suggest dpdk transplant to window platform in windows platform 1G NIC only can use 10% line rate in small packet mode, in big concurrent TCP link environment, It's too slow, I have high concurrent task system(in windows), Performance is very good, It's can solve c10k problem, in windows, It's can reach 10 million TCP link , However iops only 100k, it's far from reaching 1G NIC's 1.4 million iops If dpdk can provide mul queue packet interface to Windows user mode , It will be exciting.
-----Original Message----- From: users [mailto:users-bounces at dpdk.org] On Behalf Of [email protected] Sent: Wednesday, June 8, 2016 6:00 PM To: users at dpdk.org Subject: users Digest, Vol 34, Issue 1 Send users mailing list submissions to users at dpdk.org To subscribe or unsubscribe via the World Wide Web, visit http://dpdk.org/ml/listinfo/users or, via email, send a message with subject or body 'help' to users-request at dpdk.org You can reach the person managing the list at users-owner at dpdk.org When replying, please edit your Subject line so it is more specific than "Re: Contents of users digest..." Today's Topics: 1. KNI app random mac generation (SAKTHIVEL ANAND S) 2. Re: FW: Looking for packet forwarding of real traffic (Jean Tourrilhes) 3. A question about the function __mempool_get_bulk() of dpdk-16.04 (Wu, Xiaoban) 4. Re: A question about the function __mempool_get_bulk() of dpdk-16.04 (Wiles, Keith) 5. Re: A question about the function __mempool_get_bulk() of dpdk-16.04 (Wu, Xiaoban) ---------------------------------------------------------------------- Message: 1 Date: Tue, 7 Jun 2016 19:38:20 +0530 From: SAKTHIVEL ANAND S <[email protected]> To: users at dpdk.org Subject: [dpdk-users] KNI app random mac generation Message-ID: <CAOP5GAzF0LO3jk6ANOdzGAonKuZLFviJeyydSvi0q-b87WbYYQ at mail.gmail.com> Content-Type: text/plain; charset=UTF-8 Hi I have recently downloaded 16.04, in this version also i could see KNI random mac generation behavior. Is there any plan to fix this or it is purposefully made like that? if i am wrongly understood this random generation, please explain its purpose. -- Thanks Sakthivel S OM ------------------------------ Message: 2 Date: Tue, 7 Jun 2016 11:35:20 -0700 From: Jean Tourrilhes <[email protected]> To: users at dpdk.org, anand.sa88 at gmail.com Subject: Re: [dpdk-users] FW: Looking for packet forwarding of real traffic Message-ID: <20160607183520.GB16956 at labs.hpe.com> Content-Type: text/plain; charset=us-ascii SAKTHIVEL ANAND S <anand.sa88 at gmail.com> wrote: > > you may try KNI app ,in the examples. > Thanks I may try that. Thanks ! Jean ------------------------------ Message: 3 Date: Tue, 7 Jun 2016 20:52:25 +0000 From: "Wu, Xiaoban" <[email protected]> To: "users at dpdk.org" <users at dpdk.org> Subject: [dpdk-users] A question about the function __mempool_get_bulk() of dpdk-16.04 Message-ID: <CY1PR0201MB0827E7B199732BB8BCDE4716BC5D0 at CY1PR0201MB0827.namprd02.prod.outlook.com> Content-Type: text/plain; charset="iso-8859-1" Dear DPDK Users, I have been reading the pktgen-3.0.02 source codes with dpdk-16.04. In the pktgen.c, function pktgen_send_pkts(), line 1098, it calls function wr_pktmbuf_alloc_bulk_noreset(). Then it calls the function rte_mempool_get_bulk() which calls the function __mempool_get_bulk(). Since in the function rte_pktmbuf_pool_create(), the "flags" is set to zero in default, this leads to that when calling __mempool_get_bulk(), the input parameter"is_mc==1". Hence the function __mempool_get_bulk() will execute the line 961-992 except that when "ret" is less that 0. My question is why at line 992, this has to return 0? In the line 941, the comment says that ">=0: Success; number of objects supplied." I am confused by this comment, since in line 985-990, we can see that the input parameter "obj_table" is supplied by new values and the number of the new values is "n". Does this means that we need to "return n;" instead of "return 0;"? Am I missing something here? Thank you so much for your help. All the best, Xiaoban ------------------------------ Message: 4 Date: Tue, 7 Jun 2016 22:30:24 +0000 From: "Wiles, Keith" <[email protected]> To: "Wu, Xiaoban" <Xiaoban_Wu at student.uml.edu>, "users at dpdk.org" <users at dpdk.org> Subject: Re: [dpdk-users] A question about the function __mempool_get_bulk() of dpdk-16.04 Message-ID: <506DDABB-6DFC-4FEB-8A8D-43B7AA2EE251 at intel.com> Content-Type: text/plain; charset="utf-8" On 6/7/16, 3:52 PM, "users on behalf of Wu, Xiaoban" <users-bounces at dpdk.org on behalf of Xiaoban_Wu at student.uml.edu> wrote: >Dear DPDK Users, > > >I have been reading the pktgen-3.0.02 source codes with