From: "Wangxiaoyun (Cloud, Network Chip Application Development Dept)" <cloud.wangxiaoyun@huawei.com>
To: "Gavin Hu (Arm Technology China)" <Gavin.Hu@arm.com>,
"ferruh.yigit@intel.com" <ferruh.yigit@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
"xuanziyang2@huawei.com" <xuanziyang2@huawei.com>,
"shahar.belkar@huawei.com" <shahar.belkar@huawei.com>,
"luoxianjun@huawei.com" <luoxianjun@huawei.com>,
"tanya.brokhman@huawei.com" <tanya.brokhman@huawei.com>,
"zhouguoyang@huawei.com" <zhouguoyang@huawei.com>,
"wulike1@huawei.com" <wulike1@huawei.com>, nd <nd@arm.com>
Subject: Re: [dpdk-dev] [PATCH v2 17/17] net/hinic: optimize tx&rx performance
Date: Mon, 30 Sep 2019 22:41:07 +0800 [thread overview]
Message-ID: <39715600-b11d-fd20-b5d5-098259454201@huawei.com> (raw)
In-Reply-To: <VI1PR08MB537642D19C51A25B7A82B00F8F810@VI1PR08MB5376.eurprd08.prod.outlook.com>
Hi Gavin,
Thanks for your comments.
+#if defined(__ARM64_NEON__)
No NEON intrinsics used, maybe RTE_ARCH_ARM64 is better.
In the following line __rte_always_inline is commonly used in DPDK, the effect is same.
/Gavin
For this patch, we don't use NEON intrinsics, but for tx& rx process, we use NEON intrinsics
for wqebb bigend conversion on arm platform, so in order to keep ingres, all intrinsics
optimization on arm we use __ARM64_NEON__ definitation.
I understand your intention is the reading of the status is observed before the following reads.
This can be fulfilled by __atomic_load_n(...) with __ATOMIC_ACQUIRE semantics.
This C11 way applies to all the arches, and you don't need the differentiation of arches.
/Gavin
Thanks, i have changed it to __atomic_load_n(...) with __ATOMIC_ACQUIRE semantics, and send a new patch V3.
Best regards
Xiaoyun Wang
在 2019/9/27 10:08, Gavin Hu (Arm Technology China) 写道:
> Hi Xiaoyun,
>
>> -----Original Message-----
>> From: dev <dev-bounces@dpdk.org> On Behalf Of Xiaoyun wang
>> Sent: Wednesday, September 25, 2019 10:31 PM
>> To: ferruh.yigit@intel.com
>> Cc: dev@dpdk.org; xuanziyang2@huawei.com; shahar.belkar@huawei.com;
>> luoxianjun@huawei.com; tanya.brokhman@huawei.com;
>> zhouguoyang@huawei.com; wulike1@huawei.com; Xiaoyun wang
>> <cloud.wangxiaoyun@huawei.com>
>> Subject: [dpdk-dev] [PATCH v2 17/17] net/hinic: optimize tx&rx
>> performance
>>
>> This patch optimizes receive packets performance
>> in arm platform.
>>
>> Signed-off-by: Xiaoyun wang <cloud.wangxiaoyun@huawei.com>
>> ---
>> drivers/net/hinic/hinic_pmd_rx.c | 17 +++++++++++++++++
>> drivers/net/hinic/hinic_pmd_rx.h | 11 +++++++++++
>> 2 files changed, 28 insertions(+)
>>
>> diff --git a/drivers/net/hinic/hinic_pmd_rx.c
>> b/drivers/net/hinic/hinic_pmd_rx.c
>> index 37b4f5c..94071ee 100644
>> --- a/drivers/net/hinic/hinic_pmd_rx.c
>> +++ b/drivers/net/hinic/hinic_pmd_rx.c
>> @@ -950,6 +950,19 @@ void hinic_rx_alloc_pkts(struct hinic_rxq *rxq)
>> }
>> }
>>
>> +#if defined(__ARM64_NEON__)
> No NEON intrinsics used, maybe RTE_ARCH_ARM64 is better.
> In the following line __rte_always_inline is commonly used in DPDK, the effect is same.
> /Gavin
>
>> +static inline uint32_t __attribute__((always_inline))
>> +hinic_read_cqe_status(uintptr_t addr)
>> +{
>> + uint32_t val;
>> +
>> + asm volatile("ldar %x[val], [%x[addr]]"
>> + : [val] "=r" (val)
>> + : [addr] "r" (addr));
>> + return val;
>> +}
>> +#endif
> I understand your intention is the reading of the status is observed before the following reads.
> This can be fulfilled by __atomic_load_n(...) with __ATOMIC_ACQUIRE semantics.
> This C11 way applies to all the arches, and you don't need the differentiation of arches.
