- 03 Oct, 2018 40 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 10GbE Intel Wired LAN Driver Updates 2018-10-03 This series contains updates to ixgbe/ixgbevf and few fixes for i40e & iavf. Shannon Nelson fixes the message length for IPsec mailbox messages. Radoslaw fixes a transmit hang that occurs when XDP_TX exceeds the queue limit. Fixes a crash when we restor flow director filters after a reset. YueHaibing cleans up dead code, which did not have any callers. Dan Carpenter fixes an "off by one" error in IPsec for ixgbe. Nathan Chancellor fixes the i40e driver to use the correct enum for link speed. Also remove a debug statement since it was not producing useful information and equated to always "TRUE". Most notably, Björn introduces zero-copy AF_XDP support for the ixgbe driver. The ixgbe zero-copy code is located in its own file ixgbe_xsk.[ch], analogous to the i40e ZC support. Again, as in i40e, code paths have been copied from the XDP path to the zero-copy path. Going forward we will try to generalize more code between the AF_XDP ZC drivers, and also reduce the heavy C&P. ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Rami Rosen authored
This trivial patch fixes a typo in iavf.h. Signed-off-by:
Rami Rosen <ramirose@gmail.com> Acked-by:
Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by:
Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Björn Töpel authored
This patch adds zero-copy Tx support for AF_XDP sockets. It implements the ndo_xsk_async_xmit netdev ndo and performs all the Tx logic from a NAPI context. This means pulling egress packets from the Tx ring, placing the frames on the NIC HW descriptor ring and completing sent frames back to the application via the completion ring. The regular XDP Tx ring is used for AF_XDP as well. This rationale for this is as follows: XDP_REDIRECT guarantees mutual exclusion between different NAPI contexts based on CPU id. In other words, a netdev can XDP_REDIRECT to another netdev with a different NAPI context, since the operation is bound to a specific core and each core has its own hardware ring. As the AF_XDP Tx action is running in the same NAPI context and using the same ring, it will also be protected from XDP_REDIRECT actions with the exact same mechanism. As with AF_XDP Rx, all AF_XDP Tx specific functions are added to ixgbe_xsk.c. Signed-off-by:
Björn Töpel <bjorn.topel@intel.com> Tested-by:
William Tu <u9012063@gmail.com> Tested-by:
-
Björn Töpel authored
This patch prepares for the upcoming zero-copy Tx functionality by moving common functions used both by the regular path and zero-copy path. Signed-off-by:
-
Björn Töpel authored
This patch adds zero-copy Rx support for AF_XDP sockets. Instead of allocating buffers of type MEM_TYPE_PAGE_SHARED, the Rx frames are allocated as MEM_TYPE_ZERO_COPY when AF_XDP is enabled for a certain queue. All AF_XDP specific functions are added to a new file, ixgbe_xsk.c. Note that when AF_XDP zero-copy is enabled, the XDP action XDP_PASS will allocate a new buffer and copy the zero-copy frame prior passing it to the kernel stack. Signed-off-by:
-
Björn Töpel authored
This patch prepares for the upcoming zero-copy Rx functionality, by moving/changing linkage of common functions, used both by the regular path and zero-copy path. Signed-off-by:
-
Björn Töpel authored
Add functions for Rx/Tx ring enable/disable. Instead of resetting the whole device, only the affected ring is disabled or enabled. This plumbing is used in later commits, when zero-copy AF_XDP support is introduced. Signed-off-by:
-
Radoslaw Tyl authored
This patch fix crash when we have restore flow director filters after reset adapter. In ixgbe_fdir_filter_restore() filter->action is outside of the rx_ring array, as it has a VF identifier in the upper 32 bits. Signed-off-by:
