From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by dpdk.org (Postfix) with ESMTP id 9D6F91B05 for ; Fri, 29 Mar 2019 21:16:37 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga003.fm.intel.com ([10.253.24.29]) by fmsmga102.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 29 Mar 2019 13:16:36 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.60,285,1549958400"; d="scan'208";a="145024768" Received: from irsmsx101.ger.corp.intel.com ([163.33.3.153]) by FMSMGA003.fm.intel.com with ESMTP; 29 Mar 2019 13:16:35 -0700 Received: from irsmsx108.ger.corp.intel.com ([169.254.11.7]) by IRSMSX101.ger.corp.intel.com ([169.254.1.230]) with mapi id 14.03.0415.000; Fri, 29 Mar 2019 20:16:34 +0000 From: "Dumitrescu, Cristian" To: "Rao, Nikhil" , "Singh, Jasvinder" CC: "Lu, Wenzhuo" , "Wu, Jingjing" , "Iremonger, Bernard" , "dev@dpdk.org" Thread-Topic: [PATCH v2] ethdev: support double precision RED queue weight Thread-Index: AQHUqQLP0bMcI3wpV062J4nN7ATwQ6YjhXEQ Date: Fri, 29 Mar 2019 20:16:34 +0000 Message-ID: <3EB4FA525960D640B5BDFFD6A3D891268E85FDD0@IRSMSX108.ger.corp.intel.com> References: <1543470882-145287-1-git-send-email-nikhil.rao@intel.com> <1547138133-54919-1-git-send-email-nikhil.rao@intel.com> In-Reply-To: <1547138133-54919-1-git-send-email-nikhil.rao@intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-titus-metadata-40: eyJDYXRlZ29yeUxhYmVscyI6IiIsIk1ldGFkYXRhIjp7Im5zIjoiaHR0cDpcL1wvd3d3LnRpdHVzLmNvbVwvbnNcL0ludGVsMyIsImlkIjoiY2ZlMGFkNTItY2VkOS00ZTE0LWEzNGEtMjA0NjVmZWVhNGYxIiwicHJvcHMiOlt7Im4iOiJDVFBDbGFzc2lmaWNhdGlvbiIsInZhbHMiOlt7InZhbHVlIjoiQ1RQX05UIn1dfV19LCJTdWJqZWN0TGFiZWxzIjpbXSwiVE1DVmVyc2lvbiI6IjE3LjEwLjE4MDQuNDkiLCJUcnVzdGVkTGFiZWxIYXNoIjoicXJSMWF5QXI3T3NIdEdUWTNxa05uN1RYSnJLeExuM0JMT2hjXC8zWDltK051dEZ1OWNoRVZTZWtIRUU1VE9FZjMifQ== x-ctpclassification: CTP_NT dlp-product: dlpe-windows dlp-version: 11.0.400.15 dlp-reaction: no-action x-originating-ip: [163.33.239.180] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2] ethdev: support double precision RED queue weight 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 20:16:38 -0000 > -----Original Message----- > From: Rao, Nikhil > Sent: Thursday, January 10, 2019 4:36 PM > To: Dumitrescu, Cristian ; Singh, Jasvinde= r > > Cc: Lu, Wenzhuo ; Wu, Jingjing > ; Iremonger, Bernard > ; dev@dpdk.org; Rao, Nikhil > > Subject: [PATCH v2] ethdev: support double precision RED queue weight >=20 > RED queue weight is currently specified as a negated log of 2. >=20 > Add support for RED queue weight to be specified in double precision > and TM capability flags for double precision and negated log2 > RED queue weight support. >=20 > Update the softnic PMD and testpmd for the new tm capability flags and > the struct rte_tm_red_params::wq_is_log2 flag. >=20 > Signed-off-by: Nikhil Rao > --- > lib/librte_ethdev/rte_tm.h | 40 > ++++++++++++++++++++++++++++++-- > app/test-pmd/cmdline_tm.c | 8 ++++++- > drivers/net/softnic/rte_eth_softnic_tm.c | 7 ++++-- > 3 files changed, 50 insertions(+), 5 deletions(-) >=20 > v2: > * added RTE_STD_C11 for anonymnous union > * testpmd: added cman_wred_wq_log2/wq_dp_supported to "show port > tm cap" cli command > * testpmd: set wq_is_log2 for WRED profiles added on testpmd cli >=20 > Cristian, > I assume the testpmd changes will be upstreamed via the dpdk-next-qos > tree. >=20 > diff --git a/lib/librte_ethdev/rte_tm.h b/lib/librte_ethdev/rte_tm.h > index c4a788b..4fad31b 100644 > --- a/lib/librte_ethdev/rte_tm.h > +++ b/lib/librte_ethdev/rte_tm.h > @@ -391,6 +391,22 @@ struct rte_tm_capabilities { > */ > int cman_wred_byte_mode_supported; >=20 > + /** Double precision RED queue weight support. When non-zero, > this > + * this parameter indicates that RED queue weight in double precision > + * format is supported. > + * @see struct rte_tm_red_params::wq_is_log2 > + * @see struct rte_tm_red_params::wq_dp > + */ > + int cman_wred_wq_dp_supported; > + > + /** Negated log2 RED queue weight support. When non-zero, this > + * parameter indicates that RED queue weight in negated log2 format > + * is supported. > + * @see struct rte_tm_red_params::wq_is_log2 > + * @see struct rte_tm_red_params::wq_log2 > + */ > + int cman_wred_wq_log2_supported; > + > /** Head drop algorithm support. When non-zero, this parameter > * indicates that there is at least one leaf node that supports the > head > * drop algorithm, which might not be true for all the leaf nodes. > @@ -839,8 +855,28 @@ struct rte_tm_red_params { > */ > uint16_t maxp_inv; >=20 > - /** Negated log2 of queue weight (wq), i.e. wq =3D 1 / (2 ^ wq_log2) > */ > - uint16_t wq_log2; > + /** When non-zero, RED queue weight (wq) is negated log2 of > queue > + * weight > + * > + * @see struct rte_tm_capabilities::cman_wred_wq_dp_supported > + * @see struct rte_tm_capabilities::cman_wred_wq_log2_supported > + */ > + int wq_is_log2; > + > + RTE_STD_C11 > + union { > + /** Double precision queue weight > + * > + * @see struct > rte_tm_capabilities::cman_wred_wq_dp_supported > + */ > + double wq_dp; > + /** Negated log2 of queue weight (wq), > + * i.e. wq =3D 1 / (2 ^ wq_log2) > + * > + * @see struct > rte_tm_capabilities::cman_wred_wq_log2_supported > + */ > + uint16_t wq_log2; > + }; > }; >=20 > /** > diff --git a/app/test-pmd/cmdline_tm.c b/app/test-pmd/cmdline_tm.c > index 1012084..afc1d63 100644 > --- a/app/test-pmd/cmdline_tm.c > +++ b/app/test-pmd/cmdline_tm.c > @@ -283,6 +283,10 @@ static void cmd_show_port_tm_cap_parsed(void > *parsed_result, > cap.sched_wfq_n_groups_max); > printf("cap.sched_wfq_weight_max %" PRIu32 "\n", > cap.sched_wfq_weight_max); > + printf("cap.cman_wred_wq_dp_supported %" PRId32 "\n", > + cap.cman_wred_wq_dp_supported); > + printf("cap.cman_wred_wq_log2_supported %" PRId32 "\n", > + cap.cman_wred_wq_log2_supported); > printf("cap.cman_head_drop_supported %" PRId32 "\n", > cap.cman_head_drop_supported); > printf("cap.cman_wred_context_n_max %" PRIu32 "\n", > @@ -1285,7 +1289,7 @@ static void > cmd_add_port_tm_node_wred_profile_parsed(void *parsed_result, > wp.red_params[color].max_th =3D res->max_th_g; > wp.red_params[color].maxp_inv =3D res->maxp_inv_g; > wp.red_params[color].wq_log2 =3D res->wq_log2_g; > - > + wp.red_params[color].wq_is_log2 =3D 1; >=20 > /* WRED Params (Yellow Color)*/ > color =3D RTE_TM_YELLOW; > @@ -1293,6 +1297,7 @@ static void > cmd_add_port_tm_node_wred_profile_parsed(void *parsed_result, > wp.red_params[color].max_th =3D res->max_th_y; > wp.red_params[color].maxp_inv =3D res->maxp_inv_y; > wp.red_params[color].wq_log2 =3D res->wq_log2_y; > + wp.red_params[color].wq_is_log2 =3D 1; >=20 > /* WRED Params (Red Color)*/ > color =3D RTE_TM_RED; > @@ -1300,6 +1305,7 @@ static void > cmd_add_port_tm_node_wred_profile_parsed(void *parsed_result, > wp.red_params[color].max_th =3D res->max_th_r; > wp.red_params[color].maxp_inv =3D res->maxp_inv_r; > wp.red_params[color].wq_log2 =3D res->wq_log2_r; > + wp.red_params[color].wq_is_log2 =3D 1; >=20 > ret =3D rte_tm_wred_profile_add(port_id, wred_profile_id, &wp, > &error); > if (ret !