blob: 5f44c5520fbc86bd9076cc220906e0a4cdee4e90 [file] [log] [blame]
<
/**************************************************************************/
/* */
/* IBM System i and System p Virtual NIC Device Driver */
/* Copyright (C) 2014 IBM Corp. */
/* Santiago Leon (santi_leon@yahoo.com) */
/* Thomas Falcon (tlfalcon@linux.vnet.ibm.com) */
/* John Allen (jallen@linux.vnet.ibm.com) */
/* */
/* This program is free software; you can redistribute it and/or modify */
/* it under the terms of the GNU General Public License as published by */
/* the Free Software Foundation; either version 2 of the License, or */
/* (at your option) any later version. */
/* */
/* This program is distributed in the hope that it will be useful, */
/* but WITHOUT ANY WARRANTY; without even the implied warranty of */
/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the */
/* GNU General Public License for more details. */
/* */
/* You should have received a copy of the GNU General Public License */
/* along with this program. */
/* */
/* This module contains the implementation of a virtual ethernet device */
/* for use with IBM i/p Series LPAR Linux. It utilizes the logical LAN */
/* option of the RS/6000 Platform Architecture to interface with virtual */
/* ethernet NICs that are presented to the partition by the hypervisor. */
/* */
/* Messages are passed between the VNIC driver and the VNIC server using */
/* Command/Response Queues (CRQs) and sub CRQs (sCRQs). CRQs are used to */
/* issue and receive commands that initiate communication with the server */
/* on driver initialization. Sub CRQs (sCRQs) are similar to CRQs, but */
/* are used by the driver to notify the server that a packet is */
/* ready for transmission or that a buffer has been added to receive a */
/* packet. Subsequently, sCRQs are used by the server to notify the */
/* driver that a packet transmission has been completed or that a packet */
/* has been received and placed in a waiting buffer. */
/* */
/* In lieu of a more conventional "on-the-fly" DMA mapping strategy in */
/* which skbs are DMA mapped and immediately unmapped when the transmit */
/* or receive has been completed, the VNIC driver is required to use */
/* "long term mapping". This entails that large, continuous DMA mapped */
/* buffers are allocated on driver initialization and these buffers are */
/* then continuously reused to pass skbs to and from the VNIC server. */
/* */
/**************************************************************************/
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/completion.h>
#include <linux/ioport.h>
#include <linux/dma-mapping.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/skbuff.h>
#include <linux/init.h>
#include <linux/delay.h>
#include <linux/mm.h>
#include <linux/ethtool.h>
#include <linux/proc_fs.h>
#include <linux/in.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/irq.h>
#include <linux/kthread.h>
#include <linux/seq_file.h>
#include <linux/debugfs.h>
#include <linux/interrupt.h>
#include <net/net_namespace.h>
#include <asm/hvcall.h>
#include <linux/atomic.h>
#include <asm/vio.h>
#include <asm/iommu.h>
#include <linux/uaccess.h>
#include <asm/firmware.h>
#include <linux/seq_file.h>
#include <linux/workqueue.h>
#include "ibmvnic.h"
static const char ibmvnic_driver_name[] = "ibmvnic";
static const char ibmvnic_driver_string[] = "IBM System i/p Virtual NIC Driver";
MODULE_AUTHOR("Santiago Leon <santi_leon@yahoo.com>");
MODULE_DESCRIPTION("IBM System i/p Virtual NIC Driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(IBMVNIC_DRIVER_VERSION);
static int ibmvnic_version = IBMVNIC_INITIAL_VERSION;
static int ibmvnic_remove(struct vio_dev *);
static void release_sub_crqs(struct ibmvnic_adapter *);
static void release_sub_crqs_no_irqs(struct ibmvnic_adapter *);
static int ibmvnic_reset_crq(struct ibmvnic_adapter *);
static int ibmvnic_send_crq_init(struct ibmvnic_adapter *);
static int ibmvnic_reenable_crq_queue(struct ibmvnic_adapter *);
static int ibmvnic_send_crq(struct ibmvnic_adapter *, union ibmvnic_crq *);
static int send_subcrq(struct ibmvnic_adapter *adapter, u64 remote_handle,
union sub_crq *sub_crq);
static int send_subcrq_indirect(struct ibmvnic_adapter *, u64, u64, u64);
static irqreturn_t ibmvnic_interrupt_rx(int irq, void *instance);
static int enable_scrq_irq(struct ibmvnic_adapter *,
struct ibmvnic_sub_crq_queue *);
static int disable_scrq_irq(struct ibmvnic_adapter *,
struct ibmvnic_sub_crq_queue *);
static int pending_scrq(struct ibmvnic_adapter *,
struct ibmvnic_sub_crq_queue *);
static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *,
struct ibmvnic_sub_crq_queue *);
static int ibmvnic_poll(struct napi_struct *napi, int data);
static void send_map_query(struct ibmvnic_adapter *adapter);
static void send_request_map(struct ibmvnic_adapter *, dma_addr_t, __be32, u8);
static void send_request_unmap(struct ibmvnic_adapter *, u8);
struct ibmvnic_stat {
char name[ETH_GSTRING_LEN];
int offset;
};
#define IBMVNIC_STAT_OFF(stat) (offsetof(struct ibmvnic_adapter, stats) + \
offsetof(struct ibmvnic_statistics, stat))
#define IBMVNIC_GET_STAT(a, off) (*((u64 *)(((unsigned long)(a)) + off)))
static const struct ibmvnic_stat ibmvnic_stats[] = {
{"rx_packets", IBMVNIC_STAT_OFF(rx_packets)},
{"rx_bytes", IBMVNIC_STAT_OFF(rx_bytes)},
{"tx_packets", IBMVNIC_STAT_OFF(tx_packets)},
{"tx_bytes", IBMVNIC_STAT_OFF(tx_bytes)},
{"ucast_tx_packets", IBMVNIC_STAT_OFF(ucast_tx_packets)},
{"ucast_rx_packets", IBMVNIC_STAT_OFF(ucast_rx_packets)},
{"mcast_tx_packets", IBMVNIC_STAT_OFF(mcast_tx_packets)},
{"mcast_rx_packets", IBMVNIC_STAT_OFF(mcast_rx_packets)},
{"bcast_tx_packets", IBMVNIC_STAT_OFF(bcast_tx_packets)},
{"bcast_rx_packets", IBMVNIC_STAT_OFF(bcast_rx_packets)},
{"align_errors", IBMVNIC_STAT_OFF(align_errors)},
{"fcs_errors", IBMVNIC_STAT_OFF(fcs_errors)},
