From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail.droids-corp.org (zoll.droids-corp.org [94.23.50.67]) by dpdk.org (Postfix) with ESMTP id 3427B1B05 for ; Fri, 29 Mar 2019 13:54:31 +0100 (CET) Received: from lfbn-1-5920-128.w90-110.abo.wanadoo.fr ([90.110.126.128] helo=droids-corp.org) by mail.droids-corp.org with esmtpsa (TLS1.0:RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1h9r43-000281-Hz; Fri, 29 Mar 2019 13:56:57 +0100 Received: by droids-corp.org (sSMTP sendmail emulation); Fri, 29 Mar 2019 13:54:27 +0100 Date: Fri, 29 Mar 2019 13:54:27 +0100 From: Olivier Matz To: Konstantin Ananyev Cc: dev@dpdk.org, akhil.goyal@nxp.com Message-ID: <20190329125427.hdwevmm4wwl73tlj@platinum> References: <20190326154320.29913-1-konstantin.ananyev@intel.com> <20190329102726.27716-1-konstantin.ananyev@intel.com> <20190329102726.27716-2-konstantin.ananyev@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190329102726.27716-2-konstantin.ananyev@intel.com> User-Agent: NeoMutt/20170113 (1.7.2) Subject: Re: [dpdk-dev] [PATCH v4 1/9] mbuf: new function to generate raw Tx offload value X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Mar 2019 12:54:31 -0000 Hi Konstantin, On Fri, Mar 29, 2019 at 10:27:18AM +0000, Konstantin Ananyev wrote: > Operations to set/update bit-fields often cause compilers > to generate suboptimal code. > To help avoid such situation for tx_offload fields: > introduce new enum for tx_offload bit-fields lengths and offsets, > and new function to generate raw tx_offload value. > > Signed-off-by: Konstantin Ananyev > Acked-by: Akhil Goyal I understand the need. Out of curiosity, do you have any performance numbers to share? Few cosmetic questions below. > --- > lib/librte_mbuf/rte_mbuf.h | 79 ++++++++++++++++++++++++++++++++++---- > 1 file changed, 72 insertions(+), 7 deletions(-) > > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h > index d961ccaf6..0b197e8ce 100644 > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -479,6 +479,31 @@ struct rte_mbuf_sched { > uint16_t reserved; /**< Reserved. */ > }; /**< Hierarchical scheduler */ > > +/** > + * enum for the tx_offload bit-fields lenghts and offsets. > + * defines the layout of rte_mbuf tx_offload field. > + */ > +enum { > + RTE_MBUF_L2_LEN_BITS = 7, > + RTE_MBUF_L3_LEN_BITS = 9, > + RTE_MBUF_L4_LEN_BITS = 8, > + RTE_MBUF_TSO_SEGSZ_BITS = 16, > + RTE_MBUF_OUTL3_LEN_BITS = 9, > + RTE_MBUF_OUTL2_LEN_BITS = 7, > + RTE_MBUF_L2_LEN_OFS = 0, > + RTE_MBUF_L3_LEN_OFS = RTE_MBUF_L2_LEN_OFS + RTE_MBUF_L2_LEN_BITS, > + RTE_MBUF_L4_LEN_OFS = RTE_MBUF_L3_LEN_OFS + RTE_MBUF_L3_LEN_BITS, > + RTE_MBUF_TSO_SEGSZ_OFS = RTE_MBUF_L4_LEN_OFS + RTE_MBUF_L4_LEN_BITS, > + RTE_MBUF_OUTL3_LEN_OFS = > + RTE_MBUF_TSO_SEGSZ_OFS + RTE_MBUF_TSO_SEGSZ_BITS, > + RTE_MBUF_OUTL2_LEN_OFS = > + RTE_MBUF_OUTL3_LEN_OFS + RTE_MBUF_OUTL3_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_OFS = > + RTE_MBUF_OUTL2_LEN_OFS + RTE_MBUF_OUTL2_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_BITS = > + sizeof(uint64_t) * CHAR_BIT - RTE_MBUF_TXOFLD_UNUSED_OFS, > +}; > + What is the advantage of defining an enum instead of #defines? In any case, I wonder if it wouldn't be clearer to change the order like this: enum { RTE_MBUF_L2_LEN_OFS = 0, RTE_MBUF_L2_LEN_BITS = 7, RTE_MBUF_L3_LEN_OFS = RTE_MBUF_L2_LEN_OFS + RTE_MBUF_L2_LEN_BITS, RTE_MBUF_L3_LEN_BITS = 9, RTE_MBUF_L4_LEN_OFS = RTE_MBUF_L3_LEN_OFS + RTE_MBUF_L3_LEN_BITS, RTE_MBUF_L4_LEN_BITS = 8, ... > /** > * The generic rte_mbuf, containing a packet mbuf. > */ > @@ -640,19 +665,24 @@ struct rte_mbuf { > uint64_t tx_offload; /**< combined for easy fetch */ > __extension__ > struct { > - uint64_t l2_len:7; > + uint64_t l2_len:RTE_MBUF_L2_LEN_BITS; > /**< L2 (MAC) Header Length for non-tunneling pkt. > * Outer_L4_len + ... + Inner_L2_len for tunneling pkt. > */ > - uint64_t l3_len:9; /**< L3 (IP) Header Length. */ > - uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ > - uint64_t tso_segsz:16; /**< TCP TSO segment size */ > + uint64_t l3_len:RTE_MBUF_L3_LEN_BITS; > + /**< L3 (IP) Header Length. */ > + uint64_t l4_len:RTE_MBUF_L4_LEN_BITS; > + /**< L4 (TCP/UDP) Header Length. */ > + uint64_t tso_segsz:RTE_MBUF_TSO_SEGSZ_BITS; > + /**< TCP TSO segment size */ > > /* fields for TX offloading of tunnels */ > - uint64_t outer_l3_len:9; /**< Outer L3 (IP) Hdr Length. */ > - uint64_t outer_l2_len:7; /**< Outer L2 (MAC) Hdr Length. */ > + uint64_t outer_l3_len:RTE_MBUF_OUTL3_LEN_BITS; > + /**< Outer L3 (IP) Hdr Length. */ > + uint64_t outer_l2_len:RTE_MBUF_OUTL2_LEN_BITS; > + /**< Outer L2 (MAC) Hdr Length. */ > > - /* uint64_t unused:8; */ > + /* uint64_t unused:RTE_MBUF_TXOFLD_UNUSED_BITS; */ > }; > }; > > @@ -2243,6 +2273,41 @@ static inline int rte_pktmbuf_chain(struct rte_mbuf *head, struct rte_mbuf *tail > return 0; > } > > +/* > + * @warning > + * @b EXPERIMENTAL: This API may change without prior notice. > + * > + * For given input values generate raw tx_offload value. > + * @param il2 > + * l2_len value. > + * @param il3 > + * l3_len value. > + * @param il4 > + * l4_len value. > + * @param tso > + * tso_segsz value. > + * @param ol3 > + * outer_l3_len value. > + * @param ol2 > + * outer_l2_len value. > + * @param unused > + * unused value. > + * @return > + * raw tx_offload value. > + */ > +static __rte_always_inline uint64_t > +rte_mbuf_tx_offload(uint64_t il2, uint64_t il3, uint64_t il4, uint64_t tso, > + uint64_t ol3, uint64_t ol2, uint64_t unused) > +{ > + return il2 << RTE_MBUF_L2_LEN_OFS | > + il3 << RTE_MBUF_L3_LEN_OFS | > + il4 << RTE_MBUF_L4_LEN_OFS | > + tso << RTE_MBUF_TSO_SEGSZ_OFS | > + ol3 << RTE_MBUF_OUTL3_LEN_OFS | > + ol2 << RTE_MBUF_OUTL2_LEN_OFS | > + unused << RTE_MBUF_TXOFLD_UNUSED_OFS; > +} > + > /** >>From what I see, the problem is quite similar to what was done with rte_mbuf_sched_set() recently. So I wondered if it was possible to declare a structure like this: struct rte_mbuf_ol_len { uint64_t l2_len:7; uint64_t l3_len:9; /**< L3 (IP) Header Length. */ uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ ... } And have the set function like this: m->l = (struct rte_mbuf_ol_len) { .l2_len = l2_len, .l3_len = l3_len, .l4_len = l4_len, ... This would avoid the definition of the offsets and bits, but I didn't find any way to declare these fields as anonymous in the mbuf structure. Did you tried that way too? Thanks, Olivier From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by dpdk.space (Postfix) with ESMTP id 97D25A05D3 for ; Fri, 29 Mar 2019 13:54:34 