From owner-p4-projects@FreeBSD.ORG Tue Apr 24 04:20:56 2007 Return-Path: X-Original-To: p4-projects@freebsd.org Delivered-To: p4-projects@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 32767) id F40BD16A406; Tue, 24 Apr 2007 04:20:55 +0000 (UTC) X-Original-To: perforce@freebsd.org Delivered-To: perforce@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id B335E16A404 for ; Tue, 24 Apr 2007 04:20:55 +0000 (UTC) (envelope-from kmacy@freebsd.org) Received: from repoman.freebsd.org (repoman.freebsd.org [69.147.83.41]) by mx1.freebsd.org (Postfix) with ESMTP id A369813C469 for ; Tue, 24 Apr 2007 04:20:55 +0000 (UTC) (envelope-from kmacy@freebsd.org) Received: from repoman.freebsd.org (localhost [127.0.0.1]) by repoman.freebsd.org (8.13.8/8.13.8) with ESMTP id l3O4Kt6K043618 for ; Tue, 24 Apr 2007 04:20:55 GMT (envelope-from kmacy@freebsd.org) Received: (from perforce@localhost) by repoman.freebsd.org (8.13.8/8.13.8/Submit) id l3O4KtU1043613 for perforce@freebsd.org; Tue, 24 Apr 2007 04:20:55 GMT (envelope-from kmacy@freebsd.org) Date: Tue, 24 Apr 2007 04:20:55 GMT Message-Id: <200704240420.l3O4KtU1043613@repoman.freebsd.org> X-Authentication-Warning: repoman.freebsd.org: perforce set sender to kmacy@freebsd.org using -f From: Kip Macy To: Perforce Change Reviews Cc: Subject: PERFORCE change 118700 for review X-BeenThere: p4-projects@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: p4 projects tree changes List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 24 Apr 2007 04:20:56 -0000 http://perforce.freebsd.org/chv.cgi?CH=118700 Change 118700 by kmacy@kmacy_vt-x:opentoe_init on 2007/04/24 04:20:10 remove bind_ithread hack now that uleintr is in offload_enqueue is chaining packets not mbufs rss_{hi,lo} have more intuitive names in cxgb Affected files ... .. //depot/projects/opentoe/sys/dev/cxgb/cxgb_sge.c#12 edit Differences ... ==== //depot/projects/opentoe/sys/dev/cxgb/cxgb_sge.c#12 (text+ko) ==== @@ -2031,7 +2031,7 @@ { m->m_next = NULL; if (q->rx_tail) - q->rx_tail->m_next = m; + q->rx_tail->m_nextpkt = m; else { #ifdef notyet struct sge_qset *qs = rspq_to_qset(q); @@ -2251,23 +2251,6 @@ ; } -/* - * This is an awful hack to bind the ithread to CPU 1 - * to work around lack of ithread affinity - */ -static void -bind_ithread(int cpu) -{ -#if 0 - KASSERT(cpu < mp_ncpus, ("invalid cpu identifier")); - if (mp_ncpus > 1) { - mtx_lock_spin(&sched_lock); - sched_bind(curthread, cpu); - mtx_unlock_spin(&sched_lock); - } -#endif -} - /** * process_responses - process responses from an SGE response queue * @adap: the adapter @@ -2300,14 +2283,7 @@ printf("next_holdoff=%d\n", rspq->holdoff_tmr); last_holdoff = rspq->holdoff_tmr; } -#endif - if (pinned[qs->rspq.cntxt_id * adap->params.nports] == 0) { - /* - * Assumes that cntxt_id < mp_ncpus - */ - bind_ithread(qs->rspq.cntxt_id); - pinned[qs->rspq.cntxt_id * adap->params.nports] = 1; - } +#endif rspq->next_holdoff = rspq->holdoff_tmr; while (__predict_true(budget_left && is_new_response(r, rspq))) { @@ -2385,8 +2361,11 @@ rspq->m = NULL; } else { #ifdef notyet - m->pkthdr.csum_data = rss_hi; - m->m_priority = rss_lo; + m->pkthdr.csum_data = rss_csum; + /* + * XXX size mismatch + */ + m->m_priority = rss_hash; rx_offload(&adap->tdev, rspq, m); #endif