- 22 5月, 2020 3 次提交
-
-
由 Björn Töpel 提交于
Continuing the path to support MEM_TYPE_XSK_BUFF_POOL, the AF_XDP zero-copy/sk_buff rx_bi rings are now separate. Functions to properly allocate the different rings are added as well. v3->v4: Made i40e_fd_handle_status() static. (kbuild test robot) v4->v5: Fix kdoc for i40e_clean_programming_status(). (Jakub) Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: intel-wired-lan@lists.osuosl.org Link: https://lore.kernel.org/bpf/20200520192103.355233-8-bjorn.topel@gmail.com
-
由 Björn Töpel 提交于
As a first step to migrate i40e to the new MEM_TYPE_XSK_BUFF_POOL APIs, code that accesses the rx_bi (SW/shadow ring) is refactored to use an accessor function. Signed-off-by: NBjörn Töpel <bjorn.topel@intel.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: intel-wired-lan@lists.osuosl.org Link: https://lore.kernel.org/bpf/20200520192103.355233-7-bjorn.topel@gmail.com
-
由 Magnus Karlsson 提交于
Move the AF_XDP zero-copy driver interface to its own include file called xdp_sock_drv.h. This, hopefully, will make it more clear for NIC driver implementors to know what functions to use for zero-copy support. v4->v5: Fix -Wmissing-prototypes by include header file. (Jakub) Signed-off-by: NMagnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/20200520192103.355233-4-bjorn.topel@gmail.com
-
- 16 5月, 2020 4 次提交
-
-
由 Rahul Lakkireddy 提交于
Rework and add support for dumping EOTID software context used by TC-MQPRIO. Also track number of EOTIDs in use. Signed-off-by: NRahul Lakkireddy <rahul.lakkireddy@chelsio.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Rahul Lakkireddy 提交于
For each traffic class, firmware handles up to 4 * MTU amount of data per burst cycle. Under heavy load, this small buffer size is a bottleneck when buffering large TSO packets in <= 1500 MTU case. Increase the burst buffer size to 8 * MTU when supported. Also, keep the driver's traffic class configuration API similar to the firmware API counterpart. Signed-off-by: NRahul Lakkireddy <rahul.lakkireddy@chelsio.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Rahul Lakkireddy 提交于
Request credit update for every half credits consumed, including the current request. Also, avoid re-trying to post packets when there are no credits left. The credit update reply via interrupt will eventually restore the credits and will invoke the Tx path again. Signed-off-by: NRahul Lakkireddy <rahul.lakkireddy@chelsio.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Ioana Ciornei 提交于
Depending on the WRIOP version, the buffer size on the RX path must by a multiple of 64 or 256. Handle this restriction properly by aligning down the buffer size to the necessary value. Also, use the new buffer size dynamically computed instead of the compile time one. Fixes: 27c87486 ("dpaa2-eth: Use a single page per Rx buffer") Signed-off-by: NIoana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
- 15 5月, 2020 33 次提交
-
-
由 Jesper Dangaard Brouer 提交于
The mlx5 driver have multiple memory models, which are also changed according to whether a XDP bpf_prog is attached. The 'rx_striding_rq' setting is adjusted via ethtool priv-flags e.g.: # ethtool --set-priv-flags mlx5p2 rx_striding_rq off On the general case with 4K page_size and regular MTU packet, then the frame_sz is 2048 and 4096 when XDP is enabled, in both modes. The info on the given frame size is stored differently depending on the RQ-mode and encoded in a union in struct mlx5e_rq union wqe/mpwqe. In rx striding mode rq->mpwqe.log_stride_sz is either 11 or 12, which corresponds to 2048 or 4096 (MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ). In non-striding mode (MLX5_WQ_TYPE_CYCLIC) the frag_stride is stored in rq->wqe.info.arr[0].frag_stride, for the first fragment, which is what the XDP case cares about. To reduce effect on fast-path, this patch determine the frame_sz at setup time, to avoid determining the memory model runtime. Variable is named frame0_sz to make it clear that this is only the frame size of the first fragment. This mlx5 driver does a DMA-sync on XDP_TX action, but grow is safe as it have done a DMA-map on the entire PAGE_SIZE. The driver also already does a XDP length check against sq->hw_mtu on the possible XDP xmit paths mlx5e_xmit_xdp_frame() + mlx5e_xmit_xdp_frame_mpwqe(). V3+4: Change variable name first_frame_sz to frame0_sz V2: Fix that frag_size need to be recalc before creating SKB. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Acked-by: NTariq Toukan <tariqt@mellanox.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Link: https://lore.kernel.org/bpf/158945348021.97035.12295039384250022883.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
