linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: David Howells <dhowells@redhat.com>
To: Matthew Wilcox <willy@infradead.org>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>
Cc: David Howells <dhowells@redhat.com>,
	Al Viro <viro@zeniv.linux.org.uk>,
	Christoph Hellwig <hch@infradead.org>,
	Jens Axboe <axboe@kernel.dk>, Jeff Layton <jlayton@kernel.org>,
	Christian Brauner <brauner@kernel.org>,
	Chuck Lever III <chuck.lever@oracle.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	netdev@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	Ayush Sawal <ayush.sawal@chelsio.com>
Subject: [PATCH v3 31/55] chelsio: Support MSG_SPLICE_PAGES
Date: Fri, 31 Mar 2023 17:08:50 +0100	[thread overview]
Message-ID: <20230331160914.1608208-32-dhowells@redhat.com> (raw)
In-Reply-To: <20230331160914.1608208-1-dhowells@redhat.com>

Make Chelsio's TLS offload sendmsg() support MSG_SPLICE_PAGES, splicing in
pages from the source iterator if possible and copying the data in
otherwise.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Ayush Sawal <ayush.sawal@chelsio.com>
cc: "David S. Miller" <davem@davemloft.net>
cc: Eric Dumazet <edumazet@google.com>
cc: Jakub Kicinski <kuba@kernel.org>
cc: Paolo Abeni <pabeni@redhat.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
cc: netdev@vger.kernel.org
---
 .../chelsio/inline_crypto/chtls/chtls_io.c    | 60 ++++++++++++++++++-
 1 file changed, 59 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c
index ae6b17b96bf1..ca3daf5df95c 100644
--- a/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c
+++ b/drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c
@@ -1092,7 +1092,65 @@ int chtls_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
 		if (copy > size)
 			copy = size;
 
-		if (skb_tailroom(skb) > 0) {
+		if (msg->msg_flags & MSG_SPLICE_PAGES) {
+			struct page *page, **pages = &page;
+			ssize_t part;
+			size_t off, spliced = 0;
+			bool put = false;
+			int i;
+
+			do {
+				i = skb_shinfo(skb)->nr_frags;
+				part = iov_iter_extract_pages(&msg->msg_iter, &pages,
+							      copy - spliced, 1, 0, &off);
+				if (part <= 0) {
+					err = part ?: -EIO;
+					goto do_fault;
+				}
+
+				if (!sendpage_ok(page)) {
+					const void *p = kmap_local_page(page);
+					void *q;
+
+					q = page_frag_memdup(NULL, p + off, part,
+							     sk->sk_allocation, ULONG_MAX);
+					kunmap_local(p);
+					if (!q) {
+						iov_iter_revert(&msg->msg_iter, part);
+						return -ENOMEM;
+					}
+					page = virt_to_page(q);
+					off = offset_in_page(q);
+					put = true;
+				}
+
+				if (skb_can_coalesce(skb, i, page, off)) {
+					skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], part);
+					spliced += part;
+					if (put)
+						put_page(page);
+				} else if (i < MAX_SKB_FRAGS) {
+					if (!put)
+						get_page(page);
+					skb_fill_page_desc(skb, i, page, off, spliced);
+					spliced += part;
+					put = false;
+				} else {
+					if (put)
+						put_page(page);
+					if (!spliced)
+						goto new_buf;
+					break;
+				}
+			} while (spliced < copy);
+
+			copy = spliced;
+			skb->len += copy;
+			skb->data_len += copy;
+			skb->truesize += copy;
+			sk->sk_wmem_queued += copy;
+			
+		} else if (skb_tailroom(skb) > 0) {
 			copy = min(copy, skb_tailroom(skb));
 			if (is_tls_tx(csk))
 				copy = min_t(int, copy, csk->tlshws.txleft);



  parent reply	other threads:[~2023-03-31 16:10 UTC|newest]

