbounded(dst_reg);
+	__update_reg_bounds(dst_reg);
+}
+
+static void scalar32_min_max_arsh(struct bpf_reg_state *dst_reg,
+				  struct bpf_reg_state *src_reg)
+{
+	u64 umin_val = src_reg->u32_min_value;
+
+	/* Upon reaching here, src_known is true and
+	 * umax_val is equal to umin_val.
+	 */
+	dst_reg->s32_min_value = (u32)(((s32)dst_reg->s32_min_value) >> umin_val);
+	dst_reg->s32_max_value = (u32)(((s32)dst_reg->s32_max_value) >> umin_val);
+
+	dst_reg->var_off = tnum_arshift(tnum_subreg(dst_reg->var_off), umin_val, 32);
+
+	/* blow away the dst_reg umin_value/umax_value and rely on
+	 * dst_reg var_off to refine the result.
+	 */
+	dst_reg->u32_min_value = 0;
+	dst_reg->u32_max_value = U32_MAX;
+
+	__mark_reg64_unbounded(dst_reg);
+	__update_reg32_bounds(dst_reg);
+}
+
+static void scalar_min_max_arsh(struct bpf_reg_state *dst_reg,
+				struct bpf_reg_state *src_reg)
+{
+	u64 umin_val = src_reg->umin_value;
+
+	/* Upon reaching here, src_known is true and umax_val is equal
+	 * to umin_val.
+	 */
+	dst_reg->smin_value >>= umin_val;
+	dst_reg->smax_value >>= umin_val;
+
+	dst_reg->var_off = tnum_arshift(dst_reg->var_off, umin_val, 64);
+
+	/* blow away the dst_reg umin_value/umax_value and rely on
+	 * dst_reg var_off to refine the result.
+	 */
+	dst_reg->umin_value = 0;
+	dst_reg->umax_value = U64_MAX;
+
+	/* Its not easy to operate on alu32 bounds here because it depends
+	 * on bits being shifted in from upper 32-bits. Take easy way out
+	 * and mark unbounded so we can recalculate later from tnum.
+	 */
+	__mark_reg32_unbounded(dst_reg);
+	__update_reg_bounds(dst_reg);
+}
+
 /* WARNING: This function does calculations on 64-bit values, but the actual
  * execution may occur on 32-bit values. Therefore, things like bitshifts
  * need extra checks in the 32-bit case.
@@ -4850,33 +5603,47 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 	bool src_known, dst_known;
 	s64 smin_val, smax_val;
 	u64 umin_val, umax_val;
+	s32 s32_min_val, s32_max_val;
+	u32 u32_min_val, u32_max_val;
 	u64 insn_bitness = (BPF_CLASS(insn->code) == BPF_ALU64) ? 64 : 32;
 	u32 dst = insn->dst_reg;
 	int ret;
-
-	if (insn_bitness == 32) {
-		/* Relevant for 32-bit RSH: Information can propagate towards
-		 * LSB, so it isn't sufficient to only truncate the output to
-		 * 32 bits.
-		 */
-		coerce_reg_to_size(dst_reg, 4);
-		coerce_reg_to_size(&src_reg, 4);
-	}
+	bool alu32 = (BPF_CLASS(insn->code) != BPF_ALU64);
 
 	smin_val = src_reg.smin_value;
 	smax_val = src_reg.smax_value;
 	umin_val = src_reg.umin_value;
 	umax_val = src_reg.umax_value;
-	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) {
-		/* Taint dst register if offset had invalid bounds derived from
-		 * e.g. dead branches.
-		 */
-		__mark_reg_unknown(env, dst_reg);
-		return 0;
+	s32_min_val = src_reg.s32_min_value;
+	s32_max_val = src_reg.s32_max_value;
+	u32_min_val = src_reg.u32_min_value;
+	u32_max_val = src_reg.u32_max_value;
+
+	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) {
+			/* Taint dst register if offset had invalid bounds
+			 * derived from e.g. dead branches.
+			 */
+			__mark_reg_unknown(env, dst_reg);
+			return 0;
+		}
+	} 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) {
+			/* Taint dst register if offset had invalid bounds
+			 * derived from e.g. dead branches.
+			 */
+			__mark_reg_unknown(env, dst_reg);
+			return 0;
+		}
 	}
 
 	if (!src_known &&
@@ -4885,6 +5652,20 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 		return 0;
 	}
 
