From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-he1eur01on0086.outbound.protection.outlook.com ([104.47.0.86]:20096 "EHLO EUR01-HE1-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751290AbeCLNUQ (ORCPT ); Mon, 12 Mar 2018 09:20:16 -0400 Subject: Re: [bpf-next V3 PATCH 13/15] mlx5: use page_pool for xdp_return_frame call To: Tariq Toukan , Jesper Dangaard Brouer , netdev@vger.kernel.org, =?UTF-8?B?QmrDtnJuVMO2cGVs?= , magnus.karlsson@intel.com Cc: eugenia@mellanox.com, Jason Wang , John Fastabend , Eran Ben Elisha , Saeed Mahameed , galp@mellanox.com, Daniel Borkmann , Alexei Starovoitov References: <152062887576.27458.8590966896888512270.stgit@firesoul> <152062896782.27458.5542026179434739900.stgit@firesoul> <247192bd-761e-426c-c462-59efe3b7ca97@mellanox.com> <22866952-6bdf-2529-91a1-fb31bd2f2c2d@mellanox.com> From: Tariq Toukan Message-ID: <6e62fa6a-53c7-57a9-0493-3a48d832b479@mellanox.com> Date: Mon, 12 Mar 2018 15:20:06 +0200 MIME-Version: 1.0 In-Reply-To: <22866952-6bdf-2529-91a1-fb31bd2f2c2d@mellanox.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 8bit Sender: netdev-owner@vger.kernel.org List-ID: On 12/03/2018 12:16 PM, Tariq Toukan wrote: > > > On 12/03/2018 12:08 PM, Tariq Toukan wrote: >> >> >> On 09/03/2018 10:56 PM, Jesper Dangaard Brouer wrote: >>> This patch shows how it is possible to have both the driver local page >>> cache, which uses elevated refcnt for "catching"/avoiding SKB >>> put_page.  And at the same time, have pages getting returned to the >>> page_pool from ndp_xdp_xmit DMA completion. >>> >>> Performance is surprisingly good. Tested DMA-TX completion on ixgbe, >>> that calls "xdp_return_frame", which call page_pool_put_page(). >>> Stats show DMA-TX-completion runs on CPU#9 and mlx5 RX runs on CPU#5. >>> (Internally page_pool uses ptr_ring, which is what gives the good >>> cross CPU performance). >>> >>> Show adapter(s) (ixgbe2 mlx5p2) statistics (ONLY that changed!) >>> Ethtool(ixgbe2  ) stat:    732863573 (    732,863,573) <= tx_bytes /sec >>> Ethtool(ixgbe2  ) stat:    781724427 (    781,724,427) <= >>> tx_bytes_nic /sec >>> Ethtool(ixgbe2  ) stat:     12214393 (     12,214,393) <= tx_packets >>> /sec >>> Ethtool(ixgbe2  ) stat:     12214435 (     12,214,435) <= tx_pkts_nic >>> /sec >>> Ethtool(mlx5p2  ) stat:     12211786 (     12,211,786) <= >>> rx3_cache_empty /sec >>> Ethtool(mlx5p2  ) stat:     36506736 (     36,506,736) <= >>> rx_64_bytes_phy /sec >>> Ethtool(mlx5p2  ) stat:   2336430575 (  2,336,430,575) <= >>> rx_bytes_phy /sec >>> Ethtool(mlx5p2  ) stat:     12211786 (     12,211,786) <= >>> rx_cache_empty /sec >>> Ethtool(mlx5p2  ) stat:     22823073 (     22,823,073) <= >>> rx_discards_phy /sec >>> Ethtool(mlx5p2  ) stat:      1471860 (      1,471,860) <= >>> rx_out_of_buffer /sec >>> Ethtool(mlx5p2  ) stat:     36506715 (     36,506,715) <= >>> rx_packets_phy /sec >>> Ethtool(mlx5p2  ) stat:   2336542282 (  2,336,542,282) <= >>> rx_prio0_bytes /sec >>> Ethtool(mlx5p2  ) stat:     13683921 (     13,683,921) <= >>> rx_prio0_packets /sec >>> Ethtool(mlx5p2  ) stat:    821015537 (    821,015,537) <= >>> rx_vport_unicast_bytes /sec >>> Ethtool(mlx5p2  ) stat:     13683608 (     13,683,608) <= >>> rx_vport_unicast_packets /sec >>> >>> Before this patch: single flow performance was 6Mpps, and if I started >>> two flows the collective performance drop to 4Mpps, because we hit the >>> page allocator lock (further negative scaling occurs). >>> >>> V2: Adjustments requested by Tariq >>>   - Changed page_pool_create return codes not return NULL, only >>>     ERR_PTR, as this simplifies err handling in drivers. >>>   - Save a branch in mlx5e_page_release >>>   - Correct page_pool size calc for MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ >>> >>> Signed-off-by: Jesper Dangaard Brouer >>> --- >> >> I am running perf tests with your series. I sense a drastic >> degradation in regular TCP flows, I'm double checking the numbers now... > > Well, there's a huge performance degradation indeed, whenever the > regular flows (non-XDP) use the new page pool. Cannot merge before > fixing this. > > If I disable the local page-cache, numbers get as low as 100's of Mbps > in TCP stream tests. It seems that the page-pool doesn't fit as a general fallback (when page in local rx cache is busy), as the refcnt is elevated/changing: [ 7343.086102] ------------[ cut here ]------------ [ 7343.086103] __page_pool_put_page() violating page_pool invariance refcnt:0 [ 7343.086114] WARNING: CPU: 1 PID: 17 at net/core/page_pool.c:291 __page_pool_put_page+0x7c/0xa0 [ 7343.086114] Modules linked in: mlx5_core(OE) netconsole nfsv3 nfs fscache rpcrdma ib_isert iscsi_target_mod ib_iser libiscsi scsi_transport_iscsi ib_srpt target_core_mod ib_srp scsi_transport_srp ib_ipoib rdma_ucm ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm mlx4_ib dm_mirror ib_core dm_region_hash dm_log dm_mod dax sb_edac x86_pkg_temp_thermal coretemp kvm ipmi_si ipmi_devintf iTCO_wdt irqbypass crc32_pclmul iTCO_vendor_support ipmi_msghandler ghash_clmulni_intel dcdbas acpi_power_meter sg wmi pcspkr lpc_ich mfd_core shpchp nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables mlx4_en sr_mod cdrom sd_mod mgag200 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm ahci drm mlx4_core libahci libata crc32c_intel megaraid_sas tg3 i2c_core [last unloaded: mlx5_core] [ 7343.086137] CPU: 1 PID: 17 Comm: ksoftirqd/1 Tainted: G W OE 4.16.0-rc4+ #7 [ 7343.086138] Hardware name: Dell Inc. PowerEdge R730/0H21J3, BIOS 1.5.4 10/002/2015 [ 7343.086139] RIP: 0010:__page_pool_put_page+0x7c/0xa0 [ 7343.086140] RSP: 0018:ffffc9000653fcc8 EFLAGS: 00010292 [ 7343.086141] RAX: 000000000000003e RBX: ffffea0080d582c0 RCX: 0000000000000000 [ 7343.086141] RDX: 0000000000000000 RSI: 0000000000000002 RDI: 0000000000000246 [ 7343.086142] RBP: ffff882033ffe000 R08: 000000000000003e R09: ffffffff8282f8b6 [ 7343.086142] R10: 00000000000050ee R11: 0000000000000001 R12: ffff881fc857c000 [ 7343.086143] R13: ffff881fc49cc800 R14: 0000000000000040 R15: ffff881fc857c140 [ 7343.086143] FS: 0000000000000000(0000) GS:ffff88203f000000(0000) knlGS:0000000000000000 [ 7343.086144] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 7343.086145] CR2: 00007f0aee0dc8b0 CR3: 000000000200a006 CR4: 00000000001606e0 [ 7343.086156] Call Trace: [ 7343.086168] mlx5e_free_rx_mpwqe+0x58/0x70 [mlx5_core] [ 7343.086176] mlx5e_handle_rx_cqe_mpwrq+0x8cf/0x9b0 [mlx5_core] [ 7343.086183] mlx5e_poll_rx_cq+0xb8/0x890 [mlx5_core] [ 7343.086190] mlx5e_napi_poll+0x88/0x640 [mlx5_core] [ 7343.086192] net_rx_action+0x286/0x3d0 [ 7343.086194] __do_softirq+0xd0/0x282 [ 7343.086196] run_ksoftirqd+0x24/0x40 [ 7343.086198] smpboot_thread_fn+0xfe/0x150 [ 7343.086199] kthread+0xf5/0x130 [ 7343.086200] ? sort_range+0x20/0x20 [ 7343.086201] ? kthread_bind+0x10/0x10 [ 7343.086203] ret_from_fork+0x35/0x40 [ 7343.086204] Code: de 48 89 ef 5b 5d e9 84 f9 ff ff f0 ff 4e 1c 74 02 eb e8 8b 56 1c 48 c7 c7 e0 56 ef 81 48 c7 c6 b0 a9 cd 81 31 c0 e8 44 0b a2 ff <0f> 0b f6 45 08 01 75 0a 48 89 df 5b 5d e9 92 1d b5 ff 48 8d 73 [ 7343.086217] ---[ end trace af3c090ef841e41d ]---