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

Alexei Starovoitov says:

====================
pull-request: bpf-next 2021-03-24

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

We've added 37 non-merge commits during the last 15 day(s) which contain
a total of 65 files changed, 3200 insertions(+), 738 deletions(-).

The main changes are:

1) Static linking of multiple BPF ELF files, from Andrii.

2) Move drop error path to devmap for XDP_REDIRECT, from Lorenzo.

3) Spelling fixes from various folks.
====================

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

65 files changed, 3200 insertions, 738 deletions

diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index f973e2ead197..63cae0476bb4 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -1209,21 +1209,67 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
 	 */
 	case BPF_STX | BPF_ATOMIC | BPF_DW:
 	case BPF_STX | BPF_ATOMIC | BPF_W:
-		if (insn->imm != BPF_ADD) {
+	{
+		bool is32 = BPF_SIZE(insn->code) == BPF_W;
+
+		switch (insn->imm) {
+/* {op32|op64} {%w0|%src},%src,off(%dst) */
+#define EMIT_ATOMIC(op32, op64) do {					\
+	EMIT6_DISP_LH(0xeb000000, is32 ? (op32) : (op64),		\
+		      (insn->imm & BPF_FETCH) ? src_reg : REG_W0,	\
+		      src_reg, dst_reg, off);				\
+	if (is32 && (insn->imm & BPF_FETCH))				\
+		EMIT_ZERO(src_reg);					\
+} while (0)
+		case BPF_ADD:
+		case BPF_ADD | BPF_FETCH:
+			/* {laal|laalg} */
+			EMIT_ATOMIC(0x00fa, 0x00ea);
+			break;
+		case BPF_AND:
+		case BPF_AND | BPF_FETCH:
+			/* {lan|lang} */
+			EMIT_ATOMIC(0x00f4, 0x00e4);
+			break;
+		case BPF_OR:
+		case BPF_OR | BPF_FETCH:
+			/* {lao|laog} */
+			EMIT_ATOMIC(0x00f6, 0x00e6);
+			break;
+		case BPF_XOR:
+		case BPF_XOR | BPF_FETCH:
+			/* {lax|laxg} */
+			EMIT_ATOMIC(0x00f7, 0x00e7);
+			break;
+#undef EMIT_ATOMIC
+		case BPF_XCHG:
+			/* {ly|lg} %w0,off(%dst) */
+			EMIT6_DISP_LH(0xe3000000,
+				      is32 ? 0x0058 : 0x0004, REG_W0, REG_0,
+				      dst_reg, off);
+			/* 0: {csy|csg} %w0,%src,off(%dst) */
+			EMIT6_DISP_LH(0xeb000000, is32 ? 0x0014 : 0x0030,
+				      REG_W0, src_reg, dst_reg, off);
+			/* brc 4,0b */
+			EMIT4_PCREL_RIC(0xa7040000, 4, jit->prg - 6);
+			/* {llgfr|lgr} %src,%w0 */
+			EMIT4(is32 ? 0xb9160000 : 0xb9040000, src_reg, REG_W0);
+			if (is32 && insn_is_zext(&insn[1]))
+				insn_count = 2;
+			break;
+		case BPF_CMPXCHG:
+			/* 0: {csy|csg} %b0,%src,off(%dst) */
+			EMIT6_DISP_LH(0xeb000000, is32 ? 0x0014 : 0x0030,
+				      BPF_REG_0, src_reg, dst_reg, off);
+			break;
+		default:
 			pr_err("Unknown atomic operation %02x\n", insn->imm);
 			return -1;
 		}
 
-		/* *(u32/u64 *)(dst + off) += src
-		 *
-		 * BFW_W:  laal  %w0,%src,off(%dst)
-		 * BPF_DW: laalg %w0,%src,off(%dst)
-		 */
-		EMIT6_DISP_LH(0xeb000000,
-			      BPF_SIZE(insn->code) == BPF_W ? 0x00fa : 0x00ea,
-			      REG_W0, src_reg, dst_reg, off);
 		jit->seen |= SEEN_MEM;
 		break;
+	}
 	/*
 	 * BPF_LDX
 	 */
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 102f2c91fdb8..5c062c51b4cb 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -300,7 +300,7 @@ static int ena_xdp_xmit_frame(struct ena_ring *xdp_ring,
 
 	rc = ena_xdp_tx_map_frame(xdp_ring, tx_info, xdpf, &push_hdr, &push_len);
 	if (unlikely(rc))
-		goto error_drop_packet;
+		return rc;
 
 	ena_tx_ctx.ena_bufs = tx_info->bufs;
 	ena_tx_ctx.push_header = push_hdr;
@@ -330,8 +330,6 @@ static int ena_xdp_xmit_frame(struct ena_ring *xdp_ring,
 error_unmap_dma:
 	ena_unmap_tx_buff(xdp_ring, tx_info);
 	tx_info->xdpf = NULL;
-error_drop_packet:
-	xdp_return_frame(xdpf);
 	return rc;
 }
 
@@ -339,8 +337,8 @@ static int ena_xdp_xmit(struct net_device *dev, int n,
 			struct xdp_frame **frames, u32 flags)
 {
 	struct ena_adapter *adapter = netdev_priv(dev);
-	int qid, i, err, drops = 0;
 	struct ena_ring *xdp_ring;
+	int qid, i, nxmit = 0;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
@@ -360,12 +358,9 @@ static int ena_xdp_xmit(struct net_device *dev, int n,
 	spin_lock(&xdp_ring->xdp_tx_lock);
 
 	for (i = 0; i < n; i++) {
-		err = ena_xdp_xmit_frame(xdp_ring, dev, frames[i], 0);
-		/* The descriptor is freed by ena_xdp_xmit_frame in case
-		 * of an error.
-		 */
-		if (err)
-			drops++;
+		if (ena_xdp_xmit_frame(xdp_ring, dev, frames[i], 0))
+			break;
+		nxmit++;
 	}
 
 	/* Ring doorbell to make device aware of the packets */
@@ -378,7 +373,7 @@ static int ena_xdp_xmit(struct net_device *dev, int n,
 	spin_unlock(&xdp_ring->xdp_tx_lock);
 
 	/* Return number of packets sent */
-	return n - drops;
+	return nxmit;
 }
 
 static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
@@ -415,7 +410,9 @@ static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp)
 		/* The XDP queues are shared between XDP_TX and XDP_REDIRECT */
 		spin_lock(&xdp_ring->xdp_tx_lock);
 
-		ena_xdp_xmit_frame(xdp_ring, rx_ring->netdev, xdpf, XDP_XMIT_FLUSH);
+		if (ena_xdp_xmit_frame(xdp_ring, rx_ring->netdev, xdpf,
+				       XDP_XMIT_FLUSH))
+			xdp_return_frame(xdpf);
 
 		spin_unlock(&xdp_ring->xdp_tx_lock);
 		xdp_stat = &rx_ring->rx_stats.xdp_tx;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
index 641303894341..ec9564e584e0 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c
@@ -217,7 +217,7 @@ int bnxt_xdp_xmit(struct net_device *dev, int num_frames,
 	struct pci_dev *pdev = bp->pdev;
 	struct bnxt_tx_ring_info *txr;
 	dma_addr_t mapping;
-	int drops = 0;
+	int nxmit = 0;
 	int ring;
 	int i;
 
@@ -233,21 +233,17 @@ int bnxt_xdp_xmit(struct net_device *dev, int num_frames,
 		struct xdp_frame *xdp = frames[i];
 
 		if (!txr || !bnxt_tx_avail(bp, txr) ||
-		    !(bp->bnapi[ring]->flags & BNXT_NAPI_FLAG_XDP)) {
-			xdp_return_frame_rx_napi(xdp);
-			drops++;
-			continue;
-		}
+		    !(bp->bnapi[ring]->flags & BNXT_NAPI_FLAG_XDP))
+			break;
 
 		mapping = dma_map_single(&pdev->dev, xdp->data, xdp->len,
 					 DMA_TO_DEVICE);
 
-		if (dma_mapping_error(&pdev->dev, mapping)) {
-			xdp_return_frame_rx_napi(xdp);
-			drops++;
-			continue;
-		}
+		if (dma_mapping_error(&pdev->dev, mapping))
+			break;
+
 		__bnxt_xmit_xdp_redirect(bp, txr, mapping, xdp->len, xdp);
+		nxmit++;
 	}
 
 	if (flags & XDP_XMIT_FLUSH) {
@@ -256,7 +252,7 @@ int bnxt_xdp_xmit(struct net_device *dev, int num_frames,
 		bnxt_db_write(bp, &txr->tx_db, txr->tx_prod);
 	}
 
-	return num_frames - drops;
+	return nxmit;
 }
 
 /* Under rtnl_lock */
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
index 720dc99bd1fc..177c020bf34a 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c
@@ -3081,7 +3081,7 @@ static int dpaa_xdp_xmit(struct net_device *net_dev, int n,
 			 struct xdp_frame **frames, u32 flags)
 {
 	struct xdp_frame *xdpf;
-	int i, err, drops = 0;
+	int i, nxmit = 0;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
@@ -3091,14 +3091,12 @@ static int dpaa_xdp_xmit(struct net_device *net_dev, int n,
 
 	for (i = 0; i < n; i++) {
 		xdpf = frames[i];
-		err = dpaa_xdp_xmit_frame(net_dev, xdpf);
-		if (err) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (dpaa_xdp_xmit_frame(net_dev, xdpf))
+			break;
+		nxmit++;
 	}
 
-	return n - drops;
+	return nxmit;
 }
 
 static int dpaa_ts_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c
index 492943bb9c48..fc0eb82cdd6a 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c
@@ -2431,8 +2431,6 @@ static int dpaa2_eth_xdp_xmit(struct net_device *net_dev, int n,
 	percpu_stats->tx_packets += enqueued;
 	for (i = 0; i < enqueued; i++)
 		percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]);
-	for (i = enqueued; i < n; i++)
-		xdp_return_frame_rx_napi(frames[i]);
 
 	return enqueued;
 }
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 8b5820921377..fc20afc23bfa 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -3835,8 +3835,8 @@ netdev_tx_t i40e_lan_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
  * @frames: array of XDP buffer pointers
  * @flags: XDP extra info
  *
- * Returns number of frames successfully sent. Frames that fail are
- * free'ed via XDP return API.
+ * Returns number of frames successfully sent. Failed frames
+ * will be free'ed by XDP core.
  *
  * For error cases, a negative errno code is returned and no-frames
  * are transmitted (caller must handle freeing frames).
@@ -3849,7 +3849,7 @@ int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
 	struct i40e_ring *xdp_ring;
-	int drops = 0;
+	int nxmit = 0;
 	int i;
 
 	if (test_bit(__I40E_VSI_DOWN, vsi->state))
@@ -3869,14 +3869,13 @@ int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		int err;
 
 		err = i40e_xmit_xdp_ring(xdpf, xdp_ring);
-		if (err != I40E_XDP_TX) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (err != I40E_XDP_TX)
+			break;
+		nxmit++;
 	}
 
 	if (unlikely(flags & XDP_XMIT_FLUSH))
 		i40e_xdp_ring_update_tail(xdp_ring);
 
-	return n - drops;
+	return nxmit;
 }
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index 3148e78adc36..c71f2fbbb262 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -554,8 +554,8 @@ ice_run_xdp(struct ice_ring *rx_ring, struct xdp_buff *xdp,
  * @frames: XDP frames to be transmitted
  * @flags: transmit flags
  *
- * Returns number of frames successfully sent. Frames that fail are
- * free'ed via XDP return API.
+ * Returns number of frames successfully sent. Failed frames
+ * will be free'ed by XDP core.
  * For error cases, a negative errno code is returned and no-frames
  * are transmitted (caller must handle freeing frames).
  */
@@ -567,7 +567,7 @@ ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 	unsigned int queue_index = smp_processor_id();
 	struct ice_vsi *vsi = np->vsi;
 	struct ice_ring *xdp_ring;
-	int drops = 0, i;
+	int nxmit = 0, i;
 
 	if (test_bit(__ICE_DOWN, vsi->state))
 		return -ENETDOWN;
@@ -584,16 +584,15 @@ ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		int err;
 
 		err = ice_xmit_xdp_ring(xdpf->data, xdpf->len, xdp_ring);
-		if (err != ICE_XDP_TX) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (err != ICE_XDP_TX)
+			break;
+		nxmit++;
 	}
 
 	if (unlikely(flags & XDP_XMIT_FLUSH))
 		ice_xdp_ring_update_tail(xdp_ring);
 
-	return n - drops;
+	return nxmit;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index b83966aa6647..c9e8c65a3cfe 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -2934,7 +2934,7 @@ static int igb_xdp_xmit(struct net_device *dev, int n,
 	int cpu = smp_processor_id();
 	struct igb_ring *tx_ring;
 	struct netdev_queue *nq;
-	int drops = 0;
+	int nxmit = 0;
 	int i;
 
 	if (unlikely(test_bit(__IGB_DOWN, &adapter->state)))
@@ -2961,10 +2961,9 @@ static int igb_xdp_xmit(struct net_device *dev, int n,
 		int err;
 
 		err = igb_xmit_xdp_ring(adapter, tx_ring, xdpf);
-		if (err != IGB_XDP_TX) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (err != IGB_XDP_TX)
+			break;
+		nxmit++;
 	}
 
 	__netif_tx_unlock(nq);
@@ -2972,7 +2971,7 @@ static int igb_xdp_xmit(struct net_device *dev, int n,
 	if (unlikely(flags & XDP_XMIT_FLUSH))
 		igb_xdp_ring_update_tail(tx_ring);
 
-	return n - drops;
+	return nxmit;
 }
 
 static const struct net_device_ops igb_netdev_ops = {
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 19fe21116fe8..7ba1c2985ef7 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -10189,7 +10189,7 @@ static int ixgbe_xdp_xmit(struct net_device *dev, int n,
 {
 	struct ixgbe_adapter *adapter = netdev_priv(dev);
 	struct ixgbe_ring *ring;
-	int drops = 0;
+	int nxmit = 0;
 	int i;
 
 	if (unlikely(test_bit(__IXGBE_DOWN, &adapter->state)))
@@ -10213,16 +10213,15 @@ static int ixgbe_xdp_xmit(struct net_device *dev, int n,
 		int err;
 
 		err = ixgbe_xmit_xdp_ring(adapter, xdpf);
-		if (err != IXGBE_XDP_TX) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (err != IXGBE_XDP_TX)
+			break;
+		nxmit++;
 	}
 
 	if (unlikely(flags & XDP_XMIT_FLUSH))
 		ixgbe_xdp_ring_update_tail(ring);
 
-	return n - drops;
+	return nxmit;
 }
 
 static const struct net_device_ops ixgbe_netdev_ops = {
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index a635cf84608a..20307eec8988 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -2137,7 +2137,7 @@ mvneta_xdp_xmit(struct net_device *dev, int num_frame,
 {
 	struct mvneta_port *pp = netdev_priv(dev);
 	struct mvneta_pcpu_stats *stats = this_cpu_ptr(pp->stats);
-	int i, nxmit_byte = 0, nxmit = num_frame;
+	int i, nxmit_byte = 0, nxmit = 0;
 	int cpu = smp_processor_id();
 	struct mvneta_tx_queue *txq;
 	struct netdev_queue *nq;
@@ -2155,12 +2155,11 @@ mvneta_xdp_xmit(struct net_device *dev, int num_frame,
 	__netif_tx_lock(nq, cpu);
 	for (i = 0; i < num_frame; i++) {
 		ret = mvneta_xdp_submit_frame(pp, txq, frames[i], true);
-		if (ret == MVNETA_XDP_TX) {
-			nxmit_byte += frames[i]->len;
-		} else {
-			xdp_return_frame_rx_napi(frames[i]);
-			nxmit--;
-		}
+		if (ret != MVNETA_XDP_TX)
+			break;
+
+		nxmit_byte += frames[i]->len;
+		nxmit++;
 	}
 
 	if (unlikely(flags & XDP_XMIT_FLUSH))
diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 1767c60056c5..ec706d614cac 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -3744,7 +3744,7 @@ mvpp2_xdp_xmit(struct net_device *dev, int num_frame,
 	       struct xdp_frame **frames, u32 flags)
 {
 	struct mvpp2_port *port = netdev_priv(dev);
-	int i, nxmit_byte = 0, nxmit = num_frame;
+	int i, nxmit_byte = 0, nxmit = 0;
 	struct mvpp2_pcpu_stats *stats;
 	u16 txq_id;
 	u32 ret;
@@ -3762,12 +3762,11 @@ mvpp2_xdp_xmit(struct net_device *dev, int num_frame,
 
 	for (i = 0; i < num_frame; i++) {
 		ret = mvpp2_xdp_submit_frame(port, txq_id, frames[i], true);
-		if (ret == MVPP2_XDP_TX) {
-			nxmit_byte += frames[i]->len;
-		} else {
-			xdp_return_frame_rx_napi(frames[i]);
-			nxmit--;
-		}
+		if (ret != MVPP2_XDP_TX)
+			break;
+
+		nxmit_byte += frames[i]->len;
+		nxmit++;
 	}
 
 	if (likely(nxmit > 0))
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
index 2e3e78b0f333..2f0df5cc1a2d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c
@@ -500,7 +500,7 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
 	struct mlx5e_xdpsq *sq;
-	int drops = 0;
+	int nxmit = 0;
 	int sq_num;
 	int i;
 
@@ -529,11 +529,8 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		xdptxd.dma_addr = dma_map_single(sq->pdev, xdptxd.data,
 						 xdptxd.len, DMA_TO_DEVICE);
 
-		if (unlikely(dma_mapping_error(sq->pdev, xdptxd.dma_addr))) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-			continue;
-		}
+		if (unlikely(dma_mapping_error(sq->pdev, xdptxd.dma_addr)))
+			break;
 
 		xdpi.mode           = MLX5E_XDP_XMIT_MODE_FRAME;
 		xdpi.frame.xdpf     = xdpf;
@@ -544,9 +541,9 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		if (unlikely(!ret)) {
 			dma_unmap_single(sq->pdev, xdptxd.dma_addr,
 					 xdptxd.len, DMA_TO_DEVICE);
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
+			break;
 		}
+		nxmit++;
 	}
 
 	if (flags & XDP_XMIT_FLUSH) {
@@ -555,7 +552,7 @@ int mlx5e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
 		mlx5e_xmit_xdp_doorbell(sq);
 	}
 
-	return n - drops;
+	return nxmit;
 }
 
 void mlx5e_xdp_rx_poll_complete(struct mlx5e_rq *rq)
diff --git a/drivers/net/ethernet/qlogic/qede/qede_fp.c b/drivers/net/ethernet/qlogic/qede/qede_fp.c
index 8c47a9d2a965..102d0e0808d5 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_fp.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_fp.c
@@ -345,7 +345,7 @@ int qede_xdp_transmit(struct net_device *dev, int n_frames,
 	struct qede_tx_queue *xdp_tx;
 	struct xdp_frame *xdpf;
 	dma_addr_t mapping;
-	int i, drops = 0;
+	int i, nxmit = 0;
 	u16 xdp_prod;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
@@ -364,18 +364,13 @@ int qede_xdp_transmit(struct net_device *dev, int n_frames,
 
 		mapping = dma_map_single(dmadev, xdpf->data, xdpf->len,
 					 DMA_TO_DEVICE);
-		if (unlikely(dma_mapping_error(dmadev, mapping))) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-
-			continue;
-		}
+		if (unlikely(dma_mapping_error(dmadev, mapping)))
+			break;
 
 		if (unlikely(qede_xdp_xmit(xdp_tx, mapping, 0, xdpf->len,
-					   NULL, xdpf))) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+					   NULL, xdpf)))
+			break;
+		nxmit++;
 	}
 
 	if (flags & XDP_XMIT_FLUSH) {
@@ -387,7 +382,7 @@ int qede_xdp_transmit(struct net_device *dev, int n_frames,
 
 	spin_unlock(&xdp_tx->xdp_tx_lock);
 
-	return n_frames - drops;
+	return nxmit;
 }
 
 int qede_txq_has_work(struct qede_tx_queue *txq)
diff --git a/drivers/net/ethernet/sfc/tx.c b/drivers/net/ethernet/sfc/tx.c
index 1665529a7271..0c6650d2e239 100644
--- a/drivers/net/ethernet/sfc/tx.c
+++ b/drivers/net/ethernet/sfc/tx.c
@@ -412,14 +412,6 @@ err:
 	return NETDEV_TX_OK;
 }
 
-static void efx_xdp_return_frames(int n,  struct xdp_frame **xdpfs)
-{
-	int i;
-
-	for (i = 0; i < n; i++)
-		xdp_return_frame_rx_napi(xdpfs[i]);
-}
-
 /* Transmit a packet from an XDP buffer
  *
  * Returns number of packets sent on success, error code otherwise.
@@ -492,12 +484,7 @@ int efx_xdp_tx_buffers(struct efx_nic *efx, int n, struct xdp_frame **xdpfs,
 	if (flush && i > 0)
 		efx_nic_push_buffers(tx_queue);
 
-	if (i == 0)
-		return -EIO;
-
-	efx_xdp_return_frames(n - i, xdpfs + i);
-
-	return i;
+	return i == 0 ? -EIO : i;
 }
 
 /* Initiate a packet transmission.  We use one channel per CPU
diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c
index 200785e703c8..dfc85cc68173 100644
--- a/drivers/net/ethernet/socionext/netsec.c
+++ b/drivers/net/ethernet/socionext/netsec.c
@@ -1760,8 +1760,7 @@ static int netsec_xdp_xmit(struct net_device *ndev, int n,
 {
 	struct netsec_priv *priv = netdev_priv(ndev);
 	struct netsec_desc_ring *tx_ring = &priv->desc_ring[NETSEC_RING_TX];
-	int drops = 0;
-	int i;
+	int i, nxmit = 0;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
@@ -1772,12 +1771,11 @@ static int netsec_xdp_xmit(struct net_device *ndev, int n,
 		int err;
 
 		err = netsec_xdp_queue_one(priv, xdpf, true);
-		if (err != NETSEC_XDP_TX) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		} else {
-			tx_ring->xdp_xmit++;
-		}
+		if (err != NETSEC_XDP_TX)
+			break;
+
+		tx_ring->xdp_xmit++;
+		nxmit++;
 	}
 	spin_unlock(&tx_ring->lock);
 
@@ -1786,7 +1784,7 @@ static int netsec_xdp_xmit(struct net_device *ndev, int n,
 		tx_ring->xdp_xmit = 0;
 	}
 
-	return n - drops;
+	return nxmit;
 }
 
 static int netsec_xdp_setup(struct netsec_priv *priv, struct bpf_prog *prog,
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index fd966567464c..074702af3dc6 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -1123,25 +1123,23 @@ static int cpsw_ndo_xdp_xmit(struct net_device *ndev, int n,
 	struct cpsw_priv *priv = netdev_priv(ndev);
 	struct cpsw_common *cpsw = priv->cpsw;
 	struct xdp_frame *xdpf;
-	int i, drops = 0, port;
+	int i, nxmit = 0, port;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
 
 	for (i = 0; i < n; i++) {
 		xdpf = frames[i];
-		if (xdpf->len < CPSW_MIN_PACKET_SIZE) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-			continue;
-		}
+		if (xdpf->len < CPSW_MIN_PACKET_SIZE)
+			break;
 
 		port = priv->emac_port + cpsw->data.dual_emac;
 		if (cpsw_xdp_tx_frame(priv, xdpf, NULL, port))
-			drops++;
+			break;
+		nxmit++;
 	}
 
-	return n - drops;
+	return nxmit;
 }
 
 #ifdef CONFIG_NET_POLL_CONTROLLER
diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c
index 58a64313ac00..0751f77de2c7 100644
--- a/drivers/net/ethernet/ti/cpsw_new.c
+++ b/drivers/net/ethernet/ti/cpsw_new.c
@@ -1093,24 +1093,22 @@ static int cpsw_ndo_xdp_xmit(struct net_device *ndev, int n,
 {
 	struct cpsw_priv *priv = netdev_priv(ndev);
 	struct xdp_frame *xdpf;
-	int i, drops = 0;
+	int i, nxmit = 0;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
 
 	for (i = 0; i < n; i++) {
 		xdpf = frames[i];
-		if (xdpf->len < CPSW_MIN_PACKET_SIZE) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-			continue;
-		}
+		if (xdpf->len < CPSW_MIN_PACKET_SIZE)
+			break;
 
 		if (cpsw_xdp_tx_frame(priv, xdpf, NULL, priv->emac_port))
-			drops++;
+			break;
+		nxmit++;
 	}
 