+	/* Calculate sign/unsigned bounds and tnum for alu32 and alu64 bit ops.
+	 * There are two classes of instructions: The first class we track both
+	 * alu32 and alu64 sign/unsigned bounds independently this provides the
+	 * greatest amount of precision when alu operations are mixed with jmp32
+	 * operations. These operations are BPF_ADD, BPF_SUB, BPF_MUL, BPF_ADD,
+	 * and BPF_OR. This is possible because these ops have fairly easy to
+	 * understand and calculate behavior in both 32-bit and 64-bit alu ops.
+	 * See alu32 verifier tests for examples. The second class of
+	 * operations, BPF_LSH, BPF_RSH, and BPF_ARSH, however are not so easy
+	 * with regards to tracking sign/unsigned bounds because the bits may
+	 * cross subreg boundaries in the alu64 case. When this happens we mark
+	 * the reg unbounded in the subreg bound space and use the resulting
+	 * tnum to calculate an approximation of the sign/unsigned bounds.
+	 */
 	switch (opcode) {
 	case BPF_ADD:
 		ret = sanitize_val_alu(env, insn);
@@ -4892,22 +5673,8 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 			verbose(env, "R%d tried to add from different pointers or scalars\n", dst);
 			return ret;
 		}
-		if (signed_add_overflows(dst_reg->smin_value, smin_val) ||
-		    signed_add_overflows(dst_reg->smax_value, smax_val)) {
-			dst_reg->smin_value = S64_MIN;
-			dst_reg->smax_value = S64_MAX;
-		} else {
-			dst_reg->smin_value += smin_val;
-			dst_reg->smax_value += smax_val;
-		}
-		if (dst_reg->umin_value + umin_val < umin_val ||
-		    dst_reg->umax_value + umax_val < umax_val) {
-			dst_reg->umin_value = 0;
-			dst_reg->umax_value = U64_MAX;
-		} else {
-			dst_reg->umin_value += umin_val;
-			dst_reg->umax_value += umax_val;
-		}
+		scalar32_min_max_add(dst_reg, &src_reg);
+		scalar_min_max_add(dst_reg, &src_reg);
 		dst_reg->var_off = tnum_add(dst_reg->var_off, src_reg.var_off);
 		break;
 	case BPF_SUB:
@@ -4916,111 +5683,24 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 			verbose(env, "R%d tried to sub from different pointers or scalars\n", dst);
 			return ret;
 		}
-		if (signed_sub_overflows(dst_reg->smin_value, smax_val) ||
-		    signed_sub_overflows(dst_reg->smax_value, smin_val)) {
-			/* Overflow possible, we know nothing */
-			dst_reg->smin_value = S64_MIN;
-			dst_reg->smax_value = S64_MAX;
-		} else {
-			dst_reg->smin_value -= smax_val;
-			dst_reg->smax_value -= smin_val;
-		}
-		if (dst_reg->umin_value < umax_val) {
-			/* Overflow possible, we know nothing */
-			dst_reg->umin_value = 0;
-			dst_reg->umax_value = U64_MAX;
-		} else {
-			/* Cannot overflow (as long as bounds are consistent) */
-			dst_reg->umin_value -= umax_val;
-			dst_reg->umax_value -= umin_val;
-		}
+		scalar32_min_max_sub(dst_reg, &src_reg);
+		scalar_min_max_sub(dst_reg, &src_reg);
 		dst_reg->var_off = tnum_sub(dst_reg->var_off, src_reg.var_off);
 		break;
 	case BPF_MUL:
 		dst_reg->var_off = tnum_mul(dst_reg->var_off, src_reg.var_off);
-		if (smin_val < 0 || dst_reg->smin_value < 0) {
-			/* Ain't nobody got time to multiply that sign */
-			__mark_reg_unbounded(dst_reg);
-			__update_reg_bounds(dst_reg);
-			break;
-		}
-		/* Both values are positive, so we can work with unsigned and
-		 * copy the result to signed (unless it exceeds S64_MAX).
-		 */
-		if (umax_val > U32_MAX || dst_reg->umax_value > U32_MAX) {
-			/* Potential overflow, we know nothing */
-			__mark_reg_unbounded(dst_reg);
-			/* (except what we can learn from the var_off) */
-			__update_reg_bounds(dst_reg);
-			break;
-		}
-		dst_reg->umin_value *= umin_val;
-		dst_reg->umax_value *= umax_val;
-		if (dst_reg->umax_value > S64_MAX) {
-			/* Overflow possible, we know nothing */
-			dst_reg->smin_value = S64_MIN;
-			dst_reg->smax_value = S64_MAX;
-		} else {
-			dst_reg->smin_value = dst_reg->umin_value;
-			dst_reg->smax_value = dst_reg->umax_value;
-		}
+		scalar32_min_max_mul(dst_reg, &src_reg);
+		scalar_min_max_mul(dst_reg, &src_reg);
 		break;
 	case BPF_AND:
-		if (src_known && dst_known) {
-			__mark_reg_known(dst_reg, dst_reg->var_off.value &
-						  src_reg.var_off.value);
-			break;
-		}
-		/* We get our minimum from the var_off, since that's inherently
-		 * bitwise.  Our maximum is the minimum of the operands' maxima.
-		 */
 		dst_reg->var_off = tnum_and(dst_reg->var_off, src_reg.var_off);
-		dst_reg->umin_value = dst_reg->var_off.value;
-		dst_reg->umax_value = min(dst_reg->umax_value, umax_val);
-		if (dst_reg->smin_value < 0 || smin_val < 0) {
-			/* Lose signed bounds when ANDing negative numbers,
-			 * ain't nobody got time for that.
-			 */
-			dst_reg->smin_value = S64_MIN;
-			dst_reg->smax_value = S64_MAX;
-		} else {
-			/* ANDing two positives gives a positive, so safe to
-			 * cast result into s64.
-			 */
-			dst_reg->smin_value = dst_reg->umin_value;
-			dst_reg->smax_value = dst_reg->umax_value;
-		}
-		/* We may learn something more from the var_off */
-		__update_reg_bounds(dst_reg);
+		scalar32_min_max_and(dst_reg, &src_reg);
+		scalar_min_max_and(dst_reg, &src_reg);
 		break;
 	case BPF_OR:
-		if (src_known && dst_known) {
-			__mark_reg_known(dst_reg, dst_reg->var_off.value |
-						  src_reg.var_off.value);
-			break;
-		}
-		/* We get our maximum from the var_off, and our minimum is the
-		 * maximum of the operands' minima
-		 */
 		dst_reg->var_off = tnum_or(dst_reg->var_off, src_reg.var_off);
-		dst_reg->umin_value = max(dst_reg->umin_value, umin_val);
-		dst_reg->umax_value = dst_reg->var_off.value |
-				      dst_reg->var_off.mask;
-		if (dst_reg->smin_value < 0 || smin_val < 0) {
-			/* Lose signed bounds when ORing negative numbers,
-			 * ain't nobody got time for that.
-			 */
-			dst_reg->smin_value = S64_MIN;
-			dst_reg->smax_value = S64_MAX;
-		} else {
-			/* ORing two positives gives a positive, so safe to
-			 * cast result into s64.
-			 */
-			dst_reg->smin_value = dst_reg->umin_value;
-			dst_reg->smax_value = dst_reg->umax_value;
-		}
-		/* We may learn something more from the var_off */
-		__update_reg_bounds(dst_reg);
+		scalar32_min_max_or(dst_reg, &src_reg);
+		scalar_min_max_or(dst_reg, &src_reg);
 		break;
 	case BPF_LSH:
 		if (umax_val >= insn_bitness) {
@@ -5030,22 +5710,10 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 			mark_reg_unknown(env, regs, insn->dst_reg);
 			break;
 		}
-		/* We lose all sign bit information (except what we can pick
-		 * up from var_off)
-		 */
-		dst_reg->smin_value = S64_MIN;
-		dst_reg->smax_value = S64_MAX;
-		/* If we might shift our top bit out, then we know nothing */
-		if (dst_reg->umax_value > 1ULL << (63 - umax_val)) {
-			dst_reg->umin_value = 0;
-			dst_reg->umax_value = U64_MAX;
-		} else {
-			dst_reg->umin_value <<= umin_val;
-			dst_reg->umax_value <<= umax_val;
-		}
-		dst_reg->var_off = tnum_lshift(dst_reg->var_off, umin_val);
-		/* We may learn something more from the var_off */
-		__update_reg_bounds(dst_reg);
+		if (alu32)
+			scalar32_min_max_lsh(dst_reg, &src_reg);
+		else
+			scalar_min_max_lsh(dst_reg, &src_reg);
 		break;
 	case BPF_RSH:
 		if (umax_val >= insn_bitness) {
@@ -5055,27 +5723,10 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 			mark_reg_unknown(env, regs, insn->dst_reg);
 			break;
 		}
-		/* BPF_RSH is an unsigned shift.  If the value in dst_reg might
-		 * be negative, then either:
-		 * 1) src_reg might be zero, so the sign bit of the result is
-		 *    unknown, so we lose our signed bounds
-		 * 2) it's known negative, thus the unsigned bounds capture the
-		 *    signed bounds
-		 * 3) the signed bounds cross zero, so they tell us nothing
-		 *    about the result
-		 * If the value in dst_reg is known nonnegative, then again the
-		 * unsigned bounts capture the signed bounds.
-		 * Thus, in all cases it suffices to blow away our signed bounds
-		 * and rely on inferring new ones from the unsigned bounds and
-		 * var_off of the result.
-		 */
-		dst_reg->smin_value = S64_MIN;
-		dst_reg->smax_value = S64_MAX;
-		dst_reg->var_off = tnum_rshift(dst_reg->var_off, umin_val);
-		dst_reg->umin_value >>= umax_val;
-		dst_reg->umax_value >>= umin_val;
-		/* We may learn something more from the var_off */
-		__update_reg_bounds(dst_reg);
+		if (alu32)
+			scalar32_min_max_rsh(dst_reg, &src_reg);
+		else
+			scalar_min_max_rsh(dst_reg, &src_reg);
 		break;
 	case BPF_ARSH:
 		if (umax_val >= insn_bitness) {
@@ -5085,38 +5736,21 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
 			mark_reg_unknown(env, regs, insn->dst_reg);
 			break;
 		}
-
-		/* Upon reaching here, src_known is true and
-		 * umax_val is equal to umin_val.
-		 */
-		if (insn_bitness == 32) {
-			dst_reg->smin_value = (u32)(((s32)dst_reg->smin_value) >> umin_val);
-			dst_reg->smax_value = (u32)(((s32)dst_reg->smax_value) >> umin_val);
-		} else {
-			dst_reg->smin_value >>= umin_val;
-			dst_reg->smax_value >>= umin_val;
-		}
-
-		dst_reg->var_off = tnum_arshift(dst_reg->var_off, umin_val,
-						insn_bitness);
-
-		/* blow away the dst_reg umin_value/umax_value and rely on
-		 * dst_reg var_off to refine the result.
-		 */
-		dst_reg->umin_value = 0;
-		dst_reg->umax_value = U64_MAX;
-		__update_reg_bounds(dst_reg);
+		if (alu32)
+			scalar32_min_max_arsh(dst_reg, &src_reg);
+		else
+			scalar_min_max_arsh(dst_reg, &src_reg);
 		break;
 	default:
 		mark_reg_unknown(env, regs, insn->dst_reg);
 		break;
 	}
 