Radoslaw Tyl <radoslawx.tyl@intel.com> Tested-by:
-
Nathan Chancellor authored
Clang warns that the address of a pointer will always evaluated as true in a boolean context. drivers/net/ethernet/intel/i40e/i40e_debugfs.c:136:9: warning: address of array 'vsi->active_vlans' will always evaluate to 'true' [-Wpointer-bool-conversion] vsi->active_vlans ? "<valid>" : "<null>"); ~~~~~^~~~~~~~~~~~ ~ ./include/linux/device.h:1431:33: note: expanded from macro 'dev_info' _dev_info(dev, dev_fmt(fmt), ##__VA_ARGS__) ^~~~~~~~~~~ 1 warning generated. Given that the statement shows that active_vlans is always valid, just remove the statement since it's not giving any useful information. Link: https://github.com/ClangBuiltLinux/linux/issues/82Signed-off-by:Nathan Chancellor <natechancellor@gmail.com> Tested-by:
-
Nathan Chancellor authored
Clang warns when one enumerated type is converted implicitly to another. drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c:4214:42: warning: implicit conversion from enumeration type 'enum i40e_aq_link_speed' to different enumeration type 'enum virtchnl_link_speed' [-Wenum-conversion] pfe.event_data.link_event.link_speed = I40E_LINK_SPEED_40GB; ~ ^~~~~~~~~~~~~~~~~~~~ 1 warning generated. Use the proper enum from virtchnl_link_speed, which has the same value as I40E_LINK_SPEED_40GB, VIRTCHNL_LINK_SPEED_40GB. This appears to be missed by commit ff3f4cc2 ("virtchnl: finish conversion to virtchnl interface"). Link: https://github.com/ClangBuiltLinux/linux/issues/81Signed-off-by: -
Dan Carpenter authored
The ipsec->tx_tbl[] array has IXGBE_IPSEC_MAX_SA_COUNT elements so the > should be a >=. Fixes: 0062e7cc ("ixgbevf: add VF IPsec offload code") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Shannon Nelson <shannon.nelson@oracle.com> Tested-by:
-
YueHaibing authored
There are no in-tree callers. Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Tested-by:
-
Radoslaw Tyl authored
We have Tx hang when number Tx and XDP queues are more than 64. In XDP always is MTQC == 0x0 (64TxQs). We need more space for Tx queues. Signed-off-by:
-
Shannon Nelson authored
Don't be fancy with message lengths, just set lengths to number of dwords, not bytes. Fixes: 0062e7cc ("ixgbevf: add VF IPsec offload code") Reported-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 100GbE Intel Wired LAN Driver Updates 2018-10-03 This series contains updates to ice and virtchnl. Yashaswini Raghuram adds a new virtchnl capability flag to support the exchange of additional supported speeds. Anirudh adds support for SR-IOV for the ice driver. Added code to initialize, configure and use mailbox queues for PF and VF communication. Updated the VSI and queue management to handle both PF and VF VSI type. Added "Adaptive Virtual Function (AVF)" support for the ice PF driver by implementing virtchnl commands. Extended the malicious driver detection logic to include the VF driver as well. Fixed the queue region size which needs to be log base 2 of the number of queues in region. Brett fixes an issue which was causing switch rules to be lost, by making a call to ice_update_pkt_fwd_rule() with the necessary changes. Fixed how the PF and VF assigned the ITR index by adding a struct member itr_idx to be used to dynamically program the correct ITR index. Dave fixed a potential NULL pointer dereference by adding checks in the filter handling. ==================== Signed-off-by: -
Ganesh Goudar authored
cxgb_set_tx_maxrate will be called holding rtnl lock, hence remove all unneeded locks. Signed-off-by:
Ganesh Goudar <ganeshgr@chelsio.com> Signed-off-by:
-
Anirudh Venkataramanan authored
Update version string to 0.7.2-k Signed-off-by:
Anirudh Venkataramanan <anirudh.venkataramanan@intel.com> Tested-by:
-
Dave Ertman authored