> /Gavin
>> +
>> u16 hinic_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, u16
>> nb_pkts)
>> {
>> struct rte_mbuf *rxm;
>> @@ -972,7 +985,11 @@ u16 hinic_recv_pkts(void *rx_queue, struct
>> rte_mbuf **rx_pkts, u16 nb_pkts)
>> while (pkts < nb_pkts) {
>> /* 2. current ci is done */
>> rx_cqe = &rxq->rx_cqe[sw_ci];
>> +#if defined(__X86_64_SSE__)
>> status = rx_cqe->status;
>> +#elif defined(__ARM64_NEON__)
>> + status = hinic_read_cqe_status((uintptr_t)&rxq-
>>> rx_cqe[sw_ci]);
>> +#endif
>> if (!HINIC_GET_RX_DONE_BE(status))
>> break;
>>
>> diff --git a/drivers/net/hinic/hinic_pmd_rx.h
>> b/drivers/net/hinic/hinic_pmd_rx.h
>> index fe2735b..fa27e91 100644
>> --- a/drivers/net/hinic/hinic_pmd_rx.h
>> +++ b/drivers/net/hinic/hinic_pmd_rx.h
>> @@ -28,6 +28,7 @@ struct hinic_rq_ctrl {
>> u32 ctrl_fmt;
>> };
>>
>> +#if defined(__X86_64_SSE__)
>> struct hinic_rq_cqe {
>> u32 status;
>> u32 vlan_len;
>> @@ -36,6 +37,16 @@ struct hinic_rq_cqe {
>>
>> u32 rsvd[4];
>> };
>> +#elif defined(__ARM64_NEON__)
>> +struct hinic_rq_cqe {
>> + u32 status;
>> + u32 vlan_len;
>> + u32 offload_type;
>> + u32 rss_hash;
>> +
>> + u32 rsvd[4];
>> +} __rte_cache_aligned;
>> +#endif
>>
>> struct hinic_rq_cqe_sect {
>> struct hinic_sge sge;
>> --
>> 1.8.3.1
>
next prev parent reply other threads:[~2019-09-30 14:41 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-25 14:26 [dpdk-dev] [PATCH v2 00/17] Add advanced features for Huawei hinic pmd Xiaoyun wang
2019-09-25 14:30 ` Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 01/17] net/hinic/base: add mbox command channel for SRIOV Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 02/17] net/hinic/base: add HW interfaces " Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 03/17] net/hinic: add VF PMD operation interfaces Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 04/17] net/hinic: add VLAN filter and offload Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 05/17] net/hinic: add allmulticast mode and MTU set Xiaoyun wang
2019-09-26 18:47 ` Ferruh Yigit
2019-09-30 8:43 ` Wangxiaoyun (Cloud, Network Chip Application Development Dept)
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 06/17] net/hinic: add unicast and multicast MAC set Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 07/17] net/hinic: add fdir config interface Xiaoyun wang
2019-09-26 18:48 ` Ferruh Yigit
2019-09-30 14:08 ` Wangxiaoyun (Cloud, Network Chip Application Development Dept)
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 08/17] net/hinic: add fdir validate flow operations Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 09/17] net/hinic: create and destroy ntuple filter Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 10/17] net/hinic: create and destroy fdir filter Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 11/17] net/hinic: flush " Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 12/17] net/hinic: set link down and up Xiaoyun wang
2019-09-26 18:47 ` Ferruh Yigit
2019-09-28 8:59 ` [dpdk-dev] 答复: " Wangxiaoyun (Cloud, Network Chip Application Development Dept)
2019-09-30 8:39 ` [dpdk-dev] " Wangxiaoyun (Cloud, Network Chip Application Development Dept)
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 13/17] net/hinic: support inner L3 checksum offload Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 14/17] net/hinic: support LRO offload Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 15/17] net/hinic: add hinic PMD doc files Xiaoyun wang
2019-09-26 18:51 ` Ferruh Yigit
2019-09-30 14:15 ` Wangxiaoyun (Cloud, Network Chip Application Development Dept)
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 16/17] net/hinic/base: optimize aeq interfaces Xiaoyun wang
2019-09-25 14:30 ` [dpdk-dev] [PATCH v2 17/17] net/hinic: optimize tx&rx performance Xiaoyun wang
2019-09-27 2:08 ` Gavin Hu (Arm Technology China)
2019-09-30 14:41 ` Wangxiaoyun (Cloud, Network Chip Application Development Dept) [this message]
2019-09-26 18:51 ` [dpdk-dev] [PATCH v2 00/17] Add advanced features for Huawei hinic pmd Ferruh Yigit
2019-09-30 14:19 ` Wangxiaoyun (Cloud, Network Chip Application Development Dept)
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=39715600-b11d-fd20-b5d5-098259454201@huawei.com \
--to=cloud.wangxiaoyun@huawei.com \
--cc=Gavin.Hu@arm.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=luoxianjun@huawei.com \
--cc=nd@arm.com \
--cc=shahar.belkar@huawei.com \
--cc=tanya.brokhman@huawei.com \
--cc=wulike1@huawei.com \
--cc=xuanziyang2@huawei.com \
--cc=zhouguoyang@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).