-	if (BPF_CLASS(insn->code) != BPF_ALU64) {
-		/* 32-bit ALU ops are (32,32)->32 */
-		coerce_reg_to_size(dst_reg, 4);
-	}
+	/* ALU32 ops are zero extended into 64bit register */
+	if (alu32)
+		zext_32_to_64(dst_reg);
 
+	__update_reg_bounds(dst_reg);
 	__reg_deduce_bounds(dst_reg);
 	__reg_bound_offset(dst_reg);
 	return 0;
@@ -5290,7 +5924,7 @@ static int check_alu_op(struct bpf_verifier_env *env, struct bpf_insn *insn)
 					mark_reg_unknown(env, regs,
 							 insn->dst_reg);
 				}
-				coerce_reg_to_size(dst_reg, 4);
+				zext_32_to_64(dst_reg);
 			}
 		} else {
 			/* case: R = imm
@@ -5460,55 +6094,83 @@ static void find_good_pkt_pointers(struct bpf_verifier_state *vstate,
 					 new_range);
 }
 
-/* compute branch direction of the expression "if (reg opcode val) goto target;"
- * and return:
- *  1 - branch will be taken and "goto target" will be executed
- *  0 - branch will not be taken and fall-through to next insn
- * -1 - unknown. Example: "if (reg < 5)" is unknown when register value range [0,10]
- */
-static int is_branch_taken(struct bpf_reg_state *reg, u64 val, u8 opcode,
-			   bool is_jmp32)
+static int is_branch32_taken(struct bpf_reg_state *reg, u32 val, u8 opcode)
 {
-	struct bpf_reg_state reg_lo;
-	s64 sval;
+	struct tnum subreg = tnum_subreg(reg->var_off);
+	s32 sval = (s32)val;
 
