linux-kernel - Re: [PATCH v3 38/55] siw: Use sendmsg(MSG_SPLICE

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <2904969.1680682716@warthog.procyon.org.uk>
Date:   Wed, 05 Apr 2023 09:18:36 +0100
From:   David Howells <dhowells@...hat.com>
To:     Bernard Metzler <BMT@...ich.ibm.com>
Cc:     dhowells@...hat.com, "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Al Viro <viro@...iv.linux.org.uk>,
        Christoph Hellwig <hch@...radead.org>,
        Jens Axboe <axboe@...nel.dk>, Jeff Layton <jlayton@...nel.org>,
        Christian Brauner <brauner@...nel.org>,
        Chuck Lever III <chuck.lever@...cle.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-mm@...ck.org" <linux-mm@...ck.org>,
        Tom Talpey <tom@...pey.com>,
        "linux-rdma@...r.kernel.org" <linux-rdma@...r.kernel.org>
Subject: Re: [PATCH v3 38/55] siw: Use sendmsg(MSG_SPLICE_PAGES) rather than sendpage to transmit

Bernard Metzler <BMT@...ich.ibm.com> wrote:

> >  	if (c_tx->state == SIW_SEND_HDR) {
> >  		if (c_tx->use_sendpage) {
> > @@ -457,10 +350,15 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx,
> > struct socket *s)
> > 
> 
> Couldn't we now collapse the two header handling paths
> into one, avoiding extra 
> 'if (c_tx->use_sendpage) {} else {}' conditions?

Okay, see the attached incremental change.

Note that the calls to page_frag_memdup() I previously added are probably not
going to be necessary as copying unspliceable data is now done in the
protocols (TCP, IP/UDP, UNIX, etc.).  See patch 08 for the TCP version.

David
---
diff --git a/drivers/infiniband/sw/siw/siw_qp_tx.c b/drivers/infiniband/sw/siw/siw_qp_tx.c
index 28076832da20..edf66a97cf5f 100644
--- a/drivers/infiniband/sw/siw/siw_qp_tx.c
+++ b/drivers/infiniband/sw/siw/siw_qp_tx.c
@@ -335,7 +335,7 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
 	struct siw_sge *sge = &wqe->sqe.sge[c_tx->sge_idx];
 	struct bio_vec bvec[MAX_ARRAY];
 	struct msghdr msg = { .msg_flags = MSG_DONTWAIT | MSG_EOR };
-	void *trl, *t;
+	void *trl;
 
 	int seg = 0, do_crc = c_tx->do_crc, is_kva = 0, rv;
 	unsigned int data_len = c_tx->bytes_unsent, hdr_len = 0, trl_len = 0,
@@ -343,25 +343,11 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
 		     pbl_idx = c_tx->pbl_idx;
 
 	if (c_tx->state == SIW_SEND_HDR) {
-		if (c_tx->use_sendpage) {
-			rv = siw_tx_ctrl(c_tx, s, MSG_DONTWAIT | MSG_MORE);
-			if (rv)
-				goto done;
+		void *hdr = &c_tx->pkt.ctrl + c_tx->ctrl_sent;
 
-			c_tx->state = SIW_SEND_DATA;
-		} else {
-			const void *hdr = &c_tx->pkt.ctrl + c_tx->ctrl_sent;
-			void *h;
-
-			rv = -ENOMEM;
-			hdr_len = c_tx->ctrl_len - c_tx->ctrl_sent;
-			h = page_frag_memdup(NULL, hdr, hdr_len, GFP_NOFS,
-					     ULONG_MAX);
-			if (!h)
-				goto done;
-			bvec_set_virt(&bvec[0], h, hdr_len);
-			seg = 1;
-		}
+		hdr_len = c_tx->ctrl_len - c_tx->ctrl_sent;
+		bvec_set_virt(&bvec[0], hdr, hdr_len);
+		seg = 1;
 	}
 
 	wqe->processed += data_len;
@@ -466,12 +452,7 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
 		trl = &c_tx->trailer.pad[c_tx->ctrl_sent];
 		trl_len = MAX_TRAILER - c_tx->ctrl_sent;
 	}
-
-	rv = -ENOMEM;
-	t = page_frag_memdup(NULL, trl, trl_len, GFP_NOFS, ULONG_MAX);
-	if (!t)
-		goto done_crc;
-	bvec_set_virt(&bvec[seg], t, trl_len);
+	bvec_set_virt(&bvec[seg], trl, trl_len);
 
 	data_len = c_tx->bytes_unsent;
 
@@ -480,7 +461,6 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
 	iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, bvec, seg + 1,
 		      hdr_len + data_len + trl_len);
 	rv = sock_sendmsg(s, &msg);
-
 	if (rv < (int)hdr_len) {
 		/* Not even complete hdr pushed or negative rv */
 		wqe->processed -= data_len;
@@ -541,10 +521,6 @@ static int siw_tx_hdt(struct siw_iwarp_tx *c_tx, struct socket *s)
 	}
 done_crc:
 	c_tx->do_crc = 0;
-	if (c_tx->state == SIW_SEND_HDR)
-		folio_put(page_folio(bvec[0].bv_page));
-	folio_put(page_folio(bvec[seg].bv_page));
-done:
 	return rv;
 }