From patchwork Mon Feb 15 11:14:56 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paul Durrant X-Patchwork-Id: 8313501 Return-Path: X-Original-To: patchwork-xen-devel@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 15E21C02AA for ; Mon, 15 Feb 2016 11:26:10 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 842BB202F2 for ; Mon, 15 Feb 2016 11:26:08 +0000 (UTC) Received: from lists.xen.org (lists.xenproject.org [50.57.142.19]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id CF5412028D for ; Mon, 15 Feb 2016 11:26:06 +0000 (UTC) Received: from localhost ([127.0.0.1] helo=lists.xen.org) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1aVHFW-0005YX-Hr; Mon, 15 Feb 2016 11:23:26 +0000 Received: from mail6.bemta5.messagelabs.com ([195.245.231.135]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1aVHFU-0005YQ-UE for xen-devel@lists.xenproject.org; Mon, 15 Feb 2016 11:23:25 +0000 Received: from [85.158.139.211] by server-2.bemta-5.messagelabs.com id AC/57-21594-C25B1C65; Mon, 15 Feb 2016 11:23:24 +0000 X-Env-Sender: prvs=846b27f3b=Paul.Durrant@citrix.com X-Msg-Ref: server-16.tower-206.messagelabs.com!1455535401!22139531!1 X-Originating-IP: [66.165.176.89] X-SpamReason: No, hits=0.0 required=7.0 tests=sa_preprocessor: VHJ1c3RlZCBJUDogNjYuMTY1LjE3Ni44OSA9PiAyMDMwMDc=\n, received_headers: No Received headers X-StarScan-Received: X-StarScan-Version: 7.35.1; banners=-,-,- X-VirusChecked: Checked Received: (qmail 50299 invoked from network); 15 Feb 2016 11:23:22 -0000 Received: from smtp.citrix.com (HELO SMTP.CITRIX.COM) (66.165.176.89) by server-16.tower-206.messagelabs.com with RC4-SHA encrypted SMTP; 15 Feb 2016 11:23:22 -0000 X-IronPort-AV: E=Sophos;i="5.22,449,1449532800"; d="scan'208";a="331702182" From: Paul Durrant To: Date: Mon, 15 Feb 2016 11:14:56 +0000 Message-ID: <1455534896-9598-1-git-send-email-paul.durrant@citrix.com> X-Mailer: git-send-email 2.1.4 MIME-Version: 1.0 X-DLP: MIA1 Cc: Keir Fraser , Ian Campbell , Tim Deegan , Ian Jackson , Paul Durrant , Jan Beulich Subject: [Xen-devel] [PATCH v2] public/io/netif.h: make control ring hash protocol more general X-BeenThere: xen-devel@lists.xen.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP This patch modified the control ring protocol (of which there is not yet an implementation) to make it more general. Most of the concepts are not limited to toeplitz hashing so it's best not to make them unnecessarily specific. Apart from changing the names of various definitions and modifying comments, this patch: - Adds a new control message type to select a hash algorithm. - Adds a reference implementation of the toeplitz hash. - Changes the 'toeplitz' extra info fragment into a 'hash' extra info fragment and replaces the octet of padding with the index of the algorithm that was used to create the hash value. The patch also fixes a few spelling typos noticed along the way. Signed-off-by: Paul Durrant Cc: Ian Campbell Cc: Ian Jackson Cc: Jan Beulich Cc: Keir Fraser Cc: Tim Deegan --- v2: - Remove 'inline' from reference hash definition and make the definition optional --- xen/include/public/io/netif.h | 247 +++++++++++++++++++++++++++++------------- 1 file changed, 169 insertions(+), 78 deletions(-) diff --git a/xen/include/public/io/netif.h b/xen/include/public/io/netif.h index 8816e0f..14af483 100644 --- a/xen/include/public/io/netif.h +++ b/xen/include/public/io/netif.h @@ -191,8 +191,8 @@ */ /* - * Toeplitz hash types - * =================== + * Hash types + * ========== * * For the purposes of the definitions below, 'Packet[]' is an array of * octets containing an IP packet without options, 'Array[X..Y]' means a @@ -206,10 +206,11 @@ * Buffer[0..8] = Packet[12..15] (source address) + * Packet[16..19] (destination address) * - * Result = ToeplitzHash(Buffer, 8) + * Result = Hash(Buffer, 8) */ -#define _NETIF_CTRL_TOEPLITZ_HASH_IPV4 0 -#define NETIF_CTRL_TOEPLITZ_HASH_IPV4 (1 << _NETIF_CTRL_TOEPLITZ_HASH_IPV4) +#define _NETIF_CTRL_HASH_TYPE_IPV4 0 +#define NETIF_CTRL_HASH_TYPE_IPV4 \ + (1 << _NETIF_CTRL_HASH_TYPE_IPV4) /* * A hash calculated over an IP version 4 header and TCP header as @@ -220,10 +221,11 @@ * Packet[20..21] (source port) + * Packet[22..23] (destination port) * - * Result = ToeplitzHash(Buffer, 12) + * Result = Hash(Buffer, 12) */ -#define _NETIF_CTRL_TOEPLITZ_HASH_IPV4_TCP 1 -#define NETIF_CTRL_TOEPLITZ_HASH_IPV4_TCP (1 << _NETIF_CTRL_TOEPLITZ_HASH_IPV4_TCP) +#define _NETIF_CTRL_HASH_TYPE_IPV4_TCP 1 +#define NETIF_CTRL_HASH_TYPE_IPV4_TCP \ + (1 << _NETIF_CTRL_HASH_TYPE_IPV4_TCP) /* * A hash calculated over an IP version 6 header as follows: @@ -231,10 +233,11 @@ * Buffer[0..32] = Packet[8..23] (source address ) + * Packet[24..39] (destination address) * - * Result = ToeplitzHash(Buffer, 32) + * Result = Hash(Buffer, 32) */ -#define _NETIF_CTRL_TOEPLITZ_HASH_IPV6 2 -#define NETIF_CTRL_TOEPLITZ_HASH_IPV6 (1 << _NETIF_CTRL_TOEPLITZ_HASH_IPV4) +#define _NETIF_CTRL_HASH_TYPE_IPV6 2 +#define NETIF_CTRL_HASH_TYPE_IPV6 \ + (1 << _NETIF_CTRL_HASH_TYPE_IPV4) /* * A hash calculated over an IP version 6 header and TCP header as @@ -245,10 +248,80 @@ * Packet[40..41] (source port) + * Packet[42..43] (destination port) * - * Result = ToeplitzHash(Buffer, 36) + * Result = Hash(Buffer, 36) + */ +#define _NETIF_CTRL_HASH_TYPE_IPV6_TCP 3 +#define NETIF_CTRL_HASH_TYPE_IPV6_TCP \ + (1 << _NETIF_CTRL_HASH_TYPE_IPV4_TCP) + +/* + * Hash algorithms + * =============== + */ + +#define NETIF_CTRL_HASH_ALGORITHM_NONE 0 + +/* + * Toeplitz hash: */ -#define _NETIF_CTRL_TOEPLITZ_HASH_IPV6_TCP 3 -#define NETIF_CTRL_TOEPLITZ_HASH_IPV6_TCP (1 << _NETIF_CTRL_TOEPLITZ_HASH_IPV4_TCP) + +#define NETIF_CTRL_HASH_ALGORITHM_TOEPLITZ 1 + +/* + * This algorithm uses a 'key' as well as the data buffer itself. + * (Buffer[] and Key[] are treated as shift-registers where the MSB of + * Buffer/Key[0] is considered 'left-most' and the LSB of Buffer/Key[N-1] + * is the 'right-most'). + * + * Value = 0 + * For number of bits in Buffer[] + * If (left-most bit of Buffer[] is 1) + * Value ^= left-most 32 bits of Key[] + * Key[] << 1 + * Buffer[] << 1 + * + * The code below is provided for convenience where an operating system + * does not already provide an implementation. + */ +#ifdef NETIF_DEFINE_TOEPLITZ +static uint32_t netif_toeplitz_hash(const uint8_t *key, + unsigned int keylen, + const uint8_t *buf, + unsigned int buflen) +{ + unsigned int keyi, bufi; + uint64_t prefix = 0; + uint64_t hash = 0; + + /* Pre-load prefix with the first 8 bytes of the key */ + for (keyi = 0; keyi < 8; keyi++) { + prefix <<= 8; + prefix |= (keyi < keylen) ? key[keyi] : 0; + } + + for (bufi = 0; bufi < buflen; bufi++) { + uint8_t byte = buf[bufi]; + unsigned int bit; + + for (bit = 0; bit < 8; bit++) { + if (byte & 0x80) + hash ^= prefix; + prefix <<= 1; + byte <<=1; + } + + /* + * 'prefix' has now been left-shifted by 8, so + * OR in the next byte. + */ + prefix |= (keyi < keylen) ? key[keyi] : 0; + keyi++; + } + + /* The valid part of the hash is in the upper 32 bits. */ + return hash >> 32; +} +#endif /* NETIF_DEFINE_TOEPLITZ */ /* * Control requests (netif_ctrl_request_t) @@ -272,13 +345,14 @@ struct netif_ctrl_request { uint16_t id; uint16_t type; -#define NETIF_CTRL_TYPE_INVALID 0 -#define NETIF_CTRL_TYPE_GET_TOEPLITZ_FLAGS 1 -#define NETIF_CTRL_TYPE_SET_TOEPLITZ_FLAGS 