-	if (__is_pointer_value(false, reg))
-		return -1;
+	switch (opcode) {
+	case BPF_JEQ:
+		if (tnum_is_const(subreg))
+			return !!tnum_equals_const(subreg, val);
+		break;
+	case BPF_JNE:
+		if (tnum_is_const(subreg))
+			return !tnum_equals_const(subreg, val);
+		break;
+	case BPF_JSET:
+		if ((~subreg.mask & subreg.value) & val)
+			return 1;
+		if (!((subreg.mask | subreg.value) & val))
+			return 0;
+		break;
+	case BPF_JGT:
+		if (reg->u32_min_value > val)
+			return 1;
+		else if (reg->u32_max_value <= val)
+			return 0;
+		break;
+	case BPF_JSGT:
+		if (reg->s32_min_value > sval)
+			return 1;
+		else if (reg->s32_max_value < sval)
+			return 0;
+		break;
+	case BPF_JLT:
+		if (reg->u32_max_value < val)
+			return 1;
+		else if (reg->u32_min_value >= val)
+			return 0;
+		break;
+	case BPF_JSLT:
+		if (reg->s32_max_value < sval)
+			return 1;
+		else if (reg->s32_min_value >= sval)
+			return 0;
+		break;
+	case BPF_JGE:
+		if (reg->u32_min_value >= val)
+			return 1;
+		else if (reg->u32_max_value < val)
+			return 0;
+		break;
+	case BPF_JSGE:
+		if (reg->s32_min_value >= sval)
+			return 1;
+		else if (reg->s32_max_value < sval)
+			return 0;
+		break;
+	case BPF_JLE:
+		if (reg->u32_max_value <= val)
+			return 1;
+		else if (reg->u32_min_value > val)
+			return 0;
+		break;
+	case BPF_JSLE:
+		if (reg->s32_max_value <= sval)
+			return 1;
+		else if (reg->s32_min_value > sval)
+			return 0;
+		break;
+	}
 
-	if (is_jmp32) {
-		reg_lo = *reg;
-		reg = &reg_lo;
-		/* For JMP32, only low 32 bits are compared, coerce_reg_to_size
-		 * could truncate high bits and update umin/umax according to
-		 * information of low bits.
-		 */
-		coerce_reg_to_size(reg, 4);
-		/* smin/smax need special handling. For example, after coerce,
-		 * if smin_value is 0x00000000ffffffffLL, the value is -1 when
-		 * used as operand to JMP32. It is a negative number from s32's
-		 * point of view, while it is a positive number when seen as
-		 * s64. The smin/smax are kept as s64, therefore, when used with
-		 * JMP32, they need to be transformed into s32, then sign
-		 * extended back to s64.
-		 *
-		 * Also, smin/smax were copied from umin/umax. If umin/umax has
-		 * different sign bit, then min/max relationship doesn't
-		 * maintain after casting into s32, for this case, set smin/smax
-		 * to safest range.
-		 */
-		if ((reg->umax_value ^ reg->umin_value) &
-		    (1ULL << 31)) {
-			reg->smin_value = S32_MIN;
-			reg->smax_value = S32_MAX;
-		}
-		reg->smin_value = (s64)(s32)reg->smin_value;
-		reg->smax_value = (s64)(s32)reg->smax_value;
+	return -1;
+}
 
-		val = (u32)val;
-		sval = (s64)(s32)val;
-	} else {
-		sval = (s64)val;
-	}
+
+static int is_branch64_taken(struct bpf_reg_state *reg, u64 val, u8 opcode)
+{
+	s64 sval = (s64)val;
 
 	switch (opcode) {
 	case BPF_JEQ:
@@ -5578,27 +6240,22 @@ static int is_branch_taken(struct bpf_reg_state *reg, u64 val, u8 opcode,
 	return -1;
 }
 
-/* Generate min value of the high 32-bit from TNUM info. */
-static u64 gen_hi_min(struct tnum var)
-{
-	return var.value & ~0xffffffffULL;
-}
-
-/* Generate max value of the high 32-bit from TNUM info. */
-static u64 gen_hi_max(struct tnum var)
-{
-	return (var.value | var.mask) & ~0xffffffffULL;
-}
-
-/* Return true if VAL is compared with a s64 sign extended from s32, and they
- * are with the same signedness.
+/* compute branch direction of the expression "if (reg opcode val) goto target;"
+ * and return:
+ *  1 - branch will be taken and "goto target" will be executed
+ *  0 - branch will not be taken and fall-through to next insn
+ * -1 - unknown. Example: "if (reg < 5)" is unknown when register value
+ *      range [0,10]
  */
-static bool cmp_val_with_extended_s64(s64 sval, struct bpf_reg_state *reg)
+static int is_branch_taken(struct bpf_reg_state *reg, u64 val, u8 opcode,
+			   bool is_jmp32)
 {
-	return ((s32)sval >= 0 &&
-		reg->smin_value >= 0 && reg->smax_value <= S32_MAX) ||
-	       ((s32)sval < 0 &&
-		reg->smax_value <= 0 && reg->smin_value >= S32_MIN);
+	if (__is_pointer_value(false, reg))
+		return -1;
+
+	if (is_jmp32)
+		return is_branch32_taken(reg, val, opcode);
+	return is_branch64_taken(reg, val, opcode);
 }
 
 /* Adjusts the register min/max values in the case that the dst_reg is the
@@ -5607,10 +6264,16 @@ static bool cmp_val_with_extended_s64(s64 sval, struct bpf_reg_state *reg)
  * In JEQ/JNE cases we also adjust the var_off values.
  */
 static void reg_set_min_max(struct bpf_reg_state *true_reg,
-			    struct bpf_reg_state *false_reg, u64 val,
+			    struct bpf_reg_state *false_reg,
+			    u64 val, u32 val32,
 			    u8 opcode, bool is_jmp32)
 {
-	s64 sval;
+	struct tnum false_32off = tnum_subreg(false_reg->var_off);
+	struct tnum false_64off = false_reg->var_off;
+	struct tnum true_32off = tnum_subreg(true_reg->var_off);
+	struct tnum true_64off = true_reg->var_off;
+	s64 sval = (s64)val;
+	s32 sval32 = (s32)val32;
 
 	/* If the dst_reg is a pointer, we can't learn anything about its
 	 * variable offset from the compare (unless src_reg were a pointer into
@@ -5621,9 +6284,6 @@ static void reg_set_min_max(struct bpf_reg_state *true_reg,
 	if (__is_pointer_value(false, false_reg))
 		return;
 
