[vpp-dev] ethernet-input on master branch
Hi Damjan, I noticed you removed the quick path from dpdk-input to ip-input/mpls-input, after you merged the patch of ethernet-input optimization. Therefore, all packets now have to go through ethernet-input. It would take a few more cpu clocks than before. Please elaborate why making this change. Regards, Kingwel -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11843): https://lists.fd.io/g/vpp-dev/message/11843 Mute This Topic: https://lists.fd.io/mt/28940805/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] Question regarding captive portal
Hi Matus, Thanks for the info. But I am not able to get it working. tcp 80 packets are not reaching the http server on tap-interface. Only way it's working is if I rewrite DNS to point all domains requests to 192.168.2.2. I am running nat-plugin in endpoint-dependent mode. All tcp 80 packets from clients connected to GigabitEthernet4/0/0.4 should be sent to tap4. Here are my configs: startup.conf: unix { nodaemon log /var/log/vpp/vpp.log full-coredump cli-listen /run/vpp/cli.sock gid vpp startup-config /home/test/vpp.conf } api-trace { on } api-segment { gid vpp } cpu { main-core 2 corelist-workers 3 } nat { endpoint-dependent } dpdk { num-mbufs 32768 uio-driver auto } vpp.conf: set int state GigabitEthernet2/0/0 up ... ... set int state GigabitEthernet4/0/0 up ... ... create loopback interface instance 4 set int l2 bridge loop4 4 bvi set int ip address loop4 192.168.2.1/24 set int state loop4 up create sub GigabitEthernet4/0/0 4 set int l2 bridge GigabitEthernet4/0/0.4 4 set int l2 tag-rewrite GigabitEthernet4/0/0.4 pop 1 set int state GigabitEthernet4/0/0.4 up create tap id 4 host-ip4-addr 192.168.2.2/24 host-if-name guest set int l2 bridge tap4 4 set int state tap4 up set interface nat44 in tap4 out GigabitEthernet4/0/0.4 nat44 add static mapping tcp local 192.168.2.2 80 external GigabitEthernet4/0/0.4 80 out2in-only nat44 add interface address GigabitEthernet2/0/0 set interface nat44 in loop1 in loop2 in loop3 in loop4 out GigabitEthernet2/0/0 Thank you and Happy New Year! On Tue, Jan 1, 2019 at 10:23 PM Matus Fabian -X (matfabia - PANTHEON TECHNOLOGIES at Cisco) wrote: > > Hi, > > > > Run NAT plugin in endpoint-dependent mode (add following to startup config > “nat { endpoint-dependent }”), enable NAT feature “set interface nat44 in > tap3 out GigabitEthernet4/0/0” and create static mapping “nat44 add static > mapping tcp local 192.168.1.2 80 external GigabitEthernet4/0/0 80 > out2in-only”. > > > > Matus > > > > > > From: vpp-dev@lists.fd.io On Behalf Of carlito nueno > Sent: Friday, December 28, 2018 10:52 PM > To: vpp-dev@lists.fd.io > Subject: Re: [vpp-dev] Question regarding captive portal > > > > NAT might be the right way to achieve this. > > This is the command I used with iptables: > iptables -t nat -A eth0 -p tcp --dport 80 -j DNAT --to-destination 192.168.1.2 > > What is a similar command on VPP-NAT when I am trying to send port 80 traffic > from main interface to tap-device: > main interface: GigabitEthernet4/0/0 > tap id: 3 (tap3) with address 192.168.1.2 and host-if-name tapcap > > Thanks -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11842): https://lists.fd.io/g/vpp-dev/message/11842 Mute This Topic: https://lists.fd.io/mt/28506160/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] graph node placement on workers
