d/numam-dpdk
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
67b38d93c9
numam-dpdk/drivers/net/e1000/igb_ethdev.c

4283 lines
118 KiB
C
Raw Normal View History

first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/*-
* BSD LICENSE
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
*
e1000: move to drivers/net/ Move e1000 pmd to drivers/net directory As part of move, rename "e1000" subdirectory, which contains the code from the "base driver", to "base". Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-05-15 15:56:49 +00:00
* Copyright(c) 2010-2015 Intel Corporation. All rights reserved.
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* All rights reserved.
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
*
doc: whitespace changes in licenses Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* are met:
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
*
doc: whitespace changes in licenses Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
* * Redistributions of source code must retain the above copyright
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* notice, this list of conditions and the following disclaimer.
doc: whitespace changes in licenses Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* distribution.
doc: whitespace changes in licenses Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
* * Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* from this software without specific prior written permission.
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
*
doc: whitespace changes in licenses Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/queue.h>
#include <stdio.h>
#include <errno.h>
#include <stdint.h>
#include <stdarg.h>
#include <rte_common.h>
#include <rte_interrupts.h>
#include <rte_byteorder.h>
#include <rte_log.h>
#include <rte_debug.h>
#include <rte_pci.h>
#include <rte_ether.h>
#include <rte_ethdev.h>
#include <rte_memory.h>
#include <rte_memzone.h>
#include <rte_eal.h>
#include <rte_atomic.h>
#include <rte_malloc.h>
igb: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igb pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igb library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:32 +00:00
#include <rte_dev.h>
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
#include "e1000_logs.h"
e1000: move to drivers/net/ Move e1000 pmd to drivers/net directory As part of move, rename "e1000" subdirectory, which contains the code from the "base driver", to "base". Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-05-15 15:56:49 +00:00
#include "base/e1000_api.h"
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
#include "e1000_ethdev.h"
igb: add access to specific device info add function to support ethtool ops: - get_reg_length - get_regs - get_eeprom_length - get_eeprom - set_eeprom Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
#include "igb_regs.h"
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: set default Rx/Tx configuration Many sample apps use duplicated code to set rte_eth_txconf and rte_eth_rxconf structures. This patch allows the user to get a default optimal RX/TX configuration through rte_eth_dev_info get, and still any parameters may be tweaked as wished, before setting up queues. Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Reviewed-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: David Marchand <david.marchand@6wind.com> [Thomas: split patch]
2014-10-01 09:49:04 +00:00
/*
* Default values for port configuration
*/
#define IGB_DEFAULT_RX_FREE_THRESH 32
#define IGB_DEFAULT_RX_PTHRESH 8
#define IGB_DEFAULT_RX_HTHRESH 8
#define IGB_DEFAULT_RX_WTHRESH 0
#define IGB_DEFAULT_TX_PTHRESH 32
#define IGB_DEFAULT_TX_HTHRESH 0
#define IGB_DEFAULT_TX_WTHRESH 0
ethdev: add the hash key size per device To support querying hash key size per port, an new field of 'hash_key_size' was added in 'struct rte_eth_dev_info' for storing hash key size in bytes. The correct hash key size in bytes should be filled into the 'struct rte_eth_dev_info', to support querying it. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-06-12 07:33:56 +00:00
#define IGB_HKEY_MAX_INDEX 10
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
/* Bit shift and mask */
#define IGB_4_BIT_WIDTH (CHAR_BIT / 2)
#define IGB_4_BIT_MASK RTE_LEN2MASK(IGB_4_BIT_WIDTH, uint8_t)
#define IGB_8_BIT_WIDTH CHAR_BIT
#define IGB_8_BIT_MASK UINT8_MAX
e1000: add ieee1588 timestamping Add e1000/igb support for new ethdev APIs to enable and read IEEE1588 PTP timestamps. Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-07-02 15:16:29 +00:00
/* Additional timesync values. */
#define E1000_ETQF_FILTER_1588 3
#define E1000_TIMINCA_INCVALUE 16000000
#define E1000_TIMINCA_INIT ((0x02 << E1000_TIMINCA_16NS_SHIFT) \
| E1000_TIMINCA_INCVALUE)
igb: fix ieee1588 timestamping Ieee1588 reads system time to set its timestamp. On 1G NICs, for example, i350, system time is disabled by default. It means the ieee1588 timestamp will always be 0. This patch enables system time when ieee1588 is enabled. Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com>
2015-07-31 01:20:30 +00:00
#define E1000_TSAUXC_DISABLE_SYSTIME 0x80000000
e1000: add ieee1588 timestamping Add e1000/igb support for new ethdev APIs to enable and read IEEE1588 PTP timestamps. Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-07-02 15:16:29 +00:00
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
static int eth_igb_configure(struct rte_eth_dev *dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int eth_igb_start(struct rte_eth_dev *dev);
static void eth_igb_stop(struct rte_eth_dev *dev);
static void eth_igb_close(struct rte_eth_dev *dev);
static void eth_igb_promiscuous_enable(struct rte_eth_dev *dev);
static void eth_igb_promiscuous_disable(struct rte_eth_dev *dev);
static void eth_igb_allmulticast_enable(struct rte_eth_dev *dev);
static void eth_igb_allmulticast_disable(struct rte_eth_dev *dev);
static int eth_igb_link_update(struct rte_eth_dev *dev,
int wait_to_complete);
static void eth_igb_stats_get(struct rte_eth_dev *dev,
struct rte_eth_stats *rte_stats);
static void eth_igb_stats_reset(struct rte_eth_dev *dev);
static void eth_igb_infos_get(struct rte_eth_dev *dev,
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
struct rte_eth_dev_info *dev_info);
static void eth_igbvf_infos_get(struct rte_eth_dev *dev,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
struct rte_eth_dev_info *dev_info);
ethdev: retrieve flow control configuration This patch adds a new function in ethdev api to retrieve current flow control configuration. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Zijie Pan <zijie.pan@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:26 +00:00
static int eth_igb_flow_ctrl_get(struct rte_eth_dev *dev,
struct rte_eth_fc_conf *fc_conf);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int eth_igb_flow_ctrl_set(struct rte_eth_dev *dev,
struct rte_eth_fc_conf *fc_conf);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
static int eth_igb_lsc_interrupt_setup(struct rte_eth_dev *dev);
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
static int eth_igb_rxq_interrupt_setup(struct rte_eth_dev *dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int eth_igb_interrupt_get_status(struct rte_eth_dev *dev);
static int eth_igb_interrupt_action(struct rte_eth_dev *dev);
static void eth_igb_interrupt_handler(struct rte_intr_handle *handle,
void *param);
static int igb_hardware_init(struct e1000_hw *hw);
static void igb_hw_control_acquire(struct e1000_hw *hw);
static void igb_hw_control_release(struct e1000_hw *hw);
static void igb_init_manageability(struct e1000_hw *hw);
static void igb_release_manageability(struct e1000_hw *hw);
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
ethdev: MTU accessors This patch adds two new functions in ethdev api to retrieve current MTU and change MTU of a port. Only .mtu_set function is pmd specific. pmd should update its max_rx_pkt_len if needed. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Samuel Gauthier <samuel.gauthier@6wind.com> Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:30 +00:00
static int eth_igb_mtu_set(struct rte_eth_dev *dev, uint16_t mtu);
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
static int eth_igb_vlan_filter_set(struct rte_eth_dev *dev,
uint16_t vlan_id, int on);
static void eth_igb_vlan_tpid_set(struct rte_eth_dev *dev, uint16_t tpid_id);
static void eth_igb_vlan_offload_set(struct rte_eth_dev *dev, int mask);
static void igb_vlan_hw_filter_enable(struct rte_eth_dev *dev);
static void igb_vlan_hw_filter_disable(struct rte_eth_dev *dev);
static void igb_vlan_hw_strip_enable(struct rte_eth_dev *dev);
static void igb_vlan_hw_strip_disable(struct rte_eth_dev *dev);
static void igb_vlan_hw_extend_enable(struct rte_eth_dev *dev);
static void igb_vlan_hw_extend_disable(struct rte_eth_dev *dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int eth_igb_led_on(struct rte_eth_dev *dev);
static int eth_igb_led_off(struct rte_eth_dev *dev);
static void igb_intr_disable(struct e1000_hw *hw);
static int igb_get_rx_buffer_size(struct e1000_hw *hw);
e1000: minor changes Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
static void eth_igb_rar_set(struct rte_eth_dev *dev,
struct ether_addr *mac_addr,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
uint32_t index, uint32_t pool);
static void eth_igb_rar_clear(struct rte_eth_dev *dev, uint32_t index);
igb: add default mac address modifier - set_mac_addr Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static void eth_igb_default_mac_addr_set(struct rte_eth_dev *dev,
struct ether_addr *addr);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
static void igbvf_intr_disable(struct e1000_hw *hw);
static int igbvf_dev_configure(struct rte_eth_dev *dev);
static int igbvf_dev_start(struct rte_eth_dev *dev);
static void igbvf_dev_stop(struct rte_eth_dev *dev);
static void igbvf_dev_close(struct rte_eth_dev *dev);
static int eth_igbvf_link_update(struct e1000_hw *hw);
static void eth_igbvf_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *rte_stats);
static void eth_igbvf_stats_reset(struct rte_eth_dev *dev);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
static int igbvf_vlan_filter_set(struct rte_eth_dev *dev,
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
uint16_t vlan_id, int on);
static int igbvf_set_vfta(struct e1000_hw *hw, uint16_t vid, bool on);
static void igbvf_set_vfta_all(struct rte_eth_dev *dev, bool on);
igb: add default mac address modifier - set_mac_addr Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static void igbvf_default_mac_addr_set(struct rte_eth_dev *dev,
struct ether_addr *addr);
igb: add access to specific device info add function to support ethtool ops: - get_reg_length - get_regs - get_eeprom_length - get_eeprom - set_eeprom Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static int igbvf_get_reg_length(struct rte_eth_dev *dev);
static int igbvf_get_regs(struct rte_eth_dev *dev,
struct rte_dev_reg_info *regs);
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
static int eth_igb_rss_reta_update(struct rte_eth_dev *dev,
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
struct rte_eth_rss_reta_entry64 *reta_conf,
uint16_t reta_size);
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
static int eth_igb_rss_reta_query(struct rte_eth_dev *dev,
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
struct rte_eth_rss_reta_entry64 *reta_conf,
uint16_t reta_size);
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
static int eth_igb_syn_filter_set(struct rte_eth_dev *dev,
struct rte_eth_syn_filter *filter,
bool add);
static int eth_igb_syn_filter_get(struct rte_eth_dev *dev,
struct rte_eth_syn_filter *filter);
static int eth_igb_syn_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg);
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
static int igb_add_2tuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter);
static int igb_remove_2tuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
static int eth_igb_add_del_flex_filter(struct rte_eth_dev *dev,
struct rte_eth_flex_filter *filter,
bool add);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
static int eth_igb_get_flex_filter(struct rte_eth_dev *dev,
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct rte_eth_flex_filter *filter);
static int eth_igb_flex_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg);
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
static int igb_add_5tuple_filter_82576(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter);
static int igb_remove_5tuple_filter_82576(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter);
static int igb_add_del_ntuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *filter,
bool add);
static int igb_get_ntuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *filter);
static int igb_ntuple_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg);
igb: use generic filter control for ethertype filter This patch removes old functions which deal with ethertype filter in igb driver. It also defines eth_igb_filter_ctrl which is binding to filter_ctrl API, and ethertype filter can be dealt with through this new entrance. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-01-12 07:16:11 +00:00
static int igb_add_del_ethertype_filter(struct rte_eth_dev *dev,
struct rte_eth_ethertype_filter *filter,
bool add);
static int igb_ethertype_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg);
static int igb_get_ethertype_filter(struct rte_eth_dev *dev,
struct rte_eth_ethertype_filter *filter);
static int eth_igb_filter_ctrl(struct rte_eth_dev *dev,
enum rte_filter_type filter_type,
enum rte_filter_op filter_op,
void *arg);
igb: add access to specific device info add function to support ethtool ops: - get_reg_length - get_regs - get_eeprom_length - get_eeprom - set_eeprom Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static int eth_igb_get_reg_length(struct rte_eth_dev *dev);
static int eth_igb_get_regs(struct rte_eth_dev *dev,
struct rte_dev_reg_info *regs);
static int eth_igb_get_eeprom_length(struct rte_eth_dev *dev);
static int eth_igb_get_eeprom(struct rte_eth_dev *dev,
struct rte_dev_eeprom_info *eeprom);
static int eth_igb_set_eeprom(struct rte_eth_dev *dev,
struct rte_dev_eeprom_info *eeprom);
e1000: add multicast MAC address filtering Support the PMD function "set_mc_addr_list" in the "igb", "igb-vf", and "em" Poll Mode Drivers. Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
2015-05-28 15:05:21 +00:00
static int eth_igb_set_mc_addr_list(struct rte_eth_dev *dev,
struct ether_addr *mc_addr_set,
uint32_t nb_mc_addr);
e1000: add ieee1588 timestamping Add e1000/igb support for new ethdev APIs to enable and read IEEE1588 PTP timestamps. Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-07-02 15:16:29 +00:00
static int igb_timesync_enable(struct rte_eth_dev *dev);
static int igb_timesync_disable(struct rte_eth_dev *dev);
static int igb_timesync_read_rx_timestamp(struct rte_eth_dev *dev,
struct timespec *timestamp,
uint32_t flags);
static int igb_timesync_read_tx_timestamp(struct rte_eth_dev *dev,
struct timespec *timestamp);
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
static int eth_igb_rx_queue_intr_enable(struct rte_eth_dev *dev,
uint16_t queue_id);
static int eth_igb_rx_queue_intr_disable(struct rte_eth_dev *dev,
uint16_t queue_id);
static void eth_igb_assign_msix_vector(struct e1000_hw *hw, int8_t direction,
uint8_t queue, uint8_t msix_vector);
static void eth_igb_write_ivar(struct e1000_hw *hw, uint8_t msix_vector,
uint8_t index, uint8_t offset);
static void eth_igb_configure_msix_intr(struct rte_eth_dev *dev);
e1000: add multicast MAC address filtering Support the PMD function "set_mc_addr_list" in the "igb", "igb-vf", and "em" Poll Mode Drivers. Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
2015-05-28 15:05:21 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* Define VF Stats MACRO for Non "cleared on read" register
*/
#define UPDATE_VF_STAT(reg, last, cur) \
{ \
u32 latest = E1000_READ_REG(hw, reg); \
cur += latest - last; \
last = latest; \
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
#define IGB_FC_PAUSE_TIME 0x0680
#define IGB_LINK_UPDATE_CHECK_TIMEOUT 90 /* 9s */
#define IGB_LINK_UPDATE_CHECK_INTERVAL 100 /* ms */
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
#define IGBVF_PMD_NAME "rte_igbvf_pmd" /* PMD name */
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static enum e1000_fc_mode igb_fc_setting = e1000_fc_full;
/*
* The set of PCI devices this driver supports
*/
pci: make device id tables const The PCI device id table is immutable and should be made const in all drivers. The pseudo drivers can initialize their local copy as necessary. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Neil Horman <nhorman@tuxdriver.com>
2015-04-16 23:23:39 +00:00
static const struct rte_pci_id pci_id_igb_map[] = {
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
#define RTE_PCI_DEV_ID_DECL_IGB(vend, dev) {RTE_PCI_DEVICE(vend, dev)},
#include "rte_pci_dev_ids.h"
use simple zero initializers To initialize a structure with zeros, one field was explicitly set to avoid "missing initializer" bug with old GCC (e.g. 4.4). This warning is now disabled (commit <insertlater>) for old versions of GCC, so the workarounds may be removed. These initializers should not be needed for static variables but they are still used to workaround an ICC bug (see commit b2595c4aa92d). There is one remaining exception where {0} initializer doesn't work cleanly, even with recent GCC: lib/librte_pmd_ixgbe/ixgbe_rxtx_vec.c:735:9: error: missing braces around initializer [-Werror=missing-braces] struct rte_mbuf mb_def = {0}; /* zeroed mbuf */ Tested with gcc-4.4.7 (CentOS), gcc-4.7.2 (Debian), gcc-4.9.2 (Arch), clang-3.6.0 and icc-13.1.1. Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Tested-by: Thomas Monjalon <thomas.monjalon@6wind.com> Tested-by: John McNamara <john.mcnamara@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Vlad Zolotarov <vladz@cloudius-systems.com>
2015-04-15 17:09:28 +00:00
{0},
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
};
/*
* The set of PCI devices this driver supports (for 82576&I350 VF)
*/
pci: make device id tables const The PCI device id table is immutable and should be made const in all drivers. The pseudo drivers can initialize their local copy as necessary. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Neil Horman <nhorman@tuxdriver.com>
2015-04-16 23:23:39 +00:00
static const struct rte_pci_id pci_id_igbvf_map[] = {
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
#define RTE_PCI_DEV_ID_DECL_IGBVF(vend, dev) {RTE_PCI_DEVICE(vend, dev)},
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
#include "rte_pci_dev_ids.h"
use simple zero initializers To initialize a structure with zeros, one field was explicitly set to avoid "missing initializer" bug with old GCC (e.g. 4.4). This warning is now disabled (commit <insertlater>) for old versions of GCC, so the workarounds may be removed. These initializers should not be needed for static variables but they are still used to workaround an ICC bug (see commit b2595c4aa92d). There is one remaining exception where {0} initializer doesn't work cleanly, even with recent GCC: lib/librte_pmd_ixgbe/ixgbe_rxtx_vec.c:735:9: error: missing braces around initializer [-Werror=missing-braces] struct rte_mbuf mb_def = {0}; /* zeroed mbuf */ Tested with gcc-4.4.7 (CentOS), gcc-4.7.2 (Debian), gcc-4.9.2 (Arch), clang-3.6.0 and icc-13.1.1. Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Tested-by: Thomas Monjalon <thomas.monjalon@6wind.com> Tested-by: John McNamara <john.mcnamara@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com> Acked-by: Vlad Zolotarov <vladz@cloudius-systems.com>
2015-04-15 17:09:28 +00:00
{0},
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
};
ethdev: make dev_ops const The ethernet device ops function table should be made const for safety and security. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Neil Horman <nhorman@tuxdriver.com>
2015-04-07 21:21:03 +00:00
static const struct eth_dev_ops eth_igb_ops = {
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.dev_configure = eth_igb_configure,
.dev_start = eth_igb_start,
.dev_stop = eth_igb_stop,
.dev_close = eth_igb_close,
.promiscuous_enable = eth_igb_promiscuous_enable,
.promiscuous_disable = eth_igb_promiscuous_disable,
.allmulticast_enable = eth_igb_allmulticast_enable,
.allmulticast_disable = eth_igb_allmulticast_disable,
.link_update = eth_igb_link_update,
.stats_get = eth_igb_stats_get,
.stats_reset = eth_igb_stats_reset,
.dev_infos_get = eth_igb_infos_get,
ethdev: MTU accessors This patch adds two new functions in ethdev api to retrieve current MTU and change MTU of a port. Only .mtu_set function is pmd specific. pmd should update its max_rx_pkt_len if needed. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Samuel Gauthier <samuel.gauthier@6wind.com> Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:30 +00:00
.mtu_set = eth_igb_mtu_set,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.vlan_filter_set = eth_igb_vlan_filter_set,
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.vlan_tpid_set = eth_igb_vlan_tpid_set,
.vlan_offload_set = eth_igb_vlan_offload_set,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.rx_queue_setup = eth_igb_rx_queue_setup,
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
.rx_queue_intr_enable = eth_igb_rx_queue_intr_enable,
.rx_queue_intr_disable = eth_igb_rx_queue_intr_disable,
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.rx_queue_release = eth_igb_rx_queue_release,
igb: retrieve RX available descriptors Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
.rx_queue_count = eth_igb_rx_queue_count,
igb: check DD bit of specific RX descriptor Signed-off-by: Intel
2013-07-22 22:00:00 +00:00
.rx_descriptor_done = eth_igb_rx_descriptor_done,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.tx_queue_setup = eth_igb_tx_queue_setup,
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.tx_queue_release = eth_igb_tx_queue_release,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.dev_led_on = eth_igb_led_on,
.dev_led_off = eth_igb_led_off,
ethdev: retrieve flow control configuration This patch adds a new function in ethdev api to retrieve current flow control configuration. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Zijie Pan <zijie.pan@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:26 +00:00
.flow_ctrl_get = eth_igb_flow_ctrl_get,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.flow_ctrl_set = eth_igb_flow_ctrl_set,
.mac_addr_add = eth_igb_rar_set,
.mac_addr_remove = eth_igb_rar_clear,
igb: add default mac address modifier - set_mac_addr Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
.mac_addr_set = eth_igb_default_mac_addr_set,
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
.reta_update = eth_igb_rss_reta_update,
.reta_query = eth_igb_rss_reta_query,
ethdev: allow to set RSS hash computation flags and/or key 1) Add a new function "rss_hash_update" in the PMD API to dynamically update the RSS flags and/or the RSS key used by a NIC to compute the RSS hash of input packets. The new function uses the existing data structure "rte_eth_rss_conf" for the argument that contains the new hash flags and/or the new hash key to use. 2) Add the ixgbe-specific function "ixgbe_dev_rss_hash_update" and the igb-specific function "eth_igb_rss_hash_update" to update the RSS hash configuration of ixgbe and igb controllers respectively. Before changing anything, these 2 functions check that the update RSS operation does not attempt to disable RSS, if RSS was enabled at port initialization time, or does not attempt to enable RSS, if RSS was disabled at port initialization time. Note: Configuring the RSS hash flags and the RSS key used by a NIC consists in updating appropriate PCI registers of the NIC. These operations have been manually tested with the interactive commands "write reg" and "write regbit" of the testpmd application. Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-05-16 08:58:40 +00:00
.rss_hash_update = eth_igb_rss_hash_update,
ethdev: allow to get RSS hash functions and key 1) Add a new function "rss_hash_conf_get" in the PMD API to retrieve the current configuration of the RSS functions and/or of the RSS key used by a NIC to compute the RSS hash of input packets. The new function uses the existing data structure "rte_eth_rss_conf" for returning the RSS hash configuration. 2) Add the ixgbe-specific function "ixgbe_dev_rss_hash_conf_get" and the igb-specific function "eth_igb_rss_hash_conf_get" to retrieve the RSS hash configuration of ixgbe and igb controllers respectively. 3) Add the command "show port X rss-hash [key]" in the testpmd application to display the RSS hash configuration of port X. Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-05-16 08:58:42 +00:00
.rss_hash_conf_get = eth_igb_rss_hash_conf_get,
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
.filter_ctrl = eth_igb_filter_ctrl,
e1000: add multicast MAC address filtering Support the PMD function "set_mc_addr_list" in the "igb", "igb-vf", and "em" Poll Mode Drivers. Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
2015-05-28 15:05:21 +00:00
.set_mc_addr_list = eth_igb_set_mc_addr_list,
e1000: add ieee1588 timestamping Add e1000/igb support for new ethdev APIs to enable and read IEEE1588 PTP timestamps. Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-07-02 15:16:29 +00:00
.timesync_enable = igb_timesync_enable,
.timesync_disable = igb_timesync_disable,
.timesync_read_rx_timestamp = igb_timesync_read_rx_timestamp,
.timesync_read_tx_timestamp = igb_timesync_read_tx_timestamp,
igb: add access to specific device info add function to support ethtool ops: - get_reg_length - get_regs - get_eeprom_length - get_eeprom - set_eeprom Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
.get_reg_length = eth_igb_get_reg_length,
.get_reg = eth_igb_get_regs,
.get_eeprom_length = eth_igb_get_eeprom_length,
.get_eeprom = eth_igb_get_eeprom,
.set_eeprom = eth_igb_set_eeprom,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
};
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* dev_ops for virtual function, bare necessities for basic vf
* operation have been implemented
*/
ethdev: make dev_ops const The ethernet device ops function table should be made const for safety and security. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Neil Horman <nhorman@tuxdriver.com>
2015-04-07 21:21:03 +00:00
static const struct eth_dev_ops igbvf_eth_dev_ops = {
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.dev_configure = igbvf_dev_configure,
.dev_start = igbvf_dev_start,
.dev_stop = igbvf_dev_stop,
.dev_close = igbvf_dev_close,
.link_update = eth_igb_link_update,
.stats_get = eth_igbvf_stats_get,
.stats_reset = eth_igbvf_stats_reset,
.vlan_filter_set = igbvf_vlan_filter_set,
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
.dev_infos_get = eth_igbvf_infos_get,
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.rx_queue_setup = eth_igb_rx_queue_setup,
.rx_queue_release = eth_igb_rx_queue_release,
.tx_queue_setup = eth_igb_tx_queue_setup,
.tx_queue_release = eth_igb_tx_queue_release,
e1000: add multicast MAC address filtering Support the PMD function "set_mc_addr_list" in the "igb", "igb-vf", and "em" Poll Mode Drivers. Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
2015-05-28 15:05:21 +00:00
.set_mc_addr_list = eth_igb_set_mc_addr_list,
igb: add default mac address modifier - set_mac_addr Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
.mac_addr_set = igbvf_default_mac_addr_set,
igb: add access to specific device info add function to support ethtool ops: - get_reg_length - get_regs - get_eeprom_length - get_eeprom - set_eeprom Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
.get_reg_length = igbvf_get_reg_length,
.get_reg = igbvf_get_regs,
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
};
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/**
* Atomically reads the link status information from global
* structure rte_eth_dev.