The ice_ena/dis_vsi should have a single differentiating factor to determine if the netdev_ops call is used or a direct call to ice_vsi_open/close. This is if the netif is running or not. If netif is running, use ndo_open/ndo_close. Else, use ice_vsi_open/ice_vsi_close. Signed-off-by:
Dave Ertman <david.m.ertman@intel.com> Signed-off-by:
-
Brett Creeley authored
This issue came about when looking at the VF function ice_vc_cfg_irq_map_msg. Currently we are assigning the itr_setting value to the itr_idx received from the AVF driver, which is not correct and is not used for the VF flow anyway. Currently the only way we set the ITR index for both the PF and VF driver is by hard coding ICE_TX_ITR or ICE_RX_ITR for the ITR index on each q_vector. To fix this, add the member itr_idx in struct ice_ring_container. This can then be used to dynamically program the correct ITR index. This change also affected the PF driver so make the necessary changes there as well. Also, removed the itr_setting member in struct ice_ring because it is not being used meaningfully and is going to be removed in a future patch that includes dynamic ITR. On another note, this will be useful moving forward if we decide to split Rx/Tx rings on different q_vectors instead of sharing them as queue pairs. Signed-off-by:
Brett Creeley <brett.creeley@intel.com> Signed-off-by:
-
Dave Ertman authored
Add checks in the filter handling flow to avoid dereferencing NULL pointers. Signed-off-by:
-
Brett Creeley authored
When a switch rule is initially created we set the filter action to ICE_FWD_TO_VSI. The filter action changes to ICE_FWD_TO_VSI_LIST whenever more than one VSI is subscribed to the same switch rule. When the switch rule goes from 2 VSIs in the list to 1 VSI we remove and delete the VSI list rule, but we currently don't update the switch rule in hardware. This is causing switch rules to be lost, so fix that by making a call to ice_update_pkt_fwd_rule() with the necessary changes. Signed-off-by:
-
Anirudh Venkataramanan authored
When adding a rule, queue region size needs to be provided as log base 2 of the number of queues in region. Fix that. Signed-off-by:
-
Anirudh Venkataramanan authored
This patch extends the existing malicious driver operation detection logic to cover malicious operations by the VF driver as well. Signed-off-by:
-
Anirudh Venkataramanan authored
When PF gets a link status change event, notify the VFs of the same. Signed-off-by:
-
Anirudh Venkataramanan authored
virtchnl is a protocol/interface specification that allows the Intel "Adaptive Virtual Function (AVF)" driver (iavf.ko) to work with more than one physical function driver. The AVF driver sends "virtchnl commands" (control plane only) to the PF driver over mailbox queues and the PF driver executes these commands and returns a result to the VF, again over mailbox. This patch adds AVF support for the ice PF driver by implementing the following virtchnl commands: VIRTCHNL_OP_VERSION VIRTCHNL_OP_GET_VF_RESOURCES VIRTCHNL_OP_RESET_VF VIRTCHNL_OP_ADD_ETH_ADDR VIRTCHNL_OP_DEL_ETH_ADDR VIRTCHNL_OP_CONFIG_VSI_QUEUES VIRTCHNL_OP_ENABLE_QUEUES VIRTCHNL_OP_DISABLE_QUEUES VIRTCHNL_OP_ADD_ETH_ADDR VIRTCHNL_OP_DEL_ETH_ADDR VIRTCHNL_OP_CONFIG_VSI_QUEUES VIRTCHNL_OP_ENABLE_QUEUES VIRTCHNL_OP_DISABLE_QUEUES VIRTCHNL_OP_REQUEST_QUEUES VIRTCHNL_OP_CONFIG_IRQ_MAP VIRTCHNL_OP_CONFIG_RSS_KEY VIRTCHNL_OP_CONFIG_RSS_LUT VIRTCHNL_OP_GET_STATS VIRTCHNL_OP_ADD_VLAN VIRTCHNL_OP_DEL_VLAN VIRTCHNL_OP_ENABLE_VLAN_STRIPPING VIRTCHNL_OP_DISABLE_VLAN_STRIPPING Signed-off-by:
-
Anirudh Venkataramanan authored
This patch implements handlers for the following NDO operations: .ndo_set_vf_spoofchk .ndo_set_vf_mac .ndo_get_vf_config .ndo_set_vf_trust .ndo_set_vf_vlan .ndo_set_vf_link_state Signed-off-by:
-
Anirudh Venkataramanan authored
