net/tls: fixes for NVMe-over-TLS
Subsystems (5)
PATCH v2 2/4 net/tls: handle MSG_EOR for tls_device TX flow
In reply to Hannes Reinecketls_push_data() MSG_MORE / MSG_SENDPAGE_NOTLAST, but bails out on MSG_EOR.
But seeing that MSG_EOR is basically the opposite of MSG_MORE / MSG_SENDPAGE_NOTLAST this patch adds handling MSG_EOR by treating it as the negation of MSG_MORE.
--- net/tls/tls_device.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
net/tls/tls_device.c +7 -1
@@ -449,7 +449,7 @@ static int tls_push_data(struct sock *sk,long timeo;if (flags & - ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL |MSG_SENDPAGE_NOTLAST )) + ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL |MSG_SENDPAGE_NOTLAST |MSG_EOR ))return -EOPNOTSUPP ;if (unlikely (sk->sk_err))
@@ -529,6 +529,10 @@ static int tls_push_data(struct sock *sk, more = true;break ; } +if (flags &MSG_EOR ) { + more = false; +break ; + } done = true; }
@@ -603,6 +607,8 @@ int tls_device_sendpage(struct sock *sk, struct page *page,if (flags &MSG_SENDPAGE_NOTLAST ) flags |=MSG_MORE ; +if (flags &MSG_EOR ) + flags &= ~(MSG_MORE |MSG_SENDPAGE_NOTLAST );mutex_lock (&tls_ctx->tx_lock);lock_sock (sk);
Signature · 1 line
2.35.3
PATCH v2 4/4 selftests/net/tls: add test for MSG_EOR
In reply to Hannes ReineckeAs the recent patch is modifying the behaviour for TLS re MSG_EOR handling we should be having a test for it.
--- tools/testing/selftests/net/tls.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
tools/testing/selftests/net/tls.c +11 -0
@@ -476,6 +476,17 @@ TEST_F(tls, msg_more_unsent)EXPECT_EQ (recv (self->cfd, buf, send_len,MSG_DONTWAIT ),-1 ); } +TEST_F(tls, msg_eor) +{ +char const *test_str ="test_read" ; +int send_len =10 ; +char buf[10 ]; + +EXPECT_EQ (send (self->fd, test_str, send_len,MSG_EOR ), send_len); +EXPECT_EQ (recv (self->cfd, buf, send_len,MSG_WAITALL ), send_len); +EXPECT_EQ (memcmp (buf, test_str, send_len),0 ); +} + TEST_F(tls, sendmsg_single) {struct msghdr msg;
Signature · 1 line
2.35.3
v2 cover net/tls: fixes for NVMe-over-TLS
Hi all,
here are some small fixes to get NVMe-over-TLS up and running. The first two are just minor modifications to have MSG_EOR handled for TLS, but the third implements the ->read_sock() callback for tls_sw and I guess could do with some reviews.
As usual, comments and reviews are welcome.
Changes to the original submission:
- Add a testcase for MSG_EOR handling
Hannes Reinecke (4):
net/tls: handle MSG_EOR for tls_sw TX flow
net/tls: handle MSG_EOR for tls_device TX flow
net/tls: implement ->read_sock()
selftests/net/tls: add test for MSG_EOR
net/tls/tls.h | 2 + net/tls/tls_device.c | 8 ++- net/tls/tls_main.c | 2 + net/tls/tls_sw.c | 82 +++++++++++++++++++++++++++++-- tools/testing/selftests/net/tls.c | 11 +++++ 5 files changed, 101 insertions(+), 4 deletions(-)
Signature · 1 line
2.35.3
PATCH v2 1/4 net/tls: handle MSG_EOR for tls_sw TX flow
tls_sw_sendmsg() / tls_do_sw_sendpage() already handles MSG_MORE / MSG_SENDPAGE_NOTLAST, but bails out on MSG_EOR. But seeing that MSG_EOR is basically the opposite of MSG_MORE / MSG_SENDPAGE_NOTLAST this patch adds handling MSG_EOR by treating it as the negation of MSG_MORE.