*
* @param dev
* - Pointer to the structure rte_eth_dev to read from.
* - Pointer to the buffer to be saved with the link status.
*
* @return
* - On success, zero.
* - On failure, negative value.
*/
static inline int
rte_igb_dev_atomic_read_link_status(struct rte_eth_dev *dev,
struct rte_eth_link *link)
{
struct rte_eth_link *dst = link;
struct rte_eth_link *src = &(dev->data->dev_link);
if (rte_atomic64_cmpset((uint64_t *)dst, *(uint64_t *)dst,
*(uint64_t *)src) == 0)
return -1;
return 0;
}
/**
* Atomically writes the link status information into global
* structure rte_eth_dev.
*
* @param dev
* - Pointer to the structure rte_eth_dev to read from.
* - Pointer to the buffer to be saved with the link status.
*
* @return
* - On success, zero.
* - On failure, negative value.
*/
static inline int
rte_igb_dev_atomic_write_link_status(struct rte_eth_dev *dev,
struct rte_eth_link *link)
{
struct rte_eth_link *dst = &(dev->data->dev_link);
struct rte_eth_link *src = link;
if (rte_atomic64_cmpset((uint64_t *)dst, *(uint64_t *)dst,
*(uint64_t *)src) == 0)
return -1;
return 0;
}
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
static inline void
igb_intr_enable(struct rte_eth_dev *dev)
{
struct e1000_interrupt *intr =
E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
E1000_WRITE_REG(hw, E1000_IMS, intr->mask);
E1000_WRITE_FLUSH(hw);
}
static void
igb_intr_disable(struct e1000_hw *hw)
{
E1000_WRITE_REG(hw, E1000_IMC, ~0);
E1000_WRITE_FLUSH(hw);
}
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
static inline int32_t
igb_pf_reset_hw(struct e1000_hw *hw)
{
uint32_t ctrl_ext;
int32_t status;
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
status = e1000_reset_hw(hw);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
ctrl_ext = E1000_READ_REG(hw, E1000_CTRL_EXT);
/* Set PF Reset Done bit so PF/VF Mail Ops can work */
ctrl_ext |= E1000_CTRL_EXT_PFRSTD;
E1000_WRITE_REG(hw, E1000_CTRL_EXT, ctrl_ext);
E1000_WRITE_FLUSH(hw);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
return status;
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static void
igb_identify_hardware(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
hw->vendor_id = dev->pci_dev->id.vendor_id;
hw->device_id = dev->pci_dev->id.device_id;
hw->subsystem_vendor_id = dev->pci_dev->id.subsystem_vendor_id;
hw->subsystem_device_id = dev->pci_dev->id.subsystem_device_id;
e1000_set_mac_type(hw);
/* need to check if it is a vf device below */
}
igb: release software locked semaphores on initialization It may happen that DPDK application gets killed while having acquired locks on the ethernet hardware, causing these locks to be never released. On next restart of the application, DPDK skip those ports because it can not acquire the lock, this may cause some ports (or even complete board if SMBI is locked) to be inaccessible from DPDK application until reboot of the hardware. This patch release locks that are supposed to be locked due to an improper exit of the application. Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-04-08 13:29:58 +00:00
static int
igb_reset_swfw_lock(struct e1000_hw *hw)
{
int ret_val;
/*
* Do mac ops initialization manually here, since we will need
* some function pointers set by this call.
*/
ret_val = e1000_init_mac_params(hw);
if (ret_val)
return ret_val;
/*
* SMBI lock should not fail in this early stage. If this is the case,
* it is due to an improper exit of the application.
* So force the release of the faulty lock.
*/
if (e1000_get_hw_semaphore_generic(hw) < 0) {
e1000: use the right debug macro - We should not use DEBUGOUT* / DEBUGFUNC macros in pmd. These macros come as compat wrappers for base driver. - We should avoid calling RTE_LOG directly as pmd provides a wrapper for logs. Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:46 +00:00
PMD_DRV_LOG(DEBUG, "SMBI lock released");
igb: release software locked semaphores on initialization It may happen that DPDK application gets killed while having acquired locks on the ethernet hardware, causing these locks to be never released. On next restart of the application, DPDK skip those ports because it can not acquire the lock, this may cause some ports (or even complete board if SMBI is locked) to be inaccessible from DPDK application until reboot of the hardware. This patch release locks that are supposed to be locked due to an improper exit of the application. Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-04-08 13:29:58 +00:00
}
e1000_put_hw_semaphore_generic(hw);
if (hw->mac.ops.acquire_swfw_sync != NULL) {
uint16_t mask;
/*
* Phy lock should not fail in this early stage. If this is the case,
* it is due to an improper exit of the application.
* So force the release of the faulty lock.
*/
mask = E1000_SWFW_PHY0_SM << hw->bus.func;
if (hw->bus.func > E1000_FUNC_1)
mask <<= 2;
if (hw->mac.ops.acquire_swfw_sync(hw, mask) < 0) {
e1000: use the right debug macro - We should not use DEBUGOUT* / DEBUGFUNC macros in pmd. These macros come as compat wrappers for base driver. - We should avoid calling RTE_LOG directly as pmd provides a wrapper for logs. Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:46 +00:00
PMD_DRV_LOG(DEBUG, "SWFW phy%d lock released",
hw->bus.func);
igb: release software locked semaphores on initialization It may happen that DPDK application gets killed while having acquired locks on the ethernet hardware, causing these locks to be never released. On next restart of the application, DPDK skip those ports because it can not acquire the lock, this may cause some ports (or even complete board if SMBI is locked) to be inaccessible from DPDK application until reboot of the hardware. This patch release locks that are supposed to be locked due to an improper exit of the application. Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-04-08 13:29:58 +00:00
}
hw->mac.ops.release_swfw_sync(hw, mask);
/*
* This one is more tricky since it is common to all ports; but
* swfw_sync retries last long enough (1s) to be almost sure that if
* lock can not be taken it is due to an improper lock of the
* semaphore.
*/
mask = E1000_SWFW_EEP_SM;
if (hw->mac.ops.acquire_swfw_sync(hw, mask) < 0) {
e1000: use the right debug macro - We should not use DEBUGOUT* / DEBUGFUNC macros in pmd. These macros come as compat wrappers for base driver. - We should avoid calling RTE_LOG directly as pmd provides a wrapper for logs. Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:46 +00:00
PMD_DRV_LOG(DEBUG, "SWFW common locks released");
igb: release software locked semaphores on initialization It may happen that DPDK application gets killed while having acquired locks on the ethernet hardware, causing these locks to be never released. On next restart of the application, DPDK skip those ports because it can not acquire the lock, this may cause some ports (or even complete board if SMBI is locked) to be inaccessible from DPDK application until reboot of the hardware. This patch release locks that are supposed to be locked due to an improper exit of the application. Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-04-08 13:29:58 +00:00
}
hw->mac.ops.release_swfw_sync(hw, mask);
}
return E1000_SUCCESS;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int
ethdev: remove useless parameter in init functions The pointer to struct eth_driver is not used and is already set in struct rte_eth_dev. It's a small cleanup in PMD API which probably needs more attention to make clear what is a driver, a PCI driver, an ethernet driver, etc. Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-03-05 18:24:59 +00:00
eth_igb_dev_init(struct rte_eth_dev *eth_dev)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
int error = 0;
struct rte_pci_device *pci_dev;
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
struct e1000_vfta * shadow_vfta =
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
E1000_DEV_PRIVATE_TO_VFTA(eth_dev->data->dev_private);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(eth_dev->data->dev_private);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(eth_dev->data->dev_private);
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
uint32_t ctrl_ext;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
pci_dev = eth_dev->pci_dev;
eth_dev->dev_ops = &eth_igb_ops;
eth_dev->rx_pkt_burst = &eth_igb_recv_pkts;
eth_dev->tx_pkt_burst = &eth_igb_xmit_pkts;
/* for secondary processes, we don't initialise any further as primary
* has already done this work. Only check we don't need a different
* RX function */
if (rte_eal_process_type() != RTE_PROC_PRIMARY){
if (eth_dev->data->scattered_rx)
eth_dev->rx_pkt_burst = &eth_igb_recv_scattered_pkts;
return 0;
}
pci: support multiple PCI regions per device Need to change PCI code to support multiple I/O regions on a single device. Some devices like VMXNET3 have multiple PCI memory regions, and some have none. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: Intel
2013-05-30 17:12:39 +00:00
hw->hw_addr= (void *)pci_dev->mem_resource[0].addr;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb_identify_hardware(eth_dev);
igb: release software locked semaphores on initialization It may happen that DPDK application gets killed while having acquired locks on the ethernet hardware, causing these locks to be never released. On next restart of the application, DPDK skip those ports because it can not acquire the lock, this may cause some ports (or even complete board if SMBI is locked) to be inaccessible from DPDK application until reboot of the hardware. This patch release locks that are supposed to be locked due to an improper exit of the application. Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-04-08 13:29:58 +00:00
if (e1000_setup_init_funcs(hw, FALSE) != E1000_SUCCESS) {
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
error = -EIO;
goto err_late;
}
e1000_get_bus_info(hw);
igb: release software locked semaphores on initialization It may happen that DPDK application gets killed while having acquired locks on the ethernet hardware, causing these locks to be never released. On next restart of the application, DPDK skip those ports because it can not acquire the lock, this may cause some ports (or even complete board if SMBI is locked) to be inaccessible from DPDK application until reboot of the hardware. This patch release locks that are supposed to be locked due to an improper exit of the application. Signed-off-by: Didier Pallard <didier.pallard@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-04-08 13:29:58 +00:00
/* Reset any pending lock */
if (igb_reset_swfw_lock(hw) != E1000_SUCCESS) {
error = -EIO;
goto err_late;
}
/* Finish initialization */
if (e1000_setup_init_funcs(hw, TRUE) != E1000_SUCCESS) {
error = -EIO;
goto err_late;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
hw->mac.autoneg = 1;
hw->phy.autoneg_wait_to_complete = 0;
hw->phy.autoneg_advertised = E1000_ALL_SPEED_DUPLEX;
/* Copper options */
if (hw->phy.media_type == e1000_media_type_copper) {
hw->phy.mdix = 0; /* AUTO_ALL_MODES */
hw->phy.disable_polarity_correction = 0;
hw->phy.ms_type = e1000_ms_hw_default;
}
/*
* Start from a known state, this is important in reading the nvm
* and mac from that.
*/
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
igb_pf_reset_hw(hw);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Make sure we have a good EEPROM before we read from it */
if (e1000_validate_nvm_checksum(hw) < 0) {
/*
* Some PCI-E parts fail the first check due to
* the link being in sleep state, call it again,
* if it fails a second time its a real issue.
*/
if (e1000_validate_nvm_checksum(hw) < 0) {
PMD_INIT_LOG(ERR, "EEPROM checksum invalid");
error = -EIO;
goto err_late;
}
}
/* Read the permanent MAC address out of the EEPROM */
if (e1000_read_mac_addr(hw) != 0) {
PMD_INIT_LOG(ERR, "EEPROM error while reading MAC address");
error = -EIO;
goto err_late;
}
/* Allocate memory for storing MAC addresses */
eth_dev->data->mac_addrs = rte_zmalloc("e1000",
ETHER_ADDR_LEN * hw->mac.rar_entry_count, 0);
if (eth_dev->data->mac_addrs == NULL) {
PMD_INIT_LOG(ERR, "Failed to allocate %d bytes needed to "
"store MAC addresses",
ETHER_ADDR_LEN * hw->mac.rar_entry_count);
error = -ENOMEM;
goto err_late;
}
/* Copy the permanent MAC address */
ether_addr_copy((struct ether_addr *)hw->mac.addr, &eth_dev->data->mac_addrs[0]);
/* initialize the vfta */
memset(shadow_vfta, 0, sizeof(*shadow_vfta));
/* Now initialize the hardware */
if (igb_hardware_init(hw) != 0) {
PMD_INIT_LOG(ERR, "Hardware initialization failed");
rte_free(eth_dev->data->mac_addrs);
eth_dev->data->mac_addrs = NULL;
error = -ENODEV;
goto err_late;
}
hw->mac.get_link_status = 1;
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
adapter->stopped = 0;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Indicate SOL/IDER usage */
if (e1000_check_reset_block(hw) < 0) {
PMD_INIT_LOG(ERR, "PHY reset is blocked due to"
"SOL/IDER session");
}
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
/* initialize PF if max_vfs not zero */
igb_pf_host_init(eth_dev);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
ctrl_ext = E1000_READ_REG(hw, E1000_CTRL_EXT);
/* Set PF Reset Done bit so PF/VF Mail Ops can work */
ctrl_ext |= E1000_CTRL_EXT_PFRSTD;
E1000_WRITE_REG(hw, E1000_CTRL_EXT, ctrl_ext);
E1000_WRITE_FLUSH(hw);
e1000: fix log level of debug messages Any debug messages about hardware should be under debug (or removed) and reduce customer visible log spam. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
2015-07-09 23:01:07 +00:00
PMD_INIT_LOG(DEBUG, "port_id %d vendorID=0x%x deviceID=0x%x",
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
eth_dev->data->port_id, pci_dev->id.vendor_id,
pci_dev->id.device_id);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
/* enable support intr */
igb_intr_enable(eth_dev);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
TAILQ_INIT(&filter_info->flex_list);
filter_info->flex_mask = 0;
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
TAILQ_INIT(&filter_info->twotuple_list);
filter_info->twotuple_mask = 0;
TAILQ_INIT(&filter_info->fivetuple_list);
filter_info->fivetuple_mask = 0;
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return 0;
err_late:
igb_hw_control_release(hw);
return (error);
}
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
static int
eth_igb_dev_uninit(struct rte_eth_dev *eth_dev)
{
struct rte_pci_device *pci_dev;
struct e1000_hw *hw;
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(eth_dev->data->dev_private);
PMD_INIT_FUNC_TRACE();
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return -EPERM;
hw = E1000_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
pci_dev = eth_dev->pci_dev;
if (adapter->stopped == 0)
eth_igb_close(eth_dev);
eth_dev->dev_ops = NULL;
eth_dev->rx_pkt_burst = NULL;
eth_dev->tx_pkt_burst = NULL;
/* Reset any pending lock */
igb_reset_swfw_lock(hw);
rte_free(eth_dev->data->mac_addrs);
eth_dev->data->mac_addrs = NULL;
/* uninitialize PF if max_vfs not zero */
igb_pf_host_uninit(eth_dev);
/* disable uio intr before callback unregister */
rte_intr_disable(&(pci_dev->intr_handle));
rte_intr_callback_unregister(&(pci_dev->intr_handle),
eth_igb_interrupt_handler, (void *)eth_dev);
return 0;
}
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* Virtual Function device init
*/
static int
ethdev: remove useless parameter in init functions The pointer to struct eth_driver is not used and is already set in struct rte_eth_dev. It's a small cleanup in PMD API which probably needs more attention to make clear what is a driver, a PCI driver, an ethernet driver, etc. Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-03-05 18:24:59 +00:00
eth_igbvf_dev_init(struct rte_eth_dev *eth_dev)
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
{
struct rte_pci_device *pci_dev;
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(eth_dev->data->dev_private);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
int diag;
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
eth_dev->dev_ops = &igbvf_eth_dev_ops;
igb: fix VF init without setup In case of multi-process application, the secondary process can initialize the driver without configuring queues. In this case the Rx/Tx functions were not initialized because it was only done in queue setup. Fix by reproducing the same behaviour as in eth_igb_dev_init(). Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
eth_dev->rx_pkt_burst = &eth_igb_recv_pkts;
eth_dev->tx_pkt_burst = &eth_igb_xmit_pkts;
/* for secondary processes, we don't initialise any further as primary
* has already done this work. Only check we don't need a different
* RX function */
if (rte_eal_process_type() != RTE_PROC_PRIMARY){
if (eth_dev->data->scattered_rx)
eth_dev->rx_pkt_burst = &eth_igb_recv_scattered_pkts;
return 0;
}
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
pci_dev = eth_dev->pci_dev;
hw->device_id = pci_dev->id.device_id;
hw->vendor_id = pci_dev->id.vendor_id;
pci: support multiple PCI regions per device Need to change PCI code to support multiple I/O regions on a single device. Some devices like VMXNET3 have multiple PCI memory regions, and some have none. Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: Intel
2013-05-30 17:12:39 +00:00
hw->hw_addr = (void *)pci_dev->mem_resource[0].addr;
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
adapter->stopped = 0;
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
explicit shared code naming as base driver Intel PMDs are built on top of the base drivers which are provided by Intel and shouldn't be modified to allow easy batch upgrade from Intel. The base driver is a "shared code" between many projects. But in DPDK, the "base driver" naming makes more sense. Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-06-24 08:03:02 +00:00
/* Initialize the shared code (base driver) */
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
diag = e1000_setup_init_funcs(hw, TRUE);
if (diag != 0) {
PMD_INIT_LOG(ERR, "Shared code init failed for igbvf: %d",
diag);
return -EIO;
}
/* init_mailbox_params */
hw->mbx.ops.init_params(hw);
/* Disable the interrupts for VF */
igbvf_intr_disable(hw);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
diag = hw->mac.ops.reset_hw(hw);
/* Allocate memory for storing MAC addresses */
eth_dev->data->mac_addrs = rte_zmalloc("igbvf", ETHER_ADDR_LEN *
hw->mac.rar_entry_count, 0);
if (eth_dev->data->mac_addrs == NULL) {
PMD_INIT_LOG(ERR,
"Failed to allocate %d bytes needed to store MAC "
"addresses",
ETHER_ADDR_LEN * hw->mac.rar_entry_count);
return -ENOMEM;
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* Copy the permanent MAC address */
ether_addr_copy((struct ether_addr *) hw->mac.perm_addr,
&eth_dev->data->mac_addrs[0]);
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(DEBUG, "port %d vendorID=0x%x deviceID=0x%x "
"mac.type=%s",
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
eth_dev->data->port_id, pci_dev->id.vendor_id,
pci_dev->id.device_id, "igb_mac_82576_vf");
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
return 0;
}
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
static int
eth_igbvf_dev_uninit(struct rte_eth_dev *eth_dev)
{
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(eth_dev->data->dev_private);
PMD_INIT_FUNC_TRACE();
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return -EPERM;
if (adapter->stopped == 0)
igbvf_dev_close(eth_dev);
eth_dev->dev_ops = NULL;
eth_dev->rx_pkt_burst = NULL;
eth_dev->tx_pkt_burst = NULL;
rte_free(eth_dev->data->mac_addrs);
eth_dev->data->mac_addrs = NULL;
return 0;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static struct eth_driver rte_igb_pmd = {
drivers: explicit initialization of pci drivers Upcoming drivers will need to be able to support other bus types. This is a transparent change to how struct eth_driver is initialized. It has not function or ABI layout impact, but makes adding a later bus type (Xen, Hyper-V, ...) much easier. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-05-29 15:47:51 +00:00
.pci_drv = {
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.name = "rte_igb_pmd",
.id_table = pci_id_igb_map,
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC |
RTE_PCI_DRV_DETACHABLE,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
},
.eth_dev_init = eth_igb_dev_init,
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
.eth_dev_uninit = eth_igb_dev_uninit,
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
.dev_private_size = sizeof(struct e1000_adapter),
};
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* virtual function driver struct
*/
static struct eth_driver rte_igbvf_pmd = {
drivers: explicit initialization of pci drivers Upcoming drivers will need to be able to support other bus types. This is a transparent change to how struct eth_driver is initialized. It has not function or ABI layout impact, but makes adding a later bus type (Xen, Hyper-V, ...) much easier. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-05-29 15:47:51 +00:00
.pci_drv = {
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.name = "rte_igbvf_pmd",
.id_table = pci_id_igbvf_map,
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_DETACHABLE,
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
},
.eth_dev_init = eth_igbvf_dev_init,
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
.eth_dev_uninit = eth_igbvf_dev_uninit,
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
.dev_private_size = sizeof(struct e1000_adapter),
};
igb: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igb pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igb library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:32 +00:00
static int
rte_igb_pmd_init(const char *name __rte_unused, const char *params __rte_unused)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
rte_eth_driver_register(&rte_igb_pmd);
return 0;
}
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
static void
igb_vmdq_vlan_hw_filter_enable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
/* RCTL: enable VLAN filter since VMDq always use VLAN filter */
uint32_t rctl = E1000_READ_REG(hw, E1000_RCTL);
rctl |= E1000_RCTL_VFE;
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
}
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* VF Driver initialization routine.
