From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from pdx1-mailman-customer002.dreamhost.com (listserver-buz.dreamhost.com [69.163.136.29]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id E9663C25B50 for ; Mon, 23 Jan 2023 23:03:31 +0000 (UTC) Received: from pdx1-mailman-customer002.dreamhost.com (localhost [127.0.0.1]) by pdx1-mailman-customer002.dreamhost.com (Postfix) with ESMTP id 4P15Fb4MtSz1y6Q; Mon, 23 Jan 2023 15:01:39 -0800 (PST) Received: from smtp3.ccs.ornl.gov (smtp3.ccs.ornl.gov [160.91.203.39]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by pdx1-mailman-customer002.dreamhost.com (Postfix) with ESMTPS id 4P15Dx63b4z1wSM for ; Mon, 23 Jan 2023 15:01:05 -0800 (PST) Received: from star.ccs.ornl.gov (star.ccs.ornl.gov [160.91.202.134]) by smtp3.ccs.ornl.gov (Postfix) with ESMTP id 202426C6; Mon, 23 Jan 2023 18:00:58 -0500 (EST) Received: by star.ccs.ornl.gov (Postfix, from userid 2004) id 1685158991; Mon, 23 Jan 2023 18:00:58 -0500 (EST) From: James Simmons To: Andreas Dilger , Oleg Drokin , NeilBrown Date: Mon, 23 Jan 2023 18:00:15 -0500 Message-Id: <1674514855-15399-3-git-send-email-jsimmons@infradead.org> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1674514855-15399-1-git-send-email-jsimmons@infradead.org> References: <1674514855-15399-1-git-send-email-jsimmons@infradead.org> Subject: [lustre-devel] [PATCH 02/42] lnet: lnet_peer_merge_data to understand large addr X-BeenThere: lustre-devel@lists.lustre.org X-Mailman-Version: 2.1.39 Precedence: list List-Id: "For discussing Lustre software development." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Lustre Development List MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: lustre-devel-bounces@lists.lustre.org Sender: "lustre-devel" From: Mr NeilBrown Large addr now understood. WC-bug-id: https://jira.whamcloud.com/browse/LU-10391 Lustre-commit: 57bcd6aa7f5f347de ("LU-10391 lnet: lnet_peer_merge_data to understand large addr") Signed-off-by: Mr NeilBrown Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/44630 Reviewed-by: Frank Sehr Reviewed-by: James Simmons Reviewed-by: Chris Horn Reviewed-by: Oleg Drokin Signed-off-by: James Simmons --- net/lnet/lnet/peer.c | 84 ++++++++++++++++++++++++++++++-------------- 1 file changed, 57 insertions(+), 27 deletions(-) diff --git a/net/lnet/lnet/peer.c b/net/lnet/lnet/peer.c index 8c603c903521..a9759860a5cd 100644 --- a/net/lnet/lnet/peer.c +++ b/net/lnet/lnet/peer.c @@ -2917,6 +2917,8 @@ u32 *ping_iter_first(struct lnet_ping_iter *pi, */ if (nid) lnet_nid4_to_nid(pbuf->pb_info.pi_ni[0].ns_nid, nid); + + pi->pos += sizeof(struct lnet_ni_status); return &pbuf->pb_info.pi_ni[0].ns_status; } @@ -2953,6 +2955,19 @@ u32 *ping_iter_next(struct lnet_ping_iter *pi, struct lnet_nid *nid) return NULL; } +static int ping_info_count_entries(struct lnet_ping_buffer *pbuf) +{ + struct lnet_ping_iter pi; + u32 *st; + int nnis = 0; + + for (st = ping_iter_first(&pi, pbuf, NULL); st; + st = ping_iter_next(&pi, NULL)) + nnis += 1; + + return nnis; +} + /* * Build a peer from incoming data. * @@ -2976,10 +2991,14 @@ static int lnet_peer_merge_data(struct lnet_peer *lp, { struct lnet_peer_net *lpn; struct lnet_peer_ni *lpni; - lnet_nid_t *curnis = NULL; - struct lnet_ni_status *addnis = NULL; - lnet_nid_t *delnis = NULL; + struct lnet_nid *curnis = NULL; + struct lnet_ni_large_status *addnis = NULL; + struct lnet_nid *delnis = NULL; + struct lnet_ping_iter pi; struct lnet_nid nid; + u32 *stp; + struct lnet_nid primary = {}; + bool want_large_primary; unsigned int flags; int ncurnis; int naddnis; @@ -3003,7 +3022,8 @@ static int lnet_peer_merge_data(struct lnet_peer *lp, lp->lp_state &= ~LNET_PEER_ROUTER_ENABLED; spin_unlock(&lp->lp_lock); - nnis = max_t(int, lp->lp_nnis, pbuf->pb_info.pi_nnis); + nnis = ping_info_count_entries(pbuf); + nnis = max_t(int, lp->lp_nnis, nnis); curnis = kmalloc_array(nnis, sizeof(*curnis), GFP_NOFS); addnis = kmalloc_array(nnis, sizeof(*addnis), GFP_NOFS); delnis = kmalloc_array(nnis, sizeof(*delnis), GFP_NOFS); @@ -3018,18 +3038,31 @@ static int lnet_peer_merge_data(struct lnet_peer *lp, /* Construct the list of NIDs present in peer. */ lpni = NULL; while ((lpni = lnet_get_next_peer_ni_locked(lp, NULL, lpni)) != NULL) - curnis[ncurnis++] = lnet_nid_to_nid4(&lpni->lpni_nid); + curnis[ncurnis++] = lpni->lpni_nid; - /* - * Check for NIDs in pbuf not present in curnis[]. - * The loop starts at 1 to skip the loopback NID. + /* Check for NIDs in pbuf not present in curnis[]. + * Skip the first, which is loop-back. Take second as + * primary, unless a large primary is found. */ - for (i = 1; i < pbuf->pb_info.pi_nnis; i++) { + ping_iter_first(&pi, pbuf, NULL); + stp = ping_iter_next(&pi, &nid); + if (stp) + primary = nid; + want_large_primary = (pbuf->pb_info.pi_features & + LNET_PING_FEAT_PRIMARY_LARGE); + for (; stp; stp = ping_iter_next(&pi, &nid)) { for (j = 0; j < ncurnis; j++) - if (pbuf->pb_info.pi_ni[i].ns_nid == curnis[j]) + if (nid_same(&nid, &curnis[j])) break; - if (j == ncurnis) - addnis[naddnis++] = pbuf->pb_info.pi_ni[i]; + if (j == ncurnis) { + addnis[naddnis].ns_nid = nid; + addnis[naddnis].ns_status = *stp; + naddnis += 1; + } + if (want_large_primary && nid.nid_size) { + primary = nid; + want_large_primary = false; + } } /* * Check for NIDs in curnis[] not present in pbuf. @@ -3039,25 +3072,24 @@ static int lnet_peer_merge_data(struct lnet_peer *lp, * present in curnis[] then this peer is for this node. */ for (i = 0; i < ncurnis; i++) { - if (curnis[i] == LNET_NID_LO_0) + if (nid_is_lo0(&curnis[i])) continue; - for (j = 1; j < pbuf->pb_info.pi_nnis; j++) { - if (curnis[i] == pbuf->pb_info.pi_ni[j].ns_nid) { + ping_iter_first(&pi, pbuf, NULL); + while ((stp = ping_iter_next(&pi, &nid)) != NULL) { + if (nid_same(&curnis[i], &nid)) { /* update the information we cache for the * peer with the latest information we * received */ - lnet_nid4_to_nid(curnis[i], &nid); - lpni = lnet_peer_ni_find_locked(&nid); + lpni = lnet_peer_ni_find_locked(&curnis[i]); if (lpni) { - lpni->lpni_ns_status = - pbuf->pb_info.pi_ni[j].ns_status; + lpni->lpni_ns_status = *stp; lnet_peer_ni_decref_locked(lpni); } break; } } - if (j == pbuf->pb_info.pi_nnis) + if (!stp) delnis[ndelnis++] = curnis[i]; } @@ -3070,16 +3102,15 @@ static int lnet_peer_merge_data(struct lnet_peer *lp, goto out; for (i = 0; i < naddnis; i++) { - lnet_nid4_to_nid(addnis[i].ns_nid, &nid); - rc = lnet_peer_add_nid(lp, &nid, flags); + rc = lnet_peer_add_nid(lp, &addnis[i].ns_nid, flags); if (rc) { CERROR("Error adding NID %s to peer %s: %d\n", - libcfs_nid2str(addnis[i].ns_nid), + libcfs_nidstr(&addnis[i].ns_nid), libcfs_nidstr(&lp->lp_primary_nid), rc); if (rc == -ENOMEM) goto out; } - lpni = lnet_peer_ni_find_locked(&nid); + lpni = lnet_peer_ni_find_locked(&addnis[i].ns_nid); if (lpni) { lpni->lpni_ns_status = addnis[i].ns_status; lnet_peer_ni_decref_locked(lpni); @@ -3092,13 +3123,12 @@ static int lnet_peer_merge_data(struct lnet_peer *lp, * being told that the router changed its primary_nid * then it's okay to delete it. */ - lnet_nid4_to_nid(delnis[i], &nid); if (lp->lp_rtr_refcount > 0) flags |= LNET_PEER_RTR_NI_FORCE_DEL; - rc = lnet_peer_del_nid(lp, &nid, flags); + rc = lnet_peer_del_nid(lp, &delnis[i], flags); if (rc) { CERROR("Error deleting NID %s from peer %s: %d\n", - libcfs_nid2str(delnis[i]), + libcfs_nidstr(&delnis[i]), libcfs_nidstr(&lp->lp_primary_nid), rc); if (rc == -ENOMEM) goto out; -- 2.27.0 _______________________________________________ lustre-devel mailing list lustre-devel@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-devel-lustre.org