io_uring/zcrx: add support for multiple ifqs

Allow the user to register multiple ifqs / zcrx contexts. With that we
can use multiple interfaces / interface queues in a single io_uring
instance.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/668b03bee03b5216564482edcfefbc2ee337dd30.1745141261.git.asml.silence@gmail.com
[axboe: fold in fix]
Signed-off-by: Jens Axboe <axboe@kernel.dk>

4 files changed, 56 insertions, 30 deletions

diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h
index 06d722289fc5..7e23e993280e 100644
--- a/include/linux/io_uring_types.h
+++ b/include/linux/io_uring_types.h
@@ -40,8 +40,6 @@ enum io_uring_cmd_flags {
 	IO_URING_F_TASK_DEAD		= (1 << 13),
 };
 
-struct io_zcrx_ifq;
-
 struct io_wq_work_node {
 	struct io_wq_work_node *next;
 };
@@ -394,7 +392,8 @@ struct io_ring_ctx {
 	struct wait_queue_head		poll_wq;
 	struct io_restriction		restrictions;
 
-	struct io_zcrx_ifq		*ifq;
+	/* Stores zcrx object pointers of type struct io_zcrx_ifq */
+	struct xarray			zcrx_ctxs;
 
 	u32			pers_next;
 	struct xarray		personalities;
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 75c022526548..0dc6c2f1295e 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -359,6 +359,7 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 	INIT_LIST_HEAD(&ctx->tctx_list);
 	ctx->submit_state.free_list.next = NULL;
 	INIT_HLIST_HEAD(&ctx->waitid_list);
+	xa_init_flags(&ctx->zcrx_ctxs, XA_FLAGS_ALLOC);
 #ifdef CONFIG_FUTEX
 	INIT_HLIST_HEAD(&ctx->futex_list);
 #endif
@@ -2889,7 +2890,7 @@ static __cold void io_ring_exit_work(struct work_struct *work)
 			io_cqring_overflow_kill(ctx);
 			mutex_unlock(&ctx->uring_lock);
 		}
-		if (ctx->ifq) {
+		if (!xa_empty(&ctx->zcrx_ctxs)) {
 			mutex_lock(&ctx->uring_lock);
 			io_shutdown_zcrx_ifqs(ctx);
 			mutex_unlock(&ctx->uring_lock);
diff --git a/io_uring/net.c b/io_uring/net.c
index 782f8e76c5c7..b3a643675ce8 100644
--- a/io_uring/net.c
+++ b/io_uring/net.c
@@ -1189,11 +1189,10 @@ int io_recvzc_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 		return -EINVAL;
 
 	ifq_idx = READ_ONCE(sqe->zcrx_ifq_idx);
-	if (ifq_idx != 0)
-		return -EINVAL;
-	zc->ifq = req->ctx->ifq;
+	zc->ifq = xa_load(&req->ctx->zcrx_ctxs, ifq_idx);
 	if (!zc->ifq)
 		return -EINVAL;
+
 	zc->len = READ_ONCE(sqe->len);
 	zc->flags = READ_ONCE(sqe->ioprio);
 	zc->msg_flags = READ_ONCE(sqe->msg_flags);
diff --git a/io_uring/zcrx.c b/io_uring/zcrx.c
index 033284b695c7..22f420d6fbb9 100644
--- a/io_uring/zcrx.c
+++ b/io_uring/zcrx.c
@@ -156,8 +156,10 @@ static void io_zcrx_get_niov_uref(struct net_iov *niov)
 
 static int io_allocate_rbuf_ring(struct io_zcrx_ifq *ifq,
 				 struct io_uring_zcrx_ifq_reg *reg,
-				 struct io_uring_region_desc *rd)
+				 struct io_uring_region_desc *rd,
+				 u32 id)
 {
+	u64 mmap_offset;
 	size_t off, size;
 	void *ptr;
 	int ret;
@@ -167,7 +169,10 @@ static int io_allocate_rbuf_ring(struct io_zcrx_ifq *ifq,
 	if (size > rd->size)
 		return -EINVAL;
 
-	ret = io_create_region(ifq->ctx, &ifq->region, rd, IORING_MAP_OFF_ZCRX_REGION);
+	mmap_offset = IORING_MAP_OFF_ZCRX_REGION;
+	mmap_offset += id << IORING_OFF_PBUF_SHIFT;
+
+	ret = io_create_region(ifq->ctx, &ifq->region, rd, mmap_offset);
 	if (ret < 0)
 		return ret;
 
@@ -179,9 +184,6 @@ static int io_allocate_rbuf_ring(struct io_zcrx_ifq *ifq,
 
 static void io_free_rbuf_ring(struct io_zcrx_ifq *ifq)
 {
-	if (WARN_ON_ONCE(ifq->ctx->ifq))
-		return;
-
 	io_free_region(ifq->ctx, &ifq->region);
 	ifq->rq_ring = NULL;
 	ifq->rqes = NULL;
@@ -343,11 +345,11 @@ static void io_zcrx_ifq_free(struct io_zcrx_ifq *ifq)
 struct io_mapped_region *io_zcrx_get_region(struct io_ring_ctx *ctx,
 					    unsigned int id)
 {
+	struct io_zcrx_ifq *ifq = xa_load(&ctx->zcrx_ctxs, id);
+
 	lockdep_assert_held(&ctx->mmap_lock);
 