* Invoked one at EAL init time.
* Register itself as the [Virtual Poll Mode] Driver of PCI IGB devices.
*/
igbvf: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igbvf pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igbvf library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:33 +00:00
static int
rte_igbvf_pmd_init(const char *name __rte_unused, const char *params __rte_unused)
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
{
e1000: use the right debug macro - We should not use DEBUGOUT* / DEBUGFUNC macros in pmd. These macros come as compat wrappers for base driver. - We should avoid calling RTE_LOG directly as pmd provides a wrapper for logs. Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:46 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
rte_eth_driver_register(&rte_igbvf_pmd);
return (0);
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
eth_igb_configure(struct rte_eth_dev *dev)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
struct e1000_interrupt *intr =
E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
intr->flags |= E1000_FLAG_NEED_LINK_UPDATE;
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return (0);
}
static int
eth_igb_start(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(dev->data->dev_private);
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
int ret, mask;
uint32_t intr_vector = 0;
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
uint32_t ctrl_ext;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Power up the phy. Needed to make the link go Up */
e1000_power_up_phy(hw);
/*
* Packet Buffer Allocation (PBA)
* Writing PBA sets the receive portion of the buffer
* the remainder is used for the transmit buffer.
*/
if (hw->mac.type == e1000_82575) {
uint32_t pba;
pba = E1000_PBA_32K; /* 32K for Rx, 16K for Tx */
E1000_WRITE_REG(hw, E1000_PBA, pba);
}
/* Put the address into the Receive Address Array */
e1000_rar_set(hw, hw->mac.addr, 0);
/* Initialize the hardware */
if (igb_hardware_init(hw)) {
PMD_INIT_LOG(ERR, "Unable to initialize the hardware");
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
return (-EIO);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
adapter->stopped = 0;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: fix dual vlan ethertype Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-02-17 20:21:03 +00:00
E1000_WRITE_REG(hw, E1000_VET, ETHER_TYPE_VLAN << 16 | ETHER_TYPE_VLAN);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
ctrl_ext = E1000_READ_REG(hw, E1000_CTRL_EXT);
/* Set PF Reset Done bit so PF/VF Mail Ops can work */
ctrl_ext |= E1000_CTRL_EXT_PFRSTD;
E1000_WRITE_REG(hw, E1000_CTRL_EXT, ctrl_ext);
E1000_WRITE_FLUSH(hw);
/* configure PF module if SRIOV enabled */
igb_pf_host_configure(dev);
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
/* check and configure queue intr-vector mapping */
if (dev->data->dev_conf.intr_conf.rxq != 0)
intr_vector = dev->data->nb_rx_queues;
if (rte_intr_efd_enable(intr_handle, intr_vector))
return -1;
if (rte_intr_dp_is_en(intr_handle)) {
intr_handle->intr_vec =
rte_zmalloc("intr_vec",
dev->data->nb_rx_queues * sizeof(int), 0);
if (intr_handle->intr_vec == NULL) {
PMD_INIT_LOG(ERR, "Failed to allocate %d rx_queues"
" intr_vec\n", dev->data->nb_rx_queues);
return -ENOMEM;
}
}
/* confiugre msix for rx interrupt */
eth_igb_configure_msix_intr(dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Configure for OS presence */
igb_init_manageability(hw);
eth_igb_tx_init(dev);
/* This can fail when allocating mbufs for descriptor rings */
ret = eth_igb_rx_init(dev);
if (ret) {
PMD_INIT_LOG(ERR, "Unable to initialize RX hardware");
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
igb_dev_clear_queues(dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return ret;
}
e1000_clear_hw_cntrs_base_generic(hw);
/*
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
* VLAN Offload Settings
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
*/
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
mask = ETH_VLAN_STRIP_MASK | ETH_VLAN_FILTER_MASK | \
ETH_VLAN_EXTEND_MASK;
eth_igb_vlan_offload_set(dev, mask);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
if (dev->data->dev_conf.rxmode.mq_mode == ETH_MQ_RX_VMDQ_ONLY) {
/* Enable VLAN filter since VMDq always use VLAN filter */
igb_vmdq_vlan_hw_filter_enable(dev);
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
if ((hw->mac.type == e1000_82576) || (hw->mac.type == e1000_82580) ||
igb: fix i211 support igb_ethdev.c contains function eth_igb_infos_get() which should set number of tx/rx queues supported by the hardware. It contains huge [switch] but there is no i211 case! Also, there are few other places which mention i210, but not mention i211. I didn't check it enough to say it is totally correct. For now I see that it just able to send and receive some packets. Signed-off-by: Sergey Mironov <grrwlf@gmail.com> Reviewed-by: Helin Zhang <helin.zhang@intel.com> [Thomas: fix indent] Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-09-04 08:35:11 +00:00
(hw->mac.type == e1000_i350) || (hw->mac.type == e1000_i210) ||
(hw->mac.type == e1000_i211)) {
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Configure EITR with the maximum possible value (0xFFFF) */
E1000_WRITE_REG(hw, E1000_EITR(0), 0xFFFF);
}
/* Setup link speed and duplex */
switch (dev->data->dev_conf.link_speed) {
case ETH_LINK_SPEED_AUTONEG:
if (dev->data->dev_conf.link_duplex == ETH_LINK_AUTONEG_DUPLEX)
hw->phy.autoneg_advertised = E1000_ALL_SPEED_DUPLEX;
else if (dev->data->dev_conf.link_duplex == ETH_LINK_HALF_DUPLEX)
hw->phy.autoneg_advertised = E1000_ALL_HALF_DUPLEX;
else if (dev->data->dev_conf.link_duplex == ETH_LINK_FULL_DUPLEX)
hw->phy.autoneg_advertised = E1000_ALL_FULL_DUPLEX;
else
goto error_invalid_config;
break;
case ETH_LINK_SPEED_10:
if (dev->data->dev_conf.link_duplex == ETH_LINK_AUTONEG_DUPLEX)
hw->phy.autoneg_advertised = E1000_ALL_10_SPEED;
else if (dev->data->dev_conf.link_duplex == ETH_LINK_HALF_DUPLEX)
hw->phy.autoneg_advertised = ADVERTISE_10_HALF;
else if (dev->data->dev_conf.link_duplex == ETH_LINK_FULL_DUPLEX)
hw->phy.autoneg_advertised = ADVERTISE_10_FULL;
else
goto error_invalid_config;
break;
case ETH_LINK_SPEED_100:
if (dev->data->dev_conf.link_duplex == ETH_LINK_AUTONEG_DUPLEX)
hw->phy.autoneg_advertised = E1000_ALL_100_SPEED;
else if (dev->data->dev_conf.link_duplex == ETH_LINK_HALF_DUPLEX)
hw->phy.autoneg_advertised = ADVERTISE_100_HALF;
else if (dev->data->dev_conf.link_duplex == ETH_LINK_FULL_DUPLEX)
hw->phy.autoneg_advertised = ADVERTISE_100_FULL;
else
goto error_invalid_config;
break;
case ETH_LINK_SPEED_1000:
if ((dev->data->dev_conf.link_duplex == ETH_LINK_AUTONEG_DUPLEX) ||
(dev->data->dev_conf.link_duplex == ETH_LINK_FULL_DUPLEX))
hw->phy.autoneg_advertised = ADVERTISE_1000_FULL;
else
goto error_invalid_config;
break;
case ETH_LINK_SPEED_10000:
default:
goto error_invalid_config;
}
e1000_setup_link(hw);
/* check if lsc interrupt feature is enabled */
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
if (dev->data->dev_conf.intr_conf.lsc != 0) {
if (rte_intr_allow_others(intr_handle)) {
rte_intr_callback_register(intr_handle,
eth_igb_interrupt_handler,
(void *)dev);
eth_igb_lsc_interrupt_setup(dev);
} else
PMD_INIT_LOG(INFO, "lsc won't enable because of"
" no intr multiplex\n");
}
/* check if rxq interrupt is enabled */
if (dev->data->dev_conf.intr_conf.rxq != 0)
eth_igb_rxq_interrupt_setup(dev);
/* enable uio/vfio intr/eventfd mapping */
rte_intr_enable(intr_handle);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
e1000: minor changes Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
/* resume enabled intr since hw reset */
igb_intr_enable(dev);
igb: various updates Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
PMD_INIT_LOG(DEBUG, "<<");
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return (0);
error_invalid_config:
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(ERR, "Invalid link_speed/link_duplex (%u/%u) for port %u",
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
dev->data->dev_conf.link_speed,
dev->data->dev_conf.link_duplex, dev->data->port_id);
e1000: update RX/TX queue configuration Queues are allocated by rte_ether. Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
igb_dev_clear_queues(dev);
return (-EINVAL);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
/*********************************************************************
*
* This routine disables all traffic on the adapter by issuing a
* global reset on the MAC.
*
**********************************************************************/
static void
eth_igb_stop(struct rte_eth_dev *dev)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
struct rte_eth_link link;
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct e1000_flex_filter *p_flex;
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
struct e1000_5tuple_filter *p_5tuple, *p_5tuple_next;
struct e1000_2tuple_filter *p_2tuple, *p_2tuple_next;
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb_intr_disable(hw);
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
/* disable intr eventfd mapping */
rte_intr_disable(intr_handle);
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
igb_pf_reset_hw(hw);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
E1000_WRITE_REG(hw, E1000_WUC, 0);
igb: force phy power up/down Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
/* Set bit for Go Link disconnect */
if (hw->mac.type >= e1000_82580) {
uint32_t phpm_reg;
phpm_reg = E1000_READ_REG(hw, E1000_82580_PHY_POWER_MGMT);
phpm_reg |= E1000_82580_PM_GO_LINKD;
E1000_WRITE_REG(hw, E1000_82580_PHY_POWER_MGMT, phpm_reg);
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Power down the phy. Needed to make the link go Down */
e1000: fix power down of fiber serdes link When stopping interface, only shutdown method for copper cable was called. Signed-off-by: Shelton Chia <jiaxt@sinogrid.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-03-07 03:57:38 +00:00
if (hw->phy.media_type == e1000_media_type_copper)
e1000_power_down_phy(hw);
else
e1000_shutdown_fiber_serdes_link(hw);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb_dev_clear_queues(dev);
/* clear the recorded link status */
memset(&link, 0, sizeof(link));
rte_igb_dev_atomic_write_link_status(dev, &link);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
/* Remove all flex filters of the device */
while ((p_flex = TAILQ_FIRST(&filter_info->flex_list))) {
TAILQ_REMOVE(&filter_info->flex_list, p_flex, entries);
rte_free(p_flex);
}
filter_info->flex_mask = 0;
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
/* Remove all ntuple filters of the device */
for (p_5tuple = TAILQ_FIRST(&filter_info->fivetuple_list);
p_5tuple != NULL; p_5tuple = p_5tuple_next) {
p_5tuple_next = TAILQ_NEXT(p_5tuple, entries);
TAILQ_REMOVE(&filter_info->fivetuple_list,
p_5tuple, entries);
rte_free(p_5tuple);
}
filter_info->fivetuple_mask = 0;
for (p_2tuple = TAILQ_FIRST(&filter_info->twotuple_list);
p_2tuple != NULL; p_2tuple = p_2tuple_next) {
p_2tuple_next = TAILQ_NEXT(p_2tuple, entries);
TAILQ_REMOVE(&filter_info->twotuple_list,
p_2tuple, entries);
rte_free(p_2tuple);
}
filter_info->twotuple_mask = 0;
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
/* Clean datapath event and queue/vec mapping */
rte_intr_efd_disable(intr_handle);
if (intr_handle->intr_vec != NULL) {
rte_free(intr_handle->intr_vec);
intr_handle->intr_vec = NULL;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
static void
eth_igb_close(struct rte_eth_dev *dev)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(dev->data->dev_private);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
struct rte_eth_link link;
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
struct rte_pci_device *pci_dev;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
eth_igb_stop(dev);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
adapter->stopped = 1;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
e1000_phy_hw_reset(hw);
igb_release_manageability(hw);
igb_hw_control_release(hw);
igb: force phy power up/down Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
/* Clear bit for Go Link disconnect */
if (hw->mac.type >= e1000_82580) {
uint32_t phpm_reg;
phpm_reg = E1000_READ_REG(hw, E1000_82580_PHY_POWER_MGMT);
phpm_reg &= ~E1000_82580_PM_GO_LINKD;
E1000_WRITE_REG(hw, E1000_82580_PHY_POWER_MGMT, phpm_reg);
}
e1000: free queue memory when closing Queues were freed in clear function called in stop function. Split clearing and freeing in separate functions to move queue freeing from stop to close function. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:27 +00:00
igb_dev_free_queues(dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
pci_dev = dev->pci_dev;
if (pci_dev->intr_handle.intr_vec) {
rte_free(pci_dev->intr_handle.intr_vec);
pci_dev->intr_handle.intr_vec = NULL;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
memset(&link, 0, sizeof(link));
rte_igb_dev_atomic_write_link_status(dev, &link);
}
static int
igb_get_rx_buffer_size(struct e1000_hw *hw)
{
uint32_t rx_buf_size;
if (hw->mac.type == e1000_82576) {
rx_buf_size = (E1000_READ_REG(hw, E1000_RXPBS) & 0xffff) << 10;
igb: more supported devices Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
} else if (hw->mac.type == e1000_82580 || hw->mac.type == e1000_i350) {
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* PBS needs to be translated according to a lookup table */
rx_buf_size = (E1000_READ_REG(hw, E1000_RXPBS) & 0xf);
rx_buf_size = (uint32_t) e1000_rxpbs_adjust_82580(rx_buf_size);
rx_buf_size = (rx_buf_size << 10);
igb: fix i211 support igb_ethdev.c contains function eth_igb_infos_get() which should set number of tx/rx queues supported by the hardware. It contains huge [switch] but there is no i211 case! Also, there are few other places which mention i210, but not mention i211. I didn't check it enough to say it is totally correct. For now I see that it just able to send and receive some packets. Signed-off-by: Sergey Mironov <grrwlf@gmail.com> Reviewed-by: Helin Zhang <helin.zhang@intel.com> [Thomas: fix indent] Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-09-04 08:35:11 +00:00
} else if (hw->mac.type == e1000_i210 || hw->mac.type == e1000_i211) {
igb: more supported devices Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
rx_buf_size = (E1000_READ_REG(hw, E1000_RXPBS) & 0x3f) << 10;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
} else {
rx_buf_size = (E1000_READ_REG(hw, E1000_PBA) & 0xffff) << 10;
}
return rx_buf_size;
}
/*********************************************************************
*
* Initialize the hardware
*
**********************************************************************/
static int
igb_hardware_init(struct e1000_hw *hw)
{
uint32_t rx_buf_size;
int diag;
/* Let the firmware know the OS is in control */
igb_hw_control_acquire(hw);
/*
* These parameters control the automatic generation (Tx) and
* response (Rx) to Ethernet PAUSE frames.
* - High water mark should allow for at least two standard size (1518)
* frames to be received after sending an XOFF.
* - Low water mark works best when it is very near the high water mark.
* This allows the receiver to restart by sending XON when it has
spelling fixes Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-05-02 23:42:52 +00:00
* drained a bit. Here we use an arbitrary value of 1500 which will
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
* restart after one full frame is pulled from the buffer. There
* could be several smaller frames in the buffer and if so they will
* not trigger the XON until their total number reduces the buffer
* by 1500.
* - The pause time is fairly large at 1000 x 512ns = 512 usec.
*/
rx_buf_size = igb_get_rx_buffer_size(hw);
hw->fc.high_water = rx_buf_size - (ETHER_MAX_LEN * 2);
hw->fc.low_water = hw->fc.high_water - 1500;
hw->fc.pause_time = IGB_FC_PAUSE_TIME;
hw->fc.send_xon = 1;
/* Set Flow control, use the tunable location if sane */
if ((igb_fc_setting != e1000_fc_none) && (igb_fc_setting < 4))
hw->fc.requested_mode = igb_fc_setting;
else
hw->fc.requested_mode = e1000_fc_none;
/* Issue a global reset */
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
igb_pf_reset_hw(hw);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
E1000_WRITE_REG(hw, E1000_WUC, 0);
diag = e1000_init_hw(hw);
if (diag < 0)
return (diag);
igb: fix dual vlan ethertype Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-02-17 20:21:03 +00:00
E1000_WRITE_REG(hw, E1000_VET, ETHER_TYPE_VLAN << 16 | ETHER_TYPE_VLAN);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
e1000_get_phy_info(hw);
e1000_check_for_link(hw);
return (0);
}
/* This function is based on igb_update_stats_counters() in igb/if_igb.c */
static void
eth_igb_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *rte_stats)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_hw_stats *stats =
E1000_DEV_PRIVATE_TO_STATS(dev->data->dev_private);
int pause_frames;
if(hw->phy.media_type == e1000_media_type_copper ||
(E1000_READ_REG(hw, E1000_STATUS) & E1000_STATUS_LU)) {
stats->symerrs +=
E1000_READ_REG(hw,E1000_SYMERRS);
stats->sec += E1000_READ_REG(hw, E1000_SEC);
}
stats->crcerrs += E1000_READ_REG(hw, E1000_CRCERRS);
stats->mpc += E1000_READ_REG(hw, E1000_MPC);
stats->scc += E1000_READ_REG(hw, E1000_SCC);
stats->ecol += E1000_READ_REG(hw, E1000_ECOL);
stats->mcc += E1000_READ_REG(hw, E1000_MCC);
stats->latecol += E1000_READ_REG(hw, E1000_LATECOL);
stats->colc += E1000_READ_REG(hw, E1000_COLC);
stats->dc += E1000_READ_REG(hw, E1000_DC);
stats->rlec += E1000_READ_REG(hw, E1000_RLEC);
stats->xonrxc += E1000_READ_REG(hw, E1000_XONRXC);
stats->xontxc += E1000_READ_REG(hw, E1000_XONTXC);