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 118E82BD3; Fri, 29 Mar 2019 13:54:33 +0100 (CET) Received: from mail.droids-corp.org (zoll.droids-corp.org [94.23.50.67]) by dpdk.org (Postfix) with ESMTP id 3427B1B05 for ; Fri, 29 Mar 2019 13:54:31 +0100 (CET) Received: from lfbn-1-5920-128.w90-110.abo.wanadoo.fr ([90.110.126.128] helo=droids-corp.org) by mail.droids-corp.org with esmtpsa (TLS1.0:RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1h9r43-000281-Hz; Fri, 29 Mar 2019 13:56:57 +0100 Received: by droids-corp.org (sSMTP sendmail emulation); Fri, 29 Mar 2019 13:54:27 +0100 Date: Fri, 29 Mar 2019 13:54:27 +0100 From: Olivier Matz To: Konstantin Ananyev Cc: dev@dpdk.org, akhil.goyal@nxp.com Message-ID: <20190329125427.hdwevmm4wwl73tlj@platinum> References: <20190326154320.29913-1-konstantin.ananyev@intel.com> <20190329102726.27716-1-konstantin.ananyev@intel.com> <20190329102726.27716-2-konstantin.ananyev@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Disposition: inline In-Reply-To: <20190329102726.27716-2-konstantin.ananyev@intel.com> User-Agent: NeoMutt/20170113 (1.7.2) Subject: Re: [dpdk-dev] [PATCH v4 1/9] mbuf: new function to generate raw Tx offload value X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Message-ID: <20190329125427.UpB6ECVyTeRGLly1FVI_mSbk2MmkP5coZF8bLkpfX1w@z> Hi Konstantin, On Fri, Mar 29, 2019 at 10:27:18AM +0000, Konstantin Ananyev wrote: > Operations to set/update bit-fields often cause compilers > to generate suboptimal code. > To help avoid such situation for tx_offload fields: > introduce new enum for tx_offload bit-fields lengths and offsets, > and new function to generate raw tx_offload value. > > Signed-off-by: Konstantin Ananyev > Acked-by: Akhil Goyal I understand the need. Out of curiosity, do you have any performance numbers to share? Few cosmetic questions below. > --- > lib/librte_mbuf/rte_mbuf.h | 79 ++++++++++++++++++++++++++++++++++---- > 1 file changed, 72 insertions(+), 7 deletions(-) > > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h > index d961ccaf6..0b197e8ce 100644 > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -479,6 +479,31 @@ struct rte_mbuf_sched { > uint16_t reserved; /**< Reserved. */ > }; /**< Hierarchical scheduler */ > > +/** > + * enum for the tx_offload bit-fields lenghts and offsets. > + * defines the layout of rte_mbuf tx_offload field. > + */ > +enum { > + RTE_MBUF_L2_LEN_BITS = 7, > + RTE_MBUF_L3_LEN_BITS = 9, > + RTE_MBUF_L4_LEN_BITS = 8, > + RTE_MBUF_TSO_SEGSZ_BITS = 16, > + RTE_MBUF_OUTL3_LEN_BITS = 9, > + RTE_MBUF_OUTL2_LEN_BITS = 7, > + RTE_MBUF_L2_LEN_OFS = 0, > + RTE_MBUF_L3_LEN_OFS = RTE_MBUF_L2_LEN_OFS + RTE_MBUF_L2_LEN_BITS, > + RTE_MBUF_L4_LEN_OFS = RTE_MBUF_L3_LEN_OFS + RTE_MBUF_L3_LEN_BITS, > + RTE_MBUF_TSO_SEGSZ_OFS = RTE_MBUF_L4_LEN_OFS + RTE_MBUF_L4_LEN_BITS, > + RTE_MBUF_OUTL3_LEN_OFS = > + RTE_MBUF_TSO_SEGSZ_OFS + RTE_MBUF_TSO_SEGSZ_BITS, > + RTE_MBUF_OUTL2_LEN_OFS = > + RTE_MBUF_OUTL3_LEN_OFS + RTE_MBUF_OUTL3_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_OFS = > + RTE_MBUF_OUTL2_LEN_OFS + RTE_MBUF_OUTL2_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_BITS = > + sizeof(uint64_t) * CHAR_BIT - RTE_MBUF_TXOFLD_UNUSED_OFS, > +}; > + What is the advantage of defining an