-	val = is_jmp32 ? (u32)val : val;
-	sval = is_jmp32 ? (s64)(s32)val : (s64)val;
-
 	switch (opcode) {
 	case BPF_JEQ:
 	case BPF_JNE:
@@ -5635,211 +6295,150 @@ static void reg_set_min_max(struct bpf_reg_state *true_reg,
 		 * if it is true we know the value for sure. Likewise for
 		 * BPF_JNE.
 		 */
-		if (is_jmp32) {
-			u64 old_v = reg->var_off.value;
-			u64 hi_mask = ~0xffffffffULL;
-
-			reg->var_off.value = (old_v & hi_mask) | val;
-			reg->var_off.mask &= hi_mask;
-		} else {
+		if (is_jmp32)
+			__mark_reg32_known(reg, val32);
+		else
 			__mark_reg_known(reg, val);
-		}
 		break;
 	}
 	case BPF_JSET:
-		false_reg->var_off = tnum_and(false_reg->var_off,
-					      tnum_const(~val));
-		if (is_power_of_2(val))
-			true_reg->var_off = tnum_or(true_reg->var_off,
-						    tnum_const(val));
+		if (is_jmp32) {
+			false_32off = tnum_and(false_32off, tnum_const(~val32));
+			if (is_power_of_2(val32))
+				true_32off = tnum_or(true_32off,
+						     tnum_const(val32));
+		} else {
+			false_64off = tnum_and(false_64off, tnum_const(~val));
+			if (is_power_of_2(val))
+				true_64off = tnum_or(true_64off,
+						     tnum_const(val));
+		}
 		break;
 	case BPF_JGE:
 	case BPF_JGT:
 	{
-		u64 false_umax = opcode == BPF_JGT ? val    : val - 1;
-		u64 true_umin = opcode == BPF_JGT ? val + 1 : val;
-
 		if (is_jmp32) {
-			false_umax += gen_hi_max(false_reg->var_off);
-			true_umin += gen_hi_min(true_reg->var_off);
+			u32 false_umax = opcode == BPF_JGT ? val32  : val32 - 1;
+			u32 true_umin = opcode == BPF_JGT ? val32 + 1 : val32;
+
+			false_reg->u32_max_value = min(false_reg->u32_max_value,
+						       false_umax);
+			true_reg->u32_min_value = max(true_reg->u32_min_value,
+						      true_umin);
+		} else {
+			u64 false_umax = opcode == BPF_JGT ? val    : val - 1;
+			u64 true_umin = opcode == BPF_JGT ? val + 1 : val;
+
+			false_reg->umax_value = min(false_reg->umax_value, false_umax);
+			true_reg->umin_value = max(true_reg->umin_value, true_umin);
 		}
-		false_reg->umax_value = min(false_reg->umax_value, false_umax);
-		true_reg->umin_value = max(true_reg->umin_value, true_umin);
 		break;
 	}
 	case BPF_JSGE:
 	case BPF_JSGT:
 	{
-		s64 false_smax = opcode == BPF_JSGT ? sval    : sval - 1;
-		s64 true_smin = opcode == BPF_JSGT ? sval + 1 : sval;
+		if (is_jmp32) {
+			s32 false_smax = opcode == BPF_JSGT ? sval32    : sval32 - 1;
+			s32 true_smin = opcode == BPF_JSGT ? sval32 + 1 : sval32;
 
-		/* If the full s64 was not sign-extended from s32 then don't
-		 * deduct further info.
-		 */
-		if (is_jmp32 && !cmp_val_with_extended_s64(sval, false_reg))
-			break;
-		false_reg->smax_value = min(false_reg->smax_value, false_smax);
-		true_reg->smin_value = max(true_reg->smin_value, true_smin);
+			false_reg->s32_max_value = min(false_reg->s32_max_value, false_smax);
+			true_reg->s32_min_value = max(true_reg->s32_min_value, true_smin);
+		} else {
+			s64 false_smax = opcode == BPF_JSGT ? sval    : sval - 1;
+			s64 true_smin = opcode == BPF_JSGT ? sval + 1 : sval;
+
+			false_reg->smax_value = min(false_reg->smax_value, false_smax);
+			true_reg->smin_value = max(true_reg->smin_value, true_smin);
+		}
 		break;
 	}
 	case BPF_JLE:
 	case BPF_JLT:
 	{
-		u64 false_umin = opcode == BPF_JLT ? val    : val + 1;
-		u64 true_umax = opcode == BPF_JLT ? val - 1 : val;
-
 		if (is_jmp32) {
-			false_umin += gen_hi_min(false_reg->var_off);
-			true_umax += gen_hi_max(true_reg->var_off);
+			u32 false_umin = opcode == BPF_JLT ? val32  : val32 + 1;
+			u32 true_umax = opcode == BPF_JLT ? val32 - 1 : val32;
+
+			false_reg->u32_min_value = max(false_reg->u32_min_value,
+						       false_umin);
+			true_reg->u32_max_value = min(true_reg->u32_max_value,
+						      true_umax);
+		} else {
+			u64 false_umin = opcode == BPF_JLT ? val    : val + 1;
+			u64 true_umax = opcode == BPF_JLT ? val - 1 : val;
+
+			false_reg->umin_value = max(false_reg->umin_value, false_umin);
+			true_reg->umax_value = min(true_reg->umax_value, true_umax);
 		}
-		false_reg->umin_value = max(false_reg->umin_value, false_umin);
-		true_reg->umax_value = min(true_reg->umax_value, true_umax);
 		break;
 	}
 	case BPF_JSLE:
 	case BPF_JSLT:
 	{
-		s64 false_smin = opcode == BPF_JSLT ? sval    : sval + 1;
-		s64 true_smax = opcode == BPF_JSLT ? sval - 1 : sval;
+		if (is_jmp32) {
+			s32 false_smin = opcode == BPF_JSLT ? sval32    : sval32 + 1;
+			s32 true_smax = opcode == BPF_JSLT ? sval32 - 1 : sval32;
 