/*
** For watchdog management we need to know if we have been
** paused during the last interval, so capture that here.
*/
pause_frames = E1000_READ_REG(hw, E1000_XOFFRXC);
stats->xoffrxc += pause_frames;
stats->xofftxc += E1000_READ_REG(hw, E1000_XOFFTXC);
stats->fcruc += E1000_READ_REG(hw, E1000_FCRUC);
stats->prc64 += E1000_READ_REG(hw, E1000_PRC64);
stats->prc127 += E1000_READ_REG(hw, E1000_PRC127);
stats->prc255 += E1000_READ_REG(hw, E1000_PRC255);
stats->prc511 += E1000_READ_REG(hw, E1000_PRC511);
stats->prc1023 += E1000_READ_REG(hw, E1000_PRC1023);
stats->prc1522 += E1000_READ_REG(hw, E1000_PRC1522);
stats->gprc += E1000_READ_REG(hw, E1000_GPRC);
stats->bprc += E1000_READ_REG(hw, E1000_BPRC);
stats->mprc += E1000_READ_REG(hw, E1000_MPRC);
stats->gptc += E1000_READ_REG(hw, E1000_GPTC);
/* For the 64-bit byte counters the low dword must be read first. */
/* Both registers clear on the read of the high dword */
stats->gorc += E1000_READ_REG(hw, E1000_GORCL);
stats->gorc += ((uint64_t)E1000_READ_REG(hw, E1000_GORCH) << 32);
stats->gotc += E1000_READ_REG(hw, E1000_GOTCL);
stats->gotc += ((uint64_t)E1000_READ_REG(hw, E1000_GOTCH) << 32);
stats->rnbc += E1000_READ_REG(hw, E1000_RNBC);
stats->ruc += E1000_READ_REG(hw, E1000_RUC);
stats->rfc += E1000_READ_REG(hw, E1000_RFC);
stats->roc += E1000_READ_REG(hw, E1000_ROC);
stats->rjc += E1000_READ_REG(hw, E1000_RJC);
e1000: fix total byte statistics This patch fixes a bug in reading the 64 bit register reading which was causing the total octets counters to show zero. Now the code reads both the lower and higher 32 bits. Tested in testpmd, byte values are correct. Fixes: 805803445a02 ("e1000: support EM devices (also known as e1000/e1000e)") Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-10-22 15:18:04 +00:00
stats->tor += E1000_READ_REG(hw, E1000_TORL);
stats->tor += ((uint64_t)E1000_READ_REG(hw, E1000_TORH) << 32);
stats->tot += E1000_READ_REG(hw, E1000_TOTL);
stats->tot += ((uint64_t)E1000_READ_REG(hw, E1000_TOTH) << 32);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
stats->tpr += E1000_READ_REG(hw, E1000_TPR);
stats->tpt += E1000_READ_REG(hw, E1000_TPT);
stats->ptc64 += E1000_READ_REG(hw, E1000_PTC64);
stats->ptc127 += E1000_READ_REG(hw, E1000_PTC127);
stats->ptc255 += E1000_READ_REG(hw, E1000_PTC255);
stats->ptc511 += E1000_READ_REG(hw, E1000_PTC511);
stats->ptc1023 += E1000_READ_REG(hw, E1000_PTC1023);
stats->ptc1522 += E1000_READ_REG(hw, E1000_PTC1522);
stats->mptc += E1000_READ_REG(hw, E1000_MPTC);
stats->bptc += E1000_READ_REG(hw, E1000_BPTC);
/* Interrupt Counts */
stats->iac += E1000_READ_REG(hw, E1000_IAC);
stats->icrxptc += E1000_READ_REG(hw, E1000_ICRXPTC);
stats->icrxatc += E1000_READ_REG(hw, E1000_ICRXATC);
stats->ictxptc += E1000_READ_REG(hw, E1000_ICTXPTC);
stats->ictxatc += E1000_READ_REG(hw, E1000_ICTXATC);
stats->ictxqec += E1000_READ_REG(hw, E1000_ICTXQEC);
stats->ictxqmtc += E1000_READ_REG(hw, E1000_ICTXQMTC);
stats->icrxdmtc += E1000_READ_REG(hw, E1000_ICRXDMTC);
stats->icrxoc += E1000_READ_REG(hw, E1000_ICRXOC);
/* Host to Card Statistics */
stats->cbtmpc += E1000_READ_REG(hw, E1000_CBTMPC);
stats->htdpmc += E1000_READ_REG(hw, E1000_HTDPMC);
stats->cbrdpc += E1000_READ_REG(hw, E1000_CBRDPC);
stats->cbrmpc += E1000_READ_REG(hw, E1000_CBRMPC);
stats->rpthc += E1000_READ_REG(hw, E1000_RPTHC);
stats->hgptc += E1000_READ_REG(hw, E1000_HGPTC);
stats->htcbdpc += E1000_READ_REG(hw, E1000_HTCBDPC);
stats->hgorc += E1000_READ_REG(hw, E1000_HGORCL);
stats->hgorc += ((uint64_t)E1000_READ_REG(hw, E1000_HGORCH) << 32);
stats->hgotc += E1000_READ_REG(hw, E1000_HGOTCL);
stats->hgotc += ((uint64_t)E1000_READ_REG(hw, E1000_HGOTCH) << 32);
stats->lenerrs += E1000_READ_REG(hw, E1000_LENERRS);
stats->scvpc += E1000_READ_REG(hw, E1000_SCVPC);
stats->hrmpc += E1000_READ_REG(hw, E1000_HRMPC);
stats->algnerrc += E1000_READ_REG(hw, E1000_ALGNERRC);
stats->rxerrc += E1000_READ_REG(hw, E1000_RXERRC);
stats->tncrs += E1000_READ_REG(hw, E1000_TNCRS);
stats->cexterr += E1000_READ_REG(hw, E1000_CEXTERR);
stats->tsctc += E1000_READ_REG(hw, E1000_TSCTC);
stats->tsctfc += E1000_READ_REG(hw, E1000_TSCTFC);
if (rte_stats == NULL)
return;
/* Rx Errors */
ethdev: add Rx error counters for missed, badcrc and badlen packets Split input error stats to have a better understanding of why packets have been dropped. Keep ierrors field untouched for backward compatibility. Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2014-06-12 21:55:41 +00:00
rte_stats->ibadcrc = stats->crcerrs;
rte_stats->ibadlen = stats->rlec + stats->ruc + stats->roc;
rte_stats->imissed = stats->mpc;
rte_stats->ierrors = rte_stats->ibadcrc +
rte_stats->ibadlen +
rte_stats->imissed +
stats->rxerrc + stats->algnerrc + stats->cexterr;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* Tx Errors */
rte_stats->oerrors = stats->ecol + stats->latecol;
ethdev: add pause frame counters for em/igb/ixgbe Add into the `rte_eth_stats` data structure 4 (64-bit) counters of XOFF/XON pause frames received and sent on a given port. Update em, igb, and ixgbe drivers to return the value of the 4 XOFF/XON counters through the `rte_eth_stats_get` function exported by the DPDK API. Display the value of the 4 XOFF/XON counters in the `testpmd` application. Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2013-06-27 12:54:08 +00:00
/* XON/XOFF pause frames */
rte_stats->tx_pause_xon = stats->xontxc;
rte_stats->rx_pause_xon = stats->xonrxc;
rte_stats->tx_pause_xoff = stats->xofftxc;
rte_stats->rx_pause_xoff = stats->xoffrxc;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
rte_stats->ipackets = stats->gprc;
rte_stats->opackets = stats->gptc;
rte_stats->ibytes = stats->gorc;
rte_stats->obytes = stats->gotc;
}
static void
eth_igb_stats_reset(struct rte_eth_dev *dev)
{
struct e1000_hw_stats *hw_stats =
E1000_DEV_PRIVATE_TO_STATS(dev->data->dev_private);
/* HW registers are cleared on read */
eth_igb_stats_get(dev, NULL);
/* Reset software totals */
memset(hw_stats, 0, sizeof(*hw_stats));
}
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
static void
eth_igbvf_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *rte_stats)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_vf_stats *hw_stats = (struct e1000_vf_stats*)
E1000_DEV_PRIVATE_TO_STATS(dev->data->dev_private);
/* Good Rx packets, include VF loopback */
UPDATE_VF_STAT(E1000_VFGPRC,
hw_stats->last_gprc, hw_stats->gprc);
/* Good Rx octets, include VF loopback */
UPDATE_VF_STAT(E1000_VFGORC,
hw_stats->last_gorc, hw_stats->gorc);
/* Good Tx packets, include VF loopback */
UPDATE_VF_STAT(E1000_VFGPTC,
hw_stats->last_gptc, hw_stats->gptc);
/* Good Tx octets, include VF loopback */
UPDATE_VF_STAT(E1000_VFGOTC,
hw_stats->last_gotc, hw_stats->gotc);
/* Rx Multicst packets */
UPDATE_VF_STAT(E1000_VFMPRC,
hw_stats->last_mprc, hw_stats->mprc);
/* Good Rx loopback packets */
UPDATE_VF_STAT(E1000_VFGPRLBC,
hw_stats->last_gprlbc, hw_stats->gprlbc);
/* Good Rx loopback octets */
UPDATE_VF_STAT(E1000_VFGORLBC,
hw_stats->last_gorlbc, hw_stats->gorlbc);
/* Good Tx loopback packets */
UPDATE_VF_STAT(E1000_VFGPTLBC,
hw_stats->last_gptlbc, hw_stats->gptlbc);
/* Good Tx loopback octets */
UPDATE_VF_STAT(E1000_VFGOTLBC,
hw_stats->last_gotlbc, hw_stats->gotlbc);
if (rte_stats == NULL)
return;
rte_stats->ipackets = hw_stats->gprc;
rte_stats->ibytes = hw_stats->gorc;
rte_stats->opackets = hw_stats->gptc;
rte_stats->obytes = hw_stats->gotc;
rte_stats->imcasts = hw_stats->mprc;
rte_stats->ilbpackets = hw_stats->gprlbc;
rte_stats->ilbbytes = hw_stats->gorlbc;
rte_stats->olbpackets = hw_stats->gptlbc;
rte_stats->olbbytes = hw_stats->gotlbc;
}
static void
eth_igbvf_stats_reset(struct rte_eth_dev *dev)
{
struct e1000_vf_stats *hw_stats = (struct e1000_vf_stats*)
E1000_DEV_PRIVATE_TO_STATS(dev->data->dev_private);
/* Sync HW register to the last stats */
eth_igbvf_stats_get(dev, NULL);
/* reset HW current stats*/
memset(&hw_stats->gprc, 0, sizeof(*hw_stats) -
offsetof(struct e1000_vf_stats, gprc));
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static void
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
eth_igb_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
dev_info->min_rx_bufsize = 256; /* See BSIZE field of RCTL register. */
dev_info->max_rx_pktlen = 0x3FFF; /* See RLPML register. */
dev_info->max_mac_addrs = hw->mac.rar_entry_count;
ethdev: add support for device offload capabilities 1) Make device RX and TX offload capabilities to be returned in the rte_eth_dev_info data structure by the function rte_eth_dev_info_get The following initial set of RX offload capabilities are defined: - VLAN header stripping - IPv4 header checksum check - UDP checksum check - TCP checksum check - TCP large receive offload (LRO) The following initial set of TX offload capabilities are defined: - VLAN header insertion - IPv4 header checksum computation - UDP checksum computation - TCP checksum computation - SCTP checksum computation - TCP segmentation offload (Transmit Segmentation Offload) - UDP segmentation offload 2) Update the eth_dev_infos_get() function of the igb and ixgbe PMDs to return the offload capabilities which are supported by the device and that are effectively managed by the driver. Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2013-04-16 12:42:33 +00:00
dev_info->rx_offload_capa =
DEV_RX_OFFLOAD_VLAN_STRIP |
DEV_RX_OFFLOAD_IPV4_CKSUM |
DEV_RX_OFFLOAD_UDP_CKSUM |
DEV_RX_OFFLOAD_TCP_CKSUM;
dev_info->tx_offload_capa =
DEV_TX_OFFLOAD_VLAN_INSERT |
DEV_TX_OFFLOAD_IPV4_CKSUM |
DEV_TX_OFFLOAD_UDP_CKSUM |
DEV_TX_OFFLOAD_TCP_CKSUM |
DEV_TX_OFFLOAD_SCTP_CKSUM;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
switch (hw->mac.type) {
case e1000_82575:
dev_info->max_rx_queues = 4;
dev_info->max_tx_queues = 4;
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
dev_info->max_vmdq_pools = 0;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
break;
case e1000_82576:
dev_info->max_rx_queues = 16;
dev_info->max_tx_queues = 16;
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
dev_info->max_vmdq_pools = ETH_8_POOLS;
igb: new VMDQ argument Assign new VMDQ arguments with correct values. Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-04 10:01:25 +00:00
dev_info->vmdq_queue_num = 16;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
break;
case e1000_82580:
dev_info->max_rx_queues = 8;
dev_info->max_tx_queues = 8;
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
dev_info->max_vmdq_pools = ETH_8_POOLS;
igb: new VMDQ argument Assign new VMDQ arguments with correct values. Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-04 10:01:25 +00:00
dev_info->vmdq_queue_num = 8;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
break;
case e1000_i350:
dev_info->max_rx_queues = 8;
dev_info->max_tx_queues = 8;
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
dev_info->max_vmdq_pools = ETH_8_POOLS;
igb: new VMDQ argument Assign new VMDQ arguments with correct values. Signed-off-by: Chen Jing D(Mark) <jing.d.chen@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-04 10:01:25 +00:00
dev_info->vmdq_queue_num = 8;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
break;
igb: add i354 support Signed-off-by: Intel
2013-11-08 02:00:00 +00:00
case e1000_i354:
dev_info->max_rx_queues = 8;
dev_info->max_tx_queues = 8;
break;
igb: more supported devices Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
case e1000_i210:
dev_info->max_rx_queues = 4;
dev_info->max_tx_queues = 4;
igb: add VMDq support Signed-off-by: Intel
2013-09-18 10:00:00 +00:00
dev_info->max_vmdq_pools = 0;
igb: more supported devices Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
break;
igb: fix i211 support igb_ethdev.c contains function eth_igb_infos_get() which should set number of tx/rx queues supported by the hardware. It contains huge [switch] but there is no i211 case! Also, there are few other places which mention i210, but not mention i211. I didn't check it enough to say it is totally correct. For now I see that it just able to send and receive some packets. Signed-off-by: Sergey Mironov <grrwlf@gmail.com> Reviewed-by: Helin Zhang <helin.zhang@intel.com> [Thomas: fix indent] Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-09-04 08:35:11 +00:00
case e1000_i211:
dev_info->max_rx_queues = 2;
dev_info->max_tx_queues = 2;
dev_info->max_vmdq_pools = 0;
break;
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
default:
/* Should not happen */
break;
}
ethdev: add the hash key size per device To support querying hash key size per port, an new field of 'hash_key_size' was added in 'struct rte_eth_dev_info' for storing hash key size in bytes. The correct hash key size in bytes should be filled into the 'struct rte_eth_dev_info', to support querying it. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-06-12 07:33:56 +00:00
dev_info->hash_key_size = IGB_HKEY_MAX_INDEX * sizeof(uint32_t);
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
dev_info->reta_size = ETH_RSS_RETA_SIZE_128;
ethdev: unification of RSS offload types RSS offload types were defined separately for 1/10G and 40G NICs, and have no relationship with flow types. The modifications are to unify all RSS offload types for all PMDs. Unified RSS offload types have new and common names which can be used for any PMD or applications, and decouple from specific hardwares. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jingjing Wu <jingjing.wu@intel.com> [Thomas: merge with fm10k] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2015-02-04 07:16:32 +00:00
dev_info->flow_type_rss_offloads = IGB_RSS_OFFLOAD_ALL;
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
dev_info->default_rxconf = (struct rte_eth_rxconf) {
.rx_thresh = {
.pthresh = IGB_DEFAULT_RX_PTHRESH,
.hthresh = IGB_DEFAULT_RX_HTHRESH,
.wthresh = IGB_DEFAULT_RX_WTHRESH,
},
.rx_free_thresh = IGB_DEFAULT_RX_FREE_THRESH,
.rx_drop_en = 0,
};
dev_info->default_txconf = (struct rte_eth_txconf) {
.tx_thresh = {
.pthresh = IGB_DEFAULT_TX_PTHRESH,
.hthresh = IGB_DEFAULT_TX_HTHRESH,
.wthresh = IGB_DEFAULT_TX_WTHRESH,
},
.txq_flags = 0,
};
}
static void
eth_igbvf_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
dev_info->min_rx_bufsize = 256; /* See BSIZE field of RCTL register. */
dev_info->max_rx_pktlen = 0x3FFF; /* See RLPML register. */
dev_info->max_mac_addrs = hw->mac.rar_entry_count;
dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP |
DEV_RX_OFFLOAD_IPV4_CKSUM |
DEV_RX_OFFLOAD_UDP_CKSUM |
DEV_RX_OFFLOAD_TCP_CKSUM;
dev_info->tx_offload_capa = DEV_TX_OFFLOAD_VLAN_INSERT |
DEV_TX_OFFLOAD_IPV4_CKSUM |
DEV_TX_OFFLOAD_UDP_CKSUM |
DEV_TX_OFFLOAD_TCP_CKSUM |
DEV_TX_OFFLOAD_SCTP_CKSUM;
switch (hw->mac.type) {
igb: more supported devices Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
case e1000_vfadapt:
dev_info->max_rx_queues = 2;
dev_info->max_tx_queues = 2;
break;
case e1000_vfadapt_i350:
dev_info->max_rx_queues = 1;
dev_info->max_tx_queues = 1;
break;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
default:
/* Should not happen */
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
break;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
igb: set default Rx/Tx configuration Many sample apps use duplicated code to set rte_eth_txconf and rte_eth_rxconf structures. This patch allows the user to get a default optimal RX/TX configuration through rte_eth_dev_info get, and still any parameters may be tweaked as wished, before setting up queues. Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com> Reviewed-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: David Marchand <david.marchand@6wind.com> [Thomas: split patch]
2014-10-01 09:49:04 +00:00
dev_info->default_rxconf = (struct rte_eth_rxconf) {
.rx_thresh = {
.pthresh = IGB_DEFAULT_RX_PTHRESH,
.hthresh = IGB_DEFAULT_RX_HTHRESH,
.wthresh = IGB_DEFAULT_RX_WTHRESH,
},
.rx_free_thresh = IGB_DEFAULT_RX_FREE_THRESH,
.rx_drop_en = 0,
};
dev_info->default_txconf = (struct rte_eth_txconf) {
.tx_thresh = {
.pthresh = IGB_DEFAULT_TX_PTHRESH,
.hthresh = IGB_DEFAULT_TX_HTHRESH,
.wthresh = IGB_DEFAULT_TX_WTHRESH,
},
.txq_flags = 0,
};
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
/* return 0 means link status changed, -1 means not changed */
static int
eth_igb_link_update(struct rte_eth_dev *dev, int wait_to_complete)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct rte_eth_link link, old;
int link_check, count;
link_check = 0;
hw->mac.get_link_status = 1;
/* possible wait-to-complete in up to 9 seconds */
for (count = 0; count < IGB_LINK_UPDATE_CHECK_TIMEOUT; count ++) {
/* Read the real link status */
switch (hw->phy.media_type) {
case e1000_media_type_copper:
/* Do the work to read phy */
e1000_check_for_link(hw);
link_check = !hw->mac.get_link_status;
break;
case e1000_media_type_fiber:
e1000_check_for_link(hw);
link_check = (E1000_READ_REG(hw, E1000_STATUS) &
E1000_STATUS_LU);
break;
case e1000_media_type_internal_serdes:
e1000_check_for_link(hw);
link_check = hw->mac.serdes_has_link;
break;
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* VF device is type_unknown */
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
case e1000_media_type_unknown:
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
eth_igbvf_link_update(hw);
link_check = !hw->mac.get_link_status;
break;
default:
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
break;
}
if (link_check || wait_to_complete == 0)
break;
rte_delay_ms(IGB_LINK_UPDATE_CHECK_INTERVAL);
}
memset(&link, 0, sizeof(link));
rte_igb_dev_atomic_read_link_status(dev, &link);
old = link;
/* Now we check if a transition has happened */
if (link_check) {
hw->mac.ops.get_link_up_info(hw, &link.link_speed,
&link.link_duplex);
link.link_status = 1;
} else if (!link_check) {
link.link_speed = 0;
link.link_duplex = 0;
link.link_status = 0;
}
rte_igb_dev_atomic_write_link_status(dev, &link);
/* not changed */
if (old.link_status == link.link_status)
return -1;
/* changed */
return 0;
}
/*
* igb_hw_control_acquire sets CTRL_EXT:DRV_LOAD bit.
* For ASF and Pass Through versions of f/w this means
* that the driver is loaded.
*/
static void
igb_hw_control_acquire(struct e1000_hw *hw)
{
uint32_t ctrl_ext;
/* Let firmware know the driver has taken over */
ctrl_ext = E1000_READ_REG(hw, E1000_CTRL_EXT);
E1000_WRITE_REG(hw, E1000_CTRL_EXT, ctrl_ext | E1000_CTRL_EXT_DRV_LOAD);
}
/*
* igb_hw_control_release resets CTRL_EXT:DRV_LOAD bit.
* For ASF and Pass Through versions of f/w this means that the
* driver is no longer loaded.
*/
static void
igb_hw_control_release(struct e1000_hw *hw)
{
uint32_t ctrl_ext;
/* Let firmware taken over control of h/w */
ctrl_ext = E1000_READ_REG(hw, E1000_CTRL_EXT);
E1000_WRITE_REG(hw, E1000_CTRL_EXT,
ctrl_ext & ~E1000_CTRL_EXT_DRV_LOAD);
}
/*
* Bit of a misnomer, what this really means is
* to enable OS management of the system... aka
* to disable special hardware management features.