After some examination, we believe the (apparently) random part is assignment of VirtualEthernet rx (not tx) queues to VPP workers. It is basically this sequence of VAT commands: create_vhost_user_if socket /tmp/sock-1-1 sw_interface_dump sw_interface_dump create_vhost_user_if socket /tmp/sock-1-2 sw_interface_dump sw_interface_dump sw_interface_set_flags sw_if_index 4 admin-up link-up sw_interface_set_flags sw_if_index 5 admin-up link-up If confirmed, we can make sure that CSIT code will move the rx queues to where we want them. Vratko. From: vpp-dev@lists.fd.io On Behalf Of Vratko Polak -X (vrpolak - PANTHEON TECHNOLOGIES at Cisco) via Lists.Fd.Io Sent: Thursday, 2019-January-03 18:13 To: vpp-dev@lists.fd.io Cc: vpp-dev@lists.fd.io Subject: [vpp-dev] graph node placement on workers Hi developers. When examining some performance test results, I started seeing patterns, so I now have multiple questions. I have noticed that (with hyper-threading on) l2-input-vtr graph node is present only on half of VPP workers. (At least for the dot1q test below.) In CSIT, the node is always seen on low numbered logical cores (lcore 2, as opposed to lcore 58). Is that expected? Also, some graph nodes ({if}-output and {if}-tx) tend to be only on some workers (which makes sense) determined apparently by random chance (which makes less sense). I believe this (apparent) randomness affects results of trending. I have two examples, both are scatter-plots on this [0] chart. The first example is the dot1q-l2bdbasemaclrn-eth-2vhostvr1024-1vm test. This test uses asymmetric load (encap one way, decap the other). There is 100% match between performance (whether the dot lands in the 6Mpps band, or 5Mpps band) and which TenGigabitEthernet tx node ends up on vpp_wk_0 (lcore 2). Due to asymmetric load, I believe one subgraph has more work to do, and it is random whether the more busy subgraph ends up on a worker also burdened with l2-input-vtr. The second example is the eth-l2xcbase-eth-4vhostvr1024-2vm test. Here the load is symmetric, but I see randomness in placement of the four VirtualEthernet tx nodes. I have not examined many runs, but I suspect the performance depends on which pair of the four end up on the same node. Finally, few less coherent questions. Is "worker handoff" involved? Should l2-input-vtr node stick to the subgraph which needs it? Can we (CSIT) give VPP a hint on which VirtualEthernet tx node should be handled by which worker? Vratko. [0] https://docs.fd.io/csit/master/trending/trending/vm_vhost_l2-2n-skx-x710-64b-base.html#t1c -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11841): https://lists.fd.io/g/vpp-dev/message/11841 Mute This Topic: https://lists.fd.io/mt/28925781/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] Regarding vpp core health monitoring
Hi, > "Health" I mean to check if the VPP worker cores are processing packets as in > ideal case and these cores are not Stuck (my code is not in infinite loop) > and also there is no mbuf leak. > Also How about master core if its not processing packets? Would the > keep-alive from client to VPP make sure at client side that master core is > intact and performing well as expected? There is a range of things you could do. Depending. You can monitor the counters from the stat segment, ensuring those are increasing to your expectations. Including interface counters, per-feature counters. And I was thinking of adding free mbuf counters… Or you could run send BFD packets, or some other specially constructed packets from an external process verifying that your particular network function behaves. Your mileage may wary. Cheers, Ole-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11840): https://lists.fd.io/g/vpp-dev/message/11840 Mute This Topic: https://lists.fd.io/mt/28932607/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] Regarding vpp core health monitoring
