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

Daniel Borkmann says:

====================
pull-request: bpf-next 2019-12-27

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

We've added 127 non-merge commits during the last 17 day(s) which contain
a total of 110 files changed, 6901 insertions(+), 2721 deletions(-).

There are three merge conflicts. Conflicts and resolution looks as follows:

1) Merge conflict in net/bpf/test_run.c:

There was a tree-wide cleanup c593642c8be0 ("treewide: Use sizeof_field() macro")
which gets in the way with b590cb5f802d ("bpf: Switch to offsetofend in
BPF_PROG_TEST_RUN"):

  <<<<<<< HEAD
          if (!range_is_zero(__skb, offsetof(struct __sk_buff, priority) +
                             sizeof_field(struct __sk_buff, priority),
  =======
          if (!range_is_zero(__skb, offsetofend(struct __sk_buff, priority),
  >>>>>>> 7c8dce4b166113743adad131b5a24c4acc12f92c

There are a few occasions that look similar to this. Always take the chunk with
offsetofend(). Note that there is one where the fields differ in here:

  <<<<<<< HEAD
          if (!range_is_zero(__skb, offsetof(struct __sk_buff, tstamp) +
                             sizeof_field(struct __sk_buff, tstamp),
  =======
          if (!range_is_zero(__skb, offsetofend(struct __sk_buff, gso_segs),
  >>>>>>> 7c8dce4b166113743adad131b5a24c4acc12f92c

Just take the one with offsetofend() /and/ gso_segs. Latter is correct due to
850a88cc4096 ("bpf: Expose __sk_buff wire_len/gso_segs to BPF_PROG_TEST_RUN").

2) Merge conflict in arch/riscv/net/bpf_jit_comp.c:

(I'm keeping Bjorn in Cc here for a double-check in case I got it wrong.)

  <<<<<<< HEAD
          if (is_13b_check(off, insn))
                  return -1;
          emit(rv_blt(tcc, RV_REG_ZERO, off >> 1), ctx);
  =======
          emit_branch(BPF_JSLT, RV_REG_T1, RV_REG_ZERO, off, ctx);
  >>>>>>> 7c8dce4b166113743adad131b5a24c4acc12f92c

Result should look like:

          emit_branch(BPF_JSLT, tcc, RV_REG_ZERO, off, ctx);

3) Merge conflict in arch/riscv/include/asm/pgtable.h:

  <<<<<<< HEAD
  =======
  #define VMALLOC_SIZE     (KERN_VIRT_SIZE >> 1)
  #define VMALLOC_END      (PAGE_OFFSET - 1)
  #define VMALLOC_START    (PAGE_OFFSET - VMALLOC_SIZE)

  #define BPF_JIT_REGION_SIZE     (SZ_128M)
  #define BPF_JIT_REGION_START    (PAGE_OFFSET - BPF_JIT_REGION_SIZE)
  #define BPF_JIT_REGION_END      (VMALLOC_END)

  /*
   * Roughly size the vmemmap space to be large enough to fit enough
   * struct pages to map half the virtual address space. Then
   * position vmemmap directly below the VMALLOC region.
   */
  #define VMEMMAP_SHIFT \
          (CONFIG_VA_BITS - PAGE_SHIFT - 1 + STRUCT_PAGE_MAX_SHIFT)
  #define VMEMMAP_SIZE    BIT(VMEMMAP_SHIFT)
  #define VMEMMAP_END     (VMALLOC_START - 1)
  #define VMEMMAP_START   (VMALLOC_START - VMEMMAP_SIZE)

  #define vmemmap         ((struct page *)VMEMMAP_START)

  >>>>>>> 7c8dce4b166113743adad131b5a24c4acc12f92c

Only take the BPF_* defines from there and move them higher up in the
same file. Remove the rest from the chunk. The VMALLOC_* etc defines
got moved via 01f52e16b868 ("riscv: define vmemmap before pfn_to_page
calls"). Result:

  [...]
  #define __S101  PAGE_READ_EXEC
  #define __S110  PAGE_SHARED_EXEC
  #define __S111  PAGE_SHARED_EXEC

  #define VMALLOC_SIZE     (KERN_VIRT_SIZE >> 1)
  #define VMALLOC_END      (PAGE_OFFSET - 1)
  #define VMALLOC_START    (PAGE_OFFSET - VMALLOC_SIZE)

  #define BPF_JIT_REGION_SIZE     (SZ_128M)
  #define BPF_JIT_REGION_START    (PAGE_OFFSET - BPF_JIT_REGION_SIZE)
  #define BPF_JIT_REGION_END      (VMALLOC_END)

  /*
   * Roughly size the vmemmap space to be large enough to fit enough
   * struct pages to map half the virtual address space. Then
   * position vmemmap directly below the VMALLOC region.
   */
  #define VMEMMAP_SHIFT \
          (CONFIG_VA_BITS - PAGE_SHIFT - 1 + STRUCT_PAGE_MAX_SHIFT)
  #define VMEMMAP_SIZE    BIT(VMEMMAP_SHIFT)
  #define VMEMMAP_END     (VMALLOC_START - 1)
  #define VMEMMAP_START   (VMALLOC_START - VMEMMAP_SIZE)

  [...]

Let me know if there are any other issues.

Anyway, the main changes are:

1) Extend bpftool to produce a struct (aka "skeleton") tailored and specific
   to a provided BPF object file. This provides an alternative, simplified API
   compared to standard libbpf interaction. Also, add libbpf extern variable
   resolution for .kconfig section to import Kconfig data, from Andrii Nakryiko.

2) Add BPF dispatcher for XDP which is a mechanism to avoid indirect calls by
   generating a branch funnel as discussed back in bpfconf'19 at LSF/MM. Also,
   add various BPF riscv JIT improvements, from Björn Töpel.

3) Extend bpftool to allow matching BPF programs and maps by name,
   from Paul Chaignon.

4) Support for replacing cgroup BPF programs attached with BPF_F_ALLOW_MULTI
   flag for allowing updates without service interruption, from Andrey Ignatov.

5) Cleanup and simplification of ring access functions for AF_XDP with a
   bonus of 0-5% performance improvement, from Magnus Karlsson.

6) Enable BPF JITs for x86-64 and arm64 by default. Also, final version of
   audit support for BPF, from Daniel Borkmann and latter with Jiri Olsa.

7) Move and extend test_select_reuseport into BPF program tests under
   BPF selftests, from Jakub Sitnicki.

8) Various BPF sample improvements for xdpsock for customizing parameters
   to set up and benchmark AF_XDP, from Jay Jayatheerthan.

9) Improve libbpf to provide a ulimit hint on permission denied errors.
   Also change XDP sample programs to attach in driver mode by default,
   from Toke Høiland-Jørgensen.

10) Extend BPF test infrastructure to allow changing skb mark from tc BPF
    programs, from Nikita V. Shirokov.

11) Optimize prologue code sequence in BPF arm32 JIT, from Russell King.

12) Fix xdp_redirect_cpu BPF sample to manually attach to tracepoints after
    libbpf conversion, from Jesper Dangaard Brouer.

13) Minor misc improvements from various others.
====================

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

110 files changed, 6901 insertions, 2721 deletions

diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c
index 97dc386e3cb8..cc29869d12a3 100644
--- a/arch/arm/net/bpf_jit_32.c
+++ b/arch/arm/net/bpf_jit_32.c
@@ -1260,12 +1260,9 @@ static inline void emit_push_r64(const s8 src[], struct jit_ctx *ctx)
 
 static void build_prologue(struct jit_ctx *ctx)
 {
-	const s8 r0 = bpf2a32[BPF_REG_0][1];
-	const s8 r2 = bpf2a32[BPF_REG_1][1];
-	const s8 r3 = bpf2a32[BPF_REG_1][0];
-	const s8 r4 = bpf2a32[BPF_REG_6][1];
-	const s8 fplo = bpf2a32[BPF_REG_FP][1];
-	const s8 fphi = bpf2a32[BPF_REG_FP][0];
+	const s8 arm_r0 = bpf2a32[BPF_REG_0][1];
+	const s8 *bpf_r1 = bpf2a32[BPF_REG_1];
+	const s8 *bpf_fp = bpf2a32[BPF_REG_FP];
 	const s8 *tcc = bpf2a32[TCALL_CNT];
 
 	/* Save callee saved registers. */
@@ -1278,8 +1275,10 @@ static void build_prologue(struct jit_ctx *ctx)
 	emit(ARM_PUSH(CALLEE_PUSH_MASK), ctx);
 	emit(ARM_MOV_R(ARM_FP, ARM_SP), ctx);
 #endif
-	/* Save frame pointer for later */
-	emit(ARM_SUB_I(ARM_IP, ARM_SP, SCRATCH_SIZE), ctx);
+	/* mov r3, #0 */
+	/* sub r2, sp, #SCRATCH_SIZE */
+	emit(ARM_MOV_I(bpf_r1[0], 0), ctx);
+	emit(ARM_SUB_I(bpf_r1[1], ARM_SP, SCRATCH_SIZE), ctx);
 
 	ctx->stack_size = imm8m(STACK_SIZE);
 
@@ -1287,18 +1286,15 @@ static void build_prologue(struct jit_ctx *ctx)
 	emit(ARM_SUB_I(ARM_SP, ARM_SP, ctx->stack_size), ctx);
 
 	/* Set up BPF prog stack base register */
-	emit_a32_mov_r(fplo, ARM_IP, ctx);
-	emit_a32_mov_i(fphi, 0, ctx);
+	emit_a32_mov_r64(true, bpf_fp, bpf_r1, ctx);
 
-	/* mov r4, 0 */
-	emit(ARM_MOV_I(r4, 0), ctx);
+	/* Initialize Tail Count */
+	emit(ARM_MOV_I(bpf_r1[1], 0), ctx);
+	emit_a32_mov_r64(true, tcc, bpf_r1, ctx);
 
 	/* Move BPF_CTX to BPF_R1 */
-	emit(ARM_MOV_R(r3, r4), ctx);
-	emit(ARM_MOV_R(r2, r0), ctx);
-	/* Initialize Tail Count */
-	emit(ARM_STR_I(r4, ARM_FP, EBPF_SCRATCH_TO_ARM_FP(tcc[0])), ctx);
-	emit(ARM_STR_I(r4, ARM_FP, EBPF_SCRATCH_TO_ARM_FP(tcc[1])), ctx);
+	emit(ARM_MOV_R(bpf_r1[1], arm_r0), ctx);
+
 	/* end of prologue */
 }
 
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index b1b4476ddb83..29d03459de20 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -69,6 +69,7 @@ config ARM64
 	select ARCH_SUPPORTS_INT128 if CC_HAS_INT128 && (GCC_VERSION >= 50000 || CC_IS_CLANG)
 	select ARCH_SUPPORTS_NUMA_BALANCING
 	select ARCH_WANT_COMPAT_IPC_PARSE_VERSION if COMPAT
+	select ARCH_WANT_DEFAULT_BPF_JIT
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_HUGE_PMD_SHARE if ARM64_4K_PAGES || (ARM64_16K_PAGES && !ARM64_VA_BITS_36)
diff --git a/arch/riscv/include/asm/perf_event.h b/arch/riscv/include/asm/perf_event.h
index aefbfaa6a781..0234048b12bc 100644
--- a/arch/riscv/include/asm/perf_event.h
+++ b/arch/riscv/include/asm/perf_event.h
@@ -82,4 +82,8 @@ struct riscv_pmu {
 	int		irq;
 };
 
+#ifdef CONFIG_PERF_EVENTS
+#define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs
+#endif
+
 #endif /* _ASM_RISCV_PERF_EVENT_H */
diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index 36ae01761352..f66b87314fa2 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/arch/riscv/include/asm/pgtable.h
@@ -94,6 +94,10 @@ extern pgd_t swapper_pg_dir[];
 #define VMALLOC_END      (PAGE_OFFSET - 1)
 #define VMALLOC_START    (PAGE_OFFSET - VMALLOC_SIZE)
 
+#define BPF_JIT_REGION_SIZE	(SZ_128M)
+#define BPF_JIT_REGION_START	(PAGE_OFFSET - BPF_JIT_REGION_SIZE)
+#define BPF_JIT_REGION_END	(VMALLOC_END)
+
 /*
  * Roughly size the vmemmap space to be large enough to fit enough
  * struct pages to map half the virtual address space. Then
diff --git a/arch/riscv/include/uapi/asm/bpf_perf_event.h b/arch/riscv/include/uapi/asm/bpf_perf_event.h
new file mode 100644
index 000000000000..6cb1c2823288
--- /dev/null
+++ b/arch/riscv/include/uapi/asm/bpf_perf_event.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#ifndef _UAPI__ASM_BPF_PERF_EVENT_H__
+#define _UAPI__ASM_BPF_PERF_EVENT_H__
+
+#include <asm/ptrace.h>
+
+typedef struct user_regs_struct bpf_user_pt_regs_t;
+
+#endif /* _UAPI__ASM_BPF_PERF_EVENT_H__ */
diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 7fbf56aab661..483f4ad7f4dc 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -120,6 +120,11 @@ static bool seen_reg(int reg, struct rv_jit_context *ctx)
 	return false;
 }
 
+static void mark_fp(struct rv_jit_context *ctx)
+{
+	__set_bit(RV_CTX_F_SEEN_S5, &ctx->flags);
+}
+
 static void mark_call(struct rv_jit_context *ctx)
 {
 	__set_bit(RV_CTX_F_SEEN_CALL, &ctx->flags);
@@ -456,6 +461,11 @@ static u32 rv_amoadd_d(u8 rd, u8 rs2, u8 rs1, u8 aq, u8 rl)
 	return rv_amo_insn(0, aq, rl, rs2, rs1, 3, rd, 0x2f);
 }
 
+static u32 rv_auipc(u8 rd, u32 imm31_12)
+{
+	return rv_u_insn(imm31_12, rd, 0x17);
+}
+
 static bool is_12b_int(s64 val)
 {
 	return -(1 << 11) <= val && val < (1 << 11);
@@ -479,27 +489,7 @@ static bool is_32b_int(s64 val)
 static int is_12b_check(int off, int insn)
 {
 	if (!is_12b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-		       insn, (int)off);
-		return -1;
-	}
-	return 0;
-}
-
-static int is_13b_check(int off, int insn)
-{
-	if (!is_13b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-		       insn, (int)off);
-		return -1;
-	}
-	return 0;
-}
-
-static int is_21b_check(int off, int insn)
-{
-	if (!is_21b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
+		pr_err("bpf-jit: insn=%d 12b < offset=%d not supported yet!\n",
 		       insn, (int)off);
 		return -1;
 	}
@@ -545,10 +535,13 @@ static void emit_imm(u8 rd, s64 val, struct rv_jit_context *ctx)
 		emit(rv_addi(rd, rd, lower), ctx);
 }
 
-static int rv_offset(int bpf_to, int bpf_from, struct rv_jit_context *ctx)
+static int rv_offset(int insn, int off, struct rv_jit_context *ctx)
 {
-	int from = ctx->offset[bpf_from] - 1, to = ctx->offset[bpf_to];
+	int from, to;
 
+	off++; /* BPF branch is from PC+1, RV is from PC */
+	from = (insn > 0) ? ctx->offset[insn - 1] : 0;
+	to = (insn + off > 0) ? ctx->offset[insn + off - 1] : 0;
 	return (to - from) << 2;
 }
 
@@ -559,7 +552,7 @@ static int epilogue_offset(struct rv_jit_context *ctx)
 	return (to - from) << 2;
 }
 
-static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
+static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx)
 {
 	int stack_adjust = ctx->stack_size, store_offset = stack_adjust - 8;
 
@@ -596,8 +589,114 @@ static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
 
 	emit(rv_addi(RV_REG_SP, RV_REG_SP, stack_adjust), ctx);
 	/* Set return value. */
-	emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
-	emit(rv_jalr(RV_REG_ZERO, reg, 0), ctx);
+	if (!is_tail_call)
+		emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
+	emit(rv_jalr(RV_REG_ZERO, is_tail_call ? RV_REG_T3 : RV_REG_RA,
+		     is_tail_call ? 4 : 0), /* skip TCC init */
+	     ctx);
+}
+
+/* return -1 or inverted cond */
+static int invert_bpf_cond(u8 cond)
+{
+	switch (cond) {
+	case BPF_JEQ:
+		return BPF_JNE;
+	case BPF_JGT:
+		return BPF_JLE;
+	case BPF_JLT:
+		return BPF_JGE;
+	case BPF_JGE:
+		return BPF_JLT;
+	case BPF_JLE:
+		return BPF_JGT;
+	case BPF_JNE:
+		return BPF_JEQ;
+	case BPF_JSGT:
+		return BPF_JSLE;
+	case BPF_JSLT:
+		return BPF_JSGE;
+	case BPF_JSGE:
+		return BPF_JSLT;
+	case BPF_JSLE:
+		return BPF_JSGT;
+	}
+	return -1;
+}
+
+static void emit_bcc(u8 cond, u8 rd, u8 rs, int rvoff,
+		     struct rv_jit_context *ctx)
+{
+	switch (cond) {
+	case BPF_JEQ:
+		emit(rv_beq(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JGT:
+		emit(rv_bltu(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JLT:
+		emit(rv_bltu(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JGE:
+		emit(rv_bgeu(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JLE:
+		emit(rv_bgeu(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JNE:
+		emit(rv_bne(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSGT:
+		emit(rv_blt(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JSLT:
+		emit(rv_blt(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSGE:
+		emit(rv_bge(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSLE:
+		emit(rv_bge(rs, rd, rvoff >> 1), ctx);
+	}
+}
+
+static void emit_branch(u8 cond, u8 rd, u8 rs, int rvoff,
+			struct rv_jit_context *ctx)
+{
+	s64 upper, lower;
+
+	if (is_13b_int(rvoff)) {
+		emit_bcc(cond, rd, rs, rvoff, ctx);
+		return;
+	}
+
+	/* Adjust for jal */
+	rvoff -= 4;
+
+	/* Transform, e.g.:
+	 *   bne rd,rs,foo
+	 * to
+	 *   beq rd,rs,<.L1>
+	 *   (auipc foo)
+	 *   jal(r) foo
+	 * .L1
+	 */
+	cond = invert_bpf_cond(cond);
+	if (is_21b_int(rvoff)) {
+		emit_bcc(cond, rd, rs, 8, ctx);
+		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		return;
+	}
+
+	/* 32b No need for an additional rvoff adjustment, since we
+	 * get that from the auipc at PC', where PC = PC' + 4.
+	 */
+	upper = (rvoff + (1 << 11)) >> 12;
+	lower = rvoff & 0xfff;
+
+	emit_bcc(cond, rd, rs, 12, ctx);
+	emit(rv_auipc(RV_REG_T1, upper), ctx);
+	emit(rv_jalr(RV_REG_ZERO, RV_REG_T1, lower), ctx);
 }
 
 static void emit_zext_32(u8 reg, struct rv_jit_context *ctx)
@@ -627,18 +726,14 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 		return -1;
 	emit(rv_lwu(RV_REG_T1, off, RV_REG_A1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_bgeu(RV_REG_A2, RV_REG_T1, off >> 1), ctx);
+	emit_branch(BPF_JGE, RV_REG_A2, RV_REG_T1, off, ctx);
 
 	/* if (TCC-- < 0)
 	 *     goto out;
 	 */
 	emit(rv_addi(RV_REG_T1, tcc, -1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_blt(tcc, RV_REG_ZERO, off >> 1), ctx);
+	emit_branch(BPF_JSLT, tcc, RV_REG_ZERO, off, ctx);
 
 	/* prog = array->ptrs[index];
 	 * if (!prog)
@@ -651,18 +746,15 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 		return -1;
 	emit(rv_ld(RV_REG_T2, off, RV_REG_T2), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_beq(RV_REG_T2, RV_REG_ZERO, off >> 1), ctx);
+	emit_branch(BPF_JEQ, RV_REG_T2, RV_REG_ZERO, off, ctx);
 
 	/* goto *(prog->bpf_func + 4); */
 	off = offsetof(struct bpf_prog, bpf_func);
 	if (is_12b_check(off, insn))
 		return -1;
 	emit(rv_ld(RV_REG_T3, off, RV_REG_T2), ctx);
-	emit(rv_addi(RV_REG_T3, RV_REG_T3, 4), ctx);
 	emit(rv_addi(RV_REG_TCC, RV_REG_T1, 0), ctx);
-	__build_epilogue(RV_REG_T3, ctx);
+	__build_epilogue(true, ctx);
 	return 0;
 }
 
@@ -687,13 +779,6 @@ static void init_regs(u8 *rd, u8 *rs, const struct bpf_insn *insn,
 		*rs = bpf_to_rv_reg(insn->src_reg, ctx);
 }
 
-static int rv_offset_check(int *rvoff, s16 off, int insn,
-			   struct rv_jit_context *ctx)
-{
-	*rvoff = rv_offset(insn + off, insn, ctx);
-	return is_13b_check(*rvoff, insn);
-}
-
 static void emit_zext_32_rd_rs(u8 *rd, u8 *rs, struct rv_jit_context *ctx)
 {
 	emit(rv_addi(RV_REG_T2, *rd, 0), ctx);
@@ -726,13 +811,57 @@ static void emit_sext_32_rd(u8 *rd, struct rv_jit_context *ctx)
 	*rd = RV_REG_T2;
 }
 
+static void emit_jump_and_link(u8 rd, s64 rvoff, bool force_jalr,
+			       struct rv_jit_context *ctx)
+{
+	s64 upper, lower;
+
+	if (rvoff && is_21b_int(rvoff) && !force_jalr) {
+		emit(rv_jal(rd, rvoff >> 1), ctx);
+		return;
+	}
+
+	upper = (rvoff + (1 << 11)) >> 12;
+	lower = rvoff & 0xfff;
+	emit(rv_auipc(RV_REG_T1, upper), ctx);
+	emit(rv_jalr(rd, RV_REG_T1, lower), ctx);
+}
+
+static bool is_signed_bpf_cond(u8 cond)
+{
+	return cond == BPF_JSGT || cond == BPF_JSLT ||
+		cond == BPF_JSGE || cond == BPF_JSLE;
+}
+
+static int emit_call(bool fixed, u64 addr, struct rv_jit_context *ctx)
+{
+	s64 off = 0;
+	u64 ip;
+	u8 rd;
+
+	if (addr && ctx->insns) {
+		ip = (u64)(long)(ctx->insns + ctx->ninsns);
+		off = addr - ip;
+		if (!is_32b_int(off)) {
+			pr_err("bpf-jit: target call addr %pK is out of range\n",
+			       (void *)addr);
+			return -ERANGE;
+		}
+	}
+
+	emit_jump_and_link(RV_REG_RA, off, !fixed, ctx);
+	rd = bpf_to_rv_reg(BPF_REG_0, ctx);
+	emit(rv_addi(rd, RV_REG_A0, 0), ctx);
+	return 0;
+}
+
 static int emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
 		     bool extra_pass)
 {
 	bool is64 = BPF_CLASS(insn->code) == BPF_ALU64 ||
 		    BPF_CLASS(insn->code) == BPF_JMP;
+	int s, e, rvoff, i = insn - ctx->prog->insnsi;
 	struct bpf_prog_aux *aux = ctx->prog->aux;
-	int rvoff, i = insn - ctx->prog->insnsi;
 	u8 rd = -1, rs = -1, code = insn->code;
 	s16 off = insn->off;
 	s32 imm = insn->imm;
@@ -1000,214 +1129,110 @@ out_be:
 
 	/* JUMP off */
 	case BPF_JMP | BPF_JA:
-		rvoff = rv_offset(i + off, i, ctx);
-		if (!is_21b_int(rvoff)) {
-			pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-			       i, rvoff);
-			return -1;
-		}
-
-		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		rvoff = rv_offset(i, off, ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, false, ctx);
 		break;
 
 	/* IF (dst COND src) JUMP off */
 	case BPF_JMP | BPF_JEQ | BPF_X:
 	case BPF_JMP32 | BPF_JEQ | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_beq(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGT | BPF_X:
 	case BPF_JMP32 | BPF_JGT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bltu(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLT | BPF_X:
 	case BPF_JMP32 | BPF_JLT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bltu(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGE | BPF_X:
 	case BPF_JMP32 | BPF_JGE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bgeu(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLE | BPF_X:
 	case BPF_JMP32 | BPF_JLE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bgeu(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JNE | BPF_X:
 	case BPF_JMP32 | BPF_JNE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bne(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGT | BPF_X:
 	case BPF_JMP32 | BPF_JSGT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_blt(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLT | BPF_X:
 	case BPF_JMP32 | BPF_JSLT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_blt(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGE | BPF_X:
 	case BPF_JMP32 | BPF_JSGE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bge(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLE | BPF_X:
 	case BPF_JMP32 | BPF_JSLE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bge(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSET | BPF_X:
 	case BPF_JMP32 | BPF_JSET | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_and(RV_REG_T1, rd, rs), ctx);
-		emit(rv_bne(RV_REG_T1, RV_REG_ZERO, rvoff >> 1), ctx);
+		rvoff = rv_offset(i, off, ctx);
+		if (!is64) {
+			s = ctx->ninsns;
+			if (is_signed_bpf_cond(BPF_OP(code)))
+				emit_sext_32_rd_rs(&rd, &rs, ctx);
+			else
+				emit_zext_32_rd_rs(&rd, &rs, ctx);
+			e = ctx->ninsns;
+
+			/* Adjust for extra insns */
+			rvoff -= (e - s) << 2;
+		}
+
+		if (BPF_OP(code) == BPF_JSET) {
+			/* Adjust for and */
+			rvoff -= 4;
+			emit(rv_and(RV_REG_T1, rd, rs), ctx);
+			emit_branch(BPF_JNE, RV_REG_T1, RV_REG_ZERO, rvoff,
+				    ctx);
+		} else {
+			emit_branch(BPF_OP(code), rd, rs, rvoff, ctx);
+		}
 		break;
 
 	/* IF (dst COND imm) JUMP off */
 	case BPF_JMP | BPF_JEQ | BPF_K:
 	case BPF_JMP32 | BPF_JEQ | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_beq(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGT | BPF_K:
 	case BPF_JMP32 | BPF_JGT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bltu(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLT | BPF_K:
 	case BPF_JMP32 | BPF_JLT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bltu(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGE | BPF_K:
 	case BPF_JMP32 | BPF_JGE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bgeu(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLE | BPF_K:
 	case BPF_JMP32 | BPF_JLE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bgeu(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JNE | BPF_K:
 	case BPF_JMP32 | BPF_JNE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bne(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGT | BPF_K:
 	case BPF_JMP32 | BPF_JSGT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_blt(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLT | BPF_K:
 	case BPF_JMP32 | BPF_JSLT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_blt(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGE | BPF_K:
 	case BPF_JMP32 | BPF_JSGE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_bge(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLE | BPF_K:
 	case BPF_JMP32 | BPF_JSLE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_bge(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSET | BPF_K:
 	case BPF_JMP32 | BPF_JSET | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
+		rvoff = rv_offset(i, off, ctx);
+		s = ctx->ninsns;
 		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_and(RV_REG_T1, rd, RV_REG_T1), ctx);
-		emit(rv_bne(RV_REG_T1, RV_REG_ZERO, rvoff >> 1), ctx);
+		if (!is64) {
+			if (is_signed_bpf_cond(BPF_OP(code)))
+				emit_sext_32_rd(&rd, ctx);
+			else
+				emit_zext_32_rd_t1(&rd, ctx);
+		}
+		e = ctx->ninsns;
+
+		/* Adjust for extra insns */
+		rvoff -= (e - s) << 2;
+
+		if (BPF_OP(code) == BPF_JSET) {
+			/* Adjust for and */
+			rvoff -= 4;
+			emit(rv_and(RV_REG_T1, rd, RV_REG_T1), ctx);
+			emit_branch(BPF_JNE, RV_REG_T1, RV_REG_ZERO, rvoff,
+				    ctx);
+		} else {
+			emit_branch(BPF_OP(code), rd, RV_REG_T1, rvoff, ctx);
+		}
 		break;
 
 	/* function call */
 	case BPF_JMP | BPF_CALL:
 	{
 		bool fixed;
-		int i, ret;
+		int ret;
 		u64 addr;
 
 		mark_call(ctx);
@@ -1215,20 +1240,9 @@ out_be:
 					    &fixed);
 		if (ret < 0)
 			return ret;
-		if (fixed) {
-			emit_imm(RV_REG_T1, addr, ctx);
-		} else {
-			i = ctx->ninsns;
-			emit_imm(RV_REG_T1, addr, ctx);
-			for (i = ctx->ninsns - i; i < 8; i++) {
-				/* nop */
-				emit(rv_addi(RV_REG_ZERO, RV_REG_ZERO, 0),
-				     ctx);
-			}
-		}
-		emit(rv_jalr(RV_REG_RA, RV_REG_T1, 0), ctx);
-		rd = bpf_to_rv_reg(BPF_REG_0, ctx);
-		emit(rv_addi(rd, RV_REG_A0, 0), ctx);
+		ret = emit_call(fixed, addr, ctx);
+		if (ret)
+			return ret;
 		break;
 	}
 	/* tail call */
@@ -1243,9 +1257,7 @@ out_be:
 			break;
 
 		rvoff = epilogue_offset(ctx);
-		if (is_21b_check(rvoff, i))
-			return -1;
-		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, false, ctx);
 		break;
 
 	/* dst = imm64 */
@@ -1426,6 +1438,10 @@ static void build_prologue(struct rv_jit_context *ctx)
 {
 	int stack_adjust = 0, store_offset, bpf_stack_adjust;
 
+	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
+	if (bpf_stack_adjust)
+		mark_fp(ctx);
+
 	if (seen_reg(RV_REG_RA, ctx))
 		stack_adjust += 8;
 	stack_adjust += 8; /* RV_REG_FP */
@@ -1443,7 +1459,6 @@ static void build_prologue(struct rv_jit_context *ctx)
 		stack_adjust += 8;
 
 	stack_adjust = round_up(stack_adjust, 16);
-	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
 	stack_adjust += bpf_stack_adjust;
 
 	store_offset = stack_adjust - 8;
@@ -1502,10 +1517,10 @@ static void build_prologue(struct rv_jit_context *ctx)
 
 static void build_epilogue(struct rv_jit_context *ctx)
 {
-	__build_epilogue(RV_REG_RA, ctx);
+	__build_epilogue(false, ctx);
 }
 
-static int build_body(struct rv_jit_context *ctx, bool extra_pass)
+static int build_body(struct rv_jit_context *ctx, bool extra_pass, int *offset)
 {
 	const struct bpf_prog *prog = ctx->prog;
 	int i;
@@ -1517,12 +1532,12 @@ static int build_body(struct rv_jit_context *ctx, bool extra_pass)
 		ret = emit_insn(insn, ctx, extra_pass);
 		if (ret > 0) {
 			i++;
-			if (ctx->insns == NULL)
-				ctx->offset[i] = ctx->ninsns;
+			if (offset)
+				offset[i] = ctx->ninsns;
 			continue;
 		}
-		if (ctx->insns == NULL)
-			ctx->offset[i] = ctx->ninsns;
+		if (offset)
+			offset[i] = ctx->ninsns;
 		if (ret)
 			return ret;
 	}
@@ -1548,9 +1563,10 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 {
 	bool tmp_blinded = false, extra_pass = false;
 	struct bpf_prog *tmp, *orig_prog = prog;
+	int pass = 0, prev_ninsns = 0, i;
 	struct rv_jit_data *jit_data;
+	unsigned int image_size = 0;
 	struct rv_jit_context *ctx;
-	unsigned int image_size;
 
 	if (!prog->jit_requested)
 		return orig_prog;
@@ -1587,33 +1603,59 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 		prog = orig_prog;
 		goto out_offset;
 	}
+	for (i = 0; i < prog->len; i++) {
+		prev_ninsns += 32;
+		ctx->offset[i] = prev_ninsns;
+	}
 
-	/* First pass generates the ctx->offset, but does not emit an image. */
-	if (build_body(ctx, extra_pass)) {
-		prog = orig_prog;
-		goto out_offset;
+	for (i = 0; i < 16; i++) {
+		pass++;
+		ctx->ninsns = 0;
+		if (build_body(ctx, extra_pass, ctx->offset)) {
+			prog = orig_prog;
+			goto out_offset;
+		}
+		build_prologue(ctx);
+		ctx->epilogue_offset = ctx->ninsns;
+		build_epilogue(ctx);
+
+		if (ctx->ninsns == prev_ninsns) {
+			if (jit_data->header)
+				break;
+
+			image_size = sizeof(u32) * ctx->ninsns;
+			jit_data->header =
+				bpf_jit_binary_alloc(image_size,
+						     &jit_data->image,
+						     sizeof(u32),
+						     bpf_fill_ill_insns);
+			if (!jit_data->header) {
+				prog = orig_prog;
+				goto out_offset;
+			}
+
+			ctx->insns = (u32 *)jit_data->image;
+			/* Now, when the image is allocated, the image
+			 * can potentially shrink more (auipc/jalr ->
+			 * jal).
+			 */
+		}
+		prev_ninsns = ctx->ninsns;
 	}
-	build_prologue(ctx);
-	ctx->epilogue_offset = ctx->ninsns;
-	build_epilogue(ctx);
 
-	/* Allocate image, now that we know the size. */
-	image_size = sizeof(u32) * ctx->ninsns;
-	jit_data->header = bpf_jit_binary_alloc(image_size, &jit_data->image,
-						sizeof(u32),
-						bpf_fill_ill_insns);
-	if (!jit_data->header) {
+	if (i == 16) {
+		pr_err("bpf-jit: image did not converge in <%d passes!\n", i);
+		bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
 	}
 
-	/* Second, real pass, that acutally emits the image. */
-	ctx->insns = (u32 *)jit_data->image;
 skip_init_ctx:
+	pass++;
 	ctx->ninsns = 0;
 
 	build_prologue(ctx);
-	if (build_body(ctx, extra_pass)) {
+	if (build_body(ctx, extra_pass, NULL)) {
 		bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
@@ -1621,7 +1663,7 @@ skip_init_ctx:
 	build_epilogue(ctx);
 
 	if (bpf_jit_enable > 1)
-		bpf_jit_dump(prog->len, image_size, 2, ctx->insns);
+		bpf_jit_dump(prog->len, image_size, pass, ctx->insns);
 
 	prog->bpf_func = (void *)ctx->insns;
 	prog->jited = 1;
@@ -1641,3 +1683,16 @@ out:
 					   tmp : orig_prog);
 	return prog;
 }
+
+void *bpf_jit_alloc_exec(unsigned long size)
+{
+	return __vmalloc_node_range(size, PAGE_SIZE, BPF_JIT_REGION_START,
+				    BPF_JIT_REGION_END, GFP_KERNEL,
+				    PAGE_KERNEL_EXEC, 0, NUMA_NO_NODE,
+				    __builtin_return_address(0));
+}
+
+void bpf_jit_free_exec(void *addr)
+{
+	return vfree(addr);
+}
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 5e8949953660..1f6a0388a65f 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -93,6 +93,7 @@ config X86
 	select ARCH_USE_QUEUED_RWLOCKS
 	select ARCH_USE_QUEUED_SPINLOCKS
 	select ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH
+	select ARCH_WANT_DEFAULT_BPF_JIT	if X86_64
 	select ARCH_WANTS_DYNAMIC_TASK_STRUCT
 	select ARCH_WANT_HUGE_PMD_SHARE
 	select ARCH_WANTS_THP_SWAP		if X86_64
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index b8be18427277..4c8a2d1f8470 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -10,10 +10,12 @@
 #include <linux/if_vlan.h>
 #include <linux/bpf.h>
 #include <linux/memory.h>
+#include <linux/sort.h>
 #include <asm/extable.h>
 #include <asm/set_memory.h>
 #include <asm/nospec-branch.h>
 #include <asm/text-patching.h>
+#include <asm/asm-prototypes.h>
 
 static u8 *emit_code(u8 *ptr, u32 bytes, unsigned int len)
 {
@@ -1530,6 +1532,154 @@ int arch_prepare_bpf_trampoline(void *image, struct btf_func_model *m, u32 flags
 	return 0;
 }
 
+static int emit_cond_near_jump(u8 **pprog, void *func, void *ip, u8 jmp_cond)
+{
+	u8 *prog = *pprog;
+	int cnt = 0;
+	s64 offset;
+
+	offset = func - (ip + 2 + 4);
+	if (!is_simm32(offset)) {
+		pr_err("Target %p is out of range\n", func);
+		return -EINVAL;
+	}
+	EMIT2_off32(0x0F, jmp_cond + 0x10, offset);
+	*pprog = prog;
+	return 0;
+}
+
+static void emit_nops(u8 **pprog, unsigned int len)
+{
+	unsigned int i, noplen;
+	u8 *prog = *pprog;
+	int cnt = 0;
+
+	while (len > 0) {
+		noplen = len;
+
+		if (noplen > ASM_NOP_MAX)
+			noplen = ASM_NOP_MAX;
+
+		for (i = 0; i < noplen; i++)
+			EMIT1(ideal_nops[noplen][i]);
+		len -= noplen;
+	}
+
+	*pprog = prog;
+}
+
+static int emit_fallback_jump(u8 **pprog)
+{
+	u8 *prog = *pprog;
+	int err = 0;
+
+#ifdef CONFIG_RETPOLINE
+	/* Note that this assumes the the compiler uses external
+	 * thunks for indirect calls. Both clang and GCC use the same
+	 * naming convention for external thunks.
+	 */
+	err = emit_jump(&prog, __x86_indirect_thunk_rdx, prog);
+#else
+	int cnt = 0;
+
+	EMIT2(0xFF, 0xE2);	/* jmp rdx */
+#endif
+	*pprog = prog;
+	return err;
+}
+
+static int emit_bpf_dispatcher(u8 **pprog, int a, int b, s64 *progs)
+{
+	u8 *jg_reloc, *jg_target, *prog = *pprog;
+	int pivot, err, jg_bytes = 1, cnt = 0;
+	s64 jg_offset;
+
+	if (a == b) {
+		/* Leaf node of recursion, i.e. not a range of indices
+		 * anymore.
+		 */
+		EMIT1(add_1mod(0x48, BPF_REG_3));	/* cmp rdx,func */
+		if (!is_simm32(progs[a]))
+			return -1;
+		EMIT2_off32(0x81, add_1reg(0xF8, BPF_REG_3),
+			    progs[a]);
+		err = emit_cond_near_jump(&prog,	/* je func */
+					  (void *)progs[a], prog,
+					  X86_JE);
+		if (err)
+			return err;
+
+		err = emit_fallback_jump(&prog);	/* jmp thunk/indirect */
+		if (err)
+			return err;
+
+		*pprog = prog;
+		return 0;
+	}
+
+	/* Not a leaf node, so we pivot, and recursively descend into
+	 * the lower and upper ranges.
+	 */
+	pivot = (b - a) / 2;
+	EMIT1(add_1mod(0x48, BPF_REG_3));		/* cmp rdx,func */
+	if (!is_simm32(progs[a + pivot]))
+		return -1;
+	EMIT2_off32(0x81, add_1reg(0xF8, BPF_REG_3), progs[a + pivot]);
+
+	if (pivot > 2) {				/* jg upper_part */
+		/* Require near jump. */
+		jg_bytes = 4;
+		EMIT2_off32(0x0F, X86_JG + 0x10, 0);
+	} else {
+		EMIT2(X86_JG, 0);
+	}
+	jg_reloc = prog;
+
+	err = emit_bpf_dispatcher(&prog, a, a + pivot,	/* emit lower_part */
+				  progs);
+	if (err)
+		return err;
+
+	/* From Intel 64 and IA-32 Architectures Optimization
+	 * Reference Manual, 3.4.1.4 Code Alignment, Assembly/Compiler
+	 * Coding Rule 11: All branch targets should be 16-byte
+	 * aligned.
+	 */
+	jg_target = PTR_ALIGN(prog, 16);
+	if (jg_target != prog)
+		emit_nops(&prog, jg_target - prog);
+	jg_offset = prog - jg_reloc;
+	emit_code(jg_reloc - jg_bytes, jg_offset, jg_bytes);
+
+	err = emit_bpf_dispatcher(&prog, a + pivot + 1,	/* emit upper_part */
+				  b, progs);
+	if (err)
+		return err;
+
+	*pprog = prog;
+	return 0;
+}
+
+static int cmp_ips(const void *a, const void *b)
+{
+	const s64 *ipa = a;
+	const s64 *ipb = b;
+
+	if (*ipa > *ipb)
+		return 1;
+	if (*ipa < *ipb)
+		return -1;
+	return 0;
+}
+
+int arch_prepare_bpf_dispatcher(void *image, s64 *funcs, int num_funcs)
+{
+	u8 *prog = image;
+
+	sort(funcs, num_funcs, sizeof(funcs[0]), cmp_ips, NULL);
+	return emit_bpf_dispatcher(&prog, 0, num_funcs - 1, funcs);
+}
+
 struct x64_jit_data {
 	struct bpf_binary_header *header;
 	int *addrs;
diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
index f73cd917c44f..42058fad6a3c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c
@@ -269,7 +269,7 @@ static bool i40e_alloc_buffer_zc(struct i40e_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr(umem);
+	xsk_umem_release_addr(umem);
 	return true;
 }
 
@@ -306,7 +306,7 @@ static bool i40e_alloc_buffer_slow_zc(struct i40e_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 	return true;
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c
index cf9b8b22d24f..0af1f0b951c0 100644
--- a/drivers/net/ethernet/intel/ice/ice_xsk.c
+++ b/drivers/net/ethernet/intel/ice/ice_xsk.c
@@ -555,7 +555,7 @@ ice_alloc_buf_fast_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
 
 	rx_buf->handle = handle + umem->headroom;
 
-	xsk_umem_discard_addr(umem);
+	xsk_umem_release_addr(umem);
 	return true;
 }
 
@@ -591,7 +591,7 @@ ice_alloc_buf_slow_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf)
 
 	rx_buf->handle = handle + umem->headroom;
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 	return true;
 }
 
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
index b43be9f14105..74b540ebb3dc 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c
@@ -277,7 +277,7 @@ static bool ixgbe_alloc_buffer_zc(struct ixgbe_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr(umem);
+	xsk_umem_release_addr(umem);
 	return true;
 }
 
@@ -304,7 +304,7 @@ static bool ixgbe_alloc_buffer_slow_zc(struct ixgbe_ring *rx_ring,
 
 	bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom);
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 	return true;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
index 475b6bd5d29b..62fc8a128a8d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c
@@ -35,7 +35,7 @@ int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq,
 	 */
 	dma_info->addr = xdp_umem_get_dma(umem, handle);
 
-	xsk_umem_discard_addr_rq(umem);
+	xsk_umem_release_addr_rq(umem);
 
 	dma_sync_single_for_device(rq->pdev, dma_info->addr, PAGE_SIZE,
 				   DMA_BIDIRECTIONAL);
diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
index 9be71c195d74..a11d5b7dbbf3 100644
--- a/include/linux/bpf-cgroup.h
+++ b/include/linux/bpf-cgroup.h
@@ -85,6 +85,7 @@ int cgroup_bpf_inherit(struct cgroup *cgrp);
 void cgroup_bpf_offline(struct cgroup *cgrp);
 
 int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+			struct bpf_prog *replace_prog,
 			enum bpf_attach_type type, u32 flags);
 int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 			enum bpf_attach_type type);
@@ -93,7 +94,8 @@ int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
 
 /* Wrapper for __cgroup_bpf_*() protected by cgroup_mutex */
 int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
-		      enum bpf_attach_type type, u32 flags);
+		      struct bpf_prog *replace_prog, enum bpf_attach_type type,
+		      u32 flags);
 int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
 		      enum bpf_attach_type type, u32 flags);
 int cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 085a59afba85..b14e51d56a82 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -471,11 +471,69 @@ struct bpf_trampoline {
 	void *image;
 	u64 selector;
 };
+
+#define BPF_DISPATCHER_MAX 48 /* Fits in 2048B */
+
+struct bpf_dispatcher_prog {
+	struct bpf_prog *prog;
+	refcount_t users;
+};
+
+struct bpf_dispatcher {
+	/* dispatcher mutex */
+	struct mutex mutex;
+	void *func;
+	struct bpf_dispatcher_prog progs[BPF_DISPATCHER_MAX];
+	int num_progs;
+	void *image;
+	u32 image_off;
+};
+
+static __always_inline unsigned int bpf_dispatcher_nopfunc(
+	const void *ctx,
+	const struct bpf_insn *insnsi,
+	unsigned int (*bpf_func)(const void *,
+				 const struct bpf_insn *))
+{
+	return bpf_func(ctx, insnsi);
+}
 #ifdef CONFIG_BPF_JIT
 struct bpf_trampoline *bpf_trampoline_lookup(u64 key);
 int bpf_trampoline_link_prog(struct bpf_prog *prog);
 int bpf_trampoline_unlink_prog(struct bpf_prog *prog);
 void bpf_trampoline_put(struct bpf_trampoline *tr);
+void *bpf_jit_alloc_exec_page(void);
+#define BPF_DISPATCHER_INIT(name) {			\
+	.mutex = __MUTEX_INITIALIZER(name.mutex),	\
+	.func = &name##func,				\
+	.progs = {},					\
+	.num_progs = 0,					\
+	.image = NULL,					\
+	.image_off = 0					\
+}
+
+#define DEFINE_BPF_DISPATCHER(name)					\
+	noinline unsigned int name##func(				\
+		const void *ctx,					\
+		const struct bpf_insn *insnsi,				\
+		unsigned int (*bpf_func)(const void *,			\
+					 const struct bpf_insn *))	\
+	{								\
+		return bpf_func(ctx, insnsi);				\
+	}								\
+	EXPORT_SYMBOL(name##func);			\
+	struct bpf_dispatcher name = BPF_DISPATCHER_INIT(name);
+#define DECLARE_BPF_DISPATCHER(name)					\
+	unsigned int name##func(					\
+		const void *ctx,					\
+		const struct bpf_insn *insnsi,				\
+		unsigned int (*bpf_func)(const void *,			\
+					 const struct bpf_insn *));	\
+	extern struct bpf_dispatcher name;
+#define BPF_DISPATCHER_FUNC(name) name##func
+#define BPF_DISPATCHER_PTR(name) (&name)
+void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
+				struct bpf_prog *to);
 #else
 static inline struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
@@ -490,6 +548,13 @@ static inline int bpf_trampoline_unlink_prog(struct bpf_prog *prog)
 	return -ENOTSUPP;
 }
 static inline void bpf_trampoline_put(struct bpf_trampoline *tr) {}
+#define DEFINE_BPF_DISPATCHER(name)
+#define DECLARE_BPF_DISPATCHER(name)
+#define BPF_DISPATCHER_FUNC(name) bpf_dispatcher_nopfunc
+#define BPF_DISPATCHER_PTR(name) NULL
+static inline void bpf_dispatcher_change_prog(struct bpf_dispatcher *d,
+					      struct bpf_prog *from,
+					      struct bpf_prog *to) {}
 #endif
 
 struct bpf_func_info_aux {
@@ -897,14 +962,14 @@ struct sk_buff;
 
 struct bpf_dtab_netdev *__dev_map_lookup_elem(struct bpf_map *map, u32 key);
 struct bpf_dtab_netdev *__dev_map_hash_lookup_elem(struct bpf_map *map, u32 key);
-void __dev_map_flush(struct bpf_map *map);
+void __dev_map_flush(void);
 int dev_map_enqueue(struct bpf_dtab_netdev *dst, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 int dev_map_generic_redirect(struct bpf_dtab_netdev *dst, struct sk_buff *skb,
 			     struct bpf_prog *xdp_prog);
 
 struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key);
-void __cpu_map_flush(struct bpf_map *map);
+void __cpu_map_flush(void);
 int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
 		    struct net_device *dev_rx);
 
@@ -943,6 +1008,8 @@ int btf_distill_func_proto(struct bpf_verifier_log *log,
 
 int btf_check_func_arg_match(struct bpf_verifier_env *env, int subprog);
 
+struct bpf_prog *bpf_prog_by_id(u32 id);
+
 #else /* !CONFIG_BPF_SYSCALL */
 static inline struct bpf_prog *bpf_prog_get(u32 ufd)
 {
@@ -1004,7 +1071,7 @@ static inline struct net_device  *__dev_map_hash_lookup_elem(struct bpf_map *map
 	return NULL;
 }
 
-static inline void __dev_map_flush(struct bpf_map *map)
+static inline void __dev_map_flush(void)
 {
 }
 
@@ -1033,7 +1100,7 @@ struct bpf_cpu_map_entry *__cpu_map_lookup_elem(struct bpf_map *map, u32 key)
 	return NULL;
 }
 
-static inline void __cpu_map_flush(struct bpf_map *map)
+static inline void __cpu_map_flush(void)
 {
 }
 
@@ -1074,6 +1141,11 @@ static inline int bpf_prog_test_run_flow_dissector(struct bpf_prog *prog,
 static inline void bpf_map_put(struct bpf_map *map)
 {
 }
+
+static inline struct bpf_prog *bpf_prog_by_id(u32 id)
+{
+	return ERR_PTR(-ENOTSUPP);
+}
 #endif /* CONFIG_BPF_SYSCALL */
 
 static inline struct bpf_prog *bpf_prog_get_type(u32 ufd,
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 345f3748e0fb..70e6dd960bca 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -559,23 +559,26 @@ struct sk_filter {
 
 DECLARE_STATIC_KEY_FALSE(bpf_stats_enabled_key);
 
-#define BPF_PROG_RUN(prog, ctx)	({				\
-	u32 ret;						\
-	cant_sleep();						\
-	if (static_branch_unlikely(&bpf_stats_enabled_key)) {	\
-		struct bpf_prog_stats *stats;			\
-		u64 start = sched_clock();			\
-		ret = (*(prog)->bpf_func)(ctx, (prog)->insnsi);	\
-		stats = this_cpu_ptr(prog->aux->stats);		\
-		u64_stats_update_begin(&stats->syncp);		\
-		stats->cnt++;					\
-		stats->nsecs += sched_clock() - start;		\
-		u64_stats_update_end(&stats->syncp);		\
-	} else {						\
-		ret = (*(prog)->bpf_func)(ctx, (prog)->insnsi);	\
-	}							\
+#define __BPF_PROG_RUN(prog, ctx, dfunc)	({			\
+	u32 ret;							\
+	cant_sleep();							\
+	if (static_branch_unlikely(&bpf_stats_enabled_key)) {		\
+		struct bpf_prog_stats *stats;				\
+		u64 start = sched_clock();				\
+		ret = dfunc(ctx, (prog)->insnsi, (prog)->bpf_func);	\
+		stats = this_cpu_ptr(prog->aux->stats);			\
+		u64_stats_update_begin(&stats->syncp);			\
+		stats->cnt++;						\
+		stats->nsecs += sched_clock() - start;			\
+		u64_stats_update_end(&stats->syncp);			\
+	} else {							\
+		ret = dfunc(ctx, (prog)->insnsi, (prog)->bpf_func);	\
+	}								\
 	ret; })
 
+#define BPF_PROG_RUN(prog, ctx) __BPF_PROG_RUN(prog, ctx,		\
+					       bpf_dispatcher_nopfunc)
+
 #define BPF_SKB_CB_LEN QDISC_CB_PRIV_LEN
 
 struct bpf_skb_data_end {
@@ -589,7 +592,6 @@ struct bpf_redirect_info {
 	u32 tgt_index;
 	void *tgt_value;
 	struct bpf_map *map;
-	struct bpf_map *map_to_flush;
 	u32 kern_flags;
 };
 
@@ -699,6 +701,8 @@ static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog,
 	return res;
 }
 
+DECLARE_BPF_DISPATCHER(bpf_dispatcher_xdp)
+
 static __always_inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog,
 					    struct xdp_buff *xdp)
 {
@@ -708,9 +712,12 @@ static __always_inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog,
 	 * already takes rcu_read_lock() when fetching the program, so
 	 * it's not necessary here anymore.
 	 */
-	return BPF_PROG_RUN(prog, xdp);
+	return __BPF_PROG_RUN(prog, xdp,
+			      BPF_DISPATCHER_FUNC(bpf_dispatcher_xdp));
 }
 
+void bpf_prog_change_xdp(struct bpf_prog *prev_prog, struct bpf_prog *prog);
+
 static inline u32 bpf_prog_insn_size(const struct bpf_prog *prog)
 {
 	return prog->len * sizeof(struct bpf_insn);
diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index e3780e4b74e1..e86ec48ef627 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -72,7 +72,6 @@ struct xdp_umem {
 
 struct xsk_map {
 	struct bpf_map map;
-	struct list_head __percpu *flush_list;
 	spinlock_t lock; /* Synchronize map updates */
 	struct xdp_sock *xsk_map[];
 };
@@ -119,8 +118,8 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp);
 bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs);
 /* Used from netdev driver */
 bool xsk_umem_has_addrs(struct xdp_umem *umem, u32 cnt);
-u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr);
-void xsk_umem_discard_addr(struct xdp_umem *umem);
+bool xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr);
+void xsk_umem_release_addr(struct xdp_umem *umem);
 void xsk_umem_complete_tx(struct xdp_umem *umem, u32 nb_entries);
 bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc);
 void xsk_umem_consume_tx_done(struct xdp_umem *umem);
@@ -139,9 +138,8 @@ void xsk_map_try_sock_delete(struct xsk_map *map, struct xdp_sock *xs,
 			     struct xdp_sock **map_entry);
 int xsk_map_inc(struct xsk_map *map);
 void xsk_map_put(struct xsk_map *map);
-int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
-		       struct xdp_sock *xs);
-void __xsk_map_flush(struct bpf_map *map);
+int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp);
+void __xsk_map_flush(void);
 
 static inline struct xdp_sock *__xsk_map_lookup_elem(struct bpf_map *map,
 						     u32 key)
@@ -199,7 +197,7 @@ static inline bool xsk_umem_has_addrs_rq(struct xdp_umem *umem, u32 cnt)
 	return xsk_umem_has_addrs(umem, cnt - rq->length);
 }
 
-static inline u64 *xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
+static inline bool xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
 {
 	struct xdp_umem_fq_reuse *rq = umem->fq_reuse;
 
@@ -210,12 +208,12 @@ static inline u64 *xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
 	return addr;
 }
 
-static inline void xsk_umem_discard_addr_rq(struct xdp_umem *umem)
+static inline void xsk_umem_release_addr_rq(struct xdp_umem *umem)
 {
 	struct xdp_umem_fq_reuse *rq = umem->fq_reuse;
 
 	if (!rq->length)
-		xsk_umem_discard_addr(umem);
+		xsk_umem_release_addr(umem);
 	else
 		rq->length--;
 }
@@ -260,7 +258,7 @@ static inline u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
 	return NULL;
 }
 
-static inline void xsk_umem_discard_addr(struct xdp_umem *umem)
+static inline void xsk_umem_release_addr(struct xdp_umem *umem)
 {
 }
 
@@ -334,7 +332,7 @@ static inline u64 *xsk_umem_peek_addr_rq(struct xdp_umem *umem, u64 *addr)
 	return NULL;
 }
 
-static inline void xsk_umem_discard_addr_rq(struct xdp_umem *umem)
+static inline void xsk_umem_release_addr_rq(struct xdp_umem *umem)
 {
 }
 
@@ -369,13 +367,12 @@ static inline u64 xsk_umem_adjust_offset(struct xdp_umem *umem, u64 handle,
 	return 0;
 }
 
-static inline int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
-				     struct xdp_sock *xs)
+static inline int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 {
 	return -EOPNOTSUPP;
 }
 
-static inline void __xsk_map_flush(struct bpf_map *map)
+static inline void __xsk_map_flush(void)
 {
 }
 
diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h
index 3ad935527177..a534d71e689a 100644
--- a/include/uapi/linux/audit.h
+++ b/include/uapi/linux/audit.h
@@ -116,6 +116,7 @@
 #define AUDIT_FANOTIFY		1331	/* Fanotify access decision */
 #define AUDIT_TIME_INJOFFSET	1332	/* Timekeeping offset injected */
 #define AUDIT_TIME_ADJNTPVAL	1333	/* NTP value adjustment */
+#define AUDIT_BPF		1334	/* BPF subsystem */
 
 #define AUDIT_AVC		1400	/* SE Linux avc denial or grant */
 #define AUDIT_SELINUX_ERR	1401	/* Internal SE Linux Errors */
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index dbbcf0b02970..7df436da542d 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -231,6 +231,11 @@ enum bpf_attach_type {
  * When children program makes decision (like picking TCP CA or sock bind)
  * parent program has a chance to override it.
  *
+ * With BPF_F_ALLOW_MULTI a new program is added to the end of the list of
+ * programs for a cgroup. Though it's possible to replace an old program at
+ * any position by also specifying BPF_F_REPLACE flag and position itself in
+ * replace_bpf_fd attribute. Old program at this position will be released.
+ *
  * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
  * A cgroup with NONE doesn't allow any programs in sub-cgroups.
  * Ex1:
@@ -249,6 +254,7 @@ enum bpf_attach_type {
  */
 #define BPF_F_ALLOW_OVERRIDE	(1U << 0)
 #define BPF_F_ALLOW_MULTI	(1U << 1)
+#define BPF_F_REPLACE		(1U << 2)
 
 /* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will perform strict alignment checking as if the kernel
@@ -442,6 +448,10 @@ union bpf_attr {
 		__u32		attach_bpf_fd;	/* eBPF program to attach */
 		__u32		attach_type;
 		__u32		attach_flags;
+		__u32		replace_bpf_fd;	/* previously attached eBPF
+						 * program to replace if
+						 * BPF_F_REPLACE is used
+						 */
 	};
 
 	struct { /* anonymous struct used by BPF_PROG_TEST_RUN command */
diff --git a/include/uapi/linux/btf.h b/include/uapi/linux/btf.h
index c02dec97e1ce..1a2898c482ee 100644
--- a/include/uapi/linux/btf.h
+++ b/include/uapi/linux/btf.h
@@ -142,7 +142,8 @@ struct btf_param {
 
 enum {
 	BTF_VAR_STATIC = 0,
-	BTF_VAR_GLOBAL_ALLOCATED,
+	BTF_VAR_GLOBAL_ALLOCATED = 1,
+	BTF_VAR_GLOBAL_EXTERN = 2,
 };
 
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe
diff --git a/init/Kconfig b/init/Kconfig
index a34064a031a5..890aaa62efde 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1604,6 +1604,9 @@ config BPF_SYSCALL
 	  Enable the bpf() system call that allows to manipulate eBPF
 	  programs and maps via file descriptors.
 