Intel drivers implement native AF_XDP zerocopy in separate C-files, that have its own invocation of bpf_prog_run_xdp(). The setup of xdp_buff is also handled in separately from normal code path. This patch update XDP frame_sz for AF_XDP zerocopy drivers i40e, ice and ixgbe, as the code changes needed are very similar. Introduce a helper function xsk_umem_xdp_frame_sz() for calculating frame size. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Acked-by: NBjörn Töpel <bjorn.topel@intel.com> Cc: intel-wired-lan@lists.osuosl.org Cc: Magnus Karlsson <magnus.karlsson@intel.com> Link: https://lore.kernel.org/bpf/158945347511.97035.8536753731329475655.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This driver uses different memory models depending on PAGE_SIZE at compile time. For PAGE_SIZE 4K it uses page splitting, meaning for normal MTU frame size is 2048 bytes (and headroom 192 bytes). For larger MTUs the driver still use page splitting, by allocating order-1 pages (8192 bytes) for RX frames. For PAGE_SIZE larger than 4K, driver instead advance its rx_buffer->page_offset with the frame size "truesize". For XDP frame size calculations, this mean that in PAGE_SIZE larger than 4K mode the frame_sz change on a per packet basis. For the page split 4K PAGE_SIZE mode, xdp.frame_sz is more constant and can be updated once outside the main NAPI loop. The default setting in the driver uses build_skb(), which provides the necessary headroom and tailroom for XDP-redirect in RX-frame (in both modes). There is one complication, which is legacy-rx mode (configurable via ethtool priv-flags). There are zero headroom in this mode, which is a requirement for XDP-redirect to work. The conversion to xdp_frame (convert_to_xdp_frame) will detect this insufficient space, and xdp_do_redirect() call will fail. This is deemed acceptable, as it allows other XDP actions to still work in legacy-mode. In legacy-mode + larger PAGE_SIZE due to lacking tailroom, we also accept that xdp_adjust_tail shrink doesn't work. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: intel-wired-lan@lists.osuosl.org Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Alexander Duyck <alexander.duyck@gmail.com> Link: https://lore.kernel.org/bpf/158945347002.97035.328088795813704587.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This driver uses different memory models depending on PAGE_SIZE at compile time. For PAGE_SIZE 4K it uses page splitting, meaning for normal MTU frame size is 2048 bytes (and headroom 192 bytes). For larger MTUs the driver still use page splitting, by allocating order-1 pages (8192 bytes) for RX frames. For PAGE_SIZE larger than 4K, driver instead advance its rx_buffer->page_offset with the frame size "truesize". For XDP frame size calculations, this mean that in PAGE_SIZE larger than 4K mode the frame_sz change on a per packet basis. For the page split 4K PAGE_SIZE mode, xdp.frame_sz is more constant and can be updated once outside the main NAPI loop. The default setting in the driver uses build_skb(), which provides the necessary headroom and tailroom for XDP-redirect in RX-frame (in both modes). There is one complication, which is legacy-rx mode (configurable via ethtool priv-flags). There are zero headroom in this mode, which is a requirement for XDP-redirect to work. The conversion to xdp_frame (convert_to_xdp_frame) will detect this insufficient space, and xdp_do_redirect() call will fail. This is deemed acceptable, as it allows other XDP actions to still work in legacy-mode. In legacy-mode + larger PAGE_SIZE due to lacking tailroom, we also accept that xdp_adjust_tail shrink doesn't work. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: intel-wired-lan@lists.osuosl.org Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Alexander Duyck <alexander.duyck@gmail.com> Link: https://lore.kernel.org/bpf/158945346494.97035.12809400414566061815.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This patch mirrors the changes to ixgbe in previous patch. This VF driver doesn't support XDP_REDIRECT, but correct tailroom is still necessary for BPF-helper xdp_adjust_tail. In legacy-mode + larger PAGE_SIZE, due to lacking tailroom, we accept that xdp_adjust_tail shrink doesn't work. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: intel-wired-lan@lists.osuosl.org Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Alexander Duyck <alexander.duyck@gmail.com> Link: https://lore.kernel.org/bpf/158945345984.97035.13518286183248025173.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This driver uses different memory models depending on PAGE_SIZE at compile time. For PAGE_SIZE 4K it uses page splitting, meaning for normal MTU frame size is 2048 bytes (and headroom 192 bytes). For larger MTUs the driver still use page splitting, by allocating order-1 pages (8192 bytes) for RX frames. For PAGE_SIZE larger than 4K, driver instead advance its rx_buffer->page_offset with the frame size "truesize". For XDP frame size calculations, this mean that in PAGE_SIZE larger than 4K mode the frame_sz change on a per packet basis. For the page split 4K PAGE_SIZE mode, xdp.frame_sz is more constant and can be updated once outside the main NAPI loop. The default setting in the driver uses build_skb(), which provides the necessary headroom and tailroom for XDP-redirect in RX-frame (in both modes). There is one complication, which is legacy-rx mode (configurable via ethtool priv-flags). There are zero headroom in this mode, which is a requirement for XDP-redirect to work. The conversion to xdp_frame (convert_to_xdp_frame) will detect this insufficient space, and xdp_do_redirect() call will fail. This is deemed acceptable, as it allows other XDP actions to still work in legacy-mode. In legacy-mode + larger PAGE_SIZE due to lacking tailroom, we also accept that xdp_adjust_tail shrink doesn't work. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: intel-wired-lan@lists.osuosl.org Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Cc: Alexander Duyck <alexander.duyck@gmail.com> Link: https://lore.kernel.org/bpf/158945345455.97035.14334355929030628741.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
The ixgbe driver have another memory model when compiled on archs with PAGE_SIZE above 4096 bytes. In this mode it doesn't split the page in two halves, but instead increment rx_buffer->page_offset by truesize of packet (which include headroom and tailroom for skb_shared_info). This is done correctly in ixgbe_build_skb(), but in ixgbe_rx_buffer_flip which is currently only called on XDP_TX and XDP_REDIRECT, it forgets to add the tailroom for skb_shared_info. This breaks XDP_REDIRECT, for veth and cpumap. Fix by adding size of skb_shared_info tailroom. Maintainers notice: This fix have been queued to Jeff. Fixes: 64530739 ("ixgbe: add initial support for xdp redirect") Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Link: https://lore.kernel.org/bpf/158945344946.97035.17031588499266605743.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
The netronome nfp driver use PAGE_SIZE when xdp_prog is set, but xdp.data_hard_start begins at offset NFP_NET_RX_BUF_HEADROOM. Thus, adjust for this when setting xdp.frame_sz, as it counts from data_hard_start. When doing XDP_TX this driver is smart and instead of a full DMA-map does a DMA-sync on with packet length. As xdp_adjust_tail can now grow packet length, add checks to make sure that grow size is within the DMA-mapped size. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Reviewed-by: NJakub Kicinski <kuba@kernel.org> Link: https://lore.kernel.org/bpf/158945342911.97035.11214251236208648808.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
To help reviewers these are the defines related to RCV_FRAG_LEN #define DMA_BUFFER_LEN 1536 /* In multiples of 128bytes */ #define RCV_FRAG_LEN (SKB_DATA_ALIGN(DMA_BUFFER_LEN + NET_SKB_PAD) + \ SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: Sunil Goutham <sgoutham@marvell.com> Cc: Robert Richter <rrichter@marvell.com> Link: https://lore.kernel.org/bpf/158945342402.97035.12649844447148990032.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
The mlx4 drivers size of memory backing the RX packet is stored in frag_stride. For XDP mode this will be PAGE_SIZE (normally 4096). For normal mode frag_stride is 2048. Also adjust MLX4_EN_MAX_XDP_MTU to take tailroom into account. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Reviewed-by: NTariq Toukan <tariqt@mellanox.com> Cc: Saeed Mahameed <saeedm@mellanox.com> Link: https://lore.kernel.org/bpf/158945341893.97035.2688142527052329942.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