-	if (id != 0 || !ctx->ifq)
-		return NULL;
-	return &ctx->ifq->region;
+	return ifq ? &ifq->region : NULL;
 }
 
 int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
@@ -359,6 +361,7 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 	struct io_uring_region_desc rd;
 	struct io_zcrx_ifq *ifq;
 	int ret;
+	u32 id;
 
 	/*
 	 * 1. Interface queue allocation.
@@ -371,8 +374,6 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 	if (!(ctx->flags & IORING_SETUP_DEFER_TASKRUN &&
 	      ctx->flags & IORING_SETUP_CQE32))
 		return -EINVAL;
-	if (ctx->ifq)
-		return -EBUSY;
 	if (copy_from_user(&reg, arg, sizeof(reg)))
 		return -EFAULT;
 	if (copy_from_user(&rd, u64_to_user_ptr(reg.region_ptr), sizeof(rd)))
@@ -396,7 +397,14 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 	if (!ifq)
 		return -ENOMEM;
 
-	ret = io_allocate_rbuf_ring(ifq, &reg, &rd);
+	scoped_guard(mutex, &ctx->mmap_lock) {
+		/* preallocate id */
+		ret = xa_alloc(&ctx->zcrx_ctxs, &id, NULL, xa_limit_31b, GFP_KERNEL);
+		if (ret)
+			goto ifq_free;
+	}
+
+	ret = io_allocate_rbuf_ring(ifq, &reg, &rd, id);
 	if (ret)
 		goto err;
 
@@ -428,6 +436,14 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 	reg.offsets.rqes = sizeof(struct io_uring);
 	reg.offsets.head = offsetof(struct io_uring, head);
 	reg.offsets.tail = offsetof(struct io_uring, tail);
+	reg.zcrx_id = id;
+
+	scoped_guard(mutex, &ctx->mmap_lock) {
+		/* publish ifq */
+		ret = -ENOMEM;
+		if (xa_store(&ctx->zcrx_ctxs, id, ifq, GFP_KERNEL))
+			goto err;
+	}
 
 	if (copy_to_user(arg, &reg, sizeof(reg)) ||
 	    copy_to_user(u64_to_user_ptr(reg.region_ptr), &rd, sizeof(rd)) ||
@@ -435,26 +451,34 @@ int io_register_zcrx_ifq(struct io_ring_ctx *ctx,
 		ret = -EFAULT;
 		goto err;
 	}
-	scoped_guard(mutex, &ctx->mmap_lock)
-		ctx->ifq = ifq;
 	return 0;
 err:
+	scoped_guard(mutex, &ctx->mmap_lock)
+		xa_erase(&ctx->zcrx_ctxs, id);
+ifq_free:
 	io_zcrx_ifq_free(ifq);
 	return ret;
 }
 
 void io_unregister_zcrx_ifqs(struct io_ring_ctx *ctx)
 {
-	struct io_zcrx_ifq *ifq = ctx->ifq;
+	struct io_zcrx_ifq *ifq;
+	unsigned long id;
 
 	lockdep_assert_held(&ctx->uring_lock);
 
-	if (!ifq)
-		return;
+	while (1) {
+		scoped_guard(mutex, &ctx->mmap_lock) {
+			ifq = xa_find(&ctx->zcrx_ctxs, &id, ULONG_MAX, XA_PRESENT);
+			if (ifq)
+				xa_erase(&ctx->zcrx_ctxs, id);
+		}
+		if (!ifq)
+			break;
+		io_zcrx_ifq_free(ifq);
+	}
 
-	scoped_guard(mutex, &ctx->mmap_lock)
-		ctx->ifq = NULL;
-	io_zcrx_ifq_free(ifq);
+	xa_destroy(&ctx->zcrx_ctxs);
 }
 
 static struct net_iov *__io_zcrx_get_free_niov(struct io_zcrx_area *area)
@@ -511,12 +535,15 @@ static void io_zcrx_scrub(struct io_zcrx_ifq *ifq)
 
 void io_shutdown_zcrx_ifqs(struct io_ring_ctx *ctx)
 {
+	struct io_zcrx_ifq *ifq;
+	unsigned long index;
+
 	lockdep_assert_held(&ctx->uring_lock);
 
-	if (!ctx->ifq)
-		return;
-	io_zcrx_scrub(ctx->ifq);
-	io_close_queue(ctx->ifq);
+	xa_for_each(&ctx->zcrx_ctxs, index, ifq) {
+		io_zcrx_scrub(ifq);
+		io_close_queue(ifq);
+	}
 }
 
 static inline u32 io_zcrx_rqring_entries(struct io_zcrx_ifq *ifq)