*/
static void
igb_init_manageability(struct e1000_hw *hw)
{
if (e1000_enable_mng_pass_thru(hw)) {
uint32_t manc2h = E1000_READ_REG(hw, E1000_MANC2H);
uint32_t manc = E1000_READ_REG(hw, E1000_MANC);
/* disable hardware interception of ARP */
manc &= ~(E1000_MANC_ARP_EN);
/* enable receiving management packets to the host */
manc |= E1000_MANC_EN_MNG2HOST;
manc2h |= 1 << 5; /* Mng Port 623 */
manc2h |= 1 << 6; /* Mng Port 664 */
E1000_WRITE_REG(hw, E1000_MANC2H, manc2h);
E1000_WRITE_REG(hw, E1000_MANC, manc);
}
}
static void
igb_release_manageability(struct e1000_hw *hw)
{
if (e1000_enable_mng_pass_thru(hw)) {
uint32_t manc = E1000_READ_REG(hw, E1000_MANC);
manc |= E1000_MANC_ARP_EN;
manc &= ~E1000_MANC_EN_MNG2HOST;
E1000_WRITE_REG(hw, E1000_MANC, manc);
}
}
static void
eth_igb_promiscuous_enable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t rctl;
rctl = E1000_READ_REG(hw, E1000_RCTL);
rctl |= (E1000_RCTL_UPE | E1000_RCTL_MPE);
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
}
static void
eth_igb_promiscuous_disable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t rctl;
rctl = E1000_READ_REG(hw, E1000_RCTL);
rctl &= (~E1000_RCTL_UPE);
if (dev->data->all_multicast == 1)
rctl |= E1000_RCTL_MPE;
else
rctl &= (~E1000_RCTL_MPE);
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
}
static void
eth_igb_allmulticast_enable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t rctl;
rctl = E1000_READ_REG(hw, E1000_RCTL);
rctl |= E1000_RCTL_MPE;
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
}
static void
eth_igb_allmulticast_disable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t rctl;
if (dev->data->promiscuous == 1)
return; /* must remain in all_multicast mode */
rctl = E1000_READ_REG(hw, E1000_RCTL);
rctl &= (~E1000_RCTL_MPE);
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
}
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
static int
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
eth_igb_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_vfta * shadow_vfta =
E1000_DEV_PRIVATE_TO_VFTA(dev->data->dev_private);
uint32_t vfta;
uint32_t vid_idx;
uint32_t vid_bit;
vid_idx = (uint32_t) ((vlan_id >> E1000_VFTA_ENTRY_SHIFT) &
E1000_VFTA_ENTRY_MASK);
vid_bit = (uint32_t) (1 << (vlan_id & E1000_VFTA_ENTRY_BIT_SHIFT_MASK));
vfta = E1000_READ_REG_ARRAY(hw, E1000_VFTA, vid_idx);
if (on)
vfta |= vid_bit;
else
vfta &= ~vid_bit;
E1000_WRITE_REG_ARRAY(hw, E1000_VFTA, vid_idx, vfta);
/* update local VFTA copy */
shadow_vfta->vfta[vid_idx] = vfta;
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
return 0;
}
static void
eth_igb_vlan_tpid_set(struct rte_eth_dev *dev, uint16_t tpid)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t reg = ETHER_TYPE_VLAN ;
reg |= (tpid << 16);
E1000_WRITE_REG(hw, E1000_VET, reg);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
static void
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
igb_vlan_hw_filter_disable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t reg;
/* Filter Table Disable */
reg = E1000_READ_REG(hw, E1000_RCTL);
reg &= ~E1000_RCTL_CFIEN;
reg &= ~E1000_RCTL_VFE;
E1000_WRITE_REG(hw, E1000_RCTL, reg);
}
static void
igb_vlan_hw_filter_enable(struct rte_eth_dev *dev)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_vfta * shadow_vfta =
E1000_DEV_PRIVATE_TO_VFTA(dev->data->dev_private);
uint32_t reg;
int i;
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* Filter Table Enable, CFI not used for packet acceptance */
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
reg = E1000_READ_REG(hw, E1000_RCTL);
reg &= ~E1000_RCTL_CFIEN;
reg |= E1000_RCTL_VFE;
E1000_WRITE_REG(hw, E1000_RCTL, reg);
/* restore VFTA table */
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
for (i = 0; i < IGB_VFTA_SIZE; i++)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
E1000_WRITE_REG_ARRAY(hw, E1000_VFTA, i, shadow_vfta->vfta[i]);
}
static void
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
igb_vlan_hw_strip_disable(struct rte_eth_dev *dev)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t reg;
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* VLAN Mode Disable */
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
reg = E1000_READ_REG(hw, E1000_CTRL);
reg &= ~E1000_CTRL_VME;
E1000_WRITE_REG(hw, E1000_CTRL, reg);
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
}
static void
igb_vlan_hw_strip_enable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t reg;
/* VLAN Mode Enable */
reg = E1000_READ_REG(hw, E1000_CTRL);
reg |= E1000_CTRL_VME;
E1000_WRITE_REG(hw, E1000_CTRL, reg);
}
static void
igb_vlan_hw_extend_disable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t reg;
/* CTRL_EXT: Extended VLAN */
reg = E1000_READ_REG(hw, E1000_CTRL_EXT);
reg &= ~E1000_CTRL_EXT_EXTEND_VLAN;
E1000_WRITE_REG(hw, E1000_CTRL_EXT, reg);
igb: fix max RX packet size and support dual VLAN When in 'normal' mode, hardware is designed to receive up to 1522 bytes. When in 'jumbo' mode, RLPML register must be updated so that hardware accepts a 802.1q vlan header. If dual vlan is enabled, then a second vlan header is expected, so update RLPML register according to dual vlan support. Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
/* Update maximum packet length */
if (dev->data->dev_conf.rxmode.jumbo_frame == 1)
E1000_WRITE_REG(hw, E1000_RLPML,
dev->data->dev_conf.rxmode.max_rx_pkt_len +
VLAN_TAG_SIZE);
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
}
static void
igb_vlan_hw_extend_enable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t reg;
/* CTRL_EXT: Extended VLAN */
reg = E1000_READ_REG(hw, E1000_CTRL_EXT);
reg |= E1000_CTRL_EXT_EXTEND_VLAN;
E1000_WRITE_REG(hw, E1000_CTRL_EXT, reg);
igb: fix max RX packet size and support dual VLAN When in 'normal' mode, hardware is designed to receive up to 1522 bytes. When in 'jumbo' mode, RLPML register must be updated so that hardware accepts a 802.1q vlan header. If dual vlan is enabled, then a second vlan header is expected, so update RLPML register according to dual vlan support. Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
/* Update maximum packet length */
if (dev->data->dev_conf.rxmode.jumbo_frame == 1)
E1000_WRITE_REG(hw, E1000_RLPML,
dev->data->dev_conf.rxmode.max_rx_pkt_len +
2 * VLAN_TAG_SIZE);
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
}
static void
eth_igb_vlan_offload_set(struct rte_eth_dev *dev, int mask)
{
if(mask & ETH_VLAN_STRIP_MASK){
if (dev->data->dev_conf.rxmode.hw_vlan_strip)
igb_vlan_hw_strip_enable(dev);
else
igb_vlan_hw_strip_disable(dev);
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
if(mask & ETH_VLAN_FILTER_MASK){
if (dev->data->dev_conf.rxmode.hw_vlan_filter)
igb_vlan_hw_filter_enable(dev);
else
igb_vlan_hw_filter_disable(dev);
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
e1000: add vlan offload support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
if(mask & ETH_VLAN_EXTEND_MASK){
if (dev->data->dev_conf.rxmode.hw_vlan_extend)
igb_vlan_hw_extend_enable(dev);
else
igb_vlan_hw_extend_disable(dev);
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
/**
* It enables the interrupt mask and then enable the interrupt.
*
* @param dev
* Pointer to struct rte_eth_dev.
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
eth_igb_lsc_interrupt_setup(struct rte_eth_dev *dev)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
struct e1000_interrupt *intr =
E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
intr->mask |= E1000_ICR_LSC;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return 0;
}
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
/* It clears the interrupt causes and enables the interrupt.
* It will be called once only during nic initialized.
*
* @param dev
* Pointer to struct rte_eth_dev.
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int eth_igb_rxq_interrupt_setup(struct rte_eth_dev *dev)
{
uint32_t mask, regval;
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct rte_eth_dev_info dev_info;
memset(&dev_info, 0, sizeof(dev_info));
eth_igb_infos_get(dev, &dev_info);
mask = 0xFFFFFFFF >> (32 - dev_info.max_rx_queues);
regval = E1000_READ_REG(hw, E1000_EIMS);
E1000_WRITE_REG(hw, E1000_EIMS, regval | mask);
return 0;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/*
* It reads ICR and gets interrupt causes, check it and set a bit flag
* to update link status.
*
* @param dev
* Pointer to struct rte_eth_dev.
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
eth_igb_interrupt_get_status(struct rte_eth_dev *dev)
{
uint32_t icr;
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_interrupt *intr =
E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
igb_intr_disable(hw);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* read-on-clear nic registers here */
icr = E1000_READ_REG(hw, E1000_ICR);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
intr->flags = 0;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
if (icr & E1000_ICR_LSC) {
intr->flags |= E1000_FLAG_NEED_LINK_UPDATE;
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
if (icr & E1000_ICR_VMMB)
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
intr->flags |= E1000_FLAG_MAILBOX;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return 0;
}
/*
* It executes link_update after knowing an interrupt is prsent.
*
* @param dev
* Pointer to struct rte_eth_dev.
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
eth_igb_interrupt_action(struct rte_eth_dev *dev)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_interrupt *intr =
E1000_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
uint32_t tctl, rctl;
struct rte_eth_link link;
int ret;
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
if (intr->flags & E1000_FLAG_MAILBOX) {
igb_pf_mbx_process(dev);
intr->flags &= ~E1000_FLAG_MAILBOX;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
igb_intr_enable(dev);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
rte_intr_enable(&(dev->pci_dev->intr_handle));
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
if (intr->flags & E1000_FLAG_NEED_LINK_UPDATE) {
intr->flags &= ~E1000_FLAG_NEED_LINK_UPDATE;
/* set get_link_status to check register later */
hw->mac.get_link_status = 1;
ret = eth_igb_link_update(dev, 0);
/* check if link has changed */
if (ret < 0)
return 0;
memset(&link, 0, sizeof(link));
rte_igb_dev_atomic_read_link_status(dev, &link);
if (link.link_status) {
PMD_INIT_LOG(INFO,
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
" Port %d: Link Up - speed %u Mbps - %s",
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
dev->data->port_id,
(unsigned)link.link_speed,
link.link_duplex == ETH_LINK_FULL_DUPLEX ?
"full-duplex" : "half-duplex");
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
} else {
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(INFO, " Port %d: Link Down",
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
dev->data->port_id);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
}
e1000: fix log level of debug messages Any debug messages about hardware should be under debug (or removed) and reduce customer visible log spam. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
2015-07-09 23:01:07 +00:00
PMD_INIT_LOG(DEBUG, "PCI Address: %04d:%02d:%02d:%d",
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
dev->pci_dev->addr.domain,
dev->pci_dev->addr.bus,
dev->pci_dev->addr.devid,
dev->pci_dev->addr.function);
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
tctl = E1000_READ_REG(hw, E1000_TCTL);
rctl = E1000_READ_REG(hw, E1000_RCTL);
if (link.link_status) {
/* enable Tx/Rx */
tctl |= E1000_TCTL_EN;
rctl |= E1000_RCTL_EN;
} else {
/* disable Tx/Rx */
tctl &= ~E1000_TCTL_EN;
rctl &= ~E1000_RCTL_EN;
}
E1000_WRITE_REG(hw, E1000_TCTL, tctl);
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
E1000_WRITE_FLUSH(hw);
_rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
return 0;
}
/**
* Interrupt handler which shall be registered at first.
*
* @param handle
* Pointer to interrupt handle.
* @param param
* The address of parameter (struct rte_eth_dev *) regsitered before.
*
* @return
* void
*/
static void
lib: fix various compilation warnings Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
eth_igb_interrupt_handler(__rte_unused struct rte_intr_handle *handle,
void *param)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
{
struct rte_eth_dev *dev = (struct rte_eth_dev *)param;
eth_igb_interrupt_get_status(dev);
eth_igb_interrupt_action(dev);
}
static int
eth_igb_led_on(struct rte_eth_dev *dev)
{
struct e1000_hw *hw;
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
return (e1000_led_on(hw) == E1000_SUCCESS ? 0 : -ENOTSUP);
}
static int
eth_igb_led_off(struct rte_eth_dev *dev)
{
struct e1000_hw *hw;
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
return (e1000_led_off(hw) == E1000_SUCCESS ? 0 : -ENOTSUP);
}
ethdev: retrieve flow control configuration This patch adds a new function in ethdev api to retrieve current flow control configuration. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Zijie Pan <zijie.pan@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:26 +00:00
static int
eth_igb_flow_ctrl_get(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
{
struct e1000_hw *hw;
uint32_t ctrl;
int tx_pause;
int rx_pause;
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
fc_conf->pause_time = hw->fc.pause_time;
fc_conf->high_water = hw->fc.high_water;
fc_conf->low_water = hw->fc.low_water;
fc_conf->send_xon = hw->fc.send_xon;
ethdev: autoneg parameter in flow control accessors Add autoneg field in flow control parameters. This makes it easier to understand why changing some parameters does not always have the expected result. Changing autoneg is not supported at the moment. Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:27 +00:00
fc_conf->autoneg = hw->mac.autoneg;
ethdev: retrieve flow control configuration This patch adds a new function in ethdev api to retrieve current flow control configuration. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Zijie Pan <zijie.pan@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:26 +00:00
/*
* Return rx_pause and tx_pause status according to actual setting of
* the TFCE and RFCE bits in the CTRL register.
*/
ctrl = E1000_READ_REG(hw, E1000_CTRL);
if (ctrl & E1000_CTRL_TFCE)
tx_pause = 1;
else
tx_pause = 0;
if (ctrl & E1000_CTRL_RFCE)
rx_pause = 1;
else
rx_pause = 0;
if (rx_pause && tx_pause)
fc_conf->mode = RTE_FC_FULL;
else if (rx_pause)
fc_conf->mode = RTE_FC_RX_PAUSE;
else if (tx_pause)
fc_conf->mode = RTE_FC_TX_PAUSE;
else
fc_conf->mode = RTE_FC_NONE;
return 0;
}
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static int
eth_igb_flow_ctrl_set(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
{
struct e1000_hw *hw;
int err;
enum e1000_fc_mode rte_fcmode_2_e1000_fcmode[] = {
e1000_fc_none,
e1000_fc_rx_pause,
e1000_fc_tx_pause,
e1000_fc_full
};
uint32_t rx_buf_size;
uint32_t max_high_water;
e1000: add MAC control forward Signed-off-by: Intel
2013-11-08 02:00:00 +00:00
uint32_t rctl;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
ethdev: autoneg parameter in flow control accessors Add autoneg field in flow control parameters. This makes it easier to understand why changing some parameters does not always have the expected result. Changing autoneg is not supported at the moment. Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:27 +00:00
if (fc_conf->autoneg != hw->mac.autoneg)
return -ENOTSUP;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
rx_buf_size = igb_get_rx_buffer_size(hw);
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(DEBUG, "Rx packet buffer size = 0x%x", rx_buf_size);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
/* At least reserve one Ethernet frame for watermark */
max_high_water = rx_buf_size - ETHER_MAX_LEN;
if ((fc_conf->high_water > max_high_water) ||
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
(fc_conf->high_water < fc_conf->low_water)) {
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(ERR, "e1000 incorrect high/low water value");
PMD_INIT_LOG(ERR, "high water must <= 0x%x", max_high_water);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return (-EINVAL);
}
hw->fc.requested_mode = rte_fcmode_2_e1000_fcmode[fc_conf->mode];
hw->fc.pause_time = fc_conf->pause_time;
hw->fc.high_water = fc_conf->high_water;
hw->fc.low_water = fc_conf->low_water;
hw->fc.send_xon = fc_conf->send_xon;
err = e1000_setup_link_generic(hw);
if (err == E1000_SUCCESS) {
e1000: add MAC control forward Signed-off-by: Intel
2013-11-08 02:00:00 +00:00
/* check if we want to forward MAC frames - driver doesn't have native
* capability to do that, so we'll write the registers ourselves */
rctl = E1000_READ_REG(hw, E1000_RCTL);
/* set or clear MFLCN.PMCF bit depending on configuration */
if (fc_conf->mac_ctrl_frame_fwd != 0)
rctl |= E1000_RCTL_PMCF;
else
rctl &= ~E1000_RCTL_PMCF;
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
E1000_WRITE_FLUSH(hw);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return 0;
}
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(ERR, "e1000_setup_link_generic = 0x%x", err);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
return (-EIO);
}
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
#define E1000_RAH_POOLSEL_SHIFT (18)
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
static void
eth_igb_rar_set(struct rte_eth_dev *dev, struct ether_addr *mac_addr,
uint32_t index, __rte_unused uint32_t pool)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
uint32_t rah;
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
e1000_rar_set(hw, mac_addr->addr_bytes, index);
igb: add PF support Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
rah = E1000_READ_REG(hw, E1000_RAH(index));
rah |= (0x1 << (E1000_RAH_POOLSEL_SHIFT + pool));
E1000_WRITE_REG(hw, E1000_RAH(index), rah);
first public release version 1.2.3 Signed-off-by: Intel
2012-09-04 12:54:00 +00:00
}
static void
eth_igb_rar_clear(struct rte_eth_dev *dev, uint32_t index)
{
uint8_t addr[ETHER_ADDR_LEN];
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
memset(addr, 0, sizeof(addr));
e1000_rar_set(hw, addr, index);
}
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
igb: add default mac address modifier - set_mac_addr Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static void
eth_igb_default_mac_addr_set(struct rte_eth_dev *dev,
struct ether_addr *addr)
{
eth_igb_rar_clear(dev, 0);
eth_igb_rar_set(dev, (void *)addr, 0, 0);
}
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* Virtual Function operations
*/
static void
igbvf_intr_disable(struct e1000_hw *hw)
{
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* Clear interrupt mask to stop from interrupts being generated */
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
E1000_WRITE_REG(hw, E1000_EIMC, 0xFFFF);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
E1000_WRITE_FLUSH(hw);
}
static void
igbvf_stop_adapter(struct rte_eth_dev *dev)
{
u32 reg_val;
u16 i;
struct rte_eth_dev_info dev_info;
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
memset(&dev_info, 0, sizeof(dev_info));
igb: add redirection table size in device info As more and more information are different between PF and VF, ops of 'dev_infos_get' has been implemented respectively. In addition, new field of 'reta_size' has been added in 'struct rte_eth_dev_info' for returning redirection table size. Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-11-15 16:03:40 +00:00
eth_igbvf_infos_get(dev, &dev_info);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* Clear interrupt mask to stop from interrupts being generated */
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
igbvf_intr_disable(hw);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* Clear any pending interrupts, flush previous writes */
E1000_READ_REG(hw, E1000_EICR);
/* Disable the transmit unit. Each queue must be disabled. */
for (i = 0; i < dev_info.max_tx_queues; i++)
E1000_WRITE_REG(hw, E1000_TXDCTL(i), E1000_TXDCTL_SWFLSH);
/* Disable the receive unit by stopping each queue */
for (i = 0; i < dev_info.max_rx_queues; i++) {
reg_val = E1000_READ_REG(hw, E1000_RXDCTL(i));
reg_val &= ~E1000_RXDCTL_QUEUE_ENABLE;
E1000_WRITE_REG(hw, E1000_RXDCTL(i), reg_val);
while (E1000_READ_REG(hw, E1000_RXDCTL(i)) & E1000_RXDCTL_QUEUE_ENABLE)
;
}
/* flush all queues disables */
E1000_WRITE_FLUSH(hw);
msec_delay(2);
}
static int eth_igbvf_link_update(struct e1000_hw *hw)
{
struct e1000_mbx_info *mbx = &hw->mbx;
struct e1000_mac_info *mac = &hw->mac;
int ret_val = E1000_SUCCESS;
PMD_INIT_LOG(DEBUG, "e1000_check_for_link_vf");
/*
* We only want to run this if there has been a rst asserted.