Frame size ENA_PAGE_SIZE is limited to 16K on systems with larger PAGE_SIZE than 16K. Change ENA_XDP_MAX_MTU to also take into account the reserved tailroom. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Acked-by: NSameeh Jubran <sameehj@amazon.com> Cc: Arthur Kiyanovski <akiyano@amazon.com> Link: https://lore.kernel.org/bpf/158945341384.97035.907403694833419456.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
The driver code cpsw.c and cpsw_new.c both use page_pool with default order-0 pages or their RX-pages. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Reviewed-by: NGrygorii Strashko <grygorii.strashko@ti.com> Cc: Ilias Apalodimas <ilias.apalodimas@linaro.org> Link: https://lore.kernel.org/bpf/158945340875.97035.752144756428532878.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
The driver qede uses a full page, when XDP is enabled. The drivers value in rx_buf_seg_size (struct qede_rx_queue) will be PAGE_SIZE when an XDP bpf_prog is attached. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: Ariel Elior <aelior@marvell.com> Cc: GR-everest-linux-l2@marvell.com Link: https://lore.kernel.org/bpf/158945340366.97035.7764939691580349618.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
The dpaa2-eth driver reserve some headroom used for hardware and software annotation area in RX/TX buffers. Thus, xdp.data_hard_start doesn't start at page boundary. When XDP is configured the area reserved via dpaa2_fd_get_offset(fd) is 448 bytes of which XDP have reserved 256 bytes. As frame_sz is calculated as an offset from xdp_buff.data_hard_start, an adjust from the full PAGE_SIZE == DPAA2_ETH_RX_BUF_RAW_SIZE. When doing XDP_REDIRECT, the driver doesn't need this reserved headroom any-longer and allows xdp_do_redirect() to use it. This is an advantage for the drivers own ndo-xdp_xmit, as it uses part of this headroom for itself. Patch also adjust frame_sz in this case. The driver cannot support XDP data_meta, because it uses the headroom just before xdp.data for struct dpaa2_eth_swa (DPAA2_ETH_SWA_SIZE=64), when transmitting the packet. When transmitting a xdp_frame in dpaa2_eth_xdp_xmit_frame (call via ndo_xdp_xmit) is uses this area to store a pointer to xdp_frame and dma_size, which is used in TX completion (free_tx_fd) to return frame via xdp_return_frame(). Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Cc: Ioana Radulescu <ruxandra.radulescu@nxp.com> Link: https://lore.kernel.org/bpf/158945339348.97035.8562488847066908856.stgit@firesoul
-
由 Ilias Apalodimas 提交于
This driver takes advantage of page_pool PP_FLAG_DMA_SYNC_DEV that can help reduce the number of cache-lines that need to be flushed when doing DMA sync for_device. Due to xdp_adjust_tail can grow the area accessible to the by the CPU (can possibly write into), then max sync length *after* bpf_prog_run_xdp() needs to be taken into account. For XDP_TX action the driver is smart and does DMA-sync. When growing tail this is still safe, because page_pool have DMA-mapped the entire page size. Signed-off-by: NIlias Apalodimas <ilias.apalodimas@linaro.org> Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Acked-by: NLorenzo Bianconi <lorenzo@kernel.org> Link: https://lore.kernel.org/bpf/158945336295.97035.15034759661036971024.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This marvell driver mvneta uses PAGE_SIZE frames, which makes it really easy to convert. Driver updates rxq and now frame_sz once per NAPI call. This driver takes advantage of page_pool PP_FLAG_DMA_SYNC_DEV that can help reduce the number of cache-lines that need to be flushed when doing DMA sync for_device. Due to xdp_adjust_tail can grow the area accessible to the by the CPU (can possibly write into), then max sync length *after* bpf_prog_run_xdp() needs to be taken into account. For XDP_TX action the driver is smart and does DMA-sync. When growing tail this is still safe, because page_pool have DMA-mapped the entire page size. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Acked-by: NLorenzo Bianconi <lorenzo@kernel.org> Cc: thomas.petazzoni@bootlin.com Link: https://lore.kernel.org/bpf/158945335786.97035.12714388304493736747.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This driver uses RX page-split when possible. It was recently fixed in commit 86e85bf6 ("sfc: fix XDP-redirect in this driver") to add needed tailroom for XDP-redirect. After the fix efx->rx_page_buf_step is the frame size, with enough head and tail-room for XDP-redirect. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Link: https://lore.kernel.org/bpf/158945335278.97035.14611425333184621652.stgit@firesoul