Hi Damjan, "Health" I mean to check if the VPP worker cores are processing packets as in ideal case and these cores are not Stuck (my code is not in infinite loop) and also there is no mbuf leak. Also How about master core if its not processing packets? Would the keep-alive from client to VPP make sure at client side that master core is intact and performing well as expected? Thanks, Kathir On Fri, Jan 4, 2019 at 6:39 PM Damjan Marion wrote: > > > On Jan 4, 2019, at 10:45 AM, Kathiresan Jayabalan > wrote: > > > > Hi Team, > > > > How can an external process monitor the health of vpp cores? Basically > we want to periodically check the health of vpp cores and eventually use it > for high availability. > > Please define “health” > > Thanks, > > — > Damjan > -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11839): https://lists.fd.io/g/vpp-dev/message/11839 Mute This Topic: https://lists.fd.io/mt/28932607/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] Regarding vpp core health monitoring
> On Jan 4, 2019, at 10:45 AM, Kathiresan Jayabalan wrote: > > Hi Team, > > How can an external process monitor the health of vpp cores? Basically we > want to periodically check the health of vpp cores and eventually use it for > high availability. Please define “health” Thanks, — Damjan -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11838): https://lists.fd.io/g/vpp-dev/message/11838 Mute This Topic: https://lists.fd.io/mt/28932607/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] I have back-port changes from vpp18.10 to vpp.18.01 (0610039fd06c760924fb92d0fc7b4d3e0ffeb8e3)
The issue probably has little-to-nothing directly to do with the patch you backported. I’d suggest trying to repro with a debug image. I suspect that the node counters haven’t been vec_validated properly. To level-set: we do not support 18.01 at this point, especially not a cherrypicked tree... HTH... Dave From: chetan bhasin Sent: Friday, January 4, 2019 3:53 AM To: vpp-dev@lists.fd.io; Dave Barach (dbarach) Subject: I have back-port changes from vpp18.10 to vpp.18.01 (0610039fd06c760924fb92d0fc7b4d3e0ffeb8e3) Hi Dave, I have back-merged changes committed in VPP18.10 to 18.01. Change details is as given below - commit 0610039fd06c760924fb92d0fc7b4d3e0ffeb8e3 Author: Dave Barach mailto:dbar...@cisco.com>> Date: Thu Jul 12 13:00:47 2018 -0400 Now we are facing a crash during init time , it looks to me if we fire vppctl while VPP is coming is causing the below crash, It is not always re-producible. Do you have any idea about this issue? Crashing at (File src/vlib/main.c) 569 void 570 vlib_node_sync_stats (vlib_main_t * vm, vlib_node_t * n) 571 { 572 vlib_node_runtime_t *rt; 573 574 if (n->type == VLIB_NODE_TYPE_PROCESS) 575 { 576 /* Nothing to do for PROCESS nodes except in main thread */ 577 if (vm != _global_main) 578 return; 579 580 vlib_process_t *p = vlib_get_process_from_node (vm, n); 581 n->stats_total.suspends += p->n_suspends; 582 p->n_suspends = 0; 583 rt = >node_runtime; 584 } 585 else 586 rt = 587 vec_elt_at_index (vm->node_main.nodes_by_type[n->type], 588 n->runtime_index); 589 590 vlib_node_runtime_sync_stats (vm, rt, 0, 0, 0); 591 592 /* Sync up runtime next frame vector counters with main node structure. */ 593 { 594 vlib_next_frame_t *nf; 595 uword i; 596 for (i = 0; i < rt->n_next_nodes; i++) 597 { 598 nf = vlib_node_runtime_get_next_frame (vm, rt, i); 599 vec_elt (n->n_vectors_by_next_node, i) += 600 nf->vectors_since_last_overflow; 601 nf->vectors_since_last_overflow = 0; 602 } 603 } 604 } Back Trace is as below - (gdb) info thr Id Target Id Frame 9Thread 0x2b7d72188700 (LWP 63125) 0x2b7825278eed in nanosleep () from /lib64/libpthread.so.0 8Thread 0x2b7d71381700 (LWP 63105) 0x2b7825789113 in epoll_wait () from /lib64/libc.so.6 7Thread 0x2b7d71f87700 (LWP 63111) vlib_worker_thread_barrier_check () /src/vlib/threads.h:403 6Thread 0x2b7d71d86700 (LWP 63110) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 5Thread 0x2b7d71b85700 (LWP 63109) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 4Thread 0x2b7d71984700 (LWP 63108) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 3Thread 0x2b7d71783700 (LWP 63107) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 2Thread 0x2b7d71582700 (LWP 63106) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 * 1Thread 0x2b78231439c0 (LWP 62940) 0x2b78256c0207 in raise () from /lib64/libc.so.6 #0 0x2b78256c0207 in raise () from /lib64/libc.so.6 #1 0x2b78256c18f8 in abort () from /lib64/libc.so.6 #2 0x00405ea9 in os_panic () src/vpp/vnet/main.c:266 #3 0x2b78237d2ad9 in unix_signal_handler (signum=, si=, uc=) src/vlib/unix/main.c:126 #4 #5 0x2b782379ab97 in vlib_node_sync_stats (vm=vm@entry=0x2b78239ed240 , n=0x2b7826196468) src/vlib/main.c:599 #6 0x2b78237bda3a in worker_thread_node_runtime_update_internal () src/vlib/threads.c:1046 #7 vlib_worker_thread_barrier_release (vm=vm@entry=0x2b78239ed240 ) src/vlib/threads.c:1528 #8 0x2b78237c8463 in unix_cli_file_add (name=, name@entry=0x2b7827082494 "local:10", fd=49, cm=0x2b78239ed0e0 , cm=0x2b78239ed0e0 ) src/vlib/unix/cli.c:2607 #9 0x2b78237cd373 in unix_cli_listen_read_ready (uf=) src/vlib/unix/cli.c:2659 #10 0x2b78237d21f0 in linux_epoll_input (vm=, node=, frame=) src/vlib/unix/input.c:203 #11 0x2b782379bc16 in dispatch_node (last_time_stamp=12704185179896535, frame=0x0, dispatch_state=VLIB_NODE_STATE_POLLING, type=VLIB_NODE_TYPE_PRE_INPUT, node=0x2b78288142c0, vm=0x2b78239ed240 ) src/vlib/main.c:988 #12 vlib_main_or_worker_loop (is_main=1, vm=0x2b78239ed240 ) src/vlib/main.c:1498 #13 vlib_main_loop (vm=0x2b78239ed240 ) src/vlib/main.c:1628 #14 vlib_main (vm=vm@entry=0x2b78239ed240 , input=input@entry=0x2b782858afa0) src/vlib/main.c:1783 #15 0x2b78237d2c13 in thread0 (arg=47794993680960) src/vlib/unix/main.c:548 #16 0x2b7824b07988 in clib_calljmp () Thanks, Chetan Bhasin -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11837): https://lists.fd.io/g/vpp-dev/message/11837
[vpp-dev] Regarding vpp core health monitoring
Hi Team, How can an external process monitor the health of vpp cores? Basically we want to periodically check the health of vpp cores and eventually use it for high availability. Thanks, Kathir -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11836): https://lists.fd.io/g/vpp-dev/message/11836 Mute This Topic: https://lists.fd.io/mt/28932607/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[vpp-dev] Regarding vpp core health monitoring
Hi Team, How can an external process monitor the health of vpp cores? Basically we want to periodically check the health of vpp cores and eventually use it for high availability. Thanks, Kathir -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11835): https://lists.fd.io/g/vpp-dev/message/11835 Mute This Topic: https://lists.fd.io/mt/28932607/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] Python Tests