+config ARCH_WANT_DEFAULT_BPF_JIT
+	bool
+
 config BPF_JIT_ALWAYS_ON
 	bool "Permanently enable BPF JIT and remove BPF interpreter"
 	depends on BPF_SYSCALL && HAVE_EBPF_JIT && BPF_JIT
@@ -1611,6 +1614,10 @@ config BPF_JIT_ALWAYS_ON
 	  Enables BPF JIT and removes BPF interpreter to avoid
 	  speculative execution of BPF instructions by the interpreter
 
+config BPF_JIT_DEFAULT_ON
+	def_bool ARCH_WANT_DEFAULT_BPF_JIT || BPF_JIT_ALWAYS_ON
+	depends on HAVE_EBPF_JIT && BPF_JIT
+
 config USERFAULTFD
 	bool "Enable userfaultfd() system call"
 	depends on MMU
diff --git a/kernel/bpf/Makefile b/kernel/bpf/Makefile
index 3f671bf617e8..d4f330351f87 100644
--- a/kernel/bpf/Makefile
+++ b/kernel/bpf/Makefile
@@ -8,6 +8,7 @@ obj-$(CONFIG_BPF_SYSCALL) += local_storage.o queue_stack_maps.o
 obj-$(CONFIG_BPF_SYSCALL) += disasm.o
 obj-$(CONFIG_BPF_JIT) += trampoline.o
 obj-$(CONFIG_BPF_SYSCALL) += btf.o
+obj-$(CONFIG_BPF_JIT) += dispatcher.o
 ifeq ($(CONFIG_NET),y)
 obj-$(CONFIG_BPF_SYSCALL) += devmap.o
 obj-$(CONFIG_BPF_SYSCALL) += cpumap.o
diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 4fb20ab179fe..ee871d4a51f7 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -103,8 +103,7 @@ static u32 prog_list_length(struct list_head *head)
  * if parent has overridable or multi-prog, allow attaching
  */
 static bool hierarchy_allows_attach(struct cgroup *cgrp,
-				    enum bpf_attach_type type,
-				    u32 new_flags)
+				    enum bpf_attach_type type)
 {
 	struct cgroup *p;
 
@@ -283,31 +282,34 @@ cleanup:
  *                         propagate the change to descendants
  * @cgrp: The cgroup which descendants to traverse
  * @prog: A program to attach
+ * @replace_prog: Previously attached program to replace if BPF_F_REPLACE is set
  * @type: Type of attach operation
  * @flags: Option flags
  *
  * Must be called with cgroup_mutex held.
  */
 int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+			struct bpf_prog *replace_prog,
 			enum bpf_attach_type type, u32 flags)
 {
+	u32 saved_flags = (flags & (BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI));
 	struct list_head *progs = &cgrp->bpf.progs[type];
 	struct bpf_prog *old_prog = NULL;
 	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE],
 		*old_storage[MAX_BPF_CGROUP_STORAGE_TYPE] = {NULL};
+	struct bpf_prog_list *pl, *replace_pl = NULL;
 	enum bpf_cgroup_storage_type stype;
-	struct bpf_prog_list *pl;
-	bool pl_was_allocated;
 	int err;
 
-	if ((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI))
+	if (((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI)) ||
+	    ((flags & BPF_F_REPLACE) && !(flags & BPF_F_ALLOW_MULTI)))
 		/* invalid combination */
 		return -EINVAL;
 
-	if (!hierarchy_allows_attach(cgrp, type, flags))
+	if (!hierarchy_allows_attach(cgrp, type))
 		return -EPERM;
 
-	if (!list_empty(progs) && cgrp->bpf.flags[type] != flags)
+	if (!list_empty(progs) && cgrp->bpf.flags[type] != saved_flags)
 		/* Disallow attaching non-overridable on top
 		 * of existing overridable in this cgroup.
 		 * Disallow attaching multi-prog if overridable or none
@@ -317,6 +319,21 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 	if (prog_list_length(progs) >= BPF_CGROUP_MAX_PROGS)
 		return -E2BIG;
 
+	if (flags & BPF_F_ALLOW_MULTI) {
+		list_for_each_entry(pl, progs, node) {
+			if (pl->prog == prog)
+				/* disallow attaching the same prog twice */
+				return -EINVAL;
+			if (pl->prog == replace_prog)
+				replace_pl = pl;
+		}
+		if ((flags & BPF_F_REPLACE) && !replace_pl)
+			/* prog to replace not found for cgroup */
+			return -ENOENT;
+	} else if (!list_empty(progs)) {
+		replace_pl = list_first_entry(progs, typeof(*pl), node);
+	}
+
 	for_each_cgroup_storage_type(stype) {
 		storage[stype] = bpf_cgroup_storage_alloc(prog, stype);
 		if (IS_ERR(storage[stype])) {
@@ -327,53 +344,28 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
 		}
 	}
 
-	if (flags & BPF_F_ALLOW_MULTI) {
-		list_for_each_entry(pl, progs, node) {
-			if (pl->prog == prog) {
-				/* disallow attaching the same prog twice */
-				for_each_cgroup_storage_type(stype)
-					bpf_cgroup_storage_free(storage[stype]);
-				return -EINVAL;
-			}
+	if (replace_pl) {
+		pl = replace_pl;
+		old_prog = pl->prog;
+		for_each_cgroup_storage_type(stype) {
+			old_storage[stype] = pl->storage[stype];
+			bpf_cgroup_storage_unlink(old_storage[stype]);
 		}
-
+	} else {
 		pl = kmalloc(sizeof(*pl), GFP_KERNEL);
 		if (!pl) {
 			for_each_cgroup_storage_type(stype)
 				bpf_cgroup_storage_free(storage[stype]);
 			return -ENOMEM;
 		}
-
-		pl_was_allocated = true;
-		pl->prog = prog;
-		for_each_cgroup_storage_type(stype)
-			pl->storage[stype] = storage[stype];
 		list_add_tail(&pl->node, progs);
-	} else {
-		if (list_empty(progs)) {
-			pl = kmalloc(sizeof(*pl), GFP_KERNEL);
-			if (!pl) {
-				for_each_cgroup_storage_type(stype)
-					bpf_cgroup_storage_free(storage[stype]);
-				return -ENOMEM;
-			}
-			pl_was_allocated = true;
-			list_add_tail(&pl->node, progs);
-		} else {
-			pl = list_first_entry(progs, typeof(*pl), node);
-			old_prog = pl->prog;
-			for_each_cgroup_storage_type(stype) {
-				old_storage[stype] = pl->storage[stype];
-				bpf_cgroup_storage_unlink(old_storage[stype]);
-			}
-			pl_was_allocated = false;
-		}
-		pl->prog = prog;
-		for_each_cgroup_storage_type(stype)
-			pl->storage[stype] = storage[stype];
 	}
 
-	cgrp->bpf.flags[type] = flags;
+	pl->prog = prog;
+	for_each_cgroup_storage_type(stype)
+		pl->storage[stype] = storage[stype];
+
+	cgrp->bpf.flags[type] = saved_flags;
 
 	err = update_effective_progs(cgrp, type);
 	if (err)
@@ -401,7 +393,7 @@ cleanup:
 		pl->storage[stype] = old_storage[stype];
 		bpf_cgroup_storage_link(old_storage[stype], cgrp, type);
 	}
-	if (pl_was_allocated) {
+	if (!replace_pl) {
 		list_del(&pl->node);
 		kfree(pl);
 	}
@@ -539,6 +531,7 @@ int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
 int cgroup_bpf_prog_attach(const union bpf_attr *attr,
 			   enum bpf_prog_type ptype, struct bpf_prog *prog)
 {
+	struct bpf_prog *replace_prog = NULL;
 	struct cgroup *cgrp;
 	int ret;
 
@@ -546,8 +539,20 @@ int cgroup_bpf_prog_attach(const union bpf_attr *attr,
 	if (IS_ERR(cgrp))
 		return PTR_ERR(cgrp);
 
-	ret = cgroup_bpf_attach(cgrp, prog, attr->attach_type,
+	if ((attr->attach_flags & BPF_F_ALLOW_MULTI) &&
+	    (attr->attach_flags & BPF_F_REPLACE)) {
+		replace_prog = bpf_prog_get_type(attr->replace_bpf_fd, ptype);
+		if (IS_ERR(replace_prog)) {
+			cgroup_put(cgrp);
+			return PTR_ERR(replace_prog);
+		}
+	}
+
+	ret = cgroup_bpf_attach(cgrp, prog, replace_prog, attr->attach_type,
 				attr->attach_flags);
+
+	if (replace_prog)
+		bpf_prog_put(replace_prog);
 	cgroup_put(cgrp);
 	return ret;
 }
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index af6b738cf435..29d47aae0dd1 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -222,8 +222,6 @@ struct bpf_prog *bpf_prog_realloc(struct bpf_prog *fp_old, unsigned int size,
 	u32 pages, delta;
 	int ret;
 
-	BUG_ON(fp_old == NULL);
-
 	size = round_up(size, PAGE_SIZE);
 	pages = size / PAGE_SIZE;
 	if (pages <= fp_old->pages)
@@ -520,9 +518,9 @@ void bpf_prog_kallsyms_del_all(struct bpf_prog *fp)
 
 #ifdef CONFIG_BPF_JIT
 /* All BPF JIT sysctl knobs here. */
-int bpf_jit_enable   __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_ALWAYS_ON);
+int bpf_jit_enable   __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_DEFAULT_ON);
+int bpf_jit_kallsyms __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_DEFAULT_ON);
 int bpf_jit_harden   __read_mostly;
-int bpf_jit_kallsyms __read_mostly;
 long bpf_jit_limit   __read_mostly;
 
 static __always_inline void
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index ef49e17ae47c..70f71b154fa5 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -72,17 +72,18 @@ struct bpf_cpu_map {
 	struct bpf_map map;
 	/* Below members specific for map type */
 	struct bpf_cpu_map_entry **cpu_map;
-	struct list_head __percpu *flush_list;
 };
 
-static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx);
+static DEFINE_PER_CPU(struct list_head, cpu_map_flush_list);
+
+static int bq_flush_to_queue(struct xdp_bulk_queue *bq);
 
 static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 {
 	struct bpf_cpu_map *cmap;
 	int err = -ENOMEM;
-	int ret, cpu;
 	u64 cost;
+	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return ERR_PTR(-EPERM);
@@ -106,7 +107,6 @@ static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 
 	/* make sure page count doesn't overflow */
 	cost = (u64) cmap->map.max_entries * sizeof(struct bpf_cpu_map_entry *);
-	cost += sizeof(struct list_head) * num_possible_cpus();
 
 	/* Notice returns -EPERM on if map size is larger than memlock limit */
 	ret = bpf_map_charge_init(&cmap->map.memory, cost);
@@ -115,23 +115,14 @@ static struct bpf_map *cpu_map_alloc(union bpf_attr *attr)
 		goto free_cmap;
 	}
 
-	cmap->flush_list = alloc_percpu(struct list_head);
-	if (!cmap->flush_list)
-		goto free_charge;
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(per_cpu_ptr(cmap->flush_list, cpu));
-
 	/* Alloc array for possible remote "destination" CPUs */
 	cmap->cpu_map = bpf_map_area_alloc(cmap->map.max_entries *
 					   sizeof(struct bpf_cpu_map_entry *),
 					   cmap->map.numa_node);
 	if (!cmap->cpu_map)
-		goto free_percpu;
+		goto free_charge;
 
 	return &cmap->map;
-free_percpu:
-	free_percpu(cmap->flush_list);
 free_charge:
 	bpf_map_charge_finish(&cmap->map.memory);
 free_cmap:
@@ -399,22 +390,14 @@ free_rcu:
 static void __cpu_map_entry_free(struct rcu_head *rcu)
 {
 	struct bpf_cpu_map_entry *rcpu;
-	int cpu;
 
 	/* This cpu_map_entry have been disconnected from map and one
-	 * RCU graze-period have elapsed.  Thus, XDP cannot queue any
+	 * RCU grace-period have elapsed.  Thus, XDP cannot queue any
 	 * new packets and cannot change/set flush_needed that can
 	 * find this entry.
 	 */
 	rcpu = container_of(rcu, struct bpf_cpu_map_entry, rcu);
 
-	/* Flush remaining packets in percpu bulkq */
-	for_each_online_cpu(cpu) {
-		struct xdp_bulk_queue *bq = per_cpu_ptr(rcpu->bulkq, cpu);
-
-		/* No concurrent bq_enqueue can run at this point */
-		bq_flush_to_queue(bq, false);
-	}
 	free_percpu(rcpu->bulkq);
 	/* Cannot kthread_stop() here, last put free rcpu resources */
 	put_cpu_map_entry(rcpu);
@@ -436,7 +419,7 @@ static void __cpu_map_entry_free(struct rcu_head *rcu)
  * percpu bulkq to queue.  Due to caller map_delete_elem() disable
  * preemption, cannot call kthread_stop() to make sure queue is empty.
  * Instead a work_queue is started for stopping kthread,
- * cpu_map_kthread_stop, which waits for an RCU graze period before
+ * cpu_map_kthread_stop, which waits for an RCU grace period before
  * stopping kthread, emptying the queue.
  */
 static void __cpu_map_entry_replace(struct bpf_cpu_map *cmap,
@@ -507,7 +490,6 @@ static int cpu_map_update_elem(struct bpf_map *map, void *key, void *value,
 static void cpu_map_free(struct bpf_map *map)
 {
 	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
-	int cpu;
 	u32 i;
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
@@ -522,18 +504,6 @@ static void cpu_map_free(struct bpf_map *map)
 	bpf_clear_redirect_map(map);
 	synchronize_rcu();
 
-	/* To ensure all pending flush operations have completed wait for flush
-	 * list be empty on _all_ cpus. Because the above synchronize_rcu()
-	 * ensures the map is disconnected from the program we can assume no new
-	 * items will be added to the list.
-	 */
-	for_each_online_cpu(cpu) {
-		struct list_head *flush_list = per_cpu_ptr(cmap->flush_list, cpu);
-
-		while (!list_empty(flush_list))
-			cond_resched();
-	}
-
 	/* For cpu_map the remote CPUs can still be using the entries
 	 * (struct bpf_cpu_map_entry).
 	 */
@@ -544,10 +514,9 @@ static void cpu_map_free(struct bpf_map *map)
 		if (!rcpu)
 			continue;
 
-		/* bq flush and cleanup happens after RCU graze-period */
+		/* bq flush and cleanup happens after RCU grace-period */
 		__cpu_map_entry_replace(cmap, i, NULL); /* call_rcu */
 	}
-	free_percpu(cmap->flush_list);
 	bpf_map_area_free(cmap->cpu_map);
 	kfree(cmap);
 }
@@ -599,7 +568,7 @@ const struct bpf_map_ops cpu_map_ops = {
 	.map_check_btf		= map_check_no_btf,
 };
 
-static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx)
+static int bq_flush_to_queue(struct xdp_bulk_queue *bq)
 {
 	struct bpf_cpu_map_entry *rcpu = bq->obj;
 	unsigned int processed = 0, drops = 0;
@@ -620,10 +589,7 @@ static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx)
 		err = __ptr_ring_produce(q, xdpf);
 		if (err) {
 			drops++;
-			if (likely(in_napi_ctx))
-				xdp_return_frame_rx_napi(xdpf);
-			else
-				xdp_return_frame(xdpf);
+			xdp_return_frame_rx_napi(xdpf);
 		}
 		processed++;
 	}
@@ -642,11 +608,11 @@ static int bq_flush_to_queue(struct xdp_bulk_queue *bq, bool in_napi_ctx)
  */
 static int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
 {
-	struct list_head *flush_list = this_cpu_ptr(rcpu->cmap->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&cpu_map_flush_list);
 	struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
 	if (unlikely(bq->count == CPU_MAP_BULK_SIZE))
-		bq_flush_to_queue(bq, true);
+		bq_flush_to_queue(bq);
 
 	/* Notice, xdp_buff/page MUST be queued here, long enough for
 	 * driver to code invoking us to finished, due to driver
@@ -681,16 +647,26 @@ int cpu_map_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_buff *xdp,
 	return 0;
 }
 
-void __cpu_map_flush(struct bpf_map *map)
+void __cpu_map_flush(void)
 {
-	struct bpf_cpu_map *cmap = container_of(map, struct bpf_cpu_map, map);
-	struct list_head *flush_list = this_cpu_ptr(cmap->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&cpu_map_flush_list);
 	struct xdp_bulk_queue *bq, *tmp;
 
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node) {
-		bq_flush_to_queue(bq, true);
+		bq_flush_to_queue(bq);
 
 		/* If already running, costs spin_lock_irqsave + smb_mb */
 		wake_up_process(bq->obj->kthread);
 	}
 }
+
+static int __init cpu_map_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		INIT_LIST_HEAD(&per_cpu(cpu_map_flush_list, cpu));
+	return 0;
+}
+
+subsys_initcall(cpu_map_init);
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 3d3d61b5985b..da9c832fc5c8 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -75,7 +75,6 @@ struct bpf_dtab_netdev {
 struct bpf_dtab {
 	struct bpf_map map;
 	struct bpf_dtab_netdev **netdev_map; /* DEVMAP type only */
-	struct list_head __percpu *flush_list;
 	struct list_head list;
 
 	/* these are only used for DEVMAP_HASH type maps */
@@ -85,6 +84,7 @@ struct bpf_dtab {
 	u32 n_buckets;
 };
 
+static DEFINE_PER_CPU(struct list_head, dev_map_flush_list);
 static DEFINE_SPINLOCK(dev_map_lock);
 static LIST_HEAD(dev_map_list);
 
@@ -109,8 +109,8 @@ static inline struct hlist_head *dev_map_index_hash(struct bpf_dtab *dtab,
 
 static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 {
-	int err, cpu;
-	u64 cost;
+	u64 cost = 0;
+	int err;
 
 	/* check sanity of attributes */
 	if (attr->max_entries == 0 || attr->key_size != 4 ||
@@ -125,9 +125,6 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 
 	bpf_map_init_from_attr(&dtab->map, attr);
 
-	/* make sure page count doesn't overflow */
-	cost = (u64) sizeof(struct list_head) * num_possible_cpus();
-
 	if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		dtab->n_buckets = roundup_pow_of_two(dtab->map.max_entries);
 
@@ -143,17 +140,10 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 	if (err)
 		return -EINVAL;
 
-	dtab->flush_list = alloc_percpu(struct list_head);
-	if (!dtab->flush_list)
-		goto free_charge;
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(per_cpu_ptr(dtab->flush_list, cpu));
-
 	if (attr->map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		dtab->dev_index_head = dev_map_create_hash(dtab->n_buckets);
 		if (!dtab->dev_index_head)
-			goto free_percpu;
+			goto free_charge;
 
 		spin_lock_init(&dtab->index_lock);
 	} else {
@@ -161,13 +151,11 @@ static int dev_map_init_map(struct bpf_dtab *dtab, union bpf_attr *attr)
 						      sizeof(struct bpf_dtab_netdev *),
 						      dtab->map.numa_node);
 		if (!dtab->netdev_map)
-			goto free_percpu;
+			goto free_charge;
 	}
 
 	return 0;
 
-free_percpu:
-	free_percpu(dtab->flush_list);
 free_charge:
 	bpf_map_charge_finish(&dtab->map.memory);
 	return -ENOMEM;
@@ -201,7 +189,7 @@ static struct bpf_map *dev_map_alloc(union bpf_attr *attr)
 static void dev_map_free(struct bpf_map *map)
 {
 	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
-	int i, cpu;
+	int i;
 
 	/* At this point bpf_prog->aux->refcnt == 0 and this map->refcnt == 0,
 	 * so the programs (can be more than one that used this map) were
@@ -221,18 +209,6 @@ static void dev_map_free(struct bpf_map *map)
 	/* Make sure prior __dev_map_entry_free() have completed. */
 	rcu_barrier();
 
-	/* To ensure all pending flush operations have completed wait for flush
-	 * list to empty on _all_ cpus.
-	 * Because the above synchronize_rcu() ensures the map is disconnected
-	 * from the program we can assume no new items will be added.
-	 */
-	for_each_online_cpu(cpu) {
-		struct list_head *flush_list = per_cpu_ptr(dtab->flush_list, cpu);
-
-		while (!list_empty(flush_list))
-			cond_resched();
-	}
-
 	if (dtab->map.map_type == BPF_MAP_TYPE_DEVMAP_HASH) {
 		for (i = 0; i < dtab->n_buckets; i++) {
 			struct bpf_dtab_netdev *dev;
@@ -266,7 +242,6 @@ static void dev_map_free(struct bpf_map *map)
 		bpf_map_area_free(dtab->netdev_map);
 	}
 
-	free_percpu(dtab->flush_list);
 	kfree(dtab);
 }
 
@@ -345,8 +320,7 @@ static int dev_map_hash_get_next_key(struct bpf_map *map, void *key,
 	return -ENOENT;
 }
 
-static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags,
-		       bool in_napi_ctx)
+static int bq_xmit_all(struct xdp_bulk_queue *bq, u32 flags)
 {
 	struct bpf_dtab_netdev *obj = bq->obj;
 	struct net_device *dev = obj->dev;
@@ -384,11 +358,7 @@ error:
 	for (i = 0; i < bq->count; i++) {
 		struct xdp_frame *xdpf = bq->q[i];
 
-		/* RX path under NAPI protection, can return frames faster */
-		if (likely(in_napi_ctx))
-			xdp_return_frame_rx_napi(xdpf);
-		else
-			xdp_return_frame(xdpf);
+		xdp_return_frame_rx_napi(xdpf);
 		drops++;
 	}
 	goto out;
@@ -401,15 +371,14 @@ error:
  * net device can be torn down. On devmap tear down we ensure the flush list
  * is empty before completing to ensure all flush operations have completed.
  */
-void __dev_map_flush(struct bpf_map *map)
+void __dev_map_flush(void)
 {
-	struct bpf_dtab *dtab = container_of(map, struct bpf_dtab, map);
-	struct list_head *flush_list = this_cpu_ptr(dtab->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
 	struct xdp_bulk_queue *bq, *tmp;
 
 	rcu_read_lock();
 	list_for_each_entry_safe(bq, tmp, flush_list, flush_node)
-		bq_xmit_all(bq, XDP_XMIT_FLUSH, true);
+		bq_xmit_all(bq, XDP_XMIT_FLUSH);
 	rcu_read_unlock();
 }
 
@@ -436,11 +405,11 @@ static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
 		      struct net_device *dev_rx)
 
 {
-	struct list_head *flush_list = this_cpu_ptr(obj->dtab->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&dev_map_flush_list);
 	struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
 
 	if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
-		bq_xmit_all(bq, 0, true);
+		bq_xmit_all(bq, 0);
 
 	/* Ingress dev_rx will be the same for all xdp_frame's in
 	 * bulk_queue, because bq stored per-CPU and must be flushed
@@ -509,27 +478,11 @@ static void *dev_map_hash_lookup_elem(struct bpf_map *map, void *key)
 	return dev ? &dev->ifindex : NULL;
 }
 
-static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
-{
-	if (dev->dev->netdev_ops->ndo_xdp_xmit) {
-		struct xdp_bulk_queue *bq;
-		int cpu;
-
-		rcu_read_lock();
-		for_each_online_cpu(cpu) {
-			bq = per_cpu_ptr(dev->bulkq, cpu);
-			bq_xmit_all(bq, XDP_XMIT_FLUSH, false);
-		}
-		rcu_read_unlock();
-	}
-}
-
 static void __dev_map_entry_free(struct rcu_head *rcu)
 {
 	struct bpf_dtab_netdev *dev;
 
 	dev = container_of(rcu, struct bpf_dtab_netdev, rcu);
-	dev_map_flush_old(dev);
 	free_percpu(dev->bulkq);
 	dev_put(dev->dev);
 	kfree(dev);
@@ -810,10 +763,15 @@ static struct notifier_block dev_map_notifier = {
 
 static int __init dev_map_init(void)
 {
+	int cpu;
+
 	/* Assure tracepoint shadow struct _bpf_dtab_netdev is in sync */
 	BUILD_BUG_ON(offsetof(struct bpf_dtab_netdev, dev) !=
 		     offsetof(struct _bpf_dtab_netdev, dev));
 	register_netdevice_notifier(&dev_map_notifier);
+
+	for_each_possible_cpu(cpu)
+		INIT_LIST_HEAD(&per_cpu(dev_map_flush_list, cpu));
 	return 0;
 }
 
diff --git a/kernel/bpf/dispatcher.c b/kernel/bpf/dispatcher.c
new file mode 100644
index 000000000000..204ee61a3904
--- /dev/null
+++ b/kernel/bpf/dispatcher.c
@@ -0,0 +1,158 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2019 Intel Corporation. */
+
+#include <linux/hash.h>
+#include <linux/bpf.h>
+#include <linux/filter.h>
+
+/* The BPF dispatcher is a multiway branch code generator. The
+ * dispatcher is a mechanism to avoid the performance penalty of an
+ * indirect call, which is expensive when retpolines are enabled. A
+ * dispatch client registers a BPF program into the dispatcher, and if
+ * there is available room in the dispatcher a direct call to the BPF
+ * program will be generated. All calls to the BPF programs called via
+ * the dispatcher will then be a direct call, instead of an
+ * indirect. The dispatcher hijacks a trampoline function it via the
+ * __fentry__ of the trampoline. The trampoline function has the
+ * following signature:
+ *
+ * unsigned int trampoline(const void *ctx, const struct bpf_insn *insnsi,
+ *                         unsigned int (*bpf_func)(const void *,
+ *                                                  const struct bpf_insn *));
+ */
+
+static struct bpf_dispatcher_prog *bpf_dispatcher_find_prog(
+	struct bpf_dispatcher *d, struct bpf_prog *prog)
+{
+	int i;
+
+	for (i = 0; i < BPF_DISPATCHER_MAX; i++) {
+		if (prog == d->progs[i].prog)
+			return &d->progs[i];
+	}
+	return NULL;
+}
+
+static struct bpf_dispatcher_prog *bpf_dispatcher_find_free(
+	struct bpf_dispatcher *d)
+{
+	return bpf_dispatcher_find_prog(d, NULL);
+}
+
+static bool bpf_dispatcher_add_prog(struct bpf_dispatcher *d,
+				    struct bpf_prog *prog)
+{
+	struct bpf_dispatcher_prog *entry;
+
+	if (!prog)
+		return false;
+
+	entry = bpf_dispatcher_find_prog(d, prog);
+	if (entry) {
+		refcount_inc(&entry->users);
+		return false;
+	}
+
+	entry = bpf_dispatcher_find_free(d);
+	if (!entry)
+		return false;
+
+	bpf_prog_inc(prog);
+	entry->prog = prog;
+	refcount_set(&entry->users, 1);
+	d->num_progs++;
+	return true;
+}
+
+static bool bpf_dispatcher_remove_prog(struct bpf_dispatcher *d,
+				       struct bpf_prog *prog)
+{
+	struct bpf_dispatcher_prog *entry;
+
+	if (!prog)
+		return false;
+
+	entry = bpf_dispatcher_find_prog(d, prog);
+	if (!entry)
+		return false;
+
+	if (refcount_dec_and_test(&entry->users)) {
+		entry->prog = NULL;
+		bpf_prog_put(prog);
+		d->num_progs--;
+		return true;
+	}
+	return false;
+}
+
+int __weak arch_prepare_bpf_dispatcher(void *image, s64 *funcs, int num_funcs)
+{
+	return -ENOTSUPP;
+}
+
+static int bpf_dispatcher_prepare(struct bpf_dispatcher *d, void *image)
+{
+	s64 ips[BPF_DISPATCHER_MAX] = {}, *ipsp = &ips[0];
+	int i;
+
+	for (i = 0; i < BPF_DISPATCHER_MAX; i++) {
+		if (d->progs[i].prog)
+			*ipsp++ = (s64)(uintptr_t)d->progs[i].prog->bpf_func;
+	}
+	return arch_prepare_bpf_dispatcher(image, &ips[0], d->num_progs);
+}
+
+static void bpf_dispatcher_update(struct bpf_dispatcher *d, int prev_num_progs)
+{
+	void *old, *new;
+	u32 noff;
+	int err;
+
+	if (!prev_num_progs) {
+		old = NULL;
+		noff = 0;
+	} else {
+		old = d->image + d->image_off;
+		noff = d->image_off ^ (PAGE_SIZE / 2);
+	}
+
+	new = d->num_progs ? d->image + noff : NULL;
+	if (new) {
+		if (bpf_dispatcher_prepare(d, new))
+			return;
+	}
+
+	err = bpf_arch_text_poke(d->func, BPF_MOD_JUMP, old, new);
+	if (err || !new)
+		return;
+
+	d->image_off = noff;
+}
+
+void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
+				struct bpf_prog *to)
+{
+	bool changed = false;
+	int prev_num_progs;
+
+	if (from == to)
+		return;
+
+	mutex_lock(&d->mutex);
+	if (!d->image) {
+		d->image = bpf_jit_alloc_exec_page();
+		if (!d->image)
+			goto out;
+	}
+
+	prev_num_progs = d->num_progs;
+	changed |= bpf_dispatcher_remove_prog(d, from);
+	changed |= bpf_dispatcher_add_prog(d, to);
+
+	if (!changed)
+		goto out;
+
+	bpf_dispatcher_update(d, prev_num_progs);
+out:
+	mutex_unlock(&d->mutex);
+}
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index e3461ec59570..81ee8595dfee 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -23,6 +23,7 @@
 #include <linux/timekeeping.h>
 #include <linux/ctype.h>
 #include <linux/nospec.h>
+#include <linux/audit.h>
 #include <uapi/linux/btf.h>
 
 #define IS_FD_ARRAY(map) ((map)->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || \
@@ -1306,6 +1307,36 @@ static int find_prog_type(enum bpf_prog_type type, struct bpf_prog *prog)
 	return 0;
 }
 
+enum bpf_audit {
+	BPF_AUDIT_LOAD,
+	BPF_AUDIT_UNLOAD,
+	BPF_AUDIT_MAX,
+};
+
+static const char * const bpf_audit_str[BPF_AUDIT_MAX] = {
+	[BPF_AUDIT_LOAD]   = "LOAD",
+	[BPF_AUDIT_UNLOAD] = "UNLOAD",
+};
+
+static void bpf_audit_prog(const struct bpf_prog *prog, unsigned int op)
+{
+	struct audit_context *ctx = NULL;
+	struct audit_buffer *ab;
+
+	if (WARN_ON_ONCE(op >= BPF_AUDIT_MAX))
+		return;
+	if (audit_enabled == AUDIT_OFF)
+		return;
+	if (op == BPF_AUDIT_LOAD)
+		ctx = audit_context();
+	ab = audit_log_start(ctx, GFP_ATOMIC, AUDIT_BPF);
+	if (unlikely(!ab))
+		return;
+	audit_log_format(ab, "prog-id=%u op=%s",
+			 prog->aux->id, bpf_audit_str[op]);
+	audit_log_end(ab);
+}
+
 int __bpf_prog_charge(struct user_struct *user, u32 pages)
 {
 	unsigned long memlock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
@@ -1421,6 +1452,7 @@ static void __bpf_prog_put(struct bpf_prog *prog, bool do_idr_lock)
 {
 	if (atomic64_dec_and_test(&prog->aux->refcnt)) {
 		perf_event_bpf_event(prog, PERF_BPF_EVENT_PROG_UNLOAD, 0);
+		bpf_audit_prog(prog, BPF_AUDIT_UNLOAD);
 		/* bpf_prog_free_id() must be called first */
 		bpf_prog_free_id(prog, do_idr_lock);
 		__bpf_prog_put_noref(prog, true);
@@ -1830,6 +1862,7 @@ static int bpf_prog_load(union bpf_attr *attr, union bpf_attr __user *uattr)
 	 */
 	bpf_prog_kallsyms_add(prog);
 	perf_event_bpf_event(prog, PERF_BPF_EVENT_PROG_LOAD, 0);
+	bpf_audit_prog(prog, BPF_AUDIT_LOAD);
 
 	err = bpf_prog_new_fd(prog);
 	if (err < 0)
@@ -2040,10 +2073,10 @@ static int bpf_prog_attach_check_attach_type(const struct bpf_prog *prog,
 	}
 }
 
-#define BPF_PROG_ATTACH_LAST_FIELD attach_flags
+#define BPF_PROG_ATTACH_LAST_FIELD replace_bpf_fd
 
 #define BPF_F_ATTACH_MASK \
-	(BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI)
+	(BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI | BPF_F_REPLACE)
 
 static int bpf_prog_attach(const union bpf_attr *attr)
 {
@@ -2305,17 +2338,12 @@ static int bpf_obj_get_next_id(const union bpf_attr *attr,
 
 #define BPF_PROG_GET_FD_BY_ID_LAST_FIELD prog_id
 
-static int bpf_prog_get_fd_by_id(const union bpf_attr *attr)
+struct bpf_prog *bpf_prog_by_id(u32 id)
 {
 	struct bpf_prog *prog;
-	u32 id = attr->prog_id;
-	int fd;
-
-	if (CHECK_ATTR(BPF_PROG_GET_FD_BY_ID))
-		return -EINVAL;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
+	if (!id)
+		return ERR_PTR(-ENOENT);
 
 	spin_lock_bh(&prog_idr_lock);
 	prog = idr_find(&prog_idr, id);
@@ -2324,7 +2352,22 @@ static int bpf_prog_get_fd_by_id(const union bpf_attr *attr)
 	else
 		prog = ERR_PTR(-ENOENT);
 	spin_unlock_bh(&prog_idr_lock);
+	return prog;
+}
+
+static int bpf_prog_get_fd_by_id(const union bpf_attr *attr)
+{
+	struct bpf_prog *prog;
+	u32 id = attr->prog_id;
+	int fd;
+
+	if (CHECK_ATTR(BPF_PROG_GET_FD_BY_ID))
+		return -EINVAL;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
 
+	prog = bpf_prog_by_id(id);
 	if (IS_ERR(prog))
 		return PTR_ERR(prog);
 
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 23b0d5cfd47e..505f4e4b31d2 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -14,6 +14,22 @@ static struct hlist_head trampoline_table[TRAMPOLINE_TABLE_SIZE];
 /* serializes access to trampoline_table */
 static DEFINE_MUTEX(trampoline_mutex);
 
+void *bpf_jit_alloc_exec_page(void)
+{
+	void *image;
+
+	image = bpf_jit_alloc_exec(PAGE_SIZE);
+	if (!image)
+		return NULL;
+
+	set_vm_flush_reset_perms(image);
+	/* Keep image as writeable. The alternative is to keep flipping ro/rw
+	 * everytime new program is attached or detached.
+	 */
+	set_memory_x((long)image, 1);
+	return image;
+}
+
 struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 {
 	struct bpf_trampoline *tr;
@@ -34,7 +50,7 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 		goto out;
 
 	/* is_root was checked earlier. No need for bpf_jit_charge_modmem() */
-	image = bpf_jit_alloc_exec(PAGE_SIZE);
+	image = bpf_jit_alloc_exec_page();
 	if (!image) {
 		kfree(tr);
 		tr = NULL;
@@ -48,12 +64,6 @@ struct bpf_trampoline *bpf_trampoline_lookup(u64 key)
 	mutex_init(&tr->mutex);
 	for (i = 0; i < BPF_TRAMP_MAX; i++)
 		INIT_HLIST_HEAD(&tr->progs_hlist[i]);
-
-	set_vm_flush_reset_perms(image);
-	/* Keep image as writeable. The alternative is to keep flipping ro/rw
-	 * everytime new program is attached or detached.
-	 */
-	set_memory_x((long)image, 1);
 	tr->image = image;
 out:
 	mutex_unlock(&trampoline_mutex);
diff --git a/kernel/bpf/xskmap.c b/kernel/bpf/xskmap.c
index 90c4fce1c981..2cc5c8f4c800 100644
--- a/kernel/bpf/xskmap.c
+++ b/kernel/bpf/xskmap.c
@@ -72,9 +72,9 @@ static void xsk_map_sock_delete(struct xdp_sock *xs,
 static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 {
 	struct bpf_map_memory mem;
-	int cpu, err, numa_node;
+	int err, numa_node;
 	struct xsk_map *m;
-	u64 cost, size;
+	u64 size;
 
 	if (!capable(CAP_NET_ADMIN))
 		return ERR_PTR(-EPERM);
@@ -86,9 +86,8 @@ static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 
 	numa_node = bpf_map_attr_numa_node(attr);
 	size = struct_size(m, xsk_map, attr->max_entries);
-	cost = size + array_size(sizeof(*m->flush_list), num_possible_cpus());
 
-	err = bpf_map_charge_init(&mem, cost);
+	err = bpf_map_charge_init(&mem, size);
 	if (err < 0)
 		return ERR_PTR(err);
 
@@ -102,16 +101,6 @@ static struct bpf_map *xsk_map_alloc(union bpf_attr *attr)
 	bpf_map_charge_move(&m->map.memory, &mem);
 	spin_lock_init(&m->lock);
 
-	m->flush_list = alloc_percpu(struct list_head);
-	if (!m->flush_list) {
-		bpf_map_charge_finish(&m->map.memory);
-		bpf_map_area_free(m);
-		return ERR_PTR(-ENOMEM);
-	}
-
-	for_each_possible_cpu(cpu)
-		INIT_LIST_HEAD(per_cpu_ptr(m->flush_list, cpu));
-
 	return &m->map;
 }
 
@@ -121,7 +110,6 @@ static void xsk_map_free(struct bpf_map *map)
 
 	bpf_clear_redirect_map(map);
 	synchronize_net();
-	free_percpu(m->flush_list);
 	bpf_map_area_free(m);
 }
 
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 735af8f15f95..725365df066d 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -6288,12 +6288,13 @@ void cgroup_sk_free(struct sock_cgroup_data *skcd)
 
 #ifdef CONFIG_CGROUP_BPF
 int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
-		      enum bpf_attach_type type, u32 flags)
+		      struct bpf_prog *replace_prog, enum bpf_attach_type type,
+		      u32 flags)
 {
 	int ret;
 
 	mutex_lock(&cgroup_mutex);
-	ret = __cgroup_bpf_attach(cgrp, prog, type, flags);
+	ret = __cgroup_bpf_attach(cgrp, prog, replace_prog, type, flags);
 	mutex_unlock(&cgroup_mutex);
 	return ret;
 }
diff --git a/net/bpf/test_run.c b/net/bpf/test_run.c
index f79205d4444f..d555c0d8657d 100644
--- a/net/bpf/test_run.c
+++ b/net/bpf/test_run.c
@@ -15,7 +15,7 @@
 #include <trace/events/bpf_test_run.h>
 
 static int bpf_test_run(struct bpf_prog *prog, void *ctx, u32 repeat,
-			u32 *retval, u32 *time)
+			u32 *retval, u32 *time, bool xdp)
 {
 	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE] = { NULL };
 	enum bpf_cgroup_storage_type stype;
@@ -41,7 +41,11 @@ static int bpf_test_run(struct bpf_prog *prog, void *ctx, u32 repeat,
 	time_start = ktime_get_ns();
 	for (i = 0; i < repeat; i++) {
 		bpf_cgroup_storage_set(storage);
-		*retval = BPF_PROG_RUN(prog, ctx);
+
+		if (xdp)
+			*retval = bpf_prog_run_xdp(prog, ctx);
+		else
+			*retval = BPF_PROG_RUN(prog, ctx);
 
 		if (signal_pending(current)) {
 			ret = -EINTR;
@@ -247,34 +251,53 @@ static int convert___skb_to_skb(struct sk_buff *skb, struct __sk_buff *__skb)
 		return 0;
 
 	/* make sure the fields we don't use are zeroed */
-	if (!range_is_zero(__skb, 0, offsetof(struct __sk_buff, priority)))
+	if (!range_is_zero(__skb, 0, offsetof(struct __sk_buff, mark)))
+		return -EINVAL;
+
+	/* mark is allowed */
+
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, mark),
+			   offsetof(struct __sk_buff, priority)))
 		return -EINVAL;
 
 	/* priority is allowed */
 
-	if (!range_is_zero(__skb, offsetof(struct __sk_buff, priority) +
-			   sizeof_field(struct __sk_buff, priority),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, priority),
 			   offsetof(struct __sk_buff, cb)))
 		return -EINVAL;
 
 	/* cb is allowed */
 
-	if (!range_is_zero(__skb, offsetof(struct __sk_buff, cb) +
-			   sizeof_field(struct __sk_buff, cb),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, cb),
 			   offsetof(struct __sk_buff, tstamp)))
 		return -EINVAL;
 
 	/* tstamp is allowed */
+	/* wire_len is allowed */
+	/* gso_segs is allowed */
 
-	if (!range_is_zero(__skb, offsetof(struct __sk_buff, tstamp) +
-			   sizeof_field(struct __sk_buff, tstamp),
+	if (!range_is_zero(__skb, offsetofend(struct __sk_buff, gso_segs),
 			   sizeof(struct __sk_buff)))
 		return -EINVAL;
 
+	skb->mark = __skb->mark;
 	skb->priority = __skb->priority;
 	skb->tstamp = __skb->tstamp;
 	memcpy(&cb->data, __skb->cb, QDISC_CB_PRIV_LEN);
 
+	if (__skb->wire_len == 0) {
+		cb->pkt_len = skb->len;
+	} else {
+		if (__skb->wire_len < skb->len ||
+		    __skb->wire_len > GSO_MAX_SIZE)
+			return -EINVAL;
+		cb->pkt_len = __skb->wire_len;
+	}
+
+	if (__skb->gso_segs > GSO_MAX_SEGS)
+		return -EINVAL;
+	skb_shinfo(skb)->gso_segs = __skb->gso_segs;
+
 	return 0;
 }
 
@@ -285,9 +308,12 @@ static void convert_skb_to___skb(struct sk_buff *skb, struct __sk_buff *__skb)
 	if (!__skb)
 		return;
 
+	__skb->mark = skb->mark;
 	__skb->priority = skb->priority;
 	__skb->tstamp = skb->tstamp;
 	memcpy(__skb->cb, &cb->data, QDISC_CB_PRIV_LEN);
+	__skb->wire_len = cb->pkt_len;
+	__skb->gso_segs = skb_shinfo(skb)->gso_segs;
 }
 
 int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
@@ -359,7 +385,7 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
 	ret = convert___skb_to_skb(skb, ctx);
 	if (ret)
 		goto out;
-	ret = bpf_test_run(prog, skb, repeat, &retval, &duration);
+	ret = bpf_test_run(prog, skb, repeat, &retval, &duration, false);
 	if (ret)
 		goto out;
 	if (!is_l2) {
@@ -416,8 +442,8 @@ int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr,
 
 	rxqueue = __netif_get_rx_queue(current->nsproxy->net_ns->loopback_dev, 0);
 	xdp.rxq = &rxqueue->xdp_rxq;
-
-	ret = bpf_test_run(prog, &xdp, repeat, &retval, &duration);
+	bpf_prog_change_xdp(NULL, prog);
+	ret = bpf_test_run(prog, &xdp, repeat, &retval, &duration, true);
 	if (ret)
 		goto out;
 	if (xdp.data != data + XDP_PACKET_HEADROOM + NET_IP_ALIGN ||
@@ -425,6 +451,7 @@ int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr,
 		size = xdp.data_end - xdp.data;
 	ret = bpf_test_finish(kattr, uattr, xdp.data, size, retval, duration);
 out:
+	bpf_prog_change_xdp(prog, NULL);
 	kfree(data);
 	return ret;
 }
@@ -437,8 +464,7 @@ static int verify_user_bpf_flow_keys(struct bpf_flow_keys *ctx)
 
 	/* flags is allowed */
 
-	if (!range_is_zero(ctx, offsetof(struct bpf_flow_keys, flags) +
-			   sizeof_field(struct bpf_flow_keys, flags),
+	if (!range_is_zero(ctx, offsetofend(struct bpf_flow_keys, flags),
 			   sizeof(struct bpf_flow_keys)))
 		return -EINVAL;
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 0ad39c87b7fd..f1b8afcfbc0f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -8542,7 +8542,17 @@ static int dev_xdp_install(struct net_device *dev, bpf_op_t bpf_op,
 			   struct netlink_ext_ack *extack, u32 flags,
 			   struct bpf_prog *prog)
 {
+	bool non_hw = !(flags & XDP_FLAGS_HW_MODE);
+	struct bpf_prog *prev_prog = NULL;
 	struct netdev_bpf xdp;
+	int err;
+
+	if (non_hw) {
+		prev_prog = bpf_prog_by_id(__dev_xdp_query(dev, bpf_op,
+							   XDP_QUERY_PROG));
+		if (IS_ERR(prev_prog))
+			prev_prog = NULL;
+	}
 
 	memset(&xdp, 0, sizeof(xdp));
 	if (flags & XDP_FLAGS_HW_MODE)
@@ -8553,7 +8563,14 @@ static int dev_xdp_install(struct net_device *dev, bpf_op_t bpf_op,
 	xdp.flags = flags;
 	xdp.prog = prog;
 
-	return bpf_op(dev, &xdp);
+	err = bpf_op(dev, &xdp);
+	if (!err && non_hw)
+		bpf_prog_change_xdp(prev_prog, prog);
+
+	if (prev_prog)
+		bpf_prog_put(prev_prog);
+
+	return err;
 }
 
 static void dev_xdp_uninstall(struct net_device *dev)
diff --git a/net/core/filter.c b/net/core/filter.c
index 28b3c258188c..42fd17c48c5f 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3511,36 +3511,16 @@ err:
 }
 
 static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
-			    struct bpf_map *map,
-			    struct xdp_buff *xdp,
-			    u32 index)
+			    struct bpf_map *map, struct xdp_buff *xdp)
 {
-	int err;
-
 	switch (map->map_type) {
 	case BPF_MAP_TYPE_DEVMAP:
-	case BPF_MAP_TYPE_DEVMAP_HASH: {
-		struct bpf_dtab_netdev *dst = fwd;
-
-		err = dev_map_enqueue(dst, xdp, dev_rx);
-		if (unlikely(err))
-			return err;
-		break;
-	}
-	case BPF_MAP_TYPE_CPUMAP: {
-		struct bpf_cpu_map_entry *rcpu = fwd;
-
-		err = cpu_map_enqueue(rcpu, xdp, dev_rx);
-		if (unlikely(err))
-			return err;
-		break;
-	}
-	case BPF_MAP_TYPE_XSKMAP: {
-		struct xdp_sock *xs = fwd;
-
-		err = __xsk_map_redirect(map, xdp, xs);
-		return err;
-	}
+	case BPF_MAP_TYPE_DEVMAP_HASH:
+		return dev_map_enqueue(fwd, xdp, dev_rx);
+	case BPF_MAP_TYPE_CPUMAP:
+		return cpu_map_enqueue(fwd, xdp, dev_rx);
+	case BPF_MAP_TYPE_XSKMAP:
+		return __xsk_map_redirect(fwd, xdp);
 	default:
 		break;
 	}
@@ -3549,26 +3529,9 @@ static int __bpf_tx_xdp_map(struct net_device *dev_rx, void *fwd,
 
 void xdp_do_flush_map(void)
 {
-	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
-	struct bpf_map *map = ri->map_to_flush;
-
-	ri->map_to_flush = NULL;
-	if (map) {
-		switch (map->map_type) {
-		case BPF_MAP_TYPE_DEVMAP:
-		case BPF_MAP_TYPE_DEVMAP_HASH:
-			__dev_map_flush(map);
-			break;
-		case BPF_MAP_TYPE_CPUMAP:
-			__cpu_map_flush(map);
-			break;
-		case BPF_MAP_TYPE_XSKMAP:
-			__xsk_map_flush(map);
-			break;
-		default:
-			break;
-		}
-	}
+	__dev_map_flush();
+	__cpu_map_flush();
+	__xsk_map_flush();
 }
 EXPORT_SYMBOL_GPL(xdp_do_flush_map);
 
@@ -3617,14 +3580,10 @@ static int xdp_do_redirect_map(struct net_device *dev, struct xdp_buff *xdp,
 	ri->tgt_value = NULL;
 	WRITE_ONCE(ri->map, NULL);
 
-	if (ri->map_to_flush && unlikely(ri->map_to_flush != map))
-		xdp_do_flush_map();
-
-	err = __bpf_tx_xdp_map(dev, fwd, map, xdp, index);
+	err = __bpf_tx_xdp_map(dev, fwd, map, xdp);
 	if (unlikely(err))
 		goto err;
 
-	ri->map_to_flush = map;
 	_trace_xdp_redirect_map(dev, xdp_prog, fwd, map, index);
 	return 0;
 err:
@@ -8941,3 +8900,11 @@ const struct bpf_verifier_ops sk_reuseport_verifier_ops = {
 const struct bpf_prog_ops sk_reuseport_prog_ops = {
 };
 #endif /* CONFIG_INET */
+
+DEFINE_BPF_DISPATCHER(bpf_dispatcher_xdp)
+
+void bpf_prog_change_xdp(struct bpf_prog *prev_prog, struct bpf_prog *prog)
+{
+	bpf_dispatcher_change_prog(BPF_DISPATCHER_PTR(bpf_dispatcher_xdp),
+				   prev_prog, prog);
+}
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 328f661b83b2..02ada7ab8c6e 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -31,6 +31,8 @@
 
 #define TX_BATCH_SIZE 16
 
+static DEFINE_PER_CPU(struct list_head, xskmap_flush_list);
+
 bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs)
 {
 	return READ_ONCE(xs->rx) &&  READ_ONCE(xs->umem) &&
@@ -39,21 +41,21 @@ bool xsk_is_setup_for_bpf_map(struct xdp_sock *xs)
 
 bool xsk_umem_has_addrs(struct xdp_umem *umem, u32 cnt)
 {
-	return xskq_has_addrs(umem->fq, cnt);
+	return xskq_cons_has_entries(umem->fq, cnt);
 }
 EXPORT_SYMBOL(xsk_umem_has_addrs);
 
-u64 *xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
+bool xsk_umem_peek_addr(struct xdp_umem *umem, u64 *addr)
 {
-	return xskq_peek_addr(umem->fq, addr, umem);
+	return xskq_cons_peek_addr(umem->fq, addr, umem);
 }
 EXPORT_SYMBOL(xsk_umem_peek_addr);
 
-void xsk_umem_discard_addr(struct xdp_umem *umem)
+void xsk_umem_release_addr(struct xdp_umem *umem)
 {
-	xskq_discard_addr(umem->fq);
+	xskq_cons_release(umem->fq);
 }
-EXPORT_SYMBOL(xsk_umem_discard_addr);
+EXPORT_SYMBOL(xsk_umem_release_addr);
 
 void xsk_set_rx_need_wakeup(struct xdp_umem *umem)
 {
@@ -124,7 +126,7 @@ static void __xsk_rcv_memcpy(struct xdp_umem *umem, u64 addr, void *from_buf,
 	void *to_buf = xdp_umem_get_data(umem, addr);
 
 	addr = xsk_umem_add_offset_to_addr(addr);
-	if (xskq_crosses_non_contig_pg(umem, addr, len + metalen)) {
+	if (xskq_cons_crosses_non_contig_pg(umem, addr, len + metalen)) {
 		void *next_pg_addr = umem->pages[(addr >> PAGE_SHIFT) + 1].addr;
 		u64 page_start = addr & ~(PAGE_SIZE - 1);
 		u64 first_len = PAGE_SIZE - (addr - page_start);
@@ -146,7 +148,7 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 	u32 metalen;
 	int err;
 
-	if (!xskq_peek_addr(xs->umem->fq, &addr, xs->umem) ||
+	if (!xskq_cons_peek_addr(xs->umem->fq, &addr, xs->umem) ||
 	    len > xs->umem->chunk_size_nohr - XDP_PACKET_HEADROOM) {
 		xs->rx_dropped++;
 		return -ENOSPC;
@@ -165,9 +167,9 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 
 	offset += metalen;
 	addr = xsk_umem_adjust_offset(xs->umem, addr, offset);
-	err = xskq_produce_batch_desc(xs->rx, addr, len);
+	err = xskq_prod_reserve_desc(xs->rx, addr, len);
 	if (!err) {
-		xskq_discard_addr(xs->umem->fq);
+		xskq_cons_release(xs->umem->fq);
 		xdp_return_buff(xdp);
 		return 0;
 	}
@@ -178,7 +180,7 @@ static int __xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 
 static int __xsk_rcv_zc(struct xdp_sock *xs, struct xdp_buff *xdp, u32 len)
 {
-	int err = xskq_produce_batch_desc(xs->rx, (u64)xdp->handle, len);
+	int err = xskq_prod_reserve_desc(xs->rx, xdp->handle, len);
 
 	if (err)
 		xs->rx_dropped++;
@@ -214,7 +216,7 @@ static int xsk_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 
 static void xsk_flush(struct xdp_sock *xs)
 {
-	xskq_produce_flush_desc(xs->rx);
+	xskq_prod_submit(xs->rx);
 	xs->sk.sk_data_ready(&xs->sk);
 }
 
@@ -234,7 +236,7 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 		goto out_unlock;
 	}
 
-	if (!xskq_peek_addr(xs->umem->fq, &addr, xs->umem) ||
+	if (!xskq_cons_peek_addr(xs->umem->fq, &addr, xs->umem) ||
 	    len > xs->umem->chunk_size_nohr - XDP_PACKET_HEADROOM) {
 		err = -ENOSPC;
 		goto out_drop;
@@ -245,12 +247,12 @@ int xsk_generic_rcv(struct xdp_sock *xs, struct xdp_buff *xdp)
 	memcpy(buffer, xdp->data_meta, len + metalen);
 
 	addr = xsk_umem_adjust_offset(xs->umem, addr, metalen);
-	err = xskq_produce_batch_desc(xs->rx, addr, len);
+	err = xskq_prod_reserve_desc(xs->rx, addr, len);
 	if (err)
 		goto out_drop;
 
-	xskq_discard_addr(xs->umem->fq);
-	xskq_produce_flush_desc(xs->rx);
+	xskq_cons_release(xs->umem->fq);
+	xskq_prod_submit(xs->rx);
 
 	spin_unlock_bh(&xs->rx_lock);
 
@@ -264,11 +266,9 @@ out_unlock:
 	return err;
 }
 
-int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
-		       struct xdp_sock *xs)
+int __xsk_map_redirect(struct xdp_sock *xs, struct xdp_buff *xdp)
 {
-	struct xsk_map *m = container_of(map, struct xsk_map, map);
-	struct list_head *flush_list = this_cpu_ptr(m->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&xskmap_flush_list);
 	int err;
 
 	err = xsk_rcv(xs, xdp);
@@ -281,10 +281,9 @@ int __xsk_map_redirect(struct bpf_map *map, struct xdp_buff *xdp,
 	return 0;
 }
 
-void __xsk_map_flush(struct bpf_map *map)
+void __xsk_map_flush(void)
 {
-	struct xsk_map *m = container_of(map, struct xsk_map, map);
-	struct list_head *flush_list = this_cpu_ptr(m->flush_list);
+	struct list_head *flush_list = this_cpu_ptr(&xskmap_flush_list);
 	struct xdp_sock *xs, *tmp;
 
 	list_for_each_entry_safe(xs, tmp, flush_list, flush_node) {
@@ -295,7 +294,7 @@ void __xsk_map_flush(struct bpf_map *map)
 
 void xsk_umem_complete_tx(struct xdp_umem *umem, u32 nb_entries)
 {
-	xskq_produce_flush_addr_n(umem->cq, nb_entries);
+	xskq_prod_submit_n(umem->cq, nb_entries);
 }
 EXPORT_SYMBOL(xsk_umem_complete_tx);
 
@@ -317,13 +316,18 @@ bool xsk_umem_consume_tx(struct xdp_umem *umem, struct xdp_desc *desc)
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(xs, &umem->xsk_list, list) {
-		if (!xskq_peek_desc(xs->tx, desc, umem))
+		if (!xskq_cons_peek_desc(xs->tx, desc, umem))
 			continue;
 
-		if (xskq_produce_addr_lazy(umem->cq, desc->addr))
+		/* This is the backpreassure mechanism for the Tx path.
+		 * Reserve space in the completion queue and only proceed
+		 * if there is space in it. This avoids having to implement
+		 * any buffering in the Tx path.
+		 */
+		if (xskq_prod_reserve_addr(umem->cq, desc->addr))
 			goto out;
 
-		xskq_discard_desc(xs->tx);
+		xskq_cons_release(xs->tx);
 		rcu_read_unlock();
 		return true;
 	}
@@ -358,7 +362,7 @@ static void xsk_destruct_skb(struct sk_buff *skb)
 	unsigned long flags;
 
 	spin_lock_irqsave(&xs->tx_completion_lock, flags);
-	WARN_ON_ONCE(xskq_produce_addr(xs->umem->cq, addr));
+	xskq_prod_submit_addr(xs->umem->cq, addr);
 	spin_unlock_irqrestore(&xs->tx_completion_lock, flags);
 
 	sock_wfree(skb);
@@ -378,7 +382,7 @@ static int xsk_generic_xmit(struct sock *sk)
 	if (xs->queue_id >= xs->dev->real_num_tx_queues)
 		goto out;
 
-	while (xskq_peek_desc(xs->tx, &desc, xs->umem)) {
+	while (xskq_cons_peek_desc(xs->tx, &desc, xs->umem)) {
 		char *buffer;
 		u64 addr;
 		u32 len;
@@ -399,7 +403,12 @@ static int xsk_generic_xmit(struct sock *sk)
 		addr = desc.addr;
 		buffer = xdp_umem_get_data(xs->umem, addr);
 		err = skb_store_bits(skb, 0, buffer, len);
-		if (unlikely(err) || xskq_reserve_addr(xs->umem->cq)) {
+		/* This is the backpreassure mechanism for the Tx path.
+		 * Reserve space in the completion queue and only proceed
+		 * if there is space in it. This avoids having to implement
+		 * any buffering in the Tx path.
+		 */
+		if (unlikely(err) || xskq_prod_reserve(xs->umem->cq)) {
 			kfree_skb(skb);
 			goto out;
 		}
@@ -411,7 +420,7 @@ static int xsk_generic_xmit(struct sock *sk)
 		skb->destructor = xsk_destruct_skb;
 
 		err = dev_direct_xmit(skb, xs->queue_id);
-		xskq_discard_desc(xs->tx);
+		xskq_cons_release(xs->tx);
 		/* Ignore NET_XMIT_CN as packet might have been sent */
 		if (err == NET_XMIT_DROP || err == NETDEV_TX_BUSY) {
 			/* SKB completed but not sent */
@@ -477,9 +486,9 @@ static __poll_t xsk_poll(struct file *file, struct socket *sock,
 			__xsk_sendmsg(sk);
 	}
 
-	if (xs->rx && !xskq_empty_desc(xs->rx))
+	if (xs->rx && !xskq_prod_is_empty(xs->rx))
 		mask |= EPOLLIN | EPOLLRDNORM;
-	if (xs->tx && !xskq_full_desc(xs->tx))
+	if (xs->tx && !xskq_cons_is_full(xs->tx))
 		mask |= EPOLLOUT | EPOLLWRNORM;
 
 	return mask;
@@ -1183,7 +1192,7 @@ static struct pernet_operations xsk_net_ops = {
 
 static int __init xsk_init(void)
 {
-	int err;
+	int err, cpu;
 
 	err = proto_register(&xsk_proto, 0 /* no slab */);
 	if (err)
@@ -1201,6 +1210,8 @@ static int __init xsk_init(void)
 	if (err)
 		goto out_pernet;
 
+	for_each_possible_cpu(cpu)
+		INIT_LIST_HEAD(&per_cpu(xskmap_flush_list, cpu));
 	return 0;
 
 out_pernet:
diff --git a/net/xdp/xsk_queue.c b/net/xdp/xsk_queue.c
index b66504592d9b..c90e9c1e3c63 100644
--- a/net/xdp/xsk_queue.c
+++ b/net/xdp/xsk_queue.c
@@ -18,14 +18,14 @@ void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask)
 	q->chunk_mask = chunk_mask;
 }
 
-static u32 xskq_umem_get_ring_size(struct xsk_queue *q)
+static size_t xskq_get_ring_size(struct xsk_queue *q, bool umem_queue)
 {
-	return sizeof(struct xdp_umem_ring) + q->nentries * sizeof(u64);
-}
+	struct xdp_umem_ring *umem_ring;
+	struct xdp_rxtx_ring *rxtx_ring;
 
-static u32 xskq_rxtx_get_ring_size(struct xsk_queue *q)
-{
-	return sizeof(struct xdp_ring) + q->nentries * sizeof(struct xdp_desc);
+	if (umem_queue)
+		return struct_size(umem_ring, desc, q->nentries);
+	return struct_size(rxtx_ring, desc, q->nentries);
 }
 
 struct xsk_queue *xskq_create(u32 nentries, bool umem_queue)
@@ -43,8 +43,7 @@ struct xsk_queue *xskq_create(u32 nentries, bool umem_queue)
 
 	gfp_flags = GFP_KERNEL | __GFP_ZERO | __GFP_NOWARN |
 		    __GFP_COMP  | __GFP_NORETRY;
-	size = umem_queue ? xskq_umem_get_ring_size(q) :
-	       xskq_rxtx_get_ring_size(q);
+	size = xskq_get_ring_size(q, umem_queue);
 
 	q->ring = (struct xdp_ring *)__get_free_pages(gfp_flags,
 						      get_order(size));
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index eddae4688862..bec2af11853a 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -10,9 +10,6 @@
 #include <linux/if_xdp.h>
 #include <net/xdp_sock.h>
 
-#define RX_BATCH_SIZE 16
-#define LAZY_UPDATE_THRESHOLD 128
-
 struct xdp_ring {
 	u32 producer ____cacheline_aligned_in_smp;
 	u32 consumer ____cacheline_aligned_in_smp;
@@ -36,10 +33,8 @@ struct xsk_queue {
 	u64 size;
 	u32 ring_mask;
 	u32 nentries;
-	u32 prod_head;
-	u32 prod_tail;
-	u32 cons_head;
-	u32 cons_tail;
+	u32 cached_prod;
+	u32 cached_cons;
 	struct xdp_ring *ring;
 	u64 invalid_descs;
 };
@@ -86,56 +81,31 @@ struct xsk_queue {
  * now and again after circling through the ring.
  */
 
-/* Common functions operating for both RXTX and umem queues */
-
-static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
-{
-	return q ? q->invalid_descs : 0;
-}
-
-static inline u32 xskq_nb_avail(struct xsk_queue *q, u32 dcnt)
-{
-	u32 entries = q->prod_tail - q->cons_tail;
-
-	if (entries == 0) {
-		/* Refresh the local pointer */
-		q->prod_tail = READ_ONCE(q->ring->producer);
-		entries = q->prod_tail - q->cons_tail;
-	}
-
-	return (entries > dcnt) ? dcnt : entries;
-}
-
-static inline u32 xskq_nb_free(struct xsk_queue *q, u32 producer, u32 dcnt)
-{
-	u32 free_entries = q->nentries - (producer - q->cons_tail);
-
-	if (free_entries >= dcnt)
-		return free_entries;
-
-	/* Refresh the local tail pointer */
-	q->cons_tail = READ_ONCE(q->ring->consumer);
-	return q->nentries - (producer - q->cons_tail);
-}
-
-static inline bool xskq_has_addrs(struct xsk_queue *q, u32 cnt)
-{
-	u32 entries = q->prod_tail - q->cons_tail;
-
-	if (entries >= cnt)
-		return true;
-
-	/* Refresh the local pointer. */
-	q->prod_tail = READ_ONCE(q->ring->producer);
-	entries = q->prod_tail - q->cons_tail;
-
-	return entries >= cnt;
-}
+/* The operations on the rings are the following:
+ *
+ * producer                           consumer
+ *
+ * RESERVE entries                    PEEK in the ring for entries
+ * WRITE data into the ring           READ data from the ring
+ * SUBMIT entries                     RELEASE entries
+ *
+ * The producer reserves one or more entries in the ring. It can then
+ * fill in these entries and finally submit them so that they can be
+ * seen and read by the consumer.
+ *
+ * The consumer peeks into the ring to see if the producer has written
+ * any new entries. If so, the producer can then read these entries
+ * and when it is done reading them release them back to the producer
+ * so that the producer can use these slots to fill in new entries.
+ *
+ * The function names below reflect these operations.
+ */
 
-/* UMEM queue */
+/* Functions that read and validate content from consumer rings. */
 
-static inline bool xskq_crosses_non_contig_pg(struct xdp_umem *umem, u64 addr,
-					      u64 length)
+static inline bool xskq_cons_crosses_non_contig_pg(struct xdp_umem *umem,
+						   u64 addr,
+						   u64 length)
 {
 	bool cross_pg = (addr & (PAGE_SIZE - 1)) + length > PAGE_SIZE;
 	bool next_pg_contig =
@@ -145,9 +115,16 @@ static inline bool xskq_crosses_non_contig_pg(struct xdp_umem *umem, u64 addr,
 	return cross_pg && !next_pg_contig;
 }
 
-static inline bool xskq_is_valid_addr(struct xsk_queue *q, u64 addr)
+static inline bool xskq_cons_is_valid_unaligned(struct xsk_queue *q,
+						u64 addr,
+						u64 length,
+						struct xdp_umem *umem)
 {
-	if (addr >= q->size) {
+	u64 base_addr = xsk_umem_extract_addr(addr);
+
+	addr = xsk_umem_add_offset_to_addr(addr);
+	if (base_addr >= q->size || addr >= q->size ||
+	    xskq_cons_crosses_non_contig_pg(umem, addr, length)) {
 		q->invalid_descs++;
 		return false;
 	}
@@ -155,15 +132,9 @@ static inline bool xskq_is_valid_addr(struct xsk_queue *q, u64 addr)
 	return true;
 }
 
-static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
-						u64 length,
-						struct xdp_umem *umem)
+static inline bool xskq_cons_is_valid_addr(struct xsk_queue *q, u64 addr)
 {
-	u64 base_addr = xsk_umem_extract_addr(addr);
-
-	addr = xsk_umem_add_offset_to_addr(addr);
-	if (base_addr >= q->size || addr >= q->size ||
-	    xskq_crosses_non_contig_pg(umem, addr, length)) {
+	if (addr >= q->size) {
 		q->invalid_descs++;
 		return false;
 	}
@@ -171,204 +142,240 @@ static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
 	return true;
 }
 
-static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
-				      struct xdp_umem *umem)
+static inline bool xskq_cons_read_addr(struct xsk_queue *q, u64 *addr,
+				       struct xdp_umem *umem)
 {
-	while (q->cons_tail != q->cons_head) {
-		struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-		unsigned int idx = q->cons_tail & q->ring_mask;
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
-		*addr = READ_ONCE(ring->desc[idx]) & q->chunk_mask;
+	while (q->cached_cons != q->cached_prod) {
+		u32 idx = q->cached_cons & q->ring_mask;
+
+		*addr = ring->desc[idx] & q->chunk_mask;
 
 		if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
-			if (xskq_is_valid_addr_unaligned(q, *addr,
+			if (xskq_cons_is_valid_unaligned(q, *addr,
 							 umem->chunk_size_nohr,
 							 umem))
-				return addr;
+				return true;
 			goto out;
 		}
 
-		if (xskq_is_valid_addr(q, *addr))
-			return addr;
+		if (xskq_cons_is_valid_addr(q, *addr))
+			return true;
 
 out:
-		q->cons_tail++;
+		q->cached_cons++;
 	}
 
-	return NULL;
+	return false;
 }
 
-static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr,
-				  struct xdp_umem *umem)
+static inline bool xskq_cons_is_valid_desc(struct xsk_queue *q,
+					   struct xdp_desc *d,
+					   struct xdp_umem *umem)
 {
-	if (q->cons_tail == q->cons_head) {
-		smp_mb(); /* D, matches A */
-		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);
+	if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
+		if (!xskq_cons_is_valid_unaligned(q, d->addr, d->len, umem))
+			return false;
+
+		if (d->len > umem->chunk_size_nohr || d->options) {
+			q->invalid_descs++;
+			return false;
+		}
 
-		/* Order consumer and data */
-		smp_rmb();
+		return true;
 	}
 
-	return xskq_validate_addr(q, addr, umem);
-}
+	if (!xskq_cons_is_valid_addr(q, d->addr))
+		return false;
 
-static inline void xskq_discard_addr(struct xsk_queue *q)
-{
-	q->cons_tail++;
+	if (((d->addr + d->len) & q->chunk_mask) != (d->addr & q->chunk_mask) ||
+	    d->options) {
+		q->invalid_descs++;
+		return false;
+	}
+
+	return true;
 }
 
-static inline int xskq_produce_addr(struct xsk_queue *q, u64 addr)
+static inline bool xskq_cons_read_desc(struct xsk_queue *q,
+				       struct xdp_desc *desc,
+				       struct xdp_umem *umem)
 {
-	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
-
-	if (xskq_nb_free(q, q->prod_tail, 1) == 0)
-		return -ENOSPC;
+	while (q->cached_cons != q->cached_prod) {
+		struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
+		u32 idx = q->cached_cons & q->ring_mask;
 
-	/* A, matches D */
-	ring->desc[q->prod_tail++ & q->ring_mask] = addr;
+		*desc = ring->desc[idx];
+		if (xskq_cons_is_valid_desc(q, desc, umem))
+			return true;
 
-	/* Order producer and data */
-	smp_wmb(); /* B, matches C */
+		q->cached_cons++;
+	}
 
-	WRITE_ONCE(q->ring->producer, q->prod_tail);
-	return 0;
+	return false;
 }
 
-static inline int xskq_produce_addr_lazy(struct xsk_queue *q, u64 addr)
+/* Functions for consumers */
+
+static inline void __xskq_cons_release(struct xsk_queue *q)
 {
-	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
+	smp_mb(); /* D, matches A */
+	WRITE_ONCE(q->ring->consumer, q->cached_cons);
+}
 
-	if (xskq_nb_free(q, q->prod_head, LAZY_UPDATE_THRESHOLD) == 0)
-		return -ENOSPC;
+static inline void __xskq_cons_peek(struct xsk_queue *q)
+{
+	/* Refresh the local pointer */
+	q->cached_prod = READ_ONCE(q->ring->producer);
+	smp_rmb(); /* C, matches B */
+}
 
-	/* A, matches D */
-	ring->desc[q->prod_head++ & q->ring_mask] = addr;
-	return 0;
+static inline void xskq_cons_get_entries(struct xsk_queue *q)
+{
+	__xskq_cons_release(q);
+	__xskq_cons_peek(q);
 }
 
-static inline void xskq_produce_flush_addr_n(struct xsk_queue *q,
-					     u32 nb_entries)
+static inline bool xskq_cons_has_entries(struct xsk_queue *q, u32 cnt)
 {
-	/* Order producer and data */
-	smp_wmb(); /* B, matches C */
+	u32 entries = q->cached_prod - q->cached_cons;
 
-	q->prod_tail += nb_entries;
-	WRITE_ONCE(q->ring->producer, q->prod_tail);
+	if (entries >= cnt)
+		return true;
+
+	__xskq_cons_peek(q);
+	entries = q->cached_prod - q->cached_cons;
+
+	return entries >= cnt;
 }
 
-static inline int xskq_reserve_addr(struct xsk_queue *q)
+static inline bool xskq_cons_peek_addr(struct xsk_queue *q, u64 *addr,
+				       struct xdp_umem *umem)
 {
-	if (xskq_nb_free(q, q->prod_head, 1) == 0)
-		return -ENOSPC;
+	if (q->cached_prod == q->cached_cons)
+		xskq_cons_get_entries(q);
+	return xskq_cons_read_addr(q, addr, umem);
+}
 
-	/* A, matches D */
-	q->prod_head++;
-	return 0;
+static inline bool xskq_cons_peek_desc(struct xsk_queue *q,
+				       struct xdp_desc *desc,
+				       struct xdp_umem *umem)
+{
+	if (q->cached_prod == q->cached_cons)
+		xskq_cons_get_entries(q);
+	return xskq_cons_read_desc(q, desc, umem);
 }
 
-/* Rx/Tx queue */
+static inline void xskq_cons_release(struct xsk_queue *q)
+{
+	/* To improve performance, only update local state here.
+	 * Reflect this to global state when we get new entries
+	 * from the ring in xskq_cons_get_entries().
+	 */
+	q->cached_cons++;
+}
 
-static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d,
-				      struct xdp_umem *umem)
+static inline bool xskq_cons_is_full(struct xsk_queue *q)
 {
-	if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
-		if (!xskq_is_valid_addr_unaligned(q, d->addr, d->len, umem))
-			return false;
+	/* No barriers needed since data is not accessed */
+	return READ_ONCE(q->ring->producer) - READ_ONCE(q->ring->consumer) ==
+		q->nentries;
+}
 
-		if (d->len > umem->chunk_size_nohr || d->options) {
-			q->invalid_descs++;
-			return false;
-		}
+/* Functions for producers */
 
-		return true;
-	}
+static inline bool xskq_prod_is_full(struct xsk_queue *q)
+{
+	u32 free_entries = q->nentries - (q->cached_prod - q->cached_cons);
 
-	if (!xskq_is_valid_addr(q, d->addr))
+	if (free_entries)
 		return false;
 
-	if (((d->addr + d->len) & q->chunk_mask) != (d->addr & q->chunk_mask) ||
-	    d->options) {
-		q->invalid_descs++;
-		return false;
-	}
+	/* Refresh the local tail pointer */
+	q->cached_cons = READ_ONCE(q->ring->consumer);
+	free_entries = q->nentries - (q->cached_prod - q->cached_cons);
 
-	return true;
+	return !free_entries;
 }
 
-static inline struct xdp_desc *xskq_validate_desc(struct xsk_queue *q,
-						  struct xdp_desc *desc,
-						  struct xdp_umem *umem)
+static inline int xskq_prod_reserve(struct xsk_queue *q)
 {
-	while (q->cons_tail != q->cons_head) {
-		struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
-		unsigned int idx = q->cons_tail & q->ring_mask;
-
-		*desc = READ_ONCE(ring->desc[idx]);
-		if (xskq_is_valid_desc(q, desc, umem))
-			return desc;
-
-		q->cons_tail++;
-	}
+	if (xskq_prod_is_full(q))
+		return -ENOSPC;
 