-	return n - drops;
+	return nxmit;
 }
 
 static int cpsw_get_port_parent_id(struct net_device *ndev,
diff --git a/drivers/net/ethernet/ti/cpsw_priv.c b/drivers/net/ethernet/ti/cpsw_priv.c
index bb59e768915e..5862f0a4a975 100644
--- a/drivers/net/ethernet/ti/cpsw_priv.c
+++ b/drivers/net/ethernet/ti/cpsw_priv.c
@@ -1305,19 +1305,15 @@ int cpsw_xdp_tx_frame(struct cpsw_priv *priv, struct xdp_frame *xdpf,
 		ret = cpdma_chan_submit_mapped(txch, cpsw_xdpf_to_handle(xdpf),
 					       dma, xdpf->len, port);
 	} else {
-		if (sizeof(*xmeta) > xdpf->headroom) {
-			xdp_return_frame_rx_napi(xdpf);
+		if (sizeof(*xmeta) > xdpf->headroom)
 			return -EINVAL;
-		}
 
 		ret = cpdma_chan_submit(txch, cpsw_xdpf_to_handle(xdpf),
 					xdpf->data, xdpf->len, port);
 	}
 
-	if (ret) {
+	if (ret)
 		priv->ndev->stats.tx_dropped++;
-		xdp_return_frame_rx_napi(xdpf);
-	}
 
 	return ret;
 }
@@ -1353,7 +1349,8 @@ int cpsw_run_xdp(struct cpsw_priv *priv, int ch, struct xdp_buff *xdp,
 		if (unlikely(!xdpf))
 			goto drop;
 
-		cpsw_xdp_tx_frame(priv, xdpf, page, port);
+		if (cpsw_xdp_tx_frame(priv, xdpf, page, port))
+			xdp_return_frame_rx_napi(xdpf);
 		break;
 	case XDP_REDIRECT:
 		if (xdp_do_redirect(ndev, xdp, prog))
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index fc86da7f1628..6e55697315de 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1181,8 +1181,7 @@ static int tun_xdp_xmit(struct net_device *dev, int n,
 	struct tun_struct *tun = netdev_priv(dev);
 	struct tun_file *tfile;
 	u32 numqueues;
-	int drops = 0;
-	int cnt = n;
+	int nxmit = 0;
 	int i;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
@@ -1212,9 +1211,9 @@ resample:
 
 		if (__ptr_ring_produce(&tfile->tx_ring, frame)) {
 			atomic_long_inc(&dev->tx_dropped);
-			xdp_return_frame_rx_napi(xdp);
-			drops++;
+			break;
 		}
+		nxmit++;
 	}
 	spin_unlock(&tfile->tx_ring.producer_lock);
 
@@ -1222,17 +1221,21 @@ resample:
 		__tun_xdp_flush_tfile(tfile);
 
 	rcu_read_unlock();
-	return cnt - drops;
+	return nxmit;
 }
 
 static int tun_xdp_tx(struct net_device *dev, struct xdp_buff *xdp)
 {
 	struct xdp_frame *frame = xdp_convert_buff_to_frame(xdp);
+	int nxmit;
 
 	if (unlikely(!frame))
 		return -EOVERFLOW;
 
-	return tun_xdp_xmit(dev, 1, &frame, XDP_XMIT_FLUSH);
+	nxmit = tun_xdp_xmit(dev, 1, &frame, XDP_XMIT_FLUSH);
+	if (!nxmit)
+		xdp_return_frame_rx_napi(frame);
+	return nxmit;
 }
 
 static const struct net_device_ops tap_netdev_ops = {
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 34e49c75db42..91b73db37555 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -433,7 +433,7 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
 			 u32 flags, bool ndo_xmit)
 {
 	struct veth_priv *rcv_priv, *priv = netdev_priv(dev);
-	int i, ret = -ENXIO, drops = 0;
+	int i, ret = -ENXIO, nxmit = 0;
 	struct net_device *rcv;
 	unsigned int max_len;
 	struct veth_rq *rq;
@@ -463,21 +463,20 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
 		void *ptr = veth_xdp_to_ptr(frame);
 
 		if (unlikely(frame->len > max_len ||
-			     __ptr_ring_produce(&rq->xdp_ring, ptr))) {
-			xdp_return_frame_rx_napi(frame);
-			drops++;
-		}
+			     __ptr_ring_produce(&rq->xdp_ring, ptr)))
+			break;
+		nxmit++;
 	}
 	spin_unlock(&rq->xdp_ring.producer_lock);
 
 	if (flags & XDP_XMIT_FLUSH)
 		__veth_xdp_flush(rq);
 
-	ret = n - drops;
+	ret = nxmit;
 	if (ndo_xmit) {
 		u64_stats_update_begin(&rq->stats.syncp);
-		rq->stats.vs.peer_tq_xdp_xmit += n - drops;
-		rq->stats.vs.peer_tq_xdp_xmit_err += drops;
+		rq->stats.vs.peer_tq_xdp_xmit += nxmit;
+		rq->stats.vs.peer_tq_xdp_xmit_err += n - nxmit;
 		u64_stats_update_end(&rq->stats.syncp);
 	}
 
@@ -504,20 +503,23 @@ static int veth_ndo_xdp_xmit(struct net_device *dev, int n,
 
 static void veth_xdp_flush_bq(struct veth_rq *rq, struct veth_xdp_tx_bq *bq)
 {
-	int sent, i, err = 0;
+	int sent, i, err = 0, drops;
 
 	sent = veth_xdp_xmit(rq->dev, bq->count, bq->q, 0, false);
 	if (sent < 0) {
 		err = sent;
 		sent = 0;
-		for (i = 0; i < bq->count; i++)
-			xdp_return_frame(bq->q[i]);
 	}
-	trace_xdp_bulk_tx(rq->dev, sent, bq->count - sent, err);
+
+	for (i = sent; unlikely(i < bq->count); i++)
+		xdp_return_frame(bq->q[i]);
+
+	drops = bq->count - sent;
+	trace_xdp_bulk_tx(rq->dev, sent, drops, err);
 
 	u64_stats_update_begin(&rq->stats.syncp);
 	rq->stats.vs.xdp_tx += sent;
-	rq->stats.vs.xdp_tx_err += bq->count - sent;
+	rq->stats.vs.xdp_tx_err += drops;
 	u64_stats_update_end(&rq->stats.syncp);
 
 	bq->count = 0;
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 584a9bd59dda..bb4ea9dbc16b 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -531,10 +531,10 @@ static int virtnet_xdp_xmit(struct net_device *dev,
 	unsigned int len;
 	int packets = 0;
 	int bytes = 0;
-	int drops = 0;
+	int nxmit = 0;
 	int kicks = 0;
-	int ret, err;
 	void *ptr;
+	int ret;
 	int i;
 
 	/* Only allow ndo_xdp_xmit if XDP is loaded on dev, as this
@@ -548,7 +548,6 @@ static int virtnet_xdp_xmit(struct net_device *dev,
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) {
 		ret = -EINVAL;
-		drops = n;
 		goto out;
 	}
 
@@ -571,13 +570,11 @@ static int virtnet_xdp_xmit(struct net_device *dev,
 	for (i = 0; i < n; i++) {
 		struct xdp_frame *xdpf = frames[i];
 
-		err = __virtnet_xdp_xmit_one(vi, sq, xdpf);
-		if (err) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (__virtnet_xdp_xmit_one(vi, sq, xdpf))
+			break;
+		nxmit++;
 	}
-	ret = n - drops;
+	ret = nxmit;
 
 	if (flags & XDP_XMIT_FLUSH) {
 		if (virtqueue_kick_prepare(sq->vq) && virtqueue_notify(sq->vq))
@@ -588,7 +585,7 @@ out:
 	sq->stats.bytes += bytes;
 	sq->stats.packets += packets;
 	sq->stats.xdp_tx += n;
-	sq->stats.xdp_tx_drops += drops;
+	sq->stats.xdp_tx_drops += n - nxmit;
 	sq->stats.kicks += kicks;
 	u64_stats_update_end(&sq->stats.syncp);
 
@@ -742,7 +739,9 @@ static struct sk_buff *receive_small(struct net_device *dev,
 			if (unlikely(!xdpf))
 				goto err_xdp;
 			err = virtnet_xdp_xmit(dev, 1, &xdpf, 0);
-			if (unlikely(err < 0)) {
+			if (unlikely(!err)) {
+				xdp_return_frame_rx_napi(xdpf);
+			} else if (unlikely(err < 0)) {
 				trace_xdp_exception(vi->dev, xdp_prog, act);
 				goto err_xdp;
 			}
@@ -929,7 +928,9 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
 			if (unlikely(!xdpf))
 				goto err_xdp;
 			err = virtnet_xdp_xmit(dev, 1, &xdpf, 0);
-			if (unlikely(err < 0)) {
+			if (unlikely(!err)) {
+				xdp_return_frame_rx_napi(xdpf);
+			} else if (unlikely(err < 0)) {
 				trace_xdp_exception(vi->dev, xdp_prog, act);
 				if (unlikely(xdp_page != page))
 					put_page(xdp_page);
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index cc19cd9203da..44275908d61a 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -608,8 +608,8 @@ static int xennet_xdp_xmit(struct net_device *dev, int n,
 	struct netfront_info *np = netdev_priv(dev);
 	struct netfront_queue *queue = NULL;
 	unsigned long irq_flags;
-	int drops = 0;
-	int i, err;
+	int nxmit = 0;
+	int i;
 
 	if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
 		return -EINVAL;
@@ -622,15 +622,13 @@ static int xennet_xdp_xmit(struct net_device *dev, int n,
 
 		if (!xdpf)
 			continue;
-		err = xennet_xdp_xmit_one(dev, queue, xdpf);
-		if (err) {
-			xdp_return_frame_rx_napi(xdpf);
-			drops++;
-		}
+		if (xennet_xdp_xmit_one(dev, queue, xdpf))
+			break;
+		nxmit++;
 	}
 	spin_unlock_irqrestore(&queue->tx_lock, irq_flags);
 
-	return n - drops;
+	return nxmit;
 }
 
 
@@ -875,7 +873,9 @@ static u32 xennet_run_xdp(struct netfront_queue *queue, struct page *pdata,
 		get_page(pdata);
 		xdpf = xdp_convert_buff_to_frame(xdp);
 		err = xennet_xdp_xmit(queue->info->netdev, 1, &xdpf, 0);
-		if (unlikely(err < 0))
+		if (unlikely(!err))
+			xdp_return_frame_rx_napi(xdpf);
+		else if (unlikely(err < 0))
 			trace_xdp_exception(queue->info->netdev, prog, act);
 		break;
 	case XDP_REDIRECT:
diff --git a/include/linux/btf.h b/include/linux/btf.h
index 7fabf1428093..9c1b52738bbe 100644
--- a/include/linux/btf.h
+++ b/include/linux/btf.h
@@ -9,6 +9,7 @@
 #include <uapi/linux/bpf.h>
 
 #define BTF_TYPE_EMIT(type) ((void)(type *)0)
+#define BTF_TYPE_EMIT_ENUM(enum_val) ((void)enum_val)
 
 struct btf;
 struct btf_member;
diff --git a/kernel/bpf/bpf_lsm.c b/kernel/bpf/bpf_lsm.c
index 9829f381b51c..f211506218ea 100644
--- a/kernel/bpf/bpf_lsm.c
+++ b/kernel/bpf/bpf_lsm.c
@@ -67,7 +67,7 @@ BPF_CALL_2(bpf_bprm_opts_set, struct linux_binprm *, bprm, u64, flags)
 
 BTF_ID_LIST_SINGLE(bpf_bprm_opts_set_btf_ids, struct, linux_binprm)
 
-const static struct bpf_func_proto bpf_bprm_opts_set_proto = {
+static const struct bpf_func_proto bpf_bprm_opts_set_proto = {
 	.func		= bpf_bprm_opts_set,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
@@ -88,7 +88,7 @@ static bool bpf_ima_inode_hash_allowed(const struct bpf_prog *prog)
 
 BTF_ID_LIST_SINGLE(bpf_ima_inode_hash_btf_ids, struct, inode)
 
-const static struct bpf_func_proto bpf_ima_inode_hash_proto = {
+static const struct bpf_func_proto bpf_ima_inode_hash_proto = {
 	.func		= bpf_ima_inode_hash,
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
diff --git a/kernel/bpf/bpf_task_storage.c b/kernel/bpf/bpf_task_storage.c
index fd3c74ef608e..3ce75758d394 100644
--- a/kernel/bpf/bpf_task_storage.c
+++ b/kernel/bpf/bpf_task_storage.c
@@ -20,7 +20,7 @@
 
 DEFINE_BPF_STORAGE_CACHE(task_cache);
 
-DEFINE_PER_CPU(int, bpf_task_storage_busy);
+static DEFINE_PER_CPU(int, bpf_task_storage_busy);
 
 static void bpf_task_storage_lock(void)
 {
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 7a5ad7331c3b..aa516472ce46 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -329,7 +329,7 @@ bool dev_map_can_have_prog(struct bpf_map *map)
 static void bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 {
 	struct net_device *dev = bq->dev;
-	int sent = 0, drops = 0, err = 0;
+	int sent = 0, err = 0;
 	int i;
 
 	if (unlikely(!bq->count))
@@ -343,29 +343,23 @@ static void bq_xmit_all(struct xdp_dev_bulk_queue *bq, u32 flags)
 
 	sent = dev->netdev_ops->ndo_xdp_xmit(dev, bq->count, bq->q, flags);
 	if (sent < 0) {
+		/* If ndo_xdp_xmit fails with an errno, no frames have
+		 * been xmit'ed.
+		 */
 		err = sent;
 		sent = 0;
-		goto error;
 	}
-	drops = bq->count - sent;
-out:
-	bq->count = 0;
 
-	trace_xdp_devmap_xmit(bq->dev_rx, dev, sent, drops, err);
-	bq->dev_rx = NULL;
-	__list_del_clearprev(&bq->flush_node);
-	return;
-error:
-	/* If ndo_xdp_xmit fails with an errno, no frames have been
-	 * xmit'ed and it's our responsibility to them free all.
+	/* If not all frames have been transmitted, it is our
+	 * responsibility to free them
 	 */
-	for (i = 0; i < bq->count; i++) {
-		struct xdp_frame *xdpf = bq->q[i];
+	for (i = sent; unlikely(i < bq->count); i++)
+		xdp_return_frame_rx_napi(bq->q[i]);
 
-		xdp_return_frame_rx_napi(xdpf);
-		drops++;
-	}
-	goto out;
+	trace_xdp_devmap_xmit(bq->dev_rx, dev, sent, bq->count - sent, err);
+	bq->dev_rx = NULL;
+	bq->count = 0;
+	__list_del_clearprev(&bq->flush_node);
 }
 
 /* __dev_flush is called from xdp_do_flush() which _must_ be signaled
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 330d721dd2af..d7ebb12ffffc 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -31,7 +31,7 @@
 /*
  * The bucket lock has two protection scopes:
  *
- * 1) Serializing concurrent operations from BPF programs on differrent
+ * 1) Serializing concurrent operations from BPF programs on different
  *    CPUs
  *
  * 2) Serializing concurrent operations from BPF programs and sys_bpf()
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 999bf36ffeb1..210169c25ead 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -11912,7 +11912,6 @@ static int do_misc_fixups(struct bpf_verifier_env *env)
 		    insn->code == (BPF_ALU64 | BPF_SUB | BPF_X)) {
 			const u8 code_add = BPF_ALU64 | BPF_ADD | BPF_X;
 			const u8 code_sub = BPF_ALU64 | BPF_SUB | BPF_X;
-			struct bpf_insn insn_buf[16];
 			struct bpf_insn *patch = &insn_buf[0];
 			bool issrc, isneg;
 			u32 off_reg;
diff --git a/net/core/filter.c b/net/core/filter.c
index f5eeebf6a16f..17dc159ec40c 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -4729,6 +4729,9 @@ static int _bpf_setsockopt(struct sock *sk, int level, int optname,
 				sk->sk_prot->keepalive(sk, valbool);
 			sock_valbool_flag(sk, SOCK_KEEPOPEN, valbool);
 			break;
+		case SO_REUSEPORT:
+			sk->sk_reuseport = valbool;
+			break;
 		default:
 			ret = -EINVAL;
 		}
@@ -4898,6 +4901,9 @@ static int _bpf_getsockopt(struct sock *sk, int level, int optname,
 		case SO_BINDTOIFINDEX:
 			*((int *)optval) = sk->sk_bound_dev_if;
 			break;
+		case SO_REUSEPORT:
+			*((int *)optval) = sk->sk_reuseport;
+			break;
 		default:
 			goto err_clear;
 		}
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index de7cc8445ac0..e14fd0c50c10 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -267,6 +267,7 @@
 #include <linux/slab.h>
 #include <linux/errqueue.h>
 #include <linux/static_key.h>
+#include <linux/btf.h>
 
 #include <net/icmp.h>
 #include <net/inet_common.h>
@@ -2587,6 +2588,17 @@ void tcp_set_state(struct sock *sk, int state)
 	BUILD_BUG_ON((int)BPF_TCP_NEW_SYN_RECV != (int)TCP_NEW_SYN_RECV);
 	BUILD_BUG_ON((int)BPF_TCP_MAX_STATES != (int)TCP_MAX_STATES);
 
+	/* bpf uapi header bpf.h defines an anonymous enum with values
+	 * BPF_TCP_* used by bpf programs. Currently gcc built vmlinux
+	 * is able to emit this enum in DWARF due to the above BUILD_BUG_ON.
+	 * But clang built vmlinux does not have this enum in DWARF
+	 * since clang removes the above code before generating IR/debuginfo.
+	 * Let us explicitly emit the type debuginfo to ensure the
+	 * above-mentioned anonymous enum in the vmlinux DWARF and hence BTF
+	 * regardless of which compiler is used.
+	 */
+	BTF_TYPE_EMIT_ENUM(BPF_TCP_ESTABLISHED);
+
 	if (BPF_SOCK_OPS_TEST_FLAG(tcp_sk(sk), BPF_SOCK_OPS_STATE_CB_FLAG))
 		tcp_call_bpf_2arg(sk, BPF_SOCK_OPS_STATE_CB, oldstate, state);
 
diff --git a/samples/bpf/do_hbm_test.sh b/samples/bpf/do_hbm_test.sh
index 21790ea5c460..38e4599350db 100755
--- a/samples/bpf/do_hbm_test.sh
+++ b/samples/bpf/do_hbm_test.sh
@@ -10,7 +10,7 @@
 Usage() {
   echo "Script for testing HBM (Host Bandwidth Manager) framework."
   echo "It creates a cgroup to use for testing and load a BPF program to limit"
-  echo "egress or ingress bandwidht. It then uses iperf3 or netperf to create"
+  echo "egress or ingress bandwidth. It then uses iperf3 or netperf to create"
   echo "loads. The output is the goodput in Mbps (unless -D was used)."
   echo ""
   echo "USAGE: $name [out] [-b=<prog>|--bpf=<prog>] [-c=<cc>|--cc=<cc>]"
diff --git a/tools/bpf/bpftool/Documentation/bpftool-gen.rst b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
index 84cf0639696f..7cd6681137f3 100644
--- a/tools/bpf/bpftool/Documentation/bpftool-gen.rst
+++ b/tools/bpf/bpftool/Documentation/bpftool-gen.rst
@@ -14,16 +14,37 @@ SYNOPSIS
 
 	*OPTIONS* := { { **-j** | **--json** } [{ **-p** | **--pretty** }] }
 
-	*COMMAND* := { **skeleton** | **help** }
+	*COMMAND* := { **object** | **skeleton** | **help** }
 
 GEN COMMANDS
 =============
 
-|	**bpftool** **gen skeleton** *FILE*
+|	**bpftool** **gen object** *OUTPUT_FILE* *INPUT_FILE* [*INPUT_FILE*...]
+|	**bpftool** **gen skeleton** *FILE* [**name** *OBJECT_NAME*]
 |	**bpftool** **gen help**
 
 DESCRIPTION
 ===========
+	**bpftool gen object** *OUTPUT_FILE* *INPUT_FILE* [*INPUT_FILE*...]
+		  Statically link (combine) together one or more *INPUT_FILE*'s
+		  into a single resulting *OUTPUT_FILE*. All the files involved
+		  are BPF ELF object files.
+
+		  The rules of BPF static linking are mostly the same as for
+		  user-space object files, but in addition to combining data
+		  and instruction sections, .BTF and .BTF.ext (if present in
+		  any of the input files) data are combined together. .BTF
+		  data is deduplicated, so all the common types across
+		  *INPUT_FILE*'s will only be represented once in the resulting
+		  BTF information.
+
+		  BPF static linking allows to partition BPF source code into
+		  individually compiled files that are then linked into
+		  a single resulting BPF object file, which can be used to
+		  generated BPF skeleton (with **gen skeleton** command) or
+		  passed directly into **libbpf** (using **bpf_object__open()**
+		  family of APIs).
+
 	**bpftool gen skeleton** *FILE*
 		  Generate BPF skeleton C header file for a given *FILE*.
 
@@ -75,10 +96,13 @@ DESCRIPTION
 		  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:
+		  Each of them is prefixed with object name. Object name can
+		  either be derived from object file name, i.e., if BPF object
+		  file name is **example.o**, BPF object name will be
+		  **example**. Object name can be also specified explicitly
+		  through **name** *OBJECT_NAME* parameter. The following
+		  custom functions are provided (assuming **example** as
+		  the object name):
 
 		  - **example__open** and **example__open_opts**.
 		    These functions are used to instantiate skeleton. It
@@ -130,26 +154,19 @@ OPTIONS
 
 EXAMPLES
 ========
-**$ cat example.c**
+**$ cat example1.bpf.c**
 
 ::
 
   #include <stdbool.h>
   #include <linux/ptrace.h>
   #include <linux/bpf.h>
-  #include "bpf_helpers.h"
+  #include <bpf/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)
   {
@@ -161,6 +178,21 @@ EXAMPLES
   	return 0;
   }
 
+**$ cat example2.bpf.c**
+
+::
+
+  #include <linux/ptrace.h>
+  #include <linux/bpf.h>
+  #include <bpf/bpf_helpers.h>
+
+  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_exit")
   int handle_sys_exit(struct pt_regs *ctx)
   {
@@ -170,9 +202,17 @@ EXAMPLES
   }
 
 This is example BPF application with two BPF programs and a mix of BPF maps
-and global variables.
+and global variables. Source code is split across two source code files.
 
-**$ bpftool gen skeleton example.o**
+**$ clang -target bpf -g example1.bpf.c -o example1.bpf.o**
+**$ clang -target bpf -g example2.bpf.c -o example2.bpf.o**
+**$ bpftool gen object example.bpf.o example1.bpf.o example2.bpf.o**
+
+This set of commands compiles *example1.bpf.c* and *example2.bpf.c*
+individually and then statically links respective object files into the final
+BPF ELF object file *example.bpf.o*.
+
+**$ bpftool gen skeleton example.bpf.o name example | tee example.skel.h**
 
 ::
 
@@ -227,7 +267,7 @@ and global variables.
 
