| Message ID | 20201007184021.27584-1-rohitm@chelsio.com |
|---|---|
| State | New |
| Headers | show |
| Series | [net,v3] net/tls: sendfile fails with ktls offload | expand |
On Thu, 8 Oct 2020 00:10:21 +0530 Rohit Maheshwari wrote: > At first when sendpage gets called, if there is more data, 'more' in > tls_push_data() gets set which later sets pending_open_record_frags, but > when there is no more data in file left, and last time tls_push_data() > gets called, pending_open_record_frags doesn't get reset. And later when > 2 bytes of encrypted alert comes as sendmsg, it first checks for > pending_open_record_frags, and since this is set, it creates a record with > 0 data bytes to encrypt, meaning record length is prepend_size + tag_size > only, which causes problem. > We should set/reset pending_open_record_frags based on more bit. > > Fixes: d829e9c4112b ("tls: convert to generic sk_msg interface") > Signed-off-by: Rohit Maheshwari <rohitm@chelsio.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
On Thu, 8 Oct 2020 00:10:21 +0530 Rohit Maheshwari wrote: > At first when sendpage gets called, if there is more data, 'more' in > tls_push_data() gets set which later sets pending_open_record_frags, but > when there is no more data in file left, and last time tls_push_data() > gets called, pending_open_record_frags doesn't get reset. And later when > 2 bytes of encrypted alert comes as sendmsg, it first checks for > pending_open_record_frags, and since this is set, it creates a record with > 0 data bytes to encrypt, meaning record length is prepend_size + tag_size > only, which causes problem. > We should set/reset pending_open_record_frags based on more bit. > > Fixes: d829e9c4112b ("tls: convert to generic sk_msg interface") > Signed-off-by: Rohit Maheshwari <rohitm@chelsio.com> Applied to net.
diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index b74e2741f74f..cec86229a6a0 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -418,14 +418,14 @@ static int tls_push_data(struct sock *sk, struct tls_context *tls_ctx = tls_get_ctx(sk); struct tls_prot_info *prot = &tls_ctx->prot_info; struct tls_offload_context_tx *ctx = tls_offload_ctx_tx(tls_ctx); - int more = flags & (MSG_SENDPAGE_NOTLAST | MSG_MORE); struct tls_record_info *record = ctx->open_record; int tls_push_record_flags; struct page_frag *pfrag; size_t orig_size = size; u32 max_open_record_len; - int copy, rc = 0; + bool more = false; bool done = false; + int copy, rc = 0; long timeo; if (flags & @@ -492,9 +492,8 @@ static int tls_push_data(struct sock *sk, if (!size) { last_record: tls_push_record_flags = flags; - if (more) { - tls_ctx->pending_open_record_frags = - !!record->num_frags; + if (flags & (MSG_SENDPAGE_NOTLAST | MSG_MORE)) { + more = true; break; } @@ -526,6 +525,8 @@ static int tls_push_data(struct sock *sk, } } while (!done); + tls_ctx->pending_open_record_frags = more; + if (orig_size - size > 0) rc = orig_size - size;
At first when sendpage gets called, if there is more data, 'more' in tls_push_data() gets set which later sets pending_open_record_frags, but when there is no more data in file left, and last time tls_push_data() gets called, pending_open_record_frags doesn't get reset. And later when 2 bytes of encrypted alert comes as sendmsg, it first checks for pending_open_record_frags, and since this is set, it creates a record with 0 data bytes to encrypt, meaning record length is prepend_size + tag_size only, which causes problem. We should set/reset pending_open_record_frags based on more bit. Fixes: d829e9c4112b ("tls: convert to generic sk_msg interface") Signed-off-by: Rohit Maheshwari <rohitm@chelsio.com> --- net/tls/tls_device.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)