-	return NULL;
+	/* A, matches D */
+	q->cached_prod++;
+	return 0;
 }
 
-static inline struct xdp_desc *xskq_peek_desc(struct xsk_queue *q,
-					      struct xdp_desc *desc,
-					      struct xdp_umem *umem)
+static inline int xskq_prod_reserve_addr(struct xsk_queue *q, u64 addr)
 {
-	if (q->cons_tail == q->cons_head) {
-		smp_mb(); /* D, matches A */
-		WRITE_ONCE(q->ring->consumer, q->cons_tail);
-		q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);
-
-		/* Order consumer and data */
-		smp_rmb(); /* C, matches B */
-	}
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
 
-	return xskq_validate_desc(q, desc, umem);
-}
+	if (xskq_prod_is_full(q))
+		return -ENOSPC;
 
-static inline void xskq_discard_desc(struct xsk_queue *q)
-{
-	q->cons_tail++;
+	/* A, matches D */
+	ring->desc[q->cached_prod++ & q->ring_mask] = addr;
+	return 0;
 }
 
-static inline int xskq_produce_batch_desc(struct xsk_queue *q,
-					  u64 addr, u32 len)
+static inline int xskq_prod_reserve_desc(struct xsk_queue *q,
+					 u64 addr, u32 len)
 {
 	struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
-	unsigned int idx;
+	u32 idx;
 
-	if (xskq_nb_free(q, q->prod_head, 1) == 0)
+	if (xskq_prod_is_full(q))
 		return -ENOSPC;
 
 	/* A, matches D */
-	idx = (q->prod_head++) & q->ring_mask;
+	idx = q->cached_prod++ & q->ring_mask;
 	ring->desc[idx].addr = addr;
 	ring->desc[idx].len = len;
 
 	return 0;
 }
 
-static inline void xskq_produce_flush_desc(struct xsk_queue *q)
+static inline void __xskq_prod_submit(struct xsk_queue *q, u32 idx)
 {
-	/* Order producer and data */
 	smp_wmb(); /* B, matches C */
 
-	q->prod_tail = q->prod_head;
-	WRITE_ONCE(q->ring->producer, q->prod_tail);
+	WRITE_ONCE(q->ring->producer, idx);
+}
+
+static inline void xskq_prod_submit(struct xsk_queue *q)
+{
+	__xskq_prod_submit(q, q->cached_prod);
+}
+
+static inline void xskq_prod_submit_addr(struct xsk_queue *q, u64 addr)
+{
+	struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
+	u32 idx = q->ring->producer;
+
+	ring->desc[idx++ & q->ring_mask] = addr;
+
+	__xskq_prod_submit(q, idx);
 }
 
-static inline bool xskq_full_desc(struct xsk_queue *q)
+static inline void xskq_prod_submit_n(struct xsk_queue *q, u32 nb_entries)
 {
-	return xskq_nb_avail(q, q->nentries) == q->nentries;
+	__xskq_prod_submit(q, q->ring->producer + nb_entries);
 }
 
-static inline bool xskq_empty_desc(struct xsk_queue *q)
+static inline bool xskq_prod_is_empty(struct xsk_queue *q)
 {
-	return xskq_nb_free(q, q->prod_tail, q->nentries) == q->nentries;
+	/* No barriers needed since data is not accessed */
+	return READ_ONCE(q->ring->consumer) == READ_ONCE(q->ring->producer);
+}
+
+/* For both producers and consumers */
+
+static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
+{
+	return q ? q->invalid_descs : 0;
 }
 
 void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask);
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 1fc42ad8ff49..5b89c0370f33 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -38,6 +38,8 @@ tprogs-y += tc_l2_redirect
 tprogs-y += lwt_len_hist
 tprogs-y += xdp_tx_iptunnel
 tprogs-y += test_map_in_map
+tprogs-y += per_socket_stats_example
+tprogs-y += xdp_redirect
 tprogs-y += xdp_redirect_map
 tprogs-y += xdp_redirect_cpu
 tprogs-y += xdp_monitor
@@ -196,7 +198,7 @@ endif
 
 TPROGCFLAGS_bpf_load.o += -Wno-unused-variable
 
-TPROGS_LDLIBS			+= $(LIBBPF) -lelf
+TPROGS_LDLIBS			+= $(LIBBPF) -lelf -lz
 TPROGLDLIBS_tracex4		+= -lrt
 TPROGLDLIBS_trace_output	+= -lrt
 TPROGLDLIBS_map_perf_test	+= -lrt
@@ -234,6 +236,7 @@ BTF_LLVM_PROBE := $(shell echo "int main() { return 0; }" | \
 			  readelf -S ./llvm_btf_verify.o | grep BTF; \
 			  /bin/rm -f ./llvm_btf_verify.o)
 
+BPF_EXTRA_CFLAGS += -fno-stack-protector
 ifneq ($(BTF_LLVM_PROBE),)
 	BPF_EXTRA_CFLAGS += -g
 else
diff --git a/samples/bpf/xdp1_user.c b/samples/bpf/xdp1_user.c
index 3e553eed95a7..38a8852cb57f 100644
--- a/samples/bpf/xdp1_user.c
+++ b/samples/bpf/xdp1_user.c
@@ -98,7 +98,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -109,6 +109,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_adjust_tail_user.c b/samples/bpf/xdp_adjust_tail_user.c
index d86e9ad0356b..008789eb6ada 100644
--- a/samples/bpf/xdp_adjust_tail_user.c
+++ b/samples/bpf/xdp_adjust_tail_user.c
@@ -120,7 +120,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -132,6 +132,9 @@ int main(int argc, char **argv)
 		opt_flags[opt] = 0;
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	for (i = 0; i < strlen(optstr); i++) {
 		if (opt_flags[(unsigned int)optstr[i]]) {
 			fprintf(stderr, "Missing argument -%c\n", optstr[i]);
diff --git a/samples/bpf/xdp_fwd_user.c b/samples/bpf/xdp_fwd_user.c
index 97ff1dad7669..c30f9acfdb84 100644
--- a/samples/bpf/xdp_fwd_user.c
+++ b/samples/bpf/xdp_fwd_user.c
@@ -27,11 +27,13 @@
 #include "libbpf.h"
 #include <bpf/bpf.h>
 
+static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
+
 static int do_attach(int idx, int prog_fd, int map_fd, const char *name)
 {
 	int err;
 
-	err = bpf_set_link_xdp_fd(idx, prog_fd, 0);
+	err = bpf_set_link_xdp_fd(idx, prog_fd, xdp_flags);
 	if (err < 0) {
 		printf("ERROR: failed to attach program to %s\n", name);
 		return err;
@@ -49,7 +51,7 @@ static int do_detach(int idx, const char *name)
 {
 	int err;
 
-	err = bpf_set_link_xdp_fd(idx, -1, 0);
+	err = bpf_set_link_xdp_fd(idx, -1, xdp_flags);
 	if (err < 0)
 		printf("ERROR: failed to detach program from %s\n", name);
 
@@ -83,11 +85,17 @@ int main(int argc, char **argv)
 	int attach = 1;
 	int ret = 0;
 
-	while ((opt = getopt(argc, argv, ":dD")) != -1) {
+	while ((opt = getopt(argc, argv, ":dDSF")) != -1) {
 		switch (opt) {
 		case 'd':
 			attach = 0;
 			break;
+		case 'S':
+			xdp_flags |= XDP_FLAGS_SKB_MODE;
+			break;
+		case 'F':
+			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
+			break;
 		case 'D':
 			prog_name = "xdp_fwd_direct";
 			break;
@@ -97,6 +105,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index 0da6e9e7132e..79a2fb7d16cb 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -16,6 +16,10 @@ static const char *__doc__ =
 #include <getopt.h>
 #include <net/if.h>
 #include <time.h>
+#include <linux/limits.h>
+
+#define __must_check
+#include <linux/err.h>
 
 #include <arpa/inet.h>
 #include <linux/if_link.h>
@@ -46,6 +50,10 @@ static int cpus_count_map_fd;
 static int cpus_iterator_map_fd;
 static int exception_cnt_map_fd;
 
+#define NUM_TP 5
+struct bpf_link *tp_links[NUM_TP] = { 0 };
+static int tp_cnt = 0;
+
 /* Exit return codes */
 #define EXIT_OK		0
 #define EXIT_FAIL		1
@@ -88,6 +96,10 @@ static void int_exit(int sig)
 			printf("program on interface changed, not removing\n");
 		}
 	}
+	/* Detach tracepoints */
+	while (tp_cnt)
+		bpf_link__destroy(tp_links[--tp_cnt]);
+
 	exit(EXIT_OK);
 }
 
@@ -588,23 +600,61 @@ static void stats_poll(int interval, bool use_separators, char *prog_name,
 	free_stats_record(prev);
 }
 
+static struct bpf_link * attach_tp(struct bpf_object *obj,
+				   const char *tp_category,
+				   const char* tp_name)
+{
+	struct bpf_program *prog;
+	struct bpf_link *link;
+	char sec_name[PATH_MAX];
+	int len;
+
+	len = snprintf(sec_name, PATH_MAX, "tracepoint/%s/%s",
+		       tp_category, tp_name);
+	if (len < 0)
+		exit(EXIT_FAIL);
+
+	prog = bpf_object__find_program_by_title(obj, sec_name);
+	if (!prog) {
+		fprintf(stderr, "ERR: finding progsec: %s\n", sec_name);
+		exit(EXIT_FAIL_BPF);
+	}
+
+	link = bpf_program__attach_tracepoint(prog, tp_category, tp_name);
+	if (IS_ERR(link))
+		exit(EXIT_FAIL_BPF);
+
+	return link;
+}
+
+static void init_tracepoints(struct bpf_object *obj) {
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_redirect_err");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_redirect_map_err");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_exception");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_cpumap_enqueue");
+	tp_links[tp_cnt++] = attach_tp(obj, "xdp", "xdp_cpumap_kthread");
+}
+
 static int init_map_fds(struct bpf_object *obj)
 {
-	cpu_map_fd = bpf_object__find_map_fd_by_name(obj, "cpu_map");
-	rx_cnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rx_cnt");
+	/* Maps updated by tracepoints */
 	redirect_err_cnt_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "redirect_err_cnt");
+	exception_cnt_map_fd =
+		bpf_object__find_map_fd_by_name(obj, "exception_cnt");
 	cpumap_enqueue_cnt_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpumap_enqueue_cnt");
 	cpumap_kthread_cnt_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpumap_kthread_cnt");
+
+	/* Maps used by XDP */
+	rx_cnt_map_fd = bpf_object__find_map_fd_by_name(obj, "rx_cnt");
+	cpu_map_fd = bpf_object__find_map_fd_by_name(obj, "cpu_map");
 	cpus_available_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpus_available");
 	cpus_count_map_fd = bpf_object__find_map_fd_by_name(obj, "cpus_count");
 	cpus_iterator_map_fd =
 		bpf_object__find_map_fd_by_name(obj, "cpus_iterator");
-	exception_cnt_map_fd =
-		bpf_object__find_map_fd_by_name(obj, "exception_cnt");
 
 	if (cpu_map_fd < 0 || rx_cnt_map_fd < 0 ||
 	    redirect_err_cnt_map_fd < 0 || cpumap_enqueue_cnt_map_fd < 0 ||
@@ -662,6 +712,7 @@ int main(int argc, char **argv)
 			strerror(errno));
 		return EXIT_FAIL;
 	}
+	init_tracepoints(obj);
 	if (init_map_fds(obj) < 0) {
 		fprintf(stderr, "bpf_object__find_map_fd_by_name failed\n");
 		return EXIT_FAIL;
@@ -728,6 +779,10 @@ int main(int argc, char **argv)
 			return EXIT_FAIL_OPTION;
 		}
 	}
+
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	/* Required option */
 	if (ifindex == -1) {
 		fprintf(stderr, "ERR: required option --dev missing\n");
diff --git a/samples/bpf/xdp_redirect_map_user.c b/samples/bpf/xdp_redirect_map_user.c
index f70ee33907fd..cc840661faab 100644
--- a/samples/bpf/xdp_redirect_map_user.c
+++ b/samples/bpf/xdp_redirect_map_user.c
@@ -116,7 +116,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -127,6 +127,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		printf("usage: %s <IFNAME|IFINDEX>_IN <IFNAME|IFINDEX>_OUT\n", argv[0]);
 		return 1;
diff --git a/samples/bpf/xdp_redirect_user.c b/samples/bpf/xdp_redirect_user.c
index 5440cd620607..71dff8e3382a 100644
--- a/samples/bpf/xdp_redirect_user.c
+++ b/samples/bpf/xdp_redirect_user.c
@@ -117,7 +117,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -128,6 +128,9 @@ int main(int argc, char **argv)
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		printf("usage: %s <IFNAME|IFINDEX>_IN <IFNAME|IFINDEX>_OUT\n", argv[0]);
 		return 1;
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
index 1469b66ebad1..fef286c5add2 100644
--- a/samples/bpf/xdp_router_ipv4_user.c
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -662,6 +662,9 @@ int main(int ac, char **argv)
 		}
 	}
 
+	if (!(flags & XDP_FLAGS_SKB_MODE))
+		flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == ac) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_rxq_info_user.c b/samples/bpf/xdp_rxq_info_user.c
index 8fc3ad01de72..fc4983fd6959 100644
--- a/samples/bpf/xdp_rxq_info_user.c
+++ b/samples/bpf/xdp_rxq_info_user.c
@@ -551,6 +551,10 @@ int main(int argc, char **argv)
 			return EXIT_FAIL_OPTION;
 		}
 	}
+
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	/* Required option */
 	if (ifindex == -1) {
 		fprintf(stderr, "ERR: required option --dev missing\n");
diff --git a/samples/bpf/xdp_sample_pkts_user.c b/samples/bpf/xdp_sample_pkts_user.c
index a5760e8bf2c4..8c1af1b7372d 100644
--- a/samples/bpf/xdp_sample_pkts_user.c
+++ b/samples/bpf/xdp_sample_pkts_user.c
@@ -52,13 +52,13 @@ static int do_detach(int idx, const char *name)
 	__u32 curr_prog_id = 0;
 	int err = 0;
 
-	err = bpf_get_link_xdp_id(idx, &curr_prog_id, 0);
+	err = bpf_get_link_xdp_id(idx, &curr_prog_id, xdp_flags);
 	if (err) {
 		printf("bpf_get_link_xdp_id failed\n");
 		return err;
 	}
 	if (prog_id == curr_prog_id) {
-		err = bpf_set_link_xdp_fd(idx, -1, 0);
+		err = bpf_set_link_xdp_fd(idx, -1, xdp_flags);
 		if (err < 0)
 			printf("ERROR: failed to detach prog from %s\n", name);
 	} else if (!curr_prog_id) {
@@ -115,7 +115,7 @@ int main(int argc, char **argv)
 		.prog_type	= BPF_PROG_TYPE_XDP,
 	};
 	struct perf_buffer_opts pb_opts = {};
-	const char *optstr = "F";
+	const char *optstr = "FS";
 	int prog_fd, map_fd, opt;
 	struct bpf_object *obj;
 	struct bpf_map *map;
@@ -127,12 +127,18 @@ int main(int argc, char **argv)
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
 			break;
+		case 'S':
+			xdp_flags |= XDP_FLAGS_SKB_MODE;
+			break;
 		default:
 			usage(basename(argv[0]));
 			return 1;
 		}
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	if (optind == argc) {
 		usage(basename(argv[0]));
 		return 1;
diff --git a/samples/bpf/xdp_tx_iptunnel_user.c b/samples/bpf/xdp_tx_iptunnel_user.c
index 2fe4c7f5ffe5..5f33b5530032 100644
--- a/samples/bpf/xdp_tx_iptunnel_user.c
+++ b/samples/bpf/xdp_tx_iptunnel_user.c
@@ -231,7 +231,7 @@ int main(int argc, char **argv)
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
 		case 'N':
-			xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'F':
 			xdp_flags &= ~XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -243,6 +243,9 @@ int main(int argc, char **argv)
 		opt_flags[opt] = 0;
 	}
 
+	if (!(xdp_flags & XDP_FLAGS_SKB_MODE))
+		xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	for (i = 0; i < strlen(optstr); i++) {
 		if (opt_flags[(unsigned int)optstr[i]]) {
 			fprintf(stderr, "Missing argument -%c\n", optstr[i]);
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index a15480010828..d74c4c83fc93 100644
--- a/samples/bpf/xdpsock_user.c
+++ b/samples/bpf/xdpsock_user.c
@@ -10,6 +10,9 @@
 #include <linux/if_link.h>
 #include <linux/if_xdp.h>
 #include <linux/if_ether.h>
+#include <linux/ip.h>
+#include <linux/udp.h>
+#include <arpa/inet.h>
 #include <locale.h>
 #include <net/ethernet.h>
 #include <net/if.h>
@@ -45,12 +48,14 @@
 #endif
 
 #define NUM_FRAMES (4 * 1024)
-#define BATCH_SIZE 64
+#define MIN_PKT_SIZE 64
 
 #define DEBUG_HEXDUMP 0
 
 typedef __u64 u64;
 typedef __u32 u32;
+typedef __u16 u16;
+typedef __u8  u8;
 
 static unsigned long prev_time;
 
@@ -65,6 +70,13 @@ static u32 opt_xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
 static const char *opt_if = "";
 static int opt_ifindex;
 static int opt_queue;
+static unsigned long opt_duration;
+static unsigned long start_time;
+static bool benchmark_done;
+static u32 opt_batch_size = 64;
+static int opt_pkt_count;
+static u16 opt_pkt_size = MIN_PKT_SIZE;
+static u32 opt_pkt_fill_pattern = 0x12345678;
 static int opt_poll;
 static int opt_interval = 1;
 static u32 opt_xdp_bind_flags = XDP_USE_NEED_WAKEUP;
@@ -167,10 +179,21 @@ static void dump_stats(void)
 	}
 }
 
+static bool is_benchmark_done(void)
+{
+	if (opt_duration > 0) {
+		unsigned long dt = (get_nsecs() - start_time);
+
+		if (dt >= opt_duration)
+			benchmark_done = true;
+	}
+	return benchmark_done;
+}
+
 static void *poller(void *arg)
 {
 	(void)arg;
-	for (;;) {
+	while (!is_benchmark_done()) {
 		sleep(opt_interval);
 		dump_stats();
 	}
@@ -196,6 +219,11 @@ static void remove_xdp_program(void)
 
 static void int_exit(int sig)
 {
+	benchmark_done = true;
+}
+
+static void xdpsock_cleanup(void)
+{
 	struct xsk_umem *umem = xsks[0]->umem->umem;
 	int i;
 
@@ -204,8 +232,6 @@ static void int_exit(int sig)
 		xsk_socket__delete(xsks[i]->xsk);
 	(void)xsk_umem__delete(umem);
 	remove_xdp_program();
-
-	exit(EXIT_SUCCESS);
 }
 
 static void __exit_with_error(int error, const char *file, const char *func,
@@ -220,13 +246,6 @@ static void __exit_with_error(int error, const char *file, const char *func,
 
 #define exit_with_error(error) __exit_with_error(error, __FILE__, __func__, \
 						 __LINE__)
-
-static const char pkt_data[] =
-	"\x3c\xfd\xfe\x9e\x7f\x71\xec\xb1\xd7\x98\x3a\xc0\x08\x00\x45\x00"
-	"\x00\x2e\x00\x00\x00\x00\x40\x11\x88\x97\x05\x08\x07\x08\xc8\x14"
-	"\x1e\x04\x10\x92\x10\x92\x00\x1a\x6d\xa3\x34\x33\x1f\x69\x40\x6b"
-	"\x54\x59\xb6\x14\x2d\x11\x44\xbf\xaf\xd9\xbe\xaa";
-
 static void swap_mac_addresses(void *data)
 {
 	struct ether_header *eth = (struct ether_header *)data;
@@ -274,11 +293,243 @@ static void hex_dump(void *pkt, size_t length, u64 addr)
 	printf("\n");
 }
 
-static size_t gen_eth_frame(struct xsk_umem_info *umem, u64 addr)
+static void *memset32_htonl(void *dest, u32 val, u32 size)
+{
+	u32 *ptr = (u32 *)dest;
+	int i;
+
+	val = htonl(val);
+
+	for (i = 0; i < (size & (~0x3)); i += 4)
+		ptr[i >> 2] = val;
+
+	for (; i < size; i++)
+		((char *)dest)[i] = ((char *)&val)[i & 3];
+
+	return dest;
+}
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+static inline unsigned short from32to16(unsigned int x)
+{
+	/* add up 16-bit and 16-bit for 16+c bit */
+	x = (x & 0xffff) + (x >> 16);
+	/* add up carry.. */
+	x = (x & 0xffff) + (x >> 16);
+	return x;
+}
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+static unsigned int do_csum(const unsigned char *buff, int len)
+{
+	unsigned int result = 0;
+	int odd;
+
+	if (len <= 0)
+		goto out;
+	odd = 1 & (unsigned long)buff;
+	if (odd) {
+#ifdef __LITTLE_ENDIAN
+		result += (*buff << 8);
+#else
+		result = *buff;
+#endif
+		len--;
+		buff++;
+	}
+	if (len >= 2) {
+		if (2 & (unsigned long)buff) {
+			result += *(unsigned short *)buff;
+			len -= 2;
+			buff += 2;
+		}
+		if (len >= 4) {
+			const unsigned char *end = buff +
+						   ((unsigned int)len & ~3);
+			unsigned int carry = 0;
+
+			do {
+				unsigned int w = *(unsigned int *)buff;
+
+				buff += 4;
+				result += carry;
+				result += w;
+				carry = (w > result);
+			} while (buff < end);
+			result += carry;
+			result = (result & 0xffff) + (result >> 16);
+		}
+		if (len & 2) {
+			result += *(unsigned short *)buff;
+			buff += 2;
+		}
+	}
+	if (len & 1)
+#ifdef __LITTLE_ENDIAN
+		result += *buff;
+#else
+		result += (*buff << 8);
+#endif
+	result = from32to16(result);
+	if (odd)
+		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
+out:
+	return result;
+}
+
+__sum16 ip_fast_csum(const void *iph, unsigned int ihl);
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ *	This function code has been taken from
+ *	Linux kernel lib/checksum.c
+ */
+__sum16 ip_fast_csum(const void *iph, unsigned int ihl)
+{
+	return (__force __sum16)~do_csum(iph, ihl * 4);
+}
+
+/*
+ * Fold a partial checksum
+ * This function code has been taken from
+ * Linux kernel include/asm-generic/checksum.h
+ */
+static inline __sum16 csum_fold(__wsum csum)
+{
+	u32 sum = (__force u32)csum;
+
+	sum = (sum & 0xffff) + (sum >> 16);
+	sum = (sum & 0xffff) + (sum >> 16);
+	return (__force __sum16)~sum;
+}
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+static inline u32 from64to32(u64 x)
+{
+	/* add up 32-bit and 32-bit for 32+c bit */
+	x = (x & 0xffffffff) + (x >> 32);
+	/* add up carry.. */
+	x = (x & 0xffffffff) + (x >> 32);
+	return (u32)x;
+}
+
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum);
+
+/*
+ * This function code has been taken from
+ * Linux kernel lib/checksum.c
+ */
+__wsum csum_tcpudp_nofold(__be32 saddr, __be32 daddr,
+			  __u32 len, __u8 proto, __wsum sum)
+{
+	unsigned long long s = (__force u32)sum;
+
+	s += (__force u32)saddr;
+	s += (__force u32)daddr;
+#ifdef __BIG_ENDIAN__
+	s += proto + len;
+#else
+	s += (proto + len) << 8;
+#endif
+	return (__force __wsum)from64to32(s);
+}
+
+/*
+ * This function has been taken from
+ * Linux kernel include/asm-generic/checksum.h
+ */
+static inline __sum16
+csum_tcpudp_magic(__be32 saddr, __be32 daddr, __u32 len,
+		  __u8 proto, __wsum sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
+}
+
+static inline u16 udp_csum(u32 saddr, u32 daddr, u32 len,
+			   u8 proto, u16 *udp_pkt)
+{
+	u32 csum = 0;
+	u32 cnt = 0;
+
+	/* udp hdr and data */
+	for (; cnt < len; cnt += 2)
+		csum += udp_pkt[cnt >> 1];
+
+	return csum_tcpudp_magic(saddr, daddr, len, proto, csum);
+}
+
+#define ETH_FCS_SIZE 4
+
+#define PKT_HDR_SIZE (sizeof(struct ethhdr) + sizeof(struct iphdr) + \
+		      sizeof(struct udphdr))
+
+#define PKT_SIZE		(opt_pkt_size - ETH_FCS_SIZE)
+#define IP_PKT_SIZE		(PKT_SIZE - sizeof(struct ethhdr))
+#define UDP_PKT_SIZE		(IP_PKT_SIZE - sizeof(struct iphdr))
+#define UDP_PKT_DATA_SIZE	(UDP_PKT_SIZE - sizeof(struct udphdr))
+
+static u8 pkt_data[XSK_UMEM__DEFAULT_FRAME_SIZE];
+
+static void gen_eth_hdr_data(void)
+{
+	struct udphdr *udp_hdr = (struct udphdr *)(pkt_data +
+						   sizeof(struct ethhdr) +
+						   sizeof(struct iphdr));
+	struct iphdr *ip_hdr = (struct iphdr *)(pkt_data +
+						sizeof(struct ethhdr));
+	struct ethhdr *eth_hdr = (struct ethhdr *)pkt_data;
+
+	/* ethernet header */
+	memcpy(eth_hdr->h_dest, "\x3c\xfd\xfe\x9e\x7f\x71", ETH_ALEN);
+	memcpy(eth_hdr->h_source, "\xec\xb1\xd7\x98\x3a\xc0", ETH_ALEN);
+	eth_hdr->h_proto = htons(ETH_P_IP);
+
+	/* IP header */
+	ip_hdr->version = IPVERSION;
+	ip_hdr->ihl = 0x5; /* 20 byte header */
+	ip_hdr->tos = 0x0;
+	ip_hdr->tot_len = htons(IP_PKT_SIZE);
+	ip_hdr->id = 0;
+	ip_hdr->frag_off = 0;
+	ip_hdr->ttl = IPDEFTTL;
+	ip_hdr->protocol = IPPROTO_UDP;
+	ip_hdr->saddr = htonl(0x0a0a0a10);
+	ip_hdr->daddr = htonl(0x0a0a0a20);
+
+	/* IP header checksum */
+	ip_hdr->check = 0;
+	ip_hdr->check = ip_fast_csum((const void *)ip_hdr, ip_hdr->ihl);
+
+	/* UDP header */
+	udp_hdr->source = htons(0x1000);
+	udp_hdr->dest = htons(0x1000);
+	udp_hdr->len = htons(UDP_PKT_SIZE);
+
+	/* UDP data */
+	memset32_htonl(pkt_data + PKT_HDR_SIZE, opt_pkt_fill_pattern,
+		       UDP_PKT_DATA_SIZE);
+
+	/* UDP header checksum */
+	udp_hdr->check = 0;
+	udp_hdr->check = udp_csum(ip_hdr->saddr, ip_hdr->daddr, UDP_PKT_SIZE,
+				  IPPROTO_UDP, (u16 *)udp_hdr);
+}
+
+static void gen_eth_frame(struct xsk_umem_info *umem, u64 addr)
 {
 	memcpy(xsk_umem__get_data(umem->buffer, addr), pkt_data,
-	       sizeof(pkt_data) - 1);
-	return sizeof(pkt_data) - 1;
+	       PKT_SIZE);
 }
 
 static struct xsk_umem_info *xsk_configure_umem(void *buffer, u64 size)
@@ -375,6 +626,11 @@ static struct option long_options[] = {
 	{"unaligned", no_argument, 0, 'u'},
 	{"shared-umem", no_argument, 0, 'M'},
 	{"force", no_argument, 0, 'F'},
+	{"duration", required_argument, 0, 'd'},
+	{"batch-size", required_argument, 0, 'b'},
+	{"tx-pkt-count", required_argument, 0, 'C'},
+	{"tx-pkt-size", required_argument, 0, 's'},
+	{"tx-pkt-pattern", required_argument, 0, 'P'},
 	{0, 0, 0, 0}
 };
 
@@ -399,8 +655,21 @@ static void usage(const char *prog)
 		"  -u, --unaligned	Enable unaligned chunk placement\n"
 		"  -M, --shared-umem	Enable XDP_SHARED_UMEM\n"
 		"  -F, --force		Force loading the XDP prog\n"
+		"  -d, --duration=n	Duration in secs to run command.\n"
+		"			Default: forever.\n"
+		"  -b, --batch-size=n	Batch size for sending or receiving\n"
+		"			packets. Default: %d\n"
+		"  -C, --tx-pkt-count=n	Number of packets to send.\n"
+		"			Default: Continuous packets.\n"
+		"  -s, --tx-pkt-size=n	Transmit packet size.\n"
+		"			(Default: %d bytes)\n"
+		"			Min size: %d, Max size %d.\n"
+		"  -P, --tx-pkt-pattern=nPacket fill pattern. Default: 0x%x\n"
 		"\n";
-	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE);
+	fprintf(stderr, str, prog, XSK_UMEM__DEFAULT_FRAME_SIZE,
+		opt_batch_size, MIN_PKT_SIZE, MIN_PKT_SIZE,
+		XSK_UMEM__DEFAULT_FRAME_SIZE, opt_pkt_fill_pattern);
+
 	exit(EXIT_FAILURE);
 }
 
@@ -411,7 +680,7 @@ static void parse_command_line(int argc, char **argv)
 	opterr = 0;
 
 	for (;;) {
-		c = getopt_long(argc, argv, "Frtli:q:psSNn:czf:muM",
+		c = getopt_long(argc, argv, "Frtli:q:pSNn:czf:muMd:b:C:s:P:",
 				long_options, &option_index);
 		if (c == -1)
 			break;
@@ -440,7 +709,7 @@ static void parse_command_line(int argc, char **argv)
 			opt_xdp_bind_flags |= XDP_COPY;
 			break;
 		case 'N':
-			opt_xdp_flags |= XDP_FLAGS_DRV_MODE;
+			/* default, set below */
 			break;
 		case 'n':
 			opt_interval = atoi(optarg);
@@ -469,11 +738,37 @@ static void parse_command_line(int argc, char **argv)
 		case 'M':
 			opt_num_xsks = MAX_SOCKS;
 			break;
+		case 'd':
+			opt_duration = atoi(optarg);
+			opt_duration *= 1000000000;
+			break;
+		case 'b':
+			opt_batch_size = atoi(optarg);
+			break;
+		case 'C':
+			opt_pkt_count = atoi(optarg);
+			break;
+		case 's':
+			opt_pkt_size = atoi(optarg);
+			if (opt_pkt_size > (XSK_UMEM__DEFAULT_FRAME_SIZE) ||
+			    opt_pkt_size < MIN_PKT_SIZE) {
+				fprintf(stderr,
+					"ERROR: Invalid frame size %d\n",
+					opt_pkt_size);
+				usage(basename(argv[0]));
+			}
+			break;
+		case 'P':
+			opt_pkt_fill_pattern = strtol(optarg, NULL, 16);
+			break;
 		default:
 			usage(basename(argv[0]));
 		}
 	}
 
+	if (!(opt_xdp_flags & XDP_FLAGS_SKB_MODE))
+		opt_xdp_flags |= XDP_FLAGS_DRV_MODE;
+
 	opt_ifindex = if_nametoindex(opt_if);
 	if (!opt_ifindex) {
 		fprintf(stderr, "ERROR: interface \"%s\" does not exist\n",
@@ -513,7 +808,7 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
 	if (!opt_need_wakeup || xsk_ring_prod__needs_wakeup(&xsk->tx))
 		kick_tx(xsk);
 
-	ndescs = (xsk->outstanding_tx > BATCH_SIZE) ? BATCH_SIZE :
+	ndescs = (xsk->outstanding_tx > opt_batch_size) ? opt_batch_size :
 		xsk->outstanding_tx;
 
 	/* re-add completed Tx buffers */
@@ -542,7 +837,8 @@ static inline void complete_tx_l2fwd(struct xsk_socket_info *xsk,
 	}
 }
 
-static inline void complete_tx_only(struct xsk_socket_info *xsk)
+static inline void complete_tx_only(struct xsk_socket_info *xsk,
+				    int batch_size)
 {
 	unsigned int rcvd;
 	u32 idx;
@@ -553,7 +849,7 @@ static inline void complete_tx_only(struct xsk_socket_info *xsk)
 	if (!opt_need_wakeup || xsk_ring_prod__needs_wakeup(&xsk->tx))
 		kick_tx(xsk);
 
-	rcvd = xsk_ring_cons__peek(&xsk->umem->cq, BATCH_SIZE, &idx);
+	rcvd = xsk_ring_cons__peek(&xsk->umem->cq, batch_size, &idx);
 	if (rcvd > 0) {
 		xsk_ring_cons__release(&xsk->umem->cq, rcvd);
 		xsk->outstanding_tx -= rcvd;
@@ -567,7 +863,7 @@ static void rx_drop(struct xsk_socket_info *xsk, struct pollfd *fds)
 	u32 idx_rx = 0, idx_fq = 0;
 	int ret;
 
-	rcvd = xsk_ring_cons__peek(&xsk->rx, BATCH_SIZE, &idx_rx);
+	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq))
 			ret = poll(fds, num_socks, opt_timeout);
@@ -619,36 +915,68 @@ static void rx_drop_all(void)
 
 		for (i = 0; i < num_socks; i++)
 			rx_drop(xsks[i], fds);
+
+		if (benchmark_done)
+			break;
 	}
 }
 
-static void tx_only(struct xsk_socket_info *xsk, u32 frame_nb)
+static void tx_only(struct xsk_socket_info *xsk, u32 frame_nb, int batch_size)
 {
 	u32 idx;
+	unsigned int i;
 
-	if (xsk_ring_prod__reserve(&xsk->tx, BATCH_SIZE, &idx) == BATCH_SIZE) {
-		unsigned int i;
-
-		for (i = 0; i < BATCH_SIZE; i++) {
-			xsk_ring_prod__tx_desc(&xsk->tx, idx + i)->addr	=
-				(frame_nb + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT;
-			xsk_ring_prod__tx_desc(&xsk->tx, idx + i)->len =
-				sizeof(pkt_data) - 1;
-		}
+	while (xsk_ring_prod__reserve(&xsk->tx, batch_size, &idx) <
+				      batch_size) {
+		complete_tx_only(xsk, batch_size);
+	}
 
-		xsk_ring_prod__submit(&xsk->tx, BATCH_SIZE);
-		xsk->outstanding_tx += BATCH_SIZE;
-		frame_nb += BATCH_SIZE;
-		frame_nb %= NUM_FRAMES;
+	for (i = 0; i < batch_size; i++) {
+		struct xdp_desc *tx_desc = xsk_ring_prod__tx_desc(&xsk->tx,
+								  idx + i);
+		tx_desc->addr = (frame_nb + i) << XSK_UMEM__DEFAULT_FRAME_SHIFT;
+		tx_desc->len = PKT_SIZE;
 	}
 
-	complete_tx_only(xsk);
+	xsk_ring_prod__submit(&xsk->tx, batch_size);
+	xsk->outstanding_tx += batch_size;
+	frame_nb += batch_size;
+	frame_nb %= NUM_FRAMES;
+	complete_tx_only(xsk, batch_size);
+}
+
+static inline int get_batch_size(int pkt_cnt)
+{
+	if (!opt_pkt_count)
+		return opt_batch_size;
+
+	if (pkt_cnt + opt_batch_size <= opt_pkt_count)
+		return opt_batch_size;
+
+	return opt_pkt_count - pkt_cnt;
+}
+
+static void complete_tx_only_all(void)
+{
+	bool pending;
+	int i;
+
+	do {
+		pending = false;
+		for (i = 0; i < num_socks; i++) {
+			if (xsks[i]->outstanding_tx) {
+				complete_tx_only(xsks[i], opt_batch_size);
+				pending = !!xsks[i]->outstanding_tx;
+			}
+		}
+	} while (pending);
 }
 
 static void tx_only_all(void)
 {
 	struct pollfd fds[MAX_SOCKS] = {};
 	u32 frame_nb[MAX_SOCKS] = {};
+	int pkt_cnt = 0;
 	int i, ret;
 
 	for (i = 0; i < num_socks; i++) {
@@ -656,7 +984,9 @@ static void tx_only_all(void)
 		fds[0].events = POLLOUT;
 	}
 
-	for (;;) {
+	while ((opt_pkt_count && pkt_cnt < opt_pkt_count) || !opt_pkt_count) {
+		int batch_size = get_batch_size(pkt_cnt);
+
 		if (opt_poll) {
 			ret = poll(fds, num_socks, opt_timeout);
 			if (ret <= 0)
@@ -667,8 +997,16 @@ static void tx_only_all(void)
 		}
 
 		for (i = 0; i < num_socks; i++)
-			tx_only(xsks[i], frame_nb[i]);
+			tx_only(xsks[i], frame_nb[i], batch_size);
+
+		pkt_cnt += batch_size;
+
+		if (benchmark_done)
+			break;
 	}
+
+	if (opt_pkt_count)
+		complete_tx_only_all();
 }
 
 static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
@@ -679,7 +1017,7 @@ static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)
 
 	complete_tx_l2fwd(xsk, fds);
 
-	rcvd = xsk_ring_cons__peek(&xsk->rx, BATCH_SIZE, &idx_rx);
+	rcvd = xsk_ring_cons__peek(&xsk->rx, opt_batch_size, &idx_rx);
 	if (!rcvd) {
 		if (xsk_ring_prod__needs_wakeup(&xsk->umem->fq))
 			ret = poll(fds, num_socks, opt_timeout);
@@ -736,6 +1074,9 @@ static void l2fwd_all(void)
 
 		for (i = 0; i < num_socks; i++)
 			l2fwd(xsks[i], fds);
+
+		if (benchmark_done)
+			break;
 	}
 }
 
@@ -831,9 +1172,12 @@ int main(int argc, char **argv)
 	for (i = 0; i < opt_num_xsks; i++)
 		xsks[num_socks++] = xsk_configure_socket(umem, rx, tx);
 
-	if (opt_bench == BENCH_TXONLY)
+	if (opt_bench == BENCH_TXONLY) {
+		gen_eth_hdr_data();
+
 		for (i = 0; i < NUM_FRAMES; i++)
 			gen_eth_frame(umem, i * opt_xsk_frame_size);
+	}
 
 	if (opt_num_xsks > 1 && opt_bench != BENCH_TXONLY)
 		enter_xsks_into_map(obj);
@@ -849,6 +1193,7 @@ int main(int argc, char **argv)
 		exit_with_error(ret);
 
 	prev_time = get_nsecs();
+	start_time = prev_time;
 
 	if (opt_bench == BENCH_RXDROP)
 		rx_drop_all();
@@ -857,5 +1202,11 @@ int main(int argc, char **argv)
 	else
 		l2fwd_all();
 
+	benchmark_done = true;
+
+	pthread_join(pt, NULL);
+
+	xdpsock_cleanup();
+
 	return 0;
 }
diff --git a/tools/bpf/bpftool/Documentation/bpftool-gen.rst b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
new file mode 100644
index 000000000000..86a87da97d0b
--- /dev/null
+++ b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
@@ -0,0 +1,305 @@
+================
+bpftool-gen
+================
+-------------------------------------------------------------------------------
+tool for BPF code-generation
+-------------------------------------------------------------------------------
+
+:Manual section: 8
+
+SYNOPSIS
+========
+
+	**bpftool** [*OPTIONS*] **gen** *COMMAND*
+
+	*OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] }
+
+	*COMMAND* := { **skeleton | **help** }
+
+GEN COMMANDS
+=============
+
+|	**bpftool** **gen skeleton** *FILE*
+|	**bpftool** **gen help**
+
+DESCRIPTION
+===========
+	**bpftool gen skeleton** *FILE*
+		  Generate BPF skeleton C header file for a given *FILE*.
+
+		  BPF skeleton is an alternative interface to existing libbpf
+		  APIs for working with BPF objects. Skeleton code is intended
+		  to significantly shorten and simplify code to load and work
+		  with BPF programs from userspace side. Generated code is
+		  tailored to specific input BPF object *FILE*, reflecting its
+		  structure by listing out available maps, program, variables,
+		  etc. Skeleton eliminates the need to lookup mentioned
+		  components by name. Instead, if skeleton instantiation
+		  succeeds, they are populated in skeleton structure as valid
+		  libbpf types (e.g., struct bpf_map pointer) and can be
+		  passed to existing generic libbpf APIs.
+
+		  In addition to simple and reliable access to maps and
+		  programs, skeleton provides a storage for BPF links (struct
+		  bpf_link) for each BPF program within BPF object. When
+		  requested, supported BPF programs will be automatically
+		  attached and resulting BPF links stored for further use by
+		  user in pre-allocated fields in skeleton struct. For BPF
+		  programs that can't be automatically attached by libbpf,
+		  user can attach them manually, but store resulting BPF link
+		  in per-program link field. All such set up links will be
+		  automatically destroyed on BPF skeleton destruction. This
+		  eliminates the need for users to manage links manually and
+		  rely on libbpf support to detach programs and free up
+		  resources.
+
+		  Another facility provided by BPF skeleton is an interface to
+		  global variables of all supported kinds: mutable, read-only,
+		  as well as extern ones. This interface allows to pre-setup
+		  initial values of variables before BPF object is loaded and
+		  verified by kernel. For non-read-only variables, the same
+		  interface can be used to fetch values of global variables on
+		  userspace side, even if they are modified by BPF code.
+
+		  During skeleton generation, contents of source BPF object
+		  *FILE* is embedded within generated code and is thus not
+		  necessary to keep around. This ensures skeleton and BPF
+		  object file are matching 1-to-1 and always stay in sync.
+		  Generated code is dual-licensed under LGPL-2.1 and
+		  BSD-2-Clause licenses.
+
+		  It is a design goal and guarantee that skeleton interfaces
+		  are interoperable with generic libbpf APIs. User should
+		  always be able to use skeleton API to create and load BPF
+		  object, and later use libbpf APIs to keep working with
+		  specific maps, programs, etc.
+
+		  As part of skeleton, few custom functions are generated.
+		  Each of them is prefixed with object name, derived from
+		  object file name. I.e., if BPF object file name is
+		  **example.o**, BPF object name will be **example**. The
+		  following custom functions are provided in such case:
+
+		  - **example__open** and **example__open_opts**.
+		    These functions are used to instantiate skeleton. It
+		    corresponds to libbpf's **bpf_object__open()** API.
+		    **_opts** variants accepts extra **bpf_object_open_opts**
+		    options.
+
+		  - **example__load**.
+		    This function creates maps, loads and verifies BPF
+		    programs, initializes global data maps. It corresponds to
+		    libppf's **bpf_object__load** API.
+
+		  - **example__open_and_load** combines **example__open** and
+		    **example__load** invocations in one commonly used
+		    operation.
+
+		  - **example__attach** and **example__detach**
+		    This pair of functions allow to attach and detach,
+		    correspondingly, already loaded BPF object. Only BPF
+		    programs of types supported by libbpf for auto-attachment
+		    will be auto-attached and their corresponding BPF links
+		    instantiated. For other BPF programs, user can manually
+		    create a BPF link and assign it to corresponding fields in
+		    skeleton struct. **example__detach** will detach both
+		    links created automatically, as well as those populated by
+		    user manually.
+
+		  - **example__destroy**
+		    Detach and unload BPF programs, free up all the resources
+		    used by skeleton and BPF object.
+
+		  If BPF object has global variables, corresponding structs
+		  with memory layout corresponding to global data data section
+		  layout will be created. Currently supported ones are: *.data*,
+		  *.bss*, *.rodata*, and *.kconfig* structs/data sections.
+		  These data sections/structs can be used to set up initial
+		  values of variables, if set before **example__load**.
+		  Afterwards, if target kernel supports memory-mapped BPF
+		  arrays, same structs can be used to fetch and update
+		  (non-read-only) data from userspace, with same simplicity
+		  as for BPF side.
+
+	**bpftool gen help**
+		  Print short help message.
+
+OPTIONS
+=======
+	-h, --help
+		  Print short generic help message (similar to **bpftool help**).
+
+	-V, --version
+		  Print version number (similar to **bpftool version**).
+
+	-j, --json
+		  Generate JSON output. For commands that cannot produce JSON,
+		  this option has no effect.
+
+	-p, --pretty
+		  Generate human-readable JSON output. Implies **-j**.
+
+	-d, --debug
+		  Print all logs available from libbpf, including debug-level
+		  information.
+
+EXAMPLES
+========
+**$ cat example.c**
+::
+
+  #include <stdbool.h>
+  #include <linux/ptrace.h>
+  #include <linux/bpf.h>
+  #include "bpf_helpers.h"
+
+  const volatile int param1 = 42;
+  bool global_flag = true;
+  struct { int x; } data = {};
+
+  struct {
+  	__uint(type, BPF_MAP_TYPE_HASH);
+  	__uint(max_entries, 128);
+  	__type(key, int);
+  	__type(value, long);
+  } my_map SEC(".maps");
+
+  SEC("raw_tp/sys_enter")
+  int handle_sys_enter(struct pt_regs *ctx)
+  {
+  	static long my_static_var;
+  	if (global_flag)
+  		my_static_var++;
+  	else
+  		data.x += param1;
+  	return 0;
+  }
+
+  SEC("raw_tp/sys_exit")
+  int handle_sys_exit(struct pt_regs *ctx)
+  {
+  	int zero = 0;
+  	bpf_map_lookup_elem(&my_map, &zero);
+  	return 0;
+  }
+
+This is example BPF application with two BPF programs and a mix of BPF maps
+and global variables.
+
+**$ bpftool gen skeleton example.o**
+::
+
+  /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+  /* THIS FILE IS AUTOGENERATED! */
+  #ifndef __EXAMPLE_SKEL_H__
+  #define __EXAMPLE_SKEL_H__
+
+  #include <stdlib.h>
+  #include <libbpf.h>
+
+  struct example {
+  	struct bpf_object_skeleton *skeleton;
+  	struct bpf_object *obj;
+  	struct {
+  		struct bpf_map *rodata;
+  		struct bpf_map *data;
+  		struct bpf_map *bss;
+  		struct bpf_map *my_map;
+  	} maps;
+  	struct {
+  		struct bpf_program *handle_sys_enter;
+  		struct bpf_program *handle_sys_exit;
+  	} progs;
+  	struct {
+  		struct bpf_link *handle_sys_enter;
+  		struct bpf_link *handle_sys_exit;
+  	} links;
+  	struct example__bss {
+  		struct {
+  			int x;
+  		} data;
+  	} *bss;
+  	struct example__data {
+  		_Bool global_flag;
+  		long int handle_sys_enter_my_static_var;
+  	} *data;
+  	struct example__rodata {
+  		int param1;
+  	} *rodata;
+  };
+
+  static void example__destroy(struct example *obj);
+  static inline struct example *example__open_opts(
+                const struct bpf_object_open_opts *opts);
+  static inline struct example *example__open();
+  static inline int example__load(struct example *obj);
+  static inline struct example *example__open_and_load();
+  static inline int example__attach(struct example *obj);
+  static inline void example__detach(struct example *obj);
+
+  #endif /* __EXAMPLE_SKEL_H__ */
+
+**$ cat example_user.c**
+::
+
+  #include "example.skel.h"
+
+  int main()
+  {
+  	struct example *skel;
+  	int err = 0;
+
+  	skel = example__open();
+  	if (!skel)
+  		goto cleanup;
+
+  	skel->rodata->param1 = 128;
+
+  	err = example__load(skel);
+  	if (err)
+  		goto cleanup;
+
+  	err = example__attach(skel);
+  	if (err)
+  		goto cleanup;
+
+  	/* all libbpf APIs are usable */
+  	printf("my_map name: %s\n", bpf_map__name(skel->maps.my_map));
+  	printf("sys_enter prog FD: %d\n",
+  	       bpf_program__fd(skel->progs.handle_sys_enter));
+
+  	/* detach and re-attach sys_exit program */
+  	bpf_link__destroy(skel->links.handle_sys_exit);
+  	skel->links.handle_sys_exit =
+  		bpf_program__attach(skel->progs.handle_sys_exit);
+
+  	printf("my_static_var: %ld\n",
+  	       skel->bss->handle_sys_enter_my_static_var);
+
+  cleanup:
+  	example__destroy(skel);
+  	return err;
+  }
+
+**# ./example_user**
+::
+
+  my_map name: my_map
+  sys_enter prog FD: 8
+  my_static_var: 7
+
+This is a stripped-out version of skeleton generated for above example code.
+
+SEE ALSO
+========
+	**bpf**\ (2),
+	**bpf-helpers**\ (7),
+	**bpftool**\ (8),
+	**bpftool-map**\ (8),
+	**bpftool-prog**\ (8),
+	**bpftool-cgroup**\ (8),
+	**bpftool-feature**\ (8),
+	**bpftool-net**\ (8),
+	**bpftool-perf**\ (8),
+	**bpftool-btf**\ (8)
diff --git a/tools/bpf/bpftool/Documentation/bpftool-map.rst b/tools/bpf/bpftool/Documentation/bpftool-map.rst
index 1c0f7146aab0..cdeae8ae90ba 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-map.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-map.rst
@@ -39,9 +39,9 @@ MAP COMMANDS
 |	**bpftool** **map freeze**     *MAP*
 |	**bpftool** **map help**
 |
-|	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
+|	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* | **name** *MAP_NAME* }
 |	*DATA* := { [**hex**] *BYTES* }
-|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* }
+|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
 |	*VALUE* := { *DATA* | *MAP* | *PROG* }
 |	*UPDATE_FLAGS* := { **any** | **exist** | **noexist** }
 |	*TYPE* := { **hash** | **array** | **prog_array** | **perf_event_array** | **percpu_hash**
@@ -55,8 +55,9 @@ DESCRIPTION
 ===========
 	**bpftool map { show | list }**   [*MAP*]
 		  Show information about loaded maps.  If *MAP* is specified
-		  show information only about given map, otherwise list all
-		  maps currently loaded on the system.
+		  show information only about given maps, otherwise list all
+		  maps currently loaded on the system.  In case of **name**,
+		  *MAP* may match several maps which will all be shown.
 
 		  Output will start with map ID followed by map type and
 		  zero or more named attributes (depending on kernel version).
@@ -66,7 +67,8 @@ DESCRIPTION
 		  as *FILE*.
 
 	**bpftool map dump**    *MAP*
-		  Dump all entries in a given *MAP*.
+		  Dump all entries in a given *MAP*.  In case of **name**,
+		  *MAP* may match several maps which will all be dumped.
 
 	**bpftool map update**  *MAP* [**key** *DATA*] [**value** *VALUE*] [*UPDATE_FLAGS*]
 		  Update map entry for a given *KEY*.
diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.rst b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
index 7a374b3c851d..64ddf8a4c518 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-prog.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-prog.rst
@@ -33,7 +33,7 @@ PROG COMMANDS
 |	**bpftool** **prog help**
 |
 |	*MAP* := { **id** *MAP_ID* | **pinned** *FILE* }
-|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* }
+|	*PROG* := { **id** *PROG_ID* | **pinned** *FILE* | **tag** *PROG_TAG* | **name** *PROG_NAME* }
 |	*TYPE* := {
 |		**socket** | **kprobe** | **kretprobe** | **classifier** | **action** |
 |		**tracepoint** | **raw_tracepoint** | **xdp** | **perf_event** | **cgroup/skb** |
@@ -53,8 +53,10 @@ DESCRIPTION
 ===========
 	**bpftool prog { show | list }** [*PROG*]
 		  Show information about loaded programs.  If *PROG* is
-		  specified show information only about given program, otherwise
-		  list all programs currently loaded on the system.
+		  specified show information only about given programs,
+		  otherwise list all programs currently loaded on the system.
+		  In case of **tag** or **name**, *PROG* may match several
+		  programs which will all be shown.
 
 		  Output will start with program ID followed by program type and
 		  zero or more named attributes (depending on kernel version).
@@ -68,11 +70,15 @@ DESCRIPTION
 		  performed via the **kernel.bpf_stats_enabled** sysctl knob.
 
 	**bpftool prog dump xlated** *PROG* [{ **file** *FILE* | **opcodes** | **visual** | **linum** }]
-		  Dump eBPF instructions of the program from the kernel. By
+		  Dump eBPF instructions of the programs from the kernel. By
 		  default, eBPF will be disassembled and printed to standard
 		  output in human-readable format. In this case, **opcodes**
 		  controls if raw opcodes should be printed as well.
 
+		  In case of **tag** or **name**, *PROG* may match several
+		  programs which will all be dumped.  However, if **file** or
+		  **visual** is specified, *PROG* must match a single program.
+
 		  If **file** is specified, the binary image will instead be
 		  written to *FILE*.
 
@@ -80,15 +86,17 @@ DESCRIPTION
 		  built instead, and eBPF instructions will be presented with
 		  CFG in DOT format, on standard output.
 
-		  If the prog has line_info available, the source line will
+		  If the programs have line_info available, the source line will
 		  be displayed by default.  If **linum** is specified,
 		  the filename, line number and line column will also be
 		  displayed on top of the source line.
 
 	**bpftool prog dump jited**  *PROG* [{ **file** *FILE* | **opcodes** | **linum** }]
 		  Dump jited image (host machine code) of the program.
+
 		  If *FILE* is specified image will be written to a file,
 		  otherwise it will be disassembled and printed to stdout.
+		  *PROG* must match a single program when **file** is specified.
 
 		  **opcodes** controls if raw opcodes will be printed.
 
diff --git a/tools/bpf/bpftool/Documentation/bpftool.rst b/tools/bpf/bpftool/Documentation/bpftool.rst
index 6a9c52ef84a9..34239fda69ed 100644
--- a/tools/bpf/bpftool/Documentation/bpftool.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool.rst
@@ -81,4 +81,5 @@ SEE ALSO
 	**bpftool-feature**\ (8),
 	**bpftool-net**\ (8),
 	**bpftool-perf**\ (8),
-	**bpftool-btf**\ (8)
+	**bpftool-btf**\ (8),
+	**bpftool-gen**\ (8),
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index 70493a6da206..754d8395e451 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -59,6 +59,21 @@ _bpftool_get_map_ids_for_type()
         command sed -n 's/.*"id": \(.*\),$/\1/p' )" -- "$cur" ) )
 }
 
+_bpftool_get_map_names()
+{
+    COMPREPLY+=( $( compgen -W "$( bpftool -jp map  2>&1 | \
+        command sed -n 's/.*"name": \(.*\),$/\1/p' )" -- "$cur" ) )
+}
+
+# Takes map type and adds matching map names to the list of suggestions.
+_bpftool_get_map_names_for_type()
+{
+    local type="$1"
+    COMPREPLY+=( $( compgen -W "$( bpftool -jp map  2>&1 | \
+        command grep -C2 "$type" | \
+        command sed -n 's/.*"name": \(.*\),$/\1/p' )" -- "$cur" ) )
+}
+
 _bpftool_get_prog_ids()
 {
     COMPREPLY+=( $( compgen -W "$( bpftool -jp prog 2>&1 | \
@@ -71,6 +86,12 @@ _bpftool_get_prog_tags()
         command sed -n 's/.*"tag": "\(.*\)",$/\1/p' )" -- "$cur" ) )
 }
 
+_bpftool_get_prog_names()
+{
+    COMPREPLY+=( $( compgen -W "$( bpftool -jp prog 2>&1 | \
+        command sed -n 's/.*"name": "\(.*\)",$/\1/p' )" -- "$cur" ) )
+}
+
 _bpftool_get_btf_ids()
 {
     COMPREPLY+=( $( compgen -W "$( bpftool -jp btf 2>&1 | \
@@ -180,6 +201,52 @@ _bpftool_map_update_get_id()
     esac
 }
 
+_bpftool_map_update_get_name()
+{
+    local command="$1"
+
+    # Is it the map to update, or a map to insert into the map to update?
+    # Search for "value" keyword.
+    local idx value
+    for (( idx=7; idx < ${#words[@]}-1; idx++ )); do
+        if [[ ${words[idx]} == "value" ]]; then
+            value=1
+            break
+        fi
+    done
+    if [[ $value -eq 0 ]]; then
+        case "$command" in
+            push)
+                _bpftool_get_map_names_for_type stack
+                ;;
+            enqueue)
+                _bpftool_get_map_names_for_type queue
+                ;;
+            *)
+                _bpftool_get_map_names
+                ;;
+        esac
+        return 0
+    fi
+
+    # Name to complete is for a value. It can be either prog name or map name. This
+    # depends on the type of the map to update.
+    local type=$(_bpftool_map_guess_map_type)
+    case $type in
+        array_of_maps|hash_of_maps)
+            _bpftool_get_map_names
+            return 0
+            ;;
+        prog_array)
+            _bpftool_get_prog_names
+            return 0
+            ;;
+        *)
+            return 0
+            ;;
+    esac
+}
+
 _bpftool()
 {
     local cur prev words objword
@@ -251,7 +318,8 @@ _bpftool()
     # Completion depends on object and command in use
     case $object in
         prog)
-            # Complete id, only for subcommands that use prog (but no map) ids
+            # Complete id and name, only for subcommands that use prog (but no
+            # map) ids/names.
             case $command in
                 show|list|dump|pin)
                     case $prev in
@@ -259,12 +327,16 @@ _bpftool()
                             _bpftool_get_prog_ids
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_prog_names
+                            return 0
+                            ;;
                     esac
                     ;;
             esac
 
-            local PROG_TYPE='id pinned tag'
-            local MAP_TYPE='id pinned'
+            local PROG_TYPE='id pinned tag name'
+            local MAP_TYPE='id pinned name'
             case $command in
                 show|list)
                     [[ $prev != "$command" ]] && return 0
@@ -315,6 +387,9 @@ _bpftool()
                                 id)
                                     _bpftool_get_prog_ids
                                     ;;
+                                name)
+                                    _bpftool_get_map_names
+                                    ;;
                                 pinned)
                                     _filedir
                                     ;;
@@ -335,6 +410,9 @@ _bpftool()
                                 id)
                                     _bpftool_get_map_ids
                                     ;;
+                                name)
+                                    _bpftool_get_map_names
+                                    ;;
                                 pinned)
                                     _filedir
                                     ;;
@@ -399,6 +477,10 @@ _bpftool()
                             _bpftool_get_map_ids
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_map_names
+                            return 0
+                            ;;
                         pinned|pinmaps)
                             _filedir
                             return 0
@@ -447,7 +529,7 @@ _bpftool()
             esac
             ;;
         map)
-            local MAP_TYPE='id pinned'
+            local MAP_TYPE='id pinned name'
             case $command in
                 show|list|dump|peek|pop|dequeue|freeze)
                     case $prev in
@@ -473,6 +555,24 @@ _bpftool()
                             esac
                             return 0
                             ;;
+                        name)
+                            case "$command" in
+                                peek)
+                                    _bpftool_get_map_names_for_type stack
+                                    _bpftool_get_map_names_for_type queue
+                                    ;;
+                                pop)
+                                    _bpftool_get_map_names_for_type stack
+                                    ;;
+                                dequeue)
+                                    _bpftool_get_map_names_for_type queue
+                                    ;;
+                                *)
+                                    _bpftool_get_map_names
+                                    ;;
+                            esac
+                            return 0
+                            ;;
                         *)
                             return 0
                             ;;
@@ -520,6 +620,10 @@ _bpftool()
                             _bpftool_get_map_ids
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_map_names
+                            return 0
+                            ;;
                         key)
                             COMPREPLY+=( $( compgen -W 'hex' -- "$cur" ) )
                             ;;
@@ -545,6 +649,10 @@ _bpftool()
                             _bpftool_map_update_get_id $command
                             return 0
                             ;;
+                        name)
+                            _bpftool_map_update_get_name $command
+                            return 0
+                            ;;
                         key)
                             COMPREPLY+=( $( compgen -W 'hex' -- "$cur" ) )
                             ;;
@@ -553,13 +661,13 @@ _bpftool()
                             # map, depending on the type of the map to update.
                             case "$(_bpftool_map_guess_map_type)" in
                                 array_of_maps|hash_of_maps)
-                                    local MAP_TYPE='id pinned'
+                                    local MAP_TYPE='id pinned name'
                                     COMPREPLY+=( $( compgen -W "$MAP_TYPE" \
                                         -- "$cur" ) )
                                     return 0
                                     ;;
                                 prog_array)
-                                    local PROG_TYPE='id pinned tag'
+                                    local PROG_TYPE='id pinned tag name'
                                     COMPREPLY+=( $( compgen -W "$PROG_TYPE" \
                                         -- "$cur" ) )
                                     return 0
@@ -621,6 +729,10 @@ _bpftool()
                             _bpftool_get_map_ids_for_type perf_event_array
                             return 0
                             ;;
+                        name)
+                            _bpftool_get_map_names_for_type perf_event_array
+                            return 0
+                            ;;
                         cpu)
                             return 0
                             ;;
@@ -644,8 +756,8 @@ _bpftool()
             esac
             ;;
         btf)
-            local PROG_TYPE='id pinned tag'
-            local MAP_TYPE='id pinned'
+            local PROG_TYPE='id pinned tag name'
+            local MAP_TYPE='id pinned name'
             case $command in
                 dump)
                     case $prev in
@@ -676,6 +788,17 @@ _bpftool()
                             esac
                             return 0
                             ;;
+                        name)
+                            case $pprev in
+                                prog)
+                                    _bpftool_get_prog_names
+                                    ;;
+                                map)
+                                    _bpftool_get_map_names
+                                    ;;
+                            esac
+                            return 0
+                            ;;
                         format)
                             COMPREPLY=( $( compgen -W "c raw" -- "$cur" ) )
                             ;;
@@ -716,6 +839,17 @@ _bpftool()
                     ;;
             esac
             ;;
+        gen)
+            case $command in
+                skeleton)
+                    _filedir
+		    ;;
+                *)
+                    [[ $prev == $object ]] && \
+                        COMPREPLY=( $( compgen -W 'skeleton help' -- "$cur" ) )
+                    ;;
+            esac
+            ;;
         cgroup)
             case $command in
                 show|list|tree)
@@ -735,7 +869,7 @@ _bpftool()
                         connect6 sendmsg4 sendmsg6 recvmsg4 recvmsg6 sysctl \
                         getsockopt setsockopt'
                     local ATTACH_FLAGS='multi override'
-                    local PROG_TYPE='id pinned tag'
+                    local PROG_TYPE='id pinned tag name'
                     case $prev in
                         $command)
                             _filedir
@@ -760,7 +894,7 @@ _bpftool()
                             elif [[ "$command" == "attach" ]]; then
                                 # We have an attach type on the command line,
                                 # but it is not the previous word, or
-                                # "id|pinned|tag" (we already checked for
+                                # "id|pinned|tag|name" (we already checked for
                                 # that). This should only leave the case when
                                 # we need attach flags for "attach" commamnd.
                                 _bpftool_one_of_list "$ATTACH_FLAGS"
@@ -786,7 +920,7 @@ _bpftool()
             esac
             ;;
         net)
-            local PROG_TYPE='id pinned tag'
+            local PROG_TYPE='id pinned tag name'
             local ATTACH_TYPES='xdp xdpgeneric xdpdrv xdpoffload'
             case $command in
                 show|list)
diff --git a/tools/bpf/bpftool/cgroup.c b/tools/bpf/bpftool/cgroup.c
index 1ef45e55039e..2f017caa678d 100644
--- a/tools/bpf/bpftool/cgroup.c
+++ b/tools/bpf/bpftool/cgroup.c
@@ -117,6 +117,25 @@ static int count_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type)
 	return prog_cnt;
 }
 
+static int cgroup_has_attached_progs(int cgroup_fd)
+{
+	enum bpf_attach_type type;
+	bool no_prog = true;
+
+	for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) {
+		int count = count_attached_bpf_progs(cgroup_fd, type);
+
+		if (count < 0 && errno != EINVAL)
+			return -1;
+
+		if (count > 0) {
+			no_prog = false;
+			break;
+		}
+	}
+
+	return no_prog ? 0 : 1;
+}
 static int show_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type,
 				   int level)
 {
@@ -161,6 +180,7 @@ static int show_attached_bpf_progs(int cgroup_fd, enum bpf_attach_type type,
 static int do_show(int argc, char **argv)
 {
 	enum bpf_attach_type type;
+	int has_attached_progs;
 	const char *path;
 	int cgroup_fd;
 	int ret = -1;
@@ -192,6 +212,16 @@ static int do_show(int argc, char **argv)
 		goto exit;
 	}
 
+	has_attached_progs = cgroup_has_attached_progs(cgroup_fd);
+	if (has_attached_progs < 0) {
+		p_err("can't query bpf programs attached to %s: %s",
+		      path, strerror(errno));
+		goto exit_cgroup;
+	} else if (!has_attached_progs) {
+		ret = 0;
+		goto exit_cgroup;
+	}
+
 	if (json_output)
 		jsonw_start_array(json_wtr);
 	else
@@ -212,6 +242,7 @@ static int do_show(int argc, char **argv)
 	if (json_output)
 		jsonw_end_array(json_wtr);
 
+exit_cgroup:
 	close(cgroup_fd);
 exit:
 	return ret;
@@ -228,7 +259,7 @@ static int do_show_tree_fn(const char *fpath, const struct stat *sb,
 			   int typeflag, struct FTW *ftw)
 {
 	enum bpf_attach_type type;
-	bool skip = true;
+	int has_attached_progs;
 	int cgroup_fd;
 
 	if (typeflag != FTW_D)
@@ -240,22 +271,13 @@ static int do_show_tree_fn(const char *fpath, const struct stat *sb,
 		return SHOW_TREE_FN_ERR;
 	}
 