   #endif /* __EXAMPLE_SKEL_H__ */
 
-**$ cat example_user.c**
+**$ cat example.c**
 
 ::
 
@@ -270,7 +310,7 @@ and global variables.
   	return err;
   }
 
-**# ./example_user**
+**# ./example**
 
 ::
 
diff --git a/tools/bpf/bpftool/bash-completion/bpftool b/tools/bpf/bpftool/bash-completion/bpftool
index fdffbc64c65c..d67518bcbd44 100644
--- a/tools/bpf/bpftool/bash-completion/bpftool
+++ b/tools/bpf/bpftool/bash-completion/bpftool
@@ -981,12 +981,25 @@ _bpftool()
             ;;
         gen)
             case $command in
-                skeleton)
+                object)
                     _filedir
+                    return 0
+                    ;;
+                skeleton)
+                    case $prev in
+                        $command)
+                            _filedir
+                            return 0
+                            ;;
+                        *)
+                            _bpftool_once_attr 'name'
+                            return 0
+                            ;;
+                    esac
                     ;;
                 *)
                     [[ $prev == $object ]] && \
-                        COMPREPLY=( $( compgen -W 'skeleton help' -- "$cur" ) )
+                        COMPREPLY=( $( compgen -W 'object skeleton help' -- "$cur" ) )
                     ;;
             esac
             ;;
diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index 985610c3f193..62953bbf68b4 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -546,6 +546,7 @@ static int do_dump(int argc, char **argv)
 			NEXT_ARG();
 			if (argc < 1) {
 				p_err("expecting value for 'format' option\n");
+				err = -EINVAL;
 				goto done;
 			}
 			if (strcmp(*argv, "c") == 0) {
@@ -555,11 +556,13 @@ static int do_dump(int argc, char **argv)
 			} else {
 				p_err("unrecognized format specifier: '%s', possible values: raw, c",
 				      *argv);
+				err = -EINVAL;
 				goto done;
 			}
 			NEXT_ARG();
 		} else {
 			p_err("unrecognized option: '%s'", *argv);
+			err = -EINVAL;
 			goto done;
 		}
 	}
diff --git a/tools/bpf/bpftool/gen.c b/tools/bpf/bpftool/gen.c
index 4033c46d83e7..31ade77f5ef8 100644
--- a/tools/bpf/bpftool/gen.c
+++ b/tools/bpf/bpftool/gen.c
@@ -273,7 +273,7 @@ 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;
+	char obj_name[MAX_OBJ_NAME_LEN] = "", *obj_data;
 	struct bpf_object *obj = NULL;
 	const char *file, *ident;
 	struct bpf_program *prog;
@@ -288,6 +288,28 @@ static int do_skeleton(int argc, char **argv)
 	}
 	file = GET_ARG();
 
+	while (argc) {
+		if (!REQ_ARGS(2))
+			return -1;
+
+		if (is_prefix(*argv, "name")) {
+			NEXT_ARG();
+
+			if (obj_name[0] != '\0') {
+				p_err("object name already specified");
+				return -1;
+			}
+
+			strncpy(obj_name, *argv, MAX_OBJ_NAME_LEN - 1);
+			obj_name[MAX_OBJ_NAME_LEN - 1] = '\0';
+		} else {
+			p_err("unknown arg %s", *argv);
+			return -1;
+		}
+
+		NEXT_ARG();
+	}
+
 	if (argc) {
 		p_err("extra unknown arguments");
 		return -1;
@@ -310,7 +332,8 @@ static int do_skeleton(int argc, char **argv)
 		p_err("failed to mmap() %s: %s", file, strerror(errno));
 		goto out;
 	}
-	get_obj_name(obj_name, file);
+	if (obj_name[0] == '\0')
+		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)) {
@@ -591,6 +614,47 @@ out:
 	return err;
 }
 
+static int do_object(int argc, char **argv)
+{
+	struct bpf_linker *linker;
+	const char *output_file, *file;
+	int err = 0;
+
+	if (!REQ_ARGS(2)) {
+		usage();
+		return -1;
+	}
+
+	output_file = GET_ARG();
+
+	linker = bpf_linker__new(output_file, NULL);
+	if (!linker) {
+		p_err("failed to create BPF linker instance");
+		return -1;
+	}
+
+	while (argc) {
+		file = GET_ARG();
+
+		err = bpf_linker__add_file(linker, file);
+		if (err) {
+			p_err("failed to link '%s': %s (%d)", file, strerror(err), err);
+			goto out;
+		}
+	}
+
+	err = bpf_linker__finalize(linker);
+	if (err) {
+		p_err("failed to finalize ELF file: %s (%d)", strerror(err), err);
+		goto out;
+	}
+
+	err = 0;
+out:
+	bpf_linker__free(linker);
+	return err;
+}
+
 static int do_help(int argc, char **argv)
 {
 	if (json_output) {
@@ -599,7 +663,8 @@ static int do_help(int argc, char **argv)
 	}
 
 	fprintf(stderr,
-		"Usage: %1$s %2$s skeleton FILE\n"
+		"Usage: %1$s %2$s object OUTPUT_FILE INPUT_FILE [INPUT_FILE...]\n"
+		"       %1$s %2$s skeleton FILE [name OBJECT_NAME]\n"
 		"       %1$s %2$s help\n"
 		"\n"
 		"       " HELP_SPEC_OPTIONS "\n"
@@ -610,6 +675,7 @@ static int do_help(int argc, char **argv)
 }
 
 static const struct cmd cmds[] = {
+	{ "object",	do_object },
 	{ "skeleton",	do_skeleton },
 	{ "help",	do_help },
 	{ 0 }
diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c
index b86f450e6fce..d9afb730136a 100644
--- a/tools/bpf/bpftool/main.c
+++ b/tools/bpf/bpftool/main.c
@@ -276,7 +276,7 @@ static int do_batch(int argc, char **argv)
 	int n_argc;
 	FILE *fp;
 	char *cp;
-	int err;
+	int err = 0;
 	int i;
 
 	if (argc < 2) {
@@ -370,7 +370,6 @@ static int do_batch(int argc, char **argv)
 	} else {
 		if (!json_output)
 			printf("processed %d commands\n", lines);
-		err = 0;
 	}
 err_close:
 	if (fp != stdin)
diff --git a/tools/bpf/bpftool/map.c b/tools/bpf/bpftool/map.c
index b400364ee054..09ae0381205b 100644
--- a/tools/bpf/bpftool/map.c
+++ b/tools/bpf/bpftool/map.c
@@ -100,7 +100,7 @@ static int do_dump_btf(const struct btf_dumper *d,
 		       void *value)
 {
 	__u32 value_id;
-	int ret;
+	int ret = 0;
 
 	/* start of key-value pair */
 	jsonw_start_object(d->jw);
diff --git a/tools/lib/bpf/Build b/tools/lib/bpf/Build
index 190366d05588..9b057cc7650a 100644
--- a/tools/lib/bpf/Build
+++ b/tools/lib/bpf/Build
@@ -1,3 +1,3 @@
 libbpf-y := libbpf.o bpf.o nlattr.o btf.o libbpf_errno.o str_error.o \
 	    netlink.o bpf_prog_linfo.o libbpf_probes.o xsk.o hashmap.o \
-	    btf_dump.o ringbuf.o
+	    btf_dump.o ringbuf.o strset.o linker.o
diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index 87b14b74d62f..e43e1896cb4b 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -228,7 +228,6 @@ install_headers: $(BPF_HELPER_DEFS)
 		$(call do_install,bpf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,libbpf.h,$(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); \
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index ae6c975e0b87..cc2e51c64a54 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -29,9 +29,10 @@
  */
 #define SEC(NAME) __attribute__((section(NAME), used))
 
-#ifndef __always_inline
+/* Avoid 'linux/stddef.h' definition of '__always_inline'. */
+#undef __always_inline
 #define __always_inline inline __attribute__((always_inline))
-#endif
+
 #ifndef __noinline
 #define __noinline __attribute__((noinline))
 #endif
@@ -39,8 +40,22 @@
 #define __weak __attribute__((weak))
 #endif
 
+/* When utilizing vmlinux.h with BPF CO-RE, user BPF programs can't include
+ * any system-level headers (such as stddef.h, linux/version.h, etc), and
+ * commonly-used macros like NULL and KERNEL_VERSION aren't available through
+ * vmlinux.h. This just adds unnecessary hurdles and forces users to re-define
+ * them on their own. So as a convenience, provide such definitions here.
+ */
+#ifndef NULL
+#define NULL ((void *)0)
+#endif
+
+#ifndef KERNEL_VERSION
+#define KERNEL_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + ((c) > 255 ? 255 : (c))
+#endif
+
 /*
- * Helper macro to manipulate data structures
+ * Helper macros to manipulate data structures
  */
 #ifndef offsetof
 #define offsetof(TYPE, MEMBER)	((unsigned long)&((TYPE *)0)->MEMBER)
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 3aa58f2ac183..d30e67e7e1e5 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -21,6 +21,7 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 #include "hashmap.h"
+#include "strset.h"
 
 #define BTF_MAX_NR_TYPES 0x7fffffffU
 #define BTF_MAX_STR_OFFSET 0x7fffffffU
@@ -67,7 +68,7 @@ struct btf {
 	 * |             |            |
 	 * hdr           |            |
 	 * types_data----+            |
-	 * strs_data------------------+
+	 * strset__data(strs_set)-----+
 	 *
 	 *               +----------+---------+-----------+
 	 *               |  Header  |  Types  |  Strings  |
@@ -105,20 +106,15 @@ struct btf {
 	 */
 	int start_str_off;
 
+	/* only one of strs_data or strs_set can be non-NULL, depending on
+	 * whether BTF is in a modifiable state (strs_set is used) or not
+	 * (strs_data points inside raw_data)
+	 */
 	void *strs_data;
-	size_t strs_data_cap; /* used size stored in hdr->str_len */
-
-	/* lookup index for each unique string in strings section */
-	struct hashmap *strs_hash;
+	/* a set of unique strings */
+	struct strset *strs_set;
 	/* whether strings are already deduplicated */
 	bool strs_deduped;
-	/* extra indirection layer to make strings hashmap work with stable
-	 * string offsets and ability to transparently choose between
-	 * btf->strs_data or btf_dedup->strs_data as a source of strings.
-	 * This is used for BTF strings dedup to transfer deduplicated strings
-	 * data back to struct btf without re-building strings index.
-	 */
-	void **strs_data_ptr;
 
 	/* BTF object FD, if loaded into kernel */
 	int fd;
@@ -142,8 +138,8 @@ static inline __u64 ptr_to_u64(const void *ptr)
  * On success, memory pointer to the beginning of unused memory is returned.
  * On error, NULL is returned.
  */
-void *btf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
-		  size_t cur_cnt, size_t max_cnt, size_t add_cnt)
+void *libbpf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
+		     size_t cur_cnt, size_t max_cnt, size_t add_cnt)
 {
 	size_t new_cnt;
 	void *new_data;
@@ -179,14 +175,14 @@ void *btf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
 /* Ensure given dynamically allocated memory region has enough allocated space
  * to accommodate *need_cnt* elements of size *elem_sz* bytes each
  */
-int btf_ensure_mem(void **data, size_t *cap_cnt, size_t elem_sz, size_t need_cnt)
+int libbpf_ensure_mem(void **data, size_t *cap_cnt, size_t elem_sz, size_t need_cnt)
 {
 	void *p;
 
 	if (need_cnt <= *cap_cnt)
 		return 0;
 
-	p = btf_add_mem(data, cap_cnt, elem_sz, *cap_cnt, SIZE_MAX, need_cnt - *cap_cnt);
+	p = libbpf_add_mem(data, cap_cnt, elem_sz, *cap_cnt, SIZE_MAX, need_cnt - *cap_cnt);
 	if (!p)
 		return -ENOMEM;
 
@@ -197,8 +193,8 @@ static int btf_add_type_idx_entry(struct btf *btf, __u32 type_off)
 {
 	__u32 *p;
 
-	p = btf_add_mem((void **)&btf->type_offs, &btf->type_offs_cap, sizeof(__u32),
-			btf->nr_types, BTF_MAX_NR_TYPES, 1);
+	p = libbpf_add_mem((void **)&btf->type_offs, &btf->type_offs_cap, sizeof(__u32),
+			   btf->nr_types, BTF_MAX_NR_TYPES, 1);
 	if (!p)
 		return -ENOMEM;
 
@@ -435,7 +431,7 @@ const struct btf *btf__base_btf(const struct btf *btf)
 }
 
 /* internal helper returning non-const pointer to a type */
-static struct btf_type *btf_type_by_id(struct btf *btf, __u32 type_id)
+struct btf_type *btf_type_by_id(struct btf *btf, __u32 type_id)
 {
 	if (type_id == 0)
 		return &btf_void;
@@ -738,7 +734,7 @@ void btf__free(struct btf *btf)
 		 */
 		free(btf->hdr);
 		free(btf->types_data);
-		free(btf->strs_data);
+		strset__free(btf->strs_set);
 	}
 	free(btf->raw_data);
 	free(btf->raw_data_swapped);
@@ -1246,6 +1242,11 @@ void btf__set_fd(struct btf *btf, int fd)
 	btf->fd = fd;
 }
 
+static const void *btf_strs_data(const struct btf *btf)
+{
+	return btf->strs_data ? btf->strs_data : strset__data(btf->strs_set);
+}
+
 static void *btf_get_raw_data(const struct btf *btf, __u32 *size, bool swap_endian)
 {
 	struct btf_header *hdr = btf->hdr;
@@ -1286,7 +1287,7 @@ static void *btf_get_raw_data(const struct btf *btf, __u32 *size, bool swap_endi
 	}
 	p += hdr->type_len;
 
-	memcpy(p, btf->strs_data, hdr->str_len);
+	memcpy(p, btf_strs_data(btf), hdr->str_len);
 	p += hdr->str_len;
 
 	*size = data_sz;
@@ -1320,7 +1321,7 @@ const char *btf__str_by_offset(const struct btf *btf, __u32 offset)
 	if (offset < btf->start_str_off)
 		return btf__str_by_offset(btf->base_btf, offset);
 	else if (offset - btf->start_str_off < btf->hdr->str_len)
-		return btf->strs_data + (offset - btf->start_str_off);
+		return btf_strs_data(btf) + (offset - btf->start_str_off);
 	else
 		return NULL;
 }
@@ -1474,25 +1475,6 @@ int btf__get_map_kv_tids(const struct btf *btf, const char *map_name,
 	return 0;
 }
 
-static size_t strs_hash_fn(const void *key, void *ctx)
-{
-	const struct btf *btf = ctx;
-	const char *strs = *btf->strs_data_ptr;
-	const char *str = strs + (long)key;
-
-	return str_hash(str);
-}
-
-static bool strs_hash_equal_fn(const void *key1, const void *key2, void *ctx)
-{
-	const struct btf *btf = ctx;
-	const char *strs = *btf->strs_data_ptr;
-	const char *str1 = strs + (long)key1;
-	const char *str2 = strs + (long)key2;
-
-	return strcmp(str1, str2) == 0;
-}
-
 static void btf_invalidate_raw_data(struct btf *btf)
 {
 	if (btf->raw_data) {
@@ -1511,10 +1493,9 @@ static void btf_invalidate_raw_data(struct btf *btf)
  */
 static int btf_ensure_modifiable(struct btf *btf)
 {
-	void *hdr, *types, *strs, *strs_end, *s;
-	struct hashmap *hash = NULL;
-	long off;
-	int err;
+	void *hdr, *types;
+	struct strset *set = NULL;
+	int err = -ENOMEM;
 
 	if (btf_is_modifiable(btf)) {
 		/* any BTF modification invalidates raw_data */
@@ -1525,44 +1506,25 @@ static int btf_ensure_modifiable(struct btf *btf)
 	/* split raw data into three memory regions */
 	hdr = malloc(btf->hdr->hdr_len);
 	types = malloc(btf->hdr->type_len);
-	strs = malloc(btf->hdr->str_len);
-	if (!hdr || !types || !strs)
+	if (!hdr || !types)
 		goto err_out;
 
 	memcpy(hdr, btf->hdr, btf->hdr->hdr_len);
 	memcpy(types, btf->types_data, btf->hdr->type_len);
-	memcpy(strs, btf->strs_data, btf->hdr->str_len);
-
-	/* make hashmap below use btf->strs_data as a source of strings */
-	btf->strs_data_ptr = &btf->strs_data;
 
 	/* build lookup index for all strings */
-	hash = hashmap__new(strs_hash_fn, strs_hash_equal_fn, btf);
-	if (IS_ERR(hash)) {
-		err = PTR_ERR(hash);
-		hash = NULL;
+	set = strset__new(BTF_MAX_STR_OFFSET, btf->strs_data, btf->hdr->str_len);
+	if (IS_ERR(set)) {
+		err = PTR_ERR(set);
 		goto err_out;
 	}
 
-	strs_end = strs + btf->hdr->str_len;
-	for (off = 0, s = strs; s < strs_end; off += strlen(s) + 1, s = strs + off) {
-		/* hashmap__add() returns EEXIST if string with the same
-		 * content already is in the hash map
-		 */
-		err = hashmap__add(hash, (void *)off, (void *)off);
-		if (err == -EEXIST)
-			continue; /* duplicate */
-		if (err)
-			goto err_out;
-	}
-
 	/* only when everything was successful, update internal state */
 	btf->hdr = hdr;
 	btf->types_data = types;
 	btf->types_data_cap = btf->hdr->type_len;
-	btf->strs_data = strs;
-	btf->strs_data_cap = btf->hdr->str_len;
-	btf->strs_hash = hash;
+	btf->strs_data = NULL;
+	btf->strs_set = set;
 	/* if BTF was created from scratch, all strings are guaranteed to be
 	 * unique and deduplicated
 	 */
@@ -1577,17 +1539,10 @@ static int btf_ensure_modifiable(struct btf *btf)
 	return 0;
 
 err_out:
-	hashmap__free(hash);
+	strset__free(set);
 	free(hdr);
 	free(types);
-	free(strs);
-	return -ENOMEM;
-}
-
-static void *btf_add_str_mem(struct btf *btf, size_t add_sz)
-{
-	return btf_add_mem(&btf->strs_data, &btf->strs_data_cap, 1,
-			   btf->hdr->str_len, BTF_MAX_STR_OFFSET, add_sz);
+	return err;
 }
 
 /* Find an offset in BTF string section that corresponds to a given string *s*.
@@ -1598,34 +1553,23 @@ static void *btf_add_str_mem(struct btf *btf, size_t add_sz)
  */
 int btf__find_str(struct btf *btf, const char *s)
 {
-	long old_off, new_off, len;
-	void *p;
+	int off;
 
 	if (btf->base_btf) {
-		int ret;
-
-		ret = btf__find_str(btf->base_btf, s);
-		if (ret != -ENOENT)
-			return ret;
+		off = btf__find_str(btf->base_btf, s);
+		if (off != -ENOENT)
+			return off;
 	}
 
 	/* BTF needs to be in a modifiable state to build string lookup index */
 	if (btf_ensure_modifiable(btf))
 		return -ENOMEM;
 
-	/* see btf__add_str() for why we do this */
-	len = strlen(s) + 1;
-	p = btf_add_str_mem(btf, len);
-	if (!p)
-		return -ENOMEM;
-
-	new_off = btf->hdr->str_len;
-	memcpy(p, s, len);
+	off = strset__find_str(btf->strs_set, s);
+	if (off < 0)
+		return off;
 
-	if (hashmap__find(btf->strs_hash, (void *)new_off, (void **)&old_off))
-		return btf->start_str_off + old_off;
-
-	return -ENOENT;
+	return btf->start_str_off + off;
 }
 
 /* Add a string s to the BTF string section.
@@ -1635,56 +1579,30 @@ int btf__find_str(struct btf *btf, const char *s)
  */
 int btf__add_str(struct btf *btf, const char *s)
 {
-	long old_off, new_off, len;
-	void *p;
-	int err;
+	int off;
 
 	if (btf->base_btf) {
-		int ret;
-
-		ret = btf__find_str(btf->base_btf, s);
-		if (ret != -ENOENT)
-			return ret;
+		off = btf__find_str(btf->base_btf, s);
+		if (off != -ENOENT)
+			return off;
 	}
 
 	if (btf_ensure_modifiable(btf))
 		return -ENOMEM;
 
-	/* Hashmap keys are always offsets within btf->strs_data, so to even
-	 * look up some string from the "outside", we need to first append it
-	 * at the end, so that it can be addressed with an offset. Luckily,
-	 * until btf->hdr->str_len is incremented, that string is just a piece
-	 * of garbage for the rest of BTF code, so no harm, no foul. On the
-	 * other hand, if the string is unique, it's already appended and
-	 * ready to be used, only a simple btf->hdr->str_len increment away.
-	 */
-	len = strlen(s) + 1;
-	p = btf_add_str_mem(btf, len);
-	if (!p)
-		return -ENOMEM;
+	off = strset__add_str(btf->strs_set, s);
+	if (off < 0)
+		return off;
 
-	new_off = btf->hdr->str_len;
-	memcpy(p, s, len);
+	btf->hdr->str_len = strset__data_size(btf->strs_set);
 
-	/* Now attempt to add the string, but only if the string with the same
-	 * contents doesn't exist already (HASHMAP_ADD strategy). If such
-	 * string exists, we'll get its offset in old_off (that's old_key).
-	 */
-	err = hashmap__insert(btf->strs_hash, (void *)new_off, (void *)new_off,
-			      HASHMAP_ADD, (const void **)&old_off, NULL);
-	if (err == -EEXIST)
-		return btf->start_str_off + old_off; /* duplicated string, return existing offset */
-	if (err)
-		return err;
-
-	btf->hdr->str_len += len; /* new unique string, adjust data length */
-	return btf->start_str_off + new_off;
+	return btf->start_str_off + off;
 }
 
 static void *btf_add_type_mem(struct btf *btf, size_t add_sz)
 {
-	return btf_add_mem(&btf->types_data, &btf->types_data_cap, 1,
-			   btf->hdr->type_len, UINT_MAX, add_sz);
+	return libbpf_add_mem(&btf->types_data, &btf->types_data_cap, 1,
+			      btf->hdr->type_len, UINT_MAX, add_sz);
 }
 
 static __u32 btf_type_info(int kind, int vlen, int kflag)
@@ -1711,6 +1629,54 @@ static int btf_commit_type(struct btf *btf, int data_sz)
 	return btf->start_id + btf->nr_types - 1;
 }
 
+struct btf_pipe {
+	const struct btf *src;
+	struct btf *dst;
+};
+
+static int btf_rewrite_str(__u32 *str_off, void *ctx)
+{
+	struct btf_pipe *p = ctx;
+	int off;
+
+	if (!*str_off) /* nothing to do for empty strings */
+		return 0;
+
+	off = btf__add_str(p->dst, btf__str_by_offset(p->src, *str_off));
+	if (off < 0)
+		return off;
+
+	*str_off = off;
+	return 0;
+}
+
+int btf__add_type(struct btf *btf, const struct btf *src_btf, const struct btf_type *src_type)
+{
+	struct btf_pipe p = { .src = src_btf, .dst = btf };
+	struct btf_type *t;
+	int sz, err;
+
+	sz = btf_type_size(src_type);
+	if (sz < 0)
+		return sz;
+
+	/* deconstruct BTF, if necessary, and invalidate raw_data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	memcpy(t, src_type, sz);
+
+	err = btf_type_visit_str_offs(t, btf_rewrite_str, &p);
+	if (err)
+		return err;
+
+	return btf_commit_type(btf, sz);
+}
+
 /*
  * Append new BTF_KIND_INT type with:
  *   - *name* - non-empty, non-NULL type name;
@@ -3016,10 +2982,7 @@ struct btf_dedup {
 	/* Various option modifying behavior of algorithm */
 	struct btf_dedup_opts opts;
 	/* temporary strings deduplication state */
-	void *strs_data;
-	size_t strs_cap;
-	size_t strs_len;
-	struct hashmap* strs_hash;
+	struct strset *strs_set;
 };
 
 static long hash_combine(long h, long value)
@@ -3155,95 +3118,28 @@ done:
 	return d;
 }
 
-typedef int (*str_off_fn_t)(__u32 *str_off_ptr, void *ctx);
-
 /*
  * Iterate over all possible places in .BTF and .BTF.ext that can reference
  * string and pass pointer to it to a provided callback `fn`.
  */
-static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
+static int btf_for_each_str_off(struct btf_dedup *d, str_off_visit_fn fn, void *ctx)
 {
-	void *line_data_cur, *line_data_end;
-	int i, j, r, rec_size;
-	struct btf_type *t;
+	int i, r;
 
 	for (i = 0; i < d->btf->nr_types; i++) {
-		t = btf_type_by_id(d->btf, d->btf->start_id + i);
-		r = fn(&t->name_off, ctx);
+		struct btf_type *t = btf_type_by_id(d->btf, d->btf->start_id + i);
+
+		r = btf_type_visit_str_offs(t, fn, ctx);
 		if (r)
 			return r;
-
-		switch (btf_kind(t)) {
-		case BTF_KIND_STRUCT:
-		case BTF_KIND_UNION: {
-			struct btf_member *m = btf_members(t);
-			__u16 vlen = btf_vlen(t);
-
-			for (j = 0; j < vlen; j++) {
-				r = fn(&m->name_off, ctx);
-				if (r)
-					return r;
-				m++;
-			}
-			break;
-		}
-		case BTF_KIND_ENUM: {
-			struct btf_enum *m = btf_enum(t);
-			__u16 vlen = btf_vlen(t);
-
-			for (j = 0; j < vlen; j++) {
-				r = fn(&m->name_off, ctx);
-				if (r)
-					return r;
-				m++;
-			}
-			break;
-		}
-		case BTF_KIND_FUNC_PROTO: {
-			struct btf_param *m = btf_params(t);
-			__u16 vlen = btf_vlen(t);
-
-			for (j = 0; j < vlen; j++) {
-				r = fn(&m->name_off, ctx);
-				if (r)
-					return r;
-				m++;
-			}
-			break;
-		}
-		default:
-			break;
-		}
 	}
 
 	if (!d->btf_ext)
 		return 0;
 
