From owner-svn-src-head@FreeBSD.ORG Sat Nov 22 18:03:26 2014 Return-Path: Delivered-To: svn-src-head@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id AF76BEAD; Sat, 22 Nov 2014 18:03:26 +0000 (UTC) Received: from mail-pd0-x231.google.com (mail-pd0-x231.google.com [IPv6:2607:f8b0:400e:c02::231]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 7A1E7F43; Sat, 22 Nov 2014 18:03:26 +0000 (UTC) Received: by mail-pd0-f177.google.com with SMTP id ft15so7226942pdb.22 for ; Sat, 22 Nov 2014 10:03:26 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:from:to:cc:subject:message-id:mail-followup-to:references :mime-version:content-type:content-disposition:in-reply-to :user-agent; bh=1+RyV8J15bT0JHQW1IlyGN7s5UKGqW7tB34CykbNzI4=; b=as4AHSDCO0hdnS4ITMNgLUHWnshhQiIRCRSqHGlbnMcuJi3rVzHw761CqLb6PrDPOe QHzmYnY57nO4Psw+b1G95JgajwnRSS32TzhIkqlAaBRfQMWdqqWZAGan0tbUrsZxI8Gi LVI5YkzMEh4NovD4/jy2AQGgVXwtwZFYU2ModeaqakPqPUnplVNGEeyHdcDU/0Ysxury JdA5SVafKkQpibjuLQRI2nYJx79z4bb5TRLBisVjgVGvkZCRm1KxoacOHy5BVCL5JapX VBHsUJ3cUUsYUBzKu2dTz4yPSPyqg4XlK+3yWJfyGdcr7Zqu9ZrXNuYeQylrY0Kqd72r Hy/Q== X-Received: by 10.68.247.5 with SMTP id ya5mr18038027pbc.42.1416679405982; Sat, 22 Nov 2014 10:03:25 -0800 (PST) Received: from ox ([24.6.44.228]) by mx.google.com with ESMTPSA id ql6sm7985450pbb.39.2014.11.22.10.03.23 for (version=TLSv1.2 cipher=RC4-SHA bits=128/128); Sat, 22 Nov 2014 10:03:24 -0800 (PST) Date: Sat, 22 Nov 2014 10:03:18 -0800 From: Navdeep Parhar To: Alexander Motin Subject: Re: svn commit: r274853 - in head: sys/dev/iscsi usr.sbin/ctld usr.sbin/iscsid Message-ID: <20141122180318.GA3921@ox> Mail-Followup-To: Alexander Motin , src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org References: <201411221509.sAMF9JIh016800@svn.freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201411221509.sAMF9JIh016800@svn.freebsd.org> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org X-BeenThere: svn-src-head@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: SVN commit messages for the src tree for head/-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 22 Nov 2014 18:03:26 -0000 On Sat, Nov 22, 2014 at 03:09:19PM +0000, Alexander Motin wrote: > Author: mav > Date: Sat Nov 22 15:09:18 2014 > New Revision: 274853 > URL: https://svnweb.freebsd.org/changeset/base/274853 > > Log: > For both iSCSI initiator and target increase socket buffer sizes before > establishing connection. > > This is a workaround for Chelsio TOE driver, that does not update socket > buffer size in hardware after connection established, and unless that is > done beforehand, kernel code will stuck, attempting to send/receive full > PDU at once. cxgbe's TOE driver does update the rcv window in the hardware in response to changes in the socket buffer size (see t4_rcvd in cxgbe/tom/t4_cpl_io.c). The bug is that the driver tries to avoid sending too many small updates to the chip, and instead tries to build up a balance of credits before an update. This doesn't interact well with some applications that set the sockbuf size to a high fixed value and then wait for the entire buffer to fill up before draining it. The driver ends up waiting for the application to drain the socket's receive buffer so that pru_rcvd (t4_rcvd) can then send a big batch of credits to the chip, the application is waiting for the receive to reach its high water mark before draining the buffer, but the TOE on the chip has shut down the receive window because it's out of credits. If you see the peer sending zero window probes and the TOE refusing to budge, then the problem you've observed is definitely the problem I just described. I'll fix this in the TOE driver. I can send you an early patch in case you'd like to try out the fix before it gets committed. Regards, Navdeep > > MFC after: 1 week > > Modified: > head/sys/dev/iscsi/icl.c > head/usr.sbin/ctld/ctld.c > head/usr.sbin/ctld/ctld.h > head/usr.sbin/iscsid/iscsid.c > head/usr.sbin/iscsid/iscsid.h > > Modified: head/sys/dev/iscsi/icl.c > ============================================================================== > --- head/sys/dev/iscsi/icl.c Sat Nov 22 14:59:50 2014 (r274852) > +++ head/sys/dev/iscsi/icl.c Sat Nov 22 15:09:18 2014 (r274853) > @@ -1199,6 +1199,8 @@ icl_conn_start(struct icl_conn *ic) > icl_conn_close(ic); > return (error); > } > + ic->ic_socket->so_snd.sb_flags |= SB_AUTOSIZE; > + ic->ic_socket->so_rcv.sb_flags |= SB_AUTOSIZE; > > /* > * Disable Nagle. > > Modified: head/usr.sbin/ctld/ctld.c > ============================================================================== > --- head/usr.sbin/ctld/ctld.c Sat Nov 22 14:59:50 2014 (r274852) > +++ head/usr.sbin/ctld/ctld.c Sat Nov 22 15:09:18 2014 (r274853) > @@ -1602,7 +1602,7 @@ conf_apply(struct conf *oldconf, struct > struct portal *oldp, *newp; > struct isns *oldns, *newns; > pid_t otherpid; > - int changed, cumulated_error = 0, error; > + int changed, cumulated_error = 0, error, sockbuf; > int one = 1; > > if (oldconf->conf_debug != newconf->conf_debug) { > @@ -1899,6 +1899,16 @@ conf_apply(struct conf *oldconf, struct > cumulated_error++; > continue; > } > + sockbuf = SOCKBUF_SIZE; > + if (setsockopt(newp->p_socket, SOL_SOCKET, SO_RCVBUF, > + &sockbuf, sizeof(sockbuf)) == -1) > + log_warn("setsockopt(SO_RCVBUF) failed " > + "for %s", newp->p_listen); > + sockbuf = SOCKBUF_SIZE; > + if (setsockopt(newp->p_socket, SOL_SOCKET, SO_SNDBUF, > + &sockbuf, sizeof(sockbuf)) == -1) > + log_warn("setsockopt(SO_SNDBUF) failed " > + "for %s", newp->p_listen); > error = setsockopt(newp->p_socket, SOL_SOCKET, > SO_REUSEADDR, &one, sizeof(one)); > if (error != 0) { > > Modified: head/usr.sbin/ctld/ctld.h > ============================================================================== > --- head/usr.sbin/ctld/ctld.h Sat Nov 22 14:59:50 2014 (r274852) > +++ head/usr.sbin/ctld/ctld.h Sat Nov 22 15:09:18 2014 (r274853) > @@ -48,6 +48,7 @@ > #define MAX_NAME_LEN 223 > #define MAX_DATA_SEGMENT_LENGTH (128 * 1024) > #define MAX_BURST_LENGTH 16776192 > +#define SOCKBUF_SIZE 1048576 > > struct auth { > TAILQ_ENTRY(auth) a_next; > > Modified: head/usr.sbin/iscsid/iscsid.c > ============================================================================== > --- head/usr.sbin/iscsid/iscsid.c Sat Nov 22 14:59:50 2014 (r274852) > +++ head/usr.sbin/iscsid/iscsid.c Sat Nov 22 15:09:18 2014 (r274853) > @@ -160,7 +160,7 @@ connection_new(unsigned int session_id, > #ifdef ICL_KERNEL_PROXY > struct iscsi_daemon_connect idc; > #endif > - int error; > + int error, sockbuf; > > conn = calloc(1, sizeof(*conn)); > if (conn == NULL) > @@ -237,6 +237,14 @@ connection_new(unsigned int session_id, > fail(conn, strerror(errno)); > log_err(1, "failed to create socket for %s", from_addr); > } > + sockbuf = SOCKBUF_SIZE; > + if (setsockopt(conn->conn_socket, SOL_SOCKET, SO_RCVBUF, > + &sockbuf, sizeof(sockbuf)) == -1) > + log_warn("setsockopt(SO_RCVBUF) failed"); > + sockbuf = SOCKBUF_SIZE; > + if (setsockopt(conn->conn_socket, SOL_SOCKET, SO_SNDBUF, > + &sockbuf, sizeof(sockbuf)) == -1) > + log_warn("setsockopt(SO_SNDBUF) failed"); > if (from_ai != NULL) { > error = bind(conn->conn_socket, from_ai->ai_addr, > from_ai->ai_addrlen); > > Modified: head/usr.sbin/iscsid/iscsid.h > ============================================================================== > --- head/usr.sbin/iscsid/iscsid.h Sat Nov 22 14:59:50 2014 (r274852) > +++ head/usr.sbin/iscsid/iscsid.h Sat Nov 22 15:09:18 2014 (r274853) > @@ -44,6 +44,7 @@ > #define CONN_DIGEST_CRC32C 1 > > #define CONN_MUTUAL_CHALLENGE_LEN 1024 > +#define SOCKBUF_SIZE 1048576 > > struct connection { > int conn_iscsi_fd; >