-	for (type = 0; type < __MAX_BPF_ATTACH_TYPE; type++) {
-		int count = count_attached_bpf_progs(cgroup_fd, type);
-
-		if (count < 0 && errno != EINVAL) {
-			p_err("can't query bpf programs attached to %s: %s",
-			      fpath, strerror(errno));
-			close(cgroup_fd);
-			return SHOW_TREE_FN_ERR;
-		}
-		if (count > 0) {
-			skip = false;
-			break;
-		}
-	}
-
-	if (skip) {
+	has_attached_progs = cgroup_has_attached_progs(cgroup_fd);
+	if (has_attached_progs < 0) {
+		p_err("can't query bpf programs attached to %s: %s",
+		      fpath, strerror(errno));
+		close(cgroup_fd);
+		return SHOW_TREE_FN_ERR;
+	} else if (!has_attached_progs) {
 		close(cgroup_fd);
 		return 0;
 	}
diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
new file mode 100644
index 000000000000..7ce09a9a6999
--- /dev/null
+++ b/tools/bpf/bpftool/gen.c
@@ -0,0 +1,609 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+/* Copyright (C) 2019 Facebook */
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif
+#include <ctype.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/err.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <string.h>
+#include <unistd.h>
+#include <bpf.h>
+#include <libbpf.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/mman.h>
+#include <unistd.h>
+
+#include "btf.h"
+#include "libbpf_internal.h"
+#include "json_writer.h"
+#include "main.h"
+
+
+#define MAX_OBJ_NAME_LEN 64
+
+static void sanitize_identifier(char *name)
+{
+	int i;
+
+	for (i = 0; name[i]; i++)
+		if (!isalnum(name[i]) && name[i] != '_')
+			name[i] = '_';
+}
+
+static bool str_has_suffix(const char *str, const char *suffix)
+{
+	size_t i, n1 = strlen(str), n2 = strlen(suffix);
+
+	if (n1 < n2)
+		return false;
+
+	for (i = 0; i < n2; i++) {
+		if (str[n1 - i - 1] != suffix[n2 - i - 1])
+			return false;
+	}
+
+	return true;
+}
+
+static void get_obj_name(char *name, const char *file)
+{
+	/* Using basename() GNU version which doesn't modify arg. */
+	strncpy(name, basename(file), MAX_OBJ_NAME_LEN - 1);
+	name[MAX_OBJ_NAME_LEN - 1] = '\0';
+	if (str_has_suffix(name, ".o"))
+		name[strlen(name) - 2] = '\0';
+	sanitize_identifier(name);
+}
+
+static void get_header_guard(char *guard, const char *obj_name)
+{
+	int i;
+
+	sprintf(guard, "__%s_SKEL_H__", obj_name);
+	for (i = 0; guard[i]; i++)
+		guard[i] = toupper(guard[i]);
+}
+
+static const char *get_map_ident(const struct bpf_map *map)
+{
+	const char *name = bpf_map__name(map);
+
+	if (!bpf_map__is_internal(map))
+		return name;
+
+	if (str_has_suffix(name, ".data"))
+		return "data";
+	else if (str_has_suffix(name, ".rodata"))
+		return "rodata";
+	else if (str_has_suffix(name, ".bss"))
+		return "bss";
+	else if (str_has_suffix(name, ".kconfig"))
+		return "kconfig";
+	else
+		return NULL;
+}
+
+static void codegen_btf_dump_printf(void *ct, const char *fmt, va_list args)
+{
+	vprintf(fmt, args);
+}
+
+static int codegen_datasec_def(struct bpf_object *obj,
+			       struct btf *btf,
+			       struct btf_dump *d,
+			       const struct btf_type *sec,
+			       const char *obj_name)
+{
+	const char *sec_name = btf__name_by_offset(btf, sec->name_off);
+	const struct btf_var_secinfo *sec_var = btf_var_secinfos(sec);
+	int i, err, off = 0, pad_cnt = 0, vlen = btf_vlen(sec);
+	const char *sec_ident;
+	char var_ident[256];
+
+	if (strcmp(sec_name, ".data") == 0)
+		sec_ident = "data";
+	else if (strcmp(sec_name, ".bss") == 0)
+		sec_ident = "bss";
+	else if (strcmp(sec_name, ".rodata") == 0)
+		sec_ident = "rodata";
+	else if (strcmp(sec_name, ".kconfig") == 0)
+		sec_ident = "kconfig";
+	else
+		return 0;
+
+	printf("	struct %s__%s {\n", obj_name, sec_ident);
+	for (i = 0; i < vlen; i++, sec_var++) {
+		const struct btf_type *var = btf__type_by_id(btf, sec_var->type);
+		const char *var_name = btf__name_by_offset(btf, var->name_off);
+		DECLARE_LIBBPF_OPTS(btf_dump_emit_type_decl_opts, opts,
+			.field_name = var_ident,
+			.indent_level = 2,
+		);
+		int need_off = sec_var->offset, align_off, align;
+		__u32 var_type_id = var->type;
+		const struct btf_type *t;
+
+		t = btf__type_by_id(btf, var_type_id);
+		while (btf_is_mod(t)) {
+			var_type_id = t->type;
+			t = btf__type_by_id(btf, var_type_id);
+		}
+
+		if (off > need_off) {
+			p_err("Something is wrong for %s's variable #%d: need offset %d, already at %d.\n",
+			      sec_name, i, need_off, off);
+			return -EINVAL;
+		}
+
+		align = btf__align_of(btf, var->type);
+		if (align <= 0) {
+			p_err("Failed to determine alignment of variable '%s': %d",
+			      var_name, align);
+			return -EINVAL;
+		}
+
+		align_off = (off + align - 1) / align * align;
+		if (align_off != need_off) {
+			printf("\t\tchar __pad%d[%d];\n",
+			       pad_cnt, need_off - off);
+			pad_cnt++;
+		}
+
+		/* sanitize variable name, e.g., for static vars inside
+		 * a function, it's name is '<function name>.<variable name>',
+		 * which we'll turn into a '<function name>_<variable name>'
+		 */
+		var_ident[0] = '\0';
+		strncat(var_ident, var_name, sizeof(var_ident) - 1);
+		sanitize_identifier(var_ident);
+
+		printf("\t\t");
+		err = btf_dump__emit_type_decl(d, var_type_id, &opts);
+		if (err)
+			return err;
+		printf(";\n");
+
+		off = sec_var->offset + sec_var->size;
+	}
+	printf("	} *%s;\n", sec_ident);
+	return 0;
+}
+
+static int codegen_datasecs(struct bpf_object *obj, const char *obj_name)
+{
+	struct btf *btf = bpf_object__btf(obj);
+	int n = btf__get_nr_types(btf);
+	struct btf_dump *d;
+	int i, err = 0;
+
+	d = btf_dump__new(btf, NULL, NULL, codegen_btf_dump_printf);
+	if (IS_ERR(d))
+		return PTR_ERR(d);
+
+	for (i = 1; i <= n; i++) {
+		const struct btf_type *t = btf__type_by_id(btf, i);
+
+		if (!btf_is_datasec(t))
+			continue;
+
+		err = codegen_datasec_def(obj, btf, d, t, obj_name);
+		if (err)
+			goto out;
+	}
+out:
+	btf_dump__free(d);
+	return err;
+}
+
+static int codegen(const char *template, ...)
+{
+	const char *src, *end;
+	int skip_tabs = 0, n;
+	char *s, *dst;
+	va_list args;
+	char c;
+
+	n = strlen(template);
+	s = malloc(n + 1);
+	if (!s)
+		return -ENOMEM;
+	src = template;
+	dst = s;
+
+	/* find out "baseline" indentation to skip */
+	while ((c = *src++)) {
+		if (c == '\t') {
+			skip_tabs++;
+		} else if (c == '\n') {
+			break;
+		} else {
+			p_err("unrecognized character at pos %td in template '%s'",
+			      src - template - 1, template);
+			return -EINVAL;
+		}
+	}
+
+	while (*src) {
+		/* skip baseline indentation tabs */
+		for (n = skip_tabs; n > 0; n--, src++) {
+			if (*src != '\t') {
+				p_err("not enough tabs at pos %td in template '%s'",
+				      src - template - 1, template);
+				return -EINVAL;
+			}
+		}
+		/* trim trailing whitespace */
+		end = strchrnul(src, '\n');
+		for (n = end - src; n > 0 && isspace(src[n - 1]); n--)
+			;
+		memcpy(dst, src, n);
+		dst += n;
+		if (*end)
+			*dst++ = '\n';
+		src = *end ? end + 1 : end;
+	}
+	*dst++ = '\0';
+
+	/* print out using adjusted template */
+	va_start(args, template);
+	n = vprintf(s, args);
+	va_end(args);
+
+	free(s);
+	return n;
+}
+
+static int do_skeleton(int argc, char **argv)
+{
+	char header_guard[MAX_OBJ_NAME_LEN + sizeof("__SKEL_H__")];
+	size_t i, map_cnt = 0, prog_cnt = 0, file_sz, mmap_sz;
+	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts);
+	char obj_name[MAX_OBJ_NAME_LEN], *obj_data;
+	struct bpf_object *obj = NULL;
+	const char *file, *ident;
+	struct bpf_program *prog;
+	int fd, len, err = -1;
+	struct bpf_map *map;
+	struct btf *btf;
+	struct stat st;
+
+	if (!REQ_ARGS(1)) {
+		usage();
+		return -1;
+	}
+	file = GET_ARG();
+
+	if (argc) {
+		p_err("extra unknown arguments");
+		return -1;
+	}
+
+	if (stat(file, &st)) {
+		p_err("failed to stat() %s: %s", file, strerror(errno));
+		return -1;
+	}
+	file_sz = st.st_size;
+	mmap_sz = roundup(file_sz, sysconf(_SC_PAGE_SIZE));
+	fd = open(file, O_RDONLY);
+	if (fd < 0) {
+		p_err("failed to open() %s: %s", file, strerror(errno));
+		return -1;
+	}
+	obj_data = mmap(NULL, mmap_sz, PROT_READ, MAP_PRIVATE, fd, 0);
+	if (obj_data == MAP_FAILED) {
+		obj_data = NULL;
+		p_err("failed to mmap() %s: %s", file, strerror(errno));
+		goto out;
+	}
+	get_obj_name(obj_name, file);
+	opts.object_name = obj_name;
+	obj = bpf_object__open_mem(obj_data, file_sz, &opts);
+	if (IS_ERR(obj)) {
+		obj = NULL;
+		p_err("failed to open BPF object file: %ld", PTR_ERR(obj));
+		goto out;
+	}
+
+	bpf_object__for_each_map(map, obj) {
+		ident = get_map_ident(map);
+		if (!ident) {
+			p_err("ignoring unrecognized internal map '%s'...",
+			      bpf_map__name(map));
+			continue;
+		}
+		map_cnt++;
+	}
+	bpf_object__for_each_program(prog, obj) {
+		prog_cnt++;
+	}
+
+	get_header_guard(header_guard, obj_name);
+	codegen("\
+		\n\
+		/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */   \n\
+									    \n\
+		/* THIS FILE IS AUTOGENERATED! */			    \n\
+		#ifndef %2$s						    \n\
+		#define %2$s						    \n\
+									    \n\
+		#include <stdlib.h>					    \n\
+		#include <libbpf.h>					    \n\
+									    \n\
+		struct %1$s {						    \n\
+			struct bpf_object_skeleton *skeleton;		    \n\
+			struct bpf_object *obj;				    \n\
+		",
+		obj_name, header_guard
+	);
+
+	if (map_cnt) {
+		printf("\tstruct {\n");
+		bpf_object__for_each_map(map, obj) {
+			ident = get_map_ident(map);
+			if (!ident)
+				continue;
+			printf("\t\tstruct bpf_map *%s;\n", ident);
+		}
+		printf("\t} maps;\n");
+	}
+
+	if (prog_cnt) {
+		printf("\tstruct {\n");
+		bpf_object__for_each_program(prog, obj) {
+			printf("\t\tstruct bpf_program *%s;\n",
+			       bpf_program__name(prog));
+		}
+		printf("\t} progs;\n");
+		printf("\tstruct {\n");
+		bpf_object__for_each_program(prog, obj) {
+			printf("\t\tstruct bpf_link *%s;\n",
+			       bpf_program__name(prog));
+		}
+		printf("\t} links;\n");
+	}
+
+	btf = bpf_object__btf(obj);
+	if (btf) {
+		err = codegen_datasecs(obj, obj_name);
+		if (err)
+			goto out;
+	}
+
+	codegen("\
+		\n\
+		};							    \n\
+									    \n\
+		static void						    \n\
+		%1$s__destroy(struct %1$s *obj)				    \n\
+		{							    \n\
+			if (!obj)					    \n\
+				return;					    \n\
+			if (obj->skeleton)				    \n\
+				bpf_object__destroy_skeleton(obj->skeleton);\n\
+			free(obj);					    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__create_skeleton(struct %1$s *obj);		    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open_opts(const struct bpf_object_open_opts *opts)    \n\
+		{							    \n\
+			struct %1$s *obj;				    \n\
+									    \n\
+			obj = (typeof(obj))calloc(1, sizeof(*obj));	    \n\
+			if (!obj)					    \n\
+				return NULL;				    \n\
+			if (%1$s__create_skeleton(obj))			    \n\
+				goto err;				    \n\
+			if (bpf_object__open_skeleton(obj->skeleton, opts)) \n\
+				goto err;				    \n\
+									    \n\
+			return obj;					    \n\
+		err:							    \n\
+			%1$s__destroy(obj);				    \n\
+			return NULL;					    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open(void)					    \n\
+		{							    \n\
+			return %1$s__open_opts(NULL);			    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__load(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__load_skeleton(obj->skeleton);    \n\
+		}							    \n\
+									    \n\
+		static inline struct %1$s *				    \n\
+		%1$s__open_and_load(void)				    \n\
+		{							    \n\
+			struct %1$s *obj;				    \n\
+									    \n\
+			obj = %1$s__open();				    \n\
+			if (!obj)					    \n\
+				return NULL;				    \n\
+			if (%1$s__load(obj)) {				    \n\
+				%1$s__destroy(obj);			    \n\
+				return NULL;				    \n\
+			}						    \n\
+			return obj;					    \n\
+		}							    \n\
+									    \n\
+		static inline int					    \n\
+		%1$s__attach(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__attach_skeleton(obj->skeleton);  \n\
+		}							    \n\
+									    \n\
+		static inline void					    \n\
+		%1$s__detach(struct %1$s *obj)				    \n\
+		{							    \n\
+			return bpf_object__detach_skeleton(obj->skeleton);  \n\
+		}							    \n\
+		",
+		obj_name
+	);
+
+	codegen("\
+		\n\
+									    \n\
+		static inline int					    \n\
+		%1$s__create_skeleton(struct %1$s *obj)			    \n\
+		{							    \n\
+			struct bpf_object_skeleton *s;			    \n\
+									    \n\
+			s = (typeof(s))calloc(1, sizeof(*s));		    \n\
+			if (!s)						    \n\
+				return -1;				    \n\
+			obj->skeleton = s;				    \n\
+									    \n\
+			s->sz = sizeof(*s);				    \n\
+			s->name = \"%1$s\";				    \n\
+			s->obj = &obj->obj;				    \n\
+		",
+		obj_name
+	);
+	if (map_cnt) {
+		codegen("\
+			\n\
+									    \n\
+				/* maps */				    \n\
+				s->map_cnt = %zu;			    \n\
+				s->map_skel_sz = sizeof(*s->maps);	    \n\
+				s->maps = (typeof(s->maps))calloc(s->map_cnt, s->map_skel_sz);\n\
+				if (!s->maps)				    \n\
+					goto err;			    \n\
+			",
+			map_cnt
+		);
+		i = 0;
+		bpf_object__for_each_map(map, obj) {
+			ident = get_map_ident(map);
+
+			if (!ident)
+				continue;
+
+			codegen("\
+				\n\
+									    \n\
+					s->maps[%zu].name = \"%s\";	    \n\
+					s->maps[%zu].map = &obj->maps.%s;   \n\
+				",
+				i, bpf_map__name(map), i, ident);
+			/* memory-mapped internal maps */
+			if (bpf_map__is_internal(map) &&
+			    (bpf_map__def(map)->map_flags & BPF_F_MMAPABLE)) {
+				printf("\ts->maps[%zu].mmaped = (void **)&obj->%s;\n",
+				       i, ident);
+			}
+			i++;
+		}
+	}
+	if (prog_cnt) {
+		codegen("\
+			\n\
+									    \n\
+				/* programs */				    \n\
+				s->prog_cnt = %zu;			    \n\
+				s->prog_skel_sz = sizeof(*s->progs);	    \n\
+				s->progs = (typeof(s->progs))calloc(s->prog_cnt, s->prog_skel_sz);\n\
+				if (!s->progs)				    \n\
+					goto err;			    \n\
+			",
+			prog_cnt
+		);
+		i = 0;
+		bpf_object__for_each_program(prog, obj) {
+			codegen("\
+				\n\
+									    \n\
+					s->progs[%1$zu].name = \"%2$s\";    \n\
+					s->progs[%1$zu].prog = &obj->progs.%2$s;\n\
+					s->progs[%1$zu].link = &obj->links.%2$s;\n\
+				",
+				i, bpf_program__name(prog));
+			i++;
+		}
+	}
+	codegen("\
+		\n\
+									    \n\
+			s->data_sz = %d;				    \n\
+			s->data = (void *)\"\\				    \n\
+		",
+		file_sz);
+
+	/* embed contents of BPF object file */
+	for (i = 0, len = 0; i < file_sz; i++) {
+		int w = obj_data[i] ? 4 : 2;
+
+		len += w;
+		if (len > 78) {
+			printf("\\\n");
+			len = w;
+		}
+		if (!obj_data[i])
+			printf("\\0");
+		else
+			printf("\\x%02x", (unsigned char)obj_data[i]);
+	}
+
+	codegen("\
+		\n\
+		\";							    \n\
+									    \n\
+			return 0;					    \n\
+		err:							    \n\
+			bpf_object__destroy_skeleton(s);		    \n\
+			return -1;					    \n\
+		}							    \n\
+									    \n\
+		#endif /* %s */						    \n\
+		",
+		header_guard);
+	err = 0;
+out:
+	bpf_object__close(obj);
+	if (obj_data)
+		munmap(obj_data, mmap_sz);
+	close(fd);
+	return err;
+}
+
+static int do_help(int argc, char **argv)
+{
+	if (json_output) {
+		jsonw_null(json_wtr);
+		return 0;
+	}
+
+	fprintf(stderr,
+		"Usage: %1$s gen skeleton FILE\n"
+		"       %1$s gen help\n"
+		"\n"
+		"       " HELP_SPEC_OPTIONS "\n"
+		"",
+		bin_name);
+
+	return 0;
+}
+
+static const struct cmd cmds[] = {
+	{ "skeleton",	do_skeleton },
+	{ "help",	do_help },
+	{ 0 }
+};
+
+int do_gen(int argc, char **argv)
+{
+	return cmd_select(cmds, argc, argv, do_help);
+}
diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c
index 4764581ff9ea..1fe91c558508 100644
--- a/tools/bpf/bpftool/main.c
+++ b/tools/bpf/bpftool/main.c
@@ -58,7 +58,7 @@ static int do_help(int argc, char **argv)
 		"       %s batch file FILE\n"
 		"       %s version\n"
 		"\n"
-		"       OBJECT := { prog | map | cgroup | perf | net | feature | btf }\n"
+		"       OBJECT := { prog | map | cgroup | perf | net | feature | btf | gen }\n"
 		"       " HELP_SPEC_OPTIONS "\n"
 		"",
 		bin_name, bin_name, bin_name);
@@ -227,6 +227,7 @@ static const struct cmd cmds[] = {
 	{ "net",	do_net },
 	{ "feature",	do_feature },
 	{ "btf",	do_btf },
+	{ "gen",	do_gen },
 	{ "version",	do_version },
 	{ 0 }
 };
diff --git a/tools/bpf/bpftool/main.h b/tools/bpf/bpftool/main.h
index 2899095f8254..4e75b58d3989 100644
--- a/tools/bpf/bpftool/main.h
+++ b/tools/bpf/bpftool/main.h
@@ -42,12 +42,12 @@
 #define BPF_TAG_FMT	"%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx%02hhx"
 
 #define HELP_SPEC_PROGRAM						\
-	"PROG := { id PROG_ID | pinned FILE | tag PROG_TAG }"
+	"PROG := { id PROG_ID | pinned FILE | tag PROG_TAG | name PROG_NAME }"
 #define HELP_SPEC_OPTIONS						\
 	"OPTIONS := { {-j|--json} [{-p|--pretty}] | {-f|--bpffs} |\n"	\
 	"\t            {-m|--mapcompat} | {-n|--nomount} }"
 #define HELP_SPEC_MAP							\
-	"MAP := { id MAP_ID | pinned FILE }"
+	"MAP := { id MAP_ID | pinned FILE | name MAP_NAME }"
 
 static const char * const prog_type_name[] = {
 	[BPF_PROG_TYPE_UNSPEC]			= "unspec",
@@ -155,6 +155,7 @@ int do_net(int argc, char **arg);
 int do_tracelog(int argc, char **arg);
 int do_feature(int argc, char **argv);
 int do_btf(int argc, char **argv);
+int do_gen(int argc, char **argv);
 
 int parse_u32_arg(int *argc, char ***argv, __u32 *val, const char *what);
 int prog_parse_fd(int *argc, char ***argv);
diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index de61d73b9030..c01f76fa6876 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -91,10 +91,66 @@ static void *alloc_value(struct bpf_map_info *info)
 		return malloc(info->value_size);
 }
 
-int map_parse_fd(int *argc, char ***argv)
+static int map_fd_by_name(char *name, int **fds)
 {
-	int fd;
+	unsigned int id = 0;
+	int fd, nb_fds = 0;
+	void *tmp;
+	int err;
+
+	while (true) {
+		struct bpf_map_info info = {};
+		__u32 len = sizeof(info);
+
+		err = bpf_map_get_next_id(id, &id);
+		if (err) {
+			if (errno != ENOENT) {
+				p_err("%s", strerror(errno));
+				goto err_close_fds;
+			}
+			return nb_fds;
+		}
+
+		fd = bpf_map_get_fd_by_id(id);
+		if (fd < 0) {
+			p_err("can't get map by id (%u): %s",
+			      id, strerror(errno));
+			goto err_close_fds;
+		}
+
+		err = bpf_obj_get_info_by_fd(fd, &info, &len);
+		if (err) {
+			p_err("can't get map info (%u): %s",
+			      id, strerror(errno));
+			goto err_close_fd;
+		}
+
+		if (strncmp(name, info.name, BPF_OBJ_NAME_LEN)) {
+			close(fd);
+			continue;
+		}
+
+		if (nb_fds > 0) {
+			tmp = realloc(*fds, (nb_fds + 1) * sizeof(int));
+			if (!tmp) {
+				p_err("failed to realloc");
+				goto err_close_fd;
+			}
+			*fds = tmp;
+		}
+		(*fds)[nb_fds++] = fd;
+	}
+
+err_close_fd:
+	close(fd);
+err_close_fds:
+	while (--nb_fds >= 0)
+		close((*fds)[nb_fds]);
+	return -1;
+}
 
+static int map_parse_fds(int *argc, char ***argv, int **fds)
+{
 	if (is_prefix(**argv, "id")) {
 		unsigned int id;
 		char *endptr;
@@ -108,10 +164,25 @@ int map_parse_fd(int *argc, char ***argv)
 		}
 		NEXT_ARGP();
 
-		fd = bpf_map_get_fd_by_id(id);
-		if (fd < 0)
+		(*fds)[0] = bpf_map_get_fd_by_id(id);
+		if ((*fds)[0] < 0) {
 			p_err("get map by id (%u): %s", id, strerror(errno));
-		return fd;
+			return -1;
+		}
+		return 1;
+	} else if (is_prefix(**argv, "name")) {
+		char *name;
+
+		NEXT_ARGP();
+
+		name = **argv;
+		if (strlen(name) > BPF_OBJ_NAME_LEN - 1) {
+			p_err("can't parse name");
+			return -1;
+		}
+		NEXT_ARGP();
+
+		return map_fd_by_name(name, fds);
 	} else if (is_prefix(**argv, "pinned")) {
 		char *path;
 
@@ -120,13 +191,43 @@ int map_parse_fd(int *argc, char ***argv)
 		path = **argv;
 		NEXT_ARGP();
 
-		return open_obj_pinned_any(path, BPF_OBJ_MAP);
+		(*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_MAP);
+		if ((*fds)[0] < 0)
+			return -1;
+		return 1;
 	}
 
-	p_err("expected 'id' or 'pinned', got: '%s'?", **argv);
+	p_err("expected 'id', 'name' or 'pinned', got: '%s'?", **argv);
 	return -1;
 }
 
+int map_parse_fd(int *argc, char ***argv)
+{
+	int *fds = NULL;
+	int nb_fds, fd;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = map_parse_fds(argc, argv, &fds);
+	if (nb_fds != 1) {
+		if (nb_fds > 1) {
+			p_err("several maps match this handle");
+			while (nb_fds--)
+				close(fds[nb_fds]);
+		}
+		fd = -1;
+		goto exit_free;
+	}
+
+	fd = fds[0];
+exit_free:
+	free(fds);
+	return fd;
+}
+
 int map_parse_fd_and_info(int *argc, char ***argv, void *info, __u32 *info_len)
 {
 	int err;
@@ -479,6 +580,21 @@ static int parse_elem(char **argv, struct bpf_map_info *info,
 	return -1;
 }
 
+static void show_map_header_json(struct bpf_map_info *info, json_writer_t *wtr)
+{
+	jsonw_uint_field(wtr, "id", info->id);
+	if (info->type < ARRAY_SIZE(map_type_name))
+		jsonw_string_field(wtr, "type", map_type_name[info->type]);
+	else
+		jsonw_uint_field(wtr, "type", info->type);
+
+	if (*info->name)
+		jsonw_string_field(wtr, "name", info->name);
+
+	jsonw_name(wtr, "flags");
+	jsonw_printf(wtr, "%d", info->map_flags);
+}
+
 static int show_map_close_json(int fd, struct bpf_map_info *info)
 {
 	char *memlock, *frozen_str;
@@ -489,18 +605,7 @@ static int show_map_close_json(int fd, struct bpf_map_info *info)
 
 	jsonw_start_object(json_wtr);
 
-	jsonw_uint_field(json_wtr, "id", info->id);
-	if (info->type < ARRAY_SIZE(map_type_name))
-		jsonw_string_field(json_wtr, "type",
-				   map_type_name[info->type]);
-	else
-		jsonw_uint_field(json_wtr, "type", info->type);
-
-	if (*info->name)
-		jsonw_string_field(json_wtr, "name", info->name);
-
-	jsonw_name(json_wtr, "flags");
-	jsonw_printf(json_wtr, "%d", info->map_flags);
+	show_map_header_json(info, json_wtr);
 
 	print_dev_json(info->ifindex, info->netns_dev, info->netns_ino);
 
@@ -561,14 +666,8 @@ static int show_map_close_json(int fd, struct bpf_map_info *info)
 	return 0;
 }
 
-static int show_map_close_plain(int fd, struct bpf_map_info *info)
+static void show_map_header_plain(struct bpf_map_info *info)
 {
-	char *memlock, *frozen_str;
-	int frozen = 0;
-
-	memlock = get_fdinfo(fd, "memlock");
-	frozen_str = get_fdinfo(fd, "frozen");
-
 	printf("%u: ", info->id);
 	if (info->type < ARRAY_SIZE(map_type_name))
 		printf("%s  ", map_type_name[info->type]);
@@ -581,6 +680,17 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info)
 	printf("flags 0x%x", info->map_flags);
 	print_dev_plain(info->ifindex, info->netns_dev, info->netns_ino);
 	printf("\n");
+}
+
+static int show_map_close_plain(int fd, struct bpf_map_info *info)
+{
+	char *memlock, *frozen_str;
+	int frozen = 0;
+
+	memlock = get_fdinfo(fd, "memlock");
+	frozen_str = get_fdinfo(fd, "frozen");
+
+	show_map_header_plain(info);
 	printf("\tkey %uB  value %uB  max_entries %u",
 	       info->key_size, info->value_size, info->max_entries);
 
@@ -642,6 +752,50 @@ static int show_map_close_plain(int fd, struct bpf_map_info *info)
 	return 0;
 }
 
+static int do_show_subset(int argc, char **argv)
+{
+	struct bpf_map_info info = {};
+	__u32 len = sizeof(info);
+	int *fds = NULL;
+	int nb_fds, i;
+	int err = -1;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = map_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (json_output && nb_fds > 1)
+		jsonw_start_array(json_wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		err = bpf_obj_get_info_by_fd(fds[i], &info, &len);
+		if (err) {
+			p_err("can't get map info: %s",
+			      strerror(errno));
+			for (; i < nb_fds; i++)
+				close(fds[i]);
+			break;
+		}
+
+		if (json_output)
+			show_map_close_json(fds[i], &info);
+		else
+			show_map_close_plain(fds[i], &info);
+
+		close(fds[i]);
+	}
+	if (json_output && nb_fds > 1)
+		jsonw_end_array(json_wtr);	/* root array */
+
+exit_free:
+	free(fds);
+	return err;
+}
+
 static int do_show(int argc, char **argv)
 {
 	struct bpf_map_info info = {};
@@ -653,16 +807,8 @@ static int do_show(int argc, char **argv)
 	if (show_pinned)
 		build_pinned_obj_table(&map_table, BPF_OBJ_MAP);
 
-	if (argc == 2) {
-		fd = map_parse_fd_and_info(&argc, &argv, &info, &len);
-		if (fd < 0)
-			return -1;
-
-		if (json_output)
-			return show_map_close_json(fd, &info);
-		else
-			return show_map_close_plain(fd, &info);
-	}
+	if (argc == 2)
+		return do_show_subset(argc, argv);
 
 	if (argc)
 		return BAD_ARG();
@@ -765,26 +911,49 @@ static int dump_map_elem(int fd, void *key, void *value,
 	return 0;
 }
 
-static int do_dump(int argc, char **argv)
+static int maps_have_btf(int *fds, int nb_fds)
 {
 	struct bpf_map_info info = {};
-	void *key, *value, *prev_key;
-	unsigned int num_elems = 0;
 	__u32 len = sizeof(info);
-	json_writer_t *btf_wtr;
 	struct btf *btf = NULL;
-	int err;
-	int fd;
+	int err, i;
 
-	if (argc != 2)
-		usage();
+	for (i = 0; i < nb_fds; i++) {
+		err = bpf_obj_get_info_by_fd(fds[i], &info, &len);
+		if (err) {
+			p_err("can't get map info: %s", strerror(errno));
+			goto err_close;
+		}
 
-	fd = map_parse_fd_and_info(&argc, &argv, &info, &len);
-	if (fd < 0)
-		return -1;
+		err = btf__get_from_id(info.btf_id, &btf);
+		if (err) {
+			p_err("failed to get btf");
+			goto err_close;
+		}
 
-	key = malloc(info.key_size);
-	value = alloc_value(&info);
+		if (!btf)
+			return 0;
+	}
+
+	return 1;
+
+err_close:
+	for (; i < nb_fds; i++)
+		close(fds[i]);
+	return -1;
+}
+
+static int
+map_dump(int fd, struct bpf_map_info *info, json_writer_t *wtr,
+	 bool enable_btf, bool show_header)
+{
+	void *key, *value, *prev_key;
+	unsigned int num_elems = 0;
+	struct btf *btf = NULL;
+	int err;
+
+	key = malloc(info->key_size);
+	value = alloc_value(info);
 	if (!key || !value) {
 		p_err("mem alloc failed");
 		err = -1;
@@ -793,30 +962,32 @@ static int do_dump(int argc, char **argv)
 
 	prev_key = NULL;
 
-	err = btf__get_from_id(info.btf_id, &btf);
-	if (err) {
-		p_err("failed to get btf");
-		goto exit_free;
+	if (enable_btf) {
+		err = btf__get_from_id(info->btf_id, &btf);
+		if (err || !btf) {
+			/* enable_btf is true only if we've already checked
+			 * that all maps have BTF information.
+			 */
+			p_err("failed to get btf");
+			goto exit_free;
+		}
 	}
 
-	if (json_output)
-		jsonw_start_array(json_wtr);
-	else
-		if (btf) {
-			btf_wtr = get_btf_writer();
-			if (!btf_wtr) {
-				p_info("failed to create json writer for btf. falling back to plain output");
-				btf__free(btf);
-				btf = NULL;
-			} else {
-				jsonw_start_array(btf_wtr);
-			}
+	if (wtr) {
+		if (show_header) {
+			jsonw_start_object(wtr);	/* map object */
+			show_map_header_json(info, wtr);
+			jsonw_name(wtr, "elements");
 		}
+		jsonw_start_array(wtr);		/* elements */
+	} else if (show_header) {
+		show_map_header_plain(info);
+	}
 
-	if (info.type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY &&
-	    info.value_size != 8)
+	if (info->type == BPF_MAP_TYPE_REUSEPORT_SOCKARRAY &&
+	    info->value_size != 8)
 		p_info("Warning: cannot read values from %s map with value_size != 8",
-		       map_type_name[info.type]);
+		       map_type_name[info->type]);
 	while (true) {
 		err = bpf_map_get_next_key(fd, prev_key, key);
 		if (err) {
@@ -824,15 +995,14 @@ static int do_dump(int argc, char **argv)
 				err = 0;
 			break;
 		}
-		num_elems += dump_map_elem(fd, key, value, &info, btf, btf_wtr);
+		num_elems += dump_map_elem(fd, key, value, info, btf, wtr);
 		prev_key = key;
 	}
 
-	if (json_output)
-		jsonw_end_array(json_wtr);
-	else if (btf) {
-		jsonw_end_array(btf_wtr);
-		jsonw_destroy(&btf_wtr);
+	if (wtr) {
+		jsonw_end_array(wtr);	/* elements */
+		if (show_header)
+			jsonw_end_object(wtr);	/* map object */
 	} else {
 		printf("Found %u element%s\n", num_elems,
 		       num_elems != 1 ? "s" : "");
@@ -847,6 +1017,72 @@ exit_free:
 	return err;
 }
 
+static int do_dump(int argc, char **argv)
+{
+	json_writer_t *wtr = NULL, *btf_wtr = NULL;
+	struct bpf_map_info info = {};
+	int nb_fds, i = 0, btf = 0;
+	__u32 len = sizeof(info);
+	int *fds = NULL;
+	int err = -1;
+
+	if (argc != 2)
+		usage();
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = map_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (json_output) {
+		wtr = json_wtr;
+	} else {
+		btf = maps_have_btf(fds, nb_fds);
+		if (btf < 0)
+			goto exit_close;
+		if (btf) {
+			btf_wtr = get_btf_writer();
+			if (btf_wtr) {
+				wtr = btf_wtr;
+			} else {
+				p_info("failed to create json writer for btf. falling back to plain output");
+				btf = 0;
+			}
+		}
+	}
+
+	if (wtr && nb_fds > 1)
+		jsonw_start_array(wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		if (bpf_obj_get_info_by_fd(fds[i], &info, &len)) {
+			p_err("can't get map info: %s", strerror(errno));
+			break;
+		}
+		err = map_dump(fds[i], &info, wtr, btf, nb_fds > 1);
+		if (!wtr && i != nb_fds - 1)
+			printf("\n");
+
+		if (err)
+			break;
+		close(fds[i]);
+	}
+	if (wtr && nb_fds > 1)
+		jsonw_end_array(wtr);	/* root array */
+
+	if (btf)
+		jsonw_destroy(&btf_wtr);
+exit_close:
+	for (; i < nb_fds; i++)
+		close(fds[i]);
+exit_free:
+	free(fds);
+	return err;
+}
+
 static int alloc_key_value(struct bpf_map_info *info, void **key, void **value)
 {
 	*key = NULL;
diff --git a/tools/bpf/bpftool/net.c b/tools/bpf/bpftool/net.c
index 4f52d3151616..d93bee298e54 100644
--- a/tools/bpf/bpftool/net.c
+++ b/tools/bpf/bpftool/net.c
@@ -18,6 +18,7 @@
 
 #include <bpf.h>
 #include <nlattr.h>
+#include "libbpf_internal.h"
 #include "main.h"
 #include "netlink_dumper.h"
 
diff --git a/tools/bpf/bpftool/prog.c b/tools/bpf/bpftool/prog.c
index 2ce9c5ba1934..47a61ac42dc0 100644
--- a/tools/bpf/bpftool/prog.c
+++ b/tools/bpf/bpftool/prog.c
@@ -25,6 +25,11 @@
 #include "main.h"
 #include "xlated_dumper.h"
 
+enum dump_mode {
+	DUMP_JITED,
+	DUMP_XLATED,
+};
+
 static const char * const attach_type_strings[] = {
 	[BPF_SK_SKB_STREAM_PARSER] = "stream_parser",
 	[BPF_SK_SKB_STREAM_VERDICT] = "stream_verdict",
@@ -77,11 +82,12 @@ static void print_boot_time(__u64 nsecs, char *buf, unsigned int size)
 		strftime(buf, size, "%FT%T%z", &load_tm);
 }
 
-static int prog_fd_by_tag(unsigned char *tag)
+static int prog_fd_by_nametag(void *nametag, int **fds, bool tag)
 {
 	unsigned int id = 0;
+	int fd, nb_fds = 0;
+	void *tmp;
 	int err;
-	int fd;
 
 	while (true) {
 		struct bpf_prog_info info = {};
@@ -89,36 +95,54 @@ static int prog_fd_by_tag(unsigned char *tag)
 
 		err = bpf_prog_get_next_id(id, &id);
 		if (err) {
-			p_err("%s", strerror(errno));
-			return -1;
+			if (errno != ENOENT) {
+				p_err("%s", strerror(errno));
+				goto err_close_fds;
+			}
+			return nb_fds;
 		}
 
 		fd = bpf_prog_get_fd_by_id(id);
 		if (fd < 0) {
 			p_err("can't get prog by id (%u): %s",
 			      id, strerror(errno));
-			return -1;
+			goto err_close_fds;
 		}
 
 		err = bpf_obj_get_info_by_fd(fd, &info, &len);
 		if (err) {
 			p_err("can't get prog info (%u): %s",
 			      id, strerror(errno));
-			close(fd);
-			return -1;
+			goto err_close_fd;
 		}
 
-		if (!memcmp(tag, info.tag, BPF_TAG_SIZE))
-			return fd;
+		if ((tag && memcmp(nametag, info.tag, BPF_TAG_SIZE)) ||
+		    (!tag && strncmp(nametag, info.name, BPF_OBJ_NAME_LEN))) {
+			close(fd);
+			continue;
+		}
 
-		close(fd);
+		if (nb_fds > 0) {
+			tmp = realloc(*fds, (nb_fds + 1) * sizeof(int));
+			if (!tmp) {
+				p_err("failed to realloc");
+				goto err_close_fd;
+			}
+			*fds = tmp;
+		}
+		(*fds)[nb_fds++] = fd;
 	}
+
+err_close_fd:
+	close(fd);
+err_close_fds:
+	while (--nb_fds >= 0)
+		close((*fds)[nb_fds]);
+	return -1;
 }
 
-int prog_parse_fd(int *argc, char ***argv)
+static int prog_parse_fds(int *argc, char ***argv, int **fds)
 {
-	int fd;
-
 	if (is_prefix(**argv, "id")) {
 		unsigned int id;
 		char *endptr;
@@ -132,10 +156,12 @@ int prog_parse_fd(int *argc, char ***argv)
 		}
 		NEXT_ARGP();
 
-		fd = bpf_prog_get_fd_by_id(id);
-		if (fd < 0)
+		(*fds)[0] = bpf_prog_get_fd_by_id(id);
+		if ((*fds)[0] < 0) {
 			p_err("get by id (%u): %s", id, strerror(errno));
-		return fd;
+			return -1;
+		}
+		return 1;
 	} else if (is_prefix(**argv, "tag")) {
 		unsigned char tag[BPF_TAG_SIZE];
 
@@ -149,7 +175,20 @@ int prog_parse_fd(int *argc, char ***argv)
 		}
 		NEXT_ARGP();
 
-		return prog_fd_by_tag(tag);
+		return prog_fd_by_nametag(tag, fds, true);
+	} else if (is_prefix(**argv, "name")) {
+		char *name;
+
+		NEXT_ARGP();
+
+		name = **argv;
+		if (strlen(name) > BPF_OBJ_NAME_LEN - 1) {
+			p_err("can't parse name");
+			return -1;
+		}
+		NEXT_ARGP();
+
+		return prog_fd_by_nametag(name, fds, false);
 	} else if (is_prefix(**argv, "pinned")) {
 		char *path;
 
@@ -158,13 +197,43 @@ int prog_parse_fd(int *argc, char ***argv)
 		path = **argv;
 		NEXT_ARGP();
 
-		return open_obj_pinned_any(path, BPF_OBJ_PROG);
+		(*fds)[0] = open_obj_pinned_any(path, BPF_OBJ_PROG);
+		if ((*fds)[0] < 0)
+			return -1;
+		return 1;
 	}
 
-	p_err("expected 'id', 'tag' or 'pinned', got: '%s'?", **argv);
+	p_err("expected 'id', 'tag', 'name' or 'pinned', got: '%s'?", **argv);
 	return -1;
 }
 
+int prog_parse_fd(int *argc, char ***argv)
+{
+	int *fds = NULL;
+	int nb_fds, fd;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = prog_parse_fds(argc, argv, &fds);
+	if (nb_fds != 1) {
+		if (nb_fds > 1) {
+			p_err("several programs match this handle");
+			while (nb_fds--)
+				close(fds[nb_fds]);
+		}
+		fd = -1;
+		goto exit_free;
+	}
+
+	fd = fds[0];
+exit_free:
+	free(fds);
+	return fd;
+}
+
 static void show_prog_maps(int fd, u32 num_maps)
 {
 	struct bpf_prog_info info = {};
@@ -194,11 +263,8 @@ static void show_prog_maps(int fd, u32 num_maps)
 	}
 }
 
-static void print_prog_json(struct bpf_prog_info *info, int fd)
+static void print_prog_header_json(struct bpf_prog_info *info)
 {
-	char *memlock;
-
-	jsonw_start_object(json_wtr);
 	jsonw_uint_field(json_wtr, "id", info->id);
 	if (info->type < ARRAY_SIZE(prog_type_name))
 		jsonw_string_field(json_wtr, "type",
@@ -219,7 +285,14 @@ static void print_prog_json(struct bpf_prog_info *info, int fd)
 		jsonw_uint_field(json_wtr, "run_time_ns", info->run_time_ns);
 		jsonw_uint_field(json_wtr, "run_cnt", info->run_cnt);
 	}
+}
 
+static void print_prog_json(struct bpf_prog_info *info, int fd)
+{
+	char *memlock;
+
+	jsonw_start_object(json_wtr);
+	print_prog_header_json(info);
 	print_dev_json(info->ifindex, info->netns_dev, info->netns_ino);
 
 	if (info->load_time) {
@@ -268,10 +341,8 @@ static void print_prog_json(struct bpf_prog_info *info, int fd)
 	jsonw_end_object(json_wtr);
 }
 
-static void print_prog_plain(struct bpf_prog_info *info, int fd)
+static void print_prog_header_plain(struct bpf_prog_info *info)
 {
-	char *memlock;
-
 	printf("%u: ", info->id);
 	if (info->type < ARRAY_SIZE(prog_type_name))
 		printf("%s  ", prog_type_name[info->type]);
@@ -289,6 +360,13 @@ static void print_prog_plain(struct bpf_prog_info *info, int fd)
 		printf(" run_time_ns %lld run_cnt %lld",
 		       info->run_time_ns, info->run_cnt);
 	printf("\n");
+}
+
+static void print_prog_plain(struct bpf_prog_info *info, int fd)
+{
+	char *memlock;
+
+	print_prog_header_plain(info);
 
 	if (info->load_time) {
 		char buf[32];
@@ -349,6 +427,40 @@ static int show_prog(int fd)
 	return 0;
 }
 
+static int do_show_subset(int argc, char **argv)
+{
+	int *fds = NULL;
+	int nb_fds, i;
+	int err = -1;
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = prog_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (json_output && nb_fds > 1)
+		jsonw_start_array(json_wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		err = show_prog(fds[i]);
+		if (err) {
+			for (; i < nb_fds; i++)
+				close(fds[i]);
+			break;
+		}
+		close(fds[i]);
+	}
+	if (json_output && nb_fds > 1)
+		jsonw_end_array(json_wtr);	/* root array */
+
+exit_free:
+	free(fds);
+	return err;
+}
+
 static int do_show(int argc, char **argv)
 {
 	__u32 id = 0;
@@ -358,15 +470,8 @@ static int do_show(int argc, char **argv)
 	if (show_pinned)
 		build_pinned_obj_table(&prog_table, BPF_OBJ_PROG);
 
-	if (argc == 2) {
-		fd = prog_parse_fd(&argc, &argv);
-		if (fd < 0)
-			return -1;
-
-		err = show_prog(fd);
-		close(fd);
-		return err;
-	}
+	if (argc == 2)
+		return do_show_subset(argc, argv);
 
 	if (argc)
 		return BAD_ARG();
@@ -408,101 +513,32 @@ static int do_show(int argc, char **argv)
 	return err;
 }
 
-static int do_dump(int argc, char **argv)
+static int
+prog_dump(struct bpf_prog_info *info, enum dump_mode mode,
+	  char *filepath, bool opcodes, bool visual, bool linum)
 {
-	struct bpf_prog_info_linear *info_linear;
 	struct bpf_prog_linfo *prog_linfo = NULL;
-	enum {DUMP_JITED, DUMP_XLATED} mode;
 	const char *disasm_opt = NULL;
-	struct bpf_prog_info *info;
 	struct dump_data dd = {};
 	void *func_info = NULL;
 	struct btf *btf = NULL;
-	char *filepath = NULL;
-	bool opcodes = false;
-	bool visual = false;
 	char func_sig[1024];
 	unsigned char *buf;
-	bool linum = false;
 	__u32 member_len;
-	__u64 arrays;
 	ssize_t n;
 	int fd;
 
-	if (is_prefix(*argv, "jited")) {
-		if (disasm_init())
-			return -1;
-		mode = DUMP_JITED;
-	} else if (is_prefix(*argv, "xlated")) {
-		mode = DUMP_XLATED;
-	} else {
-		p_err("expected 'xlated' or 'jited', got: %s", *argv);
-		return -1;
-	}
-	NEXT_ARG();
-
-	if (argc < 2)
-		usage();
-
-	fd = prog_parse_fd(&argc, &argv);
-	if (fd < 0)
-		return -1;
-
-	if (is_prefix(*argv, "file")) {
-		NEXT_ARG();
-		if (!argc) {
-			p_err("expected file path");
-			return -1;
-		}
-
-		filepath = *argv;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "opcodes")) {
-		opcodes = true;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "visual")) {
-		visual = true;
-		NEXT_ARG();
-	} else if (is_prefix(*argv, "linum")) {
-		linum = true;
-		NEXT_ARG();
-	}
-
-	if (argc) {
-		usage();
-		return -1;
-	}
-
-	if (mode == DUMP_JITED)
-		arrays = 1UL << BPF_PROG_INFO_JITED_INSNS;
-	else
-		arrays = 1UL << BPF_PROG_INFO_XLATED_INSNS;
-
-	arrays |= 1UL << BPF_PROG_INFO_JITED_KSYMS;
-	arrays |= 1UL << BPF_PROG_INFO_JITED_FUNC_LENS;
-	arrays |= 1UL << BPF_PROG_INFO_FUNC_INFO;
-	arrays |= 1UL << BPF_PROG_INFO_LINE_INFO;
-	arrays |= 1UL << BPF_PROG_INFO_JITED_LINE_INFO;
-
-	info_linear = bpf_program__get_prog_info_linear(fd, arrays);
-	close(fd);
-	if (IS_ERR_OR_NULL(info_linear)) {
-		p_err("can't get prog info: %s", strerror(errno));
-		return -1;
-	}
-
-	info = &info_linear->info;
 	if (mode == DUMP_JITED) {
 		if (info->jited_prog_len == 0 || !info->jited_prog_insns) {
 			p_info("no instructions returned");
-			goto err_free;
+			return -1;
 		}
 		buf = (unsigned char *)(info->jited_prog_insns);
 		member_len = info->jited_prog_len;
 	} else {	/* DUMP_XLATED */
 		if (info->xlated_prog_len == 0) {
 			p_err("error retrieving insn dump: kernel.kptr_restrict set?");
-			goto err_free;
+			return -1;
 		}
 		buf = (unsigned char *)info->xlated_prog_insns;
 		member_len = info->xlated_prog_len;
@@ -510,7 +546,7 @@ static int do_dump(int argc, char **argv)
 
 	if (info->btf_id && btf__get_from_id(info->btf_id, &btf)) {
 		p_err("failed to get btf");
-		goto err_free;
+		return -1;
 	}
 
 	func_info = (void *)info->func_info;
@@ -526,7 +562,7 @@ static int do_dump(int argc, char **argv)
 		if (fd < 0) {
 			p_err("can't open file %s: %s", filepath,
 			      strerror(errno));
-			goto err_free;
+			return -1;
 		}
 
 		n = write(fd, buf, member_len);
@@ -534,7 +570,7 @@ static int do_dump(int argc, char **argv)
 		if (n != member_len) {
 			p_err("error writing output file: %s",
 			      n < 0 ? strerror(errno) : "short write");
-			goto err_free;
+			return -1;
 		}
 
 		if (json_output)
@@ -548,7 +584,7 @@ static int do_dump(int argc, char **argv)
 						     info->netns_ino,
 						     &disasm_opt);
 			if (!name)
-				goto err_free;
+				return -1;
 		}
 
 		if (info->nr_jited_func_lens && info->jited_func_lens) {
@@ -643,12 +679,130 @@ static int do_dump(int argc, char **argv)
 		kernel_syms_destroy(&dd);
 	}
 
-	free(info_linear);
 	return 0;
+}
 
-err_free:
-	free(info_linear);
-	return -1;
+static int do_dump(int argc, char **argv)
+{
+	struct bpf_prog_info_linear *info_linear;
+	char *filepath = NULL;
+	bool opcodes = false;
+	bool visual = false;
+	enum dump_mode mode;
+	bool linum = false;
+	int *fds = NULL;
+	int nb_fds, i = 0;
+	int err = -1;
+	__u64 arrays;
+
+	if (is_prefix(*argv, "jited")) {
+		if (disasm_init())
+			return -1;
+		mode = DUMP_JITED;
+	} else if (is_prefix(*argv, "xlated")) {
+		mode = DUMP_XLATED;
+	} else {
+		p_err("expected 'xlated' or 'jited', got: %s", *argv);
+		return -1;
+	}
+	NEXT_ARG();
+
+	if (argc < 2)
+		usage();
+
+	fds = malloc(sizeof(int));
+	if (!fds) {
+		p_err("mem alloc failed");
+		return -1;
+	}
+	nb_fds = prog_parse_fds(&argc, &argv, &fds);
+	if (nb_fds < 1)
+		goto exit_free;
+
+	if (is_prefix(*argv, "file")) {
+		NEXT_ARG();
+		if (!argc) {
+			p_err("expected file path");
+			goto exit_close;
+		}
+		if (nb_fds > 1) {
+			p_err("several programs matched");
+			goto exit_close;
+		}
+
+		filepath = *argv;
+		NEXT_ARG();
+	} else if (is_prefix(*argv, "opcodes")) {
+		opcodes = true;
+		NEXT_ARG();
+	} else if (is_prefix(*argv, "visual")) {
+		if (nb_fds > 1) {
+			p_err("several programs matched");
+			goto exit_close;
+		}
+
+		visual = true;
+		NEXT_ARG();
+	} else if (is_prefix(*argv, "linum")) {
+		linum = true;
+		NEXT_ARG();
+	}
+
+	if (argc) {
+		usage();
+		goto exit_close;
+	}
+
+	if (mode == DUMP_JITED)
+		arrays = 1UL << BPF_PROG_INFO_JITED_INSNS;
+	else
+		arrays = 1UL << BPF_PROG_INFO_XLATED_INSNS;
+
+	arrays |= 1UL << BPF_PROG_INFO_JITED_KSYMS;
+	arrays |= 1UL << BPF_PROG_INFO_JITED_FUNC_LENS;
+	arrays |= 1UL << BPF_PROG_INFO_FUNC_INFO;
+	arrays |= 1UL << BPF_PROG_INFO_LINE_INFO;
+	arrays |= 1UL << BPF_PROG_INFO_JITED_LINE_INFO;
+
+	if (json_output && nb_fds > 1)
+		jsonw_start_array(json_wtr);	/* root array */
+	for (i = 0; i < nb_fds; i++) {
+		info_linear = bpf_program__get_prog_info_linear(fds[i], arrays);
+		if (IS_ERR_OR_NULL(info_linear)) {
+			p_err("can't get prog info: %s", strerror(errno));
+			break;
+		}
+
+		if (json_output && nb_fds > 1) {
+			jsonw_start_object(json_wtr);	/* prog object */
+			print_prog_header_json(&info_linear->info);
+			jsonw_name(json_wtr, "insns");
+		} else if (nb_fds > 1) {
+			print_prog_header_plain(&info_linear->info);
+		}
+
+		err = prog_dump(&info_linear->info, mode, filepath, opcodes,
+				visual, linum);
+
+		if (json_output && nb_fds > 1)
+			jsonw_end_object(json_wtr);	/* prog object */
+		else if (i != nb_fds - 1 && nb_fds > 1)
+			printf("\n");
+
+		free(info_linear);
+		if (err)
+			break;
+		close(fds[i]);
+	}
+	if (json_output && nb_fds > 1)
+		jsonw_end_array(json_wtr);	/* root array */
+
+exit_close:
+	for (; i < nb_fds; i++)
+		close(fds[i]);
+exit_free:
+	free(fds);
+	return err;
 }
 
 static int do_pin(int argc, char **argv)
diff --git a/tools/include/uapi/asm/bpf_perf_event.h b/tools/include/uapi/asm/bpf_perf_event.h
index 13a58531e6fa..39acc149d843 100644
--- a/tools/include/uapi/asm/bpf_perf_event.h
+++ b/tools/include/uapi/asm/bpf_perf_event.h
@@ -2,6 +2,8 @@
 #include "../../arch/arm64/include/uapi/asm/bpf_perf_event.h"
 #elif defined(__s390__)
 #include "../../arch/s390/include/uapi/asm/bpf_perf_event.h"
+#elif defined(__riscv)
+#include "../../arch/riscv/include/uapi/asm/bpf_perf_event.h"
 #else
 #include <uapi/asm-generic/bpf_perf_event.h>
 #endif
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index dbbcf0b02970..7df436da542d 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -231,6 +231,11 @@ enum bpf_attach_type {
  * When children program makes decision (like picking TCP CA or sock bind)
  * parent program has a chance to override it.
  *
+ * With BPF_F_ALLOW_MULTI a new program is added to the end of the list of
+ * programs for a cgroup. Though it's possible to replace an old program at
+ * any position by also specifying BPF_F_REPLACE flag and position itself in
+ * replace_bpf_fd attribute. Old program at this position will be released.
+ *
  * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
  * A cgroup with NONE doesn't allow any programs in sub-cgroups.
  * Ex1:
@@ -249,6 +254,7 @@ enum bpf_attach_type {
  */
 #define BPF_F_ALLOW_OVERRIDE	(1U << 0)
 #define BPF_F_ALLOW_MULTI	(1U << 1)
+#define BPF_F_REPLACE		(1U << 2)
 
 /* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will perform strict alignment checking as if the kernel
@@ -442,6 +448,10 @@ union bpf_attr {
 		__u32		attach_bpf_fd;	/* eBPF program to attach */
 		__u32		attach_type;
 		__u32		attach_flags;
+		__u32		replace_bpf_fd;	/* previously attached eBPF
+						 * program to replace if
+						 * BPF_F_REPLACE is used
+						 */
 	};
 
 	struct { /* anonymous struct used by BPF_PROG_TEST_RUN command */
diff --git a/tools/include/uapi/linux/btf.h b/tools/include/uapi/linux/btf.h
index 63ae4a39e58b..1a2898c482ee 100644
--- a/tools/include/uapi/linux/btf.h
+++ b/tools/include/uapi/linux/btf.h
@@ -22,9 +22,9 @@ struct btf_header {
 };
 
 /* Max # of type identifier */
-#define BTF_MAX_TYPE	0x0000ffff
+#define BTF_MAX_TYPE	0x000fffff
 /* Max offset into the string section */
-#define BTF_MAX_NAME_OFFSET	0x0000ffff
+#define BTF_MAX_NAME_OFFSET	0x00ffffff
 /* Max # of struct/union/enum members or func args */
 #define BTF_MAX_VLEN	0xffff
 
@@ -142,7 +142,8 @@ struct btf_param {
 
 enum {
 	BTF_VAR_STATIC = 0,
-	BTF_VAR_GLOBAL_ALLOCATED,
+	BTF_VAR_GLOBAL_ALLOCATED = 1,
+	BTF_VAR_GLOBAL_EXTERN = 2,
 };
 
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index defae23a0169..d4790121adf4 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -56,8 +56,8 @@ ifndef VERBOSE
 endif
 
 FEATURE_USER = .libbpf
-FEATURE_TESTS = libelf libelf-mmap bpf reallocarray
-FEATURE_DISPLAY = libelf bpf
+FEATURE_TESTS = libelf libelf-mmap zlib bpf reallocarray
+FEATURE_DISPLAY = libelf zlib bpf
 
 INCLUDES = -I. -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(ARCH)/include/uapi -I$(srctree)/tools/include/uapi
 FEATURE_CHECK_CFLAGS-bpf = $(INCLUDES)
@@ -147,6 +147,7 @@ TAGS_PROG := $(if $(shell which etags 2>/dev/null),etags,ctags)
 
 GLOBAL_SYM_COUNT = $(shell readelf -s --wide $(BPF_IN_SHARED) | \
 			   cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' | \
+			   sed 's/\[.*\]//' | \
 			   awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$NF}' | \
 			   sort -u | wc -l)
 VERSIONED_SYM_COUNT = $(shell readelf -s --wide $(OUTPUT)libbpf.so | \
@@ -159,7 +160,7 @@ all: fixdep
 
 all_cmd: $(CMD_TARGETS) check
 
-$(BPF_IN_SHARED): force elfdep bpfdep bpf_helper_defs.h
+$(BPF_IN_SHARED): force elfdep zdep bpfdep bpf_helper_defs.h
 	@(test -f ../../include/uapi/linux/bpf.h -a -f ../../../include/uapi/linux/bpf.h && ( \
 	(diff -B ../../include/uapi/linux/bpf.h ../../../include/uapi/linux/bpf.h >/dev/null) || \
 	echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/bpf.h' differs from latest version at 'include/uapi/linux/bpf.h'" >&2 )) || true
@@ -177,7 +178,7 @@ $(BPF_IN_SHARED): force elfdep bpfdep bpf_helper_defs.h
 	echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/if_xdp.h' differs from latest version at 'include/uapi/linux/if_xdp.h'" >&2 )) || true
 	$(Q)$(MAKE) $(build)=libbpf OUTPUT=$(SHARED_OBJDIR) CFLAGS="$(CFLAGS) $(SHLIB_FLAGS)"
 
-$(BPF_IN_STATIC): force elfdep bpfdep bpf_helper_defs.h
+$(BPF_IN_STATIC): force elfdep zdep bpfdep bpf_helper_defs.h
 	$(Q)$(MAKE) $(build)=libbpf OUTPUT=$(STATIC_OBJDIR)
 
 bpf_helper_defs.h: $(srctree)/tools/include/uapi/linux/bpf.h
@@ -189,7 +190,7 @@ $(OUTPUT)libbpf.so: $(OUTPUT)libbpf.so.$(LIBBPF_VERSION)
 $(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN_SHARED)
 	$(QUIET_LINK)$(CC) $(LDFLAGS) \
 		--shared -Wl,-soname,libbpf.so.$(LIBBPF_MAJOR_VERSION) \
-		-Wl,--version-script=$(VERSION_SCRIPT) $^ -lelf -o $@
+		-Wl,--version-script=$(VERSION_SCRIPT) $^ -lelf -lz -o $@
 	@ln -sf $(@F) $(OUTPUT)libbpf.so
 	@ln -sf $(@F) $(OUTPUT)libbpf.so.$(LIBBPF_MAJOR_VERSION)
 
@@ -213,6 +214,7 @@ check_abi: $(OUTPUT)libbpf.so
 		     "versioned in $(VERSION_SCRIPT)." >&2;		 \
 		readelf -s --wide $(BPF_IN_SHARED) |			 \
 		    cut -d "@" -f1 | sed 's/_v[0-9]_[0-9]_[0-9].*//' |	 \
+		    sed 's/\[.*\]//' |					 \
 		    awk '/GLOBAL/ && /DEFAULT/ && !/UND/ {print $$NF}'|  \
 		    sort -u > $(OUTPUT)libbpf_global_syms.tmp;		 \
 		readelf -s --wide $(OUTPUT)libbpf.so |			 \
@@ -249,6 +251,7 @@ install_headers: bpf_helper_defs.h
 		$(call do_install,libbpf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,btf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,libbpf_util.h,$(prefix)/include/bpf,644); \
+		$(call do_install,libbpf_common.h,$(prefix)/include/bpf,644); \
 		$(call do_install,xsk.h,$(prefix)/include/bpf,644); \
 		$(call do_install,bpf_helpers.h,$(prefix)/include/bpf,644); \
 		$(call do_install,bpf_helper_defs.h,$(prefix)/include/bpf,644); \
@@ -277,12 +280,15 @@ clean:
 
 
 
-PHONY += force elfdep bpfdep cscope tags
+PHONY += force elfdep zdep bpfdep cscope tags
 force:
 
 elfdep:
 	@if [ "$(feature-libelf)" != "1" ]; then echo "No libelf found"; exit 1 ; fi
 
+zdep:
+	@if [ "$(feature-zlib)" != "1" ]; then echo "No zlib found"; exit 1 ; fi
+
 bpfdep:
 	@if [ "$(feature-bpf)" != "1" ]; then echo "BPF API too old"; exit 1 ; fi
 
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 98596e15390f..a787d53699c8 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -467,13 +467,28 @@ int bpf_obj_get(const char *pathname)
 int bpf_prog_attach(int prog_fd, int target_fd, enum bpf_attach_type type,
 		    unsigned int flags)
 {
+	DECLARE_LIBBPF_OPTS(bpf_prog_attach_opts, opts,
+		.flags = flags,
+	);
+
+	return bpf_prog_attach_xattr(prog_fd, target_fd, type, &opts);
+}
+
+int bpf_prog_attach_xattr(int prog_fd, int target_fd,
+			  enum bpf_attach_type type,
+			  const struct bpf_prog_attach_opts *opts)
+{
 	union bpf_attr attr;
 
+	if (!OPTS_VALID(opts, bpf_prog_attach_opts))
+		return -EINVAL;
+
 	memset(&attr, 0, sizeof(attr));
 	attr.target_fd	   = target_fd;
 	attr.attach_bpf_fd = prog_fd;
 	attr.attach_type   = type;
-	attr.attach_flags  = flags;
+	attr.attach_flags  = OPTS_GET(opts, flags, 0);
+	attr.replace_bpf_fd = OPTS_GET(opts, replace_prog_fd, 0);
 
 	return sys_bpf(BPF_PROG_ATTACH, &attr, sizeof(attr));
 }
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 3c791fa8e68e..f0ab8519986e 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -28,14 +28,12 @@
 #include <stddef.h>
 #include <stdint.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 struct bpf_create_map_attr {
 	const char *name;
 	enum bpf_map_type map_type;
@@ -128,8 +126,19 @@ LIBBPF_API int bpf_map_get_next_key(int fd, const void *key, void *next_key);
 LIBBPF_API int bpf_map_freeze(int fd);
 LIBBPF_API int bpf_obj_pin(int fd, const char *pathname);
 LIBBPF_API int bpf_obj_get(const char *pathname);
+
+struct bpf_prog_attach_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	unsigned int flags;
+	int replace_prog_fd;
+};
+#define bpf_prog_attach_opts__last_field replace_prog_fd
+
 LIBBPF_API int bpf_prog_attach(int prog_fd, int attachable_fd,
 			       enum bpf_attach_type type, unsigned int flags);
+LIBBPF_API int bpf_prog_attach_xattr(int prog_fd, int attachable_fd,
+				     enum bpf_attach_type type,
+				     const struct bpf_prog_attach_opts *opts);
 LIBBPF_API int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
 LIBBPF_API int bpf_prog_detach2(int prog_fd, int attachable_fd,
 				enum bpf_attach_type type);
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index 0c7d28292898..f69cc208778a 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -25,6 +25,9 @@
 #ifndef __always_inline
 #define __always_inline __attribute__((always_inline))
 #endif
+#ifndef __weak
+#define __weak __attribute__((weak))
+#endif
 
 /*
  * Helper structure used by eBPF C program
@@ -44,4 +47,12 @@ enum libbpf_pin_type {
 	LIBBPF_PIN_BY_NAME,
 };
 
+enum libbpf_tristate {
+	TRI_NO = 0,
+	TRI_YES = 1,
+	TRI_MODULE = 2,
+};
+
+#define __kconfig __attribute__((section(".kconfig")))
+
 #endif
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 88efa2bb7137..5f04f56e1eb6 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -278,6 +278,45 @@ done:
 	return nelems * size;
 }
 
+int btf__align_of(const struct btf *btf, __u32 id)
+{
+	const struct btf_type *t = btf__type_by_id(btf, id);
+	__u16 kind = btf_kind(t);
+
+	switch (kind) {
+	case BTF_KIND_INT:
+	case BTF_KIND_ENUM:
+		return min(sizeof(void *), t->size);
+	case BTF_KIND_PTR:
+		return sizeof(void *);
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_CONST:
+	case BTF_KIND_RESTRICT:
+		return btf__align_of(btf, t->type);
+	case BTF_KIND_ARRAY:
+		return btf__align_of(btf, btf_array(t)->type);
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION: {
+		const struct btf_member *m = btf_members(t);
+		__u16 vlen = btf_vlen(t);
+		int i, max_align = 1, align;
+
+		for (i = 0; i < vlen; i++, m++) {
+			align = btf__align_of(btf, m->type);
+			if (align <= 0)
+				return align;
+			max_align = max(max_align, align);
+		}
+
+		return max_align;
+	}
+	default:
+		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
+		return 0;
+	}
+}
+
 int btf__resolve_type(const struct btf *btf, __u32 type_id)
 {
 	const struct btf_type *t;
@@ -539,6 +578,12 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 		return -ENOENT;
 	}
 