enum instead of #defines? In any case, I wonder if it wouldn't be clearer to change the order like this: enum { RTE_MBUF_L2_LEN_OFS = 0, RTE_MBUF_L2_LEN_BITS = 7, RTE_MBUF_L3_LEN_OFS = RTE_MBUF_L2_LEN_OFS + RTE_MBUF_L2_LEN_BITS, RTE_MBUF_L3_LEN_BITS = 9, RTE_MBUF_L4_LEN_OFS = RTE_MBUF_L3_LEN_OFS + RTE_MBUF_L3_LEN_BITS, RTE_MBUF_L4_LEN_BITS = 8, ... > /** > * The generic rte_mbuf, containing a packet mbuf. > */ > @@ -640,19 +665,24 @@ struct rte_mbuf { > uint64_t tx_offload; /**< combined for easy fetch */ > __extension__ > struct { > - uint64_t l2_len:7; > + uint64_t l2_len:RTE_MBUF_L2_LEN_BITS; > /**< L2 (MAC) Header Length for non-tunneling pkt. > * Outer_L4_len + ... + Inner_L2_len for tunneling pkt. > */ > - uint64_t l3_len:9; /**< L3 (IP) Header Length. */ > - uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ > - uint64_t tso_segsz:16; /**< TCP TSO segment size */ > + uint64_t l3_len:RTE_MBUF_L3_LEN_BITS; > + /**< L3 (IP) Header Length. */ > + uint64_t l4_len:RTE_MBUF_L4_LEN_BITS; > + /**< L4 (TCP/UDP) Header Length. */ > + uint64_t tso_segsz:RTE_MBUF_TSO_SEGSZ_BITS; > + /**< TCP TSO segment size */ > > /* fields for TX offloading of tunnels */ > - uint64_t outer_l3_len:9; /**< Outer L3 (IP) Hdr Length. */ > - uint64_t outer_l2_len:7; /**< Outer L2 (MAC) Hdr Length. */ > + uint64_t outer_l3_len:RTE_MBUF_OUTL3_LEN_BITS; > + /**< Outer L3 (IP) Hdr Length. */ > + uint64_t outer_l2_len:RTE_MBUF_OUTL2_LEN_BITS; > + /**< Outer L2 (MAC) Hdr Length. */ > > - /* uint64_t unused:8; */ > + /* uint64_t unused:RTE_MBUF_TXOFLD_UNUSED_BITS; */ > }; > }; > > @@ -2243,6 +2273,41 @@ static inline int rte_pktmbuf_chain(struct rte_mbuf *head, struct rte_mbuf *tail > return 0; > } > > +/* > + * @warning > + * @b EXPERIMENTAL: This API may change without prior notice. > + * > + * For given input values generate raw tx_offload value. > + * @param il2 > + * l2_len value. > + * @param il3 > + * l3_len value. > + * @param il4 > + * l4_len value. > + * @param tso > + * tso_segsz value. > + * @param ol3 > + * outer_l3_len value. > + * @param ol2 > + * outer_l2_len value. > + * @param unused > + * unused value. > + * @return > + * raw tx_offload value. > + */ > +static __rte_always_inline uint64_t > +rte_mbuf_tx_offload(uint64_t il2, uint64_t il3, uint64_t il4, uint64_t tso, > + uint64_t ol3, uint64_t ol2, uint64_t unused) > +{ > + return il2 << RTE_MBUF_L2_LEN_OFS | > + il3 << RTE_MBUF_L3_LEN_OFS | > + il4 << RTE_MBUF_L4_LEN_OFS | > + tso << RTE_MBUF_TSO_SEGSZ_OFS | > + ol3 << RTE_MBUF_OUTL3_LEN_OFS | > + ol2 << RTE_MBUF_OUTL2_LEN_OFS | > + unused << RTE_MBUF_TXOFLD_UNUSED_OFS; > +} > + > /** >From what I see, the problem is quite similar to what was done with rte_mbuf_sched_set() recently. So I wondered if it was possible to declare a structure like this: struct rte_mbuf_ol_len { uint64_t l2_len:7; uint64_t l3_len:9; /**< L3 (IP) Header Length. */ uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ ... } And have the set function like this: m->l = (struct rte_mbuf_ol_len) { .l2_len = l2_len, .l3_len = l3_len, .l4_len = l4_len, ... This would avoid the definition of the offsets and bits, but I didn't find any way to declare these fields as anonymous in the mbuf structure. Did you tried that way too? Thanks, Olivier