Re: [RFC PATCH net-next v6 04/14] af_vsock: generalize bind table functions
From: Stefano Garzarella <sgarzare@redhat.com>
Date: 2024-07-30 08:00:34
Also in:
bpf, kvm, linux-hyperv, lkml, virtualization
On Sun, Jul 28, 2024 at 11:52:54AM GMT, Amery Hung wrote:
On Tue, Jul 23, 2024 at 7:40 AM Stefano Garzarella [off-list ref] wrote:quoted
On Wed, Jul 10, 2024 at 09:25:45PM GMT, Amery Hung wrote:quoted
From: Bobby Eshleman <redacted> This commit makes the bind table management functions in vsock usable for different bind tables. Future work will introduce a new table for datagrams to avoid address collisions, and these functions will be used there. Signed-off-by: Bobby Eshleman <redacted> --- net/vmw_vsock/af_vsock.c | 34 +++++++++++++++++++++++++++------- 1 file changed, 27 insertions(+), 7 deletions(-)diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index acc15e11700c..d571be9cdbf0 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c@@ -232,11 +232,12 @@ static void __vsock_remove_connected(struct vsock_sock *vsk) sock_put(&vsk->sk);} -static struct sock *__vsock_find_bound_socket(struct sockaddr_vm *addr) +static struct sock *vsock_find_bound_socket_common(struct sockaddr_vm *addr, + struct list_head *bind_table) { struct vsock_sock *vsk; - list_for_each_entry(vsk, vsock_bound_sockets(addr), bound_table) { + list_for_each_entry(vsk, bind_table, bound_table) { if (vsock_addr_equals_addr(addr, &vsk->local_addr)) return sk_vsock(vsk);@@ -249,6 +250,11 @@ static struct sock *__vsock_find_bound_socket(struct sockaddr_vm *addr) return NULL;} +static struct sock *__vsock_find_bound_socket(struct sockaddr_vm *addr) +{ + return vsock_find_bound_socket_common(addr, vsock_bound_sockets(addr)); +} + static struct sock *__vsock_find_connected_socket(struct sockaddr_vm *src, struct sockaddr_vm *dst) {@@ -671,12 +677,18 @@ static void vsock_pending_work(struct work_struct *work)/**** SOCKET OPERATIONS ****/ -static int __vsock_bind_connectible(struct vsock_sock *vsk, - struct sockaddr_vm *addr) +static int vsock_bind_common(struct vsock_sock *vsk, + struct sockaddr_vm *addr, + struct list_head *bind_table, + size_t table_size) { static u32 port; struct sockaddr_vm new_addr; + if (WARN_ONCE(table_size < VSOCK_HASH_SIZE, + "table size too small, may cause overflow")) + return -EINVAL; +I'd add this in another commit.quoted
if (!port) port = get_random_u32_above(LAST_RESERVED_PORT);@@ -692,7 +704,8 @@ static int __vsock_bind_connectible(structvsock_sock *vsk, new_addr.svm_port = port++; - if (!__vsock_find_bound_socket(&new_addr)) { + if (!vsock_find_bound_socket_common(&new_addr, + &bind_table[VSOCK_HASH(addr)])) {Can we add a macro for `&bind_table[VSOCK_HASH(addr)])` ?Definitely. I will add the following macro: #define vsock_bound_sockets_in_table(bind_table, addr) \ (&bind_table[VSOCK_HASH(addr)])
yeah.
quoted
quoted
found = true; break; }@@ -709,7 +722,8 @@ static int __vsock_bind_connectible(struct vsock_sock *vsk, return -EACCES; } - if (__vsock_find_bound_socket(&new_addr)) + if (vsock_find_bound_socket_common(&new_addr, + &bind_table[VSOCK_HASH(addr)])) return -EADDRINUSE; }@@ -721,11 +735,17 @@ static int __vsock_bind_connectible(struct vsock_sock *vsk, * by AF_UNIX. */ __vsock_remove_bound(vsk); - __vsock_insert_bound(vsock_bound_sockets(&vsk->local_addr), vsk); + __vsock_insert_bound(&bind_table[VSOCK_HASH(&vsk->local_addr)], vsk); return 0;} +static int __vsock_bind_connectible(struct vsock_sock *vsk, + struct sockaddr_vm *addr) +{ + return vsock_bind_common(vsk, addr, vsock_bind_table, VSOCK_HASH_SIZE + 1);What about using ARRAY_SIZE(x) ? BTW we are using that size just to check it, but all the arrays we use are statically allocated, so what about a compile time check like BUILD_BUG_ON()?I will remove the table_size check you mentioned earlier and the argument here as the arrays are allocated statically like you mentioned. If you think this check may be a good addition, I can add a BUILD_BUG_ON() in the new vsock_bound_sockets_in_table() macro.
If you want to add it, we need to do it in a separate commit. But since we already have so many changes and both arrays are statically allocated in the same file, IMHO we can avoid the check. Stefano