Post VF initialization, there are a couple of different ways in which a VF reset can be triggered. One is when the underlying PF itself goes through a reset and other is via a VFLR interrupt. ice_reset_vf introduced in this patch handles both these cases. Also introduced in this patch is a helper function ice_aq_send_msg_to_vf to send messages to VF over the mailbox queue. The PF uses this to send reset notifications to VFs. Signed-off-by:
-
Anirudh Venkataramanan authored
Until now, all the VSI and queue management code supported only the PF VSI type (ICE_VSI_PF). Update these flows to handle the VF VSI type (ICE_VSI_VF) type as well. Signed-off-by:
-
Anirudh Venkataramanan authored
This patch implements parts of ice_sriov_configure and VF reset flow. To create virtual functions (VFs), the user sets a value in num_vfs through sysfs. This results in the kernel calling the handler for .sriov_configure which is ice_sriov_configure. VF setup first starts with a VF reset, followed by allocation of the VF VSI using ice_vf_vsi_setup. Once the VF setup is complete a state bit ICE_VF_STATE_INIT is set in the vf->states bitmap to indicate that the VF is ready to go. Also for VF reset to go into effect, it's necessary to issue a disable queue command (ice_aqc_opc_dis_txqs). So this patch updates multiple functions in the disable queue flow to take additional parameters that distinguish if queues are being disabled due to VF reset. Signed-off-by:
-
Anirudh Venkataramanan authored
Mailbox queue is a type of control queue that's used for communication between PF and VF. This patch adds code to initialize, configure and use mailbox queues. This patch also adds support to detect and parse SR-IOV capabilities returned by the hardware. Signed-off-by:
-
Introduced a new virtchnl capability flag and a struct to support exchange of additional supported speeds. Signed-off-by:
Yashaswini Raghuram Prathivadi Bhayankaram <yashaswini.raghuram.prathivadi.bhayankaram@intel.com> Signed-off-by:
-
Vakul Garg authored
Presently, for non-zero copy case, separate pages are allocated for storing plaintext and encrypted text of records. These pages are stored in sg_plaintext_data and sg_encrypted_data scatterlists inside record structure. Further, sg_plaintext_data & sg_encrypted_data are passed to cryptoapis for record encryption. Allocating separate pages for plaintext and encrypted text is inefficient from both required memory and performance point of view. This patch adds support of inplace encryption of records. For non-zero copy case, we reuse the pages from sg_encrypted_data scatterlist to copy the application's plaintext data. For the movement of pages from sg_encrypted_data to sg_plaintext_data scatterlists, we introduce a new function move_to_plaintext_sg(). This function add pages into sg_plaintext_data from sg_encrypted_data scatterlists. tls_do_encryption() is modified to pass the same scatterlist as both source and destination into aead_request_set_crypt() if inplace crypto has been enabled. A new ariable 'inplace_crypto' has been introduced in record structure to signify whether the same scatterlist can be used. By default, the inplace_crypto is enabled in get_rec(). If zero-copy is used (i.e. plaintext data is not copied), inplace_crypto is set to '0'. Signed-off-by:
Vakul Garg <vakul.garg@nxp.com> Reviewed-by:
Dave Watson <davejwatson@fb.com> Signed-off-by:
-
Hauke Mehrtens authored
The MAINTAINERS file contained the wrong file name of the driver. Fixes: 14fceff4 ("net: dsa: Add Lantiq / Intel DSA driver for vrx200") Reported-by:
Joe Perches <joe@perches.com> Signed-off-by:
Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by:
-
Stephen Hemminger authored