* in this case that could mean a link change, device reset,
* or a virtual function reset
*/
/* If we were hit with a reset or timeout drop the link */
if (!e1000_check_for_rst(hw, 0) || !mbx->timeout)
mac->get_link_status = TRUE;
if (!mac->get_link_status)
goto out;
/* if link status is down no point in checking to see if pf is up */
if (!(E1000_READ_REG(hw, E1000_STATUS) & E1000_STATUS_LU))
goto out;
/* if we passed all the tests above then the link is up and we no
* longer need to check for link */
mac->get_link_status = FALSE;
out:
return ret_val;
}
static int
igbvf_dev_configure(struct rte_eth_dev *dev)
{
struct rte_eth_conf* conf = &dev->data->dev_conf;
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_LOG(DEBUG, "Configured Virtual Function port id: %d",
e1000: indent logs sections Prepare for next commit, indent sections where log messages will be modified so that next patch is only about \n. Signed-off-by: David Marchand <david.marchand@6wind.com>
2014-09-17 13:46:48 +00:00
dev->data->port_id);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*
* VF has no ability to enable/disable HW CRC
* Keep the persistent behavior the same as Host PF
*/
#ifndef RTE_LIBRTE_E1000_PF_DISABLE_STRIP_CRC
if (!conf->rxmode.hw_strip_crc) {
e1000: raise log level of significant events Any message about incorrect API usage should be at NOTICE level or above. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
2015-07-09 23:01:08 +00:00
PMD_INIT_LOG(NOTICE, "VF can't disable HW CRC Strip");
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
conf->rxmode.hw_strip_crc = 1;
}
#else
if (conf->rxmode.hw_strip_crc) {
e1000: raise log level of significant events Any message about incorrect API usage should be at NOTICE level or above. Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
2015-07-09 23:01:08 +00:00
PMD_INIT_LOG(NOTICE, "VF can't enable HW CRC Strip");
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
conf->rxmode.hw_strip_crc = 0;
}
#endif
return 0;
}
static int
igbvf_dev_start(struct rte_eth_dev *dev)
{
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
struct e1000_hw *hw =
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(dev->data->dev_private);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
int ret;
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
hw->mac.ops.reset_hw(hw);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
adapter->stopped = 0;
pci: rework interrupt handling Signed-off-by: Intel <intel.com>
2013-06-03 00:00:00 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/* Set all vfta */
igbvf_set_vfta_all(dev,1);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
eth_igbvf_tx_init(dev);
/* This can fail when allocating mbufs for descriptor rings */
ret = eth_igbvf_rx_init(dev);
if (ret) {
PMD_INIT_LOG(ERR, "Unable to initialize RX hardware");
igb_dev_clear_queues(dev);
return ret;
}
return 0;
}
static void
igbvf_dev_stop(struct rte_eth_dev *dev)
{
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
igbvf_stop_adapter(dev);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
/*
* Clear what we set, but we still keep shadow_vfta to
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
* restore after device starts
*/
igbvf_set_vfta_all(dev,0);
igb_dev_clear_queues(dev);
}
static void
igbvf_dev_close(struct rte_eth_dev *dev)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
struct e1000_adapter *adapter =
E1000_DEV_PRIVATE(dev->data->dev_private);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
e1000_reset_hw(hw);
igbvf_dev_stop(dev);
e1000: support port hotplug This patch depends on the Port Hotplug Framework. It implements the eth_dev_uninit functions for rte_em_pmd, rte_igb_pmd and rte_igbvf_pmd. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:26 +00:00
adapter->stopped = 1;
e1000: free queue memory when closing Queues were freed in clear function called in stop function. Split clearing and freeing in separate functions to move queue freeing from stop to close function. Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-07-03 14:38:27 +00:00
igb_dev_free_queues(dev);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
}
static int igbvf_set_vfta(struct e1000_hw *hw, uint16_t vid, bool on)
{
struct e1000_mbx_info *mbx = &hw->mbx;
uint32_t msgbuf[2];
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
/* After set vlan, vlan strip will also be enabled in igb driver*/
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
msgbuf[0] = E1000_VF_SET_VLAN;
msgbuf[1] = vid;
/* Setting the 8 bit field MSG INFO to TRUE indicates "add" */
if (on)
msgbuf[0] |= E1000_VF_SET_VLAN_ADD;
return (mbx->ops.write_posted(hw, msgbuf, 2, 0));
}
static void igbvf_set_vfta_all(struct rte_eth_dev *dev, bool on)
{
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
struct e1000_hw *hw =
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_vfta * shadow_vfta =
E1000_DEV_PRIVATE_TO_VFTA(dev->data->dev_private);
int i = 0, j = 0, vfta = 0, mask = 1;
for (i = 0; i < IGB_VFTA_SIZE; i++){
vfta = shadow_vfta->vfta[i];
if(vfta){
mask = 1;
for (j = 0; j < 32; j++){
if(vfta & mask)
lib: fix various compilation warnings Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
igbvf_set_vfta(hw,
(uint16_t)((i<<5)+j), on);
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
mask<<=1;
}
}
}
}
static int
igbvf_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on)
{
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
struct e1000_hw *hw =
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_vfta * shadow_vfta =
E1000_DEV_PRIVATE_TO_VFTA(dev->data->dev_private);
uint32_t vid_idx = 0;
uint32_t vid_bit = 0;
int ret = 0;
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
e1000: clean log messages - remove leading \n in some messages, - remove trailing \n in some messages, - split multi lines messages, - introduce PMD_INIT_FUNC_TRACE macro and use it instead of PMD_INIT_LOG(DEBUG, "some_func") Signed-off-by: David Marchand <david.marchand@6wind.com> Reviewed-by: Jay Rolette <rolette@infiniteio.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
2014-09-17 13:46:49 +00:00
PMD_INIT_FUNC_TRACE();
igb: add VF support Signed-off-by: Intel
2012-12-19 23:00:00 +00:00
/*vind is not used in VF driver, set to 0, check ixgbe_set_vfta_vf*/
ret = igbvf_set_vfta(hw, vlan_id, !!on);
if(ret){
PMD_INIT_LOG(ERR, "Unable to set VF vlan");
return ret;
}
vid_idx = (uint32_t) ((vlan_id >> 5) & 0x7F);
vid_bit = (uint32_t) (1 << (vlan_id & 0x1F));
/*Save what we set and retore it after device reset*/
if (on)
shadow_vfta->vfta[vid_idx] |= vid_bit;
else
shadow_vfta->vfta[vid_idx] &= ~vid_bit;
return 0;
}
igb: add default mac address modifier - set_mac_addr Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static void
igbvf_default_mac_addr_set(struct rte_eth_dev *dev, struct ether_addr *addr)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
/* index is not used by rar_set() */
hw->mac.ops.rar_set(hw, (void *)addr, 0);
}
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
static int
eth_igb_rss_reta_update(struct rte_eth_dev *dev,
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
struct rte_eth_rss_reta_entry64 *reta_conf,
uint16_t reta_size)
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
{
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
uint8_t i, j, mask;
uint32_t reta, r;
uint16_t idx, shift;
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
if (reta_size != ETH_RSS_RETA_SIZE_128) {
PMD_DRV_LOG(ERR, "The size of hash lookup table configured "
"(%d) doesn't match the number hardware can supported "
"(%d)\n", reta_size, ETH_RSS_RETA_SIZE_128);
return -EINVAL;
}
for (i = 0; i < reta_size; i += IGB_4_BIT_WIDTH) {
idx = i / RTE_RETA_GROUP_SIZE;
shift = i % RTE_RETA_GROUP_SIZE;
mask = (uint8_t)((reta_conf[idx].mask >> shift) &
IGB_4_BIT_MASK);
if (!mask)
continue;
if (mask == IGB_4_BIT_MASK)
r = 0;
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
else
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
r = E1000_READ_REG(hw, E1000_RETA(i >> 2));
for (j = 0, reta = 0; j < IGB_4_BIT_WIDTH; j++) {
if (mask & (0x1 << j))
reta |= reta_conf[idx].reta[shift + j] <<
(CHAR_BIT * j);
else
reta |= r & (IGB_8_BIT_MASK << (CHAR_BIT * j));
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
}
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
E1000_WRITE_REG(hw, E1000_RETA(i >> 2), reta);
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
}
return 0;
}
static int
eth_igb_rss_reta_query(struct rte_eth_dev *dev,
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
struct rte_eth_rss_reta_entry64 *reta_conf,
uint16_t reta_size)
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
{
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
uint8_t i, j, mask;
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
uint32_t reta;
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
uint16_t idx, shift;
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
ethdev: support multiple sizes of redirection table As 40G NIC supports different sizes (128/512/64 entries) of redirection table from that (128 entries) of 1G and 10G NICs, support of multiple sizes of redirection table is needed. It includes, * Redefine 'struct rte_eth_rss_reta' in ethdev. - To 'struct rte_eth_rss_reta_entry64' which contains 64 entries and 64 bits mask. - Array of above new structure can be used for any number of redirection table entries, as long as the number is multiple of 64. This is quite flexible for the future expanding of redirection table. * Redefinition of relevant interfaces in ethdev. - Interface of reta update has been redefined with new parameters. - Interface of reta query has been redefined with new parameters. * Rework of 1G PMD in igb. - reta update has been reworked. - reta query has been reworked. * Rework of 10G PMD in ixgbe. - reta update has been reworked. - reta query has been reworked. * Rework of 40G PMD (PF only) in i40e. - reta update has been reworked. - reta query has been reworked. * Implement relevant commands in testpmd. Test report: http://dpdk.org/ml/archives/dev/2014-November/008362.html Signed-off-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com> Tested-by: Erlu Chen <erlu.chen@intel.com>
2014-11-15 16:03:43 +00:00
if (reta_size != ETH_RSS_RETA_SIZE_128) {
PMD_DRV_LOG(ERR, "The size of hash lookup table configured "
"(%d) doesn't match the number hardware can supported "
"(%d)\n", reta_size, ETH_RSS_RETA_SIZE_128);
return -EINVAL;
}
for (i = 0; i < reta_size; i += IGB_4_BIT_WIDTH) {
idx = i / RTE_RETA_GROUP_SIZE;
shift = i % RTE_RETA_GROUP_SIZE;
mask = (uint8_t)((reta_conf[idx].mask >> shift) &
IGB_4_BIT_MASK);
if (!mask)
continue;
reta = E1000_READ_REG(hw, E1000_RETA(i >> 2));
for (j = 0; j < IGB_4_BIT_WIDTH; j++) {
if (mask & (0x1 << j))
reta_conf[idx].reta[shift + j] =
((reta >> (CHAR_BIT * j)) &
IGB_8_BIT_MASK);
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
}
}
remove trailing whitespaces This commit removes trailing whitespace from lines in files. Almost all files are affected, as the BSD license copyright header had trailing whitespace on 4 lines in it [hence the number of files reporting 8 lines changed in the diffstat]. Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> [Thomas: remove spaces before tabs in libs] [Thomas: remove more trailing spaces in non-C files] Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-03 23:42:50 +00:00
igb: RSS RETA configuration Signed-off-by: Intel
2013-06-03 00:00:00 +00:00
return 0;
}
igb: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igb pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igb library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:32 +00:00
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
#define MAC_TYPE_FILTER_SUP(type) do {\
if ((type) != e1000_82580 && (type) != e1000_i350 &&\
(type) != e1000_82576)\
igb: use generic filter control for ethertype filter This patch removes old functions which deal with ethertype filter in igb driver. It also defines eth_igb_filter_ctrl which is binding to filter_ctrl API, and ethertype filter can be dealt with through this new entrance. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-01-12 07:16:11 +00:00
return -ENOTSUP;\
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
} while (0)
static int
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
eth_igb_syn_filter_set(struct rte_eth_dev *dev,
struct rte_eth_syn_filter *filter,
bool add)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t synqf, rfctl;
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
if (filter->queue >= IGB_MAX_RX_QUEUE_NUM)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return -EINVAL;
synqf = E1000_READ_REG(hw, E1000_SYNQF(0));
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
if (add) {
if (synqf & E1000_SYN_FILTER_ENABLE)
return -EINVAL;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
synqf = (uint32_t)(((filter->queue << E1000_SYN_FILTER_QUEUE_SHIFT) &
E1000_SYN_FILTER_QUEUE) | E1000_SYN_FILTER_ENABLE);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
rfctl = E1000_READ_REG(hw, E1000_RFCTL);
if (filter->hig_pri)
rfctl |= E1000_RFCTL_SYNQFP;
else
rfctl &= ~E1000_RFCTL_SYNQFP;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
E1000_WRITE_REG(hw, E1000_RFCTL, rfctl);
} else {
if (!(synqf & E1000_SYN_FILTER_ENABLE))
return -ENOENT;
synqf = 0;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
E1000_WRITE_REG(hw, E1000_SYNQF(0), synqf);
E1000_WRITE_FLUSH(hw);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return 0;
}
static int
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
eth_igb_syn_filter_get(struct rte_eth_dev *dev,
struct rte_eth_syn_filter *filter)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t synqf, rfctl;
synqf = E1000_READ_REG(hw, E1000_SYNQF(0));
if (synqf & E1000_SYN_FILTER_ENABLE) {
rfctl = E1000_READ_REG(hw, E1000_RFCTL);
filter->hig_pri = (rfctl & E1000_RFCTL_SYNQFP) ? 1 : 0;
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
filter->queue = (uint8_t)((synqf & E1000_SYN_FILTER_QUEUE) >>
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
E1000_SYN_FILTER_QUEUE_SHIFT);
return 0;
}
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return -ENOENT;
}
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
static int
eth_igb_syn_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
int ret;
MAC_TYPE_FILTER_SUP(hw->mac.type);
if (filter_op == RTE_ETH_FILTER_NOP)
return 0;
if (arg == NULL) {
PMD_DRV_LOG(ERR, "arg shouldn't be NULL for operation %u",
filter_op);
return -EINVAL;
}
switch (filter_op) {
case RTE_ETH_FILTER_ADD:
ret = eth_igb_syn_filter_set(dev,
(struct rte_eth_syn_filter *)arg,
TRUE);
break;
case RTE_ETH_FILTER_DELETE:
ret = eth_igb_syn_filter_set(dev,
(struct rte_eth_syn_filter *)arg,
FALSE);
break;
case RTE_ETH_FILTER_GET:
ret = eth_igb_syn_filter_get(dev,
(struct rte_eth_syn_filter *)arg);
break;
default:
PMD_DRV_LOG(ERR, "unsupported operation %u\n", filter_op);
ret = -EINVAL;
break;
}
return ret;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
#define MAC_TYPE_FILTER_SUP_EXT(type) do {\
if ((type) != e1000_82580 && (type) != e1000_i350)\
return -ENOSYS; \
} while (0)
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
/* translate elements in struct rte_eth_ntuple_filter to struct e1000_2tuple_filter_info*/
static inline int
ntuple_filter_to_2tuple(struct rte_eth_ntuple_filter *filter,
struct e1000_2tuple_filter_info *filter_info)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
if (filter->queue >= IGB_MAX_RX_QUEUE_NUM)
return -EINVAL;
if (filter->priority > E1000_2TUPLE_MAX_PRI)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return -EINVAL; /* filter index is out of range. */
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
if (filter->tcp_flags > TCP_FLAG_ALL)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return -EINVAL; /* flags is invalid. */
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->dst_port_mask) {
case UINT16_MAX:
filter_info->dst_port_mask = 0;
filter_info->dst_port = filter->dst_port;
break;
case 0:
filter_info->dst_port_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid dst_port mask.");
return -EINVAL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->proto_mask) {
case UINT8_MAX:
filter_info->proto_mask = 0;
filter_info->proto = filter->proto;
break;
case 0:
filter_info->proto_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid protocol mask.");
return -EINVAL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter_info->priority = (uint8_t)filter->priority;
if (filter->flags & RTE_NTUPLE_FLAGS_TCP_FLAG)
filter_info->tcp_flags = filter->tcp_flags;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
else
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter_info->tcp_flags = 0;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return 0;
}
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
static inline struct e1000_2tuple_filter *
igb_2tuple_filter_lookup(struct e1000_2tuple_filter_list *filter_list,
struct e1000_2tuple_filter_info *key)
{
struct e1000_2tuple_filter *it;
TAILQ_FOREACH(it, filter_list, entries) {
if (memcmp(key, &it->filter_info,
sizeof(struct e1000_2tuple_filter_info)) == 0) {
return it;
}
}
return NULL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
/*
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* igb_add_2tuple_filter - add a 2tuple filter
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @param
* dev: Pointer to struct rte_eth_dev.
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* ntuple_filter: ponter to the filter that will be added.
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
igb_add_2tuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_2tuple_filter *filter;
uint32_t ttqf = E1000_TTQF_DISABLE_MASK;
uint32_t imir, imir_ext = E1000_IMIREXT_SIZE_BP;
int i, ret;
filter = rte_zmalloc("e1000_2tuple_filter",
sizeof(struct e1000_2tuple_filter), 0);
if (filter == NULL)
return -ENOMEM;
ret = ntuple_filter_to_2tuple(ntuple_filter,
&filter->filter_info);
if (ret < 0) {
rte_free(filter);
return ret;
}
if (igb_2tuple_filter_lookup(&filter_info->twotuple_list,
&filter->filter_info) != NULL) {
PMD_DRV_LOG(ERR, "filter exists.");
rte_free(filter);
return -EEXIST;
}
filter->queue = ntuple_filter->queue;
/*
* look for an unused 2tuple filter index,
* and insert the filter to list.
*/
for (i = 0; i < E1000_MAX_TTQF_FILTERS; i++) {
if (!(filter_info->twotuple_mask & (1 << i))) {
filter_info->twotuple_mask |= 1 << i;
filter->index = i;
TAILQ_INSERT_TAIL(&filter_info->twotuple_list,
filter,
entries);
break;
}
}
if (i >= E1000_MAX_TTQF_FILTERS) {
PMD_DRV_LOG(ERR, "2tuple filters are full.");
rte_free(filter);
return -ENOSYS;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
imir = (uint32_t)(filter->filter_info.dst_port & E1000_IMIR_DSTPORT);
if (filter->filter_info.dst_port_mask == 1) /* 1b means not compare. */
imir |= E1000_IMIR_PORT_BP;
else
imir &= ~E1000_IMIR_PORT_BP;
imir |= filter->filter_info.priority << E1000_IMIR_PRIORITY_SHIFT;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
ttqf |= E1000_TTQF_QUEUE_ENABLE;
ttqf |= (uint32_t)(filter->queue << E1000_TTQF_QUEUE_SHIFT);
ttqf |= (uint32_t)(filter->filter_info.proto & E1000_TTQF_PROTOCOL_MASK);
if (filter->filter_info.proto_mask == 0)
ttqf &= ~E1000_TTQF_MASK_ENABLE;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
/* tcp flags bits setting. */
if (filter->filter_info.tcp_flags & TCP_FLAG_ALL) {
if (filter->filter_info.tcp_flags & TCP_URG_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_URG;
if (filter->filter_info.tcp_flags & TCP_ACK_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_ACK;
if (filter->filter_info.tcp_flags & TCP_PSH_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_PSH;
if (filter->filter_info.tcp_flags & TCP_RST_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_RST;
if (filter->filter_info.tcp_flags & TCP_SYN_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_SYN;
if (filter->filter_info.tcp_flags & TCP_FIN_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_FIN;
} else
imir_ext |= E1000_IMIREXT_CTRL_BP;
E1000_WRITE_REG(hw, E1000_IMIR(i), imir);
E1000_WRITE_REG(hw, E1000_TTQF(i), ttqf);
E1000_WRITE_REG(hw, E1000_IMIREXT(i), imir_ext);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return 0;
}
/*
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* igb_remove_2tuple_filter - remove a 2tuple filter
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @param
* dev: Pointer to struct rte_eth_dev.
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* ntuple_filter: ponter to the filter that will be removed.
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
igb_remove_2tuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_2tuple_filter_info filter_2tuple;
struct e1000_2tuple_filter *filter;
int ret;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
memset(&filter_2tuple, 0, sizeof(struct e1000_2tuple_filter_info));
ret = ntuple_filter_to_2tuple(ntuple_filter,
&filter_2tuple);
if (ret < 0)
return ret;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter = igb_2tuple_filter_lookup(&filter_info->twotuple_list,
&filter_2tuple);
if (filter == NULL) {
PMD_DRV_LOG(ERR, "filter doesn't exist.");
return -ENOENT;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
}
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter_info->twotuple_mask &= ~(1 << filter->index);
TAILQ_REMOVE(&filter_info->twotuple_list, filter, entries);
rte_free(filter);
E1000_WRITE_REG(hw, E1000_TTQF(filter->index), E1000_TTQF_DISABLE_MASK);
E1000_WRITE_REG(hw, E1000_IMIR(filter->index), 0);
E1000_WRITE_REG(hw, E1000_IMIREXT(filter->index), 0);
return 0;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
}
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
static inline struct e1000_flex_filter *
eth_igb_flex_filter_lookup(struct e1000_flex_filter_list *filter_list,
struct e1000_flex_filter_info *key)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct e1000_flex_filter *it;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
TAILQ_FOREACH(it, filter_list, entries) {
if (memcmp(key, &it->filter_info,
sizeof(struct e1000_flex_filter_info)) == 0)
return it;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
return NULL;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
}
static int
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
eth_igb_add_del_flex_filter(struct rte_eth_dev *dev,
struct rte_eth_flex_filter *filter,
bool add)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_flex_filter *flex_filter, *it;
uint32_t wufc, queueing, mask;
uint32_t reg_off;
uint8_t shift, i, j = 0;
flex_filter = rte_zmalloc("e1000_flex_filter",
sizeof(struct e1000_flex_filter), 0);
if (flex_filter == NULL)
return -ENOMEM;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
flex_filter->filter_info.len = filter->len;
flex_filter->filter_info.priority = filter->priority;
memcpy(flex_filter->filter_info.dwords, filter->bytes, filter->len);
for (i = 0; i < RTE_ALIGN(filter->len, CHAR_BIT) / CHAR_BIT; i++) {
mask = 0;
/* reverse bits in flex filter's mask*/
for (shift = 0; shift < CHAR_BIT; shift++) {
if (filter->mask[i] & (0x01 << shift))
mask |= (0x80 >> shift);
}
flex_filter->filter_info.mask[i] = mask;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
wufc = E1000_READ_REG(hw, E1000_WUFC);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
if (flex_filter->index < E1000_MAX_FHFT)
reg_off = E1000_FHFT(flex_filter->index);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
else
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
reg_off = E1000_FHFT_EXT(flex_filter->index - E1000_MAX_FHFT);
if (add) {
if (eth_igb_flex_filter_lookup(&filter_info->flex_list,
&flex_filter->filter_info) != NULL) {
PMD_DRV_LOG(ERR, "filter exists.");
rte_free(flex_filter);
return -EEXIST;
}
flex_filter->queue = filter->queue;
/*
* look for an unused flex filter index
* and insert the filter into the list.