-	line_data_cur = d->btf_ext->line_info.info;
-	line_data_end = d->btf_ext->line_info.info + d->btf_ext->line_info.len;
-	rec_size = d->btf_ext->line_info.rec_size;
-
-	while (line_data_cur < line_data_end) {
-		struct btf_ext_info_sec *sec = line_data_cur;
-		struct bpf_line_info_min *line_info;
-		__u32 num_info = sec->num_info;
-
-		r = fn(&sec->sec_name_off, ctx);
-		if (r)
-			return r;
-
-		line_data_cur += sizeof(struct btf_ext_info_sec);
-		for (i = 0; i < num_info; i++) {
-			line_info = line_data_cur;
-			r = fn(&line_info->file_name_off, ctx);
-			if (r)
-				return r;
-			r = fn(&line_info->line_off, ctx);
-			if (r)
-				return r;
-			line_data_cur += rec_size;
-		}
-	}
+	r = btf_ext_visit_str_offs(d->btf_ext, fn, ctx);
+	if (r)
+		return r;
 
 	return 0;
 }
@@ -3252,10 +3148,8 @@ static int strs_dedup_remap_str_off(__u32 *str_off_ptr, void *ctx)
 {
 	struct btf_dedup *d = ctx;
 	__u32 str_off = *str_off_ptr;
-	long old_off, new_off, len;
 	const char *s;
-	void *p;
-	int err;
+	int off, err;
 
 	/* don't touch empty string or string in main BTF */
 	if (str_off == 0 || str_off < d->btf->start_str_off)
@@ -3272,29 +3166,11 @@ static int strs_dedup_remap_str_off(__u32 *str_off_ptr, void *ctx)
 			return err;
 	}
 
-	len = strlen(s) + 1;
-
-	new_off = d->strs_len;
-	p = btf_add_mem(&d->strs_data, &d->strs_cap, 1, new_off, BTF_MAX_STR_OFFSET, len);
-	if (!p)
-		return -ENOMEM;
-
-	memcpy(p, s, len);
+	off = strset__add_str(d->strs_set, s);
+	if (off < 0)
+		return off;
 
-	/* Now attempt to add the string, but only if the string with the same
-	 * contents doesn't exist already (HASHMAP_ADD strategy). If such
-	 * string exists, we'll get its offset in old_off (that's old_key).
-	 */
-	err = hashmap__insert(d->strs_hash, (void *)new_off, (void *)new_off,
-			      HASHMAP_ADD, (const void **)&old_off, NULL);
-	if (err == -EEXIST) {
-		*str_off_ptr = d->btf->start_str_off + old_off;
-	} else if (err) {
-		return err;
-	} else {
-		*str_off_ptr = d->btf->start_str_off + new_off;
-		d->strs_len += len;
-	}
+	*str_off_ptr = d->btf->start_str_off + off;
 	return 0;
 }
 
@@ -3311,39 +3187,23 @@ static int strs_dedup_remap_str_off(__u32 *str_off_ptr, void *ctx)
  */
 static int btf_dedup_strings(struct btf_dedup *d)
 {
-	char *s;
 	int err;
 
 	if (d->btf->strs_deduped)
 		return 0;
 
-	/* temporarily switch to use btf_dedup's strs_data for strings for hash
-	 * functions; later we'll just transfer hashmap to struct btf as is,
-	 * along the strs_data
-	 */
-	d->btf->strs_data_ptr = &d->strs_data;
-
-	d->strs_hash = hashmap__new(strs_hash_fn, strs_hash_equal_fn, d->btf);
-	if (IS_ERR(d->strs_hash)) {
-		err = PTR_ERR(d->strs_hash);
-		d->strs_hash = NULL;
+	d->strs_set = strset__new(BTF_MAX_STR_OFFSET, NULL, 0);
+	if (IS_ERR(d->strs_set)) {
+		err = PTR_ERR(d->strs_set);
 		goto err_out;
 	}
 
 	if (!d->btf->base_btf) {
-		s = btf_add_mem(&d->strs_data, &d->strs_cap, 1, d->strs_len, BTF_MAX_STR_OFFSET, 1);
-		if (!s)
-			return -ENOMEM;
-		/* initial empty string */
-		s[0] = 0;
-		d->strs_len = 1;
-
 		/* insert empty string; we won't be looking it up during strings
 		 * dedup, but it's good to have it for generic BTF string lookups
 		 */
-		err = hashmap__insert(d->strs_hash, (void *)0, (void *)0,
-				      HASHMAP_ADD, NULL, NULL);
-		if (err)
+		err = strset__add_str(d->strs_set, "");
+		if (err < 0)
 			goto err_out;
 	}
 
@@ -3353,28 +3213,16 @@ static int btf_dedup_strings(struct btf_dedup *d)
 		goto err_out;
 
 	/* replace BTF string data and hash with deduped ones */
-	free(d->btf->strs_data);
-	hashmap__free(d->btf->strs_hash);
-	d->btf->strs_data = d->strs_data;
-	d->btf->strs_data_cap = d->strs_cap;
-	d->btf->hdr->str_len = d->strs_len;
-	d->btf->strs_hash = d->strs_hash;
-	/* now point strs_data_ptr back to btf->strs_data */
-	d->btf->strs_data_ptr = &d->btf->strs_data;
-
-	d->strs_data = d->strs_hash = NULL;
-	d->strs_len = d->strs_cap = 0;
+	strset__free(d->btf->strs_set);
+	d->btf->hdr->str_len = strset__data_size(d->strs_set);
+	d->btf->strs_set = d->strs_set;
+	d->strs_set = NULL;
 	d->btf->strs_deduped = true;
 	return 0;
 
 err_out:
-	free(d->strs_data);
-	hashmap__free(d->strs_hash);
-	d->strs_data = d->strs_hash = NULL;
-	d->strs_len = d->strs_cap = 0;
-
-	/* restore strings pointer for existing d->btf->strs_hash back */
-	d->btf->strs_data_ptr = &d->strs_data;
+	strset__free(d->strs_set);
+	d->strs_set = NULL;
 
 	return err;
 }
@@ -4498,15 +4346,18 @@ static int btf_dedup_compact_types(struct btf_dedup *d)
  * then mapping it to a deduplicated type ID, stored in btf_dedup->hypot_map,
  * which is populated during compaction phase.
  */
-static int btf_dedup_remap_type_id(struct btf_dedup *d, __u32 type_id)
+static int btf_dedup_remap_type_id(__u32 *type_id, void *ctx)
 {
+	struct btf_dedup *d = ctx;
 	__u32 resolved_type_id, new_type_id;
 
-	resolved_type_id = resolve_type_id(d, type_id);
+	resolved_type_id = resolve_type_id(d, *type_id);
 	new_type_id = d->hypot_map[resolved_type_id];
 	if (new_type_id > BTF_MAX_NR_TYPES)
 		return -EINVAL;
-	return new_type_id;
+
+	*type_id = new_type_id;
+	return 0;
 }
 
 /*
@@ -4519,109 +4370,25 @@ static int btf_dedup_remap_type_id(struct btf_dedup *d, __u32 type_id)
  * referenced from any BTF type (e.g., struct fields, func proto args, etc) to
  * their final deduped type IDs.
  */
-static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
+static int btf_dedup_remap_types(struct btf_dedup *d)
 {
-	struct btf_type *t = btf_type_by_id(d->btf, type_id);
 	int i, r;
 
-	switch (btf_kind(t)) {
-	case BTF_KIND_INT:
-	case BTF_KIND_ENUM:
-	case BTF_KIND_FLOAT:
-		break;
-
-	case BTF_KIND_FWD:
-	case BTF_KIND_CONST:
-	case BTF_KIND_VOLATILE:
-	case BTF_KIND_RESTRICT:
-	case BTF_KIND_PTR:
-	case BTF_KIND_TYPEDEF:
-	case BTF_KIND_FUNC:
-	case BTF_KIND_VAR:
-		r = btf_dedup_remap_type_id(d, t->type);
-		if (r < 0)
-			return r;
-		t->type = r;
-		break;
-
-	case BTF_KIND_ARRAY: {
-		struct btf_array *arr_info = btf_array(t);
-
-		r = btf_dedup_remap_type_id(d, arr_info->type);
-		if (r < 0)
-			return r;
-		arr_info->type = r;
-		r = btf_dedup_remap_type_id(d, arr_info->index_type);
-		if (r < 0)
-			return r;
-		arr_info->index_type = r;
-		break;
-	}
-
-	case BTF_KIND_STRUCT:
-	case BTF_KIND_UNION: {
-		struct btf_member *member = btf_members(t);
-		__u16 vlen = btf_vlen(t);
-
-		for (i = 0; i < vlen; i++) {
-			r = btf_dedup_remap_type_id(d, member->type);
-			if (r < 0)
-				return r;
-			member->type = r;
-			member++;
-		}
-		break;
-	}
-
-	case BTF_KIND_FUNC_PROTO: {
-		struct btf_param *param = btf_params(t);
-		__u16 vlen = btf_vlen(t);
+	for (i = 0; i < d->btf->nr_types; i++) {
+		struct btf_type *t = btf_type_by_id(d->btf, d->btf->start_id + i);
 
-		r = btf_dedup_remap_type_id(d, t->type);
-		if (r < 0)
+		r = btf_type_visit_type_ids(t, btf_dedup_remap_type_id, d);
+		if (r)
 			return r;
-		t->type = r;
-
-		for (i = 0; i < vlen; i++) {
-			r = btf_dedup_remap_type_id(d, param->type);
-			if (r < 0)
-				return r;
-			param->type = r;
-			param++;
-		}
-		break;
-	}
-
-	case BTF_KIND_DATASEC: {
-		struct btf_var_secinfo *var = btf_var_secinfos(t);
-		__u16 vlen = btf_vlen(t);
-
-		for (i = 0; i < vlen; i++) {
-			r = btf_dedup_remap_type_id(d, var->type);
-			if (r < 0)
-				return r;
-			var->type = r;
-			var++;
-		}
-		break;
-	}
-
-	default:
-		return -EINVAL;
 	}
 
-	return 0;
-}
+	if (!d->btf_ext)
+		return 0;
 
-static int btf_dedup_remap_types(struct btf_dedup *d)
-{
-	int i, r;
+	r = btf_ext_visit_type_ids(d->btf_ext, btf_dedup_remap_type_id, d);
+	if (r)
+		return r;
 
-	for (i = 0; i < d->btf->nr_types; i++) {
-		r = btf_dedup_remap_type(d, d->btf->start_id + i);
-		if (r < 0)
-			return r;
-	}
 	return 0;
 }
 
@@ -4675,3 +4442,200 @@ struct btf *libbpf_find_kernel_btf(void)
 	pr_warn("failed to find valid kernel BTF\n");
 	return ERR_PTR(-ESRCH);
 }
+
+int btf_type_visit_type_ids(struct btf_type *t, type_id_visit_fn visit, void *ctx)
+{
+	int i, n, err;
+
+	switch (btf_kind(t)) {
+	case BTF_KIND_INT:
+	case BTF_KIND_FLOAT:
+	case BTF_KIND_ENUM:
+		return 0;
+
+	case BTF_KIND_FWD:
+	case BTF_KIND_CONST:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_RESTRICT:
+	case BTF_KIND_PTR:
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_FUNC:
+	case BTF_KIND_VAR:
+		return visit(&t->type, ctx);
+
+	case BTF_KIND_ARRAY: {
+		struct btf_array *a = btf_array(t);
+
+		err = visit(&a->type, ctx);
+		err = err ?: visit(&a->index_type, ctx);
+		return err;
+	}
+
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION: {
+		struct btf_member *m = btf_members(t);
+
+		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
+			err = visit(&m->type, ctx);
+			if (err)
+				return err;
+		}
+		return 0;
+	}
+
+	case BTF_KIND_FUNC_PROTO: {
+		struct btf_param *m = btf_params(t);
+
+		err = visit(&t->type, ctx);
+		if (err)
+			return err;
+		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
+			err = visit(&m->type, ctx);
+			if (err)
+				return err;
+		}
+		return 0;
+	}
+
+	case BTF_KIND_DATASEC: {
+		struct btf_var_secinfo *m = btf_var_secinfos(t);
+
+		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
+			err = visit(&m->type, ctx);
+			if (err)
+				return err;
+		}
+		return 0;
+	}
+
+	default:
+		return -EINVAL;
+	}
+}
+
+int btf_type_visit_str_offs(struct btf_type *t, str_off_visit_fn visit, void *ctx)
+{
+	int i, n, err;
+
+	err = visit(&t->name_off, ctx);
+	if (err)
+		return err;
+
+	switch (btf_kind(t)) {
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION: {
+		struct btf_member *m = btf_members(t);
+
+		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
+			err = visit(&m->name_off, ctx);
+			if (err)
+				return err;
+		}
+		break;
+	}
+	case BTF_KIND_ENUM: {
+		struct btf_enum *m = btf_enum(t);
+
+		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
+			err = visit(&m->name_off, ctx);
+			if (err)
+				return err;
+		}
+		break;
+	}
+	case BTF_KIND_FUNC_PROTO: {
+		struct btf_param *m = btf_params(t);
+
+		for (i = 0, n = btf_vlen(t); i < n; i++, m++) {
+			err = visit(&m->name_off, ctx);
+			if (err)
+				return err;
+		}
+		break;
+	}
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+int btf_ext_visit_type_ids(struct btf_ext *btf_ext, type_id_visit_fn visit, void *ctx)
+{
+	const struct btf_ext_info *seg;
+	struct btf_ext_info_sec *sec;
+	int i, err;
+
+	seg = &btf_ext->func_info;
+	for_each_btf_ext_sec(seg, sec) {
+		struct bpf_func_info_min *rec;
+
+		for_each_btf_ext_rec(seg, sec, i, rec) {
+			err = visit(&rec->type_id, ctx);
+			if (err < 0)
+				return err;
+		}
+	}
+
+	seg = &btf_ext->core_relo_info;
+	for_each_btf_ext_sec(seg, sec) {
+		struct bpf_core_relo *rec;
+
+		for_each_btf_ext_rec(seg, sec, i, rec) {
+			err = visit(&rec->type_id, ctx);
+			if (err < 0)
+				return err;
+		}
+	}
+
+	return 0;
+}
+
+int btf_ext_visit_str_offs(struct btf_ext *btf_ext, str_off_visit_fn visit, void *ctx)
+{
+	const struct btf_ext_info *seg;
+	struct btf_ext_info_sec *sec;
+	int i, err;
+
+	seg = &btf_ext->func_info;
+	for_each_btf_ext_sec(seg, sec) {
+		err = visit(&sec->sec_name_off, ctx);
+		if (err)
+			return err;
+	}
+
+	seg = &btf_ext->line_info;
+	for_each_btf_ext_sec(seg, sec) {
+		struct bpf_line_info_min *rec;
+
+		err = visit(&sec->sec_name_off, ctx);
+		if (err)
+			return err;
+
+		for_each_btf_ext_rec(seg, sec, i, rec) {
+			err = visit(&rec->file_name_off, ctx);
+			if (err)
+				return err;
+			err = visit(&rec->line_off, ctx);
+			if (err)
+				return err;
+		}
+	}
+
+	seg = &btf_ext->core_relo_info;
+	for_each_btf_ext_sec(seg, sec) {
+		struct bpf_core_relo *rec;
+
+		err = visit(&sec->sec_name_off, ctx);
+		if (err)
+			return err;
+
+		for_each_btf_ext_rec(seg, sec, i, rec) {
+			err = visit(&rec->access_str_off, ctx);
+			if (err)
+				return err;
+		}
+	}
+
+	return 0;
+}
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 029a9cfc8c2d..b54f1c3ebd57 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -93,6 +93,8 @@ LIBBPF_API struct btf *libbpf_find_kernel_btf(void);
 
 LIBBPF_API int btf__find_str(struct btf *btf, const char *s);
 LIBBPF_API int btf__add_str(struct btf *btf, const char *s);
+LIBBPF_API int btf__add_type(struct btf *btf, const struct btf *src_btf,
+			     const struct btf_type *src_type);
 
 LIBBPF_API int btf__add_int(struct btf *btf, const char *name, size_t byte_sz, int encoding);
 LIBBPF_API int btf__add_float(struct btf *btf, const char *name, size_t byte_sz);
@@ -174,6 +176,7 @@ struct btf_dump_emit_type_decl_opts {
 	int indent_level;
 	/* strip all the const/volatile/restrict mods */
 	bool strip_mods;
+	size_t :0;
 };
 #define btf_dump_emit_type_decl_opts__last_field strip_mods
 
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index 7b53a484860f..5e2809d685bf 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -166,11 +166,11 @@ static int btf_dump_resize(struct btf_dump *d)
 	if (last_id <= d->last_id)
 		return 0;
 
-	if (btf_ensure_mem((void **)&d->type_states, &d->type_states_cap,
-			   sizeof(*d->type_states), last_id + 1))
+	if (libbpf_ensure_mem((void **)&d->type_states, &d->type_states_cap,
+			      sizeof(*d->type_states), last_id + 1))
 		return -ENOMEM;
-	if (btf_ensure_mem((void **)&d->cached_names, &d->cached_names_cap,
-			   sizeof(*d->cached_names), last_id + 1))
+	if (libbpf_ensure_mem((void **)&d->cached_names, &d->cached_names_cap,
+			      sizeof(*d->cached_names), last_id + 1))
 		return -ENOMEM;
 
 	if (d->last_id == 0) {
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 8d610259f4be..c0fd2c718a7d 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -55,10 +55,6 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 
-#ifndef EM_BPF
-#define EM_BPF 247
-#endif
-
 #ifndef BPF_FS_MAGIC
 #define BPF_FS_MAGIC		0xcafe4a11
 #endif
@@ -1134,11 +1130,6 @@ static void bpf_object__elf_finish(struct bpf_object *obj)
 	obj->efile.obj_buf_sz = 0;
 }
 
-/* if libelf is old and doesn't support mmap(), fall back to read() */
-#ifndef ELF_C_READ_MMAP
-#define ELF_C_READ_MMAP ELF_C_READ
-#endif
-
 static int bpf_object__elf_init(struct bpf_object *obj)
 {
 	int err = 0;
@@ -2808,7 +2799,7 @@ static bool ignore_elf_section(GElf_Shdr *hdr, const char *name)
 		return true;
 
 	/* ignore .llvm_addrsig section as well */
-	if (hdr->sh_type == 0x6FFF4C03 /* SHT_LLVM_ADDRSIG */)
+	if (hdr->sh_type == SHT_LLVM_ADDRSIG)
 		return true;
 
 	/* no subprograms will lead to an empty .text section, ignore it */
@@ -4868,8 +4859,8 @@ static int load_module_btfs(struct bpf_object *obj)
 			goto err_out;
 		}
 
-		err = btf_ensure_mem((void **)&obj->btf_modules, &obj->btf_module_cap,
-				     sizeof(*obj->btf_modules), obj->btf_module_cnt + 1);
+		err = libbpf_ensure_mem((void **)&obj->btf_modules, &obj->btf_module_cap,
+				        sizeof(*obj->btf_modules), obj->btf_module_cnt + 1);
 		if (err)
 			goto err_out;
 
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 3c35eb401931..a1a424b9b8ff 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -507,6 +507,7 @@ struct xdp_link_info {
 struct bpf_xdp_set_link_opts {
 	size_t sz;
 	int old_fd;
+	size_t :0;
 };
 #define bpf_xdp_set_link_opts__last_field old_fd
 
@@ -759,6 +760,19 @@ enum libbpf_tristate {
 	TRI_MODULE = 2,
 };
 
+struct bpf_linker_opts {
+	/* size of this struct, for forward/backward compatiblity */
+	size_t sz;
+};
+#define bpf_linker_opts__last_field sz
+
+struct bpf_linker;
+
+LIBBPF_API struct bpf_linker *bpf_linker__new(const char *filename, struct bpf_linker_opts *opts);
+LIBBPF_API int bpf_linker__add_file(struct bpf_linker *linker, const char *filename);
+LIBBPF_API int bpf_linker__finalize(struct bpf_linker *linker);
+LIBBPF_API void bpf_linker__free(struct bpf_linker *linker);
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index ec898f464ab9..279ae861f568 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -354,4 +354,9 @@ LIBBPF_0.3.0 {
 LIBBPF_0.4.0 {
 	global:
 		btf__add_float;
+		btf__add_type;
+		bpf_linker__add_file;
+		bpf_linker__finalize;
+		bpf_linker__free;
+		bpf_linker__new;
 } LIBBPF_0.3.0;
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 343f6eb05637..6017902c687e 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -20,6 +20,26 @@
 
 #include "libbpf.h"
 
+#ifndef EM_BPF
+#define EM_BPF 247
+#endif
+
+#ifndef R_BPF_64_64
+#define R_BPF_64_64 1
+#endif
+#ifndef R_BPF_64_32
+#define R_BPF_64_32 10
+#endif
+
+#ifndef SHT_LLVM_ADDRSIG
+#define SHT_LLVM_ADDRSIG 0x6FFF4C03
+#endif
+
+/* if libelf is old and doesn't support mmap(), fall back to read() */
+#ifndef ELF_C_READ_MMAP
+#define ELF_C_READ_MMAP ELF_C_READ
+#endif
+
 #define BTF_INFO_ENC(kind, kind_flag, vlen) \
 	((!!(kind_flag) << 31) | ((kind) << 24) | ((vlen) & BTF_MAX_VLEN))
 #define BTF_TYPE_ENC(name, info, size_or_type) (name), (info), (size_or_type)
@@ -107,9 +127,14 @@ static inline void *libbpf_reallocarray(void *ptr, size_t nmemb, size_t size)
 	return realloc(ptr, total);
 }
 
-void *btf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
-		  size_t cur_cnt, size_t max_cnt, size_t add_cnt);
-int btf_ensure_mem(void **data, size_t *cap_cnt, size_t elem_sz, size_t need_cnt);
+struct btf;
+struct btf_type;
+
+struct btf_type *btf_type_by_id(struct btf *btf, __u32 type_id);
+
+void *libbpf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
+		     size_t cur_cnt, size_t max_cnt, size_t add_cnt);
+int libbpf_ensure_mem(void **data, size_t *cap_cnt, size_t elem_sz, size_t need_cnt);
 
 static inline bool libbpf_validate_opts(const char *opts,
 					size_t opts_sz, size_t user_sz,
@@ -351,4 +376,11 @@ struct bpf_core_relo {
 	enum bpf_core_relo_kind kind;
 };
 
