From owner-svn-src-all@freebsd.org Sun Jul 24 14:42:13 2016 Return-Path: Delivered-To: svn-src-all@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 01CF9BA3E74; Sun, 24 Jul 2016 14:42:13 +0000 (UTC) (envelope-from mav@FreeBSD.org) Received: from repo.freebsd.org (repo.freebsd.org [IPv6:2610:1c1:1:6068::e6a:0]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id C85821A44; Sun, 24 Jul 2016 14:42:12 +0000 (UTC) (envelope-from mav@FreeBSD.org) Received: from repo.freebsd.org ([127.0.1.37]) by repo.freebsd.org (8.15.2/8.15.2) with ESMTP id u6OEgCrD017988; Sun, 24 Jul 2016 14:42:12 GMT (envelope-from mav@FreeBSD.org) Received: (from mav@localhost) by repo.freebsd.org (8.15.2/8.15.2/Submit) id u6OEgCQb017987; Sun, 24 Jul 2016 14:42:12 GMT (envelope-from mav@FreeBSD.org) Message-Id: <201607241442.u6OEgCQb017987@repo.freebsd.org> X-Authentication-Warning: repo.freebsd.org: mav set sender to mav@FreeBSD.org using -f From: Alexander Motin Date: Sun, 24 Jul 2016 14:42:12 +0000 (UTC) To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r303266 - head/sys/dev/ntb/ntb_hw X-SVN-Group: head MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Jul 2016 14:42:13 -0000 Author: mav Date: Sun Jul 24 14:42:11 2016 New Revision: 303266 URL: https://svnweb.freebsd.org/changeset/base/303266 Log: Postpone ntb_get_msix_info() till we need to negotiate MSIX. Calling it earlier increases the window when MSIX info may change. This change does not solve the problem completely, but seems logical. Complete solution should probably include link reset in case of MSIX remap to trigger new negotiation, but we have no way to get notified about that now. Modified: head/sys/dev/ntb/ntb_hw/ntb_hw.c Modified: head/sys/dev/ntb/ntb_hw/ntb_hw.c ============================================================================== --- head/sys/dev/ntb/ntb_hw/ntb_hw.c Sun Jul 24 11:05:23 2016 (r303265) +++ head/sys/dev/ntb/ntb_hw/ntb_hw.c Sun Jul 24 14:42:11 2016 (r303266) @@ -1084,8 +1084,6 @@ ntb_init_isr(struct ntb_softc *ntb) ntb_create_msix_vec(ntb, num_vectors); rc = ntb_setup_msix(ntb, num_vectors); - if (rc == 0 && HAS_FEATURE(ntb, NTB_SB01BASE_LOCKUP)) - ntb_get_msix_info(ntb); } if (rc != 0) { device_printf(ntb->device, @@ -2715,6 +2713,7 @@ ntb_exchange_msix(void *ctx) if (ntb->peer_msix_done) goto msix_done; + ntb_get_msix_info(ntb); for (i = 0; i < XEON_NONLINK_DB_MSIX_BITS; i++) { ntb_peer_spad_write(ntb->device, NTB_MSIX_DATA0 + i, ntb->msix_data[i].nmd_data);