{"single_collision_frames", IBMVNIC_STAT_OFF(single_collision_frames)},
{"multi_collision_frames", IBMVNIC_STAT_OFF(multi_collision_frames)},
{"sqe_test_errors", IBMVNIC_STAT_OFF(sqe_test_errors)},
{"deferred_tx", IBMVNIC_STAT_OFF(deferred_tx)},
{"late_collisions", IBMVNIC_STAT_OFF(late_collisions)},
{"excess_collisions", IBMVNIC_STAT_OFF(excess_collisions)},
{"internal_mac_tx_errors", IBMVNIC_STAT_OFF(internal_mac_tx_errors)},
{"carrier_sense", IBMVNIC_STAT_OFF(carrier_sense)},
{"too_long_frames", IBMVNIC_STAT_OFF(too_long_frames)},
{"internal_mac_rx_errors", IBMVNIC_STAT_OFF(internal_mac_rx_errors)},
};
static long h_reg_sub_crq(unsigned long unit_address, unsigned long token,
unsigned long length, unsigned long *number,
unsigned long *irq)
{
unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
long rc;
rc = plpar_hcall(H_REG_SUB_CRQ, retbuf, unit_address, token, length);
*number = retbuf[0];
*irq = retbuf[1];
return rc;
}
/* net_device_ops functions */
static void init_rx_pool(struct ibmvnic_adapter *adapter,
struct ibmvnic_rx_pool *rx_pool, int num, int index,
int buff_size, int active)
{
netdev_dbg(adapter->netdev,
"Initializing rx_pool %d, %d buffs, %d bytes each\n",
index, num, buff_size);
rx_pool->size = num;
rx_pool->index = index;
rx_pool->buff_size = buff_size;
rx_pool->active = active;
}
static int alloc_long_term_buff(struct ibmvnic_adapter *adapter,
struct ibmvnic_long_term_buff *ltb, int size)
{
struct device *dev = &adapter->vdev->dev;
ltb->size = size;
ltb->buff = dma_alloc_coherent(dev, ltb->size, &ltb->addr,
GFP_KERNEL);
if (!ltb->buff) {
dev_err(dev, "Couldn't alloc long term buffer\n");
return -ENOMEM;
}
ltb->map_id = adapter->map_id;
adapter->map_id++;
send_request_map(adapter, ltb->addr,
ltb->size, ltb->map_id);
init_completion(&adapter->fw_done);
wait_for_completion(&adapter->fw_done);
return 0;
}
static void free_long_term_buff(struct ibmvnic_adapter *adapter,
struct ibmvnic_long_term_buff *ltb)
{
struct device *dev = &adapter->vdev->dev;
dma_free_coherent(dev, ltb->size, ltb->buff, ltb->addr);
if (!adapter->failover)
send_request_unmap(adapter, ltb->map_id);
}
static int alloc_rx_pool(struct ibmvnic_adapter *adapter,
struct ibmvnic_rx_pool *pool)
{
struct device *dev = &adapter->vdev->dev;
int i;
pool->free_map = kcalloc(pool->size, sizeof(int), GFP_KERNEL);
if (!pool->free_map)
return -ENOMEM;
pool->rx_buff = kcalloc(pool->size, sizeof(struct ibmvnic_rx_buff),
GFP_KERNEL);
if (!pool->rx_buff) {
dev_err(dev, "Couldn't alloc rx buffers\n");
kfree(pool->free_map);
return -ENOMEM;
}
if (alloc_long_term_buff(adapter, &pool->long_term_buff,
pool->size * pool->buff_size)) {
kfree(pool->free_map);
kfree(pool->rx_buff);
return -ENOMEM;
}
for (i = 0; i < pool->size; ++i)
pool->free_map[i] = i;
atomic_set(&pool->available, 0);
pool->next_alloc = 0;
pool->next_free = 0;
return 0;
}
static void replenish_rx_pool(struct ibmvnic_adapter *adapter,
struct ibmvnic_rx_pool *pool)
{
int count = pool->size - atomic_read(&pool->available);
struct device *dev = &adapter->vdev->dev;
int buffers_added = 0;
unsigned long lpar_rc;
union sub_crq sub_crq;
struct sk_buff *skb;
unsigned int offset;
dma_addr_t dma_addr;
unsigned char *dst;
u64 *handle_array;
int shift = 0;
int index;
int i;
handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
be32_to_cpu(adapter->login_rsp_buf->
off_rxadd_subcrqs));
for (i = 0; i < count; ++i) {
skb = alloc_skb(pool->buff_size, GFP_ATOMIC);
if (!skb) {
dev_err(dev, "Couldn't replenish rx buff\n");
adapter->replenish_no_mem++;
break;
}
index = pool->free_map[pool->next_free];
if (pool->rx_buff[index].skb)
dev_err(dev, "Inconsistent free_map!\n");
/* Copy the skb to the long term mapped DMA buffer */
offset = index * pool->buff_size;
dst = pool->long_term_buff.buff + offset;
memset(dst, 0, pool->buff_size);
dma_addr = pool->long_term_buff.addr + offset;
pool->rx_buff[index].data = dst;
pool->free_map[pool->next_free] = IBMVNIC_INVALID_MAP;
pool->rx_buff[index].dma = dma_addr;
pool->rx_buff[index].skb = skb;
pool->rx_buff[index].pool_index = pool->index;
pool->rx_buff[index].size = pool->buff_size;
memset(&sub_crq, 0, sizeof(sub_crq));
sub_crq.rx_add.first = IBMVNIC_CRQ_CMD;
sub_crq.rx_add.correlator =
cpu_to_be64((u64)&pool->rx_buff[index]);
sub_crq.rx_add.ioba = cpu_to_be32(dma_addr);
sub_crq.rx_add.map_id = pool->long_term_buff.map_id;
/* The length field of the sCRQ is defined to be 24 bits so the
* buffer size needs to be left shifted by a byte before it is
* converted to big endian to prevent the last byte from being
* truncated.
*/
#ifdef __LITTLE_ENDIAN__
shift = 8;
#endif
sub_crq.rx_add.len = cpu_to_be32(pool->buff_size << shift);
lpar_rc = send_subcrq(adapter, handle_array[pool->index],
&sub_crq);
if (lpar_rc != H_SUCCESS)
goto failure;
buffers_added++;
adapter->replenish_add_buff_success++;
pool->next_free = (pool->next_free + 1) % pool->size;
}
atomic_add(buffers_added, &pool->available);
return;
failure:
dev_info(dev, "replenish pools failure\n");
pool->free_map[pool->next_free] = index;
pool->rx_buff[index].skb = NULL;
if (!dma_mapping_error(dev, dma_addr))
dma_unmap_single(dev, dma_addr, pool->buff_size,
DMA_FROM_DEVICE);
dev_kfree_skb_any(skb);
adapter->replenish_add_buff_failure++;
atomic_add(buffers_added, &pool->available);
}
static void replenish_pools(struct ibmvnic_adapter *adapter)
{
int i;
if (adapter->migrated)
return;
adapter->replenish_task_cycles++;
for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
i++) {
if (adapter->rx_pool[i].active)
replenish_rx_pool(adapter, &adapter->rx_pool[i]);
}
}
static void free_rx_pool(struct ibmvnic_adapter *adapter,
struct ibmvnic_rx_pool *pool)
{
int i;
kfree(pool->free_map);
pool->free_map = NULL;
if (!pool->rx_buff)
return;
for (i = 0; i < pool->size; i++) {
if (pool->rx_buff[i].skb) {
dev_kfree_skb_any(pool->rx_buff[i].skb);
