Update Linux to v5.4.2
Change-Id: Idf6911045d9d382da2cfe01b1edff026404ac8fd
diff --git a/net/xdp/xsk_queue.h b/net/xdp/xsk_queue.h
index 8a64b15..eddae46 100644
--- a/net/xdp/xsk_queue.h
+++ b/net/xdp/xsk_queue.h
@@ -16,6 +16,7 @@
struct xdp_ring {
u32 producer ____cacheline_aligned_in_smp;
u32 consumer ____cacheline_aligned_in_smp;
+ u32 flags;
};
/* Used for the RX and TX queues for packets */
@@ -31,7 +32,8 @@
};
struct xsk_queue {
- struct xdp_umem_props umem_props;
+ u64 chunk_mask;
+ u64 size;
u32 ring_mask;
u32 nentries;
u32 prod_head;
@@ -42,6 +44,48 @@
u64 invalid_descs;
};
+/* The structure of the shared state of the rings are the same as the
+ * ring buffer in kernel/events/ring_buffer.c. For the Rx and completion
+ * ring, the kernel is the producer and user space is the consumer. For
+ * the Tx and fill rings, the kernel is the consumer and user space is
+ * the producer.
+ *
+ * producer consumer
+ *
+ * if (LOAD ->consumer) { LOAD ->producer
+ * (A) smp_rmb() (C)
+ * STORE $data LOAD $data
+ * smp_wmb() (B) smp_mb() (D)
+ * STORE ->producer STORE ->consumer
+ * }
+ *
+ * (A) pairs with (D), and (B) pairs with (C).
+ *
+ * Starting with (B), it protects the data from being written after
+ * the producer pointer. If this barrier was missing, the consumer
+ * could observe the producer pointer being set and thus load the data
+ * before the producer has written the new data. The consumer would in
+ * this case load the old data.
+ *
+ * (C) protects the consumer from speculatively loading the data before
+ * the producer pointer actually has been read. If we do not have this
+ * barrier, some architectures could load old data as speculative loads
+ * are not discarded as the CPU does not know there is a dependency
+ * between ->producer and data.
+ *
+ * (A) is a control dependency that separates the load of ->consumer
+ * from the stores of $data. In case ->consumer indicates there is no
+ * room in the buffer to store $data we do not. So no barrier is needed.
+ *
+ * (D) protects the load of the data to be observed to happen after the
+ * store of the consumer pointer. If we did not have this memory
+ * barrier, the producer could observe the consumer pointer being set
+ * and overwrite the data with a new value before the consumer got the
+ * chance to read the old value. The consumer would thus miss reading
+ * the old entry and very likely read the new entry twice, once right
+ * now and again after circling through the ring.
+ */
+
/* Common functions operating for both RXTX and umem queues */
static inline u64 xskq_nb_invalid_descs(struct xsk_queue *q)
@@ -74,11 +118,36 @@
return q->nentries - (producer - q->cons_tail);
}
+static inline bool xskq_has_addrs(struct xsk_queue *q, u32 cnt)
+{
+ u32 entries = q->prod_tail - q->cons_tail;
+
+ if (entries >= cnt)
+ return true;
+
+ /* Refresh the local pointer. */
+ q->prod_tail = READ_ONCE(q->ring->producer);
+ entries = q->prod_tail - q->cons_tail;
+
+ return entries >= cnt;
+}
+
/* UMEM queue */
+static inline bool xskq_crosses_non_contig_pg(struct xdp_umem *umem, u64 addr,
+ u64 length)
+{
+ bool cross_pg = (addr & (PAGE_SIZE - 1)) + length > PAGE_SIZE;
+ bool next_pg_contig =
+ (unsigned long)umem->pages[(addr >> PAGE_SHIFT)].addr &
+ XSK_NEXT_PG_CONTIG_MASK;
+
+ return cross_pg && !next_pg_contig;
+}
+
static inline bool xskq_is_valid_addr(struct xsk_queue *q, u64 addr)
{
- if (addr >= q->umem_props.size) {
+ if (addr >= q->size) {
q->invalid_descs++;
return false;
}
@@ -86,25 +155,54 @@
return true;
}
-static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr)
+static inline bool xskq_is_valid_addr_unaligned(struct xsk_queue *q, u64 addr,
+ u64 length,
+ struct xdp_umem *umem)
+{
+ u64 base_addr = xsk_umem_extract_addr(addr);
+
+ addr = xsk_umem_add_offset_to_addr(addr);
+ if (base_addr >= q->size || addr >= q->size ||
+ xskq_crosses_non_contig_pg(umem, addr, length)) {
+ q->invalid_descs++;
+ return false;
+ }
+
+ return true;
+}
+
+static inline u64 *xskq_validate_addr(struct xsk_queue *q, u64 *addr,
+ struct xdp_umem *umem)
{
while (q->cons_tail != q->cons_head) {
struct xdp_umem_ring *ring = (struct xdp_umem_ring *)q->ring;
unsigned int idx = q->cons_tail & q->ring_mask;
- *addr = READ_ONCE(ring->desc[idx]) & q->umem_props.chunk_mask;