-		if (is_jmp32 && !cmp_val_with_extended_s64(sval, false_reg))
-			break;
-		false_reg->smin_value = max(false_reg->smin_value, false_smin);
-		true_reg->smax_value = min(true_reg->smax_value, true_smax);
+			false_reg->s32_min_value = max(false_reg->s32_min_value, false_smin);
+			true_reg->s32_max_value = min(true_reg->s32_max_value, true_smax);
+		} else {
+			s64 false_smin = opcode == BPF_JSLT ? sval    : sval + 1;
+			s64 true_smax = opcode == BPF_JSLT ? sval - 1 : sval;
+
+			false_reg->smin_value = max(false_reg->smin_value, false_smin);
+			true_reg->smax_value = min(true_reg->smax_value, true_smax);
+		}
 		break;
 	}
 	default:
-		break;
+		return;
 	}
 
-	__reg_deduce_bounds(false_reg);
-	__reg_deduce_bounds(true_reg);
-	/* We might have learned some bits from the bounds. */
-	__reg_bound_offset(false_reg);
-	__reg_bound_offset(true_reg);
 	if (is_jmp32) {
-		__reg_bound_offset32(false_reg);
-		__reg_bound_offset32(true_reg);
+		false_reg->var_off = tnum_or(tnum_clear_subreg(false_64off),
+					     tnum_subreg(false_32off));
+		true_reg->var_off = tnum_or(tnum_clear_subreg(true_64off),
+					    tnum_subreg(true_32off));
+		__reg_combine_32_into_64(false_reg);
+		__reg_combine_32_into_64(true_reg);
+	} else {
+		false_reg->var_off = false_64off;
+		true_reg->var_off = true_64off;
+		__reg_combine_64_into_32(false_reg);
+		__reg_combine_64_into_32(true_reg);
 	}
-	/* Intersecting with the old var_off might have improved our bounds
-	 * slightly.  e.g. if umax was 0x7f...f and var_off was (0; 0xf...fc),
-	 * then new var_off is (0; 0x7f...fc) which improves our umax.
-	 */
-	__update_reg_bounds(false_reg);
-	__update_reg_bounds(true_reg);
 }
 
 /* Same as above, but for the case that dst_reg holds a constant and src_reg is
  * the variable reg.
  */
 static void reg_set_min_max_inv(struct bpf_reg_state *true_reg,
-				struct bpf_reg_state *false_reg, u64 val,
+				struct bpf_reg_state *false_reg,
+				u64 val, u32 val32,
 				u8 opcode, bool is_jmp32)
 {
-	s64 sval;
-
-	if (__is_pointer_value(false, false_reg))
-		return;
-
-	val = is_jmp32 ? (u32)val : val;
-	sval = is_jmp32 ? (s64)(s32)val : (s64)val;
-
-	switch (opcode) {
-	case BPF_JEQ:
-	case BPF_JNE:
-	{
-		struct bpf_reg_state *reg =
-			opcode == BPF_JEQ ? true_reg : false_reg;
-
-		if (is_jmp32) {
-			u64 old_v = reg->var_off.value;
-			u64 hi_mask = ~0xffffffffULL;
-
-			reg->var_off.value = (old_v & hi_mask) | val;
-			reg->var_off.mask &= hi_mask;
-		} else {
-			__mark_reg_known(reg, val);
-		}
-		break;
-	}
-	case BPF_JSET:
-		false_reg->var_off = tnum_and(false_reg->var_off,
-					      tnum_const(~val));
-		if (is_power_of_2(val))
-			true_reg->var_off = tnum_or(true_reg->var_off,
-						    tnum_const(val));
-		break;
-	case BPF_JGE:
-	case BPF_JGT:
-	{
-		u64 false_umin = opcode == BPF_JGT ? val    : val + 1;
-		u64 true_umax = opcode == BPF_JGT ? val - 1 : val;
-
-		if (is_jmp32) {
-			false_umin += gen_hi_min(false_reg->var_off);
-			true_umax += gen_hi_max(true_reg->var_off);
-		}
-		false_reg->umin_value = max(false_reg->umin_value, false_umin);
-		true_reg->umax_value = min(true_reg->umax_value, true_umax);
-		break;
-	}
-	case BPF_JSGE:
-	case BPF_JSGT:
-	{
-		s64 false_smin = opcode == BPF_JSGT ? sval    : sval + 1;
-		s64 true_smax = opcode == BPF_JSGT ? sval - 1 : sval;
-
-		if (is_jmp32 && !cmp_val_with_extended_s64(sval, false_reg))
-			break;
-		false_reg->smin_value = max(false_reg->smin_value, false_smin);
-		true_reg->smax_value = min(true_reg->smax_value, true_smax);
-		break;
-	}
-	case BPF_JLE:
-	case BPF_JLT:
-	{
-		u64 false_umax = opcode == BPF_JLT ? val    : val - 1;
-		u64 true_umin = opcode == BPF_JLT ? val + 1 : val;
-
-		if (is_jmp32) {
-			false_umax += gen_hi_max(false_reg->var_off);
-			true_umin += gen_hi_min(true_reg->var_off);
-		}
-		false_reg->umax_value = min(false_reg->umax_value, false_umax);
-		true_reg->umin_value = max(true_reg->umin_value, true_umin);
-		break;
-	}
-	case BPF_JSLE:
-	case BPF_JSLT:
-	{
-		s64 false_smax = opcode == BPF_JSLT ? sval    : sval - 1;
-		s64 true_smin = opcode == BPF_JSLT ? sval + 1 : sval;
-
-		if (is_jmp32 && !cmp_val_with_extended_s64(sval, false_reg))
-			break;
-		false_reg->smax_value = min(false_reg->smax_value, false_smax);
-		true_reg->smin_value = max(true_reg->smin_value, true_smin);
-		break;
-	}
-	default:
-		break;
-	}
-
-	__reg_deduce_bounds(false_reg);
-	__reg_deduce_bounds(true_reg);
-	/* We might have learned some bits from the bounds. */
-	__reg_bound_offset(false_reg);
-	__reg_bound_offset(true_reg);
-	if (is_jmp32) {
-		__reg_bound_offset32(false_reg);
-		__reg_bound_offset32(true_reg);
-	}
-	/* Intersecting with the old var_off might have improved our bounds
-	 * slightly.  e.g. if umax was 0x7f...f and var_off was (0; 0xf...fc),
-	 * then new var_off is (0; 0x7f...fc) which improves our umax.
+	/* How can we transform "a <op> b" into "b <op> a"? */
+	static const u8 opcode_flip[16] = {
+		/* these stay the same */
+		[BPF_JEQ  >> 4] = BPF_JEQ,
+		[BPF_JNE  >> 4] = BPF_JNE,
+		[BPF_JSET >> 4] = BPF_JSET,
+		/* these swap "lesser" and "greater" (L and G in the opcodes) */
+		[BPF_JGE  >> 4] = BPF_JLE,
+		[BPF_JGT  >> 4] = BPF_JLT,
+		[BPF_JLE  >> 4] = BPF_JGE,
+		[BPF_JLT  >> 4] = BPF_JGT,
+		[BPF_JSGE >> 4] = BPF_JSLE,
+		[BPF_JSGT >> 4] = BPF_JSLT,
+		[BPF_JSLE >> 4] = BPF_JSGE,
+		[BPF_JSLT >> 4] = BPF_JSGT
+	};
+	opcode = opcode_flip[opcode >> 4];
+	/* This uses zero as "not present in table"; luckily the zero opcode,
+	 * BPF_JA, can't get here.
 	 */
-	__update_reg_bounds(false_reg);
-	__update_reg_bounds(true_reg);
+	if (opcode)
+		reg_set_min_max(true_reg, false_reg, val, val32, opcode, is_jmp32);
 }
 