+typedef int (*type_id_visit_fn)(__u32 *type_id, void *ctx);
+typedef int (*str_off_visit_fn)(__u32 *str_off, void *ctx);
+int btf_type_visit_type_ids(struct btf_type *t, type_id_visit_fn visit, void *ctx);
+int btf_type_visit_str_offs(struct btf_type *t, str_off_visit_fn visit, void *ctx);
+int btf_ext_visit_type_ids(struct btf_ext *btf_ext, type_id_visit_fn visit, void *ctx);
+int btf_ext_visit_str_offs(struct btf_ext *btf_ext, str_off_visit_fn visit, void *ctx);
+
 #endif /* __LIBBPF_LIBBPF_INTERNAL_H */
diff --git a/tools/lib/bpf/libbpf_util.h b/tools/lib/bpf/libbpf_util.h
deleted file mode 100644
index cfbcfc063c81..000000000000
--- a/tools/lib/bpf/libbpf_util.h
+++ /dev/null
@@ -1,75 +0,0 @@
-/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
-/* Copyright (c) 2019 Facebook */
-
-#ifndef __LIBBPF_LIBBPF_UTIL_H
-#define __LIBBPF_LIBBPF_UTIL_H
-
-#include <stdbool.h>
-#include <linux/compiler.h>
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-/* Use these barrier functions instead of smp_[rw]mb() when they are
- * used in a libbpf header file. That way they can be built into the
- * application that uses libbpf.
- */
-#if defined(__i386__) || defined(__x86_64__)
-# define libbpf_smp_store_release(p, v)					\
-	do {								\
-		asm volatile("" : : : "memory");			\
-		WRITE_ONCE(*p, v);					\
-	} while (0)
-# define libbpf_smp_load_acquire(p)					\
-	({								\
-		typeof(*p) ___p1 = READ_ONCE(*p);			\
-		asm volatile("" : : : "memory");			\
-		___p1;							\
-	})
-#elif defined(__aarch64__)
-# define libbpf_smp_store_release(p, v)					\
-		asm volatile ("stlr %w1, %0" : "=Q" (*p) : "r" (v) : "memory")
-# define libbpf_smp_load_acquire(p)					\
-	({								\
-		typeof(*p) ___p1;					\
-		asm volatile ("ldar %w0, %1"				\
-			      : "=r" (___p1) : "Q" (*p) : "memory");	\
-		___p1;							\
-	})
-#elif defined(__riscv)
-# define libbpf_smp_store_release(p, v)					\
-	do {								\
-		asm volatile ("fence rw,w" : : : "memory");		\
-		WRITE_ONCE(*p, v);					\
-	} while (0)
-# define libbpf_smp_load_acquire(p)					\
-	({								\
-		typeof(*p) ___p1 = READ_ONCE(*p);			\
-		asm volatile ("fence r,rw" : : : "memory");		\
-		___p1;							\
-	})
-#endif
-
-#ifndef libbpf_smp_store_release
-#define libbpf_smp_store_release(p, v)					\
-	do {								\
-		__sync_synchronize();					\
-		WRITE_ONCE(*p, v);					\
-	} while (0)
-#endif
-
-#ifndef libbpf_smp_load_acquire
-#define libbpf_smp_load_acquire(p)					\
-	({								\
-		typeof(*p) ___p1 = READ_ONCE(*p);			\
-		__sync_synchronize();					\
-		___p1;							\
-	})
-#endif
-
-#ifdef __cplusplus
-} /* extern "C" */
-#endif
-
-#endif
diff --git a/tools/lib/bpf/linker.c b/tools/lib/bpf/linker.c
new file mode 100644
index 000000000000..5e0aa2f2c0ca
--- /dev/null
+++ b/tools/lib/bpf/linker.c
@@ -0,0 +1,1944 @@
+// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
+/*
+ * BPF static linker
+ *
+ * Copyright (c) 2021 Facebook
+ */
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <errno.h>
+#include <linux/err.h>
+#include <linux/btf.h>
+#include <elf.h>
+#include <libelf.h>
+#include <gelf.h>
+#include <fcntl.h>
+#include "libbpf.h"
+#include "btf.h"
+#include "libbpf_internal.h"
+#include "strset.h"
+
+struct src_sec {
+	const char *sec_name;
+	/* positional (not necessarily ELF) index in an array of sections */
+	int id;
+	/* positional (not necessarily ELF) index of a matching section in a final object file */
+	int dst_id;
+	/* section data offset in a matching output section */
+	int dst_off;
+	/* whether section is omitted from the final ELF file */
+	bool skipped;
+	/* whether section is an ephemeral section, not mapped to an ELF section */
+	bool ephemeral;
+
+	/* ELF info */
+	size_t sec_idx;
+	Elf_Scn *scn;
+	Elf64_Shdr *shdr;
+	Elf_Data *data;
+
+	/* corresponding BTF DATASEC type ID */
+	int sec_type_id;
+};
+
+struct src_obj {
+	const char *filename;
+	int fd;
+	Elf *elf;
+	/* Section header strings section index */
+	size_t shstrs_sec_idx;
+	/* SYMTAB section index */
+	size_t symtab_sec_idx;
+
+	struct btf *btf;
+	struct btf_ext *btf_ext;
+
+	/* List of sections (including ephemeral). Slot zero is unused. */
+	struct src_sec *secs;
+	int sec_cnt;
+
+	/* mapping of symbol indices from src to dst ELF */
+	int *sym_map;
+	/* mapping from the src BTF type IDs to dst ones */
+	int *btf_type_map;
+};
+
+/* single .BTF.ext data section */
+struct btf_ext_sec_data {
+	size_t rec_cnt;
+	__u32 rec_sz;
+	void *recs;
+};
+
+struct dst_sec {
+	char *sec_name;
+	/* positional (not necessarily ELF) index in an array of sections */
+	int id;
+
+	/* ELF info */
+	size_t sec_idx;
+	Elf_Scn *scn;
+	Elf64_Shdr *shdr;
+	Elf_Data *data;
+
+	/* final output section size */
+	int sec_sz;
+	/* final output contents of the section */
+	void *raw_data;
+
+	/* corresponding STT_SECTION symbol index in SYMTAB */
+	int sec_sym_idx;
+
+	/* section's DATASEC variable info, emitted on BTF finalization */
+	int sec_var_cnt;
+	struct btf_var_secinfo *sec_vars;
+
+	/* section's .BTF.ext data */
+	struct btf_ext_sec_data func_info;
+	struct btf_ext_sec_data line_info;
+	struct btf_ext_sec_data core_relo_info;
+};
+
+struct bpf_linker {
+	char *filename;
+	int fd;
+	Elf *elf;
+	Elf64_Ehdr *elf_hdr;
+
+	/* Output sections metadata */
+	struct dst_sec *secs;
+	int sec_cnt;
+
+	struct strset *strtab_strs; /* STRTAB unique strings */
+	size_t strtab_sec_idx; /* STRTAB section index */
+	size_t symtab_sec_idx; /* SYMTAB section index */
+
+	struct btf *btf;
+	struct btf_ext *btf_ext;
+};
+
+#define pr_warn_elf(fmt, ...)									\
+do {												\
+	libbpf_print(LIBBPF_WARN, "libbpf: " fmt ": %s\n", ##__VA_ARGS__, elf_errmsg(-1));	\
+} while (0)
+
+static int init_output_elf(struct bpf_linker *linker, const char *file);
+
+static int linker_load_obj_file(struct bpf_linker *linker, const char *filename, struct src_obj *obj);
+static int linker_sanity_check_elf(struct src_obj *obj);
+static int linker_sanity_check_btf(struct src_obj *obj);
+static int linker_sanity_check_btf_ext(struct src_obj *obj);
+static int linker_fixup_btf(struct src_obj *obj);
+static int linker_append_sec_data(struct bpf_linker *linker, struct src_obj *obj);
+static int linker_append_elf_syms(struct bpf_linker *linker, struct src_obj *obj);
+static int linker_append_elf_relos(struct bpf_linker *linker, struct src_obj *obj);
+static int linker_append_btf(struct bpf_linker *linker, struct src_obj *obj);
+static int linker_append_btf_ext(struct bpf_linker *linker, struct src_obj *obj);
+
+static int finalize_btf(struct bpf_linker *linker);
+static int finalize_btf_ext(struct bpf_linker *linker);
+
+void bpf_linker__free(struct bpf_linker *linker)
+{
+	int i;
+
+	if (!linker)
+		return;
+
+	free(linker->filename);
+
+	if (linker->elf)
+		elf_end(linker->elf);
+
+	if (linker->fd >= 0)
+		close(linker->fd);
+
+	strset__free(linker->strtab_strs);
+
+	btf__free(linker->btf);
+	btf_ext__free(linker->btf_ext);
+
+	for (i = 1; i < linker->sec_cnt; i++) {
+		struct dst_sec *sec = &linker->secs[i];
+
+		free(sec->sec_name);
+		free(sec->raw_data);
+		free(sec->sec_vars);
+
+		free(sec->func_info.recs);
+		free(sec->line_info.recs);
+		free(sec->core_relo_info.recs);
+	}
+	free(linker->secs);
+
+	free(linker);
+}
+
+struct bpf_linker *bpf_linker__new(const char *filename, struct bpf_linker_opts *opts)
+{
+	struct bpf_linker *linker;
+	int err;
+
+	if (!OPTS_VALID(opts, bpf_linker_opts))
+		return NULL;
+
+	if (elf_version(EV_CURRENT) == EV_NONE) {
+		pr_warn_elf("libelf initialization failed");
+		return NULL;
+	}
+
+	linker = calloc(1, sizeof(*linker));
+	if (!linker)
+		return NULL;
+
+	linker->fd = -1;
+
+	err = init_output_elf(linker, filename);
+	if (err)
+		goto err_out;
+
+	return linker;
+
+err_out:
+	bpf_linker__free(linker);
+	return NULL;
+}
+
+static struct dst_sec *add_dst_sec(struct bpf_linker *linker, const char *sec_name)
+{
+	struct dst_sec *secs = linker->secs, *sec;
+	size_t new_cnt = linker->sec_cnt ? linker->sec_cnt + 1 : 2;
+
+	secs = libbpf_reallocarray(secs, new_cnt, sizeof(*secs));
+	if (!secs)
+		return NULL;
+
+	/* zero out newly allocated memory */
+	memset(secs + linker->sec_cnt, 0, (new_cnt - linker->sec_cnt) * sizeof(*secs));
+
+	linker->secs = secs;
+	linker->sec_cnt = new_cnt;
+
+	sec = &linker->secs[new_cnt - 1];
+	sec->id = new_cnt - 1;
+	sec->sec_name = strdup(sec_name);
+	if (!sec->sec_name)
+		return NULL;
+
+	return sec;
+}
+
+static Elf64_Sym *add_new_sym(struct bpf_linker *linker, size_t *sym_idx)
+{
+	struct dst_sec *symtab = &linker->secs[linker->symtab_sec_idx];
+	Elf64_Sym *syms, *sym;
+	size_t sym_cnt = symtab->sec_sz / sizeof(*sym);
+
+	syms = libbpf_reallocarray(symtab->raw_data, sym_cnt + 1, sizeof(*sym));
+	if (!syms)
+		return NULL;
+
+	sym = &syms[sym_cnt];
+	memset(sym, 0, sizeof(*sym));
+
+	symtab->raw_data = syms;
+	symtab->sec_sz += sizeof(*sym);
+	symtab->shdr->sh_size += sizeof(*sym);
+	symtab->data->d_size += sizeof(*sym);
+
+	if (sym_idx)
+		*sym_idx = sym_cnt;
+
+	return sym;
+}
+
+static int init_output_elf(struct bpf_linker *linker, const char *file)
+{
+	int err, str_off;
+	Elf64_Sym *init_sym;
+	struct dst_sec *sec;
+
+	linker->filename = strdup(file);
+	if (!linker->filename)
+		return -ENOMEM;
+
+	linker->fd = open(file, O_WRONLY | O_CREAT | O_TRUNC, 0644);
+	if (linker->fd < 0) {
+		err = -errno;
+		pr_warn("failed to create '%s': %d\n", file, err);
+		return err;
+	}
+
+	linker->elf = elf_begin(linker->fd, ELF_C_WRITE, NULL);
+	if (!linker->elf) {
+		pr_warn_elf("failed to create ELF object");
+		return -EINVAL;
+	}
+
+	/* ELF header */
+	linker->elf_hdr = elf64_newehdr(linker->elf);
+	if (!linker->elf_hdr){
+		pr_warn_elf("failed to create ELF header");
+		return -EINVAL;
+	}
+
+	linker->elf_hdr->e_machine = EM_BPF;
+	linker->elf_hdr->e_type = ET_REL;
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+	linker->elf_hdr->e_ident[EI_DATA] = ELFDATA2LSB;
+#elif __BYTE_ORDER == __BIG_ENDIAN
+	linker->elf_hdr->e_ident[EI_DATA] = ELFDATA2MSB;
+#else
+#error "Unknown __BYTE_ORDER"
+#endif
+
+	/* STRTAB */
+	/* initialize strset with an empty string to conform to ELF */
+	linker->strtab_strs = strset__new(INT_MAX, "", sizeof(""));
+	if (libbpf_get_error(linker->strtab_strs))
+		return libbpf_get_error(linker->strtab_strs);
+
+	sec = add_dst_sec(linker, ".strtab");
+	if (!sec)
+		return -ENOMEM;
+
+	sec->scn = elf_newscn(linker->elf);
+	if (!sec->scn) {
+		pr_warn_elf("failed to create STRTAB section");
+		return -EINVAL;
+	}
+
+	sec->shdr = elf64_getshdr(sec->scn);
+	if (!sec->shdr)
+		return -EINVAL;
+
+	sec->data = elf_newdata(sec->scn);
+	if (!sec->data) {
+		pr_warn_elf("failed to create STRTAB data");
+		return -EINVAL;
+	}
+
+	str_off = strset__add_str(linker->strtab_strs, sec->sec_name);
+	if (str_off < 0)
+		return str_off;
+
+	sec->sec_idx = elf_ndxscn(sec->scn);
+	linker->elf_hdr->e_shstrndx = sec->sec_idx;
+	linker->strtab_sec_idx = sec->sec_idx;
+
+	sec->shdr->sh_name = str_off;
+	sec->shdr->sh_type = SHT_STRTAB;
+	sec->shdr->sh_flags = SHF_STRINGS;
+	sec->shdr->sh_offset = 0;
+	sec->shdr->sh_link = 0;
+	sec->shdr->sh_info = 0;
+	sec->shdr->sh_addralign = 1;
+	sec->shdr->sh_size = sec->sec_sz = 0;
+	sec->shdr->sh_entsize = 0;
+
+	/* SYMTAB */
+	sec = add_dst_sec(linker, ".symtab");
+	if (!sec)
+		return -ENOMEM;
+
+	sec->scn = elf_newscn(linker->elf);
+	if (!sec->scn) {
+		pr_warn_elf("failed to create SYMTAB section");
+		return -EINVAL;
+	}
+
+	sec->shdr = elf64_getshdr(sec->scn);
+	if (!sec->shdr)
+		return -EINVAL;
+
+	sec->data = elf_newdata(sec->scn);
+	if (!sec->data) {
+		pr_warn_elf("failed to create SYMTAB data");
+		return -EINVAL;
+	}
+
+	str_off = strset__add_str(linker->strtab_strs, sec->sec_name);
+	if (str_off < 0)
+		return str_off;
+
+	sec->sec_idx = elf_ndxscn(sec->scn);
+	linker->symtab_sec_idx = sec->sec_idx;
+
+	sec->shdr->sh_name = str_off;
+	sec->shdr->sh_type = SHT_SYMTAB;
+	sec->shdr->sh_flags = 0;
+	sec->shdr->sh_offset = 0;
+	sec->shdr->sh_link = linker->strtab_sec_idx;
+	/* sh_info should be one greater than the index of the last local
+	 * symbol (i.e., binding is STB_LOCAL). But why and who cares?
+	 */
+	sec->shdr->sh_info = 0;
+	sec->shdr->sh_addralign = 8;
+	sec->shdr->sh_entsize = sizeof(Elf64_Sym);
+
+	/* .BTF */
+	linker->btf = btf__new_empty();
+	err = libbpf_get_error(linker->btf);
+	if (err)
+		return err;
+
+	/* add the special all-zero symbol */
+	init_sym = add_new_sym(linker, NULL);
+	if (!init_sym)
+		return -EINVAL;
+
+	init_sym->st_name = 0;
+	init_sym->st_info = 0;
+	init_sym->st_other = 0;
+	init_sym->st_shndx = SHN_UNDEF;
+	init_sym->st_value = 0;
+	init_sym->st_size = 0;
+
+	return 0;
+}
+
+int bpf_linker__add_file(struct bpf_linker *linker, const char *filename)
+{
+	struct src_obj obj = {};
+	int err = 0;
+
+	if (!linker->elf)
+		return -EINVAL;
+
+	err = err ?: linker_load_obj_file(linker, filename, &obj);
+	err = err ?: linker_append_sec_data(linker, &obj);
+	err = err ?: linker_append_elf_syms(linker, &obj);
+	err = err ?: linker_append_elf_relos(linker, &obj);
+	err = err ?: linker_append_btf(linker, &obj);
+	err = err ?: linker_append_btf_ext(linker, &obj);
+
+	/* free up src_obj resources */
+	free(obj.btf_type_map);
+	btf__free(obj.btf);
+	btf_ext__free(obj.btf_ext);
+	free(obj.secs);
+	free(obj.sym_map);
+	if (obj.elf)
+		elf_end(obj.elf);
+	if (obj.fd >= 0)
+		close(obj.fd);
+
+	return err;
+}
+
+static bool is_dwarf_sec_name(const char *name)
+{
+	/* approximation, but the actual list is too long */
+	return strncmp(name, ".debug_", sizeof(".debug_") - 1) == 0;
+}
+
+static bool is_ignored_sec(struct src_sec *sec)
+{
+	Elf64_Shdr *shdr = sec->shdr;
+	const char *name = sec->sec_name;
+
+	/* no special handling of .strtab */
+	if (shdr->sh_type == SHT_STRTAB)
+		return true;
+
+	/* ignore .llvm_addrsig section as well */
+	if (shdr->sh_type == SHT_LLVM_ADDRSIG)
+		return true;
+
+	/* no subprograms will lead to an empty .text section, ignore it */
+	if (shdr->sh_type == SHT_PROGBITS && shdr->sh_size == 0 &&
+	    strcmp(sec->sec_name, ".text") == 0)
+		return true;
+
+	/* DWARF sections */
+	if (is_dwarf_sec_name(sec->sec_name))
+		return true;
+
+	if (strncmp(name, ".rel", sizeof(".rel") - 1) == 0) {
+		name += sizeof(".rel") - 1;
+		/* DWARF section relocations */
+		if (is_dwarf_sec_name(name))
+			return true;
+
+		/* .BTF and .BTF.ext don't need relocations */
+		if (strcmp(name, BTF_ELF_SEC) == 0 ||
+		    strcmp(name, BTF_EXT_ELF_SEC) == 0)
+			return true;
+	}
+
+	return false;
+}
+
+static struct src_sec *add_src_sec(struct src_obj *obj, const char *sec_name)
+{
+	struct src_sec *secs = obj->secs, *sec;
+	size_t new_cnt = obj->sec_cnt ? obj->sec_cnt + 1 : 2;
+
+	secs = libbpf_reallocarray(secs, new_cnt, sizeof(*secs));
+	if (!secs)
+		return NULL;
+
+	/* zero out newly allocated memory */
+	memset(secs + obj->sec_cnt, 0, (new_cnt - obj->sec_cnt) * sizeof(*secs));
+
+	obj->secs = secs;
+	obj->sec_cnt = new_cnt;
+
+	sec = &obj->secs[new_cnt - 1];
+	sec->id = new_cnt - 1;
+	sec->sec_name = sec_name;
+
+	return sec;
+}
+
+static int linker_load_obj_file(struct bpf_linker *linker, const char *filename, struct src_obj *obj)
+{
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+	const int host_endianness = ELFDATA2LSB;
+#elif __BYTE_ORDER == __BIG_ENDIAN
+	const int host_endianness = ELFDATA2MSB;
+#else
+#error "Unknown __BYTE_ORDER"
+#endif
+	int err = 0;
+	Elf_Scn *scn;
+	Elf_Data *data;
+	Elf64_Ehdr *ehdr;
+	Elf64_Shdr *shdr;
+	struct src_sec *sec;
+
+	pr_debug("linker: adding object file '%s'...\n", filename);
+
+	obj->filename = filename;
+
+	obj->fd = open(filename, O_RDONLY);
+	if (obj->fd < 0) {
+		err = -errno;
+		pr_warn("failed to open file '%s': %d\n", filename, err);
+		return err;
+	}
+	obj->elf = elf_begin(obj->fd, ELF_C_READ_MMAP, NULL);
+	if (!obj->elf) {
+		err = -errno;
+		pr_warn_elf("failed to parse ELF file '%s'", filename);
+		return err;
+	}
+
+	/* Sanity check ELF file high-level properties */
+	ehdr = elf64_getehdr(obj->elf);
+	if (!ehdr) {
+		err = -errno;
+		pr_warn_elf("failed to get ELF header for %s", filename);
+		return err;
+	}
+	if (ehdr->e_ident[EI_DATA] != host_endianness) {
+		err = -EOPNOTSUPP;
+		pr_warn_elf("unsupported byte order of ELF file %s", filename);
+		return err;
+	}
+	if (ehdr->e_type != ET_REL
+	    || ehdr->e_machine != EM_BPF
+	    || ehdr->e_ident[EI_CLASS] != ELFCLASS64) {
+		err = -EOPNOTSUPP;
+		pr_warn_elf("unsupported kind of ELF file %s", filename);
+		return err;
+	}
+
+	if (elf_getshdrstrndx(obj->elf, &obj->shstrs_sec_idx)) {
+		err = -errno;
+		pr_warn_elf("failed to get SHSTRTAB section index for %s", filename);
+		return err;
+	}
+
+	scn = NULL;
+	while ((scn = elf_nextscn(obj->elf, scn)) != NULL) {
+		size_t sec_idx = elf_ndxscn(scn);
+		const char *sec_name;
+
+		shdr = elf64_getshdr(scn);
+		if (!shdr) {
+			err = -errno;
+			pr_warn_elf("failed to get section #%zu header for %s",
+				    sec_idx, filename);
+			return err;
+		}
+
+		sec_name = elf_strptr(obj->elf, obj->shstrs_sec_idx, shdr->sh_name);
+		if (!sec_name) {
+			err = -errno;
+			pr_warn_elf("failed to get section #%zu name for %s",
+				    sec_idx, filename);
+			return err;
+		}
+
+		data = elf_getdata(scn, 0);
+		if (!data) {
+			err = -errno;
+			pr_warn_elf("failed to get section #%zu (%s) data from %s",
+				    sec_idx, sec_name, filename);
+			return err;
+		}
+
+		sec = add_src_sec(obj, sec_name);
+		if (!sec)
+			return -ENOMEM;
+
+		sec->scn = scn;
+		sec->shdr = shdr;
+		sec->data = data;
+		sec->sec_idx = elf_ndxscn(scn);
+
+		if (is_ignored_sec(sec)) {
+			sec->skipped = true;
+			continue;
+		}
+
+		switch (shdr->sh_type) {
+		case SHT_SYMTAB:
+			if (obj->symtab_sec_idx) {
+				err = -EOPNOTSUPP;
+				pr_warn("multiple SYMTAB sections found, not supported\n");
+				return err;
+			}
+			obj->symtab_sec_idx = sec_idx;
+			break;
+		case SHT_STRTAB:
+			/* we'll construct our own string table */
+			break;
+		case SHT_PROGBITS:
+			if (strcmp(sec_name, BTF_ELF_SEC) == 0) {
+				obj->btf = btf__new(data->d_buf, shdr->sh_size);
+				err = libbpf_get_error(obj->btf);
+				if (err) {
+					pr_warn("failed to parse .BTF from %s: %d\n", filename, err);
+					return err;
+				}
+				sec->skipped = true;
+				continue;
+			}
+			if (strcmp(sec_name, BTF_EXT_ELF_SEC) == 0) {
+				obj->btf_ext = btf_ext__new(data->d_buf, shdr->sh_size);
+				err = libbpf_get_error(obj->btf_ext);
+				if (err) {
+					pr_warn("failed to parse .BTF.ext from '%s': %d\n", filename, err);
+					return err;
+				}
+				sec->skipped = true;
+				continue;
+			}
+
+			/* data & code */
+			break;
+		case SHT_NOBITS:
+			/* BSS */
+			break;
+		case SHT_REL:
+			/* relocations */
+			break;
+		default:
+			pr_warn("unrecognized section #%zu (%s) in %s\n",
+				sec_idx, sec_name, filename);
+			err = -EINVAL;
+			return err;
+		}
+	}
+
+	err = err ?: linker_sanity_check_elf(obj);
+	err = err ?: linker_sanity_check_btf(obj);
+	err = err ?: linker_sanity_check_btf_ext(obj);
+	err = err ?: linker_fixup_btf(obj);
+
+	return err;
+}
+
+static bool is_pow_of_2(size_t x)
+{
+	return x && (x & (x - 1)) == 0;
+}
+
+static int linker_sanity_check_elf(struct src_obj *obj)
+{
+	struct src_sec *sec, *link_sec;
+	int i, j, n;
+
+	if (!obj->symtab_sec_idx) {
+		pr_warn("ELF is missing SYMTAB section in %s\n", obj->filename);
+		return -EINVAL;
+	}
+	if (!obj->shstrs_sec_idx) {
+		pr_warn("ELF is missing section headers STRTAB section in %s\n", obj->filename);
+		return -EINVAL;
+	}
+
+	for (i = 1; i < obj->sec_cnt; i++) {
+		sec = &obj->secs[i];
+
+		if (sec->sec_name[0] == '\0') {
+			pr_warn("ELF section #%zu has empty name in %s\n", sec->sec_idx, obj->filename);
+			return -EINVAL;
+		}
+
+		if (sec->shdr->sh_addralign && !is_pow_of_2(sec->shdr->sh_addralign))
+			return -EINVAL;
+		if (sec->shdr->sh_addralign != sec->data->d_align)
+			return -EINVAL;
+