+ *addr = READ_ONCE(ring->desc[idx]) & q->chunk_mask;
+
+ if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
+ if (xskq_is_valid_addr_unaligned(q, *addr,
+ umem->chunk_size_nohr,
+ umem))
+ return addr;
+ goto out;
+ }
+
if (xskq_is_valid_addr(q, *addr))
return addr;
+out:
q->cons_tail++;
}
return NULL;
}
-static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr)
+static inline u64 *xskq_peek_addr(struct xsk_queue *q, u64 *addr,
+ struct xdp_umem *umem)
{
if (q->cons_tail == q->cons_head) {
+ smp_mb(); /* D, matches A */
WRITE_ONCE(q->ring->consumer, q->cons_tail);
q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);
@@ -112,7 +210,7 @@
smp_rmb();
}
- return xskq_validate_addr(q, addr);
+ return xskq_validate_addr(q, addr, umem);
}
static inline void xskq_discard_addr(struct xsk_queue *q)
@@ -127,10 +225,11 @@
if (xskq_nb_free(q, q->prod_tail, 1) == 0)
return -ENOSPC;
+ /* A, matches D */
ring->desc[q->prod_tail++ & q->ring_mask] = addr;
/* Order producer and data */
- smp_wmb();
+ smp_wmb(); /* B, matches C */
WRITE_ONCE(q->ring->producer, q->prod_tail);
return 0;
@@ -143,6 +242,7 @@
if (xskq_nb_free(q, q->prod_head, LAZY_UPDATE_THRESHOLD) == 0)
return -ENOSPC;
+ /* A, matches D */
ring->desc[q->prod_head++ & q->ring_mask] = addr;
return 0;
}
@@ -151,7 +251,7 @@
u32 nb_entries)
{
/* Order producer and data */
- smp_wmb();
+ smp_wmb(); /* B, matches C */
q->prod_tail += nb_entries;
WRITE_ONCE(q->ring->producer, q->prod_tail);
@@ -162,19 +262,33 @@
if (xskq_nb_free(q, q->prod_head, 1) == 0)
return -ENOSPC;
+ /* A, matches D */
q->prod_head++;
return 0;
}
/* Rx/Tx queue */
-static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d)
+static inline bool xskq_is_valid_desc(struct xsk_queue *q, struct xdp_desc *d,
+ struct xdp_umem *umem)
{
+ if (umem->flags & XDP_UMEM_UNALIGNED_CHUNK_FLAG) {
+ if (!xskq_is_valid_addr_unaligned(q, d->addr, d->len, umem))
+ return false;
+
+ if (d->len > umem->chunk_size_nohr || d->options) {
+ q->invalid_descs++;
+ return false;
+ }
+
+ return true;
+ }
+
if (!xskq_is_valid_addr(q, d->addr))
return false;
- if (((d->addr + d->len) & q->umem_props.chunk_mask) !=
- (d->addr & q->umem_props.chunk_mask)) {
+ if (((d->addr + d->len) & q->chunk_mask) != (d->addr & q->chunk_mask) ||
+ d->options) {
q->invalid_descs++;
return false;
}
@@ -183,14 +297,15 @@
}
static inline struct xdp_desc *xskq_validate_desc(struct xsk_queue *q,
- struct xdp_desc *desc)
+ struct xdp_desc *desc,
+ struct xdp_umem *umem)
{
while (q->cons_tail != q->cons_head) {
struct xdp_rxtx_ring *ring = (struct xdp_rxtx_ring *)q->ring;
unsigned int idx = q->cons_tail & q->ring_mask;
*desc = READ_ONCE(ring->desc[idx]);
- if (xskq_is_valid_desc(q, desc))
+ if (xskq_is_valid_desc(q, desc, umem))
return desc;
q->cons_tail++;
@@ -200,17 +315,19 @@
}
static inline struct xdp_desc *xskq_peek_desc(struct xsk_queue *q,
- struct xdp_desc *desc)
+ struct xdp_desc *desc,
+ struct xdp_umem *umem)
{
if (q->cons_tail == q->cons_head) {
+ smp_mb(); /* D, matches A */
WRITE_ONCE(q->ring->consumer, q->cons_tail);
q->cons_head = q->cons_tail + xskq_nb_avail(q, RX_BATCH_SIZE);
/* Order consumer and data */
- smp_rmb();
+ smp_rmb(); /* C, matches B */
}
- return xskq_validate_desc(q, desc);
+ return xskq_validate_desc(q, desc, umem);
}
static inline void xskq_discard_desc(struct xsk_queue *q)
@@ -227,6 +344,7 @@
if (xskq_nb_free(q, q->prod_head, 1) == 0)
return -ENOSPC;
+ /* A, matches D */
idx = (q->prod_head++) & q->ring_mask;
ring->desc[idx].addr = addr;
ring->desc[idx].len = len;
@@ -237,9 +355,9 @@
static inline void xskq_produce_flush_desc(struct xsk_queue *q)
{
/* Order producer and data */
- smp_wmb();
+ smp_wmb(); /* B, matches C */
- q->prod_tail = q->prod_head,
+ q->prod_tail = q->prod_head;
WRITE_ONCE(q->ring->producer, q->prod_tail);
}
@@ -253,8 +371,11 @@
return xskq_nb_free(q, q->prod_tail, q->nentries) == q->nentries;
}
-void xskq_set_umem(struct xsk_queue *q, struct xdp_umem_props *umem_props);
+void xskq_set_umem(struct xsk_queue *q, u64 size, u64 chunk_mask);
struct xsk_queue *xskq_create(u32 nentries, bool umem_queue);
void xskq_destroy(struct xsk_queue *q_ops);
+/* Executed by the core when the entire UMEM gets freed */
+void xsk_reuseq_destroy(struct xdp_umem *umem);
+
#endif /* _LINUX_XSK_QUEUE_H */