+	/* .extern datasec size and var offsets were set correctly during
+	 * extern collection step, so just skip straight to sorting variables
+	 */
+	if (t->size)
+		goto sort_vars;
+
 	ret = bpf_object__section_size(obj, name, &size);
 	if (ret || !size || (t->size && t->size != size)) {
 		pr_debug("Invalid size for section %s: %u bytes\n", name, size);
@@ -575,7 +620,8 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 		vsi->offset = off;
 	}
 
-	qsort(t + 1, vars, sizeof(*vsi), compare_vsi_off);
+sort_vars:
+	qsort(btf_var_secinfos(t), vars, sizeof(*vsi), compare_vsi_off);
 	return 0;
 }
 
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index d9ac73a02cde..8d73f7f5551f 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -8,14 +8,12 @@
 #include <linux/btf.h>
 #include <linux/types.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 #define BTF_ELF_SEC ".BTF"
 #define BTF_EXT_ELF_SEC ".BTF.ext"
 #define MAPS_ELF_SEC ".maps"
@@ -79,6 +77,7 @@ LIBBPF_API const struct btf_type *btf__type_by_id(const struct btf *btf,
 						  __u32 id);
 LIBBPF_API __s64 btf__resolve_size(const struct btf *btf, __u32 type_id);
 LIBBPF_API int btf__resolve_type(const struct btf *btf, __u32 type_id);
+LIBBPF_API int btf__align_of(const struct btf *btf, __u32 id);
 LIBBPF_API int btf__fd(const struct btf *btf);
 LIBBPF_API const void *btf__get_raw_data(const struct btf *btf, __u32 *size);
 LIBBPF_API const char *btf__name_by_offset(const struct btf *btf, __u32 offset);
@@ -127,6 +126,28 @@ LIBBPF_API void btf_dump__free(struct btf_dump *d);
 
 LIBBPF_API int btf_dump__dump_type(struct btf_dump *d, __u32 id);
 
+struct btf_dump_emit_type_decl_opts {
+	/* size of this struct, for forward/backward compatiblity */
+	size_t sz;
+	/* optional field name for type declaration, e.g.:
+	 * - struct my_struct <FNAME>
+	 * - void (*<FNAME>)(int)
+	 * - char (*<FNAME>)[123]
+	 */
+	const char *field_name;
+	/* extra indentation level (in number of tabs) to emit for multi-line
+	 * type declarations (e.g., anonymous struct); applies for lines
+	 * starting from the second one (first line is assumed to have
+	 * necessary indentation already
+	 */
+	int indent_level;
+};
+#define btf_dump_emit_type_decl_opts__last_field indent_level
+
+LIBBPF_API int
+btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
+			 const struct btf_dump_emit_type_decl_opts *opts);
+
 /*
  * A set of helpers for easier BTF types handling
  */
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index cb126d8fcf75..e95f7710f210 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -116,6 +116,8 @@ static void btf_dump_printf(const struct btf_dump *d, const char *fmt, ...)
 	va_end(args);
 }
 
+static int btf_dump_mark_referenced(struct btf_dump *d);
+
 struct btf_dump *btf_dump__new(const struct btf *btf,
 			       const struct btf_ext *btf_ext,
 			       const struct btf_dump_opts *opts,
@@ -137,18 +139,39 @@ struct btf_dump *btf_dump__new(const struct btf *btf,
 	if (IS_ERR(d->type_names)) {
 		err = PTR_ERR(d->type_names);
 		d->type_names = NULL;
-		btf_dump__free(d);
-		return ERR_PTR(err);
 	}
 	d->ident_names = hashmap__new(str_hash_fn, str_equal_fn, NULL);
 	if (IS_ERR(d->ident_names)) {
 		err = PTR_ERR(d->ident_names);
 		d->ident_names = NULL;
-		btf_dump__free(d);
-		return ERR_PTR(err);
+		goto err;
+	}
+	d->type_states = calloc(1 + btf__get_nr_types(d->btf),
+				sizeof(d->type_states[0]));
+	if (!d->type_states) {
+		err = -ENOMEM;
+		goto err;
+	}
+	d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
+				 sizeof(d->cached_names[0]));
+	if (!d->cached_names) {
+		err = -ENOMEM;
+		goto err;
 	}
 
+	/* VOID is special */
+	d->type_states[0].order_state = ORDERED;
+	d->type_states[0].emit_state = EMITTED;
+
+	/* eagerly determine referenced types for anon enums */
+	err = btf_dump_mark_referenced(d);
+	if (err)
+		goto err;
+
 	return d;
+err:
+	btf_dump__free(d);
+	return ERR_PTR(err);
 }
 
 void btf_dump__free(struct btf_dump *d)
@@ -175,7 +198,6 @@ void btf_dump__free(struct btf_dump *d)
 	free(d);
 }
 
-static int btf_dump_mark_referenced(struct btf_dump *d);
 static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr);
 static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id);
 
@@ -202,27 +224,6 @@ int btf_dump__dump_type(struct btf_dump *d, __u32 id)
 	if (id > btf__get_nr_types(d->btf))
 		return -EINVAL;
 
-	/* type states are lazily allocated, as they might not be needed */
-	if (!d->type_states) {
-		d->type_states = calloc(1 + btf__get_nr_types(d->btf),
-					sizeof(d->type_states[0]));
-		if (!d->type_states)
-			return -ENOMEM;
-		d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
-					 sizeof(d->cached_names[0]));
-		if (!d->cached_names)
-			return -ENOMEM;
-
-		/* VOID is special */
-		d->type_states[0].order_state = ORDERED;
-		d->type_states[0].emit_state = EMITTED;
-
-		/* eagerly determine referenced types for anon enums */
-		err = btf_dump_mark_referenced(d);
-		if (err)
-			return err;
-	}
-
 	d->emit_queue_cnt = 0;
 	err = btf_dump_order_type(d, id, false);
 	if (err < 0)
@@ -752,41 +753,6 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 	}
 }
 
-static int btf_align_of(const struct btf *btf, __u32 id)
-{
-	const struct btf_type *t = btf__type_by_id(btf, id);
-	__u16 kind = btf_kind(t);
-
-	switch (kind) {
-	case BTF_KIND_INT:
-	case BTF_KIND_ENUM:
-		return min(sizeof(void *), t->size);
-	case BTF_KIND_PTR:
-		return sizeof(void *);
-	case BTF_KIND_TYPEDEF:
-	case BTF_KIND_VOLATILE:
-	case BTF_KIND_CONST:
-	case BTF_KIND_RESTRICT:
-		return btf_align_of(btf, t->type);
-	case BTF_KIND_ARRAY:
-		return btf_align_of(btf, btf_array(t)->type);
-	case BTF_KIND_STRUCT:
-	case BTF_KIND_UNION: {
-		const struct btf_member *m = btf_members(t);
-		__u16 vlen = btf_vlen(t);
-		int i, align = 1;
-
-		for (i = 0; i < vlen; i++, m++)
-			align = max(align, btf_align_of(btf, m->type));
-
-		return align;
-	}
-	default:
-		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
-		return 1;
-	}
-}
-
 static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 				 const struct btf_type *t)
 {
@@ -794,18 +760,18 @@ static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 	int align, i, bit_sz;
 	__u16 vlen;
 
-	align = btf_align_of(btf, id);
+	align = btf__align_of(btf, id);
 	/* size of a non-packed struct has to be a multiple of its alignment*/
-	if (t->size % align)
+	if (align && t->size % align)
 		return true;
 
 	m = btf_members(t);
 	vlen = btf_vlen(t);
 	/* all non-bitfield fields have to be naturally aligned */
 	for (i = 0; i < vlen; i++, m++) {
-		align = btf_align_of(btf, m->type);
+		align = btf__align_of(btf, m->type);
 		bit_sz = btf_member_bitfield_size(t, i);
-		if (bit_sz == 0 && m->offset % (8 * align) != 0)
+		if (align && bit_sz == 0 && m->offset % (8 * align) != 0)
 			return true;
 	}
 
@@ -889,7 +855,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 		fname = btf_name_of(d, m->name_off);
 		m_sz = btf_member_bitfield_size(t, i);
 		m_off = btf_member_bit_offset(t, i);
-		align = packed ? 1 : btf_align_of(d->btf, m->type);
+		align = packed ? 1 : btf__align_of(d->btf, m->type);
 
 		btf_dump_emit_bit_padding(d, off, m_off, m_sz, align, lvl + 1);
 		btf_dump_printf(d, "\n%s", pfx(lvl + 1));
@@ -907,7 +873,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 
 	/* pad at the end, if necessary */
 	if (is_struct) {
-		align = packed ? 1 : btf_align_of(d->btf, id);
+		align = packed ? 1 : btf__align_of(d->btf, id);
 		btf_dump_emit_bit_padding(d, off, t->size * 8, 0, align,
 					  lvl + 1);
 	}
@@ -1051,6 +1017,21 @@ static int btf_dump_push_decl_stack_id(struct btf_dump *d, __u32 id)
  * of a stack frame. Some care is required to "pop" stack frames after
  * processing type declaration chain.
  */
+int btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
+			     const struct btf_dump_emit_type_decl_opts *opts)
+{
+	const char *fname;
+	int lvl;
+
+	if (!OPTS_VALID(opts, btf_dump_emit_type_decl_opts))
+		return -EINVAL;
+
+	fname = OPTS_GET(opts, field_name, NULL);
+	lvl = OPTS_GET(opts, indent_level, 0);
+	btf_dump_emit_type_decl(d, id, fname, lvl);
+	return 0;
+}
+
 static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 				    const char *fname, int lvl)
 {
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 3f09772192f1..7513165b104f 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -18,6 +18,7 @@
 #include <stdarg.h>
 #include <libgen.h>
 #include <inttypes.h>
+#include <limits.h>
 #include <string.h>
 #include <unistd.h>
 #include <endian.h>
@@ -41,9 +42,11 @@
 #include <sys/types.h>
 #include <sys/vfs.h>
 #include <sys/utsname.h>
+#include <sys/resource.h>
 #include <tools/libc_compat.h>
 #include <libelf.h>
 #include <gelf.h>
+#include <zlib.h>
 
 #include "libbpf.h"
 #include "bpf.h"
@@ -99,14 +102,33 @@ void libbpf_print(enum libbpf_print_level level, const char *format, ...)
 	va_end(args);
 }
 
-#define STRERR_BUFSIZE  128
+static void pr_perm_msg(int err)
+{
+	struct rlimit limit;
+	char buf[100];
+
+	if (err != -EPERM || geteuid() != 0)
+		return;
+
+	err = getrlimit(RLIMIT_MEMLOCK, &limit);
+	if (err)
+		return;
+
+	if (limit.rlim_cur == RLIM_INFINITY)
+		return;
+
+	if (limit.rlim_cur < 1024)
+		snprintf(buf, sizeof(buf), "%zu bytes", (size_t)limit.rlim_cur);
+	else if (limit.rlim_cur < 1024*1024)
+		snprintf(buf, sizeof(buf), "%.1f KiB", (double)limit.rlim_cur / 1024);
+	else
+		snprintf(buf, sizeof(buf), "%.1f MiB", (double)limit.rlim_cur / (1024*1024));
 
-#define CHECK_ERR(action, err, out) do {	\
-	err = action;			\
-	if (err)			\
-		goto out;		\
-} while (0)
+	pr_warn("permission error while running as root; try raising 'ulimit -l'? current value: %s\n",
+		buf);
+}
 
+#define STRERR_BUFSIZE  128
 
 /* Copied from tools/perf/util/util.h */
 #ifndef zfree
@@ -146,6 +168,20 @@ struct bpf_capabilities {
 	__u32 array_mmap:1;
 };
 
+enum reloc_type {
+	RELO_LD64,
+	RELO_CALL,
+	RELO_DATA,
+	RELO_EXTERN,
+};
+
+struct reloc_desc {
+	enum reloc_type type;
+	int insn_idx;
+	int map_idx;
+	int sym_off;
+};
+
 /*
  * bpf_prog should be a better name but it has been used in
  * linux/filter.h.
@@ -164,16 +200,7 @@ struct bpf_program {
 	size_t insns_cnt, main_prog_cnt;
 	enum bpf_prog_type type;
 
-	struct reloc_desc {
-		enum {
-			RELO_LD64,
-			RELO_CALL,
-			RELO_DATA,
-		} type;
-		int insn_idx;
-		int map_idx;
-		int sym_off;
-	} *reloc_desc;
+	struct reloc_desc *reloc_desc;
 	int nr_reloc;
 	int log_level;
 
@@ -202,22 +229,29 @@ struct bpf_program {
 	__u32 prog_flags;
 };
 
+#define DATA_SEC ".data"
+#define BSS_SEC ".bss"
+#define RODATA_SEC ".rodata"
+#define KCONFIG_SEC ".kconfig"
+
 enum libbpf_map_type {
 	LIBBPF_MAP_UNSPEC,
 	LIBBPF_MAP_DATA,
 	LIBBPF_MAP_BSS,
 	LIBBPF_MAP_RODATA,
+	LIBBPF_MAP_KCONFIG,
 };
 
 static const char * const libbpf_type_to_btf_name[] = {
-	[LIBBPF_MAP_DATA]	= ".data",
-	[LIBBPF_MAP_BSS]	= ".bss",
-	[LIBBPF_MAP_RODATA]	= ".rodata",
+	[LIBBPF_MAP_DATA]	= DATA_SEC,
+	[LIBBPF_MAP_BSS]	= BSS_SEC,
+	[LIBBPF_MAP_RODATA]	= RODATA_SEC,
+	[LIBBPF_MAP_KCONFIG]	= KCONFIG_SEC,
 };
 
 struct bpf_map {
-	int fd;
 	char *name;
+	int fd;
 	int sec_idx;
 	size_t sec_offset;
 	int map_ifindex;
@@ -228,14 +262,32 @@ struct bpf_map {
 	void *priv;
 	bpf_map_clear_priv_t clear_priv;
 	enum libbpf_map_type libbpf_type;
+	void *mmaped;
 	char *pin_path;
 	bool pinned;
 	bool reused;
 };
 
-struct bpf_secdata {
-	void *rodata;
-	void *data;
+enum extern_type {
+	EXT_UNKNOWN,
+	EXT_CHAR,
+	EXT_BOOL,
+	EXT_INT,
+	EXT_TRISTATE,
+	EXT_CHAR_ARR,
+};
+
+struct extern_desc {
+	const char *name;
+	int sym_idx;
+	int btf_id;
+	enum extern_type type;
+	int sz;
+	int align;
+	int data_off;
+	bool is_signed;
+	bool is_weak;
+	bool is_set;
 };
 
 static LIST_HEAD(bpf_objects_list);
@@ -250,7 +302,11 @@ struct bpf_object {
 	struct bpf_map *maps;
 	size_t nr_maps;
 	size_t maps_cap;
-	struct bpf_secdata sections;
+
+	char *kconfig;
+	struct extern_desc *externs;
+	int nr_extern;
+	int kconfig_map_idx;
 
 	bool loaded;
 	bool has_pseudo_calls;
@@ -279,6 +335,7 @@ struct bpf_object {
 		int maps_shndx;
 		int btf_maps_shndx;
 		int text_shndx;
+		int symbols_shndx;
 		int data_shndx;
 		int rodata_shndx;
 		int bss_shndx;
@@ -550,6 +607,7 @@ static struct bpf_object *bpf_object__new(const char *path,
 	obj->efile.data_shndx = -1;
 	obj->efile.rodata_shndx = -1;
 	obj->efile.bss_shndx = -1;
+	obj->kconfig_map_idx = -1;
 
 	obj->kern_version = get_kernel_version();
 	obj->loaded = false;
@@ -748,13 +806,13 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name,
 	*size = 0;
 	if (!name) {
 		return -EINVAL;
-	} else if (!strcmp(name, ".data")) {
+	} else if (!strcmp(name, DATA_SEC)) {
 		if (obj->efile.data)
 			*size = obj->efile.data->d_size;
-	} else if (!strcmp(name, ".bss")) {
+	} else if (!strcmp(name, BSS_SEC)) {
 		if (obj->efile.bss)
 			*size = obj->efile.bss->d_size;
-	} else if (!strcmp(name, ".rodata")) {
+	} else if (!strcmp(name, RODATA_SEC)) {
 		if (obj->efile.rodata)
 			*size = obj->efile.rodata->d_size;
 	} else {
@@ -835,13 +893,38 @@ static struct bpf_map *bpf_object__add_map(struct bpf_object *obj)
 	return &obj->maps[obj->nr_maps++];
 }
 
+static size_t bpf_map_mmap_sz(const struct bpf_map *map)
+{
+	long page_sz = sysconf(_SC_PAGE_SIZE);
+	size_t map_sz;
+
+	map_sz = roundup(map->def.value_size, 8) * map->def.max_entries;
+	map_sz = roundup(map_sz, page_sz);
+	return map_sz;
+}
+
+static char *internal_map_name(struct bpf_object *obj,
+			       enum libbpf_map_type type)
+{
+	char map_name[BPF_OBJ_NAME_LEN];
+	const char *sfx = libbpf_type_to_btf_name[type];
+	int sfx_len = max((size_t)7, strlen(sfx));
+	int pfx_len = min((size_t)BPF_OBJ_NAME_LEN - sfx_len - 1,
+			  strlen(obj->name));
+
+	snprintf(map_name, sizeof(map_name), "%.*s%.*s", pfx_len, obj->name,
+		 sfx_len, libbpf_type_to_btf_name[type]);
+
+	return strdup(map_name);
+}
+
 static int
 bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
-			      int sec_idx, Elf_Data *data, void **data_buff)
+			      int sec_idx, void *data, size_t data_sz)
 {
-	char map_name[BPF_OBJ_NAME_LEN];
 	struct bpf_map_def *def;
 	struct bpf_map *map;
+	int err;
 
 	map = bpf_object__add_map(obj);
 	if (IS_ERR(map))
@@ -850,9 +933,7 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	map->libbpf_type = type;
 	map->sec_idx = sec_idx;
 	map->sec_offset = 0;
-	snprintf(map_name, sizeof(map_name), "%.8s%.7s", obj->name,
-		 libbpf_type_to_btf_name[type]);
-	map->name = strdup(map_name);
+	map->name = internal_map_name(obj, type);
 	if (!map->name) {
 		pr_warn("failed to alloc map name\n");
 		return -ENOMEM;
@@ -861,25 +942,29 @@ bpf_object__init_internal_map(struct bpf_object *obj, enum libbpf_map_type type,
 	def = &map->def;
 	def->type = BPF_MAP_TYPE_ARRAY;
 	def->key_size = sizeof(int);
-	def->value_size = data->d_size;
+	def->value_size = data_sz;
 	def->max_entries = 1;
-	def->map_flags = type == LIBBPF_MAP_RODATA ? BPF_F_RDONLY_PROG : 0;
-	if (obj->caps.array_mmap)
-		def->map_flags |= BPF_F_MMAPABLE;
+	def->map_flags = type == LIBBPF_MAP_RODATA || type == LIBBPF_MAP_KCONFIG
+			 ? BPF_F_RDONLY_PROG : 0;
+	def->map_flags |= BPF_F_MMAPABLE;
 
 	pr_debug("map '%s' (global data): at sec_idx %d, offset %zu, flags %x.\n",
-		 map_name, map->sec_idx, map->sec_offset, def->map_flags);
+		 map->name, map->sec_idx, map->sec_offset, def->map_flags);
 
-	if (data_buff) {
-		*data_buff = malloc(data->d_size);
-		if (!*data_buff) {
-			zfree(&map->name);
-			pr_warn("failed to alloc map content buffer\n");
-			return -ENOMEM;
-		}
-		memcpy(*data_buff, data->d_buf, data->d_size);
+	map->mmaped = mmap(NULL, bpf_map_mmap_sz(map), PROT_READ | PROT_WRITE,
+			   MAP_SHARED | MAP_ANONYMOUS, -1, 0);
+	if (map->mmaped == MAP_FAILED) {
+		err = -errno;
+		map->mmaped = NULL;
+		pr_warn("failed to alloc map '%s' content buffer: %d\n",
+			map->name, err);
+		zfree(&map->name);
+		return err;
 	}
 
+	if (data)
+		memcpy(map->mmaped, data, data_sz);
+
 	pr_debug("map %td is \"%s\"\n", map - obj->maps, map->name);
 	return 0;
 }
@@ -888,37 +973,332 @@ static int bpf_object__init_global_data_maps(struct bpf_object *obj)
 {
 	int err;
 
-	if (!obj->caps.global_data)
-		return 0;
 	/*
 	 * Populate obj->maps with libbpf internal maps.
 	 */
 	if (obj->efile.data_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_DATA,
 						    obj->efile.data_shndx,
-						    obj->efile.data,
-						    &obj->sections.data);
+						    obj->efile.data->d_buf,
+						    obj->efile.data->d_size);
 		if (err)
 			return err;
 	}
 	if (obj->efile.rodata_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_RODATA,
 						    obj->efile.rodata_shndx,
-						    obj->efile.rodata,
-						    &obj->sections.rodata);
+						    obj->efile.rodata->d_buf,
+						    obj->efile.rodata->d_size);
 		if (err)
 			return err;
 	}
 	if (obj->efile.bss_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_BSS,
 						    obj->efile.bss_shndx,
-						    obj->efile.bss, NULL);
+						    NULL,
+						    obj->efile.bss->d_size);
 		if (err)
 			return err;
 	}
 	return 0;
 }
 
+
+static struct extern_desc *find_extern_by_name(const struct bpf_object *obj,
+					       const void *name)
+{
+	int i;
+
+	for (i = 0; i < obj->nr_extern; i++) {
+		if (strcmp(obj->externs[i].name, name) == 0)
+			return &obj->externs[i];
+	}
+	return NULL;
+}
+
+static int set_ext_value_tri(struct extern_desc *ext, void *ext_val,
+			     char value)
+{
+	switch (ext->type) {
+	case EXT_BOOL:
+		if (value == 'm') {
+			pr_warn("extern %s=%c should be tristate or char\n",
+				ext->name, value);
+			return -EINVAL;
+		}
+		*(bool *)ext_val = value == 'y' ? true : false;
+		break;
+	case EXT_TRISTATE:
+		if (value == 'y')
+			*(enum libbpf_tristate *)ext_val = TRI_YES;
+		else if (value == 'm')
+			*(enum libbpf_tristate *)ext_val = TRI_MODULE;
+		else /* value == 'n' */
+			*(enum libbpf_tristate *)ext_val = TRI_NO;
+		break;
+	case EXT_CHAR:
+		*(char *)ext_val = value;
+		break;
+	case EXT_UNKNOWN:
+	case EXT_INT:
+	case EXT_CHAR_ARR:
+	default:
+		pr_warn("extern %s=%c should be bool, tristate, or char\n",
+			ext->name, value);
+		return -EINVAL;
+	}
+	ext->is_set = true;
+	return 0;
+}
+
+static int set_ext_value_str(struct extern_desc *ext, char *ext_val,
+			     const char *value)
+{
+	size_t len;
+
+	if (ext->type != EXT_CHAR_ARR) {
+		pr_warn("extern %s=%s should char array\n", ext->name, value);
+		return -EINVAL;
+	}
+
+	len = strlen(value);
+	if (value[len - 1] != '"') {
+		pr_warn("extern '%s': invalid string config '%s'\n",
+			ext->name, value);
+		return -EINVAL;
+	}
+
+	/* strip quotes */
+	len -= 2;
+	if (len >= ext->sz) {
+		pr_warn("extern '%s': long string config %s of (%zu bytes) truncated to %d bytes\n",
+			ext->name, value, len, ext->sz - 1);
+		len = ext->sz - 1;
+	}
+	memcpy(ext_val, value + 1, len);
+	ext_val[len] = '\0';
+	ext->is_set = true;
+	return 0;
+}
+
+static int parse_u64(const char *value, __u64 *res)
+{
+	char *value_end;
+	int err;
+
+	errno = 0;
+	*res = strtoull(value, &value_end, 0);
+	if (errno) {
+		err = -errno;
+		pr_warn("failed to parse '%s' as integer: %d\n", value, err);
+		return err;
+	}
+	if (*value_end) {
+		pr_warn("failed to parse '%s' as integer completely\n", value);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static bool is_ext_value_in_range(const struct extern_desc *ext, __u64 v)
+{
+	int bit_sz = ext->sz * 8;
+
+	if (ext->sz == 8)
+		return true;
+
+	/* Validate that value stored in u64 fits in integer of `ext->sz`
+	 * bytes size without any loss of information. If the target integer
+	 * is signed, we rely on the following limits of integer type of
+	 * Y bits and subsequent transformation:
+	 *
+	 *     -2^(Y-1) <= X           <= 2^(Y-1) - 1
+	 *            0 <= X + 2^(Y-1) <= 2^Y - 1
+	 *            0 <= X + 2^(Y-1) <  2^Y
+	 *
+	 *  For unsigned target integer, check that all the (64 - Y) bits are
+	 *  zero.
+	 */
+	if (ext->is_signed)
+		return v + (1ULL << (bit_sz - 1)) < (1ULL << bit_sz);
+	else
+		return (v >> bit_sz) == 0;
+}
+
+static int set_ext_value_num(struct extern_desc *ext, void *ext_val,
+			     __u64 value)
+{
+	if (ext->type != EXT_INT && ext->type != EXT_CHAR) {
+		pr_warn("extern %s=%llu should be integer\n",
+			ext->name, (unsigned long long)value);
+		return -EINVAL;
+	}
+	if (!is_ext_value_in_range(ext, value)) {
+		pr_warn("extern %s=%llu value doesn't fit in %d bytes\n",
+			ext->name, (unsigned long long)value, ext->sz);
+		return -ERANGE;
+	}
+	switch (ext->sz) {
+		case 1: *(__u8 *)ext_val = value; break;
+		case 2: *(__u16 *)ext_val = value; break;
+		case 4: *(__u32 *)ext_val = value; break;
+		case 8: *(__u64 *)ext_val = value; break;
+		default:
+			return -EINVAL;
+	}
+	ext->is_set = true;
+	return 0;
+}
+
+static int bpf_object__process_kconfig_line(struct bpf_object *obj,
+					    char *buf, void *data)
+{
+	struct extern_desc *ext;
+	char *sep, *value;
+	int len, err = 0;
+	void *ext_val;
+	__u64 num;
+
+	if (strncmp(buf, "CONFIG_", 7))
+		return 0;
+
+	sep = strchr(buf, '=');
+	if (!sep) {
+		pr_warn("failed to parse '%s': no separator\n", buf);
+		return -EINVAL;
+	}
+
+	/* Trim ending '\n' */
+	len = strlen(buf);
+	if (buf[len - 1] == '\n')
+		buf[len - 1] = '\0';
+	/* Split on '=' and ensure that a value is present. */
+	*sep = '\0';
+	if (!sep[1]) {
+		*sep = '=';
+		pr_warn("failed to parse '%s': no value\n", buf);
+		return -EINVAL;
+	}
+
+	ext = find_extern_by_name(obj, buf);
+	if (!ext || ext->is_set)
+		return 0;
+
+	ext_val = data + ext->data_off;
+	value = sep + 1;
+
+	switch (*value) {
+	case 'y': case 'n': case 'm':
+		err = set_ext_value_tri(ext, ext_val, *value);
+		break;
+	case '"':
+		err = set_ext_value_str(ext, ext_val, value);
+		break;
+	default:
+		/* assume integer */
+		err = parse_u64(value, &num);
+		if (err) {
+			pr_warn("extern %s=%s should be integer\n",
+				ext->name, value);
+			return err;
+		}
+		err = set_ext_value_num(ext, ext_val, num);
+		break;
+	}
+	if (err)
+		return err;
+	pr_debug("extern %s=%s\n", ext->name, value);
+	return 0;
+}
+
+static int bpf_object__read_kconfig_file(struct bpf_object *obj, void *data)
+{
+	char buf[PATH_MAX];
+	struct utsname uts;
+	int len, err = 0;
+	gzFile file;
+
+	uname(&uts);
+	len = snprintf(buf, PATH_MAX, "/boot/config-%s", uts.release);
+	if (len < 0)
+		return -EINVAL;
+	else if (len >= PATH_MAX)
+		return -ENAMETOOLONG;
+
+	/* gzopen also accepts uncompressed files. */
+	file = gzopen(buf, "r");
+	if (!file)
+		file = gzopen("/proc/config.gz", "r");
+
+	if (!file) {
+		pr_warn("failed to open system Kconfig\n");
+		return -ENOENT;
+	}
+
+	while (gzgets(file, buf, sizeof(buf))) {
+		err = bpf_object__process_kconfig_line(obj, buf, data);
+		if (err) {
+			pr_warn("error parsing system Kconfig line '%s': %d\n",
+				buf, err);
+			goto out;
+		}
+	}
+
+out:
+	gzclose(file);
+	return err;
+}
+
+static int bpf_object__read_kconfig_mem(struct bpf_object *obj,
+					const char *config, void *data)
+{
+	char buf[PATH_MAX];
+	int err = 0;
+	FILE *file;
+
+	file = fmemopen((void *)config, strlen(config), "r");
+	if (!file) {
+		err = -errno;
+		pr_warn("failed to open in-memory Kconfig: %d\n", err);
+		return err;
+	}
+
+	while (fgets(buf, sizeof(buf), file)) {
+		err = bpf_object__process_kconfig_line(obj, buf, data);
+		if (err) {
+			pr_warn("error parsing in-memory Kconfig line '%s': %d\n",
+				buf, err);
+			break;
+		}
+	}
+
+	fclose(file);
+	return err;
+}
+
+static int bpf_object__init_kconfig_map(struct bpf_object *obj)
+{
+	struct extern_desc *last_ext;
+	size_t map_sz;
+	int err;
+
+	if (obj->nr_extern == 0)
+		return 0;
+
+	last_ext = &obj->externs[obj->nr_extern - 1];
+	map_sz = last_ext->data_off + last_ext->sz;
+
+	err = bpf_object__init_internal_map(obj, LIBBPF_MAP_KCONFIG,
+					    obj->efile.symbols_shndx,
+					    NULL, map_sz);
+	if (err)
+		return err;
+
+	obj->kconfig_map_idx = obj->nr_maps - 1;
+
+	return 0;
+}
+
 static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
 {
 	Elf_Data *symbols = obj->efile.symbols;
@@ -1242,15 +1622,15 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
 			}
 			sz = btf__resolve_size(obj->btf, t->type);
 			if (sz < 0) {
-				pr_warn("map '%s': can't determine key size for type [%u]: %lld.\n",
-					map_name, t->type, sz);
+				pr_warn("map '%s': can't determine key size for type [%u]: %zd.\n",
+					map_name, t->type, (ssize_t)sz);
 				return sz;
 			}
-			pr_debug("map '%s': found key [%u], sz = %lld.\n",
-				 map_name, t->type, sz);
+			pr_debug("map '%s': found key [%u], sz = %zd.\n",
+				 map_name, t->type, (ssize_t)sz);
 			if (map->def.key_size && map->def.key_size != sz) {
-				pr_warn("map '%s': conflicting key size %u != %lld.\n",
-					map_name, map->def.key_size, sz);
+				pr_warn("map '%s': conflicting key size %u != %zd.\n",
+					map_name, map->def.key_size, (ssize_t)sz);
 				return -EINVAL;
 			}
 			map->def.key_size = sz;
@@ -1285,15 +1665,15 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
 			}
 			sz = btf__resolve_size(obj->btf, t->type);
 			if (sz < 0) {
-				pr_warn("map '%s': can't determine value size for type [%u]: %lld.\n",
-					map_name, t->type, sz);
+				pr_warn("map '%s': can't determine value size for type [%u]: %zd.\n",
+					map_name, t->type, (ssize_t)sz);
 				return sz;
 			}
-			pr_debug("map '%s': found value [%u], sz = %lld.\n",
-				 map_name, t->type, sz);
+			pr_debug("map '%s': found value [%u], sz = %zd.\n",
+				 map_name, t->type, (ssize_t)sz);
 			if (map->def.value_size && map->def.value_size != sz) {
-				pr_warn("map '%s': conflicting value size %u != %lld.\n",
-					map_name, map->def.value_size, sz);
+				pr_warn("map '%s': conflicting value size %u != %zd.\n",
+					map_name, map->def.value_size, (ssize_t)sz);
 				return -EINVAL;
 			}
 			map->def.value_size = sz;
@@ -1393,21 +1773,20 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict,
 	return 0;
 }
 
-static int bpf_object__init_maps(struct bpf_object *obj, bool relaxed_maps,
-				 const char *pin_root_path)
+static int bpf_object__init_maps(struct bpf_object *obj,
+				 const struct bpf_object_open_opts *opts)
 {
-	bool strict = !relaxed_maps;
+	const char *pin_root_path;
+	bool strict;
 	int err;
 
-	err = bpf_object__init_user_maps(obj, strict);
-	if (err)
-		return err;
-
-	err = bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
-	if (err)
-		return err;
+	strict = !OPTS_GET(opts, relaxed_maps, false);
+	pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
 
-	err = bpf_object__init_global_data_maps(obj);
+	err = bpf_object__init_user_maps(obj, strict);
+	err = err ?: bpf_object__init_user_btf_maps(obj, strict, pin_root_path);
+	err = err ?: bpf_object__init_global_data_maps(obj);
+	err = err ?: bpf_object__init_kconfig_map(obj);
 	if (err)
 		return err;
 
@@ -1509,7 +1888,8 @@ static void bpf_object__sanitize_btf_ext(struct bpf_object *obj)
 
 static bool bpf_object__is_btf_mandatory(const struct bpf_object *obj)
 {
-	return obj->efile.btf_maps_shndx >= 0;
+	return obj->efile.btf_maps_shndx >= 0 ||
+	       obj->nr_extern > 0;
 }
 
 static int bpf_object__init_btf(struct bpf_object *obj,
@@ -1526,11 +1906,6 @@ static int bpf_object__init_btf(struct bpf_object *obj,
 				BTF_ELF_SEC, err);
 			goto out;
 		}
-		err = btf__finalize_data(obj, obj->btf);
-		if (err) {
-			pr_warn("Error finalizing %s: %d.\n", BTF_ELF_SEC, err);
-			goto out;
-		}
 	}
 	if (btf_ext_data) {
 		if (!obj->btf) {
@@ -1564,6 +1939,30 @@ out:
 	return 0;
 }
 
+static int bpf_object__finalize_btf(struct bpf_object *obj)
+{
+	int err;
+
+	if (!obj->btf)
+		return 0;
+
+	err = btf__finalize_data(obj, obj->btf);
+	if (!err)
+		return 0;
+
+	pr_warn("Error finalizing %s: %d.\n", BTF_ELF_SEC, err);
+	btf__free(obj->btf);
+	obj->btf = NULL;
+	btf_ext__free(obj->btf_ext);
+	obj->btf_ext = NULL;
+
+	if (bpf_object__is_btf_mandatory(obj)) {
+		pr_warn("BTF is required, but is missing or corrupted.\n");
+		return -ENOENT;
+	}
+	return 0;
+}
+
 static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 {
 	int err = 0;
@@ -1592,8 +1991,7 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 	return 0;
 }
 
-static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
-				   const char *pin_root_path)
+static int bpf_object__elf_collect(struct bpf_object *obj)
 {
 	Elf *elf = obj->efile.elf;
 	GElf_Ehdr *ep = &obj->efile.ehdr;
@@ -1665,6 +2063,7 @@ static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
 				return -LIBBPF_ERRNO__FORMAT;
 			}
 			obj->efile.symbols = data;
+			obj->efile.symbols_shndx = idx;
 			obj->efile.strtabidx = sh.sh_link;
 		} else if (sh.sh_type == SHT_PROGBITS && data->d_size > 0) {
 			if (sh.sh_flags & SHF_EXECINSTR) {
@@ -1683,10 +2082,10 @@ static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
 						name, obj->path, cp);
 					return err;
 				}
-			} else if (strcmp(name, ".data") == 0) {
+			} else if (strcmp(name, DATA_SEC) == 0) {
 				obj->efile.data = data;
 				obj->efile.data_shndx = idx;
-			} else if (strcmp(name, ".rodata") == 0) {
+			} else if (strcmp(name, RODATA_SEC) == 0) {
 				obj->efile.rodata = data;
 				obj->efile.rodata_shndx = idx;
 			} else {
@@ -1716,7 +2115,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
 
 			obj->efile.reloc_sects[nr_sects].shdr = sh;
 			obj->efile.reloc_sects[nr_sects].data = data;
-		} else if (sh.sh_type == SHT_NOBITS && strcmp(name, ".bss") == 0) {
+		} else if (sh.sh_type == SHT_NOBITS &&
+			   strcmp(name, BSS_SEC) == 0) {
 			obj->efile.bss = data;
 			obj->efile.bss_shndx = idx;
 		} else {
@@ -1728,14 +2128,217 @@ static int bpf_object__elf_collect(struct bpf_object *obj, bool relaxed_maps,
 		pr_warn("Corrupted ELF file: index of strtab invalid\n");
 		return -LIBBPF_ERRNO__FORMAT;
 	}
-	err = bpf_object__init_btf(obj, btf_data, btf_ext_data);
-	if (!err)
-		err = bpf_object__init_maps(obj, relaxed_maps, pin_root_path);
-	if (!err)
-		err = bpf_object__sanitize_and_load_btf(obj);
-	if (!err)
-		err = bpf_object__init_prog_names(obj);
-	return err;
+	return bpf_object__init_btf(obj, btf_data, btf_ext_data);
+}
+
+static bool sym_is_extern(const GElf_Sym *sym)
+{
+	int bind = GELF_ST_BIND(sym->st_info);
+	/* externs are symbols w/ type=NOTYPE, bind=GLOBAL|WEAK, section=UND */
+	return sym->st_shndx == SHN_UNDEF &&
+	       (bind == STB_GLOBAL || bind == STB_WEAK) &&
+	       GELF_ST_TYPE(sym->st_info) == STT_NOTYPE;
+}
+
+static int find_extern_btf_id(const struct btf *btf, const char *ext_name)
+{
+	const struct btf_type *t;
+	const char *var_name;
+	int i, n;
+
+	if (!btf)
+		return -ESRCH;
+
+	n = btf__get_nr_types(btf);
+	for (i = 1; i <= n; i++) {
+		t = btf__type_by_id(btf, i);
+
+		if (!btf_is_var(t))
+			continue;
+
+		var_name = btf__name_by_offset(btf, t->name_off);
+		if (strcmp(var_name, ext_name))
+			continue;
+
+		if (btf_var(t)->linkage != BTF_VAR_GLOBAL_EXTERN)
+			return -EINVAL;
+
+		return i;
+	}
+
+	return -ENOENT;
+}
+
+static enum extern_type find_extern_type(const struct btf *btf, int id,
+					 bool *is_signed)
+{
+	const struct btf_type *t;
+	const char *name;
+
+	t = skip_mods_and_typedefs(btf, id, NULL);
+	name = btf__name_by_offset(btf, t->name_off);
+
+	if (is_signed)
+		*is_signed = false;
+	switch (btf_kind(t)) {
+	case BTF_KIND_INT: {
+		int enc = btf_int_encoding(t);
+
+		if (enc & BTF_INT_BOOL)
+			return t->size == 1 ? EXT_BOOL : EXT_UNKNOWN;
+		if (is_signed)
+			*is_signed = enc & BTF_INT_SIGNED;
+		if (t->size == 1)
+			return EXT_CHAR;
+		if (t->size < 1 || t->size > 8 || (t->size & (t->size - 1)))
+			return EXT_UNKNOWN;
+		return EXT_INT;
+	}
+	case BTF_KIND_ENUM:
+		if (t->size != 4)
+			return EXT_UNKNOWN;
+		if (strcmp(name, "libbpf_tristate"))
+			return EXT_UNKNOWN;
+		return EXT_TRISTATE;
+	case BTF_KIND_ARRAY:
+		if (btf_array(t)->nelems == 0)
+			return EXT_UNKNOWN;
+		if (find_extern_type(btf, btf_array(t)->type, NULL) != EXT_CHAR)
+			return EXT_UNKNOWN;
+		return EXT_CHAR_ARR;
+	default:
+		return EXT_UNKNOWN;
+	}
+}
+
+static int cmp_externs(const void *_a, const void *_b)
+{
+	const struct extern_desc *a = _a;
+	const struct extern_desc *b = _b;
+
+	/* descending order by alignment requirements */
+	if (a->align != b->align)
+		return a->align > b->align ? -1 : 1;
+	/* ascending order by size, within same alignment class */
+	if (a->sz != b->sz)
+		return a->sz < b->sz ? -1 : 1;
+	/* resolve ties by name */
+	return strcmp(a->name, b->name);
+}
+
+static int bpf_object__collect_externs(struct bpf_object *obj)
+{
+	const struct btf_type *t;
+	struct extern_desc *ext;
+	int i, n, off, btf_id;
+	struct btf_type *sec;
+	const char *ext_name;
+	Elf_Scn *scn;
+	GElf_Shdr sh;
+
+	if (!obj->efile.symbols)
+		return 0;
+
+	scn = elf_getscn(obj->efile.elf, obj->efile.symbols_shndx);
+	if (!scn)
+		return -LIBBPF_ERRNO__FORMAT;
+	if (gelf_getshdr(scn, &sh) != &sh)
+		return -LIBBPF_ERRNO__FORMAT;
+	n = sh.sh_size / sh.sh_entsize;
+
+	pr_debug("looking for externs among %d symbols...\n", n);
+	for (i = 0; i < n; i++) {
+		GElf_Sym sym;
+
+		if (!gelf_getsym(obj->efile.symbols, i, &sym))
+			return -LIBBPF_ERRNO__FORMAT;
+		if (!sym_is_extern(&sym))
+			continue;
+		ext_name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
+				      sym.st_name);
+		if (!ext_name || !ext_name[0])
+			continue;
+
+		ext = obj->externs;
+		ext = reallocarray(ext, obj->nr_extern + 1, sizeof(*ext));
+		if (!ext)
+			return -ENOMEM;
+		obj->externs = ext;
+		ext = &ext[obj->nr_extern];
+		memset(ext, 0, sizeof(*ext));
+		obj->nr_extern++;
+
+		ext->btf_id = find_extern_btf_id(obj->btf, ext_name);
+		if (ext->btf_id <= 0) {
+			pr_warn("failed to find BTF for extern '%s': %d\n",
+				ext_name, ext->btf_id);
+			return ext->btf_id;
+		}
+		t = btf__type_by_id(obj->btf, ext->btf_id);
+		ext->name = btf__name_by_offset(obj->btf, t->name_off);
+		ext->sym_idx = i;
+		ext->is_weak = GELF_ST_BIND(sym.st_info) == STB_WEAK;
+		ext->sz = btf__resolve_size(obj->btf, t->type);
+		if (ext->sz <= 0) {
+			pr_warn("failed to resolve size of extern '%s': %d\n",
+				ext_name, ext->sz);
+			return ext->sz;
+		}
+		ext->align = btf__align_of(obj->btf, t->type);
+		if (ext->align <= 0) {
+			pr_warn("failed to determine alignment of extern '%s': %d\n",
+				ext_name, ext->align);
+			return -EINVAL;
+		}
+		ext->type = find_extern_type(obj->btf, t->type,
+					     &ext->is_signed);
+		if (ext->type == EXT_UNKNOWN) {
+			pr_warn("extern '%s' type is unsupported\n", ext_name);
+			return -ENOTSUP;
+		}
+	}
+	pr_debug("collected %d externs total\n", obj->nr_extern);
+
+	if (!obj->nr_extern)
+		return 0;
+
+	/* sort externs by (alignment, size, name) and calculate their offsets
+	 * within a map */
+	qsort(obj->externs, obj->nr_extern, sizeof(*ext), cmp_externs);
+	off = 0;
+	for (i = 0; i < obj->nr_extern; i++) {
+		ext = &obj->externs[i];
+		ext->data_off = roundup(off, ext->align);
+		off = ext->data_off + ext->sz;
+		pr_debug("extern #%d: symbol %d, off %u, name %s\n",
+			 i, ext->sym_idx, ext->data_off, ext->name);
+	}
+
+	btf_id = btf__find_by_name(obj->btf, KCONFIG_SEC);
+	if (btf_id <= 0) {
+		pr_warn("no BTF info found for '%s' datasec\n", KCONFIG_SEC);
+		return -ESRCH;
+	}
+
+	sec = (struct btf_type *)btf__type_by_id(obj->btf, btf_id);
+	sec->size = off;
+	n = btf_vlen(sec);
+	for (i = 0; i < n; i++) {
+		struct btf_var_secinfo *vs = btf_var_secinfos(sec) + i;
+
+		t = btf__type_by_id(obj->btf, vs->type);
+		ext_name = btf__name_by_offset(obj->btf, t->name_off);
+		ext = find_extern_by_name(obj, ext_name);
+		if (!ext) {
+			pr_warn("failed to find extern definition for BTF var '%s'\n",
+				ext_name);
+			return -ESRCH;
+		}
+		vs->offset = ext->data_off;
+		btf_var(t)->linkage = BTF_VAR_GLOBAL_ALLOCATED;
+	}
+
+	return 0;
 }
 
 static struct bpf_program *
@@ -1765,6 +2368,19 @@ bpf_object__find_program_by_title(const struct bpf_object *obj,
 	return NULL;
 }
 
+struct bpf_program *
+bpf_object__find_program_by_name(const struct bpf_object *obj,
+				 const char *name)
+{
+	struct bpf_program *prog;
+
+	bpf_object__for_each_program(prog, obj) {
+		if (!strcmp(prog->name, name))
+			return prog;
+	}
+	return NULL;
+}
+
 static bool bpf_object__shndx_is_data(const struct bpf_object *obj,
 				      int shndx)
 {
@@ -1789,6 +2405,8 @@ bpf_object__section_to_libbpf_map_type(const struct bpf_object *obj, int shndx)
 		return LIBBPF_MAP_BSS;
 	else if (shndx == obj->efile.rodata_shndx)
 		return LIBBPF_MAP_RODATA;
+	else if (shndx == obj->efile.symbols_shndx)
+		return LIBBPF_MAP_KCONFIG;
 	else
 		return LIBBPF_MAP_UNSPEC;
 }
@@ -1817,7 +2435,8 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		if (sym->st_value % 8) {
-			pr_warn("bad call relo offset: %llu\n", (__u64)sym->st_value);
+			pr_warn("bad call relo offset: %zu\n",
+				(size_t)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_CALL;
@@ -1832,6 +2451,30 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			insn_idx, insn->code);
 		return -LIBBPF_ERRNO__RELOC;
 	}
+
+	if (sym_is_extern(sym)) {
+		int sym_idx = GELF_R_SYM(rel->r_info);
+		int i, n = obj->nr_extern;
+		struct extern_desc *ext;
+
+		for (i = 0; i < n; i++) {
+			ext = &obj->externs[i];
+			if (ext->sym_idx == sym_idx)
+				break;
+		}
+		if (i >= n) {
+			pr_warn("extern relo failed to find extern for sym %d\n",
+				sym_idx);
+			return -LIBBPF_ERRNO__RELOC;
+		}
+		pr_debug("found extern #%d '%s' (sym %d, off %u) for insn %u\n",
+			 i, ext->name, ext->sym_idx, ext->data_off, insn_idx);
+		reloc_desc->type = RELO_EXTERN;
+		reloc_desc->insn_idx = insn_idx;
+		reloc_desc->sym_off = ext->data_off;
+		return 0;
+	}
+
 	if (!shdr_idx || shdr_idx >= SHN_LORESERVE) {
 		pr_warn("invalid relo for \'%s\' in special section 0x%x; forgot to initialize global var?..\n",
 			name, shdr_idx);
@@ -1859,8 +2502,8 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			break;
 		}
 		if (map_idx >= nr_maps) {
-			pr_warn("map relo failed to find map for sec %u, off %llu\n",
-				shdr_idx, (__u64)sym->st_value);
+			pr_warn("map relo failed to find map for sec %u, off %zu\n",
+				shdr_idx, (size_t)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_LD64;
@@ -1875,11 +2518,6 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 		pr_warn("bad data relo against section %u\n", shdr_idx);
 		return -LIBBPF_ERRNO__RELOC;
 	}
-	if (!obj->caps.global_data) {
-		pr_warn("relocation: kernel does not support global \'%s\' variable access in insns[%d]\n",
-			name, insn_idx);
-		return -LIBBPF_ERRNO__RELOC;
-	}
 	for (map_idx = 0; map_idx < nr_maps; map_idx++) {
 		map = &obj->maps[map_idx];
 		if (map->libbpf_type != type)
@@ -1941,9 +2579,9 @@ bpf_program__collect_reloc(struct bpf_program *prog, GElf_Shdr *shdr,
 		name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
 				  sym.st_name) ? : "<?>";
 
-		pr_debug("relo for shdr %u, symb %llu, value %llu, type %d, bind %d, name %d (\'%s\'), insn %u\n",
-			 (__u32)sym.st_shndx, (__u64)GELF_R_SYM(rel.r_info),
-			 (__u64)sym.st_value, GELF_ST_TYPE(sym.st_info),
+		pr_debug("relo for shdr %u, symb %zu, value %zu, type %d, bind %d, name %d (\'%s\'), insn %u\n",
+			 (__u32)sym.st_shndx, (size_t)GELF_R_SYM(rel.r_info),
+			 (size_t)sym.st_value, GELF_ST_TYPE(sym.st_info),
 			 GELF_ST_BIND(sym.st_info), sym.st_name, name,
 			 insn_idx);
 
@@ -2298,29 +2936,35 @@ bpf_object__reuse_map(struct bpf_map *map)
 static int
 bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
 {
+	enum libbpf_map_type map_type = map->libbpf_type;
 	char *cp, errmsg[STRERR_BUFSIZE];
 	int err, zero = 0;
-	__u8 *data;
 
-	/* Nothing to do here since kernel already zero-initializes .bss map. */
-	if (map->libbpf_type == LIBBPF_MAP_BSS)
+	/* kernel already zero-initializes .bss map. */
+	if (map_type == LIBBPF_MAP_BSS)
 		return 0;
 
-	data = map->libbpf_type == LIBBPF_MAP_DATA ?
-	       obj->sections.data : obj->sections.rodata;
+	err = bpf_map_update_elem(map->fd, &zero, map->mmaped, 0);
+	if (err) {
+		err = -errno;
+		cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
+		pr_warn("Error setting initial map(%s) contents: %s\n",
+			map->name, cp);
+		return err;
+	}
 
-	err = bpf_map_update_elem(map->fd, &zero, data, 0);
-	/* Freeze .rodata map as read-only from syscall side. */
-	if (!err && map->libbpf_type == LIBBPF_MAP_RODATA) {
+	/* Freeze .rodata and .kconfig map as read-only from syscall side. */
+	if (map_type == LIBBPF_MAP_RODATA || map_type == LIBBPF_MAP_KCONFIG) {
 		err = bpf_map_freeze(map->fd);
 		if (err) {
-			cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
+			err = -errno;
+			cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
 			pr_warn("Error freezing map(%s) as read-only: %s\n",
 				map->name, cp);
-			err = 0;
+			return err;
 		}
 	}
-	return err;
+	return 0;
 }
 
 static int
@@ -2411,6 +3055,7 @@ err_out:
 			cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
 			pr_warn("failed to create map (name: '%s'): %s(%d)\n",
 				map->name, cp, err);
+			pr_perm_msg(err);
 			for (j = 0; j < i; j++)
 				zclose(obj->maps[j].fd);
 			return err;
@@ -2536,6 +3181,21 @@ static bool str_is_empty(const char *s)
 	return !s || !s[0];
 }
 
+static bool is_flex_arr(const struct btf *btf,
+			const struct bpf_core_accessor *acc,
+			const struct btf_array *arr)
+{
+	const struct btf_type *t;
+
+	/* not a flexible array, if not inside a struct or has non-zero size */
+	if (!acc->name || arr->nelems > 0)
+		return false;
+
+	/* has to be the last member of enclosing struct */
+	t = btf__type_by_id(btf, acc->type_id);
+	return acc->idx == btf_vlen(t) - 1;
+}
+
 /*
  * Turn bpf_field_reloc into a low- and high-level spec representation,
  * validating correctness along the way, as well as calculating resulting
@@ -2573,6 +3233,7 @@ static int bpf_core_spec_parse(const struct btf *btf,
 			       struct bpf_core_spec *spec)
 {
 	int access_idx, parsed_len, i;
+	struct bpf_core_accessor *acc;
 	const struct btf_type *t;
 	const char *name;
 	__u32 id;
@@ -2620,6 +3281,7 @@ static int bpf_core_spec_parse(const struct btf *btf,
 			return -EINVAL;
 
 		access_idx = spec->raw_spec[i];
+		acc = &spec->spec[spec->len];
 
 		if (btf_is_composite(t)) {
 			const struct btf_member *m;
@@ -2637,18 +3299,23 @@ static int bpf_core_spec_parse(const struct btf *btf,
 				if (str_is_empty(name))
 					return -EINVAL;
 
-				spec->spec[spec->len].type_id = id;
-				spec->spec[spec->len].idx = access_idx;
-				spec->spec[spec->len].name = name;
+				acc->type_id = id;
+				acc->idx = access_idx;
+				acc->name = name;
 				spec->len++;
 			}
 
 			id = m->type;
 		} else if (btf_is_array(t)) {
 			const struct btf_array *a = btf_array(t);
+			bool flex;
 
 			t = skip_mods_and_typedefs(btf, a->type, &id);
-			if (!t || access_idx >= a->nelems)
+			if (!t)
+				return -EINVAL;
+
+			flex = is_flex_arr(btf, acc - 1, a);
+			if (!flex && access_idx >= a->nelems)
 				return -EINVAL;
 
 			spec->spec[spec->len].type_id = id;
@@ -2953,12 +3620,14 @@ static int bpf_core_spec_match(struct bpf_core_spec *local_spec,
 			 */
 			if (i > 0) {
 				const struct btf_array *a;
+				bool flex;
 
 				if (!btf_is_array(targ_type))
 					return 0;
 
 				a = btf_array(targ_type);
-				if (local_acc->idx >= a->nelems)
+				flex = is_flex_arr(targ_btf, targ_acc - 1, a);
+				if (!flex && local_acc->idx >= a->nelems)
 					return 0;
 				if (!skip_mods_and_typedefs(targ_btf, a->type,
 							    &targ_id))
@@ -3142,11 +3811,13 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 	insn = &prog->insns[insn_idx];
 	class = BPF_CLASS(insn->code);
 
-	if (class == BPF_ALU || class == BPF_ALU64) {
+	switch (class) {
+	case BPF_ALU:
+	case BPF_ALU64:
 		if (BPF_SRC(insn->code) != BPF_K)
 			return -EINVAL;
 		if (!failed && validate && insn->imm != orig_val) {
-			pr_warn("prog '%s': unexpected insn #%d value: got %u, exp %u -> %u\n",
+			pr_warn("prog '%s': unexpected insn #%d (ALU/ALU64) value: got %u, exp %u -> %u\n",
 				bpf_program__title(prog, false), insn_idx,
 				insn->imm, orig_val, new_val);
 			return -EINVAL;
@@ -3156,7 +3827,29 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 		pr_debug("prog '%s': patched insn #%d (ALU/ALU64)%s imm %u -> %u\n",
 			 bpf_program__title(prog, false), insn_idx,
 			 failed ? " w/ failed reloc" : "", orig_val, new_val);
-	} else {
+		break;
+	case BPF_LDX:
+	case BPF_ST:
+	case BPF_STX:
+		if (!failed && validate && insn->off != orig_val) {
+			pr_warn("prog '%s': unexpected insn #%d (LD/LDX/ST/STX) value: got %u, exp %u -> %u\n",
+				bpf_program__title(prog, false), insn_idx,
+				insn->off, orig_val, new_val);
+			return -EINVAL;
+		}
+		if (new_val > SHRT_MAX) {
+			pr_warn("prog '%s': insn #%d (LD/LDX/ST/STX) value too big: %u\n",
+				bpf_program__title(prog, false), insn_idx,
+				new_val);
+			return -ERANGE;
+		}
+		orig_val = insn->off;
+		insn->off = new_val;
+		pr_debug("prog '%s': patched insn #%d (LD/LDX/ST/STX)%s off %u -> %u\n",
+			 bpf_program__title(prog, false), insn_idx,
+			 failed ? " w/ failed reloc" : "", orig_val, new_val);
+		break;
+	default:
 		pr_warn("prog '%s': trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n",
 			bpf_program__title(prog, false),
 			insn_idx, insn->code, insn->src_reg, insn->dst_reg,
@@ -3559,9 +4252,6 @@ bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
 	size_t new_cnt;
 	int err;
 
-	if (relo->type != RELO_CALL)
-		return -LIBBPF_ERRNO__RELOC;
-
 	if (prog->idx == obj->efile.text_shndx) {
 		pr_warn("relo in .text insn %d into off %d (insn #%d)\n",
 			relo->insn_idx, relo->sym_off, relo->sym_off / 8);
@@ -3623,27 +4313,37 @@ bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
 
 	for (i = 0; i < prog->nr_reloc; i++) {
 		struct reloc_desc *relo = &prog->reloc_desc[i];
+		struct bpf_insn *insn = &prog->insns[relo->insn_idx];
 
-		if (relo->type == RELO_LD64 || relo->type == RELO_DATA) {
-			struct bpf_insn *insn = &prog->insns[relo->insn_idx];
-
-			if (relo->insn_idx + 1 >= (int)prog->insns_cnt) {
-				pr_warn("relocation out of range: '%s'\n",
-					prog->section_name);
-				return -LIBBPF_ERRNO__RELOC;
-			}
+		if (relo->insn_idx + 1 >= (int)prog->insns_cnt) {
+			pr_warn("relocation out of range: '%s'\n",
+				prog->section_name);
+			return -LIBBPF_ERRNO__RELOC;
+		}
 
-			if (relo->type != RELO_DATA) {
-				insn[0].src_reg = BPF_PSEUDO_MAP_FD;
-			} else {
-				insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
-				insn[1].imm = insn[0].imm + relo->sym_off;
-			}
+		switch (relo->type) {
+		case RELO_LD64:
+			insn[0].src_reg = BPF_PSEUDO_MAP_FD;
+			insn[0].imm = obj->maps[relo->map_idx].fd;
+			break;
+		case RELO_DATA:
+			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+			insn[1].imm = insn[0].imm + relo->sym_off;
 			insn[0].imm = obj->maps[relo->map_idx].fd;
-		} else if (relo->type == RELO_CALL) {
+			break;
+		case RELO_EXTERN:
+			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+			insn[0].imm = obj->maps[obj->kconfig_map_idx].fd;
+			insn[1].imm = relo->sym_off;
+			break;
+		case RELO_CALL:
 			err = bpf_program__reloc_text(prog, obj, relo);
 			if (err)
 				return err;
+			break;
+		default:
+			pr_warn("relo #%d: bad relo type %d\n", i, relo->type);
+			return -EINVAL;
 		}
 	}
 
@@ -3778,6 +4478,7 @@ retry_load:
 	ret = -errno;
 	cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
 	pr_warn("load bpf program failed: %s\n", cp);
+	pr_perm_msg(ret);
 
 	if (log_buf && log_buf[0] != '\0') {
 		ret = -LIBBPF_ERRNO__VERIFY;
@@ -3807,11 +4508,22 @@ out:
 	return ret;
 }
 
-int
-bpf_program__load(struct bpf_program *prog,
-		  char *license, __u32 kern_version)
+static int libbpf_find_attach_btf_id(const char *name,
+				     enum bpf_attach_type attach_type,
+				     __u32 attach_prog_fd);
+
+int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 {
-	int err = 0, fd, i;
+	int err = 0, fd, i, btf_id;
+
+	if (prog->type == BPF_PROG_TYPE_TRACING) {
+		btf_id = libbpf_find_attach_btf_id(prog->section_name,
+						   prog->expected_attach_type,
+						   prog->attach_prog_fd);
+		if (btf_id <= 0)
+			return btf_id;
+		prog->attach_btf_id = btf_id;
+	}
 
 	if (prog->instances.nr < 0 || !prog->instances.fds) {
 		if (prog->preprocessor) {
@@ -3835,7 +4547,7 @@ bpf_program__load(struct bpf_program *prog,
 				prog->section_name, prog->instances.nr);
 		}
 		err = load_program(prog, prog->insns, prog->insns_cnt,
-				   license, kern_version, &fd);
+				   license, kern_ver, &fd);
 		if (!err)
 			prog->instances.fds[0] = fd;
 		goto out;
@@ -3864,9 +4576,7 @@ bpf_program__load(struct bpf_program *prog,
 		}
 
 		err = load_program(prog, result.new_insn_ptr,
-				   result.new_insn_cnt,
-				   license, kern_version, &fd);
-
+				   result.new_insn_cnt, license, kern_ver, &fd);
 		if (err) {
 			pr_warn("Loading the %dth instance of program '%s' failed\n",
 				i, prog->section_name);
@@ -3910,20 +4620,14 @@ bpf_object__load_progs(struct bpf_object *obj, int log_level)
 	return 0;
 }
 
-static int libbpf_find_attach_btf_id(const char *name,
-				     enum bpf_attach_type attach_type,
-				     __u32 attach_prog_fd);
 static struct bpf_object *
 __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
-		   struct bpf_object_open_opts *opts)
+		   const struct bpf_object_open_opts *opts)
 {
-	const char *pin_root_path;
+	const char *obj_name, *kconfig;
 	struct bpf_program *prog;
 	struct bpf_object *obj;
-	const char *obj_name;
 	char tmp_name[64];
-	bool relaxed_maps;
-	__u32 attach_prog_fd;
 	int err;
 
 	if (elf_version(EV_CURRENT) == EV_NONE) {
@@ -3952,16 +4656,23 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 		return obj;
 
 	obj->relaxed_core_relocs = OPTS_GET(opts, relaxed_core_relocs, false);
-	relaxed_maps = OPTS_GET(opts, relaxed_maps, false);
-	pin_root_path = OPTS_GET(opts, pin_root_path, NULL);
-	attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0);
-
-	CHECK_ERR(bpf_object__elf_init(obj), err, out);
-	CHECK_ERR(bpf_object__check_endianness(obj), err, out);
-	CHECK_ERR(bpf_object__probe_caps(obj), err, out);
-	CHECK_ERR(bpf_object__elf_collect(obj, relaxed_maps, pin_root_path),
-		  err, out);
-	CHECK_ERR(bpf_object__collect_reloc(obj), err, out);
+	kconfig = OPTS_GET(opts, kconfig, NULL);
+	if (kconfig) {
+		obj->kconfig = strdup(kconfig);
+		if (!obj->kconfig)
+			return ERR_PTR(-ENOMEM);
+	}
+
+	err = bpf_object__elf_init(obj);
+	err = err ? : bpf_object__check_endianness(obj);
+	err = err ? : bpf_object__elf_collect(obj);
+	err = err ? : bpf_object__collect_externs(obj);
+	err = err ? : bpf_object__finalize_btf(obj);
+	err = err ? : bpf_object__init_maps(obj, opts);
+	err = err ? : bpf_object__init_prog_names(obj);
+	err = err ? : bpf_object__collect_reloc(obj);
+	if (err)
+		goto out;
 	bpf_object__elf_finish(obj);
 
 	bpf_object__for_each_program(prog, obj) {
@@ -3978,15 +4689,8 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 
 		bpf_program__set_type(prog, prog_type);
 		bpf_program__set_expected_attach_type(prog, attach_type);
-		if (prog_type == BPF_PROG_TYPE_TRACING) {
-			err = libbpf_find_attach_btf_id(prog->section_name,
-							attach_type,
-							attach_prog_fd);
-			if (err <= 0)
-				goto out;
-			prog->attach_btf_id = err;
-			prog->attach_prog_fd = attach_prog_fd;
-		}
+		if (prog_type == BPF_PROG_TYPE_TRACING)
+			prog->attach_prog_fd = OPTS_GET(opts, attach_prog_fd, 0);
 	}
 
 	return obj;
@@ -4026,7 +4730,7 @@ struct bpf_object *bpf_object__open(const char *path)
 }
 
 struct bpf_object *
-bpf_object__open_file(const char *path, struct bpf_object_open_opts *opts)
+bpf_object__open_file(const char *path, const struct bpf_object_open_opts *opts)
 {
 	if (!path)
 		return ERR_PTR(-EINVAL);
@@ -4038,7 +4742,7 @@ bpf_object__open_file(const char *path, struct bpf_object_open_opts *opts)
 
 struct bpf_object *
 bpf_object__open_mem(const void *obj_buf, size_t obj_buf_sz,
-		     struct bpf_object_open_opts *opts)
+		     const struct bpf_object_open_opts *opts)
 {
 	if (!obj_buf || obj_buf_sz == 0)
 		return ERR_PTR(-EINVAL);
@@ -4079,6 +4783,92 @@ int bpf_object__unload(struct bpf_object *obj)
 	return 0;
 }
 
+static int bpf_object__sanitize_maps(struct bpf_object *obj)
+{
+	struct bpf_map *m;
+
+	bpf_object__for_each_map(m, obj) {
+		if (!bpf_map__is_internal(m))
+			continue;
+		if (!obj->caps.global_data) {
+			pr_warn("kernel doesn't support global data\n");
+			return -ENOTSUP;
+		}
+		if (!obj->caps.array_mmap)
+			m->def.map_flags ^= BPF_F_MMAPABLE;
+	}
+
+	return 0;
+}
+
+static int bpf_object__resolve_externs(struct bpf_object *obj,
+				       const char *extra_kconfig)
+{
+	bool need_config = false;
+	struct extern_desc *ext;
+	int err, i;
+	void *data;
+
+	if (obj->nr_extern == 0)
+		return 0;
+
+	data = obj->maps[obj->kconfig_map_idx].mmaped;
+
+	for (i = 0; i < obj->nr_extern; i++) {
+		ext = &obj->externs[i];
+
+		if (strcmp(ext->name, "LINUX_KERNEL_VERSION") == 0) {
+			void *ext_val = data + ext->data_off;
+			__u32 kver = get_kernel_version();
+
+			if (!kver) {
+				pr_warn("failed to get kernel version\n");
+				return -EINVAL;
+			}
+			err = set_ext_value_num(ext, ext_val, kver);
+			if (err)
+				return err;
+			pr_debug("extern %s=0x%x\n", ext->name, kver);
+		} else if (strncmp(ext->name, "CONFIG_", 7) == 0) {
+			need_config = true;
+		} else {
+			pr_warn("unrecognized extern '%s'\n", ext->name);
+			return -EINVAL;
+		}
+	}
+	if (need_config && extra_kconfig) {
+		err = bpf_object__read_kconfig_mem(obj, extra_kconfig, data);
+		if (err)
+			return -EINVAL;
+		need_config = false;
+		for (i = 0; i < obj->nr_extern; i++) {
+			ext = &obj->externs[i];
+			if (!ext->is_set) {
+				need_config = true;
+				break;
+			}
+		}
+	}
+	if (need_config) {
+		err = bpf_object__read_kconfig_file(obj, data);
+		if (err)
+			return -EINVAL;
+	}
+	for (i = 0; i < obj->nr_extern; i++) {
+		ext = &obj->externs[i];
+
+		if (!ext->is_set && !ext->is_weak) {
+			pr_warn("extern %s (strong) not resolved\n", ext->name);
+			return -ESRCH;
+		} else if (!ext->is_set) {
+			pr_debug("extern %s (weak) not resolved, defaulting to zero\n",
+				 ext->name);
+		}
+	}
+
+	return 0;
+}
+
 int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 {
 	struct bpf_object *obj;
@@ -4097,9 +4887,15 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 
 	obj->loaded = true;
 