+		if (sec->shdr->sh_size != sec->data->d_size)
+			return -EINVAL;
+
+		switch (sec->shdr->sh_type) {
+		case SHT_SYMTAB: {
+			Elf64_Sym *sym;
+
+			if (sec->shdr->sh_entsize != sizeof(Elf64_Sym))
+				return -EINVAL;
+			if (sec->shdr->sh_size % sec->shdr->sh_entsize != 0)
+				return -EINVAL;
+
+			if (!sec->shdr->sh_link || sec->shdr->sh_link >= obj->sec_cnt) {
+				pr_warn("ELF SYMTAB section #%zu points to missing STRTAB section #%zu in %s\n",
+					sec->sec_idx, (size_t)sec->shdr->sh_link, obj->filename);
+				return -EINVAL;
+			}
+			link_sec = &obj->secs[sec->shdr->sh_link];
+			if (link_sec->shdr->sh_type != SHT_STRTAB) {
+				pr_warn("ELF SYMTAB section #%zu points to invalid STRTAB section #%zu in %s\n",
+					sec->sec_idx, (size_t)sec->shdr->sh_link, obj->filename);
+				return -EINVAL;
+			}
+
+			n = sec->shdr->sh_size / sec->shdr->sh_entsize;
+			sym = sec->data->d_buf;
+			for (j = 0; j < n; j++, sym++) {
+				if (sym->st_shndx
+				    && sym->st_shndx < SHN_LORESERVE
+				    && sym->st_shndx >= obj->sec_cnt) {
+					pr_warn("ELF sym #%d in section #%zu points to missing section #%zu in %s\n",
+						j, sec->sec_idx, (size_t)sym->st_shndx, obj->filename);
+					return -EINVAL;
+				}
+				if (ELF64_ST_TYPE(sym->st_info) == STT_SECTION) {
+					if (sym->st_value != 0)
+						return -EINVAL;
+				}
+			}
+			break;
+		}
+		case SHT_STRTAB:
+			break;
+		case SHT_PROGBITS:
+			if (sec->shdr->sh_flags & SHF_EXECINSTR) {
+				if (sec->shdr->sh_size % sizeof(struct bpf_insn) != 0)
+					return -EINVAL;
+			}
+			break;
+		case SHT_NOBITS:
+			break;
+		case SHT_REL: {
+			Elf64_Rel *relo;
+			struct src_sec *sym_sec;
+
+			if (sec->shdr->sh_entsize != sizeof(Elf64_Rel))
+				return -EINVAL;
+			if (sec->shdr->sh_size % sec->shdr->sh_entsize != 0)
+				return -EINVAL;
+
+			/* SHT_REL's sh_link should point to SYMTAB */
+			if (sec->shdr->sh_link != obj->symtab_sec_idx) {
+				pr_warn("ELF relo section #%zu points to invalid SYMTAB section #%zu in %s\n",
+					sec->sec_idx, (size_t)sec->shdr->sh_link, obj->filename);
+				return -EINVAL;
+			}
+
+			/* SHT_REL's sh_info points to relocated section */
+			if (!sec->shdr->sh_info || sec->shdr->sh_info >= obj->sec_cnt) {
+				pr_warn("ELF relo section #%zu points to missing section #%zu in %s\n",
+					sec->sec_idx, (size_t)sec->shdr->sh_info, obj->filename);
+				return -EINVAL;
+			}
+			link_sec = &obj->secs[sec->shdr->sh_info];
+
+			/* .rel<secname> -> <secname> pattern is followed */
+			if (strncmp(sec->sec_name, ".rel", sizeof(".rel") - 1) != 0
+			    || strcmp(sec->sec_name + sizeof(".rel") - 1, link_sec->sec_name) != 0) {
+				pr_warn("ELF relo section #%zu name has invalid name in %s\n",
+					sec->sec_idx, obj->filename);
+				return -EINVAL;
+			}
+
+			/* don't further validate relocations for ignored sections */
+			if (link_sec->skipped)
+				break;
+
+			/* relocatable section is data or instructions */
+			if (link_sec->shdr->sh_type != SHT_PROGBITS
+			    && link_sec->shdr->sh_type != SHT_NOBITS) {
+				pr_warn("ELF relo section #%zu points to invalid section #%zu in %s\n",
+					sec->sec_idx, (size_t)sec->shdr->sh_info, obj->filename);
+				return -EINVAL;
+			}
+
+			/* check sanity of each relocation */
+			n = sec->shdr->sh_size / sec->shdr->sh_entsize;
+			relo = sec->data->d_buf;
+			sym_sec = &obj->secs[obj->symtab_sec_idx];
+			for (j = 0; j < n; j++, relo++) {
+				size_t sym_idx = ELF64_R_SYM(relo->r_info);
+				size_t sym_type = ELF64_R_TYPE(relo->r_info);
+
+				if (sym_type != R_BPF_64_64 && sym_type != R_BPF_64_32) {
+					pr_warn("ELF relo #%d in section #%zu has unexpected type %zu in %s\n",
+						j, sec->sec_idx, sym_type, obj->filename);
+					return -EINVAL;
+				}
+
+				if (!sym_idx || sym_idx * sizeof(Elf64_Sym) >= sym_sec->shdr->sh_size) {
+					pr_warn("ELF relo #%d in section #%zu points to invalid symbol #%zu in %s\n",
+						j, sec->sec_idx, sym_idx, obj->filename);
+					return -EINVAL;
+				}
+
+				if (link_sec->shdr->sh_flags & SHF_EXECINSTR) {
+					if (relo->r_offset % sizeof(struct bpf_insn) != 0) {
+						pr_warn("ELF relo #%d in section #%zu points to missing symbol #%zu in %s\n",
+							j, sec->sec_idx, sym_idx, obj->filename);
+						return -EINVAL;
+					}
+				}
+			}
+			break;
+		}
+		case SHT_LLVM_ADDRSIG:
+			break;
+		default:
+			pr_warn("ELF section #%zu (%s) has unrecognized type %zu in %s\n",
+				sec->sec_idx, sec->sec_name, (size_t)sec->shdr->sh_type, obj->filename);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int check_btf_type_id(__u32 *type_id, void *ctx)
+{
+	struct btf *btf = ctx;
+
+	if (*type_id > btf__get_nr_types(btf))
+		return -EINVAL;
+
+	return 0;
+}
+
+static int check_btf_str_off(__u32 *str_off, void *ctx)
+{
+	struct btf *btf = ctx;
+	const char *s;
+
+	s = btf__str_by_offset(btf, *str_off);
+
+	if (!s)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int linker_sanity_check_btf(struct src_obj *obj)
+{
+	struct btf_type *t;
+	int i, n, err = 0;
+
+	if (!obj->btf)
+		return 0;
+
+	n = btf__get_nr_types(obj->btf);
+	for (i = 1; i <= n; i++) {
+		t = btf_type_by_id(obj->btf, i);
+
+		err = err ?: btf_type_visit_type_ids(t, check_btf_type_id, obj->btf);
+		err = err ?: btf_type_visit_str_offs(t, check_btf_str_off, obj->btf);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int linker_sanity_check_btf_ext(struct src_obj *obj)
+{
+	int err = 0;
+
+	if (!obj->btf_ext)
+		return 0;
+
+	/* can't use .BTF.ext without .BTF */
+	if (!obj->btf)
+		return -EINVAL;
+
+	err = err ?: btf_ext_visit_type_ids(obj->btf_ext, check_btf_type_id, obj->btf);
+	err = err ?: btf_ext_visit_str_offs(obj->btf_ext, check_btf_str_off, obj->btf);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+static int init_sec(struct bpf_linker *linker, struct dst_sec *dst_sec, struct src_sec *src_sec)
+{
+	Elf_Scn *scn;
+	Elf_Data *data;
+	Elf64_Shdr *shdr;
+	int name_off;
+
+	dst_sec->sec_sz = 0;
+	dst_sec->sec_idx = 0;
+
+	/* ephemeral sections are just thin section shells lacking most parts */
+	if (src_sec->ephemeral)
+		return 0;
+
+	scn = elf_newscn(linker->elf);
+	if (!scn)
+		return -1;
+	data = elf_newdata(scn);
+	if (!data)
+		return -1;
+	shdr = elf64_getshdr(scn);
+	if (!shdr)
+		return -1;
+
+	dst_sec->scn = scn;
+	dst_sec->shdr = shdr;
+	dst_sec->data = data;
+	dst_sec->sec_idx = elf_ndxscn(scn);
+
+	name_off = strset__add_str(linker->strtab_strs, src_sec->sec_name);
+	if (name_off < 0)
+		return name_off;
+
+	shdr->sh_name = name_off;
+	shdr->sh_type = src_sec->shdr->sh_type;
+	shdr->sh_flags = src_sec->shdr->sh_flags;
+	shdr->sh_size = 0;
+	/* sh_link and sh_info have different meaning for different types of
+	 * sections, so we leave it up to the caller code to fill them in, if
+	 * necessary
+	 */
+	shdr->sh_link = 0;
+	shdr->sh_info = 0;
+	shdr->sh_addralign = src_sec->shdr->sh_addralign;
+	shdr->sh_entsize = src_sec->shdr->sh_entsize;
+
+	data->d_type = src_sec->data->d_type;
+	data->d_size = 0;
+	data->d_buf = NULL;
+	data->d_align = src_sec->data->d_align;
+	data->d_off = 0;
+
+	return 0;
+}
+
+static struct dst_sec *find_dst_sec_by_name(struct bpf_linker *linker, const char *sec_name)
+{
+	struct dst_sec *sec;
+	int i;
+
+	for (i = 1; i < linker->sec_cnt; i++) {
+		sec = &linker->secs[i];
+
+		if (strcmp(sec->sec_name, sec_name) == 0)
+			return sec;
+	}
+
+	return NULL;
+}
+
+static bool secs_match(struct dst_sec *dst, struct src_sec *src)
+{
+	if (dst->shdr->sh_type != src->shdr->sh_type) {
+		pr_warn("sec %s types mismatch\n", dst->sec_name);
+		return false;
+	}
+	if (dst->shdr->sh_flags != src->shdr->sh_flags) {
+		pr_warn("sec %s flags mismatch\n", dst->sec_name);
+		return false;
+	}
+	if (dst->shdr->sh_entsize != src->shdr->sh_entsize) {
+		pr_warn("sec %s entsize mismatch\n", dst->sec_name);
+		return false;
+	}
+
+	return true;
+}
+
+static bool sec_content_is_same(struct dst_sec *dst_sec, struct src_sec *src_sec)
+{
+	if (dst_sec->sec_sz != src_sec->shdr->sh_size)
+		return false;
+	if (memcmp(dst_sec->raw_data, src_sec->data->d_buf, dst_sec->sec_sz) != 0)
+		return false;
+	return true;
+}
+
+static int extend_sec(struct dst_sec *dst, struct src_sec *src)
+{
+	void *tmp;
+	size_t dst_align = dst->shdr->sh_addralign;
+	size_t src_align = src->shdr->sh_addralign;
+	size_t dst_align_sz, dst_final_sz;
+
+	if (dst_align == 0)
+		dst_align = 1;
+	if (dst_align < src_align)
+		dst_align = src_align;
+
+	dst_align_sz = (dst->sec_sz + dst_align - 1) / dst_align * dst_align;
+
+	/* no need to re-align final size */
+	dst_final_sz = dst_align_sz + src->shdr->sh_size;
+
+	if (src->shdr->sh_type != SHT_NOBITS) {
+		tmp = realloc(dst->raw_data, dst_final_sz);
+		if (!tmp)
+			return -ENOMEM;
+		dst->raw_data = tmp;
+
+		/* pad dst section, if it's alignment forced size increase */
+		memset(dst->raw_data + dst->sec_sz, 0, dst_align_sz - dst->sec_sz);
+		/* now copy src data at a properly aligned offset */
+		memcpy(dst->raw_data + dst_align_sz, src->data->d_buf, src->shdr->sh_size);
+	}
+
+	dst->sec_sz = dst_final_sz;
+	dst->shdr->sh_size = dst_final_sz;
+	dst->data->d_size = dst_final_sz;
+
+	dst->shdr->sh_addralign = dst_align;
+	dst->data->d_align = dst_align;
+
+	src->dst_off = dst_align_sz;
+
+	return 0;
+}
+
+static bool is_data_sec(struct src_sec *sec)
+{
+	if (!sec || sec->skipped)
+		return false;
+	/* ephemeral sections are data sections, e.g., .kconfig, .ksyms */
+	if (sec->ephemeral)
+		return true;
+	return sec->shdr->sh_type == SHT_PROGBITS || sec->shdr->sh_type == SHT_NOBITS;
+}
+
+static bool is_relo_sec(struct src_sec *sec)
+{
+	if (!sec || sec->skipped || sec->ephemeral)
+		return false;
+	return sec->shdr->sh_type == SHT_REL;
+}
+
+static int linker_append_sec_data(struct bpf_linker *linker, struct src_obj *obj)
+{
+	int i, err;
+
+	for (i = 1; i < obj->sec_cnt; i++) {
+		struct src_sec *src_sec;
+		struct dst_sec *dst_sec;
+
+		src_sec = &obj->secs[i];
+		if (!is_data_sec(src_sec))
+			continue;
+
+		dst_sec = find_dst_sec_by_name(linker, src_sec->sec_name);
+		if (!dst_sec) {
+			dst_sec = add_dst_sec(linker, src_sec->sec_name);
+			if (!dst_sec)
+				return -ENOMEM;
+			err = init_sec(linker, dst_sec, src_sec);
+			if (err) {
+				pr_warn("failed to init section '%s'\n", src_sec->sec_name);
+				return err;
+			}
+		} else {
+			if (!secs_match(dst_sec, src_sec)) {
+				pr_warn("ELF sections %s are incompatible\n", src_sec->sec_name);
+				return -1;
+			}
+
+			/* "license" and "version" sections are deduped */
+			if (strcmp(src_sec->sec_name, "license") == 0
+			    || strcmp(src_sec->sec_name, "version") == 0) {
+				if (!sec_content_is_same(dst_sec, src_sec)) {
+					pr_warn("non-identical contents of section '%s' are not supported\n", src_sec->sec_name);
+					return -EINVAL;
+				}
+				src_sec->skipped = true;
+				src_sec->dst_id = dst_sec->id;
+				continue;
+			}
+		}
+
+		/* record mapped section index */
+		src_sec->dst_id = dst_sec->id;
+
+		if (src_sec->ephemeral)
+			continue;
+
+		err = extend_sec(dst_sec, src_sec);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static int linker_append_elf_syms(struct bpf_linker *linker, struct src_obj *obj)
+{
+	struct src_sec *symtab = &obj->secs[obj->symtab_sec_idx];
+	Elf64_Sym *sym = symtab->data->d_buf, *dst_sym;
+	int i, n = symtab->shdr->sh_size / symtab->shdr->sh_entsize;
+	int str_sec_idx = symtab->shdr->sh_link;
+
+	obj->sym_map = calloc(n + 1, sizeof(*obj->sym_map));
+	if (!obj->sym_map)
+		return -ENOMEM;
+
+	for (i = 0; i < n; i++, sym++) {
+		struct src_sec *src_sec = NULL;
+		struct dst_sec *dst_sec = NULL;
+		const char *sym_name;
+		size_t dst_sym_idx;
+		int name_off;
+
+		/* we already have all-zero initial symbol */
+		if (sym->st_name == 0 && sym->st_info == 0 &&
+		    sym->st_other == 0 && sym->st_shndx == SHN_UNDEF &&
+		    sym->st_value == 0 && sym->st_size ==0)
+			continue;
+
+		sym_name = elf_strptr(obj->elf, str_sec_idx, sym->st_name);
+		if (!sym_name) {
+			pr_warn("can't fetch symbol name for symbol #%d in '%s'\n", i, obj->filename);
+			return -1;
+		}
+
+		if (sym->st_shndx && sym->st_shndx < SHN_LORESERVE) {
+			src_sec = &obj->secs[sym->st_shndx];
+			if (src_sec->skipped)
+				continue;
+			dst_sec = &linker->secs[src_sec->dst_id];
+
+			/* allow only one STT_SECTION symbol per section */
+			if (ELF64_ST_TYPE(sym->st_info) == STT_SECTION && dst_sec->sec_sym_idx) {
+				obj->sym_map[i] = dst_sec->sec_sym_idx;
+				continue;
+			}
+		}
+
+		name_off = strset__add_str(linker->strtab_strs, sym_name);
+		if (name_off < 0)
+			return name_off;
+
+		dst_sym = add_new_sym(linker, &dst_sym_idx);
+		if (!dst_sym)
+			return -ENOMEM;
+
+		dst_sym->st_name = name_off;
+		dst_sym->st_info = sym->st_info;
+		dst_sym->st_other = sym->st_other;
+		dst_sym->st_shndx = src_sec ? dst_sec->sec_idx : sym->st_shndx;
+		dst_sym->st_value = (src_sec ? src_sec->dst_off : 0) + sym->st_value;
+		dst_sym->st_size = sym->st_size;
+
+		obj->sym_map[i] = dst_sym_idx;
+
+		if (ELF64_ST_TYPE(sym->st_info) == STT_SECTION && dst_sym) {
+			dst_sec->sec_sym_idx = dst_sym_idx;
+			dst_sym->st_value = 0;
+		}
+
+	}
+
+	return 0;
+}
+
+static int linker_append_elf_relos(struct bpf_linker *linker, struct src_obj *obj)
+{
+	struct src_sec *src_symtab = &obj->secs[obj->symtab_sec_idx];
+	struct dst_sec *dst_symtab = &linker->secs[linker->symtab_sec_idx];
+	int i, err;
+
+	for (i = 1; i < obj->sec_cnt; i++) {
+		struct src_sec *src_sec, *src_linked_sec;
+		struct dst_sec *dst_sec, *dst_linked_sec;
+		Elf64_Rel *src_rel, *dst_rel;
+		int j, n;
+
+		src_sec = &obj->secs[i];
+		if (!is_relo_sec(src_sec))
+			continue;
+
+		/* shdr->sh_info points to relocatable section */
+		src_linked_sec = &obj->secs[src_sec->shdr->sh_info];
+		if (src_linked_sec->skipped)
+			continue;
+
+		dst_sec = find_dst_sec_by_name(linker, src_sec->sec_name);
+		if (!dst_sec) {
+			dst_sec = add_dst_sec(linker, src_sec->sec_name);
+			if (!dst_sec)
+				return -ENOMEM;
+			err = init_sec(linker, dst_sec, src_sec);
+			if (err) {
+				pr_warn("failed to init section '%s'\n", src_sec->sec_name);
+				return err;
+			}
+		} else if (!secs_match(dst_sec, src_sec)) {
+			pr_warn("Secs %s are not compatible\n", src_sec->sec_name);
+			return -1;
+		}
+
+		/* shdr->sh_link points to SYMTAB */
+		dst_sec->shdr->sh_link = linker->symtab_sec_idx;
+
+		/* shdr->sh_info points to relocated section */
+		dst_linked_sec = &linker->secs[src_linked_sec->dst_id];
+		dst_sec->shdr->sh_info = dst_linked_sec->sec_idx;
+
+		src_sec->dst_id = dst_sec->id;
+		err = extend_sec(dst_sec, src_sec);
+		if (err)
+			return err;
+
+		src_rel = src_sec->data->d_buf;
+		dst_rel = dst_sec->raw_data + src_sec->dst_off;
+		n = src_sec->shdr->sh_size / src_sec->shdr->sh_entsize;
+		for (j = 0; j < n; j++, src_rel++, dst_rel++) {
+			size_t src_sym_idx = ELF64_R_SYM(src_rel->r_info);
+			size_t sym_type = ELF64_R_TYPE(src_rel->r_info);
+			Elf64_Sym *src_sym, *dst_sym;
+			size_t dst_sym_idx;
+
+			src_sym_idx = ELF64_R_SYM(src_rel->r_info);
+			src_sym = src_symtab->data->d_buf + sizeof(*src_sym) * src_sym_idx;
+
+			dst_sym_idx = obj->sym_map[src_sym_idx];
+			dst_sym = dst_symtab->raw_data + sizeof(*dst_sym) * dst_sym_idx;
+			dst_rel->r_offset += src_linked_sec->dst_off;
+			sym_type = ELF64_R_TYPE(src_rel->r_info);
+			dst_rel->r_info = ELF64_R_INFO(dst_sym_idx, sym_type);
+
+			if (ELF64_ST_TYPE(src_sym->st_info) == STT_SECTION) {
+				struct src_sec *sec = &obj->secs[src_sym->st_shndx];
+				struct bpf_insn *insn;
+
+				if (src_linked_sec->shdr->sh_flags & SHF_EXECINSTR) {
+					/* calls to the very first static function inside
+					 * .text section at offset 0 will
+					 * reference section symbol, not the
+					 * function symbol. Fix that up,
+					 * otherwise it won't be possible to
+					 * relocate calls to two different
+					 * static functions with the same name
+					 * (rom two different object files)
+					 */
+					insn = dst_linked_sec->raw_data + dst_rel->r_offset;
+					if (insn->code == (BPF_JMP | BPF_CALL))
+						insn->imm += sec->dst_off / sizeof(struct bpf_insn);
+					else
+						insn->imm += sec->dst_off;
+				} else {
+					pr_warn("relocation against STT_SECTION in non-exec section is not supported!\n");
+					return -EINVAL;
+				}
+			}
+
+		}
+	}
+
+	return 0;
+}
+
+static struct src_sec *find_src_sec_by_name(struct src_obj *obj, const char *sec_name)
+{
+	struct src_sec *sec;
+	int i;
+
+	for (i = 1; i < obj->sec_cnt; i++) {
+		sec = &obj->secs[i];
+
+		if (strcmp(sec->sec_name, sec_name) == 0)
+			return sec;
+	}
+
+	return NULL;
+}
+
+static Elf64_Sym *find_sym_by_name(struct src_obj *obj, size_t sec_idx,
+				   int sym_type, const char *sym_name)
+{
+	struct src_sec *symtab = &obj->secs[obj->symtab_sec_idx];
+	Elf64_Sym *sym = symtab->data->d_buf;
+	int i, n = symtab->shdr->sh_size / symtab->shdr->sh_entsize;
+	int str_sec_idx = symtab->shdr->sh_link;
+	const char *name;
+
+	for (i = 0; i < n; i++, sym++) {
+		if (sym->st_shndx != sec_idx)
+			continue;
+		if (ELF64_ST_TYPE(sym->st_info) != sym_type)
+			continue;
+
+		name = elf_strptr(obj->elf, str_sec_idx, sym->st_name);
+		if (!name)
+			return NULL;
+
+		if (strcmp(sym_name, name) != 0)
+			continue;
+
+		return sym;
+	}
+
+	return NULL;
+}
+
+static int linker_fixup_btf(struct src_obj *obj)
+{
+	const char *sec_name;
+	struct src_sec *sec;
+	int i, j, n, m;
+
+	if (!obj->btf)
+		return 0;
+
+	n = btf__get_nr_types(obj->btf);
+	for (i = 1; i <= n; i++) {
+		struct btf_var_secinfo *vi;
+		struct btf_type *t;
+
+		t = btf_type_by_id(obj->btf, i);
+		if (btf_kind(t) != BTF_KIND_DATASEC)
+			continue;
+
+		sec_name = btf__str_by_offset(obj->btf, t->name_off);
+		sec = find_src_sec_by_name(obj, sec_name);
+		if (sec) {
+			/* record actual section size, unless ephemeral */
+			if (sec->shdr)
+				t->size = sec->shdr->sh_size;
+		} else {
+			/* BTF can have some sections that are not represented
+			 * in ELF, e.g., .kconfig and .ksyms, which are used
+			 * for special extern variables.  Here we'll
+			 * pre-create "section shells" for them to be able to
+			 * keep track of extra per-section metadata later
+			 * (e.g., BTF variables).
+			 */
+			sec = add_src_sec(obj, sec_name);
+			if (!sec)
+				return -ENOMEM;
+
+			sec->ephemeral = true;
+			sec->sec_idx = 0; /* will match UNDEF shndx in ELF */
+		}
+
+		/* remember ELF section and its BTF type ID match */
+		sec->sec_type_id = i;
+
+		/* fix up variable offsets */
+		vi = btf_var_secinfos(t);
+		for (j = 0, m = btf_vlen(t); j < m; j++, vi++) {
+			const struct btf_type *vt = btf__type_by_id(obj->btf, vi->type);
+			const char *var_name = btf__str_by_offset(obj->btf, vt->name_off);
+			int var_linkage = btf_var(vt)->linkage;
+			Elf64_Sym *sym;
+
+			/* no need to patch up static or extern vars */
+			if (var_linkage != BTF_VAR_GLOBAL_ALLOCATED)
+				continue;
+
+			sym = find_sym_by_name(obj, sec->sec_idx, STT_OBJECT, var_name);
+			if (!sym) {
+				pr_warn("failed to find symbol for variable '%s' in section '%s'\n", var_name, sec_name);
+				return -ENOENT;
+			}
+
+			vi->offset = sym->st_value;
+		}
+	}
+
+	return 0;
+}
+
+static int remap_type_id(__u32 *type_id, void *ctx)
+{
+	int *id_map = ctx;
+
+	*type_id = id_map[*type_id];
+
+	return 0;
+}
+
+static int linker_append_btf(struct bpf_linker *linker, struct src_obj *obj)
+{
+	const struct btf_type *t;
+	int i, j, n, start_id, id;
+
+	if (!obj->btf)
+		return 0;
+
+	start_id = btf__get_nr_types(linker->btf) + 1;
+	n = btf__get_nr_types(obj->btf);
+
+	obj->btf_type_map = calloc(n + 1, sizeof(int));
+	if (!obj->btf_type_map)
+		return -ENOMEM;
+
+	for (i = 1; i <= n; i++) {
+		t = btf__type_by_id(obj->btf, i);
+
+		/* DATASECs are handled specially below */
+		if (btf_kind(t) == BTF_KIND_DATASEC)
+			continue;
+
+		id = btf__add_type(linker->btf, obj->btf, t);
+		if (id < 0) {
+			pr_warn("failed to append BTF type #%d from file '%s'\n", i, obj->filename);