 /* Regs are known to be equal, so intersect their min/max/var_off */
@@ -6128,13 +6727,22 @@ static int check_cond_jmp_op(struct bpf_verifier_env *env,
 	dst_reg = &regs[insn->dst_reg];
 	is_jmp32 = BPF_CLASS(insn->code) == BPF_JMP32;
 
-	if (BPF_SRC(insn->code) == BPF_K)
-		pred = is_branch_taken(dst_reg, insn->imm,
-				       opcode, is_jmp32);
-	else if (src_reg->type == SCALAR_VALUE &&
-		 tnum_is_const(src_reg->var_off))
-		pred = is_branch_taken(dst_reg, src_reg->var_off.value,
-				       opcode, is_jmp32);
+	if (BPF_SRC(insn->code) == BPF_K) {
+		pred = is_branch_taken(dst_reg, insn->imm, opcode, is_jmp32);
+	} else if (src_reg->type == SCALAR_VALUE &&
+		   is_jmp32 && tnum_is_const(tnum_subreg(src_reg->var_off))) {
+		pred = is_branch_taken(dst_reg,
+				       tnum_subreg(src_reg->var_off).value,
+				       opcode,
+				       is_jmp32);
+	} else if (src_reg->type == SCALAR_VALUE &&
+		   !is_jmp32 && tnum_is_const(src_reg->var_off)) {
+		pred = is_branch_taken(dst_reg,
+				       src_reg->var_off.value,
+				       opcode,
+				       is_jmp32);
+	}
+
 	if (pred >= 0) {
 		err = mark_chain_precision(env, insn->dst_reg);
 		if (BPF_SRC(insn->code) == BPF_X && !err)
@@ -6168,32 +6776,24 @@ static int check_cond_jmp_op(struct bpf_verifier_env *env,
 	 */
 	if (BPF_SRC(insn->code) == BPF_X) {
 		struct bpf_reg_state *src_reg = &regs[insn->src_reg];
-		struct bpf_reg_state lo_reg0 = *dst_reg;
-		struct bpf_reg_state lo_reg1 = *src_reg;
-		struct bpf_reg_state *src_lo, *dst_lo;
-
-		dst_lo = &lo_reg0;
-		src_lo = &lo_reg1;
-		coerce_reg_to_size(dst_lo, 4);
-		coerce_reg_to_size(src_lo, 4);
 
 		if (dst_reg->type == SCALAR_VALUE &&
 		    src_reg->type == SCALAR_VALUE) {
 			if (tnum_is_const(src_reg->var_off) ||
-			    (is_jmp32 && tnum_is_const(src_lo->var_off)))
+			    (is_jmp32 &&
+			     tnum_is_const(tnum_subreg(src_reg->var_off))))
 				reg_set_min_max(&other_branch_regs[insn->dst_reg],
 						dst_reg,
-						is_jmp32
-						? src_lo->var_off.value
-						: src_reg->var_off.value,
+						src_reg->var_off.value,
+						tnum_subreg(src_reg->var_off).value,
 						opcode, is_jmp32);
 			else if (tnum_is_const(dst_reg->var_off) ||
-				 (is_jmp32 && tnum_is_const(dst_lo->var_off)))
+				 (is_jmp32 &&
+				  tnum_is_const(tnum_subreg(dst_reg->var_off))))
 				reg_set_min_max_inv(&other_branch_regs[insn->src_reg],
 						    src_reg,
-						    is_jmp32
-						    ? dst_lo->var_off.value
-						    : dst_reg->var_off.value,
+						    dst_reg->var_off.value,
+						    tnum_subreg(dst_reg->var_off).value,
 						    opcode, is_jmp32);
 			else if (!is_jmp32 &&
 				 (opcode == BPF_JEQ || opcode == BPF_JNE))
@@ -6204,7 +6804,8 @@ static int check_cond_jmp_op(struct bpf_verifier_env *env,
 		}
 	} else if (dst_reg->type == SCALAR_VALUE) {
 		reg_set_min_max(&other_branch_regs[insn->dst_reg],
-					dst_reg, insn->imm, opcode, is_jmp32);
+					dst_reg, insn->imm, (u32)insn->imm,
+					opcode, is_jmp32);
 	}
 
 	/* detect if R == 0 where R is returned from bpf_map_lookup_elem().
@@ -6405,8 +7006,9 @@ static int check_return_code(struct bpf_verifier_env *env)
 	struct tnum range = tnum_range(0, 1);
 	int err;
 