Similar to other patches from ERic. As diagnosed by Song Liu, ndo_poll_controller() can be very dangerous on loaded hosts, since the cpu calling ndo_poll_controller() might steal all NAPI contexts (for all RX/TX queues of the NIC). This capture can last for unlimited amount of time, since one cpu is generally not able to drain all the queues under load. In netvsc driver it uses NAPI for TX completions. The default poll_napi will do this for us now and avoid the capture. Signed-off-by:
Stephen Hemminger <sthemmin@microsoft.com> Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Eric Dumazet <edumazet@google.com> Reviewed-by:
Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by:
-
Andrew Lunn authored
Fix the patch for the mv88e6xxx.h header file in MAINTAINERS Reported-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
David S. Miller authored
Merge branch 'for-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next Johan Hedberg says: ==================== pull request: bluetooth-next 2018-09-30 Here's the first bluetooth-next pull request for the 4.20 kernel. - Fixes & cleanups to hci_qca driver - NULL dereference fix to debugfs - Improved L2CAP Connection-oriented Channel MTU & MPS handling - Added support for USB-based RTL8822C controller - Added device ID for BCM4335C0 UART-based controller - Various other smaller cleanups & fixes Please let me know if there are any issues pulling. Thanks. ==================== Signed-off-by:
-
Nathan Chancellor authored
Clang warns when one enumerated type is implicitly converted to another. drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c:390:4: warning: implicit conversion from enumeration type 'enum cxgb4_dcb_state' to different enumeration type 'enum cxgb4_dcb_state_input' [-Wenum-conversion] IEEE_FAUX_SYNC(dev, dcb); ^~~~~~~~~~~~~~~~~~~~~~~~ drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.h:70:10: note: expanded from macro 'IEEE_FAUX_SYNC' CXGB4_DCB_STATE_FW_ALLSYNCED); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ Use the equivalent value of the expected type to silence Clang while resulting in no functional change. CXGB4_DCB_STATE_FW_ALLSYNCED = CXGB4_DCB_INPUT_FW_ALLSYNCED = 3 Signed-off-by:Nick Desaulniers <ndesaulniers@google.com> Signed-off-by:
-
Nathan Chancellor authored
Clang warns when one enumerated type is implicitly converted to another. drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c:303:7: warning: implicit conversion from enumeration type 'enum cxgb4_dcb_state' to different enumeration type 'enum cxgb4_dcb_state_input' [-Wenum-conversion] ? CXGB4_DCB_STATE_FW_ALLSYNCED ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/net/ethernet/chelsio/cxgb4/cxgb4_dcb.c:304:7: warning: implicit conversion from enumeration type 'enum cxgb4_dcb_state' to different enumeration type 'enum cxgb4_dcb_state_input' [-Wenum-conversion] : CXGB4_DCB_STATE_FW_INCOMPLETE); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2 warnings generated. Use the equivalent value of the expected type to silence Clang while resulting in no functional change. CXGB4_DCB_STATE_FW_INCOMPLETE = CXGB4_DCB_INPUT_FW_INCOMPLETE = 2 CXGB4_DCB_STATE_FW_ALLSYNCED = CXGB4_DCB_INPUT_FW_ALLSYNCED = 3 Signed-off-by: -
Nathan Chancellor authored
Clang warns: drivers/net/ethernet/freescale/dpaa/dpaa_eth.c:2734:34: warning: tentative array definition assumed to have one element static const struct of_device_id dpaa_match[]; ^ 1 warning generated. Turns out that since this driver was introduced in commit 9ad1a374 ("dpaa_eth: add support for DPAA Ethernet"), this declaration has been unused. Remove it to silence the warning. Signed-off-by:Madalin Bucur <madalin.bucur@nxp.com> Signed-off-by:
-
Patrick Ruddy authored
The code to obtain the correct table for the incoming interface was missing for IPv6. This has been added along with the table creation notification to fib rules for the RTNL_FAMILY_IP6MR address family. Signed-off-by:
Patrick Ruddy <pruddy@vyatta.att-mail.com> Signed-off-by:
Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