-	CHECK_ERR(bpf_object__create_maps(obj), err, out);
-	CHECK_ERR(bpf_object__relocate(obj, attr->target_btf_path), err, out);
-	CHECK_ERR(bpf_object__load_progs(obj, attr->log_level), err, out);
+	err = bpf_object__probe_caps(obj);
+	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig);
+	err = err ? : bpf_object__sanitize_and_load_btf(obj);
+	err = err ? : bpf_object__sanitize_maps(obj);
+	err = err ? : bpf_object__create_maps(obj);
+	err = err ? : bpf_object__relocate(obj, attr->target_btf_path);
+	err = err ? : bpf_object__load_progs(obj, attr->log_level);
+	if (err)
+		goto out;
 
 	return 0;
 out:
@@ -4670,17 +5466,26 @@ void bpf_object__close(struct bpf_object *obj)
 	btf_ext__free(obj->btf_ext);
 
 	for (i = 0; i < obj->nr_maps; i++) {
-		zfree(&obj->maps[i].name);
-		zfree(&obj->maps[i].pin_path);
-		if (obj->maps[i].clear_priv)
-			obj->maps[i].clear_priv(&obj->maps[i],
-						obj->maps[i].priv);
-		obj->maps[i].priv = NULL;
-		obj->maps[i].clear_priv = NULL;
+		struct bpf_map *map = &obj->maps[i];
+
+		if (map->clear_priv)
+			map->clear_priv(map, map->priv);
+		map->priv = NULL;
+		map->clear_priv = NULL;
+
+		if (map->mmaped) {
+			munmap(map->mmaped, bpf_map_mmap_sz(map));
+			map->mmaped = NULL;
+		}
+
+		zfree(&map->name);
+		zfree(&map->pin_path);
 	}
 
-	zfree(&obj->sections.rodata);
-	zfree(&obj->sections.data);
+	zfree(&obj->kconfig);
+	zfree(&obj->externs);
+	obj->nr_extern = 0;
+
 	zfree(&obj->maps);
 	obj->nr_maps = 0;
 
@@ -4820,6 +5625,11 @@ void bpf_program__set_ifindex(struct bpf_program *prog, __u32 ifindex)
 	prog->prog_ifindex = ifindex;
 }
 
+const char *bpf_program__name(const struct bpf_program *prog)
+{
+	return prog->name;
+}
+
 const char *bpf_program__title(const struct bpf_program *prog, bool needs_copy)
 {
 	const char *title;
@@ -4972,7 +5782,28 @@ void bpf_program__set_expected_attach_type(struct bpf_program *prog,
  */
 #define BPF_APROG_COMPAT(string, ptype) BPF_PROG_SEC(string, ptype)
 
-static const struct {
+#define SEC_DEF(sec_pfx, ptype, ...) {					    \
+	.sec = sec_pfx,							    \
+	.len = sizeof(sec_pfx) - 1,					    \
+	.prog_type = BPF_PROG_TYPE_##ptype,				    \
+	__VA_ARGS__							    \
+}
+
+struct bpf_sec_def;
+
+typedef struct bpf_link *(*attach_fn_t)(const struct bpf_sec_def *sec,
+					struct bpf_program *prog);
+
+static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog);
+static struct bpf_link *attach_tp(const struct bpf_sec_def *sec,
+				  struct bpf_program *prog);
+static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog);
+static struct bpf_link *attach_trace(const struct bpf_sec_def *sec,
+				     struct bpf_program *prog);
+
+struct bpf_sec_def {
 	const char *sec;
 	size_t len;
 	enum bpf_prog_type prog_type;
@@ -4980,24 +5811,40 @@ static const struct {
 	bool is_attachable;
 	bool is_attach_btf;
 	enum bpf_attach_type attach_type;
-} section_names[] = {
+	attach_fn_t attach_fn;
+};
+
+static const struct bpf_sec_def section_defs[] = {
 	BPF_PROG_SEC("socket",			BPF_PROG_TYPE_SOCKET_FILTER),
-	BPF_PROG_SEC("kprobe/",			BPF_PROG_TYPE_KPROBE),
+	BPF_PROG_SEC("sk_reuseport",		BPF_PROG_TYPE_SK_REUSEPORT),
+	SEC_DEF("kprobe/", KPROBE,
+		.attach_fn = attach_kprobe),
 	BPF_PROG_SEC("uprobe/",			BPF_PROG_TYPE_KPROBE),
-	BPF_PROG_SEC("kretprobe/",		BPF_PROG_TYPE_KPROBE),
+	SEC_DEF("kretprobe/", KPROBE,
+		.attach_fn = attach_kprobe),
 	BPF_PROG_SEC("uretprobe/",		BPF_PROG_TYPE_KPROBE),
 	BPF_PROG_SEC("classifier",		BPF_PROG_TYPE_SCHED_CLS),
 	BPF_PROG_SEC("action",			BPF_PROG_TYPE_SCHED_ACT),
-	BPF_PROG_SEC("tracepoint/",		BPF_PROG_TYPE_TRACEPOINT),
-	BPF_PROG_SEC("tp/",			BPF_PROG_TYPE_TRACEPOINT),
-	BPF_PROG_SEC("raw_tracepoint/",		BPF_PROG_TYPE_RAW_TRACEPOINT),
-	BPF_PROG_SEC("raw_tp/",			BPF_PROG_TYPE_RAW_TRACEPOINT),
-	BPF_PROG_BTF("tp_btf/",			BPF_PROG_TYPE_TRACING,
-						BPF_TRACE_RAW_TP),
-	BPF_PROG_BTF("fentry/",			BPF_PROG_TYPE_TRACING,
-						BPF_TRACE_FENTRY),
-	BPF_PROG_BTF("fexit/",			BPF_PROG_TYPE_TRACING,
-						BPF_TRACE_FEXIT),
+	SEC_DEF("tracepoint/", TRACEPOINT,
+		.attach_fn = attach_tp),
+	SEC_DEF("tp/", TRACEPOINT,
+		.attach_fn = attach_tp),
+	SEC_DEF("raw_tracepoint/", RAW_TRACEPOINT,
+		.attach_fn = attach_raw_tp),
+	SEC_DEF("raw_tp/", RAW_TRACEPOINT,
+		.attach_fn = attach_raw_tp),
+	SEC_DEF("tp_btf/", TRACING,
+		.expected_attach_type = BPF_TRACE_RAW_TP,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
+	SEC_DEF("fentry/", TRACING,
+		.expected_attach_type = BPF_TRACE_FENTRY,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
+	SEC_DEF("fexit/", TRACING,
+		.expected_attach_type = BPF_TRACE_FEXIT,
+		.is_attach_btf = true,
+		.attach_fn = attach_trace),
 	BPF_PROG_SEC("xdp",			BPF_PROG_TYPE_XDP),
 	BPF_PROG_SEC("perf_event",		BPF_PROG_TYPE_PERF_EVENT),
 	BPF_PROG_SEC("lwt_in",			BPF_PROG_TYPE_LWT_IN),
@@ -5059,12 +5906,26 @@ static const struct {
 #undef BPF_APROG_SEC
 #undef BPF_EAPROG_SEC
 #undef BPF_APROG_COMPAT
+#undef SEC_DEF
 
 #define MAX_TYPE_NAME_SIZE 32
 
+static const struct bpf_sec_def *find_sec_def(const char *sec_name)
+{
+	int i, n = ARRAY_SIZE(section_defs);
+
+	for (i = 0; i < n; i++) {
+		if (strncmp(sec_name,
+			    section_defs[i].sec, section_defs[i].len))
+			continue;
+		return &section_defs[i];
+	}
+	return NULL;
+}
+
 static char *libbpf_get_type_names(bool attach_type)
 {
-	int i, len = ARRAY_SIZE(section_names) * MAX_TYPE_NAME_SIZE;
+	int i, len = ARRAY_SIZE(section_defs) * MAX_TYPE_NAME_SIZE;
 	char *buf;
 
 	buf = malloc(len);
@@ -5073,16 +5934,16 @@ static char *libbpf_get_type_names(bool attach_type)
 
 	buf[0] = '\0';
 	/* Forge string buf with all available names */
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (attach_type && !section_names[i].is_attachable)
+	for (i = 0; i < ARRAY_SIZE(section_defs); i++) {
+		if (attach_type && !section_defs[i].is_attachable)
 			continue;
 
-		if (strlen(buf) + strlen(section_names[i].sec) + 2 > len) {
+		if (strlen(buf) + strlen(section_defs[i].sec) + 2 > len) {
 			free(buf);
 			return NULL;
 		}
 		strcat(buf, " ");
-		strcat(buf, section_names[i].sec);
+		strcat(buf, section_defs[i].sec);
 	}
 
 	return buf;
@@ -5091,23 +5952,23 @@ static char *libbpf_get_type_names(bool attach_type)
 int libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 			     enum bpf_attach_type *expected_attach_type)
 {
+	const struct bpf_sec_def *sec_def;
 	char *type_names;
-	int i;
 
 	if (!name)
 		return -EINVAL;
 
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (strncmp(name, section_names[i].sec, section_names[i].len))
-			continue;
-		*prog_type = section_names[i].prog_type;
-		*expected_attach_type = section_names[i].expected_attach_type;
+	sec_def = find_sec_def(name);
+	if (sec_def) {
+		*prog_type = sec_def->prog_type;
+		*expected_attach_type = sec_def->expected_attach_type;
 		return 0;
 	}
-	pr_warn("failed to guess program type from ELF section '%s'\n", name);
+
+	pr_debug("failed to guess program type from ELF section '%s'\n", name);
 	type_names = libbpf_get_type_names(false);
 	if (type_names != NULL) {
-		pr_info("supported section(type) names are:%s\n", type_names);
+		pr_debug("supported section(type) names are:%s\n", type_names);
 		free(type_names);
 	}
 
@@ -5186,16 +6047,16 @@ static int libbpf_find_attach_btf_id(const char *name,
 	if (!name)
 		return -EINVAL;
 
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (!section_names[i].is_attach_btf)
+	for (i = 0; i < ARRAY_SIZE(section_defs); i++) {
+		if (!section_defs[i].is_attach_btf)
 			continue;
-		if (strncmp(name, section_names[i].sec, section_names[i].len))
+		if (strncmp(name, section_defs[i].sec, section_defs[i].len))
 			continue;
 		if (attach_prog_fd)
-			err = libbpf_find_prog_btf_id(name + section_names[i].len,
+			err = libbpf_find_prog_btf_id(name + section_defs[i].len,
 						      attach_prog_fd);
 		else
-			err = libbpf_find_vmlinux_btf_id(name + section_names[i].len,
+			err = libbpf_find_vmlinux_btf_id(name + section_defs[i].len,
 							 attach_type);
 		if (err <= 0)
 			pr_warn("%s is not found in vmlinux BTF\n", name);
@@ -5214,18 +6075,18 @@ int libbpf_attach_type_by_name(const char *name,
 	if (!name)
 		return -EINVAL;
 
-	for (i = 0; i < ARRAY_SIZE(section_names); i++) {
-		if (strncmp(name, section_names[i].sec, section_names[i].len))
+	for (i = 0; i < ARRAY_SIZE(section_defs); i++) {
+		if (strncmp(name, section_defs[i].sec, section_defs[i].len))
 			continue;
-		if (!section_names[i].is_attachable)
+		if (!section_defs[i].is_attachable)
 			return -EINVAL;
-		*attach_type = section_names[i].attach_type;
+		*attach_type = section_defs[i].attach_type;
 		return 0;
 	}
-	pr_warn("failed to guess attach type based on ELF section name '%s'\n", name);
+	pr_debug("failed to guess attach type based on ELF section name '%s'\n", name);
 	type_names = libbpf_get_type_names(true);
 	if (type_names != NULL) {
-		pr_info("attachable section(type) names are:%s\n", type_names);
+		pr_debug("attachable section(type) names are:%s\n", type_names);
 		free(type_names);
 	}
 
@@ -5466,17 +6327,37 @@ int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
 }
 
 struct bpf_link {
+	int (*detach)(struct bpf_link *link);
 	int (*destroy)(struct bpf_link *link);
+	bool disconnected;
 };
 
+/* Release "ownership" of underlying BPF resource (typically, BPF program
+ * attached to some BPF hook, e.g., tracepoint, kprobe, etc). Disconnected
+ * link, when destructed through bpf_link__destroy() call won't attempt to
+ * detach/unregisted that BPF resource. This is useful in situations where,
+ * say, attached BPF program has to outlive userspace program that attached it
+ * in the system. Depending on type of BPF program, though, there might be
+ * additional steps (like pinning BPF program in BPF FS) necessary to ensure
+ * exit of userspace program doesn't trigger automatic detachment and clean up
+ * inside the kernel.
+ */
+void bpf_link__disconnect(struct bpf_link *link)
+{
+	link->disconnected = true;
+}
+
 int bpf_link__destroy(struct bpf_link *link)
 {
-	int err;
+	int err = 0;
 
 	if (!link)
 		return 0;
 
-	err = link->destroy(link);
+	if (!link->disconnected && link->detach)
+		err = link->detach(link);
+	if (link->destroy)
+		link->destroy(link);
 	free(link);
 
 	return err;
@@ -5487,7 +6368,7 @@ struct bpf_link_fd {
 	int fd; /* hook FD */
 };
 
-static int bpf_link__destroy_perf_event(struct bpf_link *link)
+static int bpf_link__detach_perf_event(struct bpf_link *link)
 {
 	struct bpf_link_fd *l = (void *)link;
 	int err;
@@ -5519,10 +6400,10 @@ struct bpf_link *bpf_program__attach_perf_event(struct bpf_program *prog,
 		return ERR_PTR(-EINVAL);
 	}
 
-	link = malloc(sizeof(*link));
+	link = calloc(1, sizeof(*link));
 	if (!link)
 		return ERR_PTR(-ENOMEM);
-	link->link.destroy = &bpf_link__destroy_perf_event;
+	link->link.detach = &bpf_link__detach_perf_event;
 	link->fd = pfd;
 
 	if (ioctl(pfd, PERF_EVENT_IOC_SET_BPF, prog_fd) < 0) {
@@ -5679,6 +6560,18 @@ struct bpf_link *bpf_program__attach_kprobe(struct bpf_program *prog,
 	return link;
 }
 
+static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog)
+{
+	const char *func_name;
+	bool retprobe;
+
+	func_name = bpf_program__title(prog, false) + sec->len;
+	retprobe = strcmp(sec->sec, "kretprobe/") == 0;
+
+	return bpf_program__attach_kprobe(prog, retprobe, func_name);
+}
+
 struct bpf_link *bpf_program__attach_uprobe(struct bpf_program *prog,
 					    bool retprobe, pid_t pid,
 					    const char *binary_path,
@@ -5791,7 +6684,33 @@ struct bpf_link *bpf_program__attach_tracepoint(struct bpf_program *prog,
 	return link;
 }
 
-static int bpf_link__destroy_fd(struct bpf_link *link)
+static struct bpf_link *attach_tp(const struct bpf_sec_def *sec,
+				  struct bpf_program *prog)
+{
+	char *sec_name, *tp_cat, *tp_name;
+	struct bpf_link *link;
+
+	sec_name = strdup(bpf_program__title(prog, false));
+	if (!sec_name)
+		return ERR_PTR(-ENOMEM);
+
+	/* extract "tp/<category>/<name>" */
+	tp_cat = sec_name + sec->len;
+	tp_name = strchr(tp_cat, '/');
+	if (!tp_name) {
+		link = ERR_PTR(-EINVAL);
+		goto out;
+	}
+	*tp_name = '\0';
+	tp_name++;
+
+	link = bpf_program__attach_tracepoint(prog, tp_cat, tp_name);
+out:
+	free(sec_name);
+	return link;
+}
+
+static int bpf_link__detach_fd(struct bpf_link *link)
 {
 	struct bpf_link_fd *l = (void *)link;
 
@@ -5812,10 +6731,10 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 		return ERR_PTR(-EINVAL);
 	}
 
-	link = malloc(sizeof(*link));
+	link = calloc(1, sizeof(*link));
 	if (!link)
 		return ERR_PTR(-ENOMEM);
-	link->link.destroy = &bpf_link__destroy_fd;
+	link->link.detach = &bpf_link__detach_fd;
 
 	pfd = bpf_raw_tracepoint_open(tp_name, prog_fd);
 	if (pfd < 0) {
@@ -5830,6 +6749,14 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 	return (struct bpf_link *)link;
 }
 
+static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec,
+				      struct bpf_program *prog)
+{
+	const char *tp_name = bpf_program__title(prog, false) + sec->len;
+
+	return bpf_program__attach_raw_tracepoint(prog, tp_name);
+}
+
 struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 {
 	char errmsg[STRERR_BUFSIZE];
@@ -5843,10 +6770,10 @@ struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 		return ERR_PTR(-EINVAL);
 	}
 
-	link = malloc(sizeof(*link));
+	link = calloc(1, sizeof(*link));
 	if (!link)
 		return ERR_PTR(-ENOMEM);
-	link->link.destroy = &bpf_link__destroy_fd;
+	link->link.detach = &bpf_link__detach_fd;
 
 	pfd = bpf_raw_tracepoint_open(NULL, prog_fd);
 	if (pfd < 0) {
@@ -5861,6 +6788,23 @@ struct bpf_link *bpf_program__attach_trace(struct bpf_program *prog)
 	return (struct bpf_link *)link;
 }
 
+static struct bpf_link *attach_trace(const struct bpf_sec_def *sec,
+				     struct bpf_program *prog)
+{
+	return bpf_program__attach_trace(prog);
+}
+
+struct bpf_link *bpf_program__attach(struct bpf_program *prog)
+{
+	const struct bpf_sec_def *sec_def;
+
+	sec_def = find_sec_def(bpf_program__title(prog, false));
+	if (!sec_def || !sec_def->attach_fn)
+		return ERR_PTR(-ESRCH);
+
+	return sec_def->attach_fn(sec_def, prog);
+}
+
 enum bpf_perf_event_ret
 bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
@@ -5944,7 +6888,7 @@ struct perf_buffer {
 	size_t mmap_size;
 	struct perf_cpu_buf **cpu_bufs;
 	struct epoll_event *events;
-	int cpu_cnt;
+	int cpu_cnt; /* number of allocated CPU buffers */
 	int epoll_fd; /* perf event FD */
 	int map_fd; /* BPF_MAP_TYPE_PERF_EVENT_ARRAY BPF map FD */
 };
@@ -6078,11 +7022,13 @@ perf_buffer__new_raw(int map_fd, size_t page_cnt,
 static struct perf_buffer *__perf_buffer__new(int map_fd, size_t page_cnt,
 					      struct perf_buffer_params *p)
 {
+	const char *online_cpus_file = "/sys/devices/system/cpu/online";
 	struct bpf_map_info map = {};
 	char msg[STRERR_BUFSIZE];
 	struct perf_buffer *pb;
+	bool *online = NULL;
 	__u32 map_info_len;
-	int err, i;
+	int err, i, j, n;
 
 	if (page_cnt & (page_cnt - 1)) {
 		pr_warn("page count should be power of two, but is %zu\n",
@@ -6151,20 +7097,32 @@ static struct perf_buffer *__perf_buffer__new(int map_fd, size_t page_cnt,
 		goto error;
 	}
 
-	for (i = 0; i < pb->cpu_cnt; i++) {
+	err = parse_cpu_mask_file(online_cpus_file, &online, &n);
+	if (err) {
+		pr_warn("failed to get online CPU mask: %d\n", err);
+		goto error;
+	}
+
+	for (i = 0, j = 0; i < pb->cpu_cnt; i++) {
 		struct perf_cpu_buf *cpu_buf;
 		int cpu, map_key;
 
 		cpu = p->cpu_cnt > 0 ? p->cpus[i] : i;
 		map_key = p->cpu_cnt > 0 ? p->map_keys[i] : i;
 
+		/* in case user didn't explicitly requested particular CPUs to
+		 * be attached to, skip offline/not present CPUs
+		 */
+		if (p->cpu_cnt <= 0 && (cpu >= n || !online[cpu]))
+			continue;
+
 		cpu_buf = perf_buffer__open_cpu_buf(pb, p->attr, cpu, map_key);
 		if (IS_ERR(cpu_buf)) {
 			err = PTR_ERR(cpu_buf);
 			goto error;
 		}
 
-		pb->cpu_bufs[i] = cpu_buf;
+		pb->cpu_bufs[j] = cpu_buf;
 
 		err = bpf_map_update_elem(pb->map_fd, &map_key,
 					  &cpu_buf->fd, 0);
@@ -6176,21 +7134,25 @@ static struct perf_buffer *__perf_buffer__new(int map_fd, size_t page_cnt,
 			goto error;
 		}
 
-		pb->events[i].events = EPOLLIN;
-		pb->events[i].data.ptr = cpu_buf;
+		pb->events[j].events = EPOLLIN;
+		pb->events[j].data.ptr = cpu_buf;
 		if (epoll_ctl(pb->epoll_fd, EPOLL_CTL_ADD, cpu_buf->fd,
-			      &pb->events[i]) < 0) {
+			      &pb->events[j]) < 0) {
 			err = -errno;
 			pr_warn("failed to epoll_ctl cpu #%d perf FD %d: %s\n",
 				cpu, cpu_buf->fd,
 				libbpf_strerror_r(err, msg, sizeof(msg)));
 			goto error;
 		}
+		j++;
 	}
+	pb->cpu_cnt = j;
+	free(online);
 
 	return pb;
 
 error:
+	free(online);
 	if (pb)
 		perf_buffer__free(pb);
 	return ERR_PTR(err);
@@ -6521,62 +7483,267 @@ void bpf_program__bpil_offs_to_addr(struct bpf_prog_info_linear *info_linear)
 	}
 }
 
-int libbpf_num_possible_cpus(void)
+int parse_cpu_mask_str(const char *s, bool **mask, int *mask_sz)
 {
-	static const char *fcpu = "/sys/devices/system/cpu/possible";
-	int len = 0, n = 0, il = 0, ir = 0;
-	unsigned int start = 0, end = 0;
-	int tmp_cpus = 0;
-	static int cpus;
-	char buf[128];
-	int error = 0;
-	int fd = -1;
+	int err = 0, n, len, start, end = -1;
+	bool *tmp;
 
-	tmp_cpus = READ_ONCE(cpus);
-	if (tmp_cpus > 0)
-		return tmp_cpus;
+	*mask = NULL;
+	*mask_sz = 0;
+
+	/* Each sub string separated by ',' has format \d+-\d+ or \d+ */
+	while (*s) {
+		if (*s == ',' || *s == '\n') {
+			s++;
+			continue;
+		}
+		n = sscanf(s, "%d%n-%d%n", &start, &len, &end, &len);
+		if (n <= 0 || n > 2) {
+			pr_warn("Failed to get CPU range %s: %d\n", s, n);
+			err = -EINVAL;
+			goto cleanup;
+		} else if (n == 1) {
+			end = start;
+		}
+		if (start < 0 || start > end) {
+			pr_warn("Invalid CPU range [%d,%d] in %s\n",
+				start, end, s);
+			err = -EINVAL;
+			goto cleanup;
+		}
+		tmp = realloc(*mask, end + 1);
+		if (!tmp) {
+			err = -ENOMEM;
+			goto cleanup;
+		}
+		*mask = tmp;
+		memset(tmp + *mask_sz, 0, start - *mask_sz);
+		memset(tmp + start, 1, end - start + 1);
+		*mask_sz = end + 1;
+		s += len;
+	}
+	if (!*mask_sz) {
+		pr_warn("Empty CPU range\n");
+		return -EINVAL;
+	}
+	return 0;
+cleanup:
+	free(*mask);
+	*mask = NULL;
+	return err;
+}
+
+int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz)
+{
+	int fd, err = 0, len;
+	char buf[128];
 
 	fd = open(fcpu, O_RDONLY);
 	if (fd < 0) {
-		error = errno;
-		pr_warn("Failed to open file %s: %s\n", fcpu, strerror(error));
-		return -error;
+		err = -errno;
+		pr_warn("Failed to open cpu mask file %s: %d\n", fcpu, err);
+		return err;
 	}
 	len = read(fd, buf, sizeof(buf));
 	close(fd);
 	if (len <= 0) {
-		error = len ? errno : EINVAL;
-		pr_warn("Failed to read # of possible cpus from %s: %s\n",
-			fcpu, strerror(error));
-		return -error;
+		err = len ? -errno : -EINVAL;
+		pr_warn("Failed to read cpu mask from %s: %d\n", fcpu, err);
+		return err;
 	}
-	if (len == sizeof(buf)) {
-		pr_warn("File %s size overflow\n", fcpu);
-		return -EOVERFLOW;
+	if (len >= sizeof(buf)) {
+		pr_warn("CPU mask is too big in file %s\n", fcpu);
+		return -E2BIG;
 	}
 	buf[len] = '\0';
 
-	for (ir = 0, tmp_cpus = 0; ir <= len; ir++) {
-		/* Each sub string separated by ',' has format \d+-\d+ or \d+ */
-		if (buf[ir] == ',' || buf[ir] == '\0') {
-			buf[ir] = '\0';
-			n = sscanf(&buf[il], "%u-%u", &start, &end);
-			if (n <= 0) {
-				pr_warn("Failed to get # CPUs from %s\n",
-					&buf[il]);
-				return -EINVAL;
-			} else if (n == 1) {
-				end = start;
-			}
-			tmp_cpus += end - start + 1;
-			il = ir + 1;
-		}
-	}
-	if (tmp_cpus <= 0) {
-		pr_warn("Invalid #CPUs %d from %s\n", tmp_cpus, fcpu);
-		return -EINVAL;
+	return parse_cpu_mask_str(buf, mask, mask_sz);
+}
+
+int libbpf_num_possible_cpus(void)
+{
+	static const char *fcpu = "/sys/devices/system/cpu/possible";
+	static int cpus;
+	int err, n, i, tmp_cpus;
+	bool *mask;
+
+	tmp_cpus = READ_ONCE(cpus);
+	if (tmp_cpus > 0)
+		return tmp_cpus;
+
+	err = parse_cpu_mask_file(fcpu, &mask, &n);
+	if (err)
+		return err;
+
+	tmp_cpus = 0;
+	for (i = 0; i < n; i++) {
+		if (mask[i])
+			tmp_cpus++;
 	}
+	free(mask);
 
 	WRITE_ONCE(cpus, tmp_cpus);
 	return tmp_cpus;
 }
+
+int bpf_object__open_skeleton(struct bpf_object_skeleton *s,
+			      const struct bpf_object_open_opts *opts)
+{
+	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, skel_opts,
+		.object_name = s->name,
+	);
+	struct bpf_object *obj;
+	int i;
+
+	/* Attempt to preserve opts->object_name, unless overriden by user
+	 * explicitly. Overwriting object name for skeletons is discouraged,
+	 * as it breaks global data maps, because they contain object name
+	 * prefix as their own map name prefix. When skeleton is generated,
+	 * bpftool is making an assumption that this name will stay the same.
+	 */
+	if (opts) {
+		memcpy(&skel_opts, opts, sizeof(*opts));
+		if (!opts->object_name)
+			skel_opts.object_name = s->name;
+	}
+
+	obj = bpf_object__open_mem(s->data, s->data_sz, &skel_opts);
+	if (IS_ERR(obj)) {
+		pr_warn("failed to initialize skeleton BPF object '%s': %ld\n",
+			s->name, PTR_ERR(obj));
+		return PTR_ERR(obj);
+	}
+
+	*s->obj = obj;
+
+	for (i = 0; i < s->map_cnt; i++) {
+		struct bpf_map **map = s->maps[i].map;
+		const char *name = s->maps[i].name;
+		void **mmaped = s->maps[i].mmaped;
+
+		*map = bpf_object__find_map_by_name(obj, name);
+		if (!*map) {
+			pr_warn("failed to find skeleton map '%s'\n", name);
+			return -ESRCH;
+		}
+
+		/* externs shouldn't be pre-setup from user code */
+		if (mmaped && (*map)->libbpf_type != LIBBPF_MAP_KCONFIG)
+			*mmaped = (*map)->mmaped;
+	}
+
+	for (i = 0; i < s->prog_cnt; i++) {
+		struct bpf_program **prog = s->progs[i].prog;
+		const char *name = s->progs[i].name;
+
+		*prog = bpf_object__find_program_by_name(obj, name);
+		if (!*prog) {
+			pr_warn("failed to find skeleton program '%s'\n", name);
+			return -ESRCH;
+		}
+	}
+
+	return 0;
+}
+
+int bpf_object__load_skeleton(struct bpf_object_skeleton *s)
+{
+	int i, err;
+
+	err = bpf_object__load(*s->obj);
+	if (err) {
+		pr_warn("failed to load BPF skeleton '%s': %d\n", s->name, err);
+		return err;
+	}
+
+	for (i = 0; i < s->map_cnt; i++) {
+		struct bpf_map *map = *s->maps[i].map;
+		size_t mmap_sz = bpf_map_mmap_sz(map);
+		int prot, map_fd = bpf_map__fd(map);
+		void **mmaped = s->maps[i].mmaped;
+
+		if (!mmaped)
+			continue;
+
+		if (!(map->def.map_flags & BPF_F_MMAPABLE)) {
+			*mmaped = NULL;
+			continue;
+		}
+
+		if (map->def.map_flags & BPF_F_RDONLY_PROG)
+			prot = PROT_READ;
+		else
+			prot = PROT_READ | PROT_WRITE;
+
+		/* Remap anonymous mmap()-ed "map initialization image" as
+		 * a BPF map-backed mmap()-ed memory, but preserving the same
+		 * memory address. This will cause kernel to change process'
+		 * page table to point to a different piece of kernel memory,
+		 * but from userspace point of view memory address (and its
+		 * contents, being identical at this point) will stay the
+		 * same. This mapping will be released by bpf_object__close()
+		 * as per normal clean up procedure, so we don't need to worry
+		 * about it from skeleton's clean up perspective.
+		 */
+		*mmaped = mmap(map->mmaped, mmap_sz, prot,
+				MAP_SHARED | MAP_FIXED, map_fd, 0);
+		if (*mmaped == MAP_FAILED) {
+			err = -errno;
+			*mmaped = NULL;
+			pr_warn("failed to re-mmap() map '%s': %d\n",
+				 bpf_map__name(map), err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
+{
+	int i;
+
+	for (i = 0; i < s->prog_cnt; i++) {
+		struct bpf_program *prog = *s->progs[i].prog;
+		struct bpf_link **link = s->progs[i].link;
+		const struct bpf_sec_def *sec_def;
+		const char *sec_name = bpf_program__title(prog, false);
+
+		sec_def = find_sec_def(sec_name);
+		if (!sec_def || !sec_def->attach_fn)
+			continue;
+
+		*link = sec_def->attach_fn(sec_def, prog);
+		if (IS_ERR(*link)) {
+			pr_warn("failed to auto-attach program '%s': %ld\n",
+				bpf_program__name(prog), PTR_ERR(*link));
+			return PTR_ERR(*link);
+		}
+	}
+
+	return 0;
+}
+
+void bpf_object__detach_skeleton(struct bpf_object_skeleton *s)
+{
+	int i;
+
+	for (i = 0; i < s->prog_cnt; i++) {
+		struct bpf_link **link = s->progs[i].link;
+
+		if (!IS_ERR_OR_NULL(*link))
+			bpf_link__destroy(*link);
+		*link = NULL;
+	}
+}
+
+void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s)
+{
+	if (s->progs)
+		bpf_object__detach_skeleton(s);
+	if (s->obj)
+		bpf_object__close(*s->obj);
+	free(s->maps);
+	free(s->progs);
+	free(s);
+}
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 0dbf4bfba0c4..fe592ef48f1b 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -17,14 +17,12 @@
 #include <sys/types.h>  // for size_t
 #include <linux/bpf.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 enum libbpf_errno {
 	__LIBBPF_ERRNO__START = 4000,
 
@@ -67,28 +65,6 @@ struct bpf_object_open_attr {
 	enum bpf_prog_type prog_type;
 };
 
-/* Helper macro to declare and initialize libbpf options struct
- *
- * This dance with uninitialized declaration, followed by memset to zero,
- * followed by assignment using compound literal syntax is done to preserve
- * ability to use a nice struct field initialization syntax and **hopefully**
- * have all the padding bytes initialized to zero. It's not guaranteed though,
- * when copying literal, that compiler won't copy garbage in literal's padding
- * bytes, but that's the best way I've found and it seems to work in practice.
- *
- * Macro declares opts struct of given type and name, zero-initializes,
- * including any extra padding, it with memset() and then assigns initial
- * values provided by users in struct initializer-syntax as varargs.
- */
-#define DECLARE_LIBBPF_OPTS(TYPE, NAME, ...)				    \
-	struct TYPE NAME = ({ 						    \
-		memset(&NAME, 0, sizeof(struct TYPE));			    \
-		(struct TYPE) {						    \
-			.sz = sizeof(struct TYPE),			    \
-			__VA_ARGS__					    \
-		};							    \
-	})
-
 struct bpf_object_open_opts {
 	/* size of this struct, for forward/backward compatiblity */
 	size_t sz;
@@ -109,15 +85,19 @@ struct bpf_object_open_opts {
 	 */
 	const char *pin_root_path;
 	__u32 attach_prog_fd;
+	/* Additional kernel config content that augments and overrides
+	 * system Kconfig for CONFIG_xxx externs.
+	 */
+	const char *kconfig;
 };
-#define bpf_object_open_opts__last_field attach_prog_fd
+#define bpf_object_open_opts__last_field kconfig
 
 LIBBPF_API struct bpf_object *bpf_object__open(const char *path);
 LIBBPF_API struct bpf_object *
-bpf_object__open_file(const char *path, struct bpf_object_open_opts *opts);
+bpf_object__open_file(const char *path, const struct bpf_object_open_opts *opts);
 LIBBPF_API struct bpf_object *
 bpf_object__open_mem(const void *obj_buf, size_t obj_buf_sz,
-		     struct bpf_object_open_opts *opts);
+		     const struct bpf_object_open_opts *opts);
 
 /* deprecated bpf_object__open variants */
 LIBBPF_API struct bpf_object *
@@ -126,11 +106,6 @@ bpf_object__open_buffer(const void *obj_buf, size_t obj_buf_sz,
 LIBBPF_API struct bpf_object *
 bpf_object__open_xattr(struct bpf_object_open_attr *attr);
 
-int bpf_object__section_size(const struct bpf_object *obj, const char *name,
-			     __u32 *size);
-int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
-				__u32 *off);
-
 enum libbpf_pin_type {
 	LIBBPF_PIN_NONE,
 	/* PIN_BY_NAME: pin maps by name (in /sys/fs/bpf by default) */
@@ -161,6 +136,7 @@ struct bpf_object_load_attr {
 LIBBPF_API int bpf_object__load(struct bpf_object *obj);
 LIBBPF_API int bpf_object__load_xattr(struct bpf_object_load_attr *attr);
 LIBBPF_API int bpf_object__unload(struct bpf_object *obj);
+
 LIBBPF_API const char *bpf_object__name(const struct bpf_object *obj);
 LIBBPF_API unsigned int bpf_object__kversion(const struct bpf_object *obj);
 
@@ -171,6 +147,9 @@ LIBBPF_API int bpf_object__btf_fd(const struct bpf_object *obj);
 LIBBPF_API struct bpf_program *
 bpf_object__find_program_by_title(const struct bpf_object *obj,
 				  const char *title);
+LIBBPF_API struct bpf_program *
+bpf_object__find_program_by_name(const struct bpf_object *obj,
+				 const char *name);
 
 LIBBPF_API struct bpf_object *bpf_object__next(struct bpf_object *prev);
 #define bpf_object__for_each_safe(pos, tmp)			\
@@ -214,6 +193,7 @@ LIBBPF_API void *bpf_program__priv(const struct bpf_program *prog);
 LIBBPF_API void bpf_program__set_ifindex(struct bpf_program *prog,
 					 __u32 ifindex);
 
+LIBBPF_API const char *bpf_program__name(const struct bpf_program *prog);
 LIBBPF_API const char *bpf_program__title(const struct bpf_program *prog,
 					  bool needs_copy);
 
@@ -235,9 +215,12 @@ LIBBPF_API void bpf_program__unload(struct bpf_program *prog);
 
 struct bpf_link;
 
+LIBBPF_API void bpf_link__disconnect(struct bpf_link *link);
 LIBBPF_API int bpf_link__destroy(struct bpf_link *link);
 
 LIBBPF_API struct bpf_link *
+bpf_program__attach(struct bpf_program *prog);
+LIBBPF_API struct bpf_link *
 bpf_program__attach_perf_event(struct bpf_program *prog, int pfd);
 LIBBPF_API struct bpf_link *
 bpf_program__attach_kprobe(struct bpf_program *prog, bool retprobe,
@@ -512,18 +495,6 @@ bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
 			   bpf_perf_event_print_t fn, void *private_data);
 
-struct nlattr;
-typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
-int libbpf_netlink_open(unsigned int *nl_pid);
-int libbpf_nl_get_link(int sock, unsigned int nl_pid,
-		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
-int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
-int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
-int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
-			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
-
 struct bpf_prog_linfo;
 struct bpf_prog_info;
 
@@ -630,6 +601,50 @@ bpf_program__bpil_offs_to_addr(struct bpf_prog_info_linear *info_linear);
  */
 LIBBPF_API int libbpf_num_possible_cpus(void);
 
+struct bpf_map_skeleton {
+	const char *name;
+	struct bpf_map **map;
+	void **mmaped;
+};
+
+struct bpf_prog_skeleton {
+	const char *name;
+	struct bpf_program **prog;
+	struct bpf_link **link;
+};
+
+struct bpf_object_skeleton {
+	size_t sz; /* size of this struct, for forward/backward compatibility */
+
+	const char *name;
+	void *data;
+	size_t data_sz;
+
+	struct bpf_object **obj;
+
+	int map_cnt;
+	int map_skel_sz; /* sizeof(struct bpf_skeleton_map) */
+	struct bpf_map_skeleton *maps;
+
+	int prog_cnt;
+	int prog_skel_sz; /* sizeof(struct bpf_skeleton_prog) */
+	struct bpf_prog_skeleton *progs;
+};
+
+LIBBPF_API int
+bpf_object__open_skeleton(struct bpf_object_skeleton *s,
+			  const struct bpf_object_open_opts *opts);
+LIBBPF_API int bpf_object__load_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API int bpf_object__attach_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API void bpf_object__detach_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s);
+
+enum libbpf_tristate {
+	TRI_NO = 0,
+	TRI_YES = 1,
+	TRI_MODULE = 2,
+};
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index 8ddc2c40e482..e9713a574243 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -208,3 +208,19 @@ LIBBPF_0.0.6 {
 		btf__find_by_name_kind;
 		libbpf_find_vmlinux_btf_id;
 } LIBBPF_0.0.5;
+
+LIBBPF_0.0.7 {
+	global:
+		btf_dump__emit_type_decl;
+		bpf_link__disconnect;
+		bpf_object__find_program_by_name;
+		bpf_object__attach_skeleton;
+		bpf_object__destroy_skeleton;
+		bpf_object__detach_skeleton;
+		bpf_object__load_skeleton;
+		bpf_object__open_skeleton;
+		bpf_prog_attach_xattr;
+		bpf_program__attach;
+		bpf_program__name;
+		btf__align_of;
+} LIBBPF_0.0.6;
diff --git a/tools/lib/bpf/libbpf.pc.template b/tools/lib/bpf/libbpf.pc.template
index ac17fcef2108..b45ed534bdfb 100644
--- a/tools/lib/bpf/libbpf.pc.template
+++ b/tools/lib/bpf/libbpf.pc.template
@@ -8,5 +8,5 @@ Name: libbpf
 Description: BPF library
 Version: @VERSION@
 Libs: -L${libdir} -lbpf
-Requires.private: libelf
+Requires.private: libelf zlib
 Cflags: -I${includedir}
diff --git a/tools/lib/bpf/libbpf_common.h b/tools/lib/bpf/libbpf_common.h
new file mode 100644
index 000000000000..a23ae1ac27eb
--- /dev/null
+++ b/tools/lib/bpf/libbpf_common.h
@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+/*
+ * Common user-facing libbpf helpers.
+ *
+ * Copyright (c) 2019 Facebook
+ */
+
+#ifndef __LIBBPF_LIBBPF_COMMON_H
+#define __LIBBPF_LIBBPF_COMMON_H
+
+#include <string.h>
+
+#ifndef LIBBPF_API
+#define LIBBPF_API __attribute__((visibility("default")))
+#endif
+
+/* Helper macro to declare and initialize libbpf options struct
+ *
+ * This dance with uninitialized declaration, followed by memset to zero,
+ * followed by assignment using compound literal syntax is done to preserve
+ * ability to use a nice struct field initialization syntax and **hopefully**
+ * have all the padding bytes initialized to zero. It's not guaranteed though,
+ * when copying literal, that compiler won't copy garbage in literal's padding
+ * bytes, but that's the best way I've found and it seems to work in practice.
+ *
+ * Macro declares opts struct of given type and name, zero-initializes,
+ * including any extra padding, it with memset() and then assigns initial
+ * values provided by users in struct initializer-syntax as varargs.
+ */
+#define DECLARE_LIBBPF_OPTS(TYPE, NAME, ...)				    \
+	struct TYPE NAME = ({ 						    \
+		memset(&NAME, 0, sizeof(struct TYPE));			    \
+		(struct TYPE) {						    \
+			.sz = sizeof(struct TYPE),			    \
+			__VA_ARGS__					    \
+		};							    \
+	})
+
+#endif /* __LIBBPF_LIBBPF_COMMON_H */
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 97ac17a64a58..8c3afbd97747 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -76,7 +76,7 @@ static inline bool libbpf_validate_opts(const char *opts,
 
 		for (i = opts_sz; i < user_sz; i++) {
 			if (opts[i]) {
-				pr_warn("%s has non-zero extra bytes",
+				pr_warn("%s has non-zero extra bytes\n",
 					type_name);
 				return false;
 			}
@@ -95,9 +95,28 @@ static inline bool libbpf_validate_opts(const char *opts,
 #define OPTS_GET(opts, field, fallback_value) \
 	(OPTS_HAS(opts, field) ? (opts)->field : fallback_value)
 
+int parse_cpu_mask_str(const char *s, bool **mask, int *mask_sz);
+int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz);
 int libbpf__load_raw_btf(const char *raw_types, size_t types_len,
 			 const char *str_sec, size_t str_len);
 
+int bpf_object__section_size(const struct bpf_object *obj, const char *name,
+			     __u32 *size);
+int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
+				__u32 *off);
+
+struct nlattr;
+typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
+int libbpf_netlink_open(unsigned int *nl_pid);
+int libbpf_nl_get_link(int sock, unsigned int nl_pid,
+		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
+int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
+			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
+int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
+			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
+int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
+			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
+
 struct btf_ext_info {
 	/*
 	 * info points to the individual info section (e.g. func_info and
diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index 419652458da4..b139b3d75ebb 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -21,7 +21,6 @@ test_lirc_mode2_user
 get_cgroup_id_user
 test_skb_cgroup_id_user
 test_socket_cookie
-test_cgroup_attach
 test_cgroup_storage
 test_select_reuseport
 test_flow_dissector
@@ -38,5 +37,7 @@ test_hashmap
 test_btf_dump
 xdping
 test_cpp
+*.skel.h
 /no_alu32
 /bpf_gcc
+/tools
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index e0fe01d9ec33..41691fb067da 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -3,10 +3,12 @@ include ../../../../scripts/Kbuild.include
 include ../../../scripts/Makefile.arch
 
 CURDIR := $(abspath .)
-LIBDIR := $(abspath ../../../lib)
+TOOLSDIR := $(abspath ../../..)
+LIBDIR := $(TOOLSDIR)/lib
 BPFDIR := $(LIBDIR)/bpf
-TOOLSDIR := $(abspath ../../../include)
-APIDIR := $(TOOLSDIR)/uapi
+TOOLSINCDIR := $(TOOLSDIR)/include
+BPFTOOLDIR := $(TOOLSDIR)/bpf/bpftool
+APIDIR := $(TOOLSINCDIR)/uapi
 GENDIR := $(abspath ../../../../include/generated)
 GENHDR := $(GENDIR)/autoconf.h
 
@@ -19,18 +21,18 @@ LLC		?= llc
 LLVM_OBJCOPY	?= llvm-objcopy
 BPF_GCC		?= $(shell command -v bpf-gcc;)
 CFLAGS += -g -Wall -O2 $(GENFLAGS) -I$(APIDIR) -I$(LIBDIR) -I$(BPFDIR)	\
-	  -I$(GENDIR) -I$(TOOLSDIR) -I$(CURDIR)				\
+	  -I$(GENDIR) -I$(TOOLSINCDIR) -I$(CURDIR)			\
 	  -Dbpf_prog_load=bpf_prog_test_load				\
 	  -Dbpf_load_program=bpf_test_load_program
-LDLIBS += -lcap -lelf -lrt -lpthread
+LDLIBS += -lcap -lelf -lz -lrt -lpthread
 
 # Order correspond to 'make run_tests' order
 TEST_GEN_PROGS = test_verifier test_tag test_maps test_lru_map test_lpm_map test_progs \
 	test_align test_verifier_log test_dev_cgroup test_tcpbpf_user \
 	test_sock test_btf test_sockmap get_cgroup_id_user test_socket_cookie \
-	test_cgroup_storage test_select_reuseport \
+	test_cgroup_storage \
 	test_netcnt test_tcpnotify_user test_sock_fields test_sysctl test_hashmap \
-	test_cgroup_attach test_progs-no_alu32
+	test_progs-no_alu32
 
 # Also test bpf-gcc, if present
 ifneq ($(BPF_GCC),)
@@ -75,6 +77,24 @@ TEST_GEN_PROGS_EXTENDED = test_sock_addr test_skb_cgroup_id_user \
 
 TEST_CUSTOM_PROGS = urandom_read
 
+# Emit succinct information message describing current building step
+# $1 - generic step name (e.g., CC, LINK, etc);
+# $2 - optional "flavor" specifier; if provided, will be emitted as [flavor];
+# $3 - target (assumed to be file); only file name will be emitted;
+# $4 - optional extra arg, emitted as-is, if provided.
+ifeq ($(V),1)
+msg =
+else
+msg = @$(info $(1)$(if $(2), [$(2)]) $(notdir $(3)))$(if $(4), $(4))
+endif
+
+# override lib.mk's default rules
+OVERRIDE_TARGETS := 1
+override define CLEAN
+	$(call msg,    CLEAN)
+	$(RM) -r $(TEST_GEN_PROGS) $(TEST_GEN_PROGS_EXTENDED) $(TEST_GEN_FILES) $(EXTRA_CLEAN)
+endef
+
 include ../lib.mk
 
 # Define simple and short `make test_progs`, `make test_sysctl`, etc targets
@@ -87,10 +107,16 @@ $(notdir $(TEST_GEN_PROGS)						\
 	 $(TEST_GEN_PROGS_EXTENDED)					\
 	 $(TEST_CUSTOM_PROGS)): %: $(OUTPUT)/% ;
 
+$(OUTPUT)/%:%.c
+	$(call msg,     BINARY,,$@)
+	$(LINK.c) $^ $(LDLIBS) -o $@
+
 $(OUTPUT)/urandom_read: urandom_read.c
+	$(call msg,     BINARY,,$@)
 	$(CC) -o $@ $< -Wl,--build-id
 
 $(OUTPUT)/test_stub.o: test_stub.c
+	$(call msg,         CC,,$@)
 	$(CC) -c $(CFLAGS) -o $@ $<
 
 BPFOBJ := $(OUTPUT)/libbpf.a
@@ -110,13 +136,18 @@ $(OUTPUT)/test_cgroup_storage: cgroup_helpers.c
 $(OUTPUT)/test_netcnt: cgroup_helpers.c
 $(OUTPUT)/test_sock_fields: cgroup_helpers.c
 $(OUTPUT)/test_sysctl: cgroup_helpers.c
-$(OUTPUT)/test_cgroup_attach: cgroup_helpers.c
 
 .PHONY: force
 
 # force a rebuild of BPFOBJ when its dependencies are updated
 force:
 
+DEFAULT_BPFTOOL := $(OUTPUT)/tools/usr/local/sbin/bpftool
+BPFTOOL ?= $(DEFAULT_BPFTOOL)
+
+$(DEFAULT_BPFTOOL): force
+	$(MAKE) -C $(BPFTOOLDIR) DESTDIR=$(OUTPUT)/tools install
+
 $(BPFOBJ): force
 	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
 
@@ -159,27 +190,33 @@ $(OUTPUT)/flow_dissector_load.o: flow_dissector_load.h
 # $3 - CFLAGS
 # $4 - LDFLAGS
 define CLANG_BPF_BUILD_RULE
+	$(call msg,  CLANG-LLC,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -target bpf -emit-llvm				\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -mattr=dwarfris -march=bpf -mcpu=probe $4 -filetype=obj -o $2
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but with disabled alu32
 define CLANG_NOALU32_BPF_BUILD_RULE
+	$(call msg,  CLANG-LLC,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -target bpf -emit-llvm				\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -march=bpf -mcpu=v2 $4 -filetype=obj -o $2
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but using native Clang and bpf LLC
 define CLANG_NATIVE_BPF_BUILD_RULE
+	$(call msg,  CLANG-BPF,$(TRUNNER_BINARY),$2)
 	($(CLANG) $3 -O2 -emit-llvm					\
 		-c $1 -o - || echo "BPF obj compilation failed") | 	\
 	$(LLC) -march=bpf -mcpu=probe $4 -filetype=obj -o $2
 endef
 # Build BPF object using GCC
 define GCC_BPF_BUILD_RULE
+	$(call msg,    GCC-BPF,$(TRUNNER_BINARY),$2)
 	$(BPF_GCC) $3 $4 -O2 -c $1 -o $2
 endef
 
+SKEL_BLACKLIST := btf__% test_pinning_invalid.c
+
 # Set up extra TRUNNER_XXX "temporary" variables in the environment (relies on
 # $eval()) and pass control to DEFINE_TEST_RUNNER_RULES.
 # Parameters:
@@ -195,8 +232,11 @@ TRUNNER_EXTRA_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o,		\
 				 $$(filter %.c,$(TRUNNER_EXTRA_SOURCES)))
 TRUNNER_EXTRA_HDRS := $$(filter %.h,$(TRUNNER_EXTRA_SOURCES))
 TRUNNER_TESTS_HDR := $(TRUNNER_TESTS_DIR)/tests.h
-TRUNNER_BPF_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o,		\
-				$$(notdir $$(wildcard $(TRUNNER_BPF_PROGS_DIR)/*.c)))
+TRUNNER_BPF_SRCS := $$(notdir $$(wildcard $(TRUNNER_BPF_PROGS_DIR)/*.c))
+TRUNNER_BPF_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o, $$(TRUNNER_BPF_SRCS))
+TRUNNER_BPF_SKELS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.skel.h,	\
+				 $$(filter-out $(SKEL_BLACKLIST),	\
+					       $$(TRUNNER_BPF_SRCS)))
 
 # Evaluate rules now with extra TRUNNER_XXX variables above already defined
 $$(eval $$(call DEFINE_TEST_RUNNER_RULES,$1,$2))
@@ -226,12 +266,19 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 	$$(call $(TRUNNER_BPF_BUILD_RULE),$$<,$$@,			\
 					  $(TRUNNER_BPF_CFLAGS),	\
 					  $(TRUNNER_BPF_LDFLAGS))
+
+$(TRUNNER_BPF_SKELS): $(TRUNNER_OUTPUT)/%.skel.h:			\
+		      $(TRUNNER_OUTPUT)/%.o				\
+		      | $(BPFTOOL) $(TRUNNER_OUTPUT)
+	$$(call msg,   GEN-SKEL,$(TRUNNER_BINARY),$$@)
+	$$(BPFTOOL) gen skeleton $$< > $$@
 endif
 
 # ensure we set up tests.h header generation rule just once
 ifeq ($($(TRUNNER_TESTS_DIR)-tests-hdr),)
 $(TRUNNER_TESTS_DIR)-tests-hdr := y
 $(TRUNNER_TESTS_HDR): $(TRUNNER_TESTS_DIR)/*.c
+	$$(call msg,   TEST-HDR,$(TRUNNER_BINARY),$$@)
 	$$(shell ( cd $(TRUNNER_TESTS_DIR);				\
 		  echo '/* Generated header, do not edit */';		\
 		  ls *.c 2> /dev/null |					\
@@ -245,7 +292,9 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_TESTS_DIR)/%.c				\
 		      $(TRUNNER_EXTRA_HDRS)				\
 		      $(TRUNNER_BPF_OBJS)				\
+		      $(TRUNNER_BPF_SKELS)				\
 		      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
+	$$(call msg,   TEST-OBJ,$(TRUNNER_BINARY),$$@)
 	cd $$(@D) && $$(CC) $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
 
 $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
@@ -253,17 +302,20 @@ $(TRUNNER_EXTRA_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 		       $(TRUNNER_EXTRA_HDRS)				\
 		       $(TRUNNER_TESTS_HDR)				\
 		       $$(BPFOBJ) | $(TRUNNER_OUTPUT)
+	$$(call msg,  EXTRA-OBJ,$(TRUNNER_BINARY),$$@)
 	$$(CC) $$(CFLAGS) -c $$< $$(LDLIBS) -o $$@
 
+# only copy extra resources if in flavored build
 $(TRUNNER_BINARY)-extras: $(TRUNNER_EXTRA_FILES) | $(TRUNNER_OUTPUT)
 ifneq ($2,)
-	# only copy extra resources if in flavored build
+	$$(call msg,  EXTRAS-CP,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
 	cp -a $$^ $(TRUNNER_OUTPUT)/
 endif
 
 $(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)			\
 			     $(TRUNNER_EXTRA_OBJS) $$(BPFOBJ)		\
 			     | $(TRUNNER_BINARY)-extras
+	$$(call msg,     BINARY,,$$@)
 	$$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) -o $$@
 
 endef
@@ -315,12 +367,15 @@ verifier/tests.h: verifier/*.c
 		  echo '#endif' \
 		) > verifier/tests.h)
 $(OUTPUT)/test_verifier: test_verifier.c verifier/tests.h $(BPFOBJ) | $(OUTPUT)
+	$(call msg,     BINARY,,$@)
 	$(CC) $(CFLAGS) $(filter %.a %.o %.c,$^) $(LDLIBS) -o $@
 
 # Make sure we are able to include and link libbpf against c++.
-$(OUTPUT)/test_cpp: test_cpp.cpp $(BPFOBJ)
+$(OUTPUT)/test_cpp: test_cpp.cpp $(OUTPUT)/test_core_extern.skel.h $(BPFOBJ)
+	$(call msg,        CXX,,$@)
 	$(CXX) $(CFLAGS) $^ $(LDLIBS) -o $@
 
 EXTRA_CLEAN := $(TEST_CUSTOM_PROGS)					\
 	prog_tests/tests.h map_tests/tests.h verifier/tests.h		\
-	feature $(OUTPUT)/*.o $(OUTPUT)/no_alu32 $(OUTPUT)/bpf_gcc
+	feature $(OUTPUT)/*.o $(OUTPUT)/no_alu32 $(OUTPUT)/bpf_gcc	\
+	tools *.skel.h
diff --git a/tools/testing/selftests/bpf/prog_tests/attach_probe.c b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
index a83111a32d4a..a0ee87c8e1ea 100644
--- a/tools/testing/selftests/bpf/prog_tests/attach_probe.c
+++ b/tools/testing/selftests/bpf/prog_tests/attach_probe.c
@@ -1,26 +1,9 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
-
-#define EMBED_FILE(NAME, PATH)						    \
-asm (									    \
-"      .pushsection \".rodata\", \"a\", @progbits              \n"	    \
-"      .global "#NAME"_data                                    \n"	    \
-#NAME"_data:                                                   \n"	    \
-"      .incbin \"" PATH "\"                                    \n"	    \
-#NAME"_data_end:                                               \n"	    \
-"      .global "#NAME"_size                                    \n"	    \
-"      .type "#NAME"_size, @object                             \n"	    \
-"      .size "#NAME"_size, 4                                   \n"	    \
-"      .align 4,                                               \n"	    \
-#NAME"_size:                                                   \n"	    \
-"      .int "#NAME"_data_end - "#NAME"_data                    \n"	    \
-"      .popsection                                             \n"	    \
-);									    \
-extern char NAME##_data[];						    \
-extern int NAME##_size;
+#include "test_attach_probe.skel.h"
 
 ssize_t get_base_addr() {
-	size_t start;
+	size_t start, offset;
 	char buf[256];
 	FILE *f;
 
@@ -28,10 +11,11 @@ ssize_t get_base_addr() {
 	if (!f)
 		return -errno;
 
-	while (fscanf(f, "%zx-%*x %s %*s\n", &start, buf) == 2) {
+	while (fscanf(f, "%zx-%*x %s %zx %*[^\n]\n",
+		      &start, buf, &offset) == 3) {
 		if (strcmp(buf, "r-xp") == 0) {
 			fclose(f);
-			return start;
+			return start - offset;
 		}
 	}
 
@@ -39,30 +23,12 @@ ssize_t get_base_addr() {
 	return -EINVAL;
 }
 
-EMBED_FILE(probe, "test_attach_probe.o");
-
 void test_attach_probe(void)
 {
-	const char *kprobe_name = "kprobe/sys_nanosleep";
-	const char *kretprobe_name = "kretprobe/sys_nanosleep";
-	const char *uprobe_name = "uprobe/trigger_func";
-	const char *uretprobe_name = "uretprobe/trigger_func";
-	const int kprobe_idx = 0, kretprobe_idx = 1;
-	const int uprobe_idx = 2, uretprobe_idx = 3;
-	const char *obj_name = "attach_probe";
-	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, open_opts,
-		.object_name = obj_name,
-		.relaxed_maps = true,
-	);
-	struct bpf_program *kprobe_prog, *kretprobe_prog;
-	struct bpf_program *uprobe_prog, *uretprobe_prog;
-	struct bpf_object *obj;
-	int err, duration = 0, res;
-	struct bpf_link *kprobe_link = NULL;
-	struct bpf_link *kretprobe_link = NULL;
-	struct bpf_link *uprobe_link = NULL;
-	struct bpf_link *uretprobe_link = NULL;
-	int results_map_fd;
+	int duration = 0;
+	struct bpf_link *kprobe_link, *kretprobe_link;
+	struct bpf_link *uprobe_link, *uretprobe_link;
+	struct test_attach_probe* skel;
 	size_t uprobe_offset;
 	ssize_t base_addr;
 
@@ -72,123 +38,68 @@ void test_attach_probe(void)
 		return;
 	uprobe_offset = (size_t)&get_base_addr - base_addr;
 
-	/* open object */
-	obj = bpf_object__open_mem(probe_data, probe_size, &open_opts);
-	if (CHECK(IS_ERR(obj), "obj_open_mem", "err %ld\n", PTR_ERR(obj)))
+	skel = test_attach_probe__open_and_load();
+	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
 		return;
-
-	if (CHECK(strcmp(bpf_object__name(obj), obj_name), "obj_name",
-		  "wrong obj name '%s', expected '%s'\n",
-		  bpf_object__name(obj), obj_name))
+	if (CHECK(!skel->bss, "check_bss", ".bss wasn't mmap()-ed\n"))
 		goto cleanup;
 
-	kprobe_prog = bpf_object__find_program_by_title(obj, kprobe_name);
-	if (CHECK(!kprobe_prog, "find_probe",
-		  "prog '%s' not found\n", kprobe_name))
-		goto cleanup;
-	kretprobe_prog = bpf_object__find_program_by_title(obj, kretprobe_name);
-	if (CHECK(!kretprobe_prog, "find_probe",
-		  "prog '%s' not found\n", kretprobe_name))
-		goto cleanup;
-	uprobe_prog = bpf_object__find_program_by_title(obj, uprobe_name);
-	if (CHECK(!uprobe_prog, "find_probe",
-		  "prog '%s' not found\n", uprobe_name))
-		goto cleanup;
-	uretprobe_prog = bpf_object__find_program_by_title(obj, uretprobe_name);
-	if (CHECK(!uretprobe_prog, "find_probe",
-		  "prog '%s' not found\n", uretprobe_name))
-		goto cleanup;
-
-	/* create maps && load programs */
-	err = bpf_object__load(obj);
-	if (CHECK(err, "obj_load", "err %d\n", err))
-		goto cleanup;
-
-	/* load maps */
-	results_map_fd = bpf_find_map(__func__, obj, "results_map");
-	if (CHECK(results_map_fd < 0, "find_results_map",
-		  "err %d\n", results_map_fd))
-		goto cleanup;
-
-	kprobe_link = bpf_program__attach_kprobe(kprobe_prog,
+	kprobe_link = bpf_program__attach_kprobe(skel->progs.handle_kprobe,
 						 false /* retprobe */,
 						 SYS_NANOSLEEP_KPROBE_NAME);
 	if (CHECK(IS_ERR(kprobe_link), "attach_kprobe",
-		  "err %ld\n", PTR_ERR(kprobe_link))) {
-		kprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(kprobe_link)))
 		goto cleanup;
-	}
-	kretprobe_link = bpf_program__attach_kprobe(kretprobe_prog,
+	skel->links.handle_kprobe = kprobe_link;
+
+	kretprobe_link = bpf_program__attach_kprobe(skel->progs.handle_kretprobe,
 						    true /* retprobe */,
 						    SYS_NANOSLEEP_KPROBE_NAME);
 	if (CHECK(IS_ERR(kretprobe_link), "attach_kretprobe",
-		  "err %ld\n", PTR_ERR(kretprobe_link))) {
-		kretprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(kretprobe_link)))
 		goto cleanup;
-	}
-	uprobe_link = bpf_program__attach_uprobe(uprobe_prog,
+	skel->links.handle_kretprobe = kretprobe_link;
+
+	uprobe_link = bpf_program__attach_uprobe(skel->progs.handle_uprobe,
 						 false /* retprobe */,
 						 0 /* self pid */,
 						 "/proc/self/exe",
 						 uprobe_offset);
 	if (CHECK(IS_ERR(uprobe_link), "attach_uprobe",
-		  "err %ld\n", PTR_ERR(uprobe_link))) {
-		uprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(uprobe_link)))
 		goto cleanup;
-	}
-	uretprobe_link = bpf_program__attach_uprobe(uretprobe_prog,
+	skel->links.handle_uprobe = uprobe_link;
+
+	uretprobe_link = bpf_program__attach_uprobe(skel->progs.handle_uretprobe,
 						    true /* retprobe */,
 						    -1 /* any pid */,
 						    "/proc/self/exe",
 						    uprobe_offset);
 	if (CHECK(IS_ERR(uretprobe_link), "attach_uretprobe",
-		  "err %ld\n", PTR_ERR(uretprobe_link))) {
-		uretprobe_link = NULL;
+		  "err %ld\n", PTR_ERR(uretprobe_link)))
 		goto cleanup;
-	}
+	skel->links.handle_uretprobe = uretprobe_link;
 
 	/* trigger & validate kprobe && kretprobe */
 	usleep(1);
 
-	err = bpf_map_lookup_elem(results_map_fd, &kprobe_idx, &res);
-	if (CHECK(err, "get_kprobe_res",
-		  "failed to get kprobe res: %d\n", err))
+	if (CHECK(skel->bss->kprobe_res != 1, "check_kprobe_res",
+		  "wrong kprobe res: %d\n", skel->bss->kprobe_res))
 		goto cleanup;
-	if (CHECK(res != kprobe_idx + 1, "check_kprobe_res",
-		  "wrong kprobe res: %d\n", res))
-		goto cleanup;
-
-	err = bpf_map_lookup_elem(results_map_fd, &kretprobe_idx, &res);
-	if (CHECK(err, "get_kretprobe_res",
-		  "failed to get kretprobe res: %d\n", err))
-		goto cleanup;
-	if (CHECK(res != kretprobe_idx + 1, "check_kretprobe_res",
-		  "wrong kretprobe res: %d\n", res))
+	if (CHECK(skel->bss->kretprobe_res != 2, "check_kretprobe_res",
+		  "wrong kretprobe res: %d\n", skel->bss->kretprobe_res))
 		goto cleanup;
 
 	/* trigger & validate uprobe & uretprobe */
 	get_base_addr();
 