2 -#define NETIF_CTRL_TYPE_SET_TOEPLITZ_KEY 3 -#define NETIF_CTRL_TYPE_GET_TOEPLITZ_MAPPING_ORDER 4 -#define NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING_ORDER 5 -#define NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING 6 +#define NETIF_CTRL_TYPE_INVALID 0 +#define NETIF_CTRL_TYPE_GET_HASH_FLAGS 1 +#define NETIF_CTRL_TYPE_SET_HASH_FLAGS 2 +#define NETIF_CTRL_TYPE_SET_HASH_KEY 3 +#define NETIF_CTRL_TYPE_GET_HASH_MAPPING_ORDER 4 +#define NETIF_CTRL_TYPE_SET_HASH_MAPPING_ORDER 5 +#define NETIF_CTRL_TYPE_SET_HASH_MAPPING 6 +#define NETIF_CTRL_TYPE_SET_HASH_ALGORITHM 7 uint32_t data[3]; }; @@ -322,15 +396,38 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * Control messages * ================ * - * NETIF_CTRL_TYPE_GET_TOEPLITZ_FLAGS + * NETIF_CTRL_TYPE_SET_HASH_ALGORITHM * ---------------------------------- * - * This is sent by the frontend to query the types of toeplitz - * hash supported by the backend. + * This is sent by the frontend to set the desired hash algorithm. * * Request: * - * type = NETIF_CTRL_TYPE_GET_TOEPLITZ_FLAGS + * type = NETIF_CTRL_TYPE_SET_HASH_ALGORITHM + * data[0] = a NETIF_CTRL_HASH_ALGORITHM_* value + * data[1] = 0 + * data[2] = 0 + * + * Response: + * + * status = NETIF_CTRL_STATUS_NOT_SUPPORTED - Operation not supported + * NETIF_CTRL_STATUS_INVALID_PARAMETER - The algorithm is not + * supported + * NETIF_CTRL_STATUS_SUCCESS - Operation successful + * + * NOTE: Setting data[0] to NETIF_CTRL_HASH_ALGORITHM_INVALID disables + * hashing and the backend is free to choose how it steers packets to + * queues (which is the default behaviour). + * + * NETIF_CTRL_TYPE_GET_HASH_FLAGS + * ------------------------------ + * + * This is sent by the frontend to query the types of hash supported by + * the backend. + * + * Request: + * + * type = NETIF_CTRL_TYPE_GET_HASH_FLAGS * data[0] = 0 * data[1] = 0 * data[2] = 0 @@ -341,11 +438,14 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * NETIF_CTRL_STATUS_SUCCESS - Operation successful * data = supported hash types (if operation was successful) * - * NETIF_CTRL_TYPE_SET_TOEPLITZ_FLAGS - * ---------------------------------- + * NOTE: A valid hash algorithm must be selected before this operation can + * succeed. * - * This is sent by the frontend to set the types of toeplitz hash that - * the backend should calculate. (See above for hash type definitions). + * NETIF_CTRL_TYPE_SET_HASH_FLAGS + * ------------------------------ + * + * This is sent by the frontend to set the types of hash that the backend + * should calculate. (See above for hash type definitions). * Note that the 'maximal' type of hash should always be chosen. For * example, if the frontend sets both IPV4 and IPV4_TCP hash types then * the latter hash type should be calculated for any TCP packet and the @@ -353,8 +453,8 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * * Request: * - * type = NETIF_CTRL_TYPE_SET_TOEPLITZ_FLAGS - * data[0] = bitwise OR of NETIF_CTRL_TOEPLITZ_HASH_* values + * type = NETIF_CTRL_TYPE_SET_HASH_FLAGS + * data[0] = bitwise OR of NETIF_CTRL_HASH_TYPE_* values * data[1] = 0 * data[2] = 0 * @@ -367,31 +467,20 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * NETIF_CTRL_STATUS_SUCCESS - Operation successful * data = 0 * - * NOTE: Setting data[0] to zero disables toeplitz hashing and the backend - * is free to choose how it steers packets to queues (which is the - * default behaviour). - * - * NETIF_CTRL_TYPE_SET_TOEPLITZ_KEY - * -------------------------------- - * - * This is sent by the frontend to set the key of the toeplitz hash that - * the backend should calculate. The toeplitz algorithm is illustrated - * by the following pseudo-code: + * NOTE: A valid hash algorithm must be selected before this operation can + * succeed. + * Also, setting data[0] to zero disables hashing and the backend + * is free to choose how it steers packets to queues. * - * (Buffer[] and Key[] are treated as shift-registers where the MSB of - * Buffer/Key[0] is considered 'left-most' and the LSB of Buffer/Key[N-1] - * is the 'right-most'). + * NETIF_CTRL_TYPE_SET_HASH_KEY + * ---------------------------- * - * Value = 0 - * For number of bits in Buffer[] - * If (left-most bit of Buffer[] is 1) - * Value ^= left-most 32 bits of Key[] - * Key[] << 1 - * Buffer[] << 1 + * This is sent by the frontend to set the key of the hash if the algorithm + * requires it. (See hash algorithms above). * * Request: * - * type = NETIF_CTRL_TYPE_SET_TOEPLITZ_KEY + * type = NETIF_CTRL_TYPE_SET_HASH_KEY * data[0] = grant reference of page containing the key (assumed to * start at beginning of grant) * data[1] = size of key in octets @@ -411,13 +500,13 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * invalidates any previous key, hence specifying a key size of * zero will clear the key (which ensures that the calculated hash * will always be zero). - * The maximum size of key is backend specific, but is also limited - * by the single grant reference. + * The maximum size of key is algorithm and backend specific, but + * is also limited by the single grant reference. * The grant reference may be read-only and must remain valid until * the response has been processed. * - * NETIF_CTRL_TYPE_GET_TOEPLITZ_MAPPING_ORDER - * ------------------------------------------ + * NETIF_CTRL_TYPE_GET_HASH_MAPPING_ORDER + * -------------------------------------- * * This is sent by the frontend to query the maximum order of mapping * table supported by the backend. The order is specified in terms of @@ -425,7 +514,7 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * * Request: * - * type = NETIF_CTRL_TYPE_GET_TOEPLITZ_MAPPING_ORDER + * type = NETIF_CTRL_TYPE_GET_HASH_MAPPING_ORDER * data[0] = 0 * data[1] = 0 * data[2] = 0 @@ -436,8 +525,8 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * NETIF_CTRL_STATUS_SUCCESS - Operation successful * data = maximum order of mapping table (if operation was successful) * - * NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING_ORDER - * ------------------------------------------ + * NETIF_CTRL_TYPE_SET_HASH_MAPPING_ORDER + * -------------------------------------- * * This is sent by the frontend to set the actual order of the mapping * table to be used by the backend. The order is specified in terms of @@ -448,7 +537,7 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * * Request: * - * type = NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING_ORDER + * type = NETIF_CTRL_TYPE_SET_HASH_MAPPING_ORDER * data[0] = order of mapping table * data[1] = 0 * data[2] = 0 @@ -460,18 +549,18 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * NETIF_CTRL_STATUS_SUCCESS - Operation successful * data = 0 * - * NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING - * ------------------------------------ + * NETIF_CTRL_TYPE_SET_HASH_MAPPING + * -------------------------------- * * This is sent by the frontend to set the content of the table mapping - * toeplitz hash value to queue number. The backend should calculate the - * hash from the packet header, use it as an index into the table (modulo - * the size of the table) and then steer the packet to the queue number - * found at that index. + * hash value to queue number. The backend should calculate the hash from + * the packet header, use it as an index into