8 files changed, 30 insertions, 20 deletions

diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c
index e0918d180f08..46f5f29605d4 100644
--- a/kernel/bpf/cpumap.c
+++ b/kernel/bpf/cpumap.c
@@ -69,7 +69,7 @@ struct bpf_cpu_map {
 };
 
 static int bq_flush_to_queue(struct bpf_cpu_map_entry *rcpu,
-                             struct xdp_bulk_queue *bq);
+                             struct xdp_bulk_queue *bq, bool in_napi_ctx);
 
 static u64 cpu_map_bitmap_size(const union bpf_attr *attr)
 {
@@ -375,7 +375,7 @@ static void __cpu_map_entry_free(struct rcu_head *rcu)
                 struct xdp_bulk_queue *bq = per_cpu_ptr(rcpu->bulkq, cpu);
 
                 /* No concurrent bq_enqueue can run at this point */
-                bq_flush_to_queue(rcpu, bq);
+                bq_flush_to_queue(rcpu, bq, false);
         }
         free_percpu(rcpu->bulkq);
         /* Cannot kthread_stop() here, last put free rcpu resources */
@@ -558,7 +558,7 @@ const struct bpf_map_ops cpu_map_ops = {
 };
 
 static int bq_flush_to_queue(struct bpf_cpu_map_entry *rcpu,
-                             struct xdp_bulk_queue *bq)
+                             struct xdp_bulk_queue *bq, bool in_napi_ctx)
 {
         unsigned int processed = 0, drops = 0;
         const int to_cpu = rcpu->cpu;
@@ -578,7 +578,10 @@ static int bq_flush_to_queue(struct bpf_cpu_map_entry *rcpu,
                 err = __ptr_ring_produce(q, xdpf);
                 if (err) {
                         drops++;
-                        xdp_return_frame_rx_napi(xdpf);
+                        if (likely(in_napi_ctx))
+                                xdp_return_frame_rx_napi(xdpf);
+                        else
+                                xdp_return_frame(xdpf);
                 }
                 processed++;
         }
@@ -598,7 +601,7 @@ static int bq_enqueue(struct bpf_cpu_map_entry *rcpu, struct xdp_frame *xdpf)
         struct xdp_bulk_queue *bq = this_cpu_ptr(rcpu->bulkq);
 
         if (unlikely(bq->count == CPU_MAP_BULK_SIZE))
-                bq_flush_to_queue(rcpu, bq);
+                bq_flush_to_queue(rcpu, bq, true);
 
         /* Notice, xdp_buff/page MUST be queued here, long enough for
          * driver to code invoking us to finished, due to driver
@@ -661,7 +664,7 @@ void __cpu_map_flush(struct bpf_map *map)
 
                 /* Flush all frames in bulkq to real queue */
                 bq = this_cpu_ptr(rcpu->bulkq);
-                bq_flush_to_queue(rcpu, bq);
+                bq_flush_to_queue(rcpu, bq, true);
 
                 /* If already running, costs spin_lock_irqsave + smb_mb */
                 wake_up_process(rcpu->kthread);
diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index d361fc1e3bf3..750d45edae79 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -217,7 +217,8 @@ void __dev_map_insert_ctx(struct bpf_map *map, u32 bit)
 }
 
 static int bq_xmit_all(struct bpf_dtab_netdev *obj,
-                       struct xdp_bulk_queue *bq, u32 flags)
+                       struct xdp_bulk_queue *bq, u32 flags,
+                       bool in_napi_ctx)
 {
         struct net_device *dev = obj->dev;
         int sent = 0, drops = 0, err = 0;
@@ -254,7 +255,10 @@ error:
                 struct xdp_frame *xdpf = bq->q[i];
 
                 /* RX path under NAPI protection, can return frames faster */
-                xdp_return_frame_rx_napi(xdpf);
+                if (likely(in_napi_ctx))
+                        xdp_return_frame_rx_napi(xdpf);
+                else
+                        xdp_return_frame(xdpf);
                 drops++;
         }
         goto out;
@@ -286,7 +290,7 @@ void __dev_map_flush(struct bpf_map *map)
                 __clear_bit(bit, bitmap);
 
                 bq = this_cpu_ptr(dev->bulkq);
-                bq_xmit_all(dev, bq, XDP_XMIT_FLUSH);
+                bq_xmit_all(dev, bq, XDP_XMIT_FLUSH, true);
         }
 }
 
@@ -316,7 +320,7 @@ static int bq_enqueue(struct bpf_dtab_netdev *obj, struct xdp_frame *xdpf,
         struct xdp_bulk_queue *bq = this_cpu_ptr(obj->bulkq);
 
         if (unlikely(bq->count == DEV_MAP_BULK_SIZE))
-                bq_xmit_all(obj, bq, 0);
+                bq_xmit_all(obj, bq, 0, true);
 