+			return id;
+		}
+
+		obj->btf_type_map[i] = id;
+	}
+
+	/* remap all the types except DATASECs */
+	n = btf__get_nr_types(linker->btf);
+	for (i = start_id; i <= n; i++) {
+		struct btf_type *dst_t = btf_type_by_id(linker->btf, i);
+
+		if (btf_type_visit_type_ids(dst_t, remap_type_id, obj->btf_type_map))
+			return -EINVAL;
+	}
+
+	/* append DATASEC info */
+	for (i = 1; i < obj->sec_cnt; i++) {
+		struct src_sec *src_sec;
+		struct dst_sec *dst_sec;
+		const struct btf_var_secinfo *src_var;
+		struct btf_var_secinfo *dst_var;
+
+		src_sec = &obj->secs[i];
+		if (!src_sec->sec_type_id || src_sec->skipped)
+			continue;
+		dst_sec = &linker->secs[src_sec->dst_id];
+
+		t = btf__type_by_id(obj->btf, src_sec->sec_type_id);
+		src_var = btf_var_secinfos(t);
+		n = btf_vlen(t);
+		for (j = 0; j < n; j++, src_var++) {
+			void *sec_vars = dst_sec->sec_vars;
+
+			sec_vars = libbpf_reallocarray(sec_vars,
+						       dst_sec->sec_var_cnt + 1,
+						       sizeof(*dst_sec->sec_vars));
+			if (!sec_vars)
+				return -ENOMEM;
+
+			dst_sec->sec_vars = sec_vars;
+			dst_sec->sec_var_cnt++;
+
+			dst_var = &dst_sec->sec_vars[dst_sec->sec_var_cnt - 1];
+			dst_var->type = obj->btf_type_map[src_var->type];
+			dst_var->size = src_var->size;
+			dst_var->offset = src_sec->dst_off + src_var->offset;
+		}
+	}
+
+	return 0;
+}
+
+static void *add_btf_ext_rec(struct btf_ext_sec_data *ext_data, const void *src_rec)
+{
+	void *tmp;
+
+	tmp = libbpf_reallocarray(ext_data->recs, ext_data->rec_cnt + 1, ext_data->rec_sz);
+	if (!tmp)
+		return NULL;
+	ext_data->recs = tmp;
+
+	tmp += ext_data->rec_cnt * ext_data->rec_sz;
+	memcpy(tmp, src_rec, ext_data->rec_sz);
+
+	ext_data->rec_cnt++;
+
+	return tmp;
+}
+
+static int linker_append_btf_ext(struct bpf_linker *linker, struct src_obj *obj)
+{
+	const struct btf_ext_info_sec *ext_sec;
+	const char *sec_name, *s;
+	struct src_sec *src_sec;
+	struct dst_sec *dst_sec;
+	int rec_sz, str_off, i;
+
+	if (!obj->btf_ext)
+		return 0;
+
+	rec_sz = obj->btf_ext->func_info.rec_size;
+	for_each_btf_ext_sec(&obj->btf_ext->func_info, ext_sec) {
+		struct bpf_func_info_min *src_rec, *dst_rec;
+
+		sec_name = btf__name_by_offset(obj->btf, ext_sec->sec_name_off);
+		src_sec = find_src_sec_by_name(obj, sec_name);
+		if (!src_sec) {
+			pr_warn("can't find section '%s' referenced from .BTF.ext\n", sec_name);
+			return -EINVAL;
+		}
+		dst_sec = &linker->secs[src_sec->dst_id];
+
+		if (dst_sec->func_info.rec_sz == 0)
+			dst_sec->func_info.rec_sz = rec_sz;
+		if (dst_sec->func_info.rec_sz != rec_sz) {
+			pr_warn("incompatible .BTF.ext record sizes for section '%s'\n", sec_name);
+			return -EINVAL;
+		}
+
+		for_each_btf_ext_rec(&obj->btf_ext->func_info, ext_sec, i, src_rec) {
+			dst_rec = add_btf_ext_rec(&dst_sec->func_info, src_rec);
+			if (!dst_rec)
+				return -ENOMEM;
+
+			dst_rec->insn_off += src_sec->dst_off;
+			dst_rec->type_id = obj->btf_type_map[dst_rec->type_id];
+		}
+	}
+
+	rec_sz = obj->btf_ext->line_info.rec_size;
+	for_each_btf_ext_sec(&obj->btf_ext->line_info, ext_sec) {
+		struct bpf_line_info_min *src_rec, *dst_rec;
+
+		sec_name = btf__name_by_offset(obj->btf, ext_sec->sec_name_off);
+		src_sec = find_src_sec_by_name(obj, sec_name);
+		if (!src_sec) {
+			pr_warn("can't find section '%s' referenced from .BTF.ext\n", sec_name);
+			return -EINVAL;
+		}
+		dst_sec = &linker->secs[src_sec->dst_id];
+
+		if (dst_sec->line_info.rec_sz == 0)
+			dst_sec->line_info.rec_sz = rec_sz;
+		if (dst_sec->line_info.rec_sz != rec_sz) {
+			pr_warn("incompatible .BTF.ext record sizes for section '%s'\n", sec_name);
+			return -EINVAL;
+		}
+
+		for_each_btf_ext_rec(&obj->btf_ext->line_info, ext_sec, i, src_rec) {
+			dst_rec = add_btf_ext_rec(&dst_sec->line_info, src_rec);
+			if (!dst_rec)
+				return -ENOMEM;
+
+			dst_rec->insn_off += src_sec->dst_off;
+
+			s = btf__str_by_offset(obj->btf, src_rec->file_name_off);
+			str_off = btf__add_str(linker->btf, s);
+			if (str_off < 0)
+				return -ENOMEM;
+			dst_rec->file_name_off = str_off;
+
+			s = btf__str_by_offset(obj->btf, src_rec->line_off);
+			str_off = btf__add_str(linker->btf, s);
+			if (str_off < 0)
+				return -ENOMEM;
+			dst_rec->line_off = str_off;
+
+			/* dst_rec->line_col is fine */
+		}
+	}
+
+	rec_sz = obj->btf_ext->core_relo_info.rec_size;
+	for_each_btf_ext_sec(&obj->btf_ext->core_relo_info, ext_sec) {
+		struct bpf_core_relo *src_rec, *dst_rec;
+
+		sec_name = btf__name_by_offset(obj->btf, ext_sec->sec_name_off);
+		src_sec = find_src_sec_by_name(obj, sec_name);
+		if (!src_sec) {
+			pr_warn("can't find section '%s' referenced from .BTF.ext\n", sec_name);
+			return -EINVAL;
+		}
+		dst_sec = &linker->secs[src_sec->dst_id];
+
+		if (dst_sec->core_relo_info.rec_sz == 0)
+			dst_sec->core_relo_info.rec_sz = rec_sz;
+		if (dst_sec->core_relo_info.rec_sz != rec_sz) {
+			pr_warn("incompatible .BTF.ext record sizes for section '%s'\n", sec_name);
+			return -EINVAL;
+		}
+
+		for_each_btf_ext_rec(&obj->btf_ext->core_relo_info, ext_sec, i, src_rec) {
+			dst_rec = add_btf_ext_rec(&dst_sec->core_relo_info, src_rec);
+			if (!dst_rec)
+				return -ENOMEM;
+
+			dst_rec->insn_off += src_sec->dst_off;
+			dst_rec->type_id = obj->btf_type_map[dst_rec->type_id];
+
+			s = btf__str_by_offset(obj->btf, src_rec->access_str_off);
+			str_off = btf__add_str(linker->btf, s);
+			if (str_off < 0)
+				return -ENOMEM;
+			dst_rec->access_str_off = str_off;
+
+			/* dst_rec->kind is fine */
+		}
+	}
+
+	return 0;
+}
+
+int bpf_linker__finalize(struct bpf_linker *linker)
+{
+	struct dst_sec *sec;
+	size_t strs_sz;
+	const void *strs;
+	int err, i;
+
+	if (!linker->elf)
+		return -EINVAL;
+
+	err = finalize_btf(linker);
+	if (err)
+		return err;
+
+	/* Finalize strings */
+	strs_sz = strset__data_size(linker->strtab_strs);
+	strs = strset__data(linker->strtab_strs);
+
+	sec = &linker->secs[linker->strtab_sec_idx];
+	sec->data->d_align = 1;
+	sec->data->d_off = 0LL;
+	sec->data->d_buf = (void *)strs;
+	sec->data->d_type = ELF_T_BYTE;
+	sec->data->d_size = strs_sz;
+	sec->shdr->sh_size = strs_sz;
+
+	for (i = 1; i < linker->sec_cnt; i++) {
+		sec = &linker->secs[i];
+
+		/* STRTAB is handled specially above */
+		if (sec->sec_idx == linker->strtab_sec_idx)
+			continue;
+
+		/* special ephemeral sections (.ksyms, .kconfig, etc) */
+		if (!sec->scn)
+			continue;
+
+		sec->data->d_buf = sec->raw_data;
+	}
+
+	/* Finalize ELF layout */
+	if (elf_update(linker->elf, ELF_C_NULL) < 0) {
+		err = -errno;
+		pr_warn_elf("failed to finalize ELF layout");
+		return err;
+	}
+
+	/* Write out final ELF contents */
+	if (elf_update(linker->elf, ELF_C_WRITE) < 0) {
+		err = -errno;
+		pr_warn_elf("failed to write ELF contents");
+		return err;
+	}
+
+	elf_end(linker->elf);
+	close(linker->fd);
+
+	linker->elf = NULL;
+	linker->fd = -1;
+
+	return 0;
+}
+
+static int emit_elf_data_sec(struct bpf_linker *linker, const char *sec_name,
+			     size_t align, const void *raw_data, size_t raw_sz)
+{
+	Elf_Scn *scn;
+	Elf_Data *data;
+	Elf64_Shdr *shdr;
+	int name_off;
+
+	name_off = strset__add_str(linker->strtab_strs, sec_name);
+	if (name_off < 0)
+		return name_off;
+
+	scn = elf_newscn(linker->elf);
+	if (!scn)
+		return -ENOMEM;
+	data = elf_newdata(scn);
+	if (!data)
+		return -ENOMEM;
+	shdr = elf64_getshdr(scn);
+	if (!shdr)
+		return -EINVAL;
+
+	shdr->sh_name = name_off;
+	shdr->sh_type = SHT_PROGBITS;
+	shdr->sh_flags = 0;
+	shdr->sh_size = raw_sz;
+	shdr->sh_link = 0;
+	shdr->sh_info = 0;
+	shdr->sh_addralign = align;
+	shdr->sh_entsize = 0;
+
+	data->d_type = ELF_T_BYTE;
+	data->d_size = raw_sz;
+	data->d_buf = (void *)raw_data;
+	data->d_align = align;
+	data->d_off = 0;
+
+	return 0;
+}
+
+static int finalize_btf(struct bpf_linker *linker)
+{
+	struct btf *btf = linker->btf;
+	const void *raw_data;
+	int i, j, id, err;
+	__u32 raw_sz;
+
+	/* bail out if no BTF data was produced */
+	if (btf__get_nr_types(linker->btf) == 0)
+		return 0;
+
+	for (i = 1; i < linker->sec_cnt; i++) {
+		struct dst_sec *sec = &linker->secs[i];
+
+		if (!sec->sec_var_cnt)
+			continue;
+
+		id = btf__add_datasec(btf, sec->sec_name, sec->sec_sz);
+		if (id < 0) {
+			pr_warn("failed to add consolidated BTF type for datasec '%s': %d\n",
+				sec->sec_name, id);
+			return id;
+		}
+
+		for (j = 0; j < sec->sec_var_cnt; j++) {
+			struct btf_var_secinfo *vi = &sec->sec_vars[j];
+
+			if (btf__add_datasec_var_info(btf, vi->type, vi->offset, vi->size))
+				return -EINVAL;
+		}
+	}
+
+	err = finalize_btf_ext(linker);
+	if (err) {
+		pr_warn(".BTF.ext generation failed: %d\n", err);
+		return err;
+	}
+
+	err = btf__dedup(linker->btf, linker->btf_ext, NULL);
+	if (err) {
+		pr_warn("BTF dedup failed: %d\n", err);
+		return err;
+	}
+
+	/* Emit .BTF section */
+	raw_data = btf__get_raw_data(linker->btf, &raw_sz);
+	if (!raw_data)
+		return -ENOMEM;
+
+	err = emit_elf_data_sec(linker, BTF_ELF_SEC, 8, raw_data, raw_sz);
+	if (err) {
+		pr_warn("failed to write out .BTF ELF section: %d\n", err);
+		return err;
+	}
+
+	/* Emit .BTF.ext section */
+	if (linker->btf_ext) {
+		raw_data = btf_ext__get_raw_data(linker->btf_ext, &raw_sz);
+		if (!raw_data)
+			return -ENOMEM;
+
+		err = emit_elf_data_sec(linker, BTF_EXT_ELF_SEC, 8, raw_data, raw_sz);
+		if (err) {
+			pr_warn("failed to write out .BTF.ext ELF section: %d\n", err);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static int emit_btf_ext_data(struct bpf_linker *linker, void *output,
+			     const char *sec_name, struct btf_ext_sec_data *sec_data)
+{
+	struct btf_ext_info_sec *sec_info;
+	void *cur = output;
+	int str_off;
+	size_t sz;
+
+	if (!sec_data->rec_cnt)
+		return 0;
+
+	str_off = btf__add_str(linker->btf, sec_name);
+	if (str_off < 0)
+		return -ENOMEM;
+
+	sec_info = cur;
+	sec_info->sec_name_off = str_off;
+	sec_info->num_info = sec_data->rec_cnt;
+	cur += sizeof(struct btf_ext_info_sec);
+
+	sz = sec_data->rec_cnt * sec_data->rec_sz;
+	memcpy(cur, sec_data->recs, sz);
+	cur += sz;
+
+	return cur - output;
+}
+
+static int finalize_btf_ext(struct bpf_linker *linker)
+{
+	size_t funcs_sz = 0, lines_sz = 0, core_relos_sz = 0, total_sz = 0;
+	size_t func_rec_sz = 0, line_rec_sz = 0, core_relo_rec_sz = 0;
+	struct btf_ext_header *hdr;
+	void *data, *cur;
+	int i, err, sz;
+
+	/* validate that all sections have the same .BTF.ext record sizes
+	 * and calculate total data size for each type of data (func info,
+	 * line info, core relos)
+	 */
+	for (i = 1; i < linker->sec_cnt; i++) {
+		struct dst_sec *sec = &linker->secs[i];
+
+		if (sec->func_info.rec_cnt) {
+			if (func_rec_sz == 0)
+				func_rec_sz = sec->func_info.rec_sz;
+			if (func_rec_sz != sec->func_info.rec_sz) {
+				pr_warn("mismatch in func_info record size %zu != %u\n",
+					func_rec_sz, sec->func_info.rec_sz);
+				return -EINVAL;
+			}
+
+			funcs_sz += sizeof(struct btf_ext_info_sec) + func_rec_sz * sec->func_info.rec_cnt;
+		}
+		if (sec->line_info.rec_cnt) {
+			if (line_rec_sz == 0)
+				line_rec_sz = sec->line_info.rec_sz;
+			if (line_rec_sz != sec->line_info.rec_sz) {
+				pr_warn("mismatch in line_info record size %zu != %u\n",
+					line_rec_sz, sec->line_info.rec_sz);
+				return -EINVAL;
+			}
+
+			lines_sz += sizeof(struct btf_ext_info_sec) + line_rec_sz * sec->line_info.rec_cnt;
+		}
+		if (sec->core_relo_info.rec_cnt) {
+			if (core_relo_rec_sz == 0)
+				core_relo_rec_sz = sec->core_relo_info.rec_sz;
+			if (core_relo_rec_sz != sec->core_relo_info.rec_sz) {
+				pr_warn("mismatch in core_relo_info record size %zu != %u\n",
+					core_relo_rec_sz, sec->core_relo_info.rec_sz);
+				return -EINVAL;
+			}
+
+			core_relos_sz += sizeof(struct btf_ext_info_sec) + core_relo_rec_sz * sec->core_relo_info.rec_cnt;
+		}
+	}
+
+	if (!funcs_sz && !lines_sz && !core_relos_sz)
+		return 0;
+
+	total_sz += sizeof(struct btf_ext_header);
+	if (funcs_sz) {
+		funcs_sz += sizeof(__u32); /* record size prefix */
+		total_sz += funcs_sz;
+	}
+	if (lines_sz) {
+		lines_sz += sizeof(__u32); /* record size prefix */
+		total_sz += lines_sz;
+	}
+	if (core_relos_sz) {
+		core_relos_sz += sizeof(__u32); /* record size prefix */
+		total_sz += core_relos_sz;
+	}
+
+	cur = data = calloc(1, total_sz);
+	if (!data)
+		return -ENOMEM;
+
+	hdr = cur;
+	hdr->magic = BTF_MAGIC;
+	hdr->version = BTF_VERSION;
+	hdr->flags = 0;
+	hdr->hdr_len = sizeof(struct btf_ext_header);
+	cur += sizeof(struct btf_ext_header);
+
+	/* All offsets are in bytes relative to the end of this header */
+	hdr->func_info_off = 0;
+	hdr->func_info_len = funcs_sz;
+	hdr->line_info_off = funcs_sz;
+	hdr->line_info_len = lines_sz;
+	hdr->core_relo_off = funcs_sz + lines_sz;;
+	hdr->core_relo_len = core_relos_sz;
+
+	if (funcs_sz) {
+		*(__u32 *)cur = func_rec_sz;
+		cur += sizeof(__u32);
+
+		for (i = 1; i < linker->sec_cnt; i++) {
+			struct dst_sec *sec = &linker->secs[i];
+
+			sz = emit_btf_ext_data(linker, cur, sec->sec_name, &sec->func_info);
+			if (sz < 0)
+				return sz;
+
+			cur += sz;
+		}
+	}
+
+	if (lines_sz) {
+		*(__u32 *)cur = line_rec_sz;
+		cur += sizeof(__u32);
+
+		for (i = 1; i < linker->sec_cnt; i++) {
+			struct dst_sec *sec = &linker->secs[i];
+
+			sz = emit_btf_ext_data(linker, cur, sec->sec_name, &sec->line_info);
+			if (sz < 0)
+				return sz;
+
+			cur += sz;
+		}
+	}
+
+	if (core_relos_sz) {
+		*(__u32 *)cur = core_relo_rec_sz;
+		cur += sizeof(__u32);
+
+		for (i = 1; i < linker->sec_cnt; i++) {
+			struct dst_sec *sec = &linker->secs[i];
+
+			sz = emit_btf_ext_data(linker, cur, sec->sec_name, &sec->core_relo_info);
+			if (sz < 0)
+				return sz;
+
+			cur += sz;
+		}
+	}
+
+	linker->btf_ext = btf_ext__new(data, total_sz);
+	err = libbpf_get_error(linker->btf_ext);
+	if (err) {
+		linker->btf_ext = NULL;
+		pr_warn("failed to parse final .BTF.ext data: %d\n", err);
+		return err;
+	}
+
+	return 0;
+}
diff --git a/tools/lib/bpf/strset.c b/tools/lib/bpf/strset.c
new file mode 100644
index 000000000000..1fb8b49de1d6
--- /dev/null
+++ b/tools/lib/bpf/strset.c
@@ -0,0 +1,176 @@
+// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
+/* Copyright (c) 2021 Facebook */
+#include <stdint.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <errno.h>
+#include <linux/err.h>
+#include "hashmap.h"
+#include "libbpf_internal.h"
+#include "strset.h"
+
+struct strset {
+	void *strs_data;
+	size_t strs_data_len;
+	size_t strs_data_cap;
+	size_t strs_data_max_len;
+
+	/* lookup index for each unique string in strings set */
+	struct hashmap *strs_hash;
+};
+
+static size_t strset_hash_fn(const void *key, void *ctx)
+{
+	const struct strset *s = ctx;
+	const char *str = s->strs_data + (long)key;
+
+	return str_hash(str);
+}
+
+static bool strset_equal_fn(const void *key1, const void *key2, void *ctx)
+{
+	const struct strset *s = ctx;
+	const char *str1 = s->strs_data + (long)key1;
+	const char *str2 = s->strs_data + (long)key2;
+
+	return strcmp(str1, str2) == 0;
+}
+
+struct strset *strset__new(size_t max_data_sz, const char *init_data, size_t init_data_sz)
+{
+	struct strset *set = calloc(1, sizeof(*set));
+	struct hashmap *hash;
+	int err = -ENOMEM;
+
+	if (!set)
+		return ERR_PTR(-ENOMEM);
+
+	hash = hashmap__new(strset_hash_fn, strset_equal_fn, set);
+	if (IS_ERR(hash))
+		goto err_out;
+
+	set->strs_data_max_len = max_data_sz;
+	set->strs_hash = hash;
+
+	if (init_data) {
+		long off;
+
+		set->strs_data = malloc(init_data_sz);
+		if (!set->strs_data)
+			goto err_out;
+
+		memcpy(set->strs_data, init_data, init_data_sz);
+		set->strs_data_len = init_data_sz;
+		set->strs_data_cap = init_data_sz;
+
+		for (off = 0; off < set->strs_data_len; off += strlen(set->strs_data + off) + 1) {
+			/* hashmap__add() returns EEXIST if string with the same
+			 * content already is in the hash map
+			 */
+			err = hashmap__add(hash, (void *)off, (void *)off);
+			if (err == -EEXIST)
+				continue; /* duplicate */
+			if (err)
+				goto err_out;
+		}
+	}
+
+	return set;
+err_out:
+	strset__free(set);
+	return ERR_PTR(err);
+}
+
+void strset__free(struct strset *set)
+{
+	if (IS_ERR_OR_NULL(set))
+		return;
+
+	hashmap__free(set->strs_hash);
+	free(set->strs_data);
+}
+
+size_t strset__data_size(const struct strset *set)
+{
+	return set->strs_data_len;
+}
+
+const char *strset__data(const struct strset *set)
+{
+	return set->strs_data;
+}
+
+static void *strset_add_str_mem(struct strset *set, size_t add_sz)
+{
+	return libbpf_add_mem(&set->strs_data, &set->strs_data_cap, 1,
+			      set->strs_data_len, set->strs_data_max_len, add_sz);
+}
+
+/* Find string offset that corresponds to a given string *s*.
+ * Returns:
+ *   - >0 offset into string data, if string is found;
+ *   - -ENOENT, if string is not in the string data;
+ *   - <0, on any other error.
+ */
+int strset__find_str(struct strset *set, const char *s)
+{
+	long old_off, new_off, len;
+	void *p;
+
+	/* see strset__add_str() for why we do this */
+	len = strlen(s) + 1;
+	p = strset_add_str_mem(set, len);
+	if (!p)
+		return -ENOMEM;
+
+	new_off = set->strs_data_len;
+	memcpy(p, s, len);
+
+	if (hashmap__find(set->strs_hash, (void *)new_off, (void **)&old_off))
+		return old_off;
+
+	return -ENOENT;
+}
+
+/* Add a string s to the string data. If the string already exists, return its
+ * offset within string data.
+ * Returns:
+ *   - > 0 offset into string data, on success;
+ *   - < 0, on error.
+ */
+int strset__add_str(struct strset *set, const char *s)
+{
+	long old_off, new_off, len;
+	void *p;
+	int err;
+
+	/* Hashmap keys are always offsets within set->strs_data, so to even
+	 * look up some string from the "outside", we need to first append it
+	 * at the end, so that it can be addressed with an offset. Luckily,
+	 * until set->strs_data_len is incremented, that string is just a piece
+	 * of garbage for the rest of the code, so no harm, no foul. On the
+	 * other hand, if the string is unique, it's already appended and
+	 * ready to be used, only a simple set->strs_data_len increment away.
+	 */
+	len = strlen(s) + 1;
+	p = strset_add_str_mem(set, len);
+	if (!p)
+		return -ENOMEM;
+
+	new_off = set->strs_data_len;
+	memcpy(p, s, len);
+
+	/* Now attempt to add the string, but only if the string with the same
+	 * contents doesn't exist already (HASHMAP_ADD strategy). If such
+	 * string exists, we'll get its offset in old_off (that's old_key).
+	 */
+	err = hashmap__insert(set->strs_hash, (void *)new_off, (void *)new_off,
+			      HASHMAP_ADD, (const void **)&old_off, NULL);
+	if (err == -EEXIST)
+		return old_off; /* duplicated string, return existing offset */
+	if (err)
+		return err;
+
+	set->strs_data_len += len; /* new unique string, adjust data length */
+	return new_off;
+}
diff --git a/tools/lib/bpf/strset.h b/tools/lib/bpf/strset.h
new file mode 100644
index 000000000000..b6ddf77a83c2
--- /dev/null
+++ b/tools/lib/bpf/strset.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+/* Copyright (c) 2021 Facebook */
+#ifndef __LIBBPF_STRSET_H
+#define __LIBBPF_STRSET_H
+
+#include <stdbool.h>
+#include <stddef.h>
+
+struct strset;
+
+struct strset *strset__new(size_t max_data_sz, const char *init_data, size_t init_data_sz);
+void strset__free(struct strset *set);
+
+const char *strset__data(const struct strset *set);
+size_t strset__data_size(const struct strset *set);
+
+int strset__find_str(struct strset *set, const char *s);
+int strset__add_str(struct strset *set, const char *s);
+
+#endif /* __LIBBPF_STRSET_H */
diff --git a/tools/lib/bpf/xsk.h b/tools/lib/bpf/xsk.h
index a9fdea87b5cd..01c12dca9c10 100644
--- a/tools/lib/bpf/xsk.h
+++ b/tools/lib/bpf/xsk.h
@@ -3,7 +3,8 @@
 /*
  * AF_XDP user-space access library.
  *
- * Copyright(c) 2018 - 2019 Intel Corporation.
+ * Copyright (c) 2018 - 2019 Intel Corporation.
+ * Copyright (c) 2019 Facebook
  *
  * Author(s): Magnus Karlsson <magnus.karlsson@intel.com>
  */
@@ -13,15 +14,80 @@
 