pool->rx_buff[i].skb = NULL;
}
}
kfree(pool->rx_buff);
pool->rx_buff = NULL;
}
static int ibmvnic_open(struct net_device *netdev)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_tx_pool *tx_pool;
union ibmvnic_crq crq;
int rxadd_subcrqs;
u64 *size_array;
int tx_subcrqs;
int i, j;
rxadd_subcrqs =
be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
tx_subcrqs =
be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
size_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
be32_to_cpu(adapter->login_rsp_buf->
off_rxadd_buff_size));
adapter->map_id = 1;
adapter->napi = kcalloc(adapter->req_rx_queues,
sizeof(struct napi_struct), GFP_KERNEL);
if (!adapter->napi)
goto alloc_napi_failed;
for (i = 0; i < adapter->req_rx_queues; i++) {
netif_napi_add(netdev, &adapter->napi[i], ibmvnic_poll,
NAPI_POLL_WEIGHT);
napi_enable(&adapter->napi[i]);
}
adapter->rx_pool =
kcalloc(rxadd_subcrqs, sizeof(struct ibmvnic_rx_pool), GFP_KERNEL);
if (!adapter->rx_pool)
goto rx_pool_arr_alloc_failed;
send_map_query(adapter);
for (i = 0; i < rxadd_subcrqs; i++) {
init_rx_pool(adapter, &adapter->rx_pool[i],
IBMVNIC_BUFFS_PER_POOL, i,
be64_to_cpu(size_array[i]), 1);
if (alloc_rx_pool(adapter, &adapter->rx_pool[i])) {
dev_err(dev, "Couldn't alloc rx pool\n");
goto rx_pool_alloc_failed;
}
}
adapter->tx_pool =
kcalloc(tx_subcrqs, sizeof(struct ibmvnic_tx_pool), GFP_KERNEL);
if (!adapter->tx_pool)
goto tx_pool_arr_alloc_failed;
for (i = 0; i < tx_subcrqs; i++) {
tx_pool = &adapter->tx_pool[i];
tx_pool->tx_buff =
kcalloc(adapter->max_tx_entries_per_subcrq,
sizeof(struct ibmvnic_tx_buff), GFP_KERNEL);
if (!tx_pool->tx_buff)
goto tx_pool_alloc_failed;
if (alloc_long_term_buff(adapter, &tx_pool->long_term_buff,
adapter->max_tx_entries_per_subcrq *
adapter->req_mtu))
goto tx_ltb_alloc_failed;
tx_pool->free_map =
kcalloc(adapter->max_tx_entries_per_subcrq,
sizeof(int), GFP_KERNEL);
if (!tx_pool->free_map)
goto tx_fm_alloc_failed;
for (j = 0; j < adapter->max_tx_entries_per_subcrq; j++)
tx_pool->free_map[j] = j;
tx_pool->consumer_index = 0;
tx_pool->producer_index = 0;
}
adapter->bounce_buffer_size =
(netdev->mtu + ETH_HLEN - 1) / PAGE_SIZE + 1;
adapter->bounce_buffer = kmalloc(adapter->bounce_buffer_size,
GFP_KERNEL);
if (!adapter->bounce_buffer)
goto bounce_alloc_failed;
adapter->bounce_buffer_dma = dma_map_single(dev, adapter->bounce_buffer,
adapter->bounce_buffer_size,
DMA_TO_DEVICE);
if (dma_mapping_error(dev, adapter->bounce_buffer_dma)) {
dev_err(dev, "Couldn't map tx bounce buffer\n");
goto bounce_map_failed;
}
replenish_pools(adapter);
/* We're ready to receive frames, enable the sub-crq interrupts and
* set the logical link state to up
*/
for (i = 0; i < adapter->req_rx_queues; i++)
enable_scrq_irq(adapter, adapter->rx_scrq[i]);
for (i = 0; i < adapter->req_tx_queues; i++)
enable_scrq_irq(adapter, adapter->tx_scrq[i]);
memset(&crq, 0, sizeof(crq));
crq.logical_link_state.first = IBMVNIC_CRQ_CMD;
crq.logical_link_state.cmd = LOGICAL_LINK_STATE;
crq.logical_link_state.link_state = IBMVNIC_LOGICAL_LNK_UP;
ibmvnic_send_crq(adapter, &crq);
netif_tx_start_all_queues(netdev);
return 0;
bounce_map_failed:
kfree(adapter->bounce_buffer);
bounce_alloc_failed:
i = tx_subcrqs - 1;
kfree(adapter->tx_pool[i].free_map);
tx_fm_alloc_failed:
free_long_term_buff(adapter, &adapter->tx_pool[i].long_term_buff);
tx_ltb_alloc_failed:
kfree(adapter->tx_pool[i].tx_buff);
tx_pool_alloc_failed:
for (j = 0; j < i; j++) {
kfree(adapter->tx_pool[j].tx_buff);
free_long_term_buff(adapter,
&adapter->tx_pool[j].long_term_buff);
kfree(adapter->tx_pool[j].free_map);
}
kfree(adapter->tx_pool);
adapter->tx_pool = NULL;
tx_pool_arr_alloc_failed:
i = rxadd_subcrqs;
rx_pool_alloc_failed:
for (j = 0; j < i; j++) {
free_rx_pool(adapter, &adapter->rx_pool[j]);
free_long_term_buff(adapter,
&adapter->rx_pool[j].long_term_buff);
}
kfree(adapter->rx_pool);
adapter->rx_pool = NULL;
rx_pool_arr_alloc_failed:
for (i = 0; i < adapter->req_rx_queues; i++)
napi_enable(&adapter->napi[i]);
alloc_napi_failed:
return -ENOMEM;
}
static int ibmvnic_close(struct net_device *netdev)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
struct device *dev = &adapter->vdev->dev;
union ibmvnic_crq crq;
int i;
adapter->closing = true;
for (i = 0; i < adapter->req_rx_queues; i++)
napi_disable(&adapter->napi[i]);
if (!adapter->failover)
netif_tx_stop_all_queues(netdev);
if (adapter->bounce_buffer) {
if (!dma_mapping_error(dev, adapter->bounce_buffer_dma)) {
dma_unmap_single(&adapter->vdev->dev,
adapter->bounce_buffer_dma,
adapter->bounce_buffer_size,
DMA_BIDIRECTIONAL);
adapter->bounce_buffer_dma = DMA_ERROR_CODE;
}
kfree(adapter->bounce_buffer);
adapter->bounce_buffer = NULL;
}
memset(&crq, 0, sizeof(crq));
crq.logical_link_state.first = IBMVNIC_CRQ_CMD;
crq.logical_link_state.cmd = LOGICAL_LINK_STATE;
crq.logical_link_state.link_state = IBMVNIC_LOGICAL_LNK_DN;
ibmvnic_send_crq(adapter, &crq);
for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
i++) {
kfree(adapter->tx_pool[i].tx_buff);
free_long_term_buff(adapter,
&adapter->tx_pool[i].long_term_buff);
kfree(adapter->tx_pool[i].free_map);
}
kfree(adapter->tx_pool);
adapter->tx_pool = NULL;
for (i = 0; i < be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
i++) {
free_rx_pool(adapter, &adapter->rx_pool[i]);
free_long_term_buff(adapter,
&adapter->rx_pool[i].long_term_buff);
}
kfree(adapter->rx_pool);
adapter->rx_pool = NULL;
adapter->closing = false;
return 0;
}
/**
* build_hdr_data - creates L2/L3/L4 header data buffer
* @hdr_field - bitfield determining needed headers
* @skb - socket buffer
* @hdr_len - array of header lengths
* @tot_len - total length of data
*
* Reads hdr_field to determine which headers are needed by firmware.
* Builds a buffer containing these headers. Saves individual header
* lengths and total buffer length to be used to build descriptors.