dpdk-16.04. > > >In the pktgen.c, function pktgen_send_pkts(), line 1098, it calls function >wr_pktmbuf_alloc_bulk_noreset(). Then it calls the function >rte_mempool_get_bulk() which calls the function __mempool_get_bulk(). > > >Since in the function rte_pktmbuf_pool_create(), the "flags" is set to zero in >default, this leads to that when calling __mempool_get_bulk(), the input >parameter"is_mc==1". Hence the function __mempool_get_bulk() will execute the >line 961-992 except that when "ret" is less that 0. I checked out dpdk v16.04 and could not get your line numbers to workout, maybe you are on a slightly different version then release 16.04?? > > >My question is why at line 992, this has to return 0? In the line 941, the >comment says that ">=0: Success; number of objects supplied." I am confused >by this comment, since in line 985-990, we can see that the input parameter >"obj_table" is supplied by new values and the number of the new values is "n". >Does this means that we need to "return n;" instead of "return 0;"? The statement ?>=0 number of objects supplied? also confused me some, but routines rte_ring_mc/sc_dequeue_bulk() returns zero on success when all requested objects are return not the number of objects returned. It seems it would be nice to have the number of objects returned, but as it is always ?n? it is best to return < 0 on error. The _bulk allocate will not return a les then ?n? objects, this is why the code in pktgen tests for zero and not something greater then zero. The code in pktgen could test for ret >= 0 but the results are the same. > > >Am I missing something here? Thank you so much for your help. > > >All the best, > >Xiaoban > ------------------------------ Message: 5 Date: Wed, 8 Jun 2016 00:12:44 +0000 From: "Wu, Xiaoban" <[email protected]> To: "Wiles, Keith" <keith.wiles at intel.com>, "users at dpdk.org" <users at dpdk.org> Subject: Re: [dpdk-users] A question about the function __mempool_get_bulk() of dpdk-16.04 Message-ID: <CY1PR0201MB08272CA7F121D0E4C354D749BC5D0 at CY1PR0201MB0827.namprd02.prod.outlook.com> Content-Type: text/plain; charset="Windows-1252" Dear Keith, Yes, I see the checking for 0 in the pktgen codes. Maybe it is more clear if returns n like the wirte() in linux, but it also means a lot of codes need to be rewritten. As long as it is consistent and we know how to work around it, it should be fine. Thanks so much for your reply. All the best, Xiaoban ________________________________________ From: Wiles, Keith <[email protected]> Sent: Tuesday, June 7, 2016 6:30:24 PM To: Wu, Xiaoban; users at dpdk.org Subject: Re: [dpdk-users] A question about the function __mempool_get_bulk() of dpdk-16.04 On 6/7/16, 3:52 PM, "users on behalf of Wu, Xiaoban" <users-bounces at dpdk.org on behalf of Xiaoban_Wu at student.uml.edu> wrote: >Dear DPDK Users, > > >I have been reading the pktgen-3.0.02 source codes with dpdk-16.04. > > >In the pktgen.c, function pktgen_send_pkts(), line 1098, it calls function >wr_pktmbuf_alloc_bulk_noreset(). Then it calls the function >rte_mempool_get_bulk() which calls the function __mempool_get_bulk(). > > >Since in the function rte_pktmbuf_pool_create(), the "flags" is set to zero in >default, this leads to that when calling __mempool_get_bulk(), the input >parameter"is_mc==1". Hence the function __mempool_get_bulk() will execute the >line 961-992 except that when "ret" is less that 0. I checked out dpdk v16.04 and could not get your line numbers to workout, maybe you are on a slightly different version then release 16.04?? > > >My question is why at line 992, this has to return 0? In the line 941, the >comment says that ">=0: Success; number of objects supplied." I am confused >by this comment, since in line 985-990, we can see that the input parameter >"obj_table" is supplied by new values and the number of the new values is "n". >Does this means that we need to "return n;" instead of "return 0;"? The statement ?>=0 number of objects supplied? also confused me some, but routines rte_ring_mc/sc_dequeue_bulk() returns zero on success when all requested objects are return not the number of objects returned. It seems it would be nice to have the number of objects returned, but as it is always ?n? it is best to return < 0 on error. The _bulk allocate will not return a les then ?n? objects, this is why the code in pktgen tests for zero and not something greater then zero. The code in pktgen could test for ret >= 0 but the results are the same. > > >Am I missing something here? Thank you so much for your help. > > >All the best, > >Xiaoban > End of users Digest, Vol 34, Issue 1 ************************************