*/
for (i = 0; i < E1000_MAX_FLEX_FILTERS; i++) {
if (!(filter_info->flex_mask & (1 << i))) {
filter_info->flex_mask |= 1 << i;
flex_filter->index = i;
TAILQ_INSERT_TAIL(&filter_info->flex_list,
flex_filter,
entries);
break;
}
}
if (i >= E1000_MAX_FLEX_FILTERS) {
PMD_DRV_LOG(ERR, "flex filters are full.");
rte_free(flex_filter);
return -ENOSYS;
}
E1000_WRITE_REG(hw, E1000_WUFC, wufc | E1000_WUFC_FLEX_HQ |
(E1000_WUFC_FLX0 << flex_filter->index));
queueing = filter->len |
(filter->queue << E1000_FHFT_QUEUEING_QUEUE_SHIFT) |
(filter->priority << E1000_FHFT_QUEUEING_PRIO_SHIFT);
E1000_WRITE_REG(hw, reg_off + E1000_FHFT_QUEUEING_OFFSET,
queueing);
for (i = 0; i < E1000_FLEX_FILTERS_MASK_SIZE; i++) {
E1000_WRITE_REG(hw, reg_off,
flex_filter->filter_info.dwords[j]);
reg_off += sizeof(uint32_t);
E1000_WRITE_REG(hw, reg_off,
flex_filter->filter_info.dwords[++j]);
reg_off += sizeof(uint32_t);
E1000_WRITE_REG(hw, reg_off,
(uint32_t)flex_filter->filter_info.mask[i]);
reg_off += sizeof(uint32_t) * 2;
++j;
}
} else {
it = eth_igb_flex_filter_lookup(&filter_info->flex_list,
&flex_filter->filter_info);
if (it == NULL) {
PMD_DRV_LOG(ERR, "filter doesn't exist.");
rte_free(flex_filter);
return -ENOENT;
}
for (i = 0; i < E1000_FHFT_SIZE_IN_DWD; i++)
E1000_WRITE_REG(hw, reg_off + i * sizeof(uint32_t), 0);
E1000_WRITE_REG(hw, E1000_WUFC, wufc &
(~(E1000_WUFC_FLX0 << it->index)));
filter_info->flex_mask &= ~(1 << it->index);
TAILQ_REMOVE(&filter_info->flex_list, it, entries);
rte_free(it);
rte_free(flex_filter);
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return 0;
}
static int
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
eth_igb_get_flex_filter(struct rte_eth_dev *dev,
struct rte_eth_flex_filter *filter)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_flex_filter flex_filter, *it;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
uint32_t wufc, queueing, wufc_en = 0;
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
memset(&flex_filter, 0, sizeof(struct e1000_flex_filter));
flex_filter.filter_info.len = filter->len;
flex_filter.filter_info.priority = filter->priority;
memcpy(flex_filter.filter_info.dwords, filter->bytes, filter->len);
memcpy(flex_filter.filter_info.mask, filter->mask,
RTE_ALIGN(filter->len, sizeof(char)) / sizeof(char));
it = eth_igb_flex_filter_lookup(&filter_info->flex_list,
&flex_filter.filter_info);
if (it == NULL) {
PMD_DRV_LOG(ERR, "filter doesn't exist.");
return -ENOENT;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
wufc = E1000_READ_REG(hw, E1000_WUFC);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
wufc_en = E1000_WUFC_FLEX_HQ | (E1000_WUFC_FLX0 << it->index);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
if ((wufc & wufc_en) == wufc_en) {
uint32_t reg_off = 0;
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
if (it->index < E1000_MAX_FHFT)
reg_off = E1000_FHFT(it->index);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
else
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
reg_off = E1000_FHFT_EXT(it->index - E1000_MAX_FHFT);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
queueing = E1000_READ_REG(hw,
reg_off + E1000_FHFT_QUEUEING_OFFSET);
filter->len = queueing & E1000_FHFT_QUEUEING_LEN;
filter->priority = (queueing & E1000_FHFT_QUEUEING_PRIO) >>
E1000_FHFT_QUEUEING_PRIO_SHIFT;
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
filter->queue = (queueing & E1000_FHFT_QUEUEING_QUEUE) >>
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
E1000_FHFT_QUEUEING_QUEUE_SHIFT;
return 0;
}
return -ENOENT;
}
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
static int
eth_igb_flex_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct rte_eth_flex_filter *filter;
int ret = 0;
igb: fix flex filter check for unsupported 82576 Flex filter is currently not supported for NIC 82576. Only i350 and 82580 support it. So The MAC_TYPE_FILTER_SUP_EXT is used to check whether the MAC type is i350 or 82580. Fixes: 231d43909a31 ("igb: migrate flex filter to new API") Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Marvin Liu <yong.liu@intel.com>
2015-03-26 04:53:21 +00:00
MAC_TYPE_FILTER_SUP_EXT(hw->mac.type);
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
if (filter_op == RTE_ETH_FILTER_NOP)
return ret;
if (arg == NULL) {
PMD_DRV_LOG(ERR, "arg shouldn't be NULL for operation %u",
filter_op);
return -EINVAL;
}
filter = (struct rte_eth_flex_filter *)arg;
if (filter->len == 0 || filter->len > E1000_MAX_FLEX_FILTER_LEN
|| filter->len % sizeof(uint64_t) != 0) {
PMD_DRV_LOG(ERR, "filter's length is out of range");
return -EINVAL;
}
if (filter->priority > E1000_MAX_FLEX_FILTER_PRI) {
PMD_DRV_LOG(ERR, "filter's priority is out of range");
return -EINVAL;
}
switch (filter_op) {
case RTE_ETH_FILTER_ADD:
ret = eth_igb_add_del_flex_filter(dev, filter, TRUE);
break;
case RTE_ETH_FILTER_DELETE:
ret = eth_igb_add_del_flex_filter(dev, filter, FALSE);
break;
case RTE_ETH_FILTER_GET:
ret = eth_igb_get_flex_filter(dev, filter);
break;
default:
PMD_DRV_LOG(ERR, "unsupported operation %u", filter_op);
ret = -EINVAL;
break;
}
return ret;
}
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
/* translate elements in struct rte_eth_ntuple_filter to struct e1000_5tuple_filter_info*/
static inline int
ntuple_filter_to_5tuple_82576(struct rte_eth_ntuple_filter *filter,
struct e1000_5tuple_filter_info *filter_info)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
if (filter->queue >= IGB_MAX_RX_QUEUE_NUM_82576)
return -EINVAL;
if (filter->priority > E1000_2TUPLE_MAX_PRI)
return -EINVAL; /* filter index is out of range. */
if (filter->tcp_flags > TCP_FLAG_ALL)
return -EINVAL; /* flags is invalid. */
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->dst_ip_mask) {
case UINT32_MAX:
filter_info->dst_ip_mask = 0;
filter_info->dst_ip = filter->dst_ip;
break;
case 0:
filter_info->dst_ip_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid dst_ip mask.");
return -EINVAL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->src_ip_mask) {
case UINT32_MAX:
filter_info->src_ip_mask = 0;
filter_info->src_ip = filter->src_ip;
break;
case 0:
filter_info->src_ip_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid src_ip mask.");
return -EINVAL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->dst_port_mask) {
case UINT16_MAX:
filter_info->dst_port_mask = 0;
filter_info->dst_port = filter->dst_port;
break;
case 0:
filter_info->dst_port_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid dst_port mask.");
return -EINVAL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->src_port_mask) {
case UINT16_MAX:
filter_info->src_port_mask = 0;
filter_info->src_port = filter->src_port;
break;
case 0:
filter_info->src_port_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid src_port mask.");
return -EINVAL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
switch (filter->proto_mask) {
case UINT8_MAX:
filter_info->proto_mask = 0;
filter_info->proto = filter->proto;
break;
case 0:
filter_info->proto_mask = 1;
break;
default:
PMD_DRV_LOG(ERR, "invalid protocol mask.");
return -EINVAL;
}
filter_info->priority = (uint8_t)filter->priority;
if (filter->flags & RTE_NTUPLE_FLAGS_TCP_FLAG)
filter_info->tcp_flags = filter->tcp_flags;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
else
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter_info->tcp_flags = 0;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return 0;
}
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
static inline struct e1000_5tuple_filter *
igb_5tuple_filter_lookup_82576(struct e1000_5tuple_filter_list *filter_list,
struct e1000_5tuple_filter_info *key)
{
struct e1000_5tuple_filter *it;
TAILQ_FOREACH(it, filter_list, entries) {
if (memcmp(key, &it->filter_info,
sizeof(struct e1000_5tuple_filter_info)) == 0) {
return it;
}
}
return NULL;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
/*
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* igb_add_5tuple_filter_82576 - add a 5tuple filter
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @param
* dev: Pointer to struct rte_eth_dev.
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* ntuple_filter: ponter to the filter that will be added.
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
igb_add_5tuple_filter_82576(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_5tuple_filter *filter;
uint32_t ftqf = E1000_FTQF_VF_BP | E1000_FTQF_MASK;
uint32_t spqf, imir, imir_ext = E1000_IMIREXT_SIZE_BP;
uint8_t i;
int ret;
filter = rte_zmalloc("e1000_5tuple_filter",
sizeof(struct e1000_5tuple_filter), 0);
if (filter == NULL)
return -ENOMEM;
ret = ntuple_filter_to_5tuple_82576(ntuple_filter,
&filter->filter_info);
if (ret < 0) {
rte_free(filter);
return ret;
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
if (igb_5tuple_filter_lookup_82576(&filter_info->fivetuple_list,
&filter->filter_info) != NULL) {
PMD_DRV_LOG(ERR, "filter exists.");
rte_free(filter);
return -EEXIST;
}
filter->queue = ntuple_filter->queue;
/*
* look for an unused 5tuple filter index,
* and insert the filter to list.
*/
for (i = 0; i < E1000_MAX_FTQF_FILTERS; i++) {
if (!(filter_info->fivetuple_mask & (1 << i))) {
filter_info->fivetuple_mask |= 1 << i;
filter->index = i;
TAILQ_INSERT_TAIL(&filter_info->fivetuple_list,
filter,
entries);
break;
}
}
if (i >= E1000_MAX_FTQF_FILTERS) {
PMD_DRV_LOG(ERR, "5tuple filters are full.");
rte_free(filter);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return -ENOSYS;
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
}
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
ftqf |= filter->filter_info.proto & E1000_FTQF_PROTOCOL_MASK;
if (filter->filter_info.src_ip_mask == 0) /* 0b means compare. */
ftqf &= ~E1000_FTQF_MASK_SOURCE_ADDR_BP;
if (filter->filter_info.dst_ip_mask == 0)
ftqf &= ~E1000_FTQF_MASK_DEST_ADDR_BP;
if (filter->filter_info.src_port_mask == 0)
ftqf &= ~E1000_FTQF_MASK_SOURCE_PORT_BP;
if (filter->filter_info.proto_mask == 0)
ftqf &= ~E1000_FTQF_MASK_PROTO_BP;
ftqf |= (filter->queue << E1000_FTQF_QUEUE_SHIFT) &
E1000_FTQF_QUEUE_MASK;
ftqf |= E1000_FTQF_QUEUE_ENABLE;
E1000_WRITE_REG(hw, E1000_FTQF(i), ftqf);
E1000_WRITE_REG(hw, E1000_DAQF(i), filter->filter_info.dst_ip);
E1000_WRITE_REG(hw, E1000_SAQF(i), filter->filter_info.src_ip);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
spqf = filter->filter_info.src_port & E1000_SPQF_SRCPORT;
E1000_WRITE_REG(hw, E1000_SPQF(i), spqf);
imir = (uint32_t)(filter->filter_info.dst_port & E1000_IMIR_DSTPORT);
if (filter->filter_info.dst_port_mask == 1) /* 1b means not compare. */
imir |= E1000_IMIR_PORT_BP;
else
imir &= ~E1000_IMIR_PORT_BP;
imir |= filter->filter_info.priority << E1000_IMIR_PRIORITY_SHIFT;
/* tcp flags bits setting. */
if (filter->filter_info.tcp_flags & TCP_FLAG_ALL) {
if (filter->filter_info.tcp_flags & TCP_URG_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_URG;
if (filter->filter_info.tcp_flags & TCP_ACK_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_ACK;
if (filter->filter_info.tcp_flags & TCP_PSH_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_PSH;
if (filter->filter_info.tcp_flags & TCP_RST_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_RST;
if (filter->filter_info.tcp_flags & TCP_SYN_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_SYN;
if (filter->filter_info.tcp_flags & TCP_FIN_FLAG)
imir_ext |= E1000_IMIREXT_CTRL_FIN;
} else
imir_ext |= E1000_IMIREXT_CTRL_BP;
E1000_WRITE_REG(hw, E1000_IMIR(i), imir);
E1000_WRITE_REG(hw, E1000_IMIREXT(i), imir_ext);
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
return 0;
}
/*
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* igb_remove_5tuple_filter_82576 - remove a 5tuple filter
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @param
* dev: Pointer to struct rte_eth_dev.
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
* ntuple_filter: ponter to the filter that will be removed.
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
igb_remove_5tuple_filter_82576(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter)
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_5tuple_filter_info filter_5tuple;
struct e1000_5tuple_filter *filter;
int ret;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
memset(&filter_5tuple, 0, sizeof(struct e1000_5tuple_filter_info));
ret = ntuple_filter_to_5tuple_82576(ntuple_filter,
&filter_5tuple);
if (ret < 0)
return ret;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter = igb_5tuple_filter_lookup_82576(&filter_info->fivetuple_list,
&filter_5tuple);
if (filter == NULL) {
PMD_DRV_LOG(ERR, "filter doesn't exist.");
return -ENOENT;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
}
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
filter_info->fivetuple_mask &= ~(1 << filter->index);
TAILQ_REMOVE(&filter_info->fivetuple_list, filter, entries);
rte_free(filter);
E1000_WRITE_REG(hw, E1000_FTQF(filter->index),
E1000_FTQF_VF_BP | E1000_FTQF_MASK);
E1000_WRITE_REG(hw, E1000_DAQF(filter->index), 0);
E1000_WRITE_REG(hw, E1000_SAQF(filter->index), 0);
E1000_WRITE_REG(hw, E1000_SPQF(filter->index), 0);
E1000_WRITE_REG(hw, E1000_IMIR(filter->index), 0);
E1000_WRITE_REG(hw, E1000_IMIREXT(filter->index), 0);
return 0;
igb: add filters This patch adds following igb NIC filters implement: syn filter, ethertype filter, 2tuple filter, flex filter for intel NIC 82580 and i350 syn filter, ethertype filter, 5tuple filter for intel NIC 82576 Signed-off-by: jingjing.wu <jingjing.wu@intel.com> Tested-by: Waterman Cao <waterman.cao@intel.com> Reviewed-by: Vladimir Medvedkin <medvedkinv@gmail.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-06-16 07:31:44 +00:00
}
ethdev: MTU accessors This patch adds two new functions in ethdev api to retrieve current MTU and change MTU of a port. Only .mtu_set function is pmd specific. pmd should update its max_rx_pkt_len if needed. This operation has been implemented for rte_em_pmd, rte_igb_pmd and rte_ixgbe_pmd. Signed-off-by: Samuel Gauthier <samuel.gauthier@6wind.com> Signed-off-by: Ivan Boule <ivan.boule@6wind.com> Signed-off-by: David Marchand <david.marchand@6wind.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2014-06-17 18:09:30 +00:00
static int
eth_igb_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
{
uint32_t rctl;
struct e1000_hw *hw;
struct rte_eth_dev_info dev_info;
uint32_t frame_size = mtu + (ETHER_HDR_LEN + ETHER_CRC_LEN +
VLAN_TAG_SIZE);
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
#ifdef RTE_LIBRTE_82571_SUPPORT
/* XXX: not bigger than max_rx_pktlen */
if (hw->mac.type == e1000_82571)
return -ENOTSUP;
#endif
eth_igb_infos_get(dev, &dev_info);
/* check that mtu is within the allowed range */
if ((mtu < ETHER_MIN_MTU) ||
(frame_size > dev_info.max_rx_pktlen))
return -EINVAL;
/* refuse mtu that requires the support of scattered packets when this
* feature has not been enabled before. */
if (!dev->data->scattered_rx &&
frame_size > dev->data->min_rx_buf_size - RTE_PKTMBUF_HEADROOM)
return -EINVAL;
rctl = E1000_READ_REG(hw, E1000_RCTL);
/* switch to jumbo mode if needed */
if (frame_size > ETHER_MAX_LEN) {
dev->data->dev_conf.rxmode.jumbo_frame = 1;
rctl |= E1000_RCTL_LPE;
} else {
dev->data->dev_conf.rxmode.jumbo_frame = 0;
rctl &= ~E1000_RCTL_LPE;
}
E1000_WRITE_REG(hw, E1000_RCTL, rctl);
/* update max frame size */
dev->data->dev_conf.rxmode.max_rx_pkt_len = frame_size;
E1000_WRITE_REG(hw, E1000_RLPML,
dev->data->dev_conf.rxmode.max_rx_pkt_len);
return 0;
}
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
/*
* igb_add_del_ntuple_filter - add or delete a ntuple filter
*
* @param
* dev: Pointer to struct rte_eth_dev.
* ntuple_filter: Pointer to struct rte_eth_ntuple_filter
* add: if true, add filter, if false, remove filter
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
igb_add_del_ntuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter,
bool add)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
int ret;
switch (ntuple_filter->flags) {
case RTE_5TUPLE_FLAGS:
case (RTE_5TUPLE_FLAGS | RTE_NTUPLE_FLAGS_TCP_FLAG):
if (hw->mac.type != e1000_82576)
return -ENOTSUP;
if (add)
ret = igb_add_5tuple_filter_82576(dev,
ntuple_filter);
else
ret = igb_remove_5tuple_filter_82576(dev,
ntuple_filter);
break;
case RTE_2TUPLE_FLAGS:
case (RTE_2TUPLE_FLAGS | RTE_NTUPLE_FLAGS_TCP_FLAG):
if (hw->mac.type != e1000_82580 && hw->mac.type != e1000_i350)
return -ENOTSUP;
if (add)
ret = igb_add_2tuple_filter(dev, ntuple_filter);
else
ret = igb_remove_2tuple_filter(dev, ntuple_filter);
break;
default:
ret = -EINVAL;
break;
}
return ret;
}
/*
* igb_get_ntuple_filter - get a ntuple filter
*
* @param
* dev: Pointer to struct rte_eth_dev.
* ntuple_filter: Pointer to struct rte_eth_ntuple_filter
*
* @return
* - On success, zero.
* - On failure, a negative value.
*/
static int
igb_get_ntuple_filter(struct rte_eth_dev *dev,
struct rte_eth_ntuple_filter *ntuple_filter)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
struct e1000_5tuple_filter_info filter_5tuple;
struct e1000_2tuple_filter_info filter_2tuple;
struct e1000_5tuple_filter *p_5tuple_filter;
struct e1000_2tuple_filter *p_2tuple_filter;
int ret;
switch (ntuple_filter->flags) {
case RTE_5TUPLE_FLAGS:
case (RTE_5TUPLE_FLAGS | RTE_NTUPLE_FLAGS_TCP_FLAG):
if (hw->mac.type != e1000_82576)
return -ENOTSUP;
memset(&filter_5tuple,
0,
sizeof(struct e1000_5tuple_filter_info));
ret = ntuple_filter_to_5tuple_82576(ntuple_filter,
&filter_5tuple);
if (ret < 0)
return ret;
p_5tuple_filter = igb_5tuple_filter_lookup_82576(
&filter_info->fivetuple_list,
&filter_5tuple);
if (p_5tuple_filter == NULL) {
PMD_DRV_LOG(ERR, "filter doesn't exist.");
return -ENOENT;
}
ntuple_filter->queue = p_5tuple_filter->queue;
break;
case RTE_2TUPLE_FLAGS:
case (RTE_2TUPLE_FLAGS | RTE_NTUPLE_FLAGS_TCP_FLAG):
if (hw->mac.type != e1000_82580 && hw->mac.type != e1000_i350)
return -ENOTSUP;
memset(&filter_2tuple,
0,
sizeof(struct e1000_2tuple_filter_info));
ret = ntuple_filter_to_2tuple(ntuple_filter, &filter_2tuple);
if (ret < 0)
return ret;
p_2tuple_filter = igb_2tuple_filter_lookup(
&filter_info->twotuple_list,
&filter_2tuple);
if (p_2tuple_filter == NULL) {
PMD_DRV_LOG(ERR, "filter doesn't exist.");
return -ENOENT;
}
ntuple_filter->queue = p_2tuple_filter->queue;
break;
default:
ret = -EINVAL;
break;
}
return 0;
}
/*
* igb_ntuple_filter_handle - Handle operations for ntuple filter.
* @dev: pointer to rte_eth_dev structure
* @filter_op:operation will be taken.
* @arg: a pointer to specific structure corresponding to the filter_op
*/
static int
igb_ntuple_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
int ret;
MAC_TYPE_FILTER_SUP(hw->mac.type);
if (filter_op == RTE_ETH_FILTER_NOP)
return 0;
if (arg == NULL) {
PMD_DRV_LOG(ERR, "arg shouldn't be NULL for operation %u.",
filter_op);
return -EINVAL;
}
switch (filter_op) {
case RTE_ETH_FILTER_ADD:
ret = igb_add_del_ntuple_filter(dev,
(struct rte_eth_ntuple_filter *)arg,
TRUE);
break;
case RTE_ETH_FILTER_DELETE:
ret = igb_add_del_ntuple_filter(dev,
(struct rte_eth_ntuple_filter *)arg,
FALSE);
break;
case RTE_ETH_FILTER_GET:
ret = igb_get_ntuple_filter(dev,
(struct rte_eth_ntuple_filter *)arg);
break;
default:
PMD_DRV_LOG(ERR, "unsupported operation %u.", filter_op);
ret = -EINVAL;
break;
}
return ret;
}
igb: use generic filter control for ethertype filter This patch removes old functions which deal with ethertype filter in igb driver. It also defines eth_igb_filter_ctrl which is binding to filter_ctrl API, and ethertype filter can be dealt with through this new entrance. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-01-12 07:16:11 +00:00
static inline int
igb_ethertype_filter_lookup(struct e1000_filter_info *filter_info,
uint16_t ethertype)
{
int i;
for (i = 0; i < E1000_MAX_ETQF_FILTERS; i++) {
if (filter_info->ethertype_filters[i] == ethertype &&
(filter_info->ethertype_mask & (1 << i)))
return i;
}
return -1;
}
static inline int
igb_ethertype_filter_insert(struct e1000_filter_info *filter_info,
uint16_t ethertype)
{
int i;
for (i = 0; i < E1000_MAX_ETQF_FILTERS; i++) {
if (!(filter_info->ethertype_mask & (1 << i))) {
filter_info->ethertype_mask |= 1 << i;
filter_info->ethertype_filters[i] = ethertype;
return i;
}
}
return -1;
}
static inline int
igb_ethertype_filter_remove(struct e1000_filter_info *filter_info,
uint8_t idx)
{
if (idx >= E1000_MAX_ETQF_FILTERS)
return -1;
filter_info->ethertype_mask &= ~(1 << idx);
filter_info->ethertype_filters[idx] = 0;
return idx;
}
static int
igb_add_del_ethertype_filter(struct rte_eth_dev *dev,
struct rte_eth_ethertype_filter *filter,
bool add)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
uint32_t etqf = 0;
int ret;
if (filter->ether_type == ETHER_TYPE_IPv4 ||
filter->ether_type == ETHER_TYPE_IPv6) {
PMD_DRV_LOG(ERR, "unsupported ether_type(0x%04x) in"
" ethertype filter.", filter->ether_type);
return -EINVAL;
}
if (filter->flags & RTE_ETHTYPE_FLAGS_MAC) {
PMD_DRV_LOG(ERR, "mac compare is unsupported.");
return -EINVAL;
}
if (filter->flags & RTE_ETHTYPE_FLAGS_DROP) {
PMD_DRV_LOG(ERR, "drop option is unsupported.");
return -EINVAL;
}
ret = igb_ethertype_filter_lookup(filter_info, filter->ether_type);
if (ret >= 0 && add) {
PMD_DRV_LOG(ERR, "ethertype (0x%04x) filter exists.",
filter->ether_type);
return -EEXIST;
}
if (ret < 0 && !add) {
PMD_DRV_LOG(ERR, "ethertype (0x%04x) filter doesn't exist.",
filter->ether_type);
return -ENOENT;
}
if (add) {
ret = igb_ethertype_filter_insert(filter_info,
filter->ether_type);
if (ret < 0) {
PMD_DRV_LOG(ERR, "ethertype filters are full.");
return -ENOSYS;
}
etqf |= E1000_ETQF_FILTER_ENABLE | E1000_ETQF_QUEUE_ENABLE;
etqf |= (uint32_t)(filter->ether_type & E1000_ETQF_ETHERTYPE);
etqf |= filter->queue << E1000_ETQF_QUEUE_SHIFT;
} else {
ret = igb_ethertype_filter_remove(filter_info, (uint8_t)ret);
if (ret < 0)
return -ENOSYS;
}
E1000_WRITE_REG(hw, E1000_ETQF(ret), etqf);
E1000_WRITE_FLUSH(hw);
return 0;
}
static int
igb_get_ethertype_filter(struct rte_eth_dev *dev,
struct rte_eth_ethertype_filter *filter)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_filter_info *filter_info =
E1000_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
uint32_t etqf;
int ret;
ret = igb_ethertype_filter_lookup(filter_info, filter->ether_type);
if (ret < 0) {
PMD_DRV_LOG(ERR, "ethertype (0x%04x) filter doesn't exist.",
filter->ether_type);
return -ENOENT;
}
etqf = E1000_READ_REG(hw, E1000_ETQF(ret));
if (etqf & E1000_ETQF_FILTER_ENABLE) {
filter->ether_type = etqf & E1000_ETQF_ETHERTYPE;
filter->flags = 0;
filter->queue = (etqf & E1000_ETQF_QUEUE) >>
E1000_ETQF_QUEUE_SHIFT;
return 0;
}
return -ENOENT;
}
/*
* igb_ethertype_filter_handle - Handle operations for ethertype filter.