         /* Ingress dev_rx will be the same for all xdp_frame's in
          * bulk_queue, because bq stored per-CPU and must be flushed
@@ -385,7 +389,7 @@ static void dev_map_flush_old(struct bpf_dtab_netdev *dev)
                         __clear_bit(dev->bit, bitmap);
 
                         bq = per_cpu_ptr(dev->bulkq, cpu);
-                        bq_xmit_all(dev, bq, XDP_XMIT_FLUSH);
+                        bq_xmit_all(dev, bq, XDP_XMIT_FLUSH, false);
                 }
         }
 }
diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index 98fb7938beea..c4d75c52b4fc 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -1048,12 +1048,12 @@ static int bpf_tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
         timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
 
         while (msg_data_left(msg)) {
-                struct sk_msg_buff *m;
+                struct sk_msg_buff *m = NULL;
                 bool enospc = false;
                 int copy;
 
                 if (sk->sk_err) {
-                        err = sk->sk_err;
+                        err = -sk->sk_err;
                         goto out_err;
                 }
 
@@ -1116,8 +1116,11 @@ wait_for_sndbuf:
                 set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
 wait_for_memory:
                 err = sk_stream_wait_memory(sk, &timeo);
-                if (err)
+                if (err) {
+                        if (m && m != psock->cork)
+                                free_start_sg(sk, m);
                         goto out_err;
+                }
         }
 out_err:
         if (err < 0)
diff --git a/samples/bpf/xdp_redirect_cpu_kern.c b/samples/bpf/xdp_redirect_cpu_kern.c
index 303e9e7161f3..4938dcbaecbf 100644
--- a/samples/bpf/xdp_redirect_cpu_kern.c
+++ b/samples/bpf/xdp_redirect_cpu_kern.c
@@ -14,7 +14,7 @@
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
 
-#define MAX_CPUS 12 /* WARNING - sync with _user.c */
+#define MAX_CPUS 64 /* WARNING - sync with _user.c */
 
 /* Special map type that can XDP_REDIRECT frames to another CPU */
 struct bpf_map_def SEC("maps") cpu_map = {
diff --git a/samples/bpf/xdp_redirect_cpu_user.c b/samples/bpf/xdp_redirect_cpu_user.c
index f6efaefd485b..4b4d78fffe30 100644
--- a/samples/bpf/xdp_redirect_cpu_user.c
+++ b/samples/bpf/xdp_redirect_cpu_user.c
@@ -19,7 +19,7 @@ static const char *__doc__ =
 #include <arpa/inet.h>
 #include <linux/if_link.h>
 
-#define MAX_CPUS 12 /* WARNING - sync with _kern.c */
+#define MAX_CPUS 64 /* WARNING - sync with _kern.c */
 
 /* How many xdp_progs are defined in _kern.c */
 #define MAX_PROG 5
@@ -527,7 +527,7 @@ static void stress_cpumap(void)
          * procedure.
          */
         create_cpu_entry(1,  1024, 0, false);
-        create_cpu_entry(1,   128, 0, false);
+        create_cpu_entry(1,     8, 0, false);
         create_cpu_entry(1, 16000, 0, false);
 }
 
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 2d270c560df3..c36a3a76986a 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -1,4 +1,4 @@
-/* SPDX-License-Identifier: GPL-2.0 */
+// SPDX-License-Identifier: LGPL-2.1
 /* Copyright (c) 2018 Facebook */
 
 #include <stdlib.h>
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index e2a09a155f84..caac3a404dc5 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -1,4 +1,4 @@
-/* SPDX-License-Identifier: GPL-2.0 */
+/* SPDX-License-Identifier: LGPL-2.1 */
 /* Copyright (c) 2018 Facebook */
 
 #ifndef __BPF_BTF_H
diff --git a/tools/testing/selftests/bpf/test_sockmap.c b/tools/testing/selftests/bpf/test_sockmap.c
index 9e78df207919..0c7d9e556b47 100644
--- a/tools/testing/selftests/bpf/test_sockmap.c
+++ b/tools/testing/selftests/bpf/test_sockmap.c
@@ -354,7 +354,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
                 while (s->bytes_recvd < total_bytes) {
                         if (txmsg_cork) {
                                 timeout.tv_sec = 0;
-                                timeout.tv_usec = 1000;
+                                timeout.tv_usec = 300000;
                         } else {
                                 timeout.tv_sec = 1;
                                 timeout.tv_usec = 0;