the table (modulo the size + * of the table) and then steer the packet to the queue number found at + * that index. * * Request: * - * type = NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING + * type = NETIF_CTRL_TYPE_SET_HASH_MAPPING * data[0] = grant reference of page containing the mapping (sub-)table * (assumed to start at beginning of grant) * data[1] = size of (sub-)table in entries @@ -501,7 +590,7 @@ typedef struct netif_ctrl_response netif_ctrl_response_t; * +-----+-----+-----+-----+-----+-----+-----+-----+ * * where N == 2^order as specified by a - * NETIF_CTRL_TYPE_SET_TOEPLITZ_MAPPING_ORDER message and each + * NETIF_CTRL_TYPE_SET_HASH_MAPPING_ORDER message and each * mapping must specifies a queue between 0 and * "multi-queue-num-queues" (see above). * The backend may support a mapping table larger than can be @@ -531,7 +620,7 @@ DEFINE_RING_TYPES(netif_ctrl, struct netif_ctrl_request, struct netif_ctrl_respo * ... * Fragment N: netif_tx_request_t - (only if fragment N-1 flags include * NETTXF_more_data - flags on preceding - * extras are not relevent here) + * extras are not relevant here) * flags = 0 * size = fragment size * @@ -589,7 +678,7 @@ DEFINE_RING_TYPES(netif_ctrl, struct netif_ctrl_request, struct netif_ctrl_respo * ... * Fragment N: netif_rx_request_t - (only if fragment N-1 flags include * NETRXF_more_data - flags on preceding - * extras are not relevent here) + * extras are not relevant here) * flags = 0 * size = fragment size * @@ -631,7 +720,7 @@ DEFINE_RING_TYPES(netif_ctrl, struct netif_ctrl_request, struct netif_ctrl_respo * described below, extra info structures overlay the id field). * Instead it assumes that responses always appear in the same ring * slot as their corresponding request. Thus, to maintain - * compatibilty, backends must make sure this is the case. + * compatibility, backends must make sure this is the case. * * Extra Info * ========== @@ -696,21 +785,23 @@ DEFINE_RING_TYPES(netif_ctrl, struct netif_ctrl_request, struct netif_ctrl_respo * flags: XEN_NETIF_EXTRA_FLAG_* * addr: address to add/remove * - * XEN_NETIF_EXTRA_TYPE_TOEPLITZ: + * XEN_NETIF_EXTRA_TYPE_HASH: * * A backend that supports teoplitz hashing is assumed to accept * this type of extra info in transmit packets. - * A frontend that enables toeplitz hashing is assumed to accept + * A frontend that enables hashing is assumed to accept * this type of extra info in receive packets. * * 0 1 2 3 4 5 6 7 octet * +-----+-----+-----+-----+-----+-----+-----+-----+ - * |type |flags|htype| pad |LSB ---- value ---- MSB| + * |type |flags|htype| alg |LSB ---- value ---- MSB| * +-----+-----+-----+-----+-----+-----+-----+-----+ * - * type: Must be XEN_NETIF_EXTRA_TYPE_TOEPLITZ + * type: Must be XEN_NETIF_EXTRA_TYPE_HASH * flags: XEN_NETIF_EXTRA_FLAG_* - * htype: Hash type (one of _NETIF_CTRL_TOEPLITZ_HASH_* - see above) + * htype: Hash type (one of _NETIF_CTRL_HASH_TYPE_* - see above) + * alg: The algorithm used to calculate the hash (one of + * NETIF_CTRL_HASH_TYPE_ALGORITHM_* - see above) * value: Hash value */ @@ -745,7 +836,7 @@ typedef struct netif_tx_request netif_tx_request_t; #define XEN_NETIF_EXTRA_TYPE_GSO (1) /* u.gso */ #define XEN_NETIF_EXTRA_TYPE_MCAST_ADD (2) /* u.mcast */ #define XEN_NETIF_EXTRA_TYPE_MCAST_DEL (3) /* u.mcast */ -#define XEN_NETIF_EXTRA_TYPE_TOEPLITZ (4) /* u.toeplitz */ +#define XEN_NETIF_EXTRA_TYPE_HASH (4) /* u.hash */ #define XEN_NETIF_EXTRA_TYPE_MAX (5) /* netif_extra_info_t flags. */ @@ -776,9 +867,9 @@ struct netif_extra_info { } mcast; struct { uint8_t type; - uint8_t pad; + uint8_t algorithm; uint8_t value[4]; - } toeplitz; + } hash; uint16_t pad[3]; } u; };