--- net/tls/tls_sw.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)
net/tls/tls_sw.c +8 -3
@@ -953,9 +953,12 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)int pending;if (msg->msg_flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL | -MSG_CMSG_COMPAT )) +MSG_EOR |MSG_CMSG_COMPAT ))return -EOPNOTSUPP ; +if (msg->msg_flags &MSG_EOR ) + eor = true; + ret =mutex_lock_interruptible (&tls_ctx->tx_lock);if (ret)return ret;
@@ -1173,6 +1176,8 @@ static int tls_sw_do_sendpage(struct sock *sk, struct page *page,bool eor; eor = !(flags &MSG_SENDPAGE_NOTLAST ); +if (flags &MSG_EOR ) + eor = true;sk_clear_bit (SOCKWQ_ASYNC_NOSPACE , sk);/* Call the sk_stream functions to manage the sndbuf mem. */
@@ -1274,7 +1279,7 @@ static int tls_sw_do_sendpage(struct sock *sk, struct page *page,int tls_sw_sendpage_locked (struct sock *sk,struct page *page,int offset,size_t size,int flags) { -if (flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL | +if (flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL |MSG_EOR |MSG_SENDPAGE_NOTLAST |MSG_SENDPAGE_NOPOLICY |MSG_NO_SHARED_FRAGS ))return -EOPNOTSUPP ;
@@ -1288,7 +1293,7 @@ int tls_sw_sendpage(struct sock *sk, struct page *page,struct tls_context *tls_ctx =tls_get_ctx (sk);int ret; -if (flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL | +if (flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL |MSG_EOR |MSG_SENDPAGE_NOTLAST |MSG_SENDPAGE_NOPOLICY ))return -EOPNOTSUPP ;
Signature · 1 line
2.35.3
PATCH v2 3/4 net/tls: implement ->read_sock()
In reply to Hannes ReineckeImplement ->read_sock() function for use with nvme-tcp.
--- net/tls/tls.h | 2 ++ net/tls/tls_main.c | 2 ++ net/tls/tls_sw.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 75 insertions(+)
net/tls/tls.h +2 -0
@@ -113,6 +113,8 @@ bool tls_sw_sock_is_readable(struct sock *sk);ssize_t tls_sw_splice_read (struct socket *sock,loff_t *ppos,struct pipe_inode_info *pipe,size_t len,unsigned int flags); +int tls_sw_read_sock (struct sock *sk,read_descriptor_t *desc, +sk_read_actor_t read_actor);int tls_device_sendmsg (struct sock *sk,struct msghdr *msg,size_t size); int tls_device_sendpage(struct sock *sk, struct page *page,
net/tls/tls_main.c +2 -0
@@ -922,9 +922,11 @@ static void build_proto_ops(struct proto_ops ops[TLS_NUM_CONFIG][TLS_NUM_CONFIG] ops[TLS_BASE ][TLS_SW ] = ops[TLS_BASE ][TLS_BASE ]; ops[TLS_BASE ][TLS_SW ].splice_read = tls_sw_splice_read; + ops[TLS_BASE ][TLS_SW ].read_sock = tls_sw_read_sock; ops[TLS_SW ][TLS_SW ] = ops[TLS_SW ][TLS_BASE ]; ops[TLS_SW ][TLS_SW ].splice_read = tls_sw_splice_read; + ops[TLS_SW ][TLS_SW ].read_sock = tls_sw_read_sock;#ifdef CONFIG_TLS_DEVICE ops[TLS_HW ][TLS_BASE ] = ops[TLS_BASE ][TLS_BASE ];
net/tls/tls_sw.c +71 -0
77 lines not on this page · raw message
Signature · 1 line
2.35.3
Sabrina Dubroca
maintainerRe: [PATCH 1/4] net/tls: handle MSG_EOR for tls_sw TX flow
In reply to Hannes ReineckeHi Hannes,
Quoted 18 lines · Hannes Reinecke
tls_sw_sendmsg()/tls_do_sw_sendpage()already handlesMSG_MORE/MSG_SENDPAGE_NOTLAST, but bails out onMSG_EOR. But seeing thatMSG_EORis basically the opposite ofMSG_MORE/MSG_SENDPAGE_NOTLASTthis patch adds handlingMSG_EORby treating it as the negation ofMSG_MORE.Cc Jakub Kicinski <[email protected]>Signed-off-by Hannes Reinecke <[email protected]>--- net/tls/tls_sw.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)net/tls/tls_sw.c +0
-0
net/tls/tls_sw.c int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) · line 960
Quoted 9 lines
net/tls/tls_sw.c +3
-1@@ -953,9 +953,12 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)int pending;if (msg->msg_flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL | -MSG_CMSG_COMPAT )) +MSG_EOR |MSG_CMSG_COMPAT ))return -EOPNOTSUPP ; +if (msg->msg_flags &MSG_EOR ) + eor = true;
Is MSG_EOR supposed to be incompatible with MSG_MORE, or is it supposed to cancel it? (ie: MSG_MORE | MSG_EOR is invalid, or MSG_MORE | MSG_EOR behaves like MSG_EOR) The current code already behaves as if _EOR was passed as long as MSG_MORE isn't passed, so _EOR is only needed to cancel out _MORE (or in your case, because NVMe-over-TLS sets it).