-	err = bpf_map_lookup_elem(results_map_fd, &uprobe_idx, &res);
-	if (CHECK(err, "get_uprobe_res",
-		  "failed to get uprobe res: %d\n", err))
-		goto cleanup;
-	if (CHECK(res != uprobe_idx + 1, "check_uprobe_res",
-		  "wrong uprobe res: %d\n", res))
-		goto cleanup;
-
-	err = bpf_map_lookup_elem(results_map_fd, &uretprobe_idx, &res);
-	if (CHECK(err, "get_uretprobe_res",
-		  "failed to get uretprobe res: %d\n", err))
+	if (CHECK(skel->bss->uprobe_res != 3, "check_uprobe_res",
+		  "wrong uprobe res: %d\n", skel->bss->uprobe_res))
 		goto cleanup;
-	if (CHECK(res != uretprobe_idx + 1, "check_uretprobe_res",
-		  "wrong uretprobe res: %d\n", res))
+	if (CHECK(skel->bss->uretprobe_res != 4, "check_uretprobe_res",
+		  "wrong uretprobe res: %d\n", skel->bss->uretprobe_res))
 		goto cleanup;
 
 cleanup:
-	bpf_link__destroy(kprobe_link);
-	bpf_link__destroy(kretprobe_link);
-	bpf_link__destroy(uprobe_link);
-	bpf_link__destroy(uretprobe_link);
-	bpf_object__close(obj);
+	test_attach_probe__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c
new file mode 100644
index 000000000000..5b13f2c6c402
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_autodetach.c
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "cgroup_helpers.h"
+
+#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
+
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+
+static int prog_load(void)
+{
+	struct bpf_insn prog[] = {
+		BPF_MOV64_IMM(BPF_REG_0, 1), /* r0 = 1 */
+		BPF_EXIT_INSN(),
+	};
+	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+
+	return bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+			       prog, insns_cnt, "GPL", 0,
+			       bpf_log_buf, BPF_LOG_BUF_SIZE);
+}
+
+void test_cgroup_attach_autodetach(void)
+{
+	__u32 duration = 0, prog_cnt = 4, attach_flags;
+	int allow_prog[2] = {-1};
+	__u32 prog_ids[2] = {0};
+	void *ptr = NULL;
+	int cg = 0, i;
+	int attempts;
+
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
+		allow_prog[i] = prog_load();
+		if (CHECK(allow_prog[i] < 0, "prog_load",
+			  "verifier output:\n%s\n-------\n", bpf_log_buf))
+			goto err;
+	}
+
+	if (CHECK_FAIL(setup_cgroup_environment()))
+		goto err;
+
+	/* create a cgroup, attach two programs and remember their ids */
+	cg = create_and_get_cgroup("/cg_autodetach");
+	if (CHECK_FAIL(cg < 0))
+		goto err;
+
+	if (CHECK_FAIL(join_cgroup("/cg_autodetach")))
+		goto err;
+
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
+		if (CHECK(bpf_prog_attach(allow_prog[i], cg,
+					  BPF_CGROUP_INET_EGRESS,
+					  BPF_F_ALLOW_MULTI),
+			  "prog_attach", "prog[%d], errno=%d\n", i, errno))
+			goto err;
+
+	/* make sure that programs are attached and run some traffic */
+	if (CHECK(bpf_prog_query(cg, BPF_CGROUP_INET_EGRESS, 0, &attach_flags,
+				 prog_ids, &prog_cnt),
+		  "prog_query", "errno=%d\n", errno))
+		goto err;
+	if (CHECK_FAIL(system(PING_CMD)))
+		goto err;
+
+	/* allocate some memory (4Mb) to pin the original cgroup */
+	ptr = malloc(4 * (1 << 20));
+	if (CHECK_FAIL(!ptr))
+		goto err;
+
+	/* close programs and cgroup fd */
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
+		close(allow_prog[i]);
+		allow_prog[i] = -1;
+	}
+
+	close(cg);
+	cg = 0;
+
+	/* leave the cgroup and remove it. don't detach programs */
+	cleanup_cgroup_environment();
+
+	/* wait for the asynchronous auto-detachment.
+	 * wait for no more than 5 sec and give up.
+	 */
+	for (i = 0; i < ARRAY_SIZE(prog_ids); i++) {
+		for (attempts = 5; attempts >= 0; attempts--) {
+			int fd = bpf_prog_get_fd_by_id(prog_ids[i]);
+
+			if (fd < 0)
+				break;
+
+			/* don't leave the fd open */
+			close(fd);
+
+			if (CHECK_FAIL(!attempts))
+				goto err;
+
+			sleep(1);
+		}
+	}
+
+err:
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
+		if (allow_prog[i] >= 0)
+			close(allow_prog[i]);
+	if (cg)
+		close(cg);
+	free(ptr);
+	cleanup_cgroup_environment();
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
new file mode 100644
index 000000000000..2ff21dbce179
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_multi.c
@@ -0,0 +1,285 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "cgroup_helpers.h"
+
+#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
+
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+
+static int map_fd = -1;
+
+static int prog_load_cnt(int verdict, int val)
+{
+	int cgroup_storage_fd, percpu_cgroup_storage_fd;
+
+	if (map_fd < 0)
+		map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, 4, 8, 1, 0);
+	if (map_fd < 0) {
+		printf("failed to create map '%s'\n", strerror(errno));
+		return -1;
+	}
+
+	cgroup_storage_fd = bpf_create_map(BPF_MAP_TYPE_CGROUP_STORAGE,
+				sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
+	if (cgroup_storage_fd < 0) {
+		printf("failed to create map '%s'\n", strerror(errno));
+		return -1;
+	}
+
+	percpu_cgroup_storage_fd = bpf_create_map(
+		BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE,
+		sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
+	if (percpu_cgroup_storage_fd < 0) {
+		printf("failed to create map '%s'\n", strerror(errno));
+		return -1;
+	}
+
+	struct bpf_insn prog[] = {
+		BPF_MOV32_IMM(BPF_REG_0, 0),
+		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), /* *(u32 *)(fp - 4) = r0 */
+		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), /* r2 = fp - 4 */
+		BPF_LD_MAP_FD(BPF_REG_1, map_fd),
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
+		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
+		BPF_MOV64_IMM(BPF_REG_1, val), /* r1 = 1 */
+		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_DW, BPF_REG_0, BPF_REG_1, 0, 0), /* xadd r0 += r1 */
+
+		BPF_LD_MAP_FD(BPF_REG_1, cgroup_storage_fd),
+		BPF_MOV64_IMM(BPF_REG_2, 0),
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
+		BPF_MOV64_IMM(BPF_REG_1, val),
+		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_W, BPF_REG_0, BPF_REG_1, 0, 0),
+
+		BPF_LD_MAP_FD(BPF_REG_1, percpu_cgroup_storage_fd),
+		BPF_MOV64_IMM(BPF_REG_2, 0),
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
+		BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 0x1),
+		BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_3, 0),
+
+		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
+		BPF_EXIT_INSN(),
+	};
+	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+	int ret;
+
+	ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+			       prog, insns_cnt, "GPL", 0,
+			       bpf_log_buf, BPF_LOG_BUF_SIZE);
+
+	close(cgroup_storage_fd);
+	return ret;
+}
+
+void test_cgroup_attach_multi(void)
+{
+	__u32 prog_ids[4], prog_cnt = 0, attach_flags, saved_prog_id;
+	int cg1 = 0, cg2 = 0, cg3 = 0, cg4 = 0, cg5 = 0, key = 0;
+	DECLARE_LIBBPF_OPTS(bpf_prog_attach_opts, attach_opts);
+	int allow_prog[7] = {-1};
+	unsigned long long value;
+	__u32 duration = 0;
+	int i = 0;
+
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
+		allow_prog[i] = prog_load_cnt(1, 1 << i);
+		if (CHECK(allow_prog[i] < 0, "prog_load",
+			  "verifier output:\n%s\n-------\n", bpf_log_buf))
+			goto err;
+	}
+
+	if (CHECK_FAIL(setup_cgroup_environment()))
+		goto err;
+
+	cg1 = create_and_get_cgroup("/cg1");
+	if (CHECK_FAIL(cg1 < 0))
+		goto err;
+	cg2 = create_and_get_cgroup("/cg1/cg2");
+	if (CHECK_FAIL(cg2 < 0))
+		goto err;
+	cg3 = create_and_get_cgroup("/cg1/cg2/cg3");
+	if (CHECK_FAIL(cg3 < 0))
+		goto err;
+	cg4 = create_and_get_cgroup("/cg1/cg2/cg3/cg4");
+	if (CHECK_FAIL(cg4 < 0))
+		goto err;
+	cg5 = create_and_get_cgroup("/cg1/cg2/cg3/cg4/cg5");
+	if (CHECK_FAIL(cg5 < 0))
+		goto err;
+
+	if (CHECK_FAIL(join_cgroup("/cg1/cg2/cg3/cg4/cg5")))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_MULTI),
+		  "prog0_attach_to_cg1_multi", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
+				   BPF_F_ALLOW_MULTI),
+		  "fail_same_prog_attach_to_cg1", "unexpected success\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[1], cg1, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_MULTI),
+		  "prog1_attach_to_cg1_multi", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[2], cg2, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog2_attach_to_cg2_override", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_MULTI),
+		  "prog3_attach_to_cg3_multi", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[4], cg4, BPF_CGROUP_INET_EGRESS,
+			    BPF_F_ALLOW_OVERRIDE),
+		  "prog4_attach_to_cg4_override", "errno=%d\n", errno))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog[5], cg5, BPF_CGROUP_INET_EGRESS, 0),
+		  "prog5_attach_to_cg5_none", "errno=%d\n", errno))
+		goto err;
+
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 1 + 2 + 8 + 32);
+
+	/* query the number of effective progs in cg5 */
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, NULL, NULL, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 4);
+	/* retrieve prog_ids of effective progs in cg5 */
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, &attach_flags,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 4);
+	CHECK_FAIL(attach_flags != 0);
+	saved_prog_id = prog_ids[0];
+	/* check enospc handling */
+	prog_ids[0] = 0;
+	prog_cnt = 2;
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, &attach_flags,
+				  prog_ids, &prog_cnt) != -1);
+	CHECK_FAIL(errno != ENOSPC);
+	CHECK_FAIL(prog_cnt != 4);
+	/* check that prog_ids are returned even when buffer is too small */
+	CHECK_FAIL(prog_ids[0] != saved_prog_id);
+	/* retrieve prog_id of single attached prog in cg5 */
+	prog_ids[0] = 0;
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0, NULL,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 1);
+	CHECK_FAIL(prog_ids[0] != saved_prog_id);
+
+	/* detach bottom program and ping again */
+	if (CHECK(bpf_prog_detach2(-1, cg5, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_from_cg5", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 1 + 2 + 8 + 16);
+
+	/* test replace */
+
+	attach_opts.flags = BPF_F_ALLOW_OVERRIDE | BPF_F_REPLACE;
+	attach_opts.replace_prog_fd = allow_prog[0];
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_override", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != EINVAL);
+
+	attach_opts.flags = BPF_F_REPLACE;
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_no_multi", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != EINVAL);
+
+	attach_opts.flags = BPF_F_ALLOW_MULTI | BPF_F_REPLACE;
+	attach_opts.replace_prog_fd = -1;
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_bad_fd", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != EBADF);
+
+	/* replacing a program that is not attached to cgroup should fail  */
+	attach_opts.replace_prog_fd = allow_prog[3];
+	if (CHECK(!bpf_prog_attach_xattr(allow_prog[6], cg1,
+					 BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "fail_prog_replace_no_ent", "unexpected success\n"))
+		goto err;
+	CHECK_FAIL(errno != ENOENT);
+
+	/* replace 1st from the top program */
+	attach_opts.replace_prog_fd = allow_prog[0];
+	if (CHECK(bpf_prog_attach_xattr(allow_prog[6], cg1,
+					BPF_CGROUP_INET_EGRESS, &attach_opts),
+		  "prog_replace", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 64 + 2 + 8 + 16);
+
+	/* detach 3rd from bottom program and ping again */
+	if (CHECK(!bpf_prog_detach2(0, cg3, BPF_CGROUP_INET_EGRESS),
+		  "fail_prog_detach_from_cg3", "unexpected success\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_detach2(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS),
+		  "prog3_detach_from_cg3", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 64 + 2 + 16);
+
+	/* detach 2nd from bottom program and ping again */
+	if (CHECK(bpf_prog_detach2(-1, cg4, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_from_cg4", "errno=%d\n", errno))
+		goto err;
+
+	value = 0;
+	CHECK_FAIL(bpf_map_update_elem(map_fd, &key, &value, 0));
+	CHECK_FAIL(system(PING_CMD));
+	CHECK_FAIL(bpf_map_lookup_elem(map_fd, &key, &value));
+	CHECK_FAIL(value != 64 + 2 + 4);
+
+	prog_cnt = 4;
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_QUERY_EFFECTIVE, &attach_flags,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 3);
+	CHECK_FAIL(attach_flags != 0);
+	CHECK_FAIL(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0, NULL,
+				  prog_ids, &prog_cnt));
+	CHECK_FAIL(prog_cnt != 0);
+
+err:
+	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
+		if (allow_prog[i] >= 0)
+			close(allow_prog[i]);
+	close(cg1);
+	close(cg2);
+	close(cg3);
+	close(cg4);
+	close(cg5);
+	cleanup_cgroup_environment();
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c
new file mode 100644
index 000000000000..9d8cb48b99de
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cgroup_attach_override.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+
+#include "cgroup_helpers.h"
+
+#define FOO		"/foo"
+#define BAR		"/foo/bar/"
+#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
+
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+
+static int prog_load(int verdict)
+{
+	struct bpf_insn prog[] = {
+		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
+		BPF_EXIT_INSN(),
+	};
+	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+
+	return bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+			       prog, insns_cnt, "GPL", 0,
+			       bpf_log_buf, BPF_LOG_BUF_SIZE);
+}
+
+void test_cgroup_attach_override(void)
+{
+	int drop_prog = -1, allow_prog = -1, foo = -1, bar = -1;
+	__u32 duration = 0;
+
+	allow_prog = prog_load(1);
+	if (CHECK(allow_prog < 0, "prog_load_allow",
+		  "verifier output:\n%s\n-------\n", bpf_log_buf))
+		goto err;
+
+	drop_prog = prog_load(0);
+	if (CHECK(drop_prog < 0, "prog_load_drop",
+		  "verifier output:\n%s\n-------\n", bpf_log_buf))
+		goto err;
+
+	foo = test__join_cgroup(FOO);
+	if (CHECK(foo < 0, "cgroup_join_foo", "cgroup setup failed\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_drop_foo_override",
+		  "attach prog to %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+	if (CHECK(!system(PING_CMD), "ping_fail",
+		  "ping unexpectedly succeeded\n"))
+		goto err;
+
+	bar = test__join_cgroup(BAR);
+	if (CHECK(bar < 0, "cgroup_join_bar", "cgroup setup failed\n"))
+		goto err;
+
+	if (CHECK(!system(PING_CMD), "ping_fail",
+		  "ping unexpectedly succeeded\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_allow_bar_override",
+		  "attach prog to %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(system(PING_CMD), "ping_ok", "ping failed\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_bar",
+		  "detach prog from %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(!system(PING_CMD), "ping_fail",
+		  "ping unexpectedly succeeded\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_allow_bar_override",
+		  "attach prog to %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_foo",
+		  "detach prog from %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+	if (CHECK(system(PING_CMD), "ping_ok", "ping failed\n"))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				  BPF_F_ALLOW_OVERRIDE),
+		  "prog_attach_allow_bar_override",
+		  "attach prog to %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0),
+		  "fail_prog_attach_allow_bar_none",
+		  "attach prog to %s unexpectedly succeeded\n", BAR))
+		goto err;
+
+	if (CHECK(bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS),
+		  "prog_detach_bar",
+		  "detach prog from %s failed, errno=%d\n", BAR, errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS),
+		  "fail_prog_detach_foo",
+		  "double detach from %s unexpectedly succeeded\n", FOO))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS, 0),
+		  "prog_attach_allow_foo_none",
+		  "attach prog to %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0),
+		  "fail_prog_attach_allow_bar_none",
+		  "attach prog to %s unexpectedly succeeded\n", BAR))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
+				   BPF_F_ALLOW_OVERRIDE),
+		  "fail_prog_attach_allow_bar_override",
+		  "attach prog to %s unexpectedly succeeded\n", BAR))
+		goto err;
+
+	if (CHECK(!bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS,
+				   BPF_F_ALLOW_OVERRIDE),
+		  "fail_prog_attach_allow_foo_override",
+		  "attach prog to %s unexpectedly succeeded\n", FOO))
+		goto err;
+
+	if (CHECK(bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS, 0),
+		  "prog_attach_drop_foo_none",
+		  "attach prog to %s failed, errno=%d\n", FOO, errno))
+		goto err;
+
+err:
+	close(foo);
+	close(bar);
+	close(allow_prog);
+	close(drop_prog);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/core_extern.c b/tools/testing/selftests/bpf/prog_tests/core_extern.c
new file mode 100644
index 000000000000..b093787e9448
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/core_extern.c
@@ -0,0 +1,169 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <test_progs.h>
+#include <sys/mman.h>
+#include <sys/utsname.h>
+#include <linux/version.h>
+#include "test_core_extern.skel.h"
+
+static uint32_t get_kernel_version(void)
+{
+	uint32_t major, minor, patch;
+	struct utsname info;
+
+	uname(&info);
+	if (sscanf(info.release, "%u.%u.%u", &major, &minor, &patch) != 3)
+		return 0;
+	return KERNEL_VERSION(major, minor, patch);
+}
+
+#define CFG "CONFIG_BPF_SYSCALL=n\n"
+
+static struct test_case {
+	const char *name;
+	const char *cfg;
+	bool fails;
+	struct test_core_extern__data data;
+} test_cases[] = {
+	{ .name = "default search path", .data = { .bpf_syscall = true } },
+	{
+		.name = "custom values",
+		.cfg = "CONFIG_BPF_SYSCALL=n\n"
+		       "CONFIG_TRISTATE=m\n"
+		       "CONFIG_BOOL=y\n"
+		       "CONFIG_CHAR=100\n"
+		       "CONFIG_USHORT=30000\n"
+		       "CONFIG_INT=123456\n"
+		       "CONFIG_ULONG=0xDEADBEEFC0DE\n"
+		       "CONFIG_STR=\"abracad\"\n"
+		       "CONFIG_MISSING=0",
+		.data = {
+			.bpf_syscall = false,
+			.tristate_val = TRI_MODULE,
+			.bool_val = true,
+			.char_val = 100,
+			.ushort_val = 30000,
+			.int_val = 123456,
+			.ulong_val = 0xDEADBEEFC0DE,
+			.str_val = "abracad",
+		},
+	},
+	/* TRISTATE */
+	{ .name = "tristate (y)", .cfg = CFG"CONFIG_TRISTATE=y\n",
+	  .data = { .tristate_val = TRI_YES } },
+	{ .name = "tristate (n)", .cfg = CFG"CONFIG_TRISTATE=n\n",
+	  .data = { .tristate_val = TRI_NO } },
+	{ .name = "tristate (m)", .cfg = CFG"CONFIG_TRISTATE=m\n",
+	  .data = { .tristate_val = TRI_MODULE } },
+	{ .name = "tristate (int)", .fails = 1, .cfg = CFG"CONFIG_TRISTATE=1" },
+	{ .name = "tristate (bad)", .fails = 1, .cfg = CFG"CONFIG_TRISTATE=M" },
+	/* BOOL */
+	{ .name = "bool (y)", .cfg = CFG"CONFIG_BOOL=y\n",
+	  .data = { .bool_val = true } },
+	{ .name = "bool (n)", .cfg = CFG"CONFIG_BOOL=n\n",
+	  .data = { .bool_val = false } },
+	{ .name = "bool (tristate)", .fails = 1, .cfg = CFG"CONFIG_BOOL=m" },
+	{ .name = "bool (int)", .fails = 1, .cfg = CFG"CONFIG_BOOL=1" },
+	/* CHAR */
+	{ .name = "char (tristate)", .cfg = CFG"CONFIG_CHAR=m\n",
+	  .data = { .char_val = 'm' } },
+	{ .name = "char (bad)", .fails = 1, .cfg = CFG"CONFIG_CHAR=q\n" },
+	{ .name = "char (empty)", .fails = 1, .cfg = CFG"CONFIG_CHAR=\n" },
+	{ .name = "char (str)", .fails = 1, .cfg = CFG"CONFIG_CHAR=\"y\"\n" },
+	/* STRING */
+	{ .name = "str (empty)", .cfg = CFG"CONFIG_STR=\"\"\n",
+	  .data = { .str_val = "\0\0\0\0\0\0\0" } },
+	{ .name = "str (padded)", .cfg = CFG"CONFIG_STR=\"abra\"\n",
+	  .data = { .str_val = "abra\0\0\0" } },
+	{ .name = "str (too long)", .cfg = CFG"CONFIG_STR=\"abracada\"\n",
+	  .data = { .str_val = "abracad" } },
+	{ .name = "str (no value)", .fails = 1, .cfg = CFG"CONFIG_STR=\n" },
+	{ .name = "str (bad value)", .fails = 1, .cfg = CFG"CONFIG_STR=bla\n" },
+	/* INTEGERS */
+	{
+		.name = "integer forms",
+		.cfg = CFG
+		       "CONFIG_CHAR=0xA\n"
+		       "CONFIG_USHORT=0462\n"
+		       "CONFIG_INT=-100\n"
+		       "CONFIG_ULONG=+1000000000000",
+		.data = {
+			.char_val = 0xA,
+			.ushort_val = 0462,
+			.int_val = -100,
+			.ulong_val = 1000000000000,
+		},
+	},
+	{ .name = "int (bad)", .fails = 1, .cfg = CFG"CONFIG_INT=abc" },
+	{ .name = "int (str)", .fails = 1, .cfg = CFG"CONFIG_INT=\"abc\"" },
+	{ .name = "int (empty)", .fails = 1, .cfg = CFG"CONFIG_INT=" },
+	{ .name = "int (mixed)", .fails = 1, .cfg = CFG"CONFIG_INT=123abc" },
+	{ .name = "int (max)", .cfg = CFG"CONFIG_INT=2147483647",
+	  .data = { .int_val = 2147483647 } },
+	{ .name = "int (min)", .cfg = CFG"CONFIG_INT=-2147483648",
+	  .data = { .int_val = -2147483648 } },
+	{ .name = "int (max+1)", .fails = 1, .cfg = CFG"CONFIG_INT=2147483648" },
+	{ .name = "int (min-1)", .fails = 1, .cfg = CFG"CONFIG_INT=-2147483649" },
+	{ .name = "ushort (max)", .cfg = CFG"CONFIG_USHORT=65535",
+	  .data = { .ushort_val = 65535 } },
+	{ .name = "ushort (min)", .cfg = CFG"CONFIG_USHORT=0",
+	  .data = { .ushort_val = 0 } },
+	{ .name = "ushort (max+1)", .fails = 1, .cfg = CFG"CONFIG_USHORT=65536" },
+	{ .name = "ushort (min-1)", .fails = 1, .cfg = CFG"CONFIG_USHORT=-1" },
+	{ .name = "u64 (max)", .cfg = CFG"CONFIG_ULONG=0xffffffffffffffff",
+	  .data = { .ulong_val = 0xffffffffffffffff } },
+	{ .name = "u64 (min)", .cfg = CFG"CONFIG_ULONG=0",
+	  .data = { .ulong_val = 0 } },
+	{ .name = "u64 (max+1)", .fails = 1, .cfg = CFG"CONFIG_ULONG=0x10000000000000000" },
+};
+
+void test_core_extern(void)
+{
+	const uint32_t kern_ver = get_kernel_version();
+	int err, duration = 0, i, j;
+	struct test_core_extern *skel = NULL;
+	uint64_t *got, *exp;
+	int n = sizeof(*skel->data) / sizeof(uint64_t);
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		struct test_case *t = &test_cases[i];
+		DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts,
+			.kconfig = t->cfg,
+		);
+
+		if (!test__start_subtest(t->name))
+			continue;
+
+		skel = test_core_extern__open_opts(&opts);
+		if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
+			goto cleanup;
+		err = test_core_extern__load(skel);
+		if (t->fails) {
+			CHECK(!err, "skel_load",
+			      "shouldn't succeed open/load of skeleton\n");
+			goto cleanup;
+		} else if (CHECK(err, "skel_load",
+				 "failed to open/load skeleton\n")) {
+			goto cleanup;
+		}
+		err = test_core_extern__attach(skel);
+		if (CHECK(err, "attach_raw_tp", "failed attach: %d\n", err))
+			goto cleanup;
+
+		usleep(1);
+
+		t->data.kern_ver = kern_ver;
+		t->data.missing_val = 0xDEADC0DE;
+		got = (uint64_t *)skel->data;
+		exp = (uint64_t *)&t->data;
+		for (j = 0; j < n; j++) {
+			CHECK(got[j] != exp[j], "check_res",
+			      "result #%d: expected %lx, but got %lx\n",
+			       j, exp[j], got[j]);
+		}
+cleanup:
+		test_core_extern__destroy(skel);
+		skel = NULL;
+	}
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/core_reloc.c b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
index 05fe85281ff7..31e177adbdf1 100644
--- a/tools/testing/selftests/bpf/prog_tests/core_reloc.c
+++ b/tools/testing/selftests/bpf/prog_tests/core_reloc.c
@@ -74,6 +74,7 @@
 		.b123 = 2,						\
 		.c1c  = 3,						\
 		.d00d = 4,						\
+		.f10c = 0,						\
 	},								\
 	.output_len = sizeof(struct core_reloc_arrays_output)		\
 }
@@ -308,12 +309,15 @@ static struct core_reloc_test_case test_cases[] = {
 	ARRAYS_CASE(arrays),
 	ARRAYS_CASE(arrays___diff_arr_dim),
 	ARRAYS_CASE(arrays___diff_arr_val_sz),
+	ARRAYS_CASE(arrays___equiv_zero_sz_arr),
+	ARRAYS_CASE(arrays___fixed_arr),
 
 	ARRAYS_ERR_CASE(arrays___err_too_small),
 	ARRAYS_ERR_CASE(arrays___err_too_shallow),
 	ARRAYS_ERR_CASE(arrays___err_non_array),
 	ARRAYS_ERR_CASE(arrays___err_wrong_val_type1),
 	ARRAYS_ERR_CASE(arrays___err_wrong_val_type2),
+	ARRAYS_ERR_CASE(arrays___err_bad_zero_sz_arr),
 
 	/* enum/ptr/int handling scenarios */
 	PRIMITIVES_CASE(primitives),
diff --git a/tools/testing/selftests/bpf/prog_tests/cpu_mask.c b/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
new file mode 100644
index 000000000000..1fa1bdbaffa9
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/cpu_mask.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <bpf/btf.h>
+#include "libbpf_internal.h"
+
+static int duration = 0;
+
+static void validate_mask(int case_nr, const char *exp, bool *mask, int n)
+{
+	int i;
+
+	for (i = 0; exp[i]; i++) {
+		if (exp[i] == '1') {
+			if (CHECK(i + 1 > n, "mask_short",
+				  "case #%d: mask too short, got n=%d, need at least %d\n",
+				  case_nr, n, i + 1))
+				return;
+			CHECK(!mask[i], "cpu_not_set",
+			      "case #%d: mask differs, expected cpu#%d SET\n",
+			      case_nr, i);
+		} else {
+			CHECK(i < n && mask[i], "cpu_set",
+			      "case #%d: mask differs, expected cpu#%d UNSET\n",
+			      case_nr, i);
+		}
+	}
+	CHECK(i < n, "mask_long",
+	      "case #%d: mask too long, got n=%d, expected at most %d\n",
+	      case_nr, n, i);
+}
+
+static struct {
+	const char *cpu_mask;
+	const char *expect;
+	bool fails;
+} test_cases[] = {
+	{ "0\n", "1", false },
+	{ "0,2\n", "101", false },
+	{ "0-2\n", "111", false },
+	{ "0-2,3-4\n", "11111", false },
+	{ "0", "1", false },
+	{ "0-2", "111", false },
+	{ "0,2", "101", false },
+	{ "0,1-3", "1111", false },
+	{ "0,1,2,3", "1111", false },
+	{ "0,2-3,5", "101101", false },
+	{ "3-3", "0001", false },
+	{ "2-4,6,9-10", "00111010011", false },
+	/* failure cases */
+	{ "", "", true },
+	{ "0-", "", true },
+	{ "0 ", "", true },
+	{ "0_1", "", true },
+	{ "1-0", "", true },
+	{ "-1", "", true },
+};
+
+void test_cpu_mask()
+{
+	int i, err, n;
+	bool *mask;
+
+	for (i = 0; i < ARRAY_SIZE(test_cases); i++) {
+		mask = NULL;
+		err = parse_cpu_mask_str(test_cases[i].cpu_mask, &mask, &n);
+		if (test_cases[i].fails) {
+			CHECK(!err, "should_fail",
+			      "case #%d: parsing should fail!\n", i + 1);
+		} else {
+			if (CHECK(err, "parse_err",
+				  "case #%d: cpu mask parsing failed: %d\n",
+				  i + 1, err))
+				continue;
+			validate_mask(i + 1, test_cases[i].expect, mask, n);
+		}
+		free(mask);
+	}
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c b/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
index 40bcff2cc274..235ac4f67f5b 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_fexit.c
@@ -1,90 +1,55 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <test_progs.h>
+#include "test_pkt_access.skel.h"
+#include "fentry_test.skel.h"
+#include "fexit_test.skel.h"
 
 void test_fentry_fexit(void)
 {
-	struct bpf_prog_load_attr attr_fentry = {
-		.file = "./fentry_test.o",
-	};
-	struct bpf_prog_load_attr attr_fexit = {
-		.file = "./fexit_test.o",
-	};
-
-	struct bpf_object *obj_fentry = NULL, *obj_fexit = NULL, *pkt_obj;
-	struct bpf_map *data_map_fentry, *data_map_fexit;
-	char fentry_name[] = "fentry/bpf_fentry_testX";
-	char fexit_name[] = "fexit/bpf_fentry_testX";
-	int err, pkt_fd, kfree_skb_fd, i;
-	struct bpf_link *link[12] = {};
-	struct bpf_program *prog[12];
-	__u32 duration, retval;
-	const int zero = 0;
-	u64 result[12];
-
-	err = bpf_prog_load("./test_pkt_access.o", BPF_PROG_TYPE_SCHED_CLS,
-			    &pkt_obj, &pkt_fd);
-	if (CHECK(err, "prog_load sched cls", "err %d errno %d\n", err, errno))
+	struct test_pkt_access *pkt_skel = NULL;
+	struct fentry_test *fentry_skel = NULL;
+	struct fexit_test *fexit_skel = NULL;
+	__u64 *fentry_res, *fexit_res;
+	__u32 duration = 0, retval;
+	int err, pkt_fd, i;
+
+	pkt_skel = test_pkt_access__open_and_load();
+	if (CHECK(!pkt_skel, "pkt_skel_load", "pkt_access skeleton failed\n"))
 		return;
-	err = bpf_prog_load_xattr(&attr_fentry, &obj_fentry, &kfree_skb_fd);
-	if (CHECK(err, "prog_load fail", "err %d errno %d\n", err, errno))
+	fentry_skel = fentry_test__open_and_load();
+	if (CHECK(!fentry_skel, "fentry_skel_load", "fentry skeleton failed\n"))
 		goto close_prog;
-	err = bpf_prog_load_xattr(&attr_fexit, &obj_fexit, &kfree_skb_fd);
-	if (CHECK(err, "prog_load fail", "err %d errno %d\n", err, errno))
+	fexit_skel = fexit_test__open_and_load();
+	if (CHECK(!fexit_skel, "fexit_skel_load", "fexit skeleton failed\n"))
 		goto close_prog;
 
-	for (i = 0; i < 6; i++) {
-		fentry_name[sizeof(fentry_name) - 2] = '1' + i;
-		prog[i] = bpf_object__find_program_by_title(obj_fentry, fentry_name);
-		if (CHECK(!prog[i], "find_prog", "prog %s not found\n", fentry_name))
-			goto close_prog;
-		link[i] = bpf_program__attach_trace(prog[i]);
-		if (CHECK(IS_ERR(link[i]), "attach_trace", "failed to link\n"))
-			goto close_prog;
-	}
-	data_map_fentry = bpf_object__find_map_by_name(obj_fentry, "fentry_t.bss");
-	if (CHECK(!data_map_fentry, "find_data_map", "data map not found\n"))
+	err = fentry_test__attach(fentry_skel);
+	if (CHECK(err, "fentry_attach", "fentry attach failed: %d\n", err))
 		goto close_prog;
-
-	for (i = 6; i < 12; i++) {
-		fexit_name[sizeof(fexit_name) - 2] = '1' + i - 6;
-		prog[i] = bpf_object__find_program_by_title(obj_fexit, fexit_name);
-		if (CHECK(!prog[i], "find_prog", "prog %s not found\n", fexit_name))
-			goto close_prog;
-		link[i] = bpf_program__attach_trace(prog[i]);
-		if (CHECK(IS_ERR(link[i]), "attach_trace", "failed to link\n"))
-			goto close_prog;
-	}
-	data_map_fexit = bpf_object__find_map_by_name(obj_fexit, "fexit_te.bss");
-	if (CHECK(!data_map_fexit, "find_data_map", "data map not found\n"))
+	err = fexit_test__attach(fexit_skel);
+	if (CHECK(err, "fexit_attach", "fexit attach failed: %d\n", err))
 		goto close_prog;
 
+	pkt_fd = bpf_program__fd(pkt_skel->progs.test_pkt_access);
 	err = bpf_prog_test_run(pkt_fd, 1, &pkt_v6, sizeof(pkt_v6),
 				NULL, NULL, &retval, &duration);
 	CHECK(err || retval, "ipv6",
 	      "err %d errno %d retval %d duration %d\n",
 	      err, errno, retval, duration);
 
-	err = bpf_map_lookup_elem(bpf_map__fd(data_map_fentry), &zero, &result);
-	if (CHECK(err, "get_result",
-		  "failed to get output data: %d\n", err))
-		goto close_prog;
-
-	err = bpf_map_lookup_elem(bpf_map__fd(data_map_fexit), &zero, result + 6);
-	if (CHECK(err, "get_result",
-		  "failed to get output data: %d\n", err))
-		goto close_prog;
-
-	for (i = 0; i < 12; i++)
-		if (CHECK(result[i] != 1, "result", "bpf_fentry_test%d failed err %ld\n",
-			  i % 6 + 1, result[i]))
-			goto close_prog;
+	fentry_res = (__u64 *)fentry_skel->bss;
+	fexit_res = (__u64 *)fexit_skel->bss;
+	printf("%lld\n", fentry_skel->bss->test1_result);
+	for (i = 0; i < 6; i++) {
+		CHECK(fentry_res[i] != 1, "result",
+		      "fentry_test%d failed err %lld\n", i + 1, fentry_res[i]);
+		CHECK(fexit_res[i] != 1, "result",
+		      "fexit_test%d failed err %lld\n", i + 1, fexit_res[i]);
+	}
 
 close_prog:
-	for (i = 0; i < 12; i++)
-		if (!IS_ERR_OR_NULL(link[i]))
-			bpf_link__destroy(link[i]);
-	bpf_object__close(obj_fentry);
-	bpf_object__close(obj_fexit);
-	bpf_object__close(pkt_obj);
+	test_pkt_access__destroy(pkt_skel);
+	fentry_test__destroy(fentry_skel);
+	fexit_test__destroy(fexit_skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/fentry_test.c b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
index 9fb103193878..e1a379f5f7d2 100644
--- a/tools/testing/selftests/bpf/prog_tests/fentry_test.c
+++ b/tools/testing/selftests/bpf/prog_tests/fentry_test.c
@@ -1,64 +1,43 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2019 Facebook */
 #include <test_progs.h>
+#include "test_pkt_access.skel.h"
+#include "fentry_test.skel.h"
 
 void test_fentry_test(void)
 {
-	struct bpf_prog_load_attr attr = {
-		.file = "./fentry_test.o",
-	};
-
-	char prog_name[] = "fentry/bpf_fentry_testX";
-	struct bpf_object *obj = NULL, *pkt_obj;
-	int err, pkt_fd, kfree_skb_fd, i;
-	struct bpf_link *link[6] = {};
-	struct bpf_program *prog[6];
+	struct test_pkt_access *pkt_skel = NULL;
+	struct fentry_test *fentry_skel = NULL;
+	int err, pkt_fd, i;
 	__u32 duration, retval;
-	struct bpf_map *data_map;
-	const int zero = 0;
-	u64 result[6];
+	__u64 *result;
 
-	err = bpf_prog_load("./test_pkt_access.o", BPF_PROG_TYPE_SCHED_CLS,
-			    &pkt_obj, &pkt_fd);
-	if (CHECK(err, "prog_load sched cls", "err %d errno %d\n", err, errno))
+	pkt_skel = test_pkt_access__open_and_load();
+	if (CHECK(!pkt_skel, "pkt_skel_load", "pkt_access skeleton failed\n"))
 		return;
-	err = bpf_prog_load_xattr(&attr, &obj, &kfree_skb_fd);
-	if (CHECK(err, "prog_load fail", "err %d errno %d\n", err, errno))
-		goto close_prog;
+	fentry_skel = fentry_test__open_and_load();
+	if (CHECK(!fentry_skel, "fentry_skel_load", "fentry skeleton failed\n"))
+		goto cleanup;
 
-	for (i = 0; i < 6; i++) {
-		prog_name[sizeof(prog_name) - 2] = '1' + i;
-		prog[i] = bpf_object__find_program_by_title(obj, prog_name);
-		if (CHECK(!prog[i], "find_prog", "prog %s not found\n", prog_name))
-			goto close_prog;
-		link[i] = bpf_program__attach_trace(prog[i]);
-		if (CHECK(IS_ERR(link[i]), "attach_trace", "failed to link\n"))
-			goto close_prog;
-	}
-	data_map = bpf_object__find_map_by_name(obj, "fentry_t.bss");
-	if (CHECK(!data_map, "find_data_map", "data map not found\n"))
-		goto close_prog;
+	err = fentry_test__attach(fentry_skel);
+	if (CHECK(err, "fentry_attach", "fentry attach failed: %d\n", err))
+		goto cleanup;
 
+	pkt_fd = bpf_program__fd(pkt_skel->progs.test_pkt_access);
 	err = bpf_prog_test_run(pkt_fd, 1, &pkt_v6, sizeof(pkt_v6),
 				NULL, NULL, &retval, &duration);
 	CHECK(err || retval, "ipv6",
 	      "err %d errno %d retval %d duration %d\n",
 	      err, errno, retval, duration);
 
-	err = bpf_map_lookup_elem(bpf_map__fd(data_map), &zero, &result);
-	if (CHECK(err, "get_result",
-		  "failed to get output data: %d\n", err))
-		goto close_prog;
-
-	for (i = 0; i < 6; i++)
-		if (CHECK(result[i] != 1, "result", "bpf_fentry_test%d failed err %ld\n",
-			  i + 1, result[i]))
-			goto close_prog;
+	result = (__u64 *)fentry_skel->bss;
+	for (i = 0; i < 6; i++) {
+		if (CHECK(result[i] != 1, "result",
+			  "fentry_test%d failed err %lld\n", i + 1, result[i]))
+			goto cleanup;
+	}
 
-close_prog:
-	for (i = 0; i < 6; i++)
-		if (!IS_ERR_OR_NULL(link[i]))
-			bpf_link__destroy(link[i]);
-	bpf_object__close(obj);
-	bpf_object__close(pkt_obj);
+cleanup:
+	fentry_test__destroy(fentry_skel);
+	test_pkt_access__destroy(pkt_skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/mmap.c b/tools/testing/selftests/bpf/prog_tests/mmap.c
index 051a6d48762c..16a814eb4d64 100644
--- a/tools/testing/selftests/bpf/prog_tests/mmap.c
+++ b/tools/testing/selftests/bpf/prog_tests/mmap.c
@@ -1,16 +1,12 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
 #include <sys/mman.h>
+#include "test_mmap.skel.h"
 
 struct map_data {
 	__u64 val[512 * 4];
 };
 
-struct bss_data {
-	__u64 in_val;
-	__u64 out_val;
-};
-
 static size_t roundup_page(size_t sz)
 {
 	long page_size = sysconf(_SC_PAGE_SIZE);
@@ -19,41 +15,25 @@ static size_t roundup_page(size_t sz)
 
 void test_mmap(void)
 {
-	const char *file = "test_mmap.o";
-	const char *probe_name = "raw_tracepoint/sys_enter";
-	const char *tp_name = "sys_enter";
-	const size_t bss_sz = roundup_page(sizeof(struct bss_data));
+	const size_t bss_sz = roundup_page(sizeof(struct test_mmap__bss));
 	const size_t map_sz = roundup_page(sizeof(struct map_data));
 	const int zero = 0, one = 1, two = 2, far = 1500;
 	const long page_size = sysconf(_SC_PAGE_SIZE);
 	int err, duration = 0, i, data_map_fd;
-	struct bpf_program *prog;
-	struct bpf_object *obj;
-	struct bpf_link *link = NULL;
 	struct bpf_map *data_map, *bss_map;
 	void *bss_mmaped = NULL, *map_mmaped = NULL, *tmp1, *tmp2;
-	volatile struct bss_data *bss_data;
-	volatile struct map_data *map_data;
+	struct test_mmap__bss *bss_data;
+	struct map_data *map_data;
+	struct test_mmap *skel;
 	__u64 val = 0;
 
-	obj = bpf_object__open_file("test_mmap.o", NULL);
-	if (CHECK(IS_ERR(obj), "obj_open", "failed to open '%s': %ld\n",
-		  file, PTR_ERR(obj)))
+
+	skel = test_mmap__open_and_load();
+	if (CHECK(!skel, "skel_open_and_load", "skeleton open/load failed\n"))
 		return;
-	prog = bpf_object__find_program_by_title(obj, probe_name);
-	if (CHECK(!prog, "find_probe", "prog '%s' not found\n", probe_name))
-		goto cleanup;
-	err = bpf_object__load(obj);
-	if (CHECK(err, "obj_load", "failed to load prog '%s': %d\n",
-		  probe_name, err))
-		goto cleanup;
 
-	bss_map = bpf_object__find_map_by_name(obj, "test_mma.bss");
-	if (CHECK(!bss_map, "find_bss_map", ".bss map not found\n"))
-		goto cleanup;
-	data_map = bpf_object__find_map_by_name(obj, "data_map");
-	if (CHECK(!data_map, "find_data_map", "data_map map not found\n"))
-		goto cleanup;
+	bss_map = skel->maps.bss;
+	data_map = skel->maps.data_map;
 	data_map_fd = bpf_map__fd(data_map);
 
 	bss_mmaped = mmap(NULL, bss_sz, PROT_READ | PROT_WRITE, MAP_SHARED,
@@ -77,13 +57,15 @@ void test_mmap(void)
 
 	CHECK_FAIL(bss_data->in_val);
 	CHECK_FAIL(bss_data->out_val);
+	CHECK_FAIL(skel->bss->in_val);
+	CHECK_FAIL(skel->bss->out_val);
 	CHECK_FAIL(map_data->val[0]);
 	CHECK_FAIL(map_data->val[1]);
 	CHECK_FAIL(map_data->val[2]);
 	CHECK_FAIL(map_data->val[far]);
 
-	link = bpf_program__attach_raw_tracepoint(prog, tp_name);
-	if (CHECK(IS_ERR(link), "attach_raw_tp", "err %ld\n", PTR_ERR(link)))
+	err = test_mmap__attach(skel);
+	if (CHECK(err, "attach_raw_tp", "err %d\n", err))
 		goto cleanup;
 
 	bss_data->in_val = 123;
@@ -94,6 +76,8 @@ void test_mmap(void)
 
 	CHECK_FAIL(bss_data->in_val != 123);
 	CHECK_FAIL(bss_data->out_val != 123);
+	CHECK_FAIL(skel->bss->in_val != 123);
+	CHECK_FAIL(skel->bss->out_val != 123);
 	CHECK_FAIL(map_data->val[0] != 111);
 	CHECK_FAIL(map_data->val[1] != 222);
 	CHECK_FAIL(map_data->val[2] != 123);
@@ -160,6 +144,8 @@ void test_mmap(void)
 	usleep(1);
 	CHECK_FAIL(bss_data->in_val != 321);
 	CHECK_FAIL(bss_data->out_val != 321);
+	CHECK_FAIL(skel->bss->in_val != 321);
+	CHECK_FAIL(skel->bss->out_val != 321);
 	CHECK_FAIL(map_data->val[0] != 111);
 	CHECK_FAIL(map_data->val[1] != 222);
 	CHECK_FAIL(map_data->val[2] != 321);
@@ -203,6 +189,8 @@ void test_mmap(void)
 	map_data = tmp2;
 	CHECK_FAIL(bss_data->in_val != 321);
 	CHECK_FAIL(bss_data->out_val != 321);
+	CHECK_FAIL(skel->bss->in_val != 321);
+	CHECK_FAIL(skel->bss->out_val != 321);
 	CHECK_FAIL(map_data->val[0] != 111);
 	CHECK_FAIL(map_data->val[1] != 222);
 	CHECK_FAIL(map_data->val[2] != 321);
@@ -214,7 +202,5 @@ cleanup:
 		CHECK_FAIL(munmap(bss_mmaped, bss_sz));
 	if (map_mmaped)
 		CHECK_FAIL(munmap(map_mmaped, map_sz));
-	if (!IS_ERR_OR_NULL(link))
-		bpf_link__destroy(link);
-	bpf_object__close(obj);
+	test_mmap__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
index 3003fddc0613..cf6c87936c69 100644
--- a/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
+++ b/tools/testing/selftests/bpf/prog_tests/perf_buffer.c
@@ -4,6 +4,7 @@
 #include <sched.h>
 #include <sys/socket.h>
 #include <test_progs.h>
+#include "libbpf_internal.h"
 
 static void on_sample(void *ctx, int cpu, void *data, __u32 size)
 {
@@ -19,7 +20,7 @@ static void on_sample(void *ctx, int cpu, void *data, __u32 size)
 
 void test_perf_buffer(void)
 {
-	int err, prog_fd, nr_cpus, i, duration = 0;
+	int err, prog_fd, on_len, nr_on_cpus = 0,  nr_cpus, i, duration = 0;
 	const char *prog_name = "kprobe/sys_nanosleep";
 	const char *file = "./test_perf_buffer.o";
 	struct perf_buffer_opts pb_opts = {};
@@ -29,15 +30,27 @@ void test_perf_buffer(void)
 	struct bpf_object *obj;
 	struct perf_buffer *pb;
 	struct bpf_link *link;
+	bool *online;
 
 	nr_cpus = libbpf_num_possible_cpus();
 	if (CHECK(nr_cpus < 0, "nr_cpus", "err %d\n", nr_cpus))
 		return;
 
+	err = parse_cpu_mask_file("/sys/devices/system/cpu/online",
+				  &online, &on_len);
+	if (CHECK(err, "nr_on_cpus", "err %d\n", err))
+		return;
+
+	for (i = 0; i < on_len; i++)
+		if (online[i])
+			nr_on_cpus++;
+
 	/* load program */
 	err = bpf_prog_load(file, BPF_PROG_TYPE_KPROBE, &obj, &prog_fd);
-	if (CHECK(err, "obj_load", "err %d errno %d\n", err, errno))
-		return;
+	if (CHECK(err, "obj_load", "err %d errno %d\n", err, errno)) {
+		obj = NULL;
+		goto out_close;
+	}
 
 	prog = bpf_object__find_program_by_title(obj, prog_name);
 	if (CHECK(!prog, "find_probe", "prog '%s' not found\n", prog_name))
@@ -64,6 +77,11 @@ void test_perf_buffer(void)
 	/* trigger kprobe on every CPU */
 	CPU_ZERO(&cpu_seen);
 	for (i = 0; i < nr_cpus; i++) {
+		if (i >= on_len || !online[i]) {
+			printf("skipping offline CPU #%d\n", i);
+			continue;
+		}
+
 		CPU_ZERO(&cpu_set);
 		CPU_SET(i, &cpu_set);
 
@@ -81,8 +99,8 @@ void test_perf_buffer(void)
 	if (CHECK(err < 0, "perf_buffer__poll", "err %d\n", err))
 		goto out_free_pb;
 
-	if (CHECK(CPU_COUNT(&cpu_seen) != nr_cpus, "seen_cpu_cnt",
-		  "expect %d, seen %d\n", nr_cpus, CPU_COUNT(&cpu_seen)))
+	if (CHECK(CPU_COUNT(&cpu_seen) != nr_on_cpus, "seen_cpu_cnt",
+		  "expect %d, seen %d\n", nr_on_cpus, CPU_COUNT(&cpu_seen)))
 		goto out_free_pb;
 
 out_free_pb:
@@ -91,4 +109,5 @@ out_detach:
 	bpf_link__destroy(link);
 out_close:
 	bpf_object__close(obj);
+	free(online);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/probe_user.c b/tools/testing/selftests/bpf/prog_tests/probe_user.c
index 8a3187dec048..7aecfd9e87d1 100644
--- a/tools/testing/selftests/bpf/prog_tests/probe_user.c
+++ b/tools/testing/selftests/bpf/prog_tests/probe_user.c
@@ -3,8 +3,7 @@
 
 void test_probe_user(void)
 {
-#define kprobe_name "__sys_connect"
-	const char *prog_name = "kprobe/" kprobe_name;
+	const char *prog_name = "kprobe/__sys_connect";
 	const char *obj_file = "./test_probe_user.o";
 	DECLARE_LIBBPF_OPTS(bpf_object_open_opts, opts, );
 	int err, results_map_fd, sock_fd, duration = 0;
@@ -33,8 +32,7 @@ void test_probe_user(void)
 		  "err %d\n", results_map_fd))
 		goto cleanup;
 
-	kprobe_link = bpf_program__attach_kprobe(kprobe_prog, false,
-						 kprobe_name);
+	kprobe_link = bpf_program__attach(kprobe_prog);
 	if (CHECK(IS_ERR(kprobe_link), "attach_kprobe",
 		  "err %ld\n", PTR_ERR(kprobe_link))) {
 		kprobe_link = NULL;
diff --git a/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c b/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c
index d90acc13d1ec..563e12120e77 100644
--- a/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c
+++ b/tools/testing/selftests/bpf/prog_tests/rdonly_maps.c
@@ -16,14 +16,11 @@ struct rdonly_map_subtest {
 
 void test_rdonly_maps(void)
 {
-	const char *prog_name_skip_loop = "raw_tracepoint/sys_enter:skip_loop";
-	const char *prog_name_part_loop = "raw_tracepoint/sys_enter:part_loop";
-	const char *prog_name_full_loop = "raw_tracepoint/sys_enter:full_loop";
 	const char *file = "test_rdonly_maps.o";
 	struct rdonly_map_subtest subtests[] = {
-		{ "skip loop", prog_name_skip_loop, 0, 0 },
-		{ "part loop", prog_name_part_loop, 3, 2 + 3 + 4 },
-		{ "full loop", prog_name_full_loop, 4, 2 + 3 + 4 + 5 },
+		{ "skip loop", "skip_loop", 0, 0 },
+		{ "part loop", "part_loop", 3, 2 + 3 + 4 },
+		{ "full loop", "full_loop", 4, 2 + 3 + 4 + 5 },
 	};
 	int i, err, zero = 0, duration = 0;
 	struct bpf_link *link = NULL;
@@ -50,7 +47,7 @@ void test_rdonly_maps(void)
 		if (!test__start_subtest(t->subtest_name))
 			continue;
 
-		prog = bpf_object__find_program_by_title(obj, t->prog_name);
+		prog = bpf_object__find_program_by_name(obj, t->prog_name);
 		if (CHECK(!prog, "find_prog", "prog '%s' not found\n",
 			  t->prog_name))
 			goto cleanup;
diff --git a/tools/testing/selftests/bpf/test_select_reuseport.c b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
index 7566c13eb51a..2c37ae7dc214 100644
--- a/tools/testing/selftests/bpf/test_select_reuseport.c
+++ b/tools/testing/selftests/bpf/prog_tests/select_reuseport.c
@@ -20,8 +20,11 @@
 #include <bpf/libbpf.h>
 #include "bpf_rlimit.h"
 #include "bpf_util.h"
+
+#include "test_progs.h"
 #include "test_select_reuseport_common.h"
 
+#define MAX_TEST_NAME 80
 #define MIN_TCPHDR_LEN 20
 #define UDPHDR_LEN 8
 
@@ -32,11 +35,11 @@
 static int result_map, tmp_index_ovr_map, linum_map, data_check_map;
 static enum result expected_results[NR_RESULTS];
 static int sk_fds[REUSEPORT_ARRAY_SIZE];
-static int reuseport_array, outer_map;
+static int reuseport_array = -1, outer_map = -1;
 static int select_by_skb_data_prog;
-static int saved_tcp_syncookie;
+static int saved_tcp_syncookie = -1;
 static struct bpf_object *obj;
-static int saved_tcp_fo;
+static int saved_tcp_fo = -1;
 static __u32 index_zero;
 static int epfd;
 
@@ -46,16 +49,21 @@ static union sa46 {
 	sa_family_t family;
 } srv_sa;
 
-#define CHECK(condition, tag, format...) ({				\
-	int __ret = !!(condition);					\
-	if (__ret) {							\
-		printf("%s(%d):FAIL:%s ", __func__, __LINE__, tag);	\
-		printf(format);						\
-		exit(-1);						\
+#define RET_IF(condition, tag, format...) ({				\
+	if (CHECK_FAIL(condition)) {					\
+		printf(tag " " format);					\
+		return;							\
+	}								\
+})
+
+#define RET_ERR(condition, tag, format...) ({				\
+	if (CHECK_FAIL(condition)) {					\
+		printf(tag " " format);					\
+		return -1;						\
 	}								\
 })
 
-static void create_maps(void)
+static int create_maps(void)
 {
 	struct bpf_create_map_attr attr = {};
 
@@ -67,8 +75,8 @@ static void create_maps(void)
 	attr.max_entries = REUSEPORT_ARRAY_SIZE;
 
 	reuseport_array = bpf_create_map_xattr(&attr);
-	CHECK(reuseport_array == -1, "creating reuseport_array",
-	      "reuseport_array:%d errno:%d\n", reuseport_array, errno);
+	RET_ERR(reuseport_array == -1, "creating reuseport_array",
+		"reuseport_array:%d errno:%d\n", reuseport_array, errno);
 
 	/* Creating outer_map */
 	attr.name = "outer_map";
@@ -78,63 +86,61 @@ static void create_maps(void)
 	attr.max_entries = 1;
 	attr.inner_map_fd = reuseport_array;
 	outer_map = bpf_create_map_xattr(&attr);
-	CHECK(outer_map == -1, "creating outer_map",
-	      "outer_map:%d errno:%d\n", outer_map, errno);
+	RET_ERR(outer_map == -1, "creating outer_map",
+		"outer_map:%d errno:%d\n", outer_map, errno);
+
+	return 0;
 }
 
-static void prepare_bpf_obj(void)
+static int prepare_bpf_obj(void)
 {
 	struct bpf_program *prog;
 	struct bpf_map *map;
 	int err;
-	struct bpf_object_open_attr attr = {
-		.file = "test_select_reuseport_kern.o",
-		.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,
-	};
 
-	obj = bpf_object__open_xattr(&attr);
-	CHECK(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
-	      "obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
-
-	prog = bpf_program__next(NULL, obj);
-	CHECK(!prog, "get first bpf_program", "!prog\n");
-	bpf_program__set_type(prog, attr.prog_type);
+	obj = bpf_object__open("test_select_reuseport_kern.o");
+	RET_ERR(IS_ERR_OR_NULL(obj), "open test_select_reuseport_kern.o",
+		"obj:%p PTR_ERR(obj):%ld\n", obj, PTR_ERR(obj));
 
 	map = bpf_object__find_map_by_name(obj, "outer_map");
-	CHECK(!map, "find outer_map", "!map\n");
+	RET_ERR(!map, "find outer_map", "!map\n");
 	err = bpf_map__reuse_fd(map, outer_map);
-	CHECK(err, "reuse outer_map", "err:%d\n", err);
+	RET_ERR(err, "reuse outer_map", "err:%d\n", err);
 
 	err = bpf_object__load(obj);
-	CHECK(err, "load bpf_object", "err:%d\n", err);
+	RET_ERR(err, "load bpf_object", "err:%d\n", err);
 
+	prog = bpf_program__next(NULL, obj);
+	RET_ERR(!prog, "get first bpf_program", "!prog\n");
 	select_by_skb_data_prog = bpf_program__fd(prog);
-	CHECK(select_by_skb_data_prog == -1, "get prog fd",
-	      "select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
+	RET_ERR(select_by_skb_data_prog == -1, "get prog fd",
+		"select_by_skb_data_prog:%d\n", select_by_skb_data_prog);
 
 	map = bpf_object__find_map_by_name(obj, "result_map");
-	CHECK(!map, "find result_map", "!map\n");
+	RET_ERR(!map, "find result_map", "!map\n");
 	result_map = bpf_map__fd(map);
-	CHECK(result_map == -1, "get result_map fd",
-	      "result_map:%d\n", result_map);
+	RET_ERR(result_map == -1, "get result_map fd",
+		"result_map:%d\n", result_map);
 
 	map = bpf_object__find_map_by_name(obj, "tmp_index_ovr_map");
-	CHECK(!map, "find tmp_index_ovr_map", "!map\n");
+	RET_ERR(!map, "find tmp_index_ovr_map\n", "!map");
 	tmp_index_ovr_map = bpf_map__fd(map);
-	CHECK(tmp_index_ovr_map == -1, "get tmp_index_ovr_map fd",
-	      "tmp_index_ovr_map:%d\n", tmp_index_ovr_map);
+	RET_ERR(tmp_index_ovr_map == -1, "get tmp_index_ovr_map fd",
+		"tmp_index_ovr_map:%d\n", tmp_index_ovr_map);
 
 	map = bpf_object__find_map_by_name(obj, "linum_map");
-	CHECK(!map, "find linum_map", "!map\n");
+	RET_ERR(!map, "find linum_map", "!map\n");
 	linum_map = bpf_map__fd(map);
-	CHECK(linum_map == -1, "get linum_map fd",
-	      "linum_map:%d\n", linum_map);
+	RET_ERR(linum_map == -1, "get linum_map fd",
+		"linum_map:%d\n", linum_map);
 
 	map = bpf_object__find_map_by_name(obj, "data_check_map");
-	CHECK(!map, "find data_check_map", "!map\n");
+	RET_ERR(!map, "find data_check_map", "!map\n");
 	data_check_map = bpf_map__fd(map);
-	CHECK(data_check_map == -1, "get data_check_map fd",
-	      "data_check_map:%d\n", data_check_map);
+	RET_ERR(data_check_map == -1, "get data_check_map fd",
+		"data_check_map:%d\n", data_check_map);
+
+	return 0;
 }
 
 static void sa46_init_loopback(union sa46 *sa, sa_family_t family)
@@ -163,65 +169,73 @@ static int read_int_sysctl(const char *sysctl)
 	int fd, ret;
 
 	fd = open(sysctl, 0);
-	CHECK(fd == -1, "open(sysctl)", "sysctl:%s fd:%d errno:%d\n",
-	      sysctl, fd, errno);
+	RET_ERR(fd == -1, "open(sysctl)",
+		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
 
 	ret = read(fd, buf, sizeof(buf));
-	CHECK(ret <= 0, "read(sysctl)", "sysctl:%s ret:%d errno:%d\n",
-	      sysctl, ret, errno);
-	close(fd);
+	RET_ERR(ret <= 0, "read(sysctl)",
+		"sysctl:%s ret:%d errno:%d\n", sysctl, ret, errno);
 
+	close(fd);
 	return atoi(buf);
 }
 
-static void write_int_sysctl(const char *sysctl, int v)
+static int write_int_sysctl(const char *sysctl, int v)
 {
 	int fd, ret, size;
 	char buf[16];
 
 	fd = open(sysctl, O_RDWR);
-	CHECK(fd == -1, "open(sysctl)", "sysctl:%s fd:%d errno:%d\n",
-	      sysctl, fd, errno);
+	RET_ERR(fd == -1, "open(sysctl)",
+		"sysctl:%s fd:%d errno:%d\n", sysctl, fd, errno);
 
 	size = snprintf(buf, sizeof(buf), "%d", v);
 	ret = write(fd, buf, size);
-	CHECK(ret != size, "write(sysctl)",
-	      "sysctl:%s ret:%d size:%d errno:%d\n", sysctl, ret, size, errno);
+	RET_ERR(ret != size, "write(sysctl)",
+		"sysctl:%s ret:%d size:%d errno:%d\n",
+		sysctl, ret, size, errno);
+
 	close(fd);
+	return 0;
 }
 
 static void restore_sysctls(void)
 {
-	write_int_sysctl(TCP_FO_SYSCTL, saved_tcp_fo);
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
+	if (saved_tcp_fo != -1)
+		write_int_sysctl(TCP_FO_SYSCTL, saved_tcp_fo);
+	if (saved_tcp_syncookie != -1)
+		write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, saved_tcp_syncookie);
 }
 
-static void enable_fastopen(void)
+static int enable_fastopen(void)
 {
 	int fo;
 
 	fo = read_int_sysctl(TCP_FO_SYSCTL);
-	write_int_sysctl(TCP_FO_SYSCTL, fo | 7);
+	if (fo < 0)
+		return -1;
+
+	return write_int_sysctl(TCP_FO_SYSCTL, fo | 7);
 }
 
-static void enable_syncookie(void)
+static int enable_syncookie(void)
 {
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 2);
+	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 2);
 }
 
-static void disable_syncookie(void)
+static int disable_syncookie(void)
 {
-	write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
+	return write_int_sysctl(TCP_SYNCOOKIE_SYSCTL, 0);
 }
 
-static __u32 get_linum(void)
+static long get_linum(void)
 {
 	__u32 linum;
 	int err;
 
 	err = bpf_map_lookup_elem(linum_map, &index_zero, &linum);
-	CHECK(err == -1, "lookup_elem(linum_map)", "err:%d errno:%d\n",
-	      err, errno);
+	RET_ERR(err == -1, "lookup_elem(linum_map)", "err:%d errno:%d\n",
+		err, errno);
 
 	return linum;
 }
@@ -237,12 +251,12 @@ static void check_data(int type, sa_family_t family, const struct cmd *cmd,
 	addrlen = sizeof(cli_sa);
 	err = getsockname(cli_fd, (struct sockaddr *)&cli_sa,
 			  &addrlen);
-	CHECK(err == -1, "getsockname(cli_fd)", "err:%d errno:%d\n",
-	      err, errno);
+	RET_IF(err == -1, "getsockname(cli_fd)", "err:%d errno:%d\n",
+	       err, errno);
 
 	err = bpf_map_lookup_elem(data_check_map, &index_zero, &result);
-	CHECK(err == -1, "lookup_elem(data_check_map)", "err:%d errno:%d\n",
-	      err, errno);
+	RET_IF(err == -1, "lookup_elem(data_check_map)", "err:%d errno:%d\n",
+	       err, errno);
 
 	if (type == SOCK_STREAM) {
 		expected.len = MIN_TCPHDR_LEN;
@@ -284,22 +298,22 @@ static void check_data(int type, sa_family_t family, const struct cmd *cmd,
 		printf("expected: (0x%x, %u, %u)\n",
 		       expected.eth_protocol, expected.ip_protocol,
 		       expected.bind_inany);
-		CHECK(1, "data_check result != expected",
-		      "bpf_prog_linum:%u\n", get_linum());
+		RET_IF(1, "data_check result != expected",
+		       "bpf_prog_linum:%ld\n", get_linum());
 	}
 
-	CHECK(!result.hash, "data_check result.hash empty",
-	      "result.hash:%u", result.hash);
+	RET_IF(!result.hash, "data_check result.hash empty",
+	       "result.hash:%u", result.hash);
 
 	expected.len += cmd ? sizeof(*cmd) : 0;
 	if (type == SOCK_STREAM)
-		CHECK(expected.len > result.len, "expected.len > result.len",
-		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
-		      expected.len, result.len, get_linum());
+		RET_IF(expected.len > result.len, "expected.len > result.len",
+		       "expected.len:%u result.len:%u bpf_prog_linum:%ld\n",
+		       expected.len, result.len, get_linum());
 	else
-		CHECK(expected.len != result.len, "expected.len != result.len",
-		      "expected.len:%u result.len:%u bpf_prog_linum:%u\n",
-		      expected.len, result.len, get_linum());
+		RET_IF(expected.len != result.len, "expected.len != result.len",
+		       "expected.len:%u result.len:%u bpf_prog_linum:%ld\n",
+		       expected.len, result.len, get_linum());
 }
 
 static void check_results(void)
@@ -310,8 +324,8 @@ static void check_results(void)
 
 	for (i = 0; i < NR_RESULTS; i++) {
 		err = bpf_map_lookup_elem(result_map, &i, &results[i]);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "lookup_elem(result_map)",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
 	}
 
 	for (i = 0; i < NR_RESULTS; i++) {
@@ -337,10 +351,10 @@ static void check_results(void)
 		printf(", %u", expected_results[i]);
 	printf("]\n");
 
-	CHECK(expected_results[broken] != results[broken],
-	      "unexpected result",
-	      "expected_results[%u] != results[%u] bpf_prog_linum:%u\n",
-	      broken, broken, get_linum());
+	RET_IF(expected_results[broken] != results[broken],
+	       "unexpected result",
+	       "expected_results[%u] != results[%u] bpf_prog_linum:%ld\n",
+	       broken, broken, get_linum());
 }
 
 static int send_data(int type, sa_family_t family, void *data, size_t len,
@@ -350,17 +364,17 @@ static int send_data(int type, sa_family_t family, void *data, size_t len,
 	int fd, err;
 
 	fd = socket(family, type, 0);
-	CHECK(fd == -1, "socket()", "fd:%d errno:%d\n", fd, errno);
+	RET_ERR(fd == -1, "socket()", "fd:%d errno:%d\n", fd, errno);
 
 	sa46_init_loopback(&cli_sa, family);
 	err = bind(fd, (struct sockaddr *)&cli_sa, sizeof(cli_sa));
-	CHECK(fd == -1, "bind(cli_sa)", "err:%d errno:%d\n", err, errno);
+	RET_ERR(fd == -1, "bind(cli_sa)", "err:%d errno:%d\n", err, errno);
 
 	err = sendto(fd, data, len, MSG_FASTOPEN, (struct sockaddr *)&srv_sa,
 		     sizeof(srv_sa));
-	CHECK(err != len && expected >= PASS,
-	      "sendto()", "family:%u err:%d errno:%d expected:%d\n",
-	      family, err, errno, expected);
+	RET_ERR(err != len && expected >= PASS,
+		"sendto()", "family:%u err:%d errno:%d expected:%d\n",
+		family, err, errno, expected);
 
 	return fd;
 }
@@ -375,47 +389,49 @@ static void do_test(int type, sa_family_t family, struct cmd *cmd,
 
 	cli_fd = send_data(type, family, cmd, cmd ? sizeof(*cmd) : 0,
 			   expected);
+	if (cli_fd < 0)
+		return;
 	nev = epoll_wait(epfd, &ev, 1, expected >= PASS ? 5 : 0);
-	CHECK((nev <= 0 && expected >= PASS) ||
-	      (nev > 0 && expected < PASS),
-	      "nev <> expected",
-	      "nev:%d expected:%d type:%d family:%d data:(%d, %d)\n",
-	      nev, expected, type, family,
-	      cmd ? cmd->reuseport_index : -1,
-	      cmd ? cmd->pass_on_failure : -1);
+	RET_IF((nev <= 0 && expected >= PASS) ||
+	       (nev > 0 && expected < PASS),
+	       "nev <> expected",
+	       "nev:%d expected:%d type:%d family:%d data:(%d, %d)\n",
+	       nev, expected, type, family,
+	       cmd ? cmd->reuseport_index : -1,
+	       cmd ? cmd->pass_on_failure : -1);
 	check_results();
 	check_data(type, family, cmd, cli_fd);
 
 	if (expected < PASS)
 		return;
 