-
由 Jesper Dangaard Brouer 提交于
This driver uses full PAGE_SIZE pages when XDP is enabled. In case of XDP uses driver uses __bnxt_alloc_rx_page which does full page DMA-map. Thus, xdp_adjust_tail grow is DMA compliant for XDP_TX action that does DMA-sync. Signed-off-by: NJesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: NAlexei Starovoitov <ast@kernel.org> Reviewed-by: NAndy Gospodarek <gospo@broadcom.com> Cc: Michael Chan <michael.chan@broadcom.com> Cc: Andy Gospodarek <andrew.gospodarek@broadcom.com> Link: https://lore.kernel.org/bpf/158945334769.97035.13437970179897613984.stgit@firesoul
-
由 Heiner Kallweit 提交于
93882c6f ("r8169: switch from netif_xxx message functions to netdev_xxx") removed the last module parameter from the driver, therefore there's no need any longer to include linux/moduleparam.h. Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Heiner Kallweit 提交于
After 9de5d235 ("net: phy: fix aneg restart in phy_ethtool_set_eee") we don't need the check for aneg being enabled any longer, and as discussed with Russell configuring the EEE advertisement should be supported even if we're in a half-duplex mode currently. Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Luo bin 提交于
support to change TX/RX queue depth with ethtool -G Signed-off-by: NLuo bin <luobin9@huawei.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Ivan Khoronzhuk 提交于
AM65 CPSW h/w supports Enhanced Scheduled Traffic (EST – defined in P802.1Qbv/D2.2 that later got included in IEEE 802.1Q-2018) configuration. EST allows express queue traffic to be scheduled (placed) on the wire at specific repeatable time intervals. In Linux kernel, EST configuration is done through tc command and the taprio scheduler in the net core implements a software only scheduler (SCH_TAPRIO). If the NIC is capable of EST configuration, user indicate "flag 2" in the command which is then parsed by taprio scheduler in net core and indicate that the command is to be offloaded to h/w. taprio then offloads the command to the driver by calling ndo_setup_tc() ndo ops. This patch implements ndo_setup_tc() to offload EST configuration to CPSW h/w. Currently driver supports only SetGateStates operation. EST operates on a repeating time interval generated by the CPTS EST function generator. Each Ethernet port has a global EST fetch RAM that can be configured as 2 buffers, each of 64 locations or one large buffer of 128 locations. In 2 buffer configuration, a ping pong mechanism is used to hold the active schedule (oper) in one buffer and new (admin) command in the other. Each 22-bit fetch command consists of a 14-bit fetch count (14 MSB’s) and an 8-bit priority fetch allow (8 LSB’s) that will be applied for the fetch count time in wireside clocks. Driver process each of the sched-entry in the offload command and update the fetch RAM. Driver configures duration in sched-entry into the fetch count and Gate mask into the priority fetch bits of the RAM. Then configures the CPTS EST function generator to activate the schedule. Currently driver supports only 2 buffer configuration which means driver supports a max cycle time of ~8 msec. CPSW supports a configurable number of priority queues (up to 8) and needs to be switched to this mode from the default round robin mode before EST can be offloaded. User configures these through ethtool commands (-L for changing number of queues and --set-priv-flags to disable round robin mode). Driver doesn't enable EST if pf_p0_rx_ptype_rrobin privat flag is set. The flag is common for all ports, and so can't be just overridden by taprio configuration w/o user involvement. Command fails if pf_p0_rx_ptype_rrobin is already set in the driver. Scheds (commands) configuration depends on interface speed so driver translates the duration to the fetch count based on link speed. Each schedule can be constructed with several command entries in fetch RAM depending on interval. For example if each sched has timer interval < ~130us on 1000 Mb link then each sched consumes one command and have 1:1 mapping. When Ethernet link goes down, driver purge the configuration if link is down for more than 1 second. The patch allows to update the timer and scheds memory only if it's really needed, and skip cases required the user to stop timer by configuring only shceds memory. Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org> Signed-off-by: NMurali Karicheri <m-karicheri2@ti.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Ivan Khoronzhuk 提交于
