From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 88B0341EC0; Fri, 17 Mar 2023 21:44:49 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 6303A42FD8; Fri, 17 Mar 2023 21:44:49 +0100 (CET) Received: from linux.microsoft.com (linux.microsoft.com [13.77.154.182]) by mails.dpdk.org (Postfix) with ESMTP id 988B042F98 for ; Fri, 17 Mar 2023 21:44:47 +0100 (CET) Received: by linux.microsoft.com (Postfix, from userid 1086) id E4A2920C32D1; Fri, 17 Mar 2023 13:44:46 -0700 (PDT) DKIM-Filter: OpenDKIM Filter v2.11.0 linux.microsoft.com E4A2920C32D1 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.microsoft.com; s=default; t=1679085886; bh=wGG8p50s2M0pHeVrNyR8I0RPc4WFbXH7yT+9pIJsL00=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=CQKAxhpeTugQntSsfHPBBeaHcwYLQWoYRJ4/VpDeGYNtK5IWHDKDDJA16nMJFQvq6 UWk2Xw0NEe09cLjQUwrLxNFFc/lVsXpda6/doZYQn7RTOpULmLPv/iPLRhT9MmtqR6 DNq7SeJfW4gFWOT4bwf9arZrWwoPTSmN/l1WlFJY= Date: Fri, 17 Mar 2023 13:44:46 -0700 From: Tyler Retzlaff To: dev@dpdk.org Cc: Honnappa.Nagarahalli@arm.com, Ruifeng.Wang@arm.com, thomas@monjalon.net Subject: Re: [PATCH 6/7] net/null: replace rte atomics with GCC builtin atomics Message-ID: <20230317204446.GC30723@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net> References: <1679084388-19267-1-git-send-email-roretzla@linux.microsoft.com> <1679084388-19267-7-git-send-email-roretzla@linux.microsoft.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1679084388-19267-7-git-send-email-roretzla@linux.microsoft.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On Fri, Mar 17, 2023 at 01:19:47PM -0700, Tyler Retzlaff wrote: > Replace the use of rte_atomic.h types and functions, instead use GCC > supplied C++11 memory model builtins. > > Signed-off-by: Tyler Retzlaff > --- > drivers/net/null/rte_eth_null.c | 20 ++++++++++---------- > 1 file changed, 10 insertions(+), 10 deletions(-) > > diff --git a/drivers/net/null/rte_eth_null.c b/drivers/net/null/rte_eth_null.c > index 47d9554..195c3bd 100644 > --- a/drivers/net/null/rte_eth_null.c > +++ b/drivers/net/null/rte_eth_null.c > @@ -37,8 +37,8 @@ struct null_queue { > struct rte_mempool *mb_pool; > struct rte_mbuf *dummy_packet; > > - rte_atomic64_t rx_pkts; > - rte_atomic64_t tx_pkts; > + int64_t rx_pkts; > + int64_t tx_pkts; > }; > > struct pmd_options { > @@ -101,7 +101,7 @@ struct pmd_internals { > bufs[i]->port = h->internals->port_id; > } > > - rte_atomic64_add(&(h->rx_pkts), i); > + __atomic_fetch_add(&h->rx_pkts, i, __ATOMIC_SEQ_CST); > > return i; > } > @@ -128,7 +128,7 @@ struct pmd_internals { > bufs[i]->port = h->internals->port_id; > } > > - rte_atomic64_add(&(h->rx_pkts), i); > + __atomic_fetch_add(&h->rx_pkts, i, __ATOMIC_SEQ_CST); > > return i; > } > @@ -152,7 +152,7 @@ struct pmd_internals { > for (i = 0; i < nb_bufs; i++) > rte_pktmbuf_free(bufs[i]); > > - rte_atomic64_add(&(h->tx_pkts), i); > + __atomic_fetch_add(&h->tx_pkts, i, __ATOMIC_SEQ_CST); > > return i; > } > @@ -174,7 +174,7 @@ struct pmd_internals { > rte_pktmbuf_free(bufs[i]); > } > > - rte_atomic64_add(&(h->tx_pkts), i); > + __atomic_fetch_add(&h->tx_pkts, i, __ATOMIC_SEQ_CST); > > return i; > } > @@ -317,7 +317,7 @@ struct pmd_internals { > RTE_DIM(internal->rx_null_queues))); > for (i = 0; i < num_stats; i++) { > igb_stats->q_ipackets[i] = > - internal->rx_null_queues[i].rx_pkts.cnt; > + internal->rx_null_queues[i].rx_pkts; > rx_total += igb_stats->q_ipackets[i]; > } > > @@ -326,7 +326,7 @@ struct pmd_internals { > RTE_DIM(internal->tx_null_queues))); > for (i = 0; i < num_stats; i++) { > igb_stats->q_opackets[i] = > - internal->tx_null_queues[i].tx_pkts.cnt; > + internal->tx_null_queues[i].tx_pkts; > tx_total += igb_stats->q_opackets[i]; > } > these variables are operated on with atomic builtins in other places yet here they are being non-atomically loaded. should probably be using _atomic_load_n(...) > @@ -347,9 +347,9 @@ struct pmd_internals { > > internal = dev->data->dev_private; > for (i = 0; i < RTE_DIM(internal->rx_null_queues); i++) > - internal->rx_null_queues[i].rx_pkts.cnt = 0; > + internal->rx_null_queues[i].rx_pkts = 0; > for (i = 0; i < RTE_DIM(internal->tx_null_queues); i++) > - internal->tx_null_queues[i].tx_pkts.cnt = 0; > + internal->tx_null_queues[i].tx_pkts = 0; same thing, these should probably be __atomic_store_n(...)