From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id B937732A5 for ; Tue, 20 Nov 2018 20:14:32 +0100 (CET) Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 2BFB6308624F; Tue, 20 Nov 2018 19:14:32 +0000 (UTC) Received: from ktraynor.remote.csb (unknown [10.36.118.7]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4DE3C60141; Tue, 20 Nov 2018 19:14:31 +0000 (UTC) From: Kevin Traynor To: Stephen Hemminger Cc: dpdk stable Date: Tue, 20 Nov 2018 19:11:59 +0000 Message-Id: <20181120191252.30277-9-ktraynor@redhat.com> In-Reply-To: <20181120191252.30277-1-ktraynor@redhat.com> References: <20181120191252.30277-1-ktraynor@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Scanned-By: MIMEDefang 2.79 on 10.5.11.11 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.49]); Tue, 20 Nov 2018 19:14:32 +0000 (UTC) Subject: [dpdk-stable] patch 'net/netvsc: resize event buffer as needed' has been queued to stable release 18.08.1 X-BeenThere: stable@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches for DPDK stable branches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 20 Nov 2018 19:14:33 -0000 Hi, FYI, your patch has been queued to stable release 18.08.1 Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet. It will be pushed if I get no objections before 11/23/18. So please shout if anyone has objections. Also note that after the patch there's a diff of the upstream commit vs the patch applied to the branch. If the code is different (ie: not only metadata diffs), due for example to a change in context or macro names, please double check it. Thanks. Kevin Traynor --- >>From 9b56b30a66a8005f3eba80dd953d2d388a88fa44 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 14 Aug 2018 09:45:25 -0700 Subject: [PATCH] net/netvsc: resize event buffer as needed [ upstream commit 1f2766b7ee02c434b554513193c2a2be6664601d ] The event buffer was changed to be a fixed size value, had a couple of issues. The big one is that rte_free was still being called for a pointer that was not setup with rte_malloc(). The event buffer was also too small to handle heavy receive traffic; and running the event buffer out would crash the application. Fix by going back to a dynamically resized event buffer. And grow it by 25% to avoid lots of realloc's. Fixes: 530af95a7849 ("bus/vmbus: avoid signalling host on read") Signed-off-by: Stephen Hemminger --- drivers/net/netvsc/hn_rxtx.c | 50 ++++++++++++++++++++++++++---------- drivers/net/netvsc/hn_var.h | 2 +- 2 files changed, 38 insertions(+), 14 deletions(-) diff --git a/drivers/net/netvsc/hn_rxtx.c b/drivers/net/netvsc/hn_rxtx.c index 02ef27e36..9b394d261 100644 --- a/drivers/net/netvsc/hn_rxtx.c +++ b/drivers/net/netvsc/hn_rxtx.c @@ -11,4 +11,5 @@ #include #include +#include #include @@ -719,14 +720,22 @@ struct hn_rx_queue *hn_rx_queue_alloc(struct hn_data *hv, struct hn_rx_queue *rxq; - rxq = rte_zmalloc_socket("HN_RXQ", - sizeof(*rxq) + HN_RXQ_EVENT_DEFAULT, + rxq = rte_zmalloc_socket("HN_RXQ", sizeof(*rxq), RTE_CACHE_LINE_SIZE, socket_id); - if (rxq) { - rxq->hv = hv; - rxq->chan = hv->channels[queue_id]; - rte_spinlock_init(&rxq->ring_lock); - rxq->port_id = hv->port_id; - rxq->queue_id = queue_id; + if (!rxq) + return NULL; + + rxq->hv = hv; + rxq->chan = hv->channels[queue_id]; + rte_spinlock_init(&rxq->ring_lock); + rxq->port_id = hv->port_id; + rxq->queue_id = queue_id; + rxq->event_sz = HN_RXQ_EVENT_DEFAULT; + rxq->event_buf = rte_malloc_socket("HN_EVENTS", HN_RXQ_EVENT_DEFAULT, + RTE_CACHE_LINE_SIZE, socket_id); + if (!rxq->event_buf) { + rte_free(rxq); + return NULL; } + return rxq; } @@ -854,17 +863,32 @@ void hn_process_events(struct hn_data *hv, uint16_t queue_id) for (;;) { const struct vmbus_chanpkt_hdr *pkt; - uint32_t len = HN_RXQ_EVENT_DEFAULT; + uint32_t len = rxq->event_sz; const void *data; +retry: ret = rte_vmbus_chan_recv_raw(rxq->chan, rxq->event_buf, &len); if (ret == -EAGAIN) break; /* ring is empty */ - else if (ret == -ENOBUFS) - rte_exit(EXIT_FAILURE, "event buffer not big enough (%u < %u)", - HN_RXQ_EVENT_DEFAULT, len); - else if (ret <= 0) + if (unlikely(ret == -ENOBUFS)) { + /* event buffer not large enough to read ring */ + + PMD_DRV_LOG(DEBUG, + "event buffer expansion (need %u)", len); + rxq->event_sz = len + len / 4; + rxq->event_buf = rte_realloc(rxq->event_buf, rxq->event_sz, + RTE_CACHE_LINE_SIZE); + if (rxq->event_buf) + goto retry; + /* out of memory, no more events now */ + rxq->event_sz = 0; + break; + } + + if (unlikely(ret <= 0)) { + /* This indicates a failure to communicate (or worse) */ rte_exit(EXIT_FAILURE, "vmbus ring buffer error: %d", ret); + } bytes_read += ret; diff --git a/drivers/net/netvsc/hn_var.h b/drivers/net/netvsc/hn_var.h index f7ff8585b..0430f450c 100644 --- a/drivers/net/netvsc/hn_var.h +++ b/drivers/net/netvsc/hn_var.h @@ -78,5 +78,5 @@ struct hn_rx_queue { uint64_t ring_full; - uint8_t event_buf[]; + void *event_buf; }; -- 2.19.0 --- Diff of the applied patch vs upstream commit (please double-check if non-empty: --- --- - 2018-11-20 17:53:07.687552366 +0000 +++ 0009-net-netvsc-resize-event-buffer-as-needed.patch 2018-11-20 17:53:07.000000000 +0000 @@ -1,8 +1,10 @@ -From 1f2766b7ee02c434b554513193c2a2be6664601d Mon Sep 17 00:00:00 2001 +From 9b56b30a66a8005f3eba80dd953d2d388a88fa44 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 14 Aug 2018 09:45:25 -0700 Subject: [PATCH] net/netvsc: resize event buffer as needed +[ upstream commit 1f2766b7ee02c434b554513193c2a2be6664601d ] + The event buffer was changed to be a fixed size value, had a couple of issues. The big one is that rte_free was still being called for a pointer that was not setup with rte_malloc(). @@ -15,7 +17,6 @@ And grow it by 25% to avoid lots of realloc's. Fixes: 530af95a7849 ("bus/vmbus: avoid signalling host on read") -Cc: stable@dpdk.org Signed-off-by: Stephen Hemminger --- @@ -24,7 +25,7 @@ 2 files changed, 38 insertions(+), 14 deletions(-) diff --git a/drivers/net/netvsc/hn_rxtx.c b/drivers/net/netvsc/hn_rxtx.c -index 24abc2a91..cb5bc6029 100644 +index 02ef27e36..9b394d261 100644 --- a/drivers/net/netvsc/hn_rxtx.c +++ b/drivers/net/netvsc/hn_rxtx.c @@ -11,4 +11,5 @@ @@ -64,7 +65,7 @@ + return rxq; } -@@ -864,17 +873,32 @@ uint32_t hn_process_events(struct hn_data *hv, uint16_t queue_id, +@@ -854,17 +863,32 @@ void hn_process_events(struct hn_data *hv, uint16_t queue_id) for (;;) { const struct vmbus_chanpkt_hdr *pkt; - uint32_t len = HN_RXQ_EVENT_DEFAULT; @@ -103,10 +104,10 @@ bytes_read += ret; diff --git a/drivers/net/netvsc/hn_var.h b/drivers/net/netvsc/hn_var.h -index fec8d7c40..b42bd97b9 100644 +index f7ff8585b..0430f450c 100644 --- a/drivers/net/netvsc/hn_var.h +++ b/drivers/net/netvsc/hn_var.h -@@ -81,5 +81,5 @@ struct hn_rx_queue { +@@ -78,5 +78,5 @@ struct hn_rx_queue { uint64_t ring_full; - uint8_t event_buf[];