Hi Paul, I believe it's okay to just add people as reviewers afterwards. Thanks, Klement Quoting Paul Vinciguerra (2019-01-03 18:16:42) >I am working on a change to cleanup some anti-patterns in the test >framework code, but the scope is significantly larger. >[1]https://gerrit.fd.io/r/#/c/16642/ > >Specifically, I'm referring to the use of: > except: >and > raise Exception(...) > >This is not in any way a dig against any of the contributors. The pattern >is recommended in the docs for people to follow. > >Does anyone have any suggestions as to how to best involve the original >contributors without stalling progress? >Is adding reviewers from git blame sufficient, or do the original >contributors prefer to know earlier in the process? > >Here are some links if anyone is interested: >[2]https://julien.danjou.info/python-exceptions-guide/ >[3]https://hynek.me/articles/hasattr/ > > References > >Visible links >1. https://gerrit.fd.io/r/#/c/16642/ > https://gerrit.fd.io/r/#/c/16642/ >2. https://julien.danjou.info/python-exceptions-guide/ >3. https://hynek.me/articles/hasattr/ -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11834): https://lists.fd.io/g/vpp-dev/message/11834 Mute This Topic: https://lists.fd.io/mt/28925838/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
[vpp-dev] I have back-port changes from vpp18.10 to vpp.18.01 (0610039fd06c760924fb92d0fc7b4d3e0ffeb8e3)
Hi Dave, I have back-merged changes committed in VPP18.10 to 18.01. Change details is as given below - commit 0610039fd06c760924fb92d0fc7b4d3e0ffeb8e3 Author: Dave Barach Date: Thu Jul 12 13:00:47 2018 -0400 Now we are facing a crash during init time , it looks to me if we fire vppctl while VPP is coming is causing the below crash, It is not always re-producible. Do you have any idea about this issue? *Crashing at* (File *src/vlib/main.c)* 569 void 570 vlib_node_sync_stats (vlib_main_t * vm, vlib_node_t * n) 571 { 572 vlib_node_runtime_t *rt; 573 574 if (n->type == VLIB_NODE_TYPE_PROCESS) 575 { 576 /* Nothing to do for PROCESS nodes except in main thread */ 577 if (vm != _global_main) 578 return; 579 580 vlib_process_t *p = vlib_get_process_from_node (vm, n); 581 n->stats_total.suspends += p->n_suspends; 582 p->n_suspends = 0; 583 rt = >node_runtime; 584 } 585 else 586 rt = 587 vec_elt_at_index (vm->node_main.nodes_by_type[n->type], 588 n->runtime_index); 589 590 vlib_node_runtime_sync_stats (vm, rt, 0, 0, 0); 591 592 /* Sync up runtime next frame vector counters with main node structure. */ 593 { 594 vlib_next_frame_t *nf; 595 uword i; 596 for (i = 0; i < rt->n_next_nodes; i++) 597 { 598 nf = vlib_node_runtime_get_next_frame (vm, rt, i); *599 vec_elt (n->n_vectors_by_next_node, i) +=600 nf->vectors_since_last_overflow;* 601 nf->vectors_since_last_overflow = 0; 602 } 603 } 604 } Back Trace is as below - (gdb) info thr Id Target Id Frame 9Thread 0x2b7d72188700 (LWP 63125) 0x2b7825278eed in nanosleep () from /lib64/libpthread.so.0 8Thread 0x2b7d71381700 (LWP 63105) 0x2b7825789113 in epoll_wait () from /lib64/libc.so.6 7Thread 0x2b7d71f87700 (LWP 63111) vlib_worker_thread_barrier_check () /src/vlib/threads.h:403 6Thread 0x2b7d71d86700 (LWP 63110) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 5Thread 0x2b7d71b85700 (LWP 63109) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 4Thread 0x2b7d71984700 (LWP 63108) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 3Thread 0x2b7d71783700 (LWP 63107) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 2Thread 0x2b7d71582700 (LWP 63106) vlib_worker_thread_barrier_check () src/vlib/threads.h:403 * 1Thread 0x2b78231439c0 (LWP 62940) 0x2b78256c0207 in raise () from /lib64/libc.so.6 #0 0x2b78256c0207 in raise () from /lib64/libc.so.6 #1 0x2b78256c18f8 in abort () from /lib64/libc.so.6 #2 0x00405ea9 in os_panic () src/vpp/vnet/main.c:266 #3 0x2b78237d2ad9 in unix_signal_handler (signum=, si=, uc=) src/vlib/unix/main.c:126 #4 *#5 0x2b782379ab97 in vlib_node_sync_stats (vm=vm@entry=0x2b78239ed240 , n=0x2b7826196468)src/vlib/main.c:599#6 0x2b78237bda3a in worker_thread_node_runtime_update_internal () src/vlib/threads.c:1046* *#7 vlib_worker_thread_barrier_release (vm=vm@entry=0x2b78239ed240 )src/vlib/threads.c:1528* #8 0x2b78237c8463 in unix_cli_file_add (name=, name@entry=0x2b7827082494 "local:10", fd=49, cm=0x2b78239ed0e0 , cm=0x2b78239ed0e0 ) src/vlib/unix/cli.c:2607 #9 0x2b78237cd373 in unix_cli_listen_read_ready (uf=) src/vlib/unix/cli.c:2659 #10 0x2b78237d21f0 in linux_epoll_input (vm=, node=, frame=) src/vlib/unix/input.c:203 #11 0x2b782379bc16 in dispatch_node (last_time_stamp=12704185179896535, frame=0x0, dispatch_state=VLIB_NODE_STATE_POLLING, type=VLIB_NODE_TYPE_PRE_INPUT, node=0x2b78288142c0, vm=0x2b78239ed240 ) src/vlib/main.c:988 #12 vlib_main_or_worker_loop (is_main=1, vm=0x2b78239ed240 ) src/vlib/main.c:1498 #13 vlib_main_loop (vm=0x2b78239ed240 ) src/vlib/main.c:1628 #14 vlib_main (vm=vm@entry=0x2b78239ed240 , input=input@entry=0x2b782858afa0) src/vlib/main.c:1783 #15 0x2b78237d2c13 in thread0 (arg=47794993680960) src/vlib/unix/main.c:548 #16 0x2b7824b07988 in clib_calljmp () Thanks, Chetan Bhasin -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11833): https://lists.fd.io/g/vpp-dev/message/11833 Mute This Topic: https://lists.fd.io/mt/28932141/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] VPP/DPDK performance with Madvise (Transparent Huge pages)
Sorry, cannot help much wit details you provided... — Damjan > On Jan 4, 2019, at 9:33 AM, chetan bhasin wrote: > > Hi Damjan, > > I want to ensure that VPP uses THP and other Linux applications don't use THP > feature. > > Thanks, > Chetan Bhasin > >> On Thu, Jan 3, 2019 at 4:46 PM Damjan Marion wrote: >> >> >>> On 2 Jan 2019, at 16:28, chetan bhasin wrote: >>> >>> Hi, >>> >>> We are using VPP 18.01. We have seen that in case Transparent Hugepage is >>> enable with Madvise . VPP does not take benefit of Annonymous Huge Pages. >>> Does anybody have any idea about the same ? >> >> Can you provide more details, What exactly you did and what exactly you >> observed >> >> -- >> Damjan >> -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11832): https://lists.fd.io/g/vpp-dev/message/11832 Mute This Topic: https://lists.fd.io/mt/28914567/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
Re: [vpp-dev] VPP/DPDK performance with Madvise (Transparent Huge pages)
Hi Damjan, I want to ensure that VPP uses THP and other Linux applications don't use THP feature. Thanks, Chetan Bhasin On Thu, Jan 3, 2019 at 4:46 PM Damjan Marion wrote: > > > On 2 Jan 2019, at 16:28, chetan bhasin wrote: > > Hi, > > We are using VPP 18.01. We have seen that in case Transparent Hugepage is > enable with Madvise . VPP does not take benefit of Annonymous Huge Pages. > Does anybody have any idea about the same ? > > > Can you provide more details, What exactly you did and what exactly you > observed > > -- > Damjan > > -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11831): https://lists.fd.io/g/vpp-dev/message/11831 Mute This Topic: https://lists.fd.io/mt/28914567/21656 Group Owner: vpp-dev+ow...@lists.fd.io Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-