*/
static int build_hdr_data(u8 hdr_field, struct sk_buff *skb,
int *hdr_len, u8 *hdr_data)
{
int len = 0;
u8 *hdr;
hdr_len[0] = sizeof(struct ethhdr);
if (skb->protocol == htons(ETH_P_IP)) {
hdr_len[1] = ip_hdr(skb)->ihl * 4;
if (ip_hdr(skb)->protocol == IPPROTO_TCP)
hdr_len[2] = tcp_hdrlen(skb);
else if (ip_hdr(skb)->protocol == IPPROTO_UDP)
hdr_len[2] = sizeof(struct udphdr);
} else if (skb->protocol == htons(ETH_P_IPV6)) {
hdr_len[1] = sizeof(struct ipv6hdr);
if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
hdr_len[2] = tcp_hdrlen(skb);
else if (ipv6_hdr(skb)->nexthdr == IPPROTO_UDP)
hdr_len[2] = sizeof(struct udphdr);
}
memset(hdr_data, 0, 120);
if ((hdr_field >> 6) & 1) {
hdr = skb_mac_header(skb);
memcpy(hdr_data, hdr, hdr_len[0]);
len += hdr_len[0];
}
if ((hdr_field >> 5) & 1) {
hdr = skb_network_header(skb);
memcpy(hdr_data + len, hdr, hdr_len[1]);
len += hdr_len[1];
}
if ((hdr_field >> 4) & 1) {
hdr = skb_transport_header(skb);
memcpy(hdr_data + len, hdr, hdr_len[2]);
len += hdr_len[2];
}
return len;
}
/**
* create_hdr_descs - create header and header extension descriptors
* @hdr_field - bitfield determining needed headers
* @data - buffer containing header data
* @len - length of data buffer
* @hdr_len - array of individual header lengths
* @scrq_arr - descriptor array
*
* Creates header and, if needed, header extension descriptors and
* places them in a descriptor array, scrq_arr
*/
static void create_hdr_descs(u8 hdr_field, u8 *hdr_data, int len, int *hdr_len,
union sub_crq *scrq_arr)
{
union sub_crq hdr_desc;
int tmp_len = len;
u8 *data, *cur;
int tmp;
while (tmp_len > 0) {
cur = hdr_data + len - tmp_len;
memset(&hdr_desc, 0, sizeof(hdr_desc));
if (cur != hdr_data) {
data = hdr_desc.hdr_ext.data;
tmp = tmp_len > 29 ? 29 : tmp_len;
hdr_desc.hdr_ext.first = IBMVNIC_CRQ_CMD;
hdr_desc.hdr_ext.type = IBMVNIC_HDR_EXT_DESC;
hdr_desc.hdr_ext.len = tmp;
} else {
data = hdr_desc.hdr.data;
tmp = tmp_len > 24 ? 24 : tmp_len;
hdr_desc.hdr.first = IBMVNIC_CRQ_CMD;
hdr_desc.hdr.type = IBMVNIC_HDR_DESC;
hdr_desc.hdr.len = tmp;
hdr_desc.hdr.l2_len = (u8)hdr_len[0];
hdr_desc.hdr.l3_len = cpu_to_be16((u16)hdr_len[1]);
hdr_desc.hdr.l4_len = (u8)hdr_len[2];
hdr_desc.hdr.flag = hdr_field << 1;
}
memcpy(data, cur, tmp);
tmp_len -= tmp;
*scrq_arr = hdr_desc;
scrq_arr++;
}
}
/**
* build_hdr_descs_arr - build a header descriptor array
* @skb - socket buffer
* @num_entries - number of descriptors to be sent
* @subcrq - first TX descriptor
* @hdr_field - bit field determining which headers will be sent
*
* This function will build a TX descriptor array with applicable
* L2/L3/L4 packet header descriptors to be sent by send_subcrq_indirect.
*/
static void build_hdr_descs_arr(struct ibmvnic_tx_buff *txbuff,
int *num_entries, u8 hdr_field)
{
int hdr_len[3] = {0, 0, 0};
int tot_len, len;
u8 *hdr_data = txbuff->hdr_data;
tot_len = build_hdr_data(hdr_field, txbuff->skb, hdr_len,
txbuff->hdr_data);
len = tot_len;
len -= 24;
if (len > 0)
num_entries += len % 29 ? len / 29 + 1 : len / 29;
create_hdr_descs(hdr_field, hdr_data, tot_len, hdr_len,
txbuff->indir_arr + 1);
}
static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
int queue_num = skb_get_queue_mapping(skb);
u8 *hdrs = (u8 *)&adapter->tx_rx_desc_req;
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_tx_buff *tx_buff = NULL;
struct ibmvnic_tx_pool *tx_pool;
unsigned int tx_send_failed = 0;
unsigned int tx_map_failed = 0;
unsigned int tx_dropped = 0;
unsigned int tx_packets = 0;
unsigned int tx_bytes = 0;
dma_addr_t data_dma_addr;
struct netdev_queue *txq;
bool used_bounce = false;
unsigned long lpar_rc;
union sub_crq tx_crq;
unsigned int offset;
int num_entries = 1;
unsigned char *dst;
u64 *handle_array;
int index = 0;
int ret = 0;
tx_pool = &adapter->tx_pool[queue_num];
txq = netdev_get_tx_queue(netdev, skb_get_queue_mapping(skb));
handle_array = (u64 *)((u8 *)(adapter->login_rsp_buf) +
be32_to_cpu(adapter->login_rsp_buf->
off_txsubm_subcrqs));
if (adapter->migrated) {
tx_send_failed++;
tx_dropped++;
ret = NETDEV_TX_BUSY;
goto out;
}
index = tx_pool->free_map[tx_pool->consumer_index];
offset = index * adapter->req_mtu;
dst = tx_pool->long_term_buff.buff + offset;
memset(dst, 0, adapter->req_mtu);
skb_copy_from_linear_data(skb, dst, skb->len);
data_dma_addr = tx_pool->long_term_buff.addr + offset;
tx_pool->consumer_index =
(tx_pool->consumer_index + 1) %
adapter->max_tx_entries_per_subcrq;
tx_buff = &tx_pool->tx_buff[index];
tx_buff->skb = skb;
tx_buff->data_dma[0] = data_dma_addr;
tx_buff->data_len[0] = skb->len;
tx_buff->index = index;
tx_buff->pool_index = queue_num;
tx_buff->last_frag = true;
tx_buff->used_bounce = used_bounce;
memset(&tx_crq, 0, sizeof(tx_crq));
tx_crq.v1.first = IBMVNIC_CRQ_CMD;
tx_crq.v1.type = IBMVNIC_TX_DESC;
tx_crq.v1.n_crq_elem = 1;
tx_crq.v1.n_sge = 1;
tx_crq.v1.flags1 = IBMVNIC_TX_COMP_NEEDED;
tx_crq.v1.correlator = cpu_to_be32(index);
tx_crq.v1.dma_reg = cpu_to_be16(tx_pool->long_term_buff.map_id);
tx_crq.v1.sge_len = cpu_to_be32(skb->len);
tx_crq.v1.ioba = cpu_to_be64(data_dma_addr);
if (adapter->vlan_header_insertion) {
