From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Hunt, David" Subject: Re: [PATCH v7 04/17] lib: add new burst oriented distributor structs Date: Wed, 1 Mar 2017 09:57:27 +0000 Message-ID: <86b189d1-14cc-7959-b286-8d3fac38aa78@intel.com> References: <1485163480-156507-2-git-send-email-david.hunt@intel.com> <1487647073-129064-1-git-send-email-david.hunt@intel.com> <1487647073-129064-5-git-send-email-david.hunt@intel.com> <20170224140850.GE106392@bricha3-MOBL3.ger.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Cc: dev@dpdk.org To: Bruce Richardson Return-path: Received: from mga02.intel.com (mga02.intel.com [134.134.136.20]) by dpdk.org (Postfix) with ESMTP id 293E12BDF for ; Wed, 1 Mar 2017 10:57:30 +0100 (CET) In-Reply-To: <20170224140850.GE106392@bricha3-MOBL3.ger.corp.intel.com> List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On 24/2/2017 2:08 PM, Bruce Richardson wrote: > On Tue, Feb 21, 2017 at 03:17:40AM +0000, David Hunt wrote: >> Signed-off-by: David Hunt >> --- >> lib/librte_distributor/rte_distributor_private.h | 61 ++++++++++++++++++++++++ >> 1 file changed, 61 insertions(+) >> >> diff --git a/lib/librte_distributor/rte_distributor_private.h b/lib/librte_distributor/rte_distributor_private.h >> index 2d85b9b..c8e0f98 100644 >> --- a/lib/librte_distributor/rte_distributor_private.h >> +++ b/lib/librte_distributor/rte_distributor_private.h >> @@ -129,6 +129,67 @@ struct rte_distributor_v20 { >> struct rte_distributor_returned_pkts returns; >> }; >> >> +/* All different signature compare functions */ >> +enum rte_distributor_match_function { >> + RTE_DIST_MATCH_SCALAR = 0, >> + RTE_DIST_MATCH_VECTOR, >> + RTE_DIST_NUM_MATCH_FNS >> +}; >> + >> +/** >> + * Buffer structure used to pass the pointer data between cores. This is cache >> + * line aligned, but to improve performance and prevent adjacent cache-line >> + * prefetches of buffers for other workers, e.g. when worker 1's buffer is on >> + * the next cache line to worker 0, we pad this out to two cache lines. >> + * We can pass up to 8 mbufs at a time in one cacheline. >> + * There is a separate cacheline for returns in the burst API. >> + */ >> +struct rte_distributor_buffer { >> + volatile int64_t bufptr64[RTE_DIST_BURST_SIZE] >> + __rte_cache_aligned; /* <= outgoing to worker */ >> + >> + int64_t pad1 __rte_cache_aligned; /* <= one cache line */ >> + >> + volatile int64_t retptr64[RTE_DIST_BURST_SIZE] >> + __rte_cache_aligned; /* <= incoming from worker */ >> + >> + int64_t pad2 __rte_cache_aligned; /* <= one cache line */ >> + >> + int count __rte_cache_aligned; /* <= number of current mbufs */ >> +}; >> + >> +struct rte_distributor { >> + TAILQ_ENTRY(rte_distributor) next; /**< Next in list. */ >> + >> + char name[RTE_DISTRIBUTOR_NAMESIZE]; /**< Name of the ring. */ >> + unsigned int num_workers; /**< Number of workers polling */ >> + unsigned int alg_type; /**< Number of alg types */ >> + >> + /**> >> + * First cache line in the this array are the tags inflight >> + * on the worker core. Second cache line are the backlog >> + * that are going to go to the worker core. >> + */ >> + uint16_t in_flight_tags[RTE_DISTRIB_MAX_WORKERS][RTE_DIST_BURST_SIZE*2] >> + __rte_cache_aligned; >> + >> + struct rte_distributor_backlog backlog[RTE_DISTRIB_MAX_WORKERS] >> + __rte_cache_aligned; >> + >> + struct rte_distributor_buffer bufs[RTE_DISTRIB_MAX_WORKERS]; >> + >> + struct rte_distributor_returned_pkts returns; >> + >> + enum rte_distributor_match_function dist_match_fn; >> + >> + struct rte_distributor_v20 *d_v20; >> +}; >> + >> +void >> +find_match_scalar(struct rte_distributor *d, >> + uint16_t *data_ptr, >> + uint16_t *output_ptr); >> + >> #ifdef __cplusplus >> } >> #endif > The last patch claimed that this header file is for structs/definitions > common between the old and new implementations. These definitions look > to apply only to the new one, so do they belong in the .c file instead? The _v20 structs are used as a fallback in the new struct, so probably best to have in a common private file.