-	/* The struct_ops func-ptr's return type could be "void" */
-	if (env->prog->type == BPF_PROG_TYPE_STRUCT_OPS &&
+	/* LSM and struct_ops func-ptr's return type could be "void" */
+	if ((env->prog->type == BPF_PROG_TYPE_STRUCT_OPS ||
+	     env->prog->type == BPF_PROG_TYPE_LSM) &&
 	    !prog->aux->attach_func_proto->type)
 		return 0;
 
@@ -8139,26 +8741,48 @@ static bool is_tracing_prog_type(enum bpf_prog_type type)
 	}
 }
 
+static bool is_preallocated_map(struct bpf_map *map)
+{
+	if (!check_map_prealloc(map))
+		return false;
+	if (map->inner_map_meta && !check_map_prealloc(map->inner_map_meta))
+		return false;
+	return true;
+}
+
 static int check_map_prog_compatibility(struct bpf_verifier_env *env,
 					struct bpf_map *map,
 					struct bpf_prog *prog)
 
 {
-	/* Make sure that BPF_PROG_TYPE_PERF_EVENT programs only use
-	 * preallocated hash maps, since doing memory allocation
-	 * in overflow_handler can crash depending on where nmi got
-	 * triggered.
+	/*
+	 * Validate that trace type programs use preallocated hash maps.
+	 *
+	 * For programs attached to PERF events this is mandatory as the
+	 * perf NMI can hit any arbitrary code sequence.
+	 *
+	 * All other trace types using preallocated hash maps are unsafe as
+	 * well because tracepoint or kprobes can be inside locked regions
+	 * of the memory allocator or at a place where a recursion into the
+	 * memory allocator would see inconsistent state.
+	 *
+	 * On RT enabled kernels run-time allocation of all trace type
+	 * programs is strictly prohibited due to lock type constraints. On
+	 * !RT kernels it is allowed for backwards compatibility reasons for
+	 * now, but warnings are emitted so developers are made aware of
+	 * the unsafety and can fix their programs before this is enforced.
 	 */
-	if (prog->type == BPF_PROG_TYPE_PERF_EVENT) {
-		if (!check_map_prealloc(map)) {
+	if (is_tracing_prog_type(prog->type) && !is_preallocated_map(map)) {
+		if (prog->type == BPF_PROG_TYPE_PERF_EVENT) {
 			verbose(env, "perf_event programs can only use preallocated hash map\n");
 			return -EINVAL;
 		}
-		if (map->inner_map_meta &&
-		    !check_map_prealloc(map->inner_map_meta)) {
-			verbose(env, "perf_event programs can only use preallocated inner hash map\n");
+		if (IS_ENABLED(CONFIG_PREEMPT_RT)) {
+			verbose(env, "trace type programs can only use preallocated hash map\n");
 			return -EINVAL;
 		}
+		WARN_ONCE(1, "trace type BPF program uses run-time allocation\n");
+		verbose(env, "trace type programs with run-time allocated hash maps are unsafe. Switch to preallocated hash maps.\n");
 	}
 
 	if ((is_tracing_prog_type(prog->type) ||
@@ -9774,6 +10398,26 @@ static int check_struct_ops_btf_id(struct bpf_verifier_env *env)
 
 	return 0;
 }
+#define SECURITY_PREFIX "security_"
+
+static int check_attach_modify_return(struct bpf_verifier_env *env)
+{
+	struct bpf_prog *prog = env->prog;
+	unsigned long addr = (unsigned long) prog->aux->trampoline->func.addr;
+
+	/* This is expected to be cleaned up in the future with the KRSI effort
+	 * introducing the LSM_HOOK macro for cleaning up lsm_hooks.h.
+	 */
+	if (within_error_injection_list(addr) ||
+	    !strncmp(SECURITY_PREFIX, prog->aux->attach_func_name,
+		     sizeof(SECURITY_PREFIX) - 1))
+		return 0;
+
+	verbose(env, "fmod_ret attach_btf_id %u (%s) is not modifiable\n",
+		prog->aux->attach_btf_id, prog->aux->attach_func_name);
+
+	return -EINVAL;
+}
 
 static int check_attach_btf_id(struct bpf_verifier_env *env)
 {
@@ -9794,7 +10438,9 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 	if (prog->type == BPF_PROG_TYPE_STRUCT_OPS)
 		return check_struct_ops_btf_id(env);
 
-	if (prog->type != BPF_PROG_TYPE_TRACING && !prog_extension)
+	if (prog->type != BPF_PROG_TYPE_TRACING &&
+	    prog->type != BPF_PROG_TYPE_LSM &&
+	    !prog_extension)
 		return 0;
 
 	if (!btf_id) {
@@ -9924,8 +10570,17 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 		if (!prog_extension)
 			return -EINVAL;
 		/* fallthrough */
+	case BPF_MODIFY_RETURN:
+	case BPF_LSM_MAC:
 	case BPF_TRACE_FENTRY:
 	case BPF_TRACE_FEXIT:
+		prog->aux->attach_func_name = tname;
+		if (prog->type == BPF_PROG_TYPE_LSM) {
+			ret = bpf_lsm_verify_prog(&env->log, prog);
+			if (ret < 0)
+				return ret;
+		}
+
 		if (!btf_type_is_func(t)) {
 			verbose(env, "attach_btf_id %u is not a function\n",
 				btf_id);
@@ -9940,7 +10595,6 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 		tr = bpf_trampoline_lookup(key);
 		if (!tr)
 			return -ENOMEM;
-		prog->aux->attach_func_name = tname;
 		/* t is either vmlinux type or another program's type */
 		prog->aux->attach_func_proto = t;
 		mutex_lock(&tr->mutex);
@@ -9973,6 +10627,9 @@ static int check_attach_btf_id(struct bpf_verifier_env *env)
 		}
 		tr->func.addr = (void *)addr;
 		prog->aux->trampoline = tr;
+
+		if (prog->expected_attach_type == BPF_MODIFY_RETURN)
+			ret = check_attach_modify_return(env);
 out:
 		mutex_unlock(&tr->mutex);
 		if (ret)