tx_crq.v1.flags2 |= IBMVNIC_TX_VLAN_INSERT;
tx_crq.v1.vlan_id = cpu_to_be16(skb->vlan_tci);
}
if (skb->protocol == htons(ETH_P_IP)) {
if (ip_hdr(skb)->version == 4)
tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_IPV4;
else if (ip_hdr(skb)->version == 6)
tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_IPV6;
if (ip_hdr(skb)->protocol == IPPROTO_TCP)
tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_TCP;
else if (ip_hdr(skb)->protocol != IPPROTO_TCP)
tx_crq.v1.flags1 |= IBMVNIC_TX_PROT_UDP;
}
if (skb->ip_summed == CHECKSUM_PARTIAL) {
tx_crq.v1.flags1 |= IBMVNIC_TX_CHKSUM_OFFLOAD;
hdrs += 2;
}
/* determine if l2/3/4 headers are sent to firmware */
if ((*hdrs >> 7) & 1 &&
(skb->protocol == htons(ETH_P_IP) ||
skb->protocol == htons(ETH_P_IPV6))) {
build_hdr_descs_arr(tx_buff, &num_entries, *hdrs);
tx_crq.v1.n_crq_elem = num_entries;
tx_buff->indir_arr[0] = tx_crq;
tx_buff->indir_dma = dma_map_single(dev, tx_buff->indir_arr,
sizeof(tx_buff->indir_arr),
DMA_TO_DEVICE);
if (dma_mapping_error(dev, tx_buff->indir_dma)) {
if (!firmware_has_feature(FW_FEATURE_CMO))
dev_err(dev, "tx: unable to map descriptor array\n");
tx_map_failed++;
tx_dropped++;
ret = NETDEV_TX_BUSY;
goto out;
}
lpar_rc = send_subcrq_indirect(adapter, handle_array[queue_num],
(u64)tx_buff->indir_dma,
(u64)num_entries);
} else {
lpar_rc = send_subcrq(adapter, handle_array[queue_num],
&tx_crq);
}
if (lpar_rc != H_SUCCESS) {
dev_err(dev, "tx failed with code %ld\n", lpar_rc);
if (tx_pool->consumer_index == 0)
tx_pool->consumer_index =
adapter->max_tx_entries_per_subcrq - 1;
else
tx_pool->consumer_index--;
tx_send_failed++;
tx_dropped++;
ret = NETDEV_TX_BUSY;
goto out;
}
tx_packets++;
tx_bytes += skb->len;
txq->trans_start = jiffies;
ret = NETDEV_TX_OK;
out:
netdev->stats.tx_dropped += tx_dropped;
netdev->stats.tx_bytes += tx_bytes;
netdev->stats.tx_packets += tx_packets;
adapter->tx_send_failed += tx_send_failed;
adapter->tx_map_failed += tx_map_failed;
return ret;
}
static void ibmvnic_set_multi(struct net_device *netdev)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
struct netdev_hw_addr *ha;
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.request_capability.first = IBMVNIC_CRQ_CMD;
crq.request_capability.cmd = REQUEST_CAPABILITY;
if (netdev->flags & IFF_PROMISC) {
if (!adapter->promisc_supported)
return;
} else {
if (netdev->flags & IFF_ALLMULTI) {
/* Accept all multicast */
memset(&crq, 0, sizeof(crq));
crq.multicast_ctrl.first = IBMVNIC_CRQ_CMD;
crq.multicast_ctrl.cmd = MULTICAST_CTRL;
crq.multicast_ctrl.flags = IBMVNIC_ENABLE_ALL;
ibmvnic_send_crq(adapter, &crq);
} else if (netdev_mc_empty(netdev)) {
/* Reject all multicast */
memset(&crq, 0, sizeof(crq));
crq.multicast_ctrl.first = IBMVNIC_CRQ_CMD;
crq.multicast_ctrl.cmd = MULTICAST_CTRL;
crq.multicast_ctrl.flags = IBMVNIC_DISABLE_ALL;
ibmvnic_send_crq(adapter, &crq);
} else {
/* Accept one or more multicast(s) */
netdev_for_each_mc_addr(ha, netdev) {
memset(&crq, 0, sizeof(crq));
crq.multicast_ctrl.first = IBMVNIC_CRQ_CMD;
crq.multicast_ctrl.cmd = MULTICAST_CTRL;
crq.multicast_ctrl.flags = IBMVNIC_ENABLE_MC;
ether_addr_copy(&crq.multicast_ctrl.mac_addr[0],
ha->addr);
ibmvnic_send_crq(adapter, &crq);
}
}
}
}
static int ibmvnic_set_mac(struct net_device *netdev, void *p)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
struct sockaddr *addr = p;
union ibmvnic_crq crq;
if (!is_valid_ether_addr(addr->sa_data))
return -EADDRNOTAVAIL;
memset(&crq, 0, sizeof(crq));
crq.change_mac_addr.first = IBMVNIC_CRQ_CMD;
crq.change_mac_addr.cmd = CHANGE_MAC_ADDR;
ether_addr_copy(&crq.change_mac_addr.mac_addr[0], addr->sa_data);
ibmvnic_send_crq(adapter, &crq);
/* netdev->dev_addr is changed in handle_change_mac_rsp function */
return 0;
}
static int ibmvnic_change_mtu(struct net_device *netdev, int new_mtu)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
if (new_mtu > adapter->req_mtu || new_mtu < adapter->min_mtu)
return -EINVAL;
netdev->mtu = new_mtu;
return 0;
}
static void ibmvnic_tx_timeout(struct net_device *dev)
{
struct ibmvnic_adapter *adapter = netdev_priv(dev);
int rc;
/* Adapter timed out, resetting it */
release_sub_crqs(adapter);
rc = ibmvnic_reset_crq(adapter);
if (rc)
dev_err(&adapter->vdev->dev, "Adapter timeout, reset failed\n");
else
ibmvnic_send_crq_init(adapter);
}
static void remove_buff_from_pool(struct ibmvnic_adapter *adapter,
struct ibmvnic_rx_buff *rx_buff)
{
struct ibmvnic_rx_pool *pool = &adapter->rx_pool[rx_buff->pool_index];
rx_buff->skb = NULL;
pool->free_map[pool->next_alloc] = (int)(rx_buff - pool->rx_buff);
pool->next_alloc = (pool->next_alloc + 1) % pool->size;
atomic_dec(&pool->available);
}
static int ibmvnic_poll(struct napi_struct *napi, int budget)
{
struct net_device *netdev = napi->dev;
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
int scrq_num = (int)(napi - adapter->napi);
int frames_processed = 0;
restart_poll:
while (frames_processed < budget) {
struct sk_buff *skb;
struct ibmvnic_rx_buff *rx_buff;
union sub_crq *next;
u32 length;
u16 offset;
u8 flags = 0;
if (!pending_scrq(adapter, adapter->rx_scrq[scrq_num]))
break;
next = ibmvnic_next_scrq(adapter, adapter->rx_scrq[scrq_num]);