* @dev: pointer to rte_eth_dev structure
* @filter_op:operation will be taken.
* @arg: a pointer to specific structure corresponding to the filter_op
*/
static int
igb_ethertype_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
int ret;
MAC_TYPE_FILTER_SUP(hw->mac.type);
if (filter_op == RTE_ETH_FILTER_NOP)
return 0;
if (arg == NULL) {
PMD_DRV_LOG(ERR, "arg shouldn't be NULL for operation %u.",
filter_op);
return -EINVAL;
}
switch (filter_op) {
case RTE_ETH_FILTER_ADD:
ret = igb_add_del_ethertype_filter(dev,
(struct rte_eth_ethertype_filter *)arg,
TRUE);
break;
case RTE_ETH_FILTER_DELETE:
ret = igb_add_del_ethertype_filter(dev,
(struct rte_eth_ethertype_filter *)arg,
FALSE);
break;
case RTE_ETH_FILTER_GET:
ret = igb_get_ethertype_filter(dev,
(struct rte_eth_ethertype_filter *)arg);
break;
default:
PMD_DRV_LOG(ERR, "unsupported operation %u.", filter_op);
ret = -EINVAL;
break;
}
return ret;
}
static int
eth_igb_filter_ctrl(struct rte_eth_dev *dev,
enum rte_filter_type filter_type,
enum rte_filter_op filter_op,
void *arg)
{
int ret = -EINVAL;
switch (filter_type) {
igb: migrate ntuple filter to new API This patch defines new functions dealing with ntuple filters which is corresponding to 2tuple filter for 82580 and i350 in HW, and to 5tuple filter for 82576 in HW. It removes old functions which deal with 2tuple and 5tuple filters in igb driver. Ntuple filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
2015-02-10 04:48:30 +00:00
case RTE_ETH_FILTER_NTUPLE:
ret = igb_ntuple_filter_handle(dev, filter_op, arg);
break;
igb: use generic filter control for ethertype filter This patch removes old functions which deal with ethertype filter in igb driver. It also defines eth_igb_filter_ctrl which is binding to filter_ctrl API, and ethertype filter can be dealt with through this new entrance. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-01-12 07:16:11 +00:00
case RTE_ETH_FILTER_ETHERTYPE:
ret = igb_ethertype_filter_handle(dev, filter_op, arg);
break;
igb: migrate syn filter to new API This patch defines new functions dealing with syn filter. It removes old functions of syn filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-11 07:51:47 +00:00
case RTE_ETH_FILTER_SYN:
ret = eth_igb_syn_filter_handle(dev, filter_op, arg);
break;
igb: migrate flex filter to new API This patch defines new functions dealing with flex filter. It removes old functions of flex filter in igb driver. Syn filter is dealt with through entrance eth_igb_filter_ctrl. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
2015-02-21 01:53:08 +00:00
case RTE_ETH_FILTER_FLEXIBLE:
ret = eth_igb_flex_filter_handle(dev, filter_op, arg);
break;
igb: use generic filter control for ethertype filter This patch removes old functions which deal with ethertype filter in igb driver. It also defines eth_igb_filter_ctrl which is binding to filter_ctrl API, and ethertype filter can be dealt with through this new entrance. Signed-off-by: Jingjing Wu <jingjing.wu@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
2015-01-12 07:16:11 +00:00
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
break;
}
return ret;
}
e1000: add multicast MAC address filtering Support the PMD function "set_mc_addr_list" in the "igb", "igb-vf", and "em" Poll Mode Drivers. Signed-off-by: Ivan Boule <ivan.boule@6wind.com>
2015-05-28 15:05:21 +00:00
static int
eth_igb_set_mc_addr_list(struct rte_eth_dev *dev,
struct ether_addr *mc_addr_set,
uint32_t nb_mc_addr)
{
struct e1000_hw *hw;
hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
e1000_update_mc_addr_list(hw, (u8 *)mc_addr_set, nb_mc_addr);
return 0;
}
e1000: add ieee1588 timestamping Add e1000/igb support for new ethdev APIs to enable and read IEEE1588 PTP timestamps. Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-07-02 15:16:29 +00:00
static int
igb_timesync_enable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t tsync_ctl;
igb: fix ieee1588 timestamping Ieee1588 reads system time to set its timestamp. On 1G NICs, for example, i350, system time is disabled by default. It means the ieee1588 timestamp will always be 0. This patch enables system time when ieee1588 is enabled. Signed-off-by: Wenzhuo Lu <wenzhuo.lu@intel.com> Acked-by: John McNamara <john.mcnamara@intel.com>
2015-07-31 01:20:30 +00:00
uint32_t tsauxc;
/* Enable system time for it isn't on by default. */
tsauxc = E1000_READ_REG(hw, E1000_TSAUXC);
tsauxc &= ~E1000_TSAUXC_DISABLE_SYSTIME;
E1000_WRITE_REG(hw, E1000_TSAUXC, tsauxc);
e1000: add ieee1588 timestamping Add e1000/igb support for new ethdev APIs to enable and read IEEE1588 PTP timestamps. Signed-off-by: John McNamara <john.mcnamara@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2015-07-02 15:16:29 +00:00
/* Start incrementing the register used to timestamp PTP packets. */
E1000_WRITE_REG(hw, E1000_TIMINCA, E1000_TIMINCA_INIT);
/* Enable L2 filtering of IEEE1588/802.1AS Ethernet frame types. */
E1000_WRITE_REG(hw, E1000_ETQF(E1000_ETQF_FILTER_1588),
(ETHER_TYPE_1588 |
E1000_ETQF_FILTER_ENABLE |
E1000_ETQF_1588));
/* Enable timestamping of received PTP packets. */
tsync_ctl = E1000_READ_REG(hw, E1000_TSYNCRXCTL);
tsync_ctl |= E1000_TSYNCRXCTL_ENABLED;
E1000_WRITE_REG(hw, E1000_TSYNCRXCTL, tsync_ctl);
/* Enable Timestamping of transmitted PTP packets. */
tsync_ctl = E1000_READ_REG(hw, E1000_TSYNCTXCTL);
tsync_ctl |= E1000_TSYNCTXCTL_ENABLED;
E1000_WRITE_REG(hw, E1000_TSYNCTXCTL, tsync_ctl);
return 0;
}
static int
igb_timesync_disable(struct rte_eth_dev *dev)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t tsync_ctl;
/* Disable timestamping of transmitted PTP packets. */
tsync_ctl = E1000_READ_REG(hw, E1000_TSYNCTXCTL);
tsync_ctl &= ~E1000_TSYNCTXCTL_ENABLED;
E1000_WRITE_REG(hw, E1000_TSYNCTXCTL, tsync_ctl);
/* Disable timestamping of received PTP packets. */
tsync_ctl = E1000_READ_REG(hw, E1000_TSYNCRXCTL);
tsync_ctl &= ~E1000_TSYNCRXCTL_ENABLED;
E1000_WRITE_REG(hw, E1000_TSYNCRXCTL, tsync_ctl);
/* Disable L2 filtering of IEEE1588/802.1AS Ethernet frame types. */
E1000_WRITE_REG(hw, E1000_ETQF(E1000_ETQF_FILTER_1588), 0);
/* Stop incrementating the System Time registers. */
E1000_WRITE_REG(hw, E1000_TIMINCA, 0);
return 0;
}
static int
igb_timesync_read_rx_timestamp(struct rte_eth_dev *dev,
struct timespec *timestamp,
uint32_t flags __rte_unused)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t tsync_rxctl;
uint32_t rx_stmpl;
uint32_t rx_stmph;
tsync_rxctl = E1000_READ_REG(hw, E1000_TSYNCRXCTL);
if ((tsync_rxctl & E1000_TSYNCRXCTL_VALID) == 0)
return -EINVAL;
rx_stmpl = E1000_READ_REG(hw, E1000_RXSTMPL);
rx_stmph = E1000_READ_REG(hw, E1000_RXSTMPH);
timestamp->tv_sec = (uint64_t)(((uint64_t)rx_stmph << 32) | rx_stmpl);
timestamp->tv_nsec = 0;
return 0;
}
static int
igb_timesync_read_tx_timestamp(struct rte_eth_dev *dev,
struct timespec *timestamp)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t tsync_txctl;
uint32_t tx_stmpl;
uint32_t tx_stmph;
tsync_txctl = E1000_READ_REG(hw, E1000_TSYNCTXCTL);
if ((tsync_txctl & E1000_TSYNCTXCTL_VALID) == 0)
return -EINVAL;
tx_stmpl = E1000_READ_REG(hw, E1000_TXSTMPL);
tx_stmph = E1000_READ_REG(hw, E1000_TXSTMPH);
timestamp->tv_sec = (uint64_t)(((uint64_t)tx_stmph << 32) | tx_stmpl);
timestamp->tv_nsec = 0;
return 0;
}
igb: add access to specific device info add function to support ethtool ops: - get_reg_length - get_regs - get_eeprom_length - get_eeprom - set_eeprom Signed-off-by: Liang-Min Larry Wang <liang-min.wang@intel.com> Acked-by: Andrew Harvey <agh@cisco.com> Acked-by: David Harton <dharton@cisco.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
2015-07-16 13:25:36 +00:00
static int
eth_igb_get_reg_length(struct rte_eth_dev *dev __rte_unused)
{
int count = 0;
int g_ind = 0;
const struct reg_info *reg_group;
while ((reg_group = igb_regs[g_ind++]))
count += igb_reg_group_count(reg_group);
return count;
}
static int
igbvf_get_reg_length(struct rte_eth_dev *dev __rte_unused)
{
int count = 0;
int g_ind = 0;
const struct reg_info *reg_group;
while ((reg_group = igbvf_regs[g_ind++]))
count += igb_reg_group_count(reg_group);
return count;
}
static int
eth_igb_get_regs(struct rte_eth_dev *dev,
struct rte_dev_reg_info *regs)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t *data = regs->data;
int g_ind = 0;
int count = 0;
const struct reg_info *reg_group;
/* Support only full register dump */
if ((regs->length == 0) ||
(regs->length == (uint32_t)eth_igb_get_reg_length(dev))) {
regs->version = hw->mac.type << 24 | hw->revision_id << 16 |
hw->device_id;
while ((reg_group = igb_regs[g_ind++]))
count += igb_read_regs_group(dev, &data[count],
reg_group);
return 0;
}
return -ENOTSUP;
}
static int
igbvf_get_regs(struct rte_eth_dev *dev,
struct rte_dev_reg_info *regs)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t *data = regs->data;
int g_ind = 0;
int count = 0;
const struct reg_info *reg_group;
/* Support only full register dump */
if ((regs->length == 0) ||
(regs->length == (uint32_t)igbvf_get_reg_length(dev))) {
regs->version = hw->mac.type << 24 | hw->revision_id << 16 |
hw->device_id;
while ((reg_group = igbvf_regs[g_ind++]))
count += igb_read_regs_group(dev, &data[count],
reg_group);
return 0;
}
return -ENOTSUP;
}
static int
eth_igb_get_eeprom_length(struct rte_eth_dev *dev)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
/* Return unit is byte count */
return hw->nvm.word_size * 2;
}
static int
eth_igb_get_eeprom(struct rte_eth_dev *dev,
struct rte_dev_eeprom_info *in_eeprom)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_nvm_info *nvm = &hw->nvm;
uint16_t *data = in_eeprom->data;
int first, length;
first = in_eeprom->offset >> 1;
length = in_eeprom->length >> 1;
if ((first >= hw->nvm.word_size) ||
((first + length) >= hw->nvm.word_size))
return -EINVAL;
in_eeprom->magic = hw->vendor_id |
((uint32_t)hw->device_id << 16);
if ((nvm->ops.read) == NULL)
return -ENOTSUP;
return nvm->ops.read(hw, first, length, data);
}
static int
eth_igb_set_eeprom(struct rte_eth_dev *dev,
struct rte_dev_eeprom_info *in_eeprom)
{
struct e1000_hw *hw = E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct e1000_nvm_info *nvm = &hw->nvm;
uint16_t *data = in_eeprom->data;
int first, length;
first = in_eeprom->offset >> 1;
length = in_eeprom->length >> 1;
if ((first >= hw->nvm.word_size) ||
((first + length) >= hw->nvm.word_size))
return -EINVAL;
in_eeprom->magic = (uint32_t)hw->vendor_id |
((uint32_t)hw->device_id << 16);
if ((nvm->ops.write) == NULL)
return -ENOTSUP;
return nvm->ops.write(hw, first, length, data);
}
igb: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igb pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igb library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:32 +00:00
static struct rte_driver pmd_igb_drv = {
igb: fix indentation of previous patchset There was an indentation error in commit d93c252e88f2b29a5 "convert to use of PMD_REGISTER_DRIVER and fix linking" Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-05-21 11:23:50 +00:00
.type = PMD_PDEV,
igb: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igb pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igb library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:32 +00:00
.init = rte_igb_pmd_init,
};
igbvf: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igbvf pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igbvf library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:33 +00:00
static struct rte_driver pmd_igbvf_drv = {
.type = PMD_PDEV,
.init = rte_igbvf_pmd_init,
};
igb: enable Rx queue interrupts for PF The patch does below for igb PF: - Setup NIC to generate MSI-X interrupts - Set the IVAR register to map interrupt causes to vectors - Implement interrupt enable/disable functions Signed-off-by: Danny Zhou <danny.zhou@intel.com> Signed-off-by: Cunming Liang <cunming.liang@intel.com>
2015-07-20 03:02:28 +00:00
static int
eth_igb_rx_queue_intr_disable(struct rte_eth_dev *dev, uint16_t queue_id)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t mask = 1 << queue_id;
E1000_WRITE_REG(hw, E1000_EIMC, mask);
E1000_WRITE_FLUSH(hw);
return 0;
}
static int
eth_igb_rx_queue_intr_enable(struct rte_eth_dev *dev, uint16_t queue_id)
{
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t mask = 1 << queue_id;
uint32_t regval;
regval = E1000_READ_REG(hw, E1000_EIMS);
E1000_WRITE_REG(hw, E1000_EIMS, regval | mask);
E1000_WRITE_FLUSH(hw);
rte_intr_enable(&dev->pci_dev->intr_handle);
return 0;
}
static void
eth_igb_write_ivar(struct e1000_hw *hw, uint8_t msix_vector,
uint8_t index, uint8_t offset)
{
uint32_t val = E1000_READ_REG_ARRAY(hw, E1000_IVAR0, index);
/* clear bits */
val &= ~((uint32_t)0xFF << offset);
/* write vector and valid bit */
val |= (msix_vector | E1000_IVAR_VALID) << offset;
E1000_WRITE_REG_ARRAY(hw, E1000_IVAR0, index, val);
}
static void
eth_igb_assign_msix_vector(struct e1000_hw *hw, int8_t direction,
uint8_t queue, uint8_t msix_vector)
{
uint32_t tmp = 0;
if (hw->mac.type == e1000_82575) {
if (direction == 0)
tmp = E1000_EICR_RX_QUEUE0 << queue;
else if (direction == 1)
tmp = E1000_EICR_TX_QUEUE0 << queue;
E1000_WRITE_REG(hw, E1000_MSIXBM(msix_vector), tmp);
} else if (hw->mac.type == e1000_82576) {
if ((direction == 0) || (direction == 1))
eth_igb_write_ivar(hw, msix_vector, queue & 0x7,
((queue & 0x8) << 1) +
8 * direction);
} else if ((hw->mac.type == e1000_82580) ||
(hw->mac.type == e1000_i350) ||
(hw->mac.type == e1000_i354) ||
(hw->mac.type == e1000_i210) ||
(hw->mac.type == e1000_i211)) {
if ((direction == 0) || (direction == 1))
eth_igb_write_ivar(hw, msix_vector,
queue >> 1,
((queue & 0x1) << 4) +
8 * direction);
}
}
/* Sets up the hardware to generate MSI-X interrupts properly
* @hw
* board private structure
*/
static void
eth_igb_configure_msix_intr(struct rte_eth_dev *dev)
{
int queue_id;
uint32_t tmpval, regval, intr_mask;
struct e1000_hw *hw =
E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
uint32_t vec = 0;
struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
/* won't configure msix register if no mapping is done
* between intr vector and event fd
*/
if (!rte_intr_dp_is_en(intr_handle))
return;
/* set interrupt vector for other causes */
if (hw->mac.type == e1000_82575) {
tmpval = E1000_READ_REG(hw, E1000_CTRL_EXT);
/* enable MSI-X PBA support */
tmpval |= E1000_CTRL_EXT_PBA_CLR;
/* Auto-Mask interrupts upon ICR read */
tmpval |= E1000_CTRL_EXT_EIAME;
tmpval |= E1000_CTRL_EXT_IRCA;
E1000_WRITE_REG(hw, E1000_CTRL_EXT, tmpval);
/* enable msix_other interrupt */
E1000_WRITE_REG_ARRAY(hw, E1000_MSIXBM(0), 0, E1000_EIMS_OTHER);
regval = E1000_READ_REG(hw, E1000_EIAC);
E1000_WRITE_REG(hw, E1000_EIAC, regval | E1000_EIMS_OTHER);
regval = E1000_READ_REG(hw, E1000_EIAM);
E1000_WRITE_REG(hw, E1000_EIMS, regval | E1000_EIMS_OTHER);
} else if ((hw->mac.type == e1000_82576) ||
(hw->mac.type == e1000_82580) ||
(hw->mac.type == e1000_i350) ||
(hw->mac.type == e1000_i354) ||
(hw->mac.type == e1000_i210) ||
(hw->mac.type == e1000_i211)) {
/* turn on MSI-X capability first */
E1000_WRITE_REG(hw, E1000_GPIE, E1000_GPIE_MSIX_MODE |
E1000_GPIE_PBA | E1000_GPIE_EIAME |
E1000_GPIE_NSICR);
intr_mask = (1 << intr_handle->max_intr) - 1;
regval = E1000_READ_REG(hw, E1000_EIAC);
E1000_WRITE_REG(hw, E1000_EIAC, regval | intr_mask);
/* enable msix_other interrupt */
regval = E1000_READ_REG(hw, E1000_EIMS);
E1000_WRITE_REG(hw, E1000_EIMS, regval | intr_mask);
tmpval = (dev->data->nb_rx_queues | E1000_IVAR_VALID) << 8;
E1000_WRITE_REG(hw, E1000_IVAR_MISC, tmpval);
}
/* use EIAM to auto-mask when MSI-X interrupt
* is asserted, this saves a register write for every interrupt
*/
intr_mask = (1 << intr_handle->nb_efd) - 1;
regval = E1000_READ_REG(hw, E1000_EIAM);
E1000_WRITE_REG(hw, E1000_EIAM, regval | intr_mask);
for (queue_id = 0; queue_id < dev->data->nb_rx_queues; queue_id++) {
eth_igb_assign_msix_vector(hw, 0, queue_id, vec);
intr_handle->intr_vec[queue_id] = vec;
if (vec < intr_handle->nb_efd - 1)
vec++;
}
E1000_WRITE_FLUSH(hw);
}
igb: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igb pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igb library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:32 +00:00
PMD_REGISTER_DRIVER(pmd_igb_drv);
igbvf: convert to use of PMD_REGISTER_DRIVER and fix linking Convert the igbvf pmd driver to use the PMD_REGISTER_DRIVER macro. This means that the test applications now have no reference to the igbvf library when building DSO's and must specify its use on the command line with the -d option. Static linking will still initalize the driver automatically. Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
2014-04-21 14:59:33 +00:00
PMD_REGISTER_DRIVER(pmd_igbvf_drv);
Reference in New Issue Copy Permalink