If _EOR and _MORE (or MSG_SENDPAGE_NOTLAST below) are supposed to be incompatible, we should return an error when they're both set. If we accept both flags being set at the same time, I think we should document the expected behavior ("_EOR overrides _MORE/_NOTLAST") and add specific selftests to avoid regressions.
Signature · 1 line
Sabrina
Re: [PATCH 1/4] net/tls: handle MSG_EOR for tls_sw TX flow
Quoted 28 lines · Sabrina Dubroca
Hi Hannes,
Quoted 25 lines · Hannes Reinecke
tls_sw_sendmsg()/tls_do_sw_sendpage()already handlesMSG_MORE/MSG_SENDPAGE_NOTLAST, but bails out onMSG_EOR. But seeing thatMSG_EORis basically the opposite ofMSG_MORE/MSG_SENDPAGE_NOTLASTthis patch adds handlingMSG_EORby treating it as the negation ofMSG_MORE.Cc Jakub Kicinski <[email protected]>Signed-off-by Hannes Reinecke <[email protected]>--- net/tls/tls_sw.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)net/tls/tls_sw.c +1
-1@@ -953,9 +953,12 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)int pending;if (msg->msg_flags & ~(MSG_MORE |MSG_DONTWAIT |MSG_NOSIGNAL | -MSG_CMSG_COMPAT )) +MSG_EOR |MSG_CMSG_COMPAT ))return -EOPNOTSUPP ;
Quoted 10 lines
net/tls/tls_sw.c int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) · line 960
Quoted 2 lines
net/tls/tls_sw.c +2
-0+if (msg->msg_flags &MSG_EOR ) + eor = true;Is
MSG_EORsupposed to be incompatible withMSG_MORE, or is it supposed to cancel it? (ie:MSG_MORE|MSG_EORis invalid, orMSG_MORE|MSG_EORbehaves likeMSG_EOR) The current code already behaves as if_EORwas passed as long asMSG_MOREisn't passed, so_EORis only needed to cancel out_MORE(or in your case, because NVMe-over-TLS sets it).
I guess the correct answer here is: Implementation defined. Personally I would veer towards the first option (ie MSG_MORE is incompatible with MSG_EOR), but then it depends on the implementation to check for that. And most don't.
Quoted 6 lines
If
_EORand_MORE(orMSG_SENDPAGE_NOTLASTbelow) are supposed to be incompatible, we should return an error when they're both set. If we accept both flags being set at the same time, I think we should document the expected behavior ("_EORoverrides _MORE/_NOTLAST") and add specific selftests to avoid regressions.
Hmm. Yes, guess you are correct. I'll be adding a check to make MSG_MORE incompatible with MSG_EOR.
Cheers,
Hannes
Signature · 5 lines
Dr. Hannes Reinecke Kernel Storage Architect [email protected] +49 911 74053 688 SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg HRB 36809 (AG Nürnberg), Geschäftsführer: Ivo Totev, Andrew Myers, Andrew McDonald, Martje Boudien Moerman
Re: [PATCH 3/4] net/tls: implement ->read_sock()
In reply to Hannes ReineckeRe: [PATCH 4/4] selftests/net/tls: add test for MSG_EOR
In reply to Hannes Reinecke