rx_buff =
(struct ibmvnic_rx_buff *)be64_to_cpu(next->
rx_comp.correlator);
/* do error checking */
if (next->rx_comp.rc) {
netdev_err(netdev, "rx error %x\n", next->rx_comp.rc);
/* free the entry */
next->rx_comp.first = 0;
remove_buff_from_pool(adapter, rx_buff);
break;
}
length = be32_to_cpu(next->rx_comp.len);
offset = be16_to_cpu(next->rx_comp.off_frame_data);
flags = next->rx_comp.flags;
skb = rx_buff->skb;
skb_copy_to_linear_data(skb, rx_buff->data + offset,
length);
skb->vlan_tci = be16_to_cpu(next->rx_comp.vlan_tci);
/* free the entry */
next->rx_comp.first = 0;
remove_buff_from_pool(adapter, rx_buff);
skb_put(skb, length);
skb->protocol = eth_type_trans(skb, netdev);
if (flags & IBMVNIC_IP_CHKSUM_GOOD &&
flags & IBMVNIC_TCP_UDP_CHKSUM_GOOD) {
skb->ip_summed = CHECKSUM_UNNECESSARY;
}
length = skb->len;
napi_gro_receive(napi, skb); /* send it up */
netdev->stats.rx_packets++;
netdev->stats.rx_bytes += length;
frames_processed++;
}
replenish_rx_pool(adapter, &adapter->rx_pool[scrq_num]);
if (frames_processed < budget) {
enable_scrq_irq(adapter, adapter->rx_scrq[scrq_num]);
napi_complete(napi);
if (pending_scrq(adapter, adapter->rx_scrq[scrq_num]) &&
napi_reschedule(napi)) {
disable_scrq_irq(adapter, adapter->rx_scrq[scrq_num]);
goto restart_poll;
}
}
return frames_processed;
}
#ifdef CONFIG_NET_POLL_CONTROLLER
static void ibmvnic_netpoll_controller(struct net_device *dev)
{
struct ibmvnic_adapter *adapter = netdev_priv(dev);
int i;
replenish_pools(netdev_priv(dev));
for (i = 0; i < adapter->req_rx_queues; i++)
ibmvnic_interrupt_rx(adapter->rx_scrq[i]->irq,
adapter->rx_scrq[i]);
}
#endif
static const struct net_device_ops ibmvnic_netdev_ops = {
.ndo_open = ibmvnic_open,
.ndo_stop = ibmvnic_close,
.ndo_start_xmit = ibmvnic_xmit,
.ndo_set_rx_mode = ibmvnic_set_multi,
.ndo_set_mac_address = ibmvnic_set_mac,
.ndo_validate_addr = eth_validate_addr,
.ndo_change_mtu = ibmvnic_change_mtu,
.ndo_tx_timeout = ibmvnic_tx_timeout,
#ifdef CONFIG_NET_POLL_CONTROLLER
.ndo_poll_controller = ibmvnic_netpoll_controller,
#endif
};
/* ethtool functions */
static int ibmvnic_get_settings(struct net_device *netdev,
struct ethtool_cmd *cmd)
{
cmd->supported = (SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg |
SUPPORTED_FIBRE);
cmd->advertising = (ADVERTISED_1000baseT_Full | ADVERTISED_Autoneg |
ADVERTISED_FIBRE);
ethtool_cmd_speed_set(cmd, SPEED_1000);
cmd->duplex = DUPLEX_FULL;
cmd->port = PORT_FIBRE;
cmd->phy_address = 0;
cmd->transceiver = XCVR_INTERNAL;
cmd->autoneg = AUTONEG_ENABLE;
cmd->maxtxpkt = 0;
cmd->maxrxpkt = 1;
return 0;
}
static void ibmvnic_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
{
strlcpy(info->driver, ibmvnic_driver_name, sizeof(info->driver));
strlcpy(info->version, IBMVNIC_DRIVER_VERSION, sizeof(info->version));
}
static u32 ibmvnic_get_msglevel(struct net_device *netdev)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
return adapter->msg_enable;
}
static void ibmvnic_set_msglevel(struct net_device *netdev, u32 data)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
adapter->msg_enable = data;
}
static u32 ibmvnic_get_link(struct net_device *netdev)
{
struct ibmvnic_adapter *adapter = netdev_priv(netdev);
/* Don't need to send a query because we request a logical link up at
* init and then we wait for link state indications
*/
return adapter->logical_link_state;
}
static void ibmvnic_get_ringparam(struct net_device *netdev,
struct ethtool_ringparam *ring)
{
ring->rx_max_pending = 0;
ring->tx_max_pending = 0;
ring->rx_mini_max_pending = 0;
ring->rx_jumbo_max_pending = 0;
ring->rx_pending = 0;
ring->tx_pending = 0;
ring->rx_mini_pending = 0;
ring->rx_jumbo_pending = 0;
}
static void ibmvnic_get_strings(struct net_device *dev, u32 stringset, u8 *data)
{
int i;
if (stringset != ETH_SS_STATS)
return;
for (i = 0; i < ARRAY_SIZE(ibmvnic_stats); i++, data += ETH_GSTRING_LEN)
memcpy(data, ibmvnic_stats[i].name, ETH_GSTRING_LEN);
}
static int ibmvnic_get_sset_count(struct net_device *dev, int sset)
{
switch (sset) {
case ETH_SS_STATS:
return ARRAY_SIZE(ibmvnic_stats);
default:
return -EOPNOTSUPP;
}
}
static void ibmvnic_get_ethtool_stats(struct net_device *dev,
struct ethtool_stats *stats, u64 *data)
{
struct ibmvnic_adapter *adapter = netdev_priv(dev);
union ibmvnic_crq crq;
int i;
memset(&crq, 0, sizeof(crq));
crq.request_statistics.first = IBMVNIC_CRQ_CMD;
crq.request_statistics.cmd = REQUEST_STATISTICS;
crq.request_statistics.ioba = cpu_to_be32(adapter->stats_token);
crq.request_statistics.len =
cpu_to_be32(sizeof(struct ibmvnic_statistics));
ibmvnic_send_crq(adapter, &crq);
/* Wait for data to be written */
init_completion(&adapter->stats_done);
wait_for_completion(&adapter->stats_done);
for (i = 0; i < ARRAY_SIZE(ibmvnic_stats); i++)
data[i] = IBMVNIC_GET_STAT(adapter, ibmvnic_stats[i].offset);
}
static const struct ethtool_ops ibmvnic_ethtool_ops = {
.get_settings = ibmvnic_get_settings,
.get_drvinfo = ibmvnic_get_drvinfo,
.get_msglevel = ibmvnic_get_msglevel,
.set_msglevel = ibmvnic_set_msglevel,
.get_link = ibmvnic_get_link,
.get_ringparam = ibmvnic_get_ringparam,