 #include <stdio.h>
 #include <stdint.h>
+#include <stdbool.h>
 #include <linux/if_xdp.h>
 
 #include "libbpf.h"
-#include "libbpf_util.h"
 
 #ifdef __cplusplus
 extern "C" {
 #endif
 
+/* Load-Acquire Store-Release barriers used by the XDP socket
+ * library. The following macros should *NOT* be considered part of
+ * the xsk.h API, and is subject to change anytime.
+ *
+ * LIBRARY INTERNAL
+ */
+
+#define __XSK_READ_ONCE(x) (*(volatile typeof(x) *)&x)
+#define __XSK_WRITE_ONCE(x, v) (*(volatile typeof(x) *)&x) = (v)
+
+#if defined(__i386__) || defined(__x86_64__)
+# define libbpf_smp_store_release(p, v)					\
+	do {								\
+		asm volatile("" : : : "memory");			\
+		__XSK_WRITE_ONCE(*p, v);				\
+	} while (0)
+# define libbpf_smp_load_acquire(p)					\
+	({								\
+		typeof(*p) ___p1 = __XSK_READ_ONCE(*p);			\
+		asm volatile("" : : : "memory");			\
+		___p1;							\
+	})
+#elif defined(__aarch64__)
+# define libbpf_smp_store_release(p, v)					\
+		asm volatile ("stlr %w1, %0" : "=Q" (*p) : "r" (v) : "memory")
+# define libbpf_smp_load_acquire(p)					\
+	({								\
+		typeof(*p) ___p1;					\
+		asm volatile ("ldar %w0, %1"				\
+			      : "=r" (___p1) : "Q" (*p) : "memory");	\
+		___p1;							\
+	})
+#elif defined(__riscv)
+# define libbpf_smp_store_release(p, v)					\
+	do {								\
+		asm volatile ("fence rw,w" : : : "memory");		\
+		__XSK_WRITE_ONCE(*p, v);				\
+	} while (0)
+# define libbpf_smp_load_acquire(p)					\
+	({								\
+		typeof(*p) ___p1 = __XSK_READ_ONCE(*p);			\
+		asm volatile ("fence r,rw" : : : "memory");		\
+		___p1;							\
+	})
+#endif
+
+#ifndef libbpf_smp_store_release
+#define libbpf_smp_store_release(p, v)					\
+	do {								\
+		__sync_synchronize();					\
+		__XSK_WRITE_ONCE(*p, v);				\
+	} while (0)
+#endif
+
+#ifndef libbpf_smp_load_acquire
+#define libbpf_smp_load_acquire(p)					\
+	({								\
+		typeof(*p) ___p1 = __XSK_READ_ONCE(*p);			\
+		__sync_synchronize();					\
+		___p1;							\
+	})
+#endif
+
+/* LIBRARY INTERNAL -- END */
+
 /* Do not access these members directly. Use the functions below. */
 #define DEFINE_XSK_RING(name) \
 struct name { \
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index c3999587bc23..6448c626498f 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -21,7 +21,7 @@ endif
 
 BPF_GCC		?= $(shell command -v bpf-gcc;)
 SAN_CFLAGS	?=
-CFLAGS += -g -rdynamic -Wall -O2 $(GENFLAGS) $(SAN_CFLAGS)		\
+CFLAGS += -g -Og -rdynamic -Wall $(GENFLAGS) $(SAN_CFLAGS)		\
 	  -I$(CURDIR) -I$(INCLUDE_DIR) -I$(GENDIR) -I$(LIBDIR)		\
 	  -I$(TOOLSINCDIR) -I$(APIDIR) -I$(OUTPUT)			\
 	  -Dbpf_prog_load=bpf_prog_test_load				\
@@ -201,6 +201,7 @@ $(DEFAULT_BPFTOOL): $(wildcard $(BPFTOOLDIR)/*.[ch] $(BPFTOOLDIR)/Makefile)    \
 		    $(HOST_BPFOBJ) | $(HOST_BUILD_DIR)/bpftool
 	$(Q)$(MAKE) $(submake_extras)  -C $(BPFTOOLDIR)			       \
 		    CC=$(HOSTCC) LD=$(HOSTLD)				       \
+		    EXTRA_CFLAGS='-g -Og'				       \
 		    OUTPUT=$(HOST_BUILD_DIR)/bpftool/			       \
 		    prefix= DESTDIR=$(HOST_SCRATCH_DIR)/ install
 
@@ -218,6 +219,7 @@ $(BPFOBJ): $(wildcard $(BPFDIR)/*.[ch] $(BPFDIR)/Makefile)		       \
 	   ../../../include/uapi/linux/bpf.h                                   \
 	   | $(INCLUDE_DIR) $(BUILD_DIR)/libbpf
 	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(BUILD_DIR)/libbpf/ \
+		    EXTRA_CFLAGS='-g -Og'					       \
 		    DESTDIR=$(SCRATCH_DIR) prefix= all install_headers
 
 ifneq ($(BPFOBJ),$(HOST_BPFOBJ))
@@ -225,11 +227,12 @@ $(HOST_BPFOBJ): $(wildcard $(BPFDIR)/*.[ch] $(BPFDIR)/Makefile)                \
 	   ../../../include/uapi/linux/bpf.h                                   \
 	   | $(INCLUDE_DIR) $(HOST_BUILD_DIR)/libbpf
 	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR)                             \
-		OUTPUT=$(HOST_BUILD_DIR)/libbpf/ CC=$(HOSTCC) LD=$(HOSTLD)     \
+		    EXTRA_CFLAGS='-g -Og'					       \
+		    OUTPUT=$(HOST_BUILD_DIR)/libbpf/ CC=$(HOSTCC) LD=$(HOSTLD) \
 		    DESTDIR=$(HOST_SCRATCH_DIR)/ prefix= all install_headers
 endif
 
-$(INCLUDE_DIR)/vmlinux.h: $(VMLINUX_BTF) | $(BPFTOOL) $(INCLUDE_DIR)
+$(INCLUDE_DIR)/vmlinux.h: $(VMLINUX_BTF) $(BPFTOOL) | $(INCLUDE_DIR)
 ifeq ($(VMLINUX_H),)
 	$(call msg,GEN,,$@)
 	$(Q)$(BPFTOOL) btf dump file $(VMLINUX_BTF) format c > $@
@@ -300,6 +303,10 @@ endef
 
 SKEL_BLACKLIST := btf__% test_pinning_invalid.c test_sk_assign.c
 
+LINKED_SKELS := test_static_linked.skel.h
+
+test_static_linked.skel.h-deps := test_static_linked1.o test_static_linked2.o
+
 # Set up extra TRUNNER_XXX "temporary" variables in the environment (relies on
 # $eval()) and pass control to DEFINE_TEST_RUNNER_RULES.
 # Parameters:
@@ -320,6 +327,7 @@ TRUNNER_BPF_OBJS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.o, $$(TRUNNER_BPF_SRCS)
 TRUNNER_BPF_SKELS := $$(patsubst %.c,$$(TRUNNER_OUTPUT)/%.skel.h,	\
 				 $$(filter-out $(SKEL_BLACKLIST),	\
 					       $$(TRUNNER_BPF_SRCS)))
+TRUNNER_BPF_SKELS_LINKED := $$(addprefix $$(TRUNNER_OUTPUT)/,$(LINKED_SKELS))
 TEST_GEN_FILES += $$(TRUNNER_BPF_OBJS)
 
 # Evaluate rules now with extra TRUNNER_XXX variables above already defined
@@ -352,11 +360,22 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.o:				\
 	$$(call $(TRUNNER_BPF_BUILD_RULE),$$<,$$@,			\
 					  $(TRUNNER_BPF_CFLAGS))
 
-$(TRUNNER_BPF_SKELS): $(TRUNNER_OUTPUT)/%.skel.h:			\
-		      $(TRUNNER_OUTPUT)/%.o				\
-		      | $(BPFTOOL) $(TRUNNER_OUTPUT)
+$(TRUNNER_BPF_SKELS): %.skel.h: %.o $(BPFTOOL) | $(TRUNNER_OUTPUT)
+	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
+	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$<
+	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o)
+	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o)
+	$(Q)diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
+	$(Q)$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.o=)) > $$@
+
+$(TRUNNER_BPF_SKELS_LINKED): $(TRUNNER_BPF_OBJS) $(BPFTOOL) | $(TRUNNER_OUTPUT)
+	$$(call msg,LINK-BPF,$(TRUNNER_BINARY),$$(@:.skel.h=.o))
+	$(Q)$$(BPFTOOL) gen object $$(@:.skel.h=.linked1.o) $$(addprefix $(TRUNNER_OUTPUT)/,$$($$(@F)-deps))
+	$(Q)$$(BPFTOOL) gen object $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked1.o)
+	$(Q)$$(BPFTOOL) gen object $$(@:.skel.h=.linked3.o) $$(@:.skel.h=.linked2.o)
+	$(Q)diff $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked3.o)
 	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
-	$(Q)$$(BPFTOOL) gen skeleton $$< > $$@
+	$(Q)$$(BPFTOOL) gen skeleton $$(@:.skel.h=.linked3.o) name $$(notdir $$(@:.skel.h=)) > $$@
 endif
 
 # ensure we set up tests.h header generation rule just once
@@ -378,6 +397,7 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_EXTRA_HDRS)				\
 		      $(TRUNNER_BPF_OBJS)				\
 		      $(TRUNNER_BPF_SKELS)				\
+		      $(TRUNNER_BPF_SKELS_LINKED)			\
 		      $$(BPFOBJ) | $(TRUNNER_OUTPUT)
 	$$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
 	$(Q)cd $$(@D) && $$(CC) -I. $$(CFLAGS) -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
diff --git a/tools/testing/selftests/bpf/get_cgroup_id_user.c b/tools/testing/selftests/bpf/get_cgroup_id_user.c
index b8d6aef99db4..99628e1a1e58 100644
--- a/tools/testing/selftests/bpf/get_cgroup_id_user.c
+++ b/tools/testing/selftests/bpf/get_cgroup_id_user.c
@@ -57,6 +57,10 @@ int main(int argc, char **argv)
 	__u32 key = 0, pid;
 	int exit_code = 1;
 	char buf[256];
+	const struct timespec req = {
+		.tv_sec = 1,
+		.tv_nsec = 0,
+	};
 
 	cgroup_fd = cgroup_setup_and_join(TEST_CGROUP);
 	if (CHECK(cgroup_fd < 0, "cgroup_setup_and_join", "err %d errno %d\n", cgroup_fd, errno))
@@ -115,7 +119,7 @@ int main(int argc, char **argv)
 		goto close_pmu;
 
 	/* trigger some syscalls */
-	sleep(1);
+	syscall(__NR_nanosleep, &req, NULL);
 
 	err = bpf_map_lookup_elem(cgidmap_fd, &key, &kcgid);
 	if (CHECK(err, "bpf_map_lookup_elem", "err %d errno %d\n", err, errno))
diff --git a/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c b/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c
index f0a64d8ac59a..e42ea1195d18 100644
--- a/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c
+++ b/tools/testing/selftests/bpf/map_tests/array_map_batch_ops.c
@@ -55,7 +55,6 @@ void test_array_map_batch_ops(void)
 	int map_fd, *keys, *values, *visited;
 	__u32 count, total, total_success;
 	const __u32 max_entries = 10;
-	bool nospace_err;
 	__u64 batch = 0;
 	int err, step;
 	DECLARE_LIBBPF_OPTS(bpf_map_batch_opts, opts,
@@ -90,7 +89,6 @@ void test_array_map_batch_ops(void)
 		 * elements each.
 		 */
 		count = step;
-		nospace_err = false;
 		while (true) {
 			err = bpf_map_lookup_batch(map_fd,
 						total ? &batch : NULL, &batch,
@@ -107,9 +105,6 @@ void test_array_map_batch_ops(void)
 
 		}
 
-		if (nospace_err == true)
-			continue;
-
 		CHECK(total != max_entries, "lookup with steps",
 		      "total = %u, max_entries = %u\n", total, max_entries);
 
diff --git a/tools/testing/selftests/bpf/prog_tests/static_linked.c b/tools/testing/selftests/bpf/prog_tests/static_linked.c
new file mode 100644
index 000000000000..46556976dccc
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/static_linked.c
@@ -0,0 +1,40 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2019 Facebook */
+
+#include <test_progs.h>
+#include "test_static_linked.skel.h"
+
+void test_static_linked(void)
+{
+	int err;
+	struct test_static_linked* skel;
+
+	skel = test_static_linked__open();
+	if (!ASSERT_OK_PTR(skel, "skel_open"))
+		return;
+
+	skel->rodata->rovar1 = 1;
+	skel->bss->static_var1 = 2;
+	skel->bss->static_var11 = 3;
+
+	skel->rodata->rovar2 = 4;
+	skel->bss->static_var2 = 5;
+	skel->bss->static_var22 = 6;
+
+	err = test_static_linked__load(skel);
+	if (!ASSERT_OK(err, "skel_load"))
+		goto cleanup;
+
+	err = test_static_linked__attach(skel);
+	if (!ASSERT_OK(err, "skel_attach"))
+		goto cleanup;
+
+	/* trigger */
+	usleep(1);
+
+	ASSERT_EQ(skel->bss->var1, 1 * 2 + 2 + 3, "var1");
+	ASSERT_EQ(skel->bss->var2, 4 * 3 + 5 + 6, "var2");
+
+cleanup:
+	test_static_linked__destroy(skel);
+}
diff --git a/tools/testing/selftests/bpf/progs/bind4_prog.c b/tools/testing/selftests/bpf/progs/bind4_prog.c
index 115a3b0ad984..474c6a62078a 100644
--- a/tools/testing/selftests/bpf/progs/bind4_prog.c
+++ b/tools/testing/selftests/bpf/progs/bind4_prog.c
@@ -57,6 +57,27 @@ static __inline int bind_to_device(struct bpf_sock_addr *ctx)
 	return 0;
 }
 
+static __inline int bind_reuseport(struct bpf_sock_addr *ctx)
+{
+	int val = 1;
+
+	if (bpf_setsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)))
+		return 1;
+	if (bpf_getsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)) || !val)
+		return 1;
+	val = 0;
+	if (bpf_setsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)))
+		return 1;
+	if (bpf_getsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)) || val)
+		return 1;
+
+	return 0;
+}
+
 static __inline int misc_opts(struct bpf_sock_addr *ctx, int opt)
 {
 	int old, tmp, new = 0xeb9f;
@@ -127,6 +148,10 @@ int bind_v4_prog(struct bpf_sock_addr *ctx)
 	if (misc_opts(ctx, SO_MARK) || misc_opts(ctx, SO_PRIORITY))
 		return 0;
 
+	/* Set reuseport and unset */
+	if (bind_reuseport(ctx))
+		return 0;
+
 	ctx->user_ip4 = bpf_htonl(SERV4_REWRITE_IP);
 	ctx->user_port = bpf_htons(SERV4_REWRITE_PORT);
 
diff --git a/tools/testing/selftests/bpf/progs/bind6_prog.c b/tools/testing/selftests/bpf/progs/bind6_prog.c
index 4c0d348034b9..c19cfa869f30 100644
--- a/tools/testing/selftests/bpf/progs/bind6_prog.c
+++ b/tools/testing/selftests/bpf/progs/bind6_prog.c
@@ -63,6 +63,27 @@ static __inline int bind_to_device(struct bpf_sock_addr *ctx)
 	return 0;
 }
 
+static __inline int bind_reuseport(struct bpf_sock_addr *ctx)
+{
+	int val = 1;
+
+	if (bpf_setsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)))
+		return 1;
+	if (bpf_getsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)) || !val)
+		return 1;
+	val = 0;
+	if (bpf_setsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)))
+		return 1;
+	if (bpf_getsockopt(ctx, SOL_SOCKET, SO_REUSEPORT,
+			   &val, sizeof(val)) || val)
+		return 1;
+
+	return 0;
+}
+
 static __inline int misc_opts(struct bpf_sock_addr *ctx, int opt)
 {
 	int old, tmp, new = 0xeb9f;
@@ -141,6 +162,10 @@ int bind_v6_prog(struct bpf_sock_addr *ctx)
 	if (misc_opts(ctx, SO_MARK) || misc_opts(ctx, SO_PRIORITY))
 		return 0;
 
+	/* Set reuseport and unset */
+	if (bind_reuseport(ctx))
+		return 0;
+
 	ctx->user_ip6[0] = bpf_htonl(SERV6_REWRITE_IP_0);
 	ctx->user_ip6[1] = bpf_htonl(SERV6_REWRITE_IP_1);
 	ctx->user_ip6[2] = bpf_htonl(SERV6_REWRITE_IP_2);
diff --git a/tools/testing/selftests/bpf/progs/fentry_test.c b/tools/testing/selftests/bpf/progs/fentry_test.c
index 5f645fdaba6f..52a550d281d9 100644
--- a/tools/testing/selftests/bpf/progs/fentry_test.c
+++ b/tools/testing/selftests/bpf/progs/fentry_test.c
@@ -64,7 +64,7 @@ __u64 test7_result = 0;
 SEC("fentry/bpf_fentry_test7")
 int BPF_PROG(test7, struct bpf_fentry_test_t *arg)
 {
-	if (arg == 0)
+	if (!arg)
 		test7_result = 1;
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/fexit_test.c b/tools/testing/selftests/bpf/progs/fexit_test.c
index 0952affb22a6..8f1ccb7302e1 100644
--- a/tools/testing/selftests/bpf/progs/fexit_test.c
+++ b/tools/testing/selftests/bpf/progs/fexit_test.c
@@ -65,7 +65,7 @@ __u64 test7_result = 0;
 SEC("fexit/bpf_fentry_test7")
 int BPF_PROG(test7, struct bpf_fentry_test_t *arg)
 {
-	if (arg == 0)
+	if (!arg)
 		test7_result = 1;
 	return 0;
 }
@@ -74,7 +74,7 @@ __u64 test8_result = 0;
 SEC("fexit/bpf_fentry_test8")
 int BPF_PROG(test8, struct bpf_fentry_test_t *arg)
 {
-	if (arg->a == 0)
+	if (!arg->a)
 		test8_result = 1;
 	return 0;
 }
diff --git a/tools/testing/selftests/bpf/progs/skb_pkt_end.c b/tools/testing/selftests/bpf/progs/skb_pkt_end.c
index cf6823f42e80..7f2eaa2f89f8 100644
--- a/tools/testing/selftests/bpf/progs/skb_pkt_end.c
+++ b/tools/testing/selftests/bpf/progs/skb_pkt_end.c
@@ -4,7 +4,6 @@
 #include <bpf/bpf_core_read.h>
 #include <bpf/bpf_helpers.h>
 
-#define NULL 0
 #define INLINE __always_inline
 
 #define skb_shorter(skb, len) ((void *)(long)(skb)->data + (len) > (void *)(long)skb->data_end)
diff --git a/tools/testing/selftests/bpf/progs/test_global_func10.c b/tools/testing/selftests/bpf/progs/test_global_func10.c
index 61c2ae92ce41..97b7031d0e22 100644
--- a/tools/testing/selftests/bpf/progs/test_global_func10.c
+++ b/tools/testing/selftests/bpf/progs/test_global_func10.c
@@ -14,7 +14,7 @@ struct Big {
 
 __noinline int foo(const struct Big *big)
 {
-	if (big == 0)
+	if (!big)
 		return 0;
 
 	return bpf_get_prandom_u32() < big->y;
diff --git a/tools/testing/selftests/bpf/progs/test_static_linked1.c b/tools/testing/selftests/bpf/progs/test_static_linked1.c
new file mode 100644
index 000000000000..ea1a6c4c7172
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_static_linked1.c
@@ -0,0 +1,30 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2021 Facebook */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+
+/* 8-byte aligned .bss */
+static volatile long static_var1;
+static volatile int static_var11;
+int var1 = 0;
+/* 4-byte aligned .rodata */
+const volatile int rovar1;
+
+/* same "subprog" name in both files */
+static __noinline int subprog(int x)
+{
+	/* but different formula */
+	return x * 2;
+}
+
+SEC("raw_tp/sys_enter")
+int handler1(const void *ctx)
+{
+	var1 = subprog(rovar1) + static_var1 + static_var11;
+
+	return 0;
+}
+
+char LICENSE[] SEC("license") = "GPL";
+int VERSION SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/progs/test_static_linked2.c b/tools/testing/selftests/bpf/progs/test_static_linked2.c
new file mode 100644
index 000000000000..54d8d1ab577c
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/test_static_linked2.c
@@ -0,0 +1,31 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) 2021 Facebook */
+
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+
+/* 4-byte aligned .bss */
+static volatile int static_var2;
+static volatile int static_var22;
+int var2 = 0;
+/* 8-byte aligned .rodata */
+const volatile long rovar2;
+
+/* same "subprog" name in both files */
+static __noinline int subprog(int x)
+{
+	/* but different formula */
+	return x * 3;
+}
+
+SEC("raw_tp/sys_enter")
+int handler2(const void *ctx)
+{
+	var2 = subprog(rovar2) + static_var2 + static_var22;
+
+	return 0;
+}
+
+/* different name and/or type of the variable doesn't matter */
+char _license[] SEC("license") = "GPL";
+int _version SEC("version") = 1;
diff --git a/tools/testing/selftests/bpf/xdpxceiver.c b/tools/testing/selftests/bpf/xdpxceiver.c
index 8b0f7fdd9003..1e21a3172687 100644
--- a/tools/testing/selftests/bpf/xdpxceiver.c
+++ b/tools/testing/selftests/bpf/xdpxceiver.c
@@ -297,7 +297,7 @@ static void xsk_configure_umem(struct ifobject *data, void *buffer, u64 size)
 static void xsk_populate_fill_ring(struct xsk_umem_info *umem)
 {
 	int ret, i;
-	u32 idx;
+	u32 idx = 0;
 
 	ret = xsk_ring_prod__reserve(&umem->fq, XSK_RING_PROD__DEFAULT_NUM_DESCS, &idx);
 	if (ret != XSK_RING_PROD__DEFAULT_NUM_DESCS)
@@ -584,7 +584,7 @@ static void rx_pkt(struct xsk_socket_info *xsk, struct pollfd *fds)
 
 static void tx_only(struct xsk_socket_info *xsk, u32 *frameptr, int batch_size)
 {
-	u32 idx;
+	u32 idx = 0;
 	unsigned int i;
 	bool tx_invalid_test = stat_test_type == STAT_TEST_TX_INVALID;
 	u32 len = tx_invalid_test ? XSK_UMEM__DEFAULT_FRAME_SIZE + 1 : PKT_SIZE;