Thread overview: 80+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-31 16:08 [PATCH v3 00/55] splice, net: Replace sendpage with sendmsg(MSG_SPLICE_PAGES) David Howells
2023-03-31 16:08 ` [PATCH v3 01/55] netfs: Fix netfs_extract_iter_to_sg() for ITER_UBUF/IOVEC David Howells
2023-03-31 19:05   ` Jeff Layton
2023-03-31 16:08 ` [PATCH v3 02/55] iov_iter: Remove last_offset member David Howells
2023-03-31 19:16   ` Jeff Layton
2023-03-31 16:08 ` [PATCH v3 03/55] net: Declare MSG_SPLICE_PAGES internal sendmsg() flag David Howells
2023-04-02 14:56   ` Willem de Bruijn
2023-03-31 16:08 ` [PATCH v3 04/55] mm: Move the page fragment allocator from page_alloc.c into its own file David Howells
2023-03-31 16:08 ` [PATCH v3 05/55] mm: Make the page_frag_cache allocator use multipage folios David Howells
2023-03-31 16:08 ` [PATCH v3 06/55] mm: Make the page_frag_cache allocator use per-cpu David Howells
2023-04-05 15:04   ` Christoph Hellwig
2023-03-31 16:08 ` [PATCH v3 07/55] tcp: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 08/55] tcp: Make sendmsg(MSG_SPLICE_PAGES) copy unspliceable data David Howells
2023-03-31 16:08 ` [PATCH v3 09/55] tcp: Convert do_tcp_sendpages() to use MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 10/55] tcp_bpf: Inline do_tcp_sendpages as it's now a wrapper around tcp_sendmsg David Howells
2023-03-31 16:08 ` [PATCH v3 11/55] espintcp: Inline do_tcp_sendpages() David Howells
2023-03-31 16:08 ` [PATCH v3 12/55] tls: " David Howells
2023-03-31 16:08 ` [PATCH v3 13/55] siw: " David Howells
2023-03-31 16:08 ` [PATCH v3 14/55] tcp: Fold do_tcp_sendpages() into tcp_sendpage_locked() David Howells
2023-03-31 16:08 ` [PATCH v3 15/55] ip, udp: Support MSG_SPLICE_PAGES David Howells
2023-04-02 15:10   ` Willem de Bruijn
2023-04-03  9:50   ` David Howells
2023-04-03 13:46     ` Willem de Bruijn
2023-04-03 22:04     ` David Howells
2023-04-04 16:58       ` Willem de Bruijn
2023-04-04 17:16       ` David Howells
2023-04-04 17:36         ` Willem de Bruijn
2023-04-03 11:18   ` David Howells
2023-03-31 16:08 ` [PATCH v3 16/55] ip, udp: Make sendmsg(MSG_SPLICE_PAGES) copy unspliceable data David Howells
2023-03-31 16:08 ` [PATCH v3 17/55] ip6, udp6: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 18/55] udp: Convert udp_sendpage() to use MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 19/55] af_unix: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 20/55] af_unix: Make sendmsg(MSG_SPLICE_PAGES) copy unspliceable data David Howells
2023-03-31 16:08 ` [PATCH v3 21/55] crypto: af_alg: Pin pages rather than ref'ing if appropriate David Howells
2023-03-31 16:08 ` [PATCH v3 22/55] crypto: af_alg: Use netfs_extract_iter_to_sg() to create scatterlists David Howells
2023-03-31 16:08 ` [PATCH v3 23/55] crypto: af_alg: Indent the loop in af_alg_sendmsg() David Howells
2023-03-31 16:08 ` [PATCH v3 24/55] crypto: af_alg: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 25/55] crypto: af_alg: Convert af_alg_sendpage() to use MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 26/55] crypto: af_alg/hash: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 27/55] tls/device: " David Howells
2023-03-31 16:08 ` [PATCH v3 28/55] tls/device: Convert tls_device_sendpage() to use MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 29/55] tls/sw: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 30/55] tls/sw: Convert tls_sw_sendpage() to use MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` David Howells [this message]