.get_strings = ibmvnic_get_strings,
.get_sset_count = ibmvnic_get_sset_count,
.get_ethtool_stats = ibmvnic_get_ethtool_stats,
};
/* Routines for managing CRQs/sCRQs */
static void release_sub_crq_queue(struct ibmvnic_adapter *adapter,
struct ibmvnic_sub_crq_queue *scrq)
{
struct device *dev = &adapter->vdev->dev;
long rc;
netdev_dbg(adapter->netdev, "Releasing sub-CRQ\n");
/* Close the sub-crqs */
do {
rc = plpar_hcall_norets(H_FREE_SUB_CRQ,
adapter->vdev->unit_address,
scrq->crq_num);
} while (rc == H_BUSY || H_IS_LONG_BUSY(rc));
dma_unmap_single(dev, scrq->msg_token, 4 * PAGE_SIZE,
DMA_BIDIRECTIONAL);
free_pages((unsigned long)scrq->msgs, 2);
kfree(scrq);
}
static struct ibmvnic_sub_crq_queue *init_sub_crq_queue(struct ibmvnic_adapter
*adapter)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_sub_crq_queue *scrq;
int rc;
scrq = kmalloc(sizeof(*scrq), GFP_ATOMIC);
if (!scrq)
return NULL;
scrq->msgs = (union sub_crq *)__get_free_pages(GFP_ATOMIC, 2);
memset(scrq->msgs, 0, 4 * PAGE_SIZE);
if (!scrq->msgs) {
dev_warn(dev, "Couldn't allocate crq queue messages page\n");
goto zero_page_failed;
}
scrq->msg_token = dma_map_single(dev, scrq->msgs, 4 * PAGE_SIZE,
DMA_BIDIRECTIONAL);
if (dma_mapping_error(dev, scrq->msg_token)) {
dev_warn(dev, "Couldn't map crq queue messages page\n");
goto map_failed;
}
rc = h_reg_sub_crq(adapter->vdev->unit_address, scrq->msg_token,
4 * PAGE_SIZE, &scrq->crq_num, &scrq->hw_irq);
if (rc == H_RESOURCE)
rc = ibmvnic_reset_crq(adapter);
if (rc == H_CLOSED) {
dev_warn(dev, "Partner adapter not ready, waiting.\n");
} else if (rc) {
dev_warn(dev, "Error %d registering sub-crq\n", rc);
goto reg_failed;
}
scrq->adapter = adapter;
scrq->size = 4 * PAGE_SIZE / sizeof(*scrq->msgs);
scrq->cur = 0;
scrq->rx_skb_top = NULL;
spin_lock_init(&scrq->lock);
netdev_dbg(adapter->netdev,
"sub-crq initialized, num %lx, hw_irq=%lx, irq=%x\n",
scrq->crq_num, scrq->hw_irq, scrq->irq);
return scrq;
reg_failed:
dma_unmap_single(dev, scrq->msg_token, 4 * PAGE_SIZE,
DMA_BIDIRECTIONAL);
map_failed:
free_pages((unsigned long)scrq->msgs, 2);
zero_page_failed:
kfree(scrq);
return NULL;
}
static void release_sub_crqs(struct ibmvnic_adapter *adapter)
{
int i;
if (adapter->tx_scrq) {
for (i = 0; i < adapter->req_tx_queues; i++)
if (adapter->tx_scrq[i]) {
free_irq(adapter->tx_scrq[i]->irq,
adapter->tx_scrq[i]);
irq_dispose_mapping(adapter->tx_scrq[i]->irq);
release_sub_crq_queue(adapter,
adapter->tx_scrq[i]);
}
adapter->tx_scrq = NULL;
}
if (adapter->rx_scrq) {
for (i = 0; i < adapter->req_rx_queues; i++)
if (adapter->rx_scrq[i]) {
free_irq(adapter->rx_scrq[i]->irq,
adapter->rx_scrq[i]);
irq_dispose_mapping(adapter->rx_scrq[i]->irq);
release_sub_crq_queue(adapter,
adapter->rx_scrq[i]);
}
adapter->rx_scrq = NULL;
}
adapter->requested_caps = 0;
}
static void release_sub_crqs_no_irqs(struct ibmvnic_adapter *adapter)
{
int i;
if (adapter->tx_scrq) {
for (i = 0; i < adapter->req_tx_queues; i++)
if (adapter->tx_scrq[i])
release_sub_crq_queue(adapter,
adapter->tx_scrq[i]);
adapter->tx_scrq = NULL;
}
if (adapter->rx_scrq) {
for (i = 0; i < adapter->req_rx_queues; i++)
if (adapter->rx_scrq[i])
release_sub_crq_queue(adapter,
adapter->rx_scrq[i]);
adapter->rx_scrq = NULL;
}
adapter->requested_caps = 0;
}
static int disable_scrq_irq(struct ibmvnic_adapter *adapter,
struct ibmvnic_sub_crq_queue *scrq)
{
struct device *dev = &adapter->vdev->dev;
unsigned long rc;
rc = plpar_hcall_norets(H_VIOCTL, adapter->vdev->unit_address,
H_DISABLE_VIO_INTERRUPT, scrq->hw_irq, 0, 0);
if (rc)
dev_err(dev, "Couldn't disable scrq irq 0x%lx. rc=%ld\n",
scrq->hw_irq, rc);
return rc;
}
static int enable_scrq_irq(struct ibmvnic_adapter *adapter,
struct ibmvnic_sub_crq_queue *scrq)
{
struct device *dev = &adapter->vdev->dev;
unsigned long rc;
if (scrq->hw_irq > 0x100000000ULL) {
dev_err(dev, "bad hw_irq = %lx\n", scrq->hw_irq);
return 1;
}
rc = plpar_hcall_norets(H_VIOCTL, adapter->vdev->unit_address,
H_ENABLE_VIO_INTERRUPT, scrq->hw_irq, 0, 0);
if (rc)
dev_err(dev, "Couldn't enable scrq irq 0x%lx. rc=%ld\n",
scrq->hw_irq, rc);
return rc;
}
static int ibmvnic_complete_tx(struct ibmvnic_adapter *adapter,
struct ibmvnic_sub_crq_queue *scrq)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_tx_buff *txbuff;
union sub_crq *next;
int index;
int i, j;
u8 first;
restart_loop:
while (pending_scrq(adapter, scrq)) {
unsigned int pool = scrq->pool_index;
next = ibmvnic_next_scrq(adapter, scrq);
for (i = 0; i < next->tx_comp.num_comps; i++) {
if (next->tx_comp.rcs[i]) {
dev_err(dev, "tx error %x\n",
next->tx_comp.rcs[i]);
continue;
}
index = be32_to_cpu(next->tx_comp.correlators[i]);
txbuff = &adapter->tx_pool[pool].tx_buff[index];
for (j = 0; j < IBMVNIC_MAX_FRAGS_PER_CRQ; j++) {
if (!txbuff->data_dma[j])
continue;
txbuff->data_dma[j] = 0;
txbuff->used_bounce = false;
}
/* if sub_crq was sent indirectly */
first = txbuff->indir_arr[0].generic.first;
if (first == IBMVNIC_CRQ_CMD) {
dma_unmap_single(dev, txbuff->indir_dma,
sizeof(txbuff->indir_arr),
DMA_TO_DEVICE);
}
if (txbuff->last_frag)
dev_kfree_skb_any(txbuff->skb);
adapter->tx_pool[pool].free_map[adapter->tx_pool[pool].