-	CHECK(expected != PASS_ERR_SK_SELECT_REUSEPORT &&
-	      cmd->reuseport_index != ev.data.u32,
-	      "check cmd->reuseport_index",
-	      "cmd:(%u, %u) ev.data.u32:%u\n",
-	      cmd->pass_on_failure, cmd->reuseport_index, ev.data.u32);
+	RET_IF(expected != PASS_ERR_SK_SELECT_REUSEPORT &&
+	       cmd->reuseport_index != ev.data.u32,
+	       "check cmd->reuseport_index",
+	       "cmd:(%u, %u) ev.data.u32:%u\n",
+	       cmd->pass_on_failure, cmd->reuseport_index, ev.data.u32);
 
 	srv_fd = sk_fds[ev.data.u32];
 	if (type == SOCK_STREAM) {
 		int new_fd = accept(srv_fd, NULL, 0);
 
-		CHECK(new_fd == -1, "accept(srv_fd)",
-		      "ev.data.u32:%u new_fd:%d errno:%d\n",
-		      ev.data.u32, new_fd, errno);
+		RET_IF(new_fd == -1, "accept(srv_fd)",
+		       "ev.data.u32:%u new_fd:%d errno:%d\n",
+		       ev.data.u32, new_fd, errno);
 
 		nread = recv(new_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
-		CHECK(nread != sizeof(rcv_cmd),
-		      "recv(new_fd)",
-		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
-		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
+		RET_IF(nread != sizeof(rcv_cmd),
+		       "recv(new_fd)",
+		       "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
+		       ev.data.u32, nread, sizeof(rcv_cmd), errno);
 
 		close(new_fd);
 	} else {
 		nread = recv(srv_fd, &rcv_cmd, sizeof(rcv_cmd), MSG_DONTWAIT);
-		CHECK(nread != sizeof(rcv_cmd),
-		      "recv(sk_fds)",
-		      "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
-		      ev.data.u32, nread, sizeof(rcv_cmd), errno);
+		RET_IF(nread != sizeof(rcv_cmd),
+		       "recv(sk_fds)",
+		       "ev.data.u32:%u nread:%zd sizeof(rcv_cmd):%zu errno:%d\n",
+		       ev.data.u32, nread, sizeof(rcv_cmd), errno);
 	}
 
 	close(cli_fd);
@@ -428,18 +444,14 @@ static void test_err_inner_map(int type, sa_family_t family)
 		.pass_on_failure = 0,
 	};
 
-	printf("%s: ", __func__);
 	expected_results[DROP_ERR_INNER_MAP]++;
 	do_test(type, family, &cmd, DROP_ERR_INNER_MAP);
-	printf("OK\n");
 }
 
 static void test_err_skb_data(int type, sa_family_t family)
 {
-	printf("%s: ", __func__);
 	expected_results[DROP_ERR_SKB_DATA]++;
 	do_test(type, family, NULL, DROP_ERR_SKB_DATA);
-	printf("OK\n");
 }
 
 static void test_err_sk_select_port(int type, sa_family_t family)
@@ -449,10 +461,8 @@ static void test_err_sk_select_port(int type, sa_family_t family)
 		.pass_on_failure = 0,
 	};
 
-	printf("%s: ", __func__);
 	expected_results[DROP_ERR_SK_SELECT_REUSEPORT]++;
 	do_test(type, family, &cmd, DROP_ERR_SK_SELECT_REUSEPORT);
-	printf("OK\n");
 }
 
 static void test_pass(int type, sa_family_t family)
@@ -460,14 +470,12 @@ static void test_pass(int type, sa_family_t family)
 	struct cmd cmd;
 	int i;
 
-	printf("%s: ", __func__);
 	cmd.pass_on_failure = 0;
 	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
 		expected_results[PASS]++;
 		cmd.reuseport_index = i;
 		do_test(type, family, &cmd, PASS);
 	}
-	printf("OK\n");
 }
 
 static void test_syncookie(int type, sa_family_t family)
@@ -481,7 +489,6 @@ static void test_syncookie(int type, sa_family_t family)
 	if (type != SOCK_STREAM)
 		return;
 
-	printf("%s: ", __func__);
 	/*
 	 * +1 for TCP-SYN and
 	 * +1 for the TCP-ACK (ack the syncookie)
@@ -497,17 +504,16 @@ static void test_syncookie(int type, sa_family_t family)
 	 */
 	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero,
 				  &tmp_index, BPF_ANY);
-	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, 1)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "update_elem(tmp_index_ovr_map, 0, 1)",
+	       "err:%d errno:%d\n", err, errno);
 	do_test(type, family, &cmd, PASS);
 	err = bpf_map_lookup_elem(tmp_index_ovr_map, &index_zero,
 				  &tmp_index);
-	CHECK(err == -1 || tmp_index != -1,
-	      "lookup_elem(tmp_index_ovr_map)",
-	      "err:%d errno:%d tmp_index:%d\n",
-	      err, errno, tmp_index);
+	RET_IF(err == -1 || tmp_index != -1,
+	       "lookup_elem(tmp_index_ovr_map)",
+	       "err:%d errno:%d tmp_index:%d\n",
+	       err, errno, tmp_index);
 	disable_syncookie();
-	printf("OK\n");
 }
 
 static void test_pass_on_err(int type, sa_family_t family)
@@ -517,10 +523,8 @@ static void test_pass_on_err(int type, sa_family_t family)
 		.pass_on_failure = 1,
 	};
 
-	printf("%s: ", __func__);
 	expected_results[PASS_ERR_SK_SELECT_REUSEPORT] += 1;
 	do_test(type, family, &cmd, PASS_ERR_SK_SELECT_REUSEPORT);
-	printf("OK\n");
 }
 
 static void test_detach_bpf(int type, sa_family_t family)
@@ -532,46 +536,47 @@ static void test_detach_bpf(int type, sa_family_t family)
 	struct cmd cmd = {};
 	int optvalue = 0;
 
-	printf("%s: ", __func__);
 	err = setsockopt(sk_fds[0], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
 			 &optvalue, sizeof(optvalue));
-	CHECK(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
+	       "err:%d errno:%d\n", err, errno);
 
 	err = setsockopt(sk_fds[1], SOL_SOCKET, SO_DETACH_REUSEPORT_BPF,
 			 &optvalue, sizeof(optvalue));
-	CHECK(err == 0 || errno != ENOENT, "setsockopt(SO_DETACH_REUSEPORT_BPF)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == 0 || errno != ENOENT,
+	       "setsockopt(SO_DETACH_REUSEPORT_BPF)",
+	       "err:%d errno:%d\n", err, errno);
 
 	for (i = 0; i < NR_RESULTS; i++) {
 		err = bpf_map_lookup_elem(result_map, &i, &tmp);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "lookup_elem(result_map)",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
 		nr_run_before += tmp;
 	}
 
 	cli_fd = send_data(type, family, &cmd, sizeof(cmd), PASS);
+	if (cli_fd < 0)
+		return;
 	nev = epoll_wait(epfd, &ev, 1, 5);
-	CHECK(nev <= 0, "nev <= 0",
-	      "nev:%d expected:1 type:%d family:%d data:(0, 0)\n",
-	      nev,  type, family);
+	RET_IF(nev <= 0, "nev <= 0",
+	       "nev:%d expected:1 type:%d family:%d data:(0, 0)\n",
+	       nev,  type, family);
 
 	for (i = 0; i < NR_RESULTS; i++) {
 		err = bpf_map_lookup_elem(result_map, &i, &tmp);
-		CHECK(err == -1, "lookup_elem(result_map)",
-		      "i:%u err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "lookup_elem(result_map)",
+		       "i:%u err:%d errno:%d\n", i, err, errno);
 		nr_run_after += tmp;
 	}
 
-	CHECK(nr_run_before != nr_run_after,
-	      "nr_run_before != nr_run_after",
-	      "nr_run_before:%u nr_run_after:%u\n",
-	      nr_run_before, nr_run_after);
+	RET_IF(nr_run_before != nr_run_after,
+	       "nr_run_before != nr_run_after",
+	       "nr_run_before:%u nr_run_after:%u\n",
+	       nr_run_before, nr_run_after);
 
-	printf("OK\n");
 	close(cli_fd);
 #else
-	printf("%s: SKIP\n", __func__);
+	test__skip();
 #endif
 }
 
@@ -594,73 +599,83 @@ static void prepare_sk_fds(int type, sa_family_t family, bool inany)
 	 */
 	for (i = first; i >= 0; i--) {
 		sk_fds[i] = socket(family, type, 0);
-		CHECK(sk_fds[i] == -1, "socket()", "sk_fds[%d]:%d errno:%d\n",
-		      i, sk_fds[i], errno);
+		RET_IF(sk_fds[i] == -1, "socket()", "sk_fds[%d]:%d errno:%d\n",
+		       i, sk_fds[i], errno);
 		err = setsockopt(sk_fds[i], SOL_SOCKET, SO_REUSEPORT,
 				 &optval, sizeof(optval));
-		CHECK(err == -1, "setsockopt(SO_REUSEPORT)",
-		      "sk_fds[%d] err:%d errno:%d\n",
-		      i, err, errno);
+		RET_IF(err == -1, "setsockopt(SO_REUSEPORT)",
+		       "sk_fds[%d] err:%d errno:%d\n",
+		       i, err, errno);
 
 		if (i == first) {
 			err = setsockopt(sk_fds[i], SOL_SOCKET,
 					 SO_ATTACH_REUSEPORT_EBPF,
 					 &select_by_skb_data_prog,
 					 sizeof(select_by_skb_data_prog));
-			CHECK(err == -1, "setsockopt(SO_ATTACH_REUEPORT_EBPF)",
-			      "err:%d errno:%d\n", err, errno);
+			RET_IF(err == -1, "setsockopt(SO_ATTACH_REUEPORT_EBPF)",
+			       "err:%d errno:%d\n", err, errno);
 		}
 
 		err = bind(sk_fds[i], (struct sockaddr *)&srv_sa, addrlen);
-		CHECK(err == -1, "bind()", "sk_fds[%d] err:%d errno:%d\n",
-		      i, err, errno);
+		RET_IF(err == -1, "bind()", "sk_fds[%d] err:%d errno:%d\n",
+		       i, err, errno);
 
 		if (type == SOCK_STREAM) {
 			err = listen(sk_fds[i], 10);
-			CHECK(err == -1, "listen()",
-			      "sk_fds[%d] err:%d errno:%d\n",
-			      i, err, errno);
+			RET_IF(err == -1, "listen()",
+			       "sk_fds[%d] err:%d errno:%d\n",
+			       i, err, errno);
 		}
 
 		err = bpf_map_update_elem(reuseport_array, &i, &sk_fds[i],
 					  BPF_NOEXIST);
-		CHECK(err == -1, "update_elem(reuseport_array)",
-		      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
+		RET_IF(err == -1, "update_elem(reuseport_array)",
+		       "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
 
 		if (i == first) {
 			socklen_t addrlen = sizeof(srv_sa);
 
 			err = getsockname(sk_fds[i], (struct sockaddr *)&srv_sa,
 					  &addrlen);
-			CHECK(err == -1, "getsockname()",
-			      "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
+			RET_IF(err == -1, "getsockname()",
+			       "sk_fds[%d] err:%d errno:%d\n", i, err, errno);
 		}
 	}
 
 	epfd = epoll_create(1);
-	CHECK(epfd == -1, "epoll_create(1)",
-	      "epfd:%d errno:%d\n", epfd, errno);
+	RET_IF(epfd == -1, "epoll_create(1)",
+	       "epfd:%d errno:%d\n", epfd, errno);
 
 	ev.events = EPOLLIN;
 	for (i = 0; i < REUSEPORT_ARRAY_SIZE; i++) {
 		ev.data.u32 = i;
 		err = epoll_ctl(epfd, EPOLL_CTL_ADD, sk_fds[i], &ev);
-		CHECK(err, "epoll_ctl(EPOLL_CTL_ADD)", "sk_fds[%d]\n", i);
+		RET_IF(err, "epoll_ctl(EPOLL_CTL_ADD)", "sk_fds[%d]\n", i);
 	}
 }
 
-static void setup_per_test(int type, unsigned short family, bool inany)
+static void setup_per_test(int type, sa_family_t family, bool inany,
+			   bool no_inner_map)
 {
 	int ovr = -1, err;
 
 	prepare_sk_fds(type, family, inany);
 	err = bpf_map_update_elem(tmp_index_ovr_map, &index_zero, &ovr,
 				  BPF_ANY);
-	CHECK(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "update_elem(tmp_index_ovr_map, 0, -1)",
+	       "err:%d errno:%d\n", err, errno);
+
+	/* Install reuseport_array to outer_map? */
+	if (no_inner_map)
+		return;
+
+	err = bpf_map_update_elem(outer_map, &index_zero, &reuseport_array,
+				  BPF_ANY);
+	RET_IF(err == -1, "update_elem(outer_map, 0, reuseport_array)",
+	       "err:%d errno:%d\n", err, errno);
 }
 
-static void cleanup_per_test(void)
+static void cleanup_per_test(bool no_inner_map)
 {
 	int i, err;
 
@@ -668,75 +683,124 @@ static void cleanup_per_test(void)
 		close(sk_fds[i]);
 	close(epfd);
 
+	/* Delete reuseport_array from outer_map? */
+	if (no_inner_map)
+		return;
+
 	err = bpf_map_delete_elem(outer_map, &index_zero);
-	CHECK(err == -1, "delete_elem(outer_map)",
-	      "err:%d errno:%d\n", err, errno);
+	RET_IF(err == -1, "delete_elem(outer_map)",
+	       "err:%d errno:%d\n", err, errno);
 }
 
 static void cleanup(void)
 {
-	close(outer_map);
-	close(reuseport_array);
-	bpf_object__close(obj);
+	if (outer_map != -1)
+		close(outer_map);
+	if (reuseport_array != -1)
+		close(reuseport_array);
+	if (obj)
+		bpf_object__close(obj);
 }
 
-static void test_all(void)
+static const char *family_str(sa_family_t family)
 {
-	/* Extra SOCK_STREAM to test bind_inany==true */
-	const int types[] = { SOCK_STREAM, SOCK_DGRAM, SOCK_STREAM };
-	const char * const type_strings[] = { "TCP", "UDP", "TCP" };
-	const char * const family_strings[] = { "IPv6", "IPv4" };
-	const unsigned short families[] = { AF_INET6, AF_INET };
-	const bool bind_inany[] = { false, false, true };
-	int t, f, err;
-
-	for (f = 0; f < ARRAY_SIZE(families); f++) {
-		unsigned short family = families[f];
-
-		for (t = 0; t < ARRAY_SIZE(types); t++) {
-			bool inany = bind_inany[t];
-			int type = types[t];
-
-			printf("######## %s/%s %s ########\n",
-			       family_strings[f], type_strings[t],
-				inany ? " INANY  " : "LOOPBACK");
-
-			setup_per_test(type, family, inany);
-
-			test_err_inner_map(type, family);
-
-			/* Install reuseport_array to the outer_map */
-			err = bpf_map_update_elem(outer_map, &index_zero,
-						  &reuseport_array, BPF_ANY);
-			CHECK(err == -1, "update_elem(outer_map)",
-			      "err:%d errno:%d\n", err, errno);
-
-			test_err_skb_data(type, family);
-			test_err_sk_select_port(type, family);
-			test_pass(type, family);
-			test_syncookie(type, family);
-			test_pass_on_err(type, family);
-			/* Must be the last test */
-			test_detach_bpf(type, family);
-
-			cleanup_per_test();
-			printf("\n");
-		}
+	switch (family) {
+	case AF_INET:
+		return "IPv4";
+	case AF_INET6:
+		return "IPv6";
+	default:
+		return "unknown";
+	}
+}
+
+static const char *sotype_str(int sotype)
+{
+	switch (sotype) {
+	case SOCK_STREAM:
+		return "TCP";
+	case SOCK_DGRAM:
+		return "UDP";
+	default:
+		return "unknown";
+	}
+}
+
+#define TEST_INIT(fn, ...) { fn, #fn, __VA_ARGS__ }
+
+static void test_config(int sotype, sa_family_t family, bool inany)
+{
+	const struct test {
+		void (*fn)(int sotype, sa_family_t family);
+		const char *name;
+		bool no_inner_map;
+	} tests[] = {
+		TEST_INIT(test_err_inner_map, true /* no_inner_map */),
+		TEST_INIT(test_err_skb_data),
+		TEST_INIT(test_err_sk_select_port),
+		TEST_INIT(test_pass),
+		TEST_INIT(test_syncookie),
+		TEST_INIT(test_pass_on_err),
+		TEST_INIT(test_detach_bpf),
+	};
+	char s[MAX_TEST_NAME];
+	const struct test *t;
+
+	for (t = tests; t < tests + ARRAY_SIZE(tests); t++) {
+		snprintf(s, sizeof(s), "%s/%s %s %s",
+			 family_str(family), sotype_str(sotype),
+			 inany ? "INANY" : "LOOPBACK", t->name);
+
+		if (!test__start_subtest(s))
+			continue;
+
+		setup_per_test(sotype, family, inany, t->no_inner_map);
+		t->fn(sotype, family);
+		cleanup_per_test(t->no_inner_map);
 	}
 }
 
-int main(int argc, const char **argv)
+#define BIND_INANY true
+
+static void test_all(void)
 {
-	create_maps();
-	prepare_bpf_obj();
+	const struct config {
+		int sotype;
+		sa_family_t family;
+		bool inany;
+	} configs[] = {
+		{ SOCK_STREAM, AF_INET },
+		{ SOCK_STREAM, AF_INET, BIND_INANY },
+		{ SOCK_STREAM, AF_INET6 },
+		{ SOCK_STREAM, AF_INET6, BIND_INANY },
+		{ SOCK_DGRAM, AF_INET },
+		{ SOCK_DGRAM, AF_INET6 },
+	};
+	const struct config *c;
+
+	for (c = configs; c < configs + ARRAY_SIZE(configs); c++)
+		test_config(c->sotype, c->family, c->inany);
+}
+
+void test_select_reuseport(void)
+{
+	if (create_maps())
+		goto out;
+	if (prepare_bpf_obj())
+		goto out;
+
 	saved_tcp_fo = read_int_sysctl(TCP_FO_SYSCTL);
 	saved_tcp_syncookie = read_int_sysctl(TCP_SYNCOOKIE_SYSCTL);
-	enable_fastopen();
-	disable_syncookie();
-	atexit(restore_sysctls);
+	if (saved_tcp_syncookie < 0 || saved_tcp_syncookie < 0)
+		goto out;
 
-	test_all();
+	if (enable_fastopen())
+		goto out;
+	if (disable_syncookie())
+		goto out;
 
+	test_all();
+out:
 	cleanup();
-	return 0;
+	restore_sysctls();
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
index a2eb8db8dafb..c6d6b685a946 100644
--- a/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
+++ b/tools/testing/selftests/bpf/prog_tests/skb_ctx.c
@@ -11,6 +11,9 @@ void test_skb_ctx(void)
 		.cb[4] = 5,
 		.priority = 6,
 		.tstamp = 7,
+		.wire_len = 100,
+		.gso_segs = 8,
+		.mark = 9,
 	};
 	struct bpf_prog_test_run_attr tattr = {
 		.data_in = &pkt_v4,
@@ -91,4 +94,8 @@ void test_skb_ctx(void)
 		   "ctx_out_tstamp",
 		   "skb->tstamp == %lld, expected %d\n",
 		   skb.tstamp, 8);
+	CHECK_ATTR(skb.mark != 10,
+		   "ctx_out_mark",
+		   "skb->mark == %u, expected %d\n",
+		   skb.mark, 10);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/skeleton.c b/tools/testing/selftests/bpf/prog_tests/skeleton.c
new file mode 100644
index 000000000000..9264a2736018
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/skeleton.c
@@ -0,0 +1,63 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <test_progs.h>
+
+struct s {
+	int a;
+	long long b;
+} __attribute__((packed));
+
+#include "test_skeleton.skel.h"
+
+void test_skeleton(void)
+{
+	int duration = 0, err;
+	struct test_skeleton* skel;
+	struct test_skeleton__bss *bss;
+	struct test_skeleton__kconfig *kcfg;
+
+	skel = test_skeleton__open();
+	if (CHECK(!skel, "skel_open", "failed to open skeleton\n"))
+		return;
+
+	if (CHECK(skel->kconfig, "skel_kconfig", "kconfig is mmaped()!\n"))
+		goto cleanup;
+
+	err = test_skeleton__load(skel);
+	if (CHECK(err, "skel_load", "failed to load skeleton: %d\n", err))
+		goto cleanup;
+
+	bss = skel->bss;
+	bss->in1 = 1;
+	bss->in2 = 2;
+	bss->in3 = 3;
+	bss->in4 = 4;
+	bss->in5.a = 5;
+	bss->in5.b = 6;
+	kcfg = skel->kconfig;
+
+	err = test_skeleton__attach(skel);
+	if (CHECK(err, "skel_attach", "skeleton attach failed: %d\n", err))
+		goto cleanup;
+
+	/* trigger tracepoint */
+	usleep(1);
+
+	CHECK(bss->out1 != 1, "res1", "got %d != exp %d\n", bss->out1, 1);
+	CHECK(bss->out2 != 2, "res2", "got %lld != exp %d\n", bss->out2, 2);
+	CHECK(bss->out3 != 3, "res3", "got %d != exp %d\n", (int)bss->out3, 3);
+	CHECK(bss->out4 != 4, "res4", "got %lld != exp %d\n", bss->out4, 4);
+	CHECK(bss->handler_out5.a != 5, "res5", "got %d != exp %d\n",
+	      bss->handler_out5.a, 5);
+	CHECK(bss->handler_out5.b != 6, "res6", "got %lld != exp %d\n",
+	      bss->handler_out5.b, 6);
+
+	CHECK(bss->bpf_syscall != kcfg->CONFIG_BPF_SYSCALL, "ext1",
+	      "got %d != exp %d\n", bss->bpf_syscall, kcfg->CONFIG_BPF_SYSCALL);
+	CHECK(bss->kern_ver != kcfg->LINUX_KERNEL_VERSION, "ext2",
+	      "got %d != exp %d\n", bss->kern_ver, kcfg->LINUX_KERNEL_VERSION);
+
+cleanup:
+	test_skeleton__destroy(skel);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
index d841dced971f..e8399ae50e77 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id.c
@@ -1,16 +1,14 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_stacktrace_build_id.skel.h"
 
 void test_stacktrace_build_id(void)
 {
+
 	int control_map_fd, stackid_hmap_fd, stackmap_fd, stack_amap_fd;
-	const char *prog_name = "tracepoint/random/urandom_read";
-	const char *file = "./test_stacktrace_build_id.o";
-	int err, prog_fd, stack_trace_len;
+	struct test_stacktrace_build_id *skel;
+	int err, stack_trace_len;
 	__u32 key, previous_key, val, duration = 0;
-	struct bpf_program *prog;
-	struct bpf_object *obj;
-	struct bpf_link *link = NULL;
 	char buf[256];
 	int i, j;
 	struct bpf_stack_build_id id_offs[PERF_MAX_STACK_DEPTH];
@@ -18,43 +16,24 @@ void test_stacktrace_build_id(void)
 	int retry = 1;
 
 retry:
-	err = bpf_prog_load(file, BPF_PROG_TYPE_TRACEPOINT, &obj, &prog_fd);
-	if (CHECK(err, "prog_load", "err %d errno %d\n", err, errno))
+	skel = test_stacktrace_build_id__open_and_load();
+	if (CHECK(!skel, "skel_open_and_load", "skeleton open/load failed\n"))
 		return;
 
-	prog = bpf_object__find_program_by_title(obj, prog_name);
-	if (CHECK(!prog, "find_prog", "prog '%s' not found\n", prog_name))
-		goto close_prog;
-
-	link = bpf_program__attach_tracepoint(prog, "random", "urandom_read");
-	if (CHECK(IS_ERR(link), "attach_tp", "err %ld\n", PTR_ERR(link)))
-		goto close_prog;
+	err = test_stacktrace_build_id__attach(skel);
+	if (CHECK(err, "attach_tp", "err %d\n", err))
+		goto cleanup;
 
 	/* find map fds */
-	control_map_fd = bpf_find_map(__func__, obj, "control_map");
-	if (CHECK(control_map_fd < 0, "bpf_find_map control_map",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackid_hmap_fd = bpf_find_map(__func__, obj, "stackid_hmap");
-	if (CHECK(stackid_hmap_fd < 0, "bpf_find_map stackid_hmap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackmap_fd = bpf_find_map(__func__, obj, "stackmap");
-	if (CHECK(stackmap_fd < 0, "bpf_find_map stackmap", "err %d errno %d\n",
-		  err, errno))
-		goto disable_pmu;
-
-	stack_amap_fd = bpf_find_map(__func__, obj, "stack_amap");
-	if (CHECK(stack_amap_fd < 0, "bpf_find_map stack_amap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+	control_map_fd = bpf_map__fd(skel->maps.control_map);
+	stackid_hmap_fd = bpf_map__fd(skel->maps.stackid_hmap);
+	stackmap_fd = bpf_map__fd(skel->maps.stackmap);
+	stack_amap_fd = bpf_map__fd(skel->maps.stack_amap);
 
 	if (CHECK_FAIL(system("dd if=/dev/urandom of=/dev/zero count=4 2> /dev/null")))
-		goto disable_pmu;
+		goto cleanup;
 	if (CHECK_FAIL(system("./urandom_read")))
-		goto disable_pmu;
+		goto cleanup;
 	/* disable stack trace collection */
 	key = 0;
 	val = 1;
@@ -66,23 +45,23 @@ retry:
 	err = compare_map_keys(stackid_hmap_fd, stackmap_fd);
 	if (CHECK(err, "compare_map_keys stackid_hmap vs. stackmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = compare_map_keys(stackmap_fd, stackid_hmap_fd);
 	if (CHECK(err, "compare_map_keys stackmap vs. stackid_hmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = extract_build_id(buf, 256);
 
 	if (CHECK(err, "get build_id with readelf",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = bpf_map_get_next_key(stackmap_fd, NULL, &key);
 	if (CHECK(err, "get_next_key from stackmap",
 		  "err %d, errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	do {
 		char build_id[64];
@@ -90,7 +69,7 @@ retry:
 		err = bpf_map_lookup_elem(stackmap_fd, &key, id_offs);
 		if (CHECK(err, "lookup_elem from stackmap",
 			  "err %d, errno %d\n", err, errno))
-			goto disable_pmu;
+			goto cleanup;
 		for (i = 0; i < PERF_MAX_STACK_DEPTH; ++i)
 			if (id_offs[i].status == BPF_STACK_BUILD_ID_VALID &&
 			    id_offs[i].offset != 0) {
@@ -108,8 +87,7 @@ retry:
 	 * try it one more time.
 	 */
 	if (build_id_matches < 1 && retry--) {
-		bpf_link__destroy(link);
-		bpf_object__close(obj);
+		test_stacktrace_build_id__destroy(skel);
 		printf("%s:WARN:Didn't find expected build ID from the map, retrying\n",
 		       __func__);
 		goto retry;
@@ -117,17 +95,14 @@ retry:
 
 	if (CHECK(build_id_matches < 1, "build id match",
 		  "Didn't find expected build ID from the map\n"))
-		goto disable_pmu;
+		goto cleanup;
 
-	stack_trace_len = PERF_MAX_STACK_DEPTH
-		* sizeof(struct bpf_stack_build_id);
+	stack_trace_len = PERF_MAX_STACK_DEPTH *
+			  sizeof(struct bpf_stack_build_id);
 	err = compare_stack_ips(stackmap_fd, stack_amap_fd, stack_trace_len);
 	CHECK(err, "compare_stack_ips stackmap vs. stack_amap",
 	      "err %d errno %d\n", err, errno);
 
-disable_pmu:
-	bpf_link__destroy(link);
-
-close_prog:
-	bpf_object__close(obj);
+cleanup:
+	test_stacktrace_build_id__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
index f62aa0eb959b..8974450a4bdb 100644
--- a/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
+++ b/tools/testing/selftests/bpf/prog_tests/stacktrace_build_id_nmi.c
@@ -1,5 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <test_progs.h>
+#include "test_stacktrace_build_id.skel.h"
 
 static __u64 read_perf_max_sample_freq(void)
 {
@@ -16,19 +17,15 @@ static __u64 read_perf_max_sample_freq(void)
 
 void test_stacktrace_build_id_nmi(void)
 {
-	int control_map_fd, stackid_hmap_fd, stackmap_fd, stack_amap_fd;
-	const char *prog_name = "tracepoint/random/urandom_read";
-	const char *file = "./test_stacktrace_build_id.o";
-	int err, pmu_fd, prog_fd;
+	int control_map_fd, stackid_hmap_fd, stackmap_fd;
+	struct test_stacktrace_build_id *skel;
+	int err, pmu_fd;
 	struct perf_event_attr attr = {
 		.freq = 1,
 		.type = PERF_TYPE_HARDWARE,
 		.config = PERF_COUNT_HW_CPU_CYCLES,
 	};
 	__u32 key, previous_key, val, duration = 0;
-	struct bpf_program *prog;
-	struct bpf_object *obj;
-	struct bpf_link *link;
 	char buf[256];
 	int i, j;
 	struct bpf_stack_build_id id_offs[PERF_MAX_STACK_DEPTH];
@@ -38,13 +35,16 @@ void test_stacktrace_build_id_nmi(void)
 	attr.sample_freq = read_perf_max_sample_freq();
 
 retry:
-	err = bpf_prog_load(file, BPF_PROG_TYPE_PERF_EVENT, &obj, &prog_fd);
-	if (CHECK(err, "prog_load", "err %d errno %d\n", err, errno))
+	skel = test_stacktrace_build_id__open();
+	if (CHECK(!skel, "skel_open", "skeleton open failed\n"))
 		return;
 
-	prog = bpf_object__find_program_by_title(obj, prog_name);
-	if (CHECK(!prog, "find_prog", "prog '%s' not found\n", prog_name))
-		goto close_prog;
+	/* override program type */
+	bpf_program__set_perf_event(skel->progs.oncpu);
+
+	err = test_stacktrace_build_id__load(skel);
+	if (CHECK(err, "skel_load", "skeleton load failed: %d\n", err))
+		goto cleanup;
 
 	pmu_fd = syscall(__NR_perf_event_open, &attr, -1 /* pid */,
 			 0 /* cpu 0 */, -1 /* group id */,
@@ -52,40 +52,25 @@ retry:
 	if (CHECK(pmu_fd < 0, "perf_event_open",
 		  "err %d errno %d. Does the test host support PERF_COUNT_HW_CPU_CYCLES?\n",
 		  pmu_fd, errno))
-		goto close_prog;
+		goto cleanup;
 
-	link = bpf_program__attach_perf_event(prog, pmu_fd);
-	if (CHECK(IS_ERR(link), "attach_perf_event",
-		  "err %ld\n", PTR_ERR(link))) {
+	skel->links.oncpu = bpf_program__attach_perf_event(skel->progs.oncpu,
+							   pmu_fd);
+	if (CHECK(IS_ERR(skel->links.oncpu), "attach_perf_event",
+		  "err %ld\n", PTR_ERR(skel->links.oncpu))) {
 		close(pmu_fd);
-		goto close_prog;
+		goto cleanup;
 	}
 
 	/* find map fds */
-	control_map_fd = bpf_find_map(__func__, obj, "control_map");
-	if (CHECK(control_map_fd < 0, "bpf_find_map control_map",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackid_hmap_fd = bpf_find_map(__func__, obj, "stackid_hmap");
-	if (CHECK(stackid_hmap_fd < 0, "bpf_find_map stackid_hmap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
-
-	stackmap_fd = bpf_find_map(__func__, obj, "stackmap");
-	if (CHECK(stackmap_fd < 0, "bpf_find_map stackmap", "err %d errno %d\n",
-		  err, errno))
-		goto disable_pmu;
-
-	stack_amap_fd = bpf_find_map(__func__, obj, "stack_amap");
-	if (CHECK(stack_amap_fd < 0, "bpf_find_map stack_amap",
-		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+	control_map_fd = bpf_map__fd(skel->maps.control_map);
+	stackid_hmap_fd = bpf_map__fd(skel->maps.stackid_hmap);
+	stackmap_fd = bpf_map__fd(skel->maps.stackmap);
 
 	if (CHECK_FAIL(system("dd if=/dev/urandom of=/dev/zero count=4 2> /dev/null")))
-		goto disable_pmu;
+		goto cleanup;
 	if (CHECK_FAIL(system("taskset 0x1 ./urandom_read 100000")))
-		goto disable_pmu;
+		goto cleanup;
 	/* disable stack trace collection */
 	key = 0;
 	val = 1;
@@ -97,23 +82,23 @@ retry:
 	err = compare_map_keys(stackid_hmap_fd, stackmap_fd);
 	if (CHECK(err, "compare_map_keys stackid_hmap vs. stackmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = compare_map_keys(stackmap_fd, stackid_hmap_fd);
 	if (CHECK(err, "compare_map_keys stackmap vs. stackid_hmap",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = extract_build_id(buf, 256);
 
 	if (CHECK(err, "get build_id with readelf",
 		  "err %d errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	err = bpf_map_get_next_key(stackmap_fd, NULL, &key);
 	if (CHECK(err, "get_next_key from stackmap",
 		  "err %d, errno %d\n", err, errno))
-		goto disable_pmu;
+		goto cleanup;
 
 	do {
 		char build_id[64];
@@ -121,7 +106,7 @@ retry:
 		err = bpf_map_lookup_elem(stackmap_fd, &key, id_offs);
 		if (CHECK(err, "lookup_elem from stackmap",
 			  "err %d, errno %d\n", err, errno))
-			goto disable_pmu;
+			goto cleanup;
 		for (i = 0; i < PERF_MAX_STACK_DEPTH; ++i)
 			if (id_offs[i].status == BPF_STACK_BUILD_ID_VALID &&
 			    id_offs[i].offset != 0) {
@@ -139,8 +124,7 @@ retry:
 	 * try it one more time.
 	 */
 	if (build_id_matches < 1 && retry--) {
-		bpf_link__destroy(link);
-		bpf_object__close(obj);
+		test_stacktrace_build_id__destroy(skel);
 		printf("%s:WARN:Didn't find expected build ID from the map, retrying\n",
 		       __func__);
 		goto retry;
@@ -148,7 +132,7 @@ retry:
 
 	if (CHECK(build_id_matches < 1, "build id match",
 		  "Didn't find expected build ID from the map\n"))
-		goto disable_pmu;
+		goto cleanup;
 
 	/*
 	 * We intentionally skip compare_stack_ips(). This is because we
@@ -157,8 +141,6 @@ retry:
 	 * BPF_STACK_BUILD_ID_IP;
 	 */
 
-disable_pmu:
-	bpf_link__destroy(link);
-close_prog:
-	bpf_object__close(obj);
+cleanup:
+	test_stacktrace_build_id__destroy(skel);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_perf.c b/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
new file mode 100644
index 000000000000..7185bee16fe4
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/xdp_perf.c
@@ -0,0 +1,25 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+
+void test_xdp_perf(void)
+{
+	const char *file = "./xdp_dummy.o";
+	__u32 duration, retval, size;
+	struct bpf_object *obj;
+	char in[128], out[128];
+	int err, prog_fd;
+
+	err = bpf_prog_load(file, BPF_PROG_TYPE_XDP, &obj, &prog_fd);
+	if (CHECK_FAIL(err))
+		return;
+
+	err = bpf_prog_test_run(prog_fd, 1000000, &in[0], 128,
+				out, &size, &retval, &duration);
+
+	CHECK(err || retval != XDP_PASS || size != 128,
+	      "xdp-perf",
+	      "err %d errno %d retval %d size %d\n",
+	      err, errno, retval, size);
+
+	bpf_object__close(obj);
+}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c
new file mode 100644
index 000000000000..65eac371b061
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___equiv_zero_sz_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___equiv_zero_sz_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c
new file mode 100644
index 000000000000..ecda2b545ac2
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___err_bad_zero_sz_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___err_bad_zero_sz_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c
new file mode 100644
index 000000000000..fe1d01232c22
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/btf__core_reloc_arrays___fixed_arr.c
@@ -0,0 +1,3 @@
+#include "core_reloc_types.h"
+
+void f(struct core_reloc_arrays___fixed_arr x) {}
diff --git a/tools/testing/selftests/bpf/progs/core_reloc_types.h b/tools/testing/selftests/bpf/progs/core_reloc_types.h
index 9311489e14b2..6d598cfbdb3e 100644
--- a/tools/testing/selftests/bpf/progs/core_reloc_types.h
+++ b/tools/testing/selftests/bpf/progs/core_reloc_types.h
@@ -327,6 +327,7 @@ struct core_reloc_arrays_output {
 	char b123;
 	int c1c;
 	int d00d;
+	int f10c;
 };
 
 struct core_reloc_arrays_substruct {
@@ -339,6 +340,7 @@ struct core_reloc_arrays {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 /* bigger array dimensions */
@@ -347,6 +349,7 @@ struct core_reloc_arrays___diff_arr_dim {
 	char b[3][4][5];
 	struct core_reloc_arrays_substruct c[4];
 	struct core_reloc_arrays_substruct d[2][3];
+	struct core_reloc_arrays_substruct f[1][3];
 };
 
 /* different size of array's value (struct) */
@@ -363,6 +366,29 @@ struct core_reloc_arrays___diff_arr_val_sz {
 		int d;
 		int __padding2;
 	} d[1][2];
+	struct {
+		int __padding1;
+		int c;
+		int __padding2;
+	} f[][2];
+};
+
+struct core_reloc_arrays___equiv_zero_sz_arr {
+	int a[5];
+	char b[2][3][4];
+	struct core_reloc_arrays_substruct c[3];
+	struct core_reloc_arrays_substruct d[1][2];
+	/* equivalent to flexible array */
+	struct core_reloc_arrays_substruct f[0][2];
+};
+
+struct core_reloc_arrays___fixed_arr {
+	int a[5];
+	char b[2][3][4];
+	struct core_reloc_arrays_substruct c[3];
+	struct core_reloc_arrays_substruct d[1][2];
+	/* not a flexible array anymore, but within access bounds */
+	struct core_reloc_arrays_substruct f[1][2];
 };
 
 struct core_reloc_arrays___err_too_small {
@@ -370,6 +396,7 @@ struct core_reloc_arrays___err_too_small {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 struct core_reloc_arrays___err_too_shallow {
@@ -377,6 +404,7 @@ struct core_reloc_arrays___err_too_shallow {
 	char b[2][3]; /* this one lacks one dimension */
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 struct core_reloc_arrays___err_non_array {
@@ -384,6 +412,7 @@ struct core_reloc_arrays___err_non_array {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 struct core_reloc_arrays___err_wrong_val_type {
@@ -391,6 +420,16 @@ struct core_reloc_arrays___err_wrong_val_type {
 	char b[2][3][4];
 	int c[3]; /* value is not a struct */
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
+};
+
+struct core_reloc_arrays___err_bad_zero_sz_arr {
+	/* zero-sized array, but not at the end */
+	struct core_reloc_arrays_substruct f[0][2];
+	int a[5];
+	char b[2][3][4];
+	struct core_reloc_arrays_substruct c[3];
+	struct core_reloc_arrays_substruct d[1][2];
 };
 
 /*
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index 534621e38906..221b69700625 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -5,46 +5,36 @@
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
 
-struct {
-	__uint(type, BPF_MAP_TYPE_ARRAY);
-	__uint(max_entries, 4);
-	__type(key, int);
-	__type(value, int);
-} results_map SEC(".maps");
+int kprobe_res = 0;
+int kretprobe_res = 0;
+int uprobe_res = 0;
+int uretprobe_res = 0;
 
 SEC("kprobe/sys_nanosleep")
-int handle_sys_nanosleep_entry(struct pt_regs *ctx)
+int handle_kprobe(struct pt_regs *ctx)
 {
-	const int key = 0, value = 1;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	kprobe_res = 1;
 	return 0;
 }
 
 SEC("kretprobe/sys_nanosleep")
-int handle_sys_getpid_return(struct pt_regs *ctx)
+int handle_kretprobe(struct pt_regs *ctx)
 {
-	const int key = 1, value = 2;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	kretprobe_res = 2;
 	return 0;
 }
 
 SEC("uprobe/trigger_func")
-int handle_uprobe_entry(struct pt_regs *ctx)
+int handle_uprobe(struct pt_regs *ctx)
 {
-	const int key = 2, value = 3;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	uprobe_res = 3;
 	return 0;
 }
 
 SEC("uretprobe/trigger_func")
-int handle_uprobe_return(struct pt_regs *ctx)
+int handle_uretprobe(struct pt_regs *ctx)
 {
-	const int key = 3, value = 4;
-
-	bpf_map_update_elem(&results_map, &key, &value, 0);
+	uretprobe_res = 4;
 	return 0;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/test_core_extern.c b/tools/testing/selftests/bpf/progs/test_core_extern.c
new file mode 100644
index 000000000000..9bfc91d9d004
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_core_extern.c
@@ -0,0 +1,62 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <stdint.h>
+#include <stdbool.h>
+#include <linux/ptrace.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+/* non-existing BPF helper, to test dead code elimination */
+static int (*bpf_missing_helper)(const void *arg1, int arg2) = (void *) 999;
+
+extern int LINUX_KERNEL_VERSION __kconfig;
+extern bool CONFIG_BPF_SYSCALL __kconfig; /* strong */
+extern enum libbpf_tristate CONFIG_TRISTATE __kconfig __weak;
+extern bool CONFIG_BOOL __kconfig __weak;
+extern char CONFIG_CHAR __kconfig __weak;
+extern uint16_t CONFIG_USHORT __kconfig __weak;
+extern int CONFIG_INT __kconfig __weak;
+extern uint64_t CONFIG_ULONG __kconfig __weak;
+extern const char CONFIG_STR[8] __kconfig __weak;
+extern uint64_t CONFIG_MISSING __kconfig __weak;
+
+uint64_t kern_ver = -1;
+uint64_t bpf_syscall = -1;
+uint64_t tristate_val = -1;
+uint64_t bool_val = -1;
+uint64_t char_val = -1;
+uint64_t ushort_val = -1;
+uint64_t int_val = -1;
+uint64_t ulong_val = -1;
+char str_val[8] = {-1, -1, -1, -1, -1, -1, -1, -1};
+uint64_t missing_val = -1;
+
+SEC("raw_tp/sys_enter")
+int handle_sys_enter(struct pt_regs *ctx)
+{
+	int i;
+
+	kern_ver = LINUX_KERNEL_VERSION;
+	bpf_syscall = CONFIG_BPF_SYSCALL;
+	tristate_val = CONFIG_TRISTATE;
+	bool_val = CONFIG_BOOL;
+	char_val = CONFIG_CHAR;
+	ushort_val = CONFIG_USHORT;
+	int_val = CONFIG_INT;
+	ulong_val = CONFIG_ULONG;
+
+	for (i = 0; i < sizeof(CONFIG_STR); i++) {
+		str_val[i] = CONFIG_STR[i];
+	}
+
+	if (CONFIG_MISSING)
+		/* invalid, but dead code - never executed */
+		missing_val = bpf_missing_helper(ctx, 123);
+	else
+		missing_val = 0xDEADC0DE;
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
index 89951b684282..053b86f6b53f 100644
--- a/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
+++ b/tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c
@@ -18,6 +18,7 @@ struct core_reloc_arrays_output {
 	char b123;
 	int c1c;
 	int d00d;
+	int f01c;
 };
 
 struct core_reloc_arrays_substruct {
@@ -30,6 +31,7 @@ struct core_reloc_arrays {
 	char b[2][3][4];
 	struct core_reloc_arrays_substruct c[3];
 	struct core_reloc_arrays_substruct d[1][2];
+	struct core_reloc_arrays_substruct f[][2];
 };
 
 #define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
@@ -40,18 +42,16 @@ int test_core_arrays(void *ctx)
 	struct core_reloc_arrays *in = (void *)&data.in;
 	struct core_reloc_arrays_output *out = (void *)&data.out;
 
-	/* in->a[2] */
 	if (CORE_READ(&out->a2, &in->a[2]))
 		return 1;
-	/* in->b[1][2][3] */
 	if (CORE_READ(&out->b123, &in->b[1][2][3]))
 		return 1;
-	/* in->c[1].c */
 	if (CORE_READ(&out->c1c, &in->c[1].c))
 		return 1;
-	/* in->d[0][0].d */
 	if (CORE_READ(&out->d00d, &in->d[0][0].d))
 		return 1;
+	if (CORE_READ(&out->f01c, &in->f[0][1].c))
+		return 1;
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
index ea7d84f01235..b1f09f5bb1cf 100644
--- a/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
+++ b/tools/testing/selftests/bpf/progs/test_select_reuseport_kern.c
@@ -62,7 +62,7 @@ struct {
 	goto done;				\
 })
 
-SEC("select_by_skb_data")
+SEC("sk_reuseport")
 int _select_by_skb_data(struct sk_reuseport_md *reuse_md)
 {
 	__u32 linum, index = 0, flags = 0, index_zero = 0;
diff --git a/tools/testing/selftests/bpf/progs/test_skb_ctx.c b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
index 2a9f4c736ebc..e18da87fe84f 100644
--- a/tools/testing/selftests/bpf/progs/test_skb_ctx.c
+++ b/tools/testing/selftests/bpf/progs/test_skb_ctx.c
@@ -17,6 +17,12 @@ int process(struct __sk_buff *skb)
 	}
 	skb->priority++;
 	skb->tstamp++;
+	skb->mark++;
+
+	if (skb->wire_len != 100)
+		return 1;
+	if (skb->gso_segs != 8)
+		return 1;
 
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/test_skeleton.c b/tools/testing/selftests/bpf/progs/test_skeleton.c
new file mode 100644
index 000000000000..4f69aac5635f
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_skeleton.c
@@ -0,0 +1,46 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <stdbool.h>
+#include <linux/bpf.h>
+#include "bpf_helpers.h"
+
+struct s {
+	int a;
+	long long b;
+} __attribute__((packed));
+
+int in1 = 0;
+long long in2 = 0;
+char in3 = '\0';
+long long in4 __attribute__((aligned(64))) = 0;
+struct s in5 = {};
+
+long long out2 = 0;
+char out3 = 0;
+long long out4 = 0;
+int out1 = 0;
+
+extern bool CONFIG_BPF_SYSCALL __kconfig;
+extern int LINUX_KERNEL_VERSION __kconfig;
+bool bpf_syscall = 0;
+int kern_ver = 0;
+
+SEC("raw_tp/sys_enter")
+int handler(const void *ctx)
+{
+	static volatile struct s out5;
+
+	out1 = in1;
+	out2 = in2;
+	out3 = in3;
+	out4 = in4;
+	out5 = in5;
+
+	bpf_syscall = CONFIG_BPF_SYSCALL;
+	kern_ver = LINUX_KERNEL_VERSION;
+
+	return 0;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/test_cgroup_attach.c b/tools/testing/selftests/bpf/test_cgroup_attach.c
deleted file mode 100644
index 7671909ee1cb..000000000000
--- a/tools/testing/selftests/bpf/test_cgroup_attach.c
+++ /dev/null
@@ -1,571 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-
-/* eBPF example program:
- *
- * - Creates arraymap in kernel with 4 bytes keys and 8 byte values
- *
- * - Loads eBPF program
- *
- *   The eBPF program accesses the map passed in to store two pieces of
- *   information. The number of invocations of the program, which maps
- *   to the number of packets received, is stored to key 0. Key 1 is
- *   incremented on each iteration by the number of bytes stored in
- *   the skb. The program also stores the number of received bytes
- *   in the cgroup storage.
- *
- * - Attaches the new program to a cgroup using BPF_PROG_ATTACH
- *
- * - Every second, reads map[0] and map[1] to see how many bytes and
- *   packets were seen on any socket of tasks in the given cgroup.
- */
-
-#define _GNU_SOURCE
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <assert.h>
-#include <sys/resource.h>
-#include <sys/time.h>
-#include <unistd.h>
-#include <linux/filter.h>
-
-#include <linux/bpf.h>
-#include <bpf/bpf.h>
-
-#include "bpf_util.h"
-#include "bpf_rlimit.h"
-#include "cgroup_helpers.h"
-
-#define FOO		"/foo"
-#define BAR		"/foo/bar/"
-#define PING_CMD	"ping -q -c1 -w1 127.0.0.1 > /dev/null"
-
-char bpf_log_buf[BPF_LOG_BUF_SIZE];
-
-#ifdef DEBUG
-#define debug(args...) printf(args)
-#else
-#define debug(args...)
-#endif
-
-static int prog_load(int verdict)
-{
-	int ret;
-	struct bpf_insn prog[] = {
-		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
-		BPF_EXIT_INSN(),
-	};
-	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
-
-	ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
-			       prog, insns_cnt, "GPL", 0,
-			       bpf_log_buf, BPF_LOG_BUF_SIZE);
-
-	if (ret < 0) {
-		log_err("Loading program");
-		printf("Output from verifier:\n%s\n-------\n", bpf_log_buf);
-		return 0;
-	}
-	return ret;
-}
-
-static int test_foo_bar(void)
-{
-	int drop_prog, allow_prog, foo = 0, bar = 0, rc = 0;
-
-	allow_prog = prog_load(1);
-	if (!allow_prog)
-		goto err;
-
-	drop_prog = prog_load(0);
-	if (!drop_prog)
-		goto err;
-
-	if (setup_cgroup_environment())
-		goto err;
-
-	/* Create cgroup /foo, get fd, and join it */
-	foo = create_and_get_cgroup(FOO);
-	if (foo < 0)
-		goto err;
-
-	if (join_cgroup(FOO))
-		goto err;
-
-	if (bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo");
-		goto err;
-	}
-
-	debug("Attached DROP prog. This ping in cgroup /foo should fail...\n");
-	assert(system(PING_CMD) != 0);
-
-	/* Create cgroup /foo/bar, get fd, and join it */
-	bar = create_and_get_cgroup(BAR);
-	if (bar < 0)
-		goto err;
-
-	if (join_cgroup(BAR))
-		goto err;
-
-	debug("Attached DROP prog. This ping in cgroup /foo/bar should fail...\n");
-	assert(system(PING_CMD) != 0);
-
-	if (bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo/bar");
-		goto err;
-	}
-
-	debug("Attached PASS prog. This ping in cgroup /foo/bar should pass...\n");
-	assert(system(PING_CMD) == 0);
-
-	if (bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching program from /foo/bar");
-		goto err;
-	}
-
-	debug("Detached PASS from /foo/bar while DROP is attached to /foo.\n"
-	       "This ping in cgroup /foo/bar should fail...\n");
-	assert(system(PING_CMD) != 0);
-
-	if (bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo/bar");
-		goto err;
-	}
-
-	if (bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching program from /foo");
-		goto err;
-	}
-
-	debug("Attached PASS from /foo/bar and detached DROP from /foo.\n"
-	       "This ping in cgroup /foo/bar should pass...\n");
-	assert(system(PING_CMD) == 0);
-
-	if (bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0)) {
-		errno = 0;
-		log_err("Unexpected success attaching prog to /foo/bar");
-		goto err;
-	}
-
-	if (bpf_prog_detach(bar, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching program from /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_detach(foo, BPF_CGROUP_INET_EGRESS)) {
-		errno = 0;
-		log_err("Unexpected success in double detach from /foo");
-		goto err;
-	}
-
-	if (bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS, 0)) {
-		log_err("Attaching non-overridable prog to /foo");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS, 0)) {
-		errno = 0;
-		log_err("Unexpected success attaching non-overridable prog to /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, bar, BPF_CGROUP_INET_EGRESS,
-			     BPF_F_ALLOW_OVERRIDE)) {
-		errno = 0;
-		log_err("Unexpected success attaching overridable prog to /foo/bar");
-		goto err;
-	}
-
-	if (!bpf_prog_attach(allow_prog, foo, BPF_CGROUP_INET_EGRESS,
-			     BPF_F_ALLOW_OVERRIDE)) {
-		errno = 0;
-		log_err("Unexpected success attaching overridable prog to /foo");
-		goto err;
-	}
-
-	if (bpf_prog_attach(drop_prog, foo, BPF_CGROUP_INET_EGRESS, 0)) {
-		log_err("Attaching different non-overridable prog to /foo");
-		goto err;
-	}
-
-	goto out;
-
-err:
-	rc = 1;
-
-out:
-	close(foo);
-	close(bar);
-	cleanup_cgroup_environment();
-	if (!rc)
-		printf("#override:PASS\n");
-	else
-		printf("#override:FAIL\n");
-	return rc;
-}
-
-static int map_fd = -1;
-
-static int prog_load_cnt(int verdict, int val)
-{
-	int cgroup_storage_fd, percpu_cgroup_storage_fd;
-
-	if (map_fd < 0)
-		map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, 4, 8, 1, 0);
-	if (map_fd < 0) {
-		printf("failed to create map '%s'\n", strerror(errno));
-		return -1;
-	}
-
-	cgroup_storage_fd = bpf_create_map(BPF_MAP_TYPE_CGROUP_STORAGE,
-				sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
-	if (cgroup_storage_fd < 0) {
-		printf("failed to create map '%s'\n", strerror(errno));
-		return -1;
-	}
-
-	percpu_cgroup_storage_fd = bpf_create_map(
-		BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE,
-		sizeof(struct bpf_cgroup_storage_key), 8, 0, 0);
-	if (percpu_cgroup_storage_fd < 0) {
-		printf("failed to create map '%s'\n", strerror(errno));
-		return -1;
-	}
-
-	struct bpf_insn prog[] = {
-		BPF_MOV32_IMM(BPF_REG_0, 0),
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), /* *(u32 *)(fp - 4) = r0 */
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), /* r2 = fp - 4 */
-		BPF_LD_MAP_FD(BPF_REG_1, map_fd),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
-		BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
-		BPF_MOV64_IMM(BPF_REG_1, val), /* r1 = 1 */
-		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_DW, BPF_REG_0, BPF_REG_1, 0, 0), /* xadd r0 += r1 */
-
-		BPF_LD_MAP_FD(BPF_REG_1, cgroup_storage_fd),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-		BPF_MOV64_IMM(BPF_REG_1, val),
-		BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_W, BPF_REG_0, BPF_REG_1, 0, 0),
-
-		BPF_LD_MAP_FD(BPF_REG_1, percpu_cgroup_storage_fd),
-		BPF_MOV64_IMM(BPF_REG_2, 0),
-		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_get_local_storage),
-		BPF_LDX_MEM(BPF_W, BPF_REG_3, BPF_REG_0, 0),
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_3, 0x1),
-		BPF_STX_MEM(BPF_W, BPF_REG_0, BPF_REG_3, 0),
-
-		BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
-		BPF_EXIT_INSN(),
-	};
-	size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
-	int ret;
-
-	ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
-			       prog, insns_cnt, "GPL", 0,
-			       bpf_log_buf, BPF_LOG_BUF_SIZE);
-
-	if (ret < 0) {
-		log_err("Loading program");
-		printf("Output from verifier:\n%s\n-------\n", bpf_log_buf);
-		return 0;
-	}
-	close(cgroup_storage_fd);
-	return ret;
-}
-
-
-static int test_multiprog(void)
-{
-	__u32 prog_ids[4], prog_cnt = 0, attach_flags, saved_prog_id;
-	int cg1 = 0, cg2 = 0, cg3 = 0, cg4 = 0, cg5 = 0, key = 0;
-	int drop_prog, allow_prog[6] = {}, rc = 0;
-	unsigned long long value;
-	int i = 0;
-
-	for (i = 0; i < 6; i++) {
-		allow_prog[i] = prog_load_cnt(1, 1 << i);
-		if (!allow_prog[i])
-			goto err;
-	}
-	drop_prog = prog_load_cnt(0, 1);
-	if (!drop_prog)
-		goto err;
-
-	if (setup_cgroup_environment())
-		goto err;
-
-	cg1 = create_and_get_cgroup("/cg1");
-	if (cg1 < 0)
-		goto err;
-	cg2 = create_and_get_cgroup("/cg1/cg2");
-	if (cg2 < 0)
-		goto err;
-	cg3 = create_and_get_cgroup("/cg1/cg2/cg3");
-	if (cg3 < 0)
-		goto err;
-	cg4 = create_and_get_cgroup("/cg1/cg2/cg3/cg4");
-	if (cg4 < 0)
-		goto err;
-	cg5 = create_and_get_cgroup("/cg1/cg2/cg3/cg4/cg5");
-	if (cg5 < 0)
-		goto err;
-
-	if (join_cgroup("/cg1/cg2/cg3/cg4/cg5"))
-		goto err;
-
-	if (bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_MULTI)) {
-		log_err("Attaching prog to cg1");
-		goto err;
-	}
-	if (!bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS,
-			     BPF_F_ALLOW_MULTI)) {
-		log_err("Unexpected success attaching the same prog to cg1");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[1], cg1, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_MULTI)) {
-		log_err("Attaching prog2 to cg1");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[2], cg2, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to cg2");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_MULTI)) {
-		log_err("Attaching prog to cg3");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[4], cg4, BPF_CGROUP_INET_EGRESS,
-			    BPF_F_ALLOW_OVERRIDE)) {
-		log_err("Attaching prog to cg4");
-		goto err;
-	}
-	if (bpf_prog_attach(allow_prog[5], cg5, BPF_CGROUP_INET_EGRESS, 0)) {
-		log_err("Attaching prog to cg5");
-		goto err;
-	}
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 8 + 32);
-
-	/* query the number of effective progs in cg5 */
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      NULL, NULL, &prog_cnt) == 0);
-	assert(prog_cnt == 4);
-	/* retrieve prog_ids of effective progs in cg5 */
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      &attach_flags, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 4);
-	assert(attach_flags == 0);
-	saved_prog_id = prog_ids[0];
-	/* check enospc handling */
-	prog_ids[0] = 0;
-	prog_cnt = 2;
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      &attach_flags, prog_ids, &prog_cnt) == -1 &&
-	       errno == ENOSPC);
-	assert(prog_cnt == 4);
-	/* check that prog_ids are returned even when buffer is too small */
-	assert(prog_ids[0] == saved_prog_id);
-	/* retrieve prog_id of single attached prog in cg5 */
-	prog_ids[0] = 0;
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0,
-			      NULL, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 1);
-	assert(prog_ids[0] == saved_prog_id);
-
-	/* detach bottom program and ping again */
-	if (bpf_prog_detach2(-1, cg5, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching prog from cg5");
-		goto err;
-	}
-	value = 0;
-	assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 8 + 16);
-
-	/* detach 3rd from bottom program and ping again */
-	errno = 0;
-	if (!bpf_prog_detach2(0, cg3, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Unexpected success on detach from cg3");
-		goto err;
-	}
-	if (bpf_prog_detach2(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching from cg3");
-		goto err;
-	}
-	value = 0;
-	assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 16);
-
-	/* detach 2nd from bottom program and ping again */
-	if (bpf_prog_detach2(-1, cg4, BPF_CGROUP_INET_EGRESS)) {
-		log_err("Detaching prog from cg4");
-		goto err;
-	}
-	value = 0;
-	assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
-	assert(system(PING_CMD) == 0);
-	assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
-	assert(value == 1 + 2 + 4);
-
-	prog_cnt = 4;
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, BPF_F_QUERY_EFFECTIVE,
-			      &attach_flags, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 3);
-	assert(attach_flags == 0);
-	assert(bpf_prog_query(cg5, BPF_CGROUP_INET_EGRESS, 0,
-			      NULL, prog_ids, &prog_cnt) == 0);
-	assert(prog_cnt == 0);
-	goto out;
-err:
-	rc = 1;
-
-out:
-	for (i = 0; i < 6; i++)
-		if (allow_prog[i] > 0)
-			close(allow_prog[i]);
-	close(cg1);
-	close(cg2);
-	close(cg3);
-	close(cg4);
-	close(cg5);
-	cleanup_cgroup_environment();
-	if (!rc)
-		printf("#multi:PASS\n");
-	else
-		printf("#multi:FAIL\n");
-	return rc;
-}
-
-static int test_autodetach(void)
-{
-	__u32 prog_cnt = 4, attach_flags;
-	int allow_prog[2] = {0};
-	__u32 prog_ids[2] = {0};
-	int cg = 0, i, rc = -1;
-	void *ptr = NULL;
-	int attempts;
-
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
-		allow_prog[i] = prog_load_cnt(1, 1 << i);
-		if (!allow_prog[i])
-			goto err;
-	}
-
-	if (setup_cgroup_environment())
-		goto err;
-
-	/* create a cgroup, attach two programs and remember their ids */
-	cg = create_and_get_cgroup("/cg_autodetach");
-	if (cg < 0)
-		goto err;
-
-	if (join_cgroup("/cg_autodetach"))
-		goto err;
-
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
-		if (bpf_prog_attach(allow_prog[i], cg, BPF_CGROUP_INET_EGRESS,
-				    BPF_F_ALLOW_MULTI)) {
-			log_err("Attaching prog[%d] to cg:egress", i);
-			goto err;
-		}
-	}
-
-	/* make sure that programs are attached and run some traffic */
-	assert(bpf_prog_query(cg, BPF_CGROUP_INET_EGRESS, 0, &attach_flags,
-			      prog_ids, &prog_cnt) == 0);
-	assert(system(PING_CMD) == 0);
-
-	/* allocate some memory (4Mb) to pin the original cgroup */
-	ptr = malloc(4 * (1 << 20));
-	if (!ptr)
-		goto err;
-
-	/* close programs and cgroup fd */
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++) {
-		close(allow_prog[i]);
-		allow_prog[i] = 0;
-	}
-
-	close(cg);
-	cg = 0;
-
-	/* leave the cgroup and remove it. don't detach programs */
-	cleanup_cgroup_environment();
-
-	/* wait for the asynchronous auto-detachment.
-	 * wait for no more than 5 sec and give up.
-	 */
-	for (i = 0; i < ARRAY_SIZE(prog_ids); i++) {
-		for (attempts = 5; attempts >= 0; attempts--) {
-			int fd = bpf_prog_get_fd_by_id(prog_ids[i]);
-
-			if (fd < 0)
-				break;
-
-			/* don't leave the fd open */
-			close(fd);
-
-			if (!attempts)
-				goto err;
-
-			sleep(1);
-		}
-	}
-
-	rc = 0;
-err:
-	for (i = 0; i < ARRAY_SIZE(allow_prog); i++)
-		if (allow_prog[i] > 0)
-			close(allow_prog[i]);
-	if (cg)
-		close(cg);
-	free(ptr);
-	cleanup_cgroup_environment();
-	if (!rc)
-		printf("#autodetach:PASS\n");
-	else
-		printf("#autodetach:FAIL\n");
-	return rc;
-}
-
-int main(void)
-{
-	int (*tests[])(void) = {
-		test_foo_bar,
-		test_multiprog,
-		test_autodetach,
-	};
-	int errors = 0;
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(tests); i++)
-		if (tests[i]())
-			errors++;
-
-	if (errors)
-		printf("test_cgroup_attach:FAIL\n");
-	else
-		printf("test_cgroup_attach:PASS\n");
-
-	return errors ? EXIT_FAILURE : EXIT_SUCCESS;
-}
diff --git a/tools/testing/selftests/bpf/test_cpp.cpp b/tools/testing/selftests/bpf/test_cpp.cpp
index f0eb2727b766..6fe23a10d48a 100644
--- a/tools/testing/selftests/bpf/test_cpp.cpp
+++ b/tools/testing/selftests/bpf/test_cpp.cpp
@@ -1,12 +1,16 @@
 /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#include <iostream>
 #include "libbpf.h"
 #include "bpf.h"
 #include "btf.h"
+#include "test_core_extern.skel.h"
 
 /* do nothing, just make sure we can link successfully */
 
 int main(int argc, char *argv[])
 {
+	struct test_core_extern *skel;
+
 	/* libbpf.h */
 	libbpf_set_print(NULL);
 
@@ -16,5 +20,11 @@ int main(int argc, char *argv[])
 	/* btf.h */
 	btf__new(NULL, 0);
 
+	/* BPF skeleton */
+	skel = test_core_extern__open_and_load();
+	test_core_extern__destroy(skel);
+
+	std::cout << "DONE!" << std::endl;
+
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 8477df835979..de1fdaa4e7b4 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -100,6 +100,7 @@ extern struct ipv6_packet pkt_v6;
 
 #define _CHECK(condition, tag, duration, format...) ({			\
 	int __ret = !!(condition);					\
+	int __save_errno = errno;					\
 	if (__ret) {							\
 		test__fail();						\
 		printf("%s:FAIL:%s ", __func__, tag);			\
@@ -108,15 +109,18 @@ extern struct ipv6_packet pkt_v6;
 		printf("%s:PASS:%s %d nsec\n",				\
 		       __func__, tag, duration);			\
 	}								\
+	errno = __save_errno;						\
 	__ret;								\
 })
 
 #define CHECK_FAIL(condition) ({					\
 	int __ret = !!(condition);					\
+	int __save_errno = errno;					\
 	if (__ret) {							\
 		test__fail();						\
 		printf("%s:FAIL:%d\n", __func__, __LINE__);		\
 	}								\
+	errno = __save_errno;						\
 	__ret;								\
 })