2023-03-31 16:08 ` [PATCH v3 32/55] chelsio: Convert chtls_sendpage() " David Howells
2023-03-31 16:08 ` [PATCH v3 33/55] kcm: Support MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 34/55] kcm: Convert kcm_sendpage() to use MSG_SPLICE_PAGES David Howells
2023-03-31 16:08 ` [PATCH v3 35/55] splice, net: Use sendmsg(MSG_SPLICE_PAGES) rather than ->sendpage() David Howells
2023-03-31 16:08 ` [PATCH v3 36/55] splice, net: Reimplement splice_to_socket() to pass multiple bufs to sendmsg() David Howells
2023-03-31 16:08 ` [PATCH v3 37/55] Remove file->f_op->sendpage David Howells
2023-03-31 16:08 ` [PATCH v3 38/55] siw: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage to transmit David Howells
2023-04-04 10:52   ` Bernard Metzler
2023-04-05  8:18   ` David Howells
2023-03-31 16:08 ` [PATCH v3 39/55] ceph: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage David Howells
2023-03-31 16:08 ` [PATCH v3 40/55] iscsi: " David Howells
2023-03-31 16:09 ` [PATCH v3 41/55] iscsi: Assume "sendpage" is okay in iscsi_tcp_segment_map() David Howells
2023-04-24 17:19   ` Fabio M. De Francesco
2023-04-25  8:30   ` David Howells
2023-04-25 13:13     ` Fabio M. De Francesco
2023-03-31 16:09 ` [PATCH v3 42/55] tcp_bpf: Make tcp_bpf_sendpage() go through tcp_bpf_sendmsg(MSG_SPLICE_PAGES) David Howells
2023-03-31 16:09 ` [PATCH v3 43/55] net: Use sendmsg(MSG_SPLICE_PAGES) not sendpage in skb_send_sock() David Howells
2023-03-31 16:09 ` [PATCH v3 44/55] algif: Remove hash_sendpage*() David Howells
2023-03-31 16:09 ` [PATCH v3 45/55] ceph: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage() David Howells
2023-04-10 12:20   ` Xiubo Li
2023-03-31 16:09 ` [PATCH v3 46/55] rds: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage David Howells
2023-03-31 16:09 ` [PATCH v3 47/55] dlm: " David Howells
2023-03-31 16:09 ` [PATCH v3 48/55] sunrpc: Use sendmsg(MSG_SPLICE_PAGES) rather then sendpage David Howells
2023-03-31 16:09 ` [PATCH v3 49/55] nvme: " David Howells
2023-03-31 16:09 ` [PATCH v3 50/55] kcm: " David Howells
2023-03-31 16:09 ` [PATCH v3 51/55] smc: Drop smc_sendpage() in favour of smc_sendmsg() + MSG_SPLICE_PAGES David Howells
2023-04-26 13:07   ` D. Wythe
2023-04-28  3:12     ` D. Wythe
2023-03-31 16:09 ` [PATCH v3 52/55] ocfs2: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage() David Howells
2023-03-31 16:09 ` [PATCH v3 53/55] drbd: Use sendmsg(MSG_SPLICE_PAGES) rather than sendmsg() David Howells
2023-03-31 16:09 ` [PATCH v3 54/55] drdb: Send an entire bio in a single sendmsg David Howells
2023-03-31 16:27 ` Trivial TLS server David Howells
2023-03-31 16:28 ` Trivial TLS client David Howells
2023-03-31 17:37 ` Test program for AF_KCM David Howells
2023-04-03  9:30 ` [PATCH v3 00/55] splice, net: Replace sendpage with sendmsg(MSG_SPLICE_PAGES) Christoph Hellwig
2023-04-03  9:34 ` Is AF_KCM functional? David Howells

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230331160914.1608208-32-dhowells@redhat.com \
    --to=dhowells@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=ayush.sawal@chelsio.com \
    --cc=brauner@kernel.org \
    --cc=chuck.lever@oracle.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=hch@infradead.org \
    --cc=jlayton@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).