TAPRIO/EST offload support in CPSW2G requires EST scheduler function enabled in CPTS. So this patch add a function to set cycle time for EST scheduler. It also add a function for getting time in ns of PHC clock for taprio qdisc configuration. Mostly to verify if timer update is needed or to get actual state of oper/admin schedule. Signed-off-by: NIvan Khoronzhuk <ivan.khoronzhuk@linaro.org> Signed-off-by: NMurali Karicheri <m-karicheri2@ti.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
MCP may signal driver about generic critical failure. Driver has to collect mdump information (get_retain), it pushes that to logs and triggers generic notification on "hardware attention" event. Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
Fan failure is sent by firmware, driver reacts on this error with newly introduced notification path. It will collect dump and shut down the device to prevent physical breakage Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Denis Bolotin 提交于
Upon tx timeout detection we do disable carrier and print TX queue info on TX timeout. We then raise hw error condition and trigger service task to handle this. This handler will capture extra debug info and then optionally trigger recovery procedure to try restore function. Signed-off-by: NDenis Bolotin <dbolotin@marvell.com> Signed-off-by: NAriel Elior <aelior@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
Driver has an ability to initiate a recovery process as a reaction to detected errors. But the codepath (recovery_process) was disabled and never active. Here we add ethtool private flag to allow user have the recovery procedure activated. We still do not enable this by default though, since in some configurations this is not desirable. E.g. this may impact other PFs/VFs. Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
On different hardware events we have to respond differently, on some of hardware indications hw attention (error condition) should be cleared by the driver to continue normal functioning. Here we introduce attention clear flags, and put them on some important events (in aeu_descs). Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
Thats probably a legacy code had double declaration of some fields. Cleanup this, removing copy and fixing references. Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
On various critical errors, notification handler should also report the err information into the management firmware. MFW can interact with server/motherboard backend agents - these are used by server manufacturers to monitor server HW health. Thus, it is important for driver to report on any faulty conditions Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
In a number of critical places not only debug trace should be printed, but the appropriate hw error condition should be raised and error handling/recovery should start. Introduce our new qed_hw_err_notify invocation in these places to record and indicate critical error conditions in hardware. Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
qede (ethernet level driver) registers a callback handler. This handler maintains eth dev state flags/bits to track error processing. It implements in place processing part for nonsleeping context (WARN_ON trigger), and a deferred (delayed work) part which triggers recovery process for recoverable errors. In later patches this atomic handler will come with more meat. We introduce err_flags on ethdevice structure, its being used to record error handling properties. Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-
由 Igor Russkikh 提交于
Here we introduce qed device error tracking flags and error types. qed_hw_err_notify is an entrace point to report errors. It'll notify higher level drivers (qede/qedr/etc) to handle and recover the error. List of posible errors comes from hardware interfaces, but could be extended in future. Signed-off-by: NAriel Elior <ariel.elior@marvell.com> Signed-off-by: NMichal Kalderon <michal.kalderon@marvell.com> Signed-off-by: NIgor Russkikh <irusskikh@marvell.com> Signed-off-by: NDavid S. Miller <davem@davemloft.net>
-