=3D 0) { > diff --git a/drivers/net/softnic/rte_eth_softnic_tm.c > b/drivers/net/softnic/rte_eth_softnic_tm.c > index baaafbe..e96ea8e 100644 > --- a/drivers/net/softnic/rte_eth_softnic_tm.c > +++ b/drivers/net/softnic/rte_eth_softnic_tm.c > @@ -469,7 +469,8 @@ struct softnic_tmgr_port * > .cman_wred_context_shared_n_max =3D 0, > .cman_wred_context_shared_n_nodes_per_context_max =3D 0, > .cman_wred_context_shared_n_contexts_per_node_max =3D 0, > - > + .cman_wred_wq_dp_supported =3D 0, > + .cman_wred_wq_log2_supported =3D WRED_SUPPORTED, > .mark_vlan_dei_supported =3D {0, 0, 0}, > .mark_ip_ecn_tcp_supported =3D {0, 0, 0}, > .mark_ip_ecn_sctp_supported =3D {0, 0, 0}, > @@ -1243,8 +1244,10 @@ struct softnic_tmgr_port * > for (color =3D RTE_TM_GREEN; color < RTE_TM_COLORS; color++) { > uint32_t min_th =3D profile->red_params[color].min_th; > uint32_t max_th =3D profile->red_params[color].max_th; > + int wq_is_log2 =3D profile->red_params[color].wq_is_log2; >=20 > - if (min_th > max_th || > + if (wq_is_log2 =3D=3D 0 || > + min_th > max_th || > max_th =3D=3D 0 || > min_th > UINT16_MAX || > max_th > UINT16_MAX) > -- > 1.8.3.1 Hi Nikhil, Code looks good to me, but there is a problem: you are introducing new API = without providing an implementation for it, and this is no longer allowed b= y the DPDK Tech Board [1][2]. Would it be possible to provide an implementation for this new API? A SW im= plementation in librte_sched/rte_red.[hc] or divers/net/softnic would be en= ough to tick this box. Thanks, Cristian [1] https://mails.dpdk.org/archives/dev/2018-December/122213.html [2] https://mails.dpdk.org/archives/dev/2018-November/118697.html 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 C4E19A05D3 for ; Fri, 29 Mar 2019 21:16:39 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id ED6912BD3; Fri, 29 Mar 2019 21:16:38 +0100 (CET) Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by dpdk.org (Postfix) with ESMTP id 9D6F91B05 for ; Fri, 29 Mar 2019 21:16:37 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga003.fm.intel.com ([10.253.24.29]) by fmsmga102.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 29 Mar 2019 13:16:36 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.60,285,1549958400"; d="scan'208";a="145024768" Received: from irsmsx101.ger.corp.intel.com ([163.33.3.153]) by FMSMGA003.fm.intel.com with ESMTP; 29 Mar 2019 13:16:35 -0700 Received: from irsmsx108.ger.corp.intel.com ([169.254.11.7]) by IRSMSX101.ger.corp.intel.com ([169.254.1.230]) with mapi id 14.03.0415.000; Fri, 29 Mar 2019 20:16:34 +0000 From: "Dumitrescu, Cristian" To: "Rao, Nikhil" , "Singh, Jasvinder" CC: "Lu, Wenzhuo" , "Wu, Jingjing" , "Iremonger, Bernard" , "dev@dpdk.org" Thread-Topic: [PATCH v2] ethdev: support double precision RED queue weight Thread-Index: AQHUqQLP0bMcI3wpV062J4nN7ATwQ6YjhXEQ Date: Fri, 29 Mar 2019 20:16:34 +0000 Message-ID: <3EB4FA525960D640B5BDFFD6A3D891268E85FDD0@IRSMSX108.ger.corp.intel.com> References: <1543470882-145287-1-git-send-email-nikhil.rao@intel.com> <1547138133-54919-1-git-send-email-nikhil.rao@intel.com> In-Reply-To: <1547138133-54919-1-git-send-email-nikhil.rao@intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-titus-metadata-40: eyJDYXRlZ29yeUxhYmVscyI6IiIsIk1ldGFkYXRhIjp7Im5zIjoiaHR0cDpcL1wvd3d3LnRpdHVzLmNvbVwvbnNcL0ludGVsMyIsImlkIjoiY2ZlMGFkNTItY2VkOS00ZTE0LWEzNGEtMjA0NjVmZWVhNGYxIiwicHJvcHMiOlt7Im4iOiJDVFBDbGFzc2lmaWNhdGlvbiIsInZhbHMiOlt7InZhbHVlIjoiQ1RQX05UIn1dfV19LCJTdWJqZWN0TGFiZWxzIjpbXSwiVE1DVmVyc2lvbiI6IjE3LjEwLjE4MDQuNDkiLCJUcnVzdGVkTGFiZWxIYXNoIjoicXJSMWF5QXI3T3NIdEdUWTNxa05uN1RYSnJLeExuM0JMT2hjXC8zWDltK051dEZ1OWNoRVZTZWtIRUU1VE9FZjMifQ== x-ctpclassification: CTP_NT dlp-product: dlpe-windows dlp-version: 11.0.400.15 dlp-reaction: no-action x-originating-ip: [163.33.239.180] Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2] ethdev: support double precision RED queue weight 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: <20190329201634.OMoPEH4MlzTLyTl6isI7F-3_hwGUOW9b4azs-Abcfm4@z> > -----Original Message----- > From: Rao, Nikhil > Sent: Thursday, January 10, 2019 4:36 PM > To: Dumitrescu, Cristian ; Singh, Jasvinde= r > > Cc: Lu, Wenzhuo ; Wu, Jingjing > ; Iremonger, Bernard > ; dev@dpdk.org; Rao, Nikhil > > Subject: [PATCH v2] ethdev: support double precision RED queue weight >=20 > RED queue weight is currently specified as a negated log of 2. >=20 > Add support for RED queue weight to be specified in double precision > and TM capability flags for double precision and negated log2 > RED queue weight support. >=20 > Update the softnic PMD and testpmd for the new tm capability flags and > the struct rte_tm_red_params::wq_is_log2 flag. >=20 > Signed-off-by: Nikhil Rao > --- > lib/librte_ethdev/rte_tm.h | 40 > ++++++++++++++++++++++++++++++-- > app/test-pmd/cmdline_tm.c | 8 ++++++- > drivers/net/softnic/rte_eth_softnic_tm.c | 7 ++++-- > 3 files changed, 50 insertions(+), 5 deletions(-) >=20 > v2: > * added RTE_STD_C11 for anonymnous union > * testpmd: added cman_wred_wq_log2/wq_dp_supported to "show port > tm cap" cli command > * testpmd: set wq_is_log2 for WRED profiles added on testpmd cli >=20 > Cristian, > I assume the testpmd changes will be upstreamed via the dpdk-next-qos > tree. >=20 > diff --git a/lib/librte_ethdev/rte_tm.h b/lib/librte_ethdev/rte_tm.h > index c4a788b..4fad31b 100644 > --- a/lib/librte_ethdev/rte_tm.h > +++ b/lib/librte_ethdev/rte_tm.h > @@ -391,6 +391,22 @@ struct rte_tm_capabilities { > */ > int cman_wred_byte_mode_supported; >=20 > + /** Double precision RED queue weight support. When non-zero, > this > + * this parameter indicates that RED queue weight in double precision > + * format is supported. > + * @see struct rte_tm_red_params::wq_is_log2 > + * @see struct rte_tm_red_params::wq_dp > + */ > + int cman_wred_wq_dp_supported; > + > + /** Negated log2 RED queue weight support. When non-zero, this > + * parameter indicates that RED queue weight in negated log2 format > + * is supported. > + * @see struct rte_tm_red_params::wq_is_log2 > + * @see struct rte_tm_red_params::wq_log2 > + */ > + int cman_wred_wq_log2_supported; > + > /** Head drop algorithm support. When non-zero, this parameter > * indicates that there is at least one leaf node that supports the > head > * drop algorithm, which might not be true for all the leaf nodes. > @@ -839,8 +855,28 @@ struct rte_tm_red_params { > */ > uint16_t maxp_inv; >=20 > - /** Negated log2 of queue weight (wq), i.e. wq =3D 1 / (2 ^ wq_log2) > */ > - uint16_t wq_log2; > + /** When non-zero, RED queue weight (wq) is negated log2 of > queue > + * weight > + * > + * @see struct rte_tm_capabilities::cman_wred_wq_dp_supported > + * @see struct rte_tm_capabilities::cman_wred_wq_log2_supported > + */ > + int wq_is_log2; > + > + RTE_STD_C11 > + union { > + /** Double precision queue weight > + * > + * @see struct > rte_tm_capabilities::cman_wred_wq_dp_supported > + */ > + double wq_dp; > + /** Negated log2 of queue weight (wq), > + * i.e. wq =3D 1 / (2 ^ wq_log2) > + * > + * @see struct > rte_tm_capabilities::cman_wred_wq_log2_supported > + */ > + uint16_t wq_log2; > + }; > }; >=20 > /** > diff --git a/app/test-pmd/cmdline_tm.c b/app/test-pmd/cmdline_tm.c > index 1012084..afc1d63 100644 > --- a/app/test-pmd/cmdline_tm.c > +++ b/app/test-pmd/cmdline_tm.c > @@ -283,6 +283,10 @@ static void cmd_show_port_tm_cap_parsed(void > *parsed_result, > cap.sched_wfq_n_groups_max); > printf("cap.sched_wfq_weight_max %" PRIu32 "\n", > cap.sched_wfq_weight_max); > + printf("cap.cman_wred_wq_dp_supported %" PRId32 "\n", > + cap.cman_wred_wq_dp_supported); > + printf("cap.cman_wred_wq_log2_supported %" PRId32 "\n", > + cap.cman_wred_wq_log2_supported); > printf("cap.cman_head_drop_supported %" PRId32 "\n", > cap.cman_head_drop_supported); > printf("cap.cman_wred_context_n_max %" PRIu32 "\n", > @@ -1285,7 +1289,7 @@ static void > cmd_add_port_tm_node_wred_profile_parsed(void *parsed_result, > wp.red_params[color].max_th =3D res->max_th_g; > wp.red_params[color].maxp_inv =3D res->maxp_inv_g; > wp.red_params[color].wq_log2 =3D res->wq_log2_g; > - > + wp.red_params[color].wq_is_log2 =3D 1; >=20 > /* WRED Params (Yellow Color)*/ > color =3D RTE_TM_YELLOW; > @@ -1293,6 +1297,7 @@ static void > cmd_add_port_tm_node_wred_profile_parsed(void *parsed_result, > wp.red_params[color].max_th =3D res->max_th_y; > wp.red_params[color].maxp_inv =3D res->maxp_inv_y; > wp.red_params[color].wq_log2 =3D res->wq_log2_y; > + wp.red_params[color].wq_is_log2 =3D 1; >=20 > /* WRED Params (Red Color)*/ > color =3D RTE_TM_RED; > @@ -1300,6 +1305,7 @@ static void > cmd_add_port_tm_node_wred_profile_parsed(void *parsed_result, > wp.red_params[color].max_th =3D res->max_th_r; > wp.red_params[color].maxp_inv =3D res->maxp_inv_r; > wp.red_params[color].wq_log2 =3D res->wq_log2_r; > + wp.red_params[color].wq_is_log2 =3D 1; >=20 > ret =3D rte_tm_wred_profile_add(port_id, wred_profile_id, &wp, > &error); > if (ret !=3D 0) { > diff --git a/drivers/net/softnic/rte_eth_softnic_tm.c > b/drivers/net/softnic/rte_eth_softnic_tm.c > index baaafbe..e96ea8e 100644 > --- a/drivers/net/softnic/rte_eth_softnic_tm.c > +++ b/drivers/net/softnic/rte_eth_softnic_tm.c > @@ -469,7 +469,8 @@ struct softnic_tmgr_port * > .cman_wred_context_shared_n_max =3D 0, > .cman_wred_context_shared_n_nodes_per_context_max =3D 0, > .cman_wred_context_shared_n_contexts_per_node_max =3D 0, > - > + .cman_wred_wq_dp_supported =3D 0, > + .cman_wred_wq_log2_supported =3D WRED_SUPPORTED, > .mark_vlan_dei_supported =3D {0, 0, 0}, > .mark_ip_ecn_tcp_supported =3D {0, 0, 0}, > .mark_ip_ecn_sctp_supported =3D {0, 0, 0}, > @@ -1243,8 +1244,10 @@ struct softnic_tmgr_port * > for (color =3D RTE_TM_GREEN; color < RTE_TM_COLORS; color++) { > uint32_t min_th =3D profile->red_params[color].min_th; > uint32_t max_th =3D profile->red_params[color].max_th; > + int wq_is_log2 =3D profile->red_params[color].wq_is_log2; >=20 > - if (min_th > max_th || > + if (wq_is_log2 =3D=3D 0 || > + min_th > max_th || > max_th =3D=3D 0 || > min_th > UINT16_MAX || > max_th > UINT16_MAX) > -- > 1.8.3.1 Hi Nikhil, Code looks good to me, but there is a problem: you are introducing new API = without providing an implementation for it, and this is no longer allowed b= y the DPDK Tech Board [1][2]. Would it be possible to provide an implementation for this new API? A SW im= plementation in librte_sched/rte_red.[hc] or divers/net/softnic would be en= ough to tick this box. Thanks, Cristian [1] https://mails.dpdk.org/archives/dev/2018-December/122213.html [2] https://mails.dpdk.org/archives/dev/2018-November/118697.html