producer_index] = index;
adapter->tx_pool[pool].producer_index =
(adapter->tx_pool[pool].producer_index + 1) %
adapter->max_tx_entries_per_subcrq;
}
/* remove tx_comp scrq*/
next->tx_comp.first = 0;
}
enable_scrq_irq(adapter, scrq);
if (pending_scrq(adapter, scrq)) {
disable_scrq_irq(adapter, scrq);
goto restart_loop;
}
return 0;
}
static irqreturn_t ibmvnic_interrupt_tx(int irq, void *instance)
{
struct ibmvnic_sub_crq_queue *scrq = instance;
struct ibmvnic_adapter *adapter = scrq->adapter;
disable_scrq_irq(adapter, scrq);
ibmvnic_complete_tx(adapter, scrq);
return IRQ_HANDLED;
}
static irqreturn_t ibmvnic_interrupt_rx(int irq, void *instance)
{
struct ibmvnic_sub_crq_queue *scrq = instance;
struct ibmvnic_adapter *adapter = scrq->adapter;
if (napi_schedule_prep(&adapter->napi[scrq->scrq_num])) {
disable_scrq_irq(adapter, scrq);
__napi_schedule(&adapter->napi[scrq->scrq_num]);
}
return IRQ_HANDLED;
}
static int init_sub_crq_irqs(struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_sub_crq_queue *scrq;
int i = 0, j = 0;
int rc = 0;
for (i = 0; i < adapter->req_tx_queues; i++) {
scrq = adapter->tx_scrq[i];
scrq->irq = irq_create_mapping(NULL, scrq->hw_irq);
if (!scrq->irq) {
rc = -EINVAL;
dev_err(dev, "Error mapping irq\n");
goto req_tx_irq_failed;
}
rc = request_irq(scrq->irq, ibmvnic_interrupt_tx,
0, "ibmvnic_tx", scrq);
if (rc) {
dev_err(dev, "Couldn't register tx irq 0x%x. rc=%d\n",
scrq->irq, rc);
irq_dispose_mapping(scrq->irq);
goto req_rx_irq_failed;
}
}
for (i = 0; i < adapter->req_rx_queues; i++) {
scrq = adapter->rx_scrq[i];
scrq->irq = irq_create_mapping(NULL, scrq->hw_irq);
if (!scrq->irq) {
rc = -EINVAL;
dev_err(dev, "Error mapping irq\n");
goto req_rx_irq_failed;
}
rc = request_irq(scrq->irq, ibmvnic_interrupt_rx,
0, "ibmvnic_rx", scrq);
if (rc) {
dev_err(dev, "Couldn't register rx irq 0x%x. rc=%d\n",
scrq->irq, rc);
irq_dispose_mapping(scrq->irq);
goto req_rx_irq_failed;
}
}
return rc;
req_rx_irq_failed:
for (j = 0; j < i; j++) {
free_irq(adapter->rx_scrq[j]->irq, adapter->rx_scrq[j]);
irq_dispose_mapping(adapter->rx_scrq[j]->irq);
}
i = adapter->req_tx_queues;
req_tx_irq_failed:
for (j = 0; j < i; j++) {
free_irq(adapter->tx_scrq[j]->irq, adapter->tx_scrq[j]);
irq_dispose_mapping(adapter->rx_scrq[j]->irq);
}
release_sub_crqs_no_irqs(adapter);
return rc;
}
static void init_sub_crqs(struct ibmvnic_adapter *adapter, int retry)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_sub_crq_queue **allqueues;
int registered_queues = 0;
union ibmvnic_crq crq;
int total_queues;
int more = 0;
int i;
if (!retry) {
/* Sub-CRQ entries are 32 byte long */
int entries_page = 4 * PAGE_SIZE / (sizeof(u64) * 4);
if (adapter->min_tx_entries_per_subcrq > entries_page ||
adapter->min_rx_add_entries_per_subcrq > entries_page) {
dev_err(dev, "Fatal, invalid entries per sub-crq\n");
goto allqueues_failed;
}
/* Get the minimum between the queried max and the entries
* that fit in our PAGE_SIZE
*/
adapter->req_tx_entries_per_subcrq =
adapter->max_tx_entries_per_subcrq > entries_page ?
entries_page : adapter->max_tx_entries_per_subcrq;
adapter->req_rx_add_entries_per_subcrq =
adapter->max_rx_add_entries_per_subcrq > entries_page ?
entries_page : adapter->max_rx_add_entries_per_subcrq;
/* Choosing the maximum number of queues supported by firmware*/
adapter->req_tx_queues = adapter->max_tx_queues;
adapter->req_rx_queues = adapter->max_rx_queues;
adapter->req_rx_add_queues = adapter->max_rx_add_queues;
adapter->req_mtu = adapter->max_mtu;
}
total_queues = adapter->req_tx_queues + adapter->req_rx_queues;
allqueues = kcalloc(total_queues, sizeof(*allqueues), GFP_ATOMIC);
if (!allqueues)
goto allqueues_failed;
for (i = 0; i < total_queues; i++) {
allqueues[i] = init_sub_crq_queue(adapter);
if (!allqueues[i]) {
dev_warn(dev, "Couldn't allocate all sub-crqs\n");
break;
}
registered_queues++;
}
/* Make sure we were able to register the minimum number of queues */
if (registered_queues <
adapter->min_tx_queues + adapter->min_rx_queues) {
dev_err(dev, "Fatal: Couldn't init min number of sub-crqs\n");
goto tx_failed;
}
/* Distribute the failed allocated queues*/
for (i = 0; i < total_queues - registered_queues + more ; i++) {
netdev_dbg(adapter->netdev, "Reducing number of queues\n");
switch (i % 3) {
case 0:
if (adapter->req_rx_queues > adapter->min_rx_queues)
adapter->req_rx_queues--;
else
more++;
break;
case 1:
if (adapter->req_tx_queues > adapter->min_tx_queues)
adapter->req_tx_queues--;
else
more++;
break;
}
}
adapter->tx_scrq = kcalloc(adapter->req_tx_queues,
sizeof(*adapter->tx_scrq), GFP_ATOMIC);
if (!adapter->tx_scrq)
goto tx_failed;
for (i = 0; i < adapter->req_tx_queues; i++) {
adapter->tx_scrq[i] = allqueues[i];
adapter->tx_scrq[i]->pool_index = i;
}
adapter->rx_scrq = kcalloc(adapter->req_rx_queues,
sizeof(*adapter->rx_scrq), GFP_ATOMIC);
if (!adapter->rx_scrq)
goto rx_failed;
for (i = 0; i < adapter->req_rx_queues; i++) {
adapter->rx_scrq[i] = allqueues[i + adapter->req_tx_queues];
adapter->rx_scrq[i]->scrq_num = i;
}
memset(&crq, 0, sizeof(crq));
crq.request_capability.first = IBMVNIC_CRQ_CMD;
crq.request_capability.cmd = REQUEST_CAPABILITY;
crq.request_capability.capability = cpu_to_be16(REQ_TX_QUEUES);
crq.request_capability.number = cpu_to_be64(adapter->req_tx_queues);
ibmvnic_send_crq(adapter, &crq);
crq.request_capability.capability = cpu_to_be16(REQ_RX_QUEUES);
crq.request_capability.number = cpu_to_be64(adapter->req_rx_queues);
ibmvnic_send_crq(adapter, &crq);
crq.request_capability.capability = cpu_to_be16(REQ_RX_ADD_QUEUES);
crq.request_capability.number = cpu_to_be64(adapter->req_rx_add_queues);
ibmvnic_send_crq(adapter, &crq);
crq.request_capability.capability =
cpu_to_be16(REQ_TX_ENTRIES_PER_SUBCRQ);
crq.request_capability.number =
cpu_to_be64(adapter->req_tx_entries_per_subcrq);
ibmvnic_send_crq(adapter, &crq);
crq.request_capability.capability =
cpu_to_be16(REQ_RX_ADD_ENTRIES_PER_SUBCRQ);
crq.request_capability.number =
cpu_to_be64(adapter->req_rx_add_entries_per_subcrq);
ibmvnic_send_crq(adapter, &crq);
crq.request_capability.capability = cpu_to_be16(REQ_MTU);
crq.request_capability.number = cpu_to_be64(adapter->req_mtu);
ibmvnic_send_crq(adapter, &crq);
if (adapter->netdev->flags & IFF_PROMISC) {
if (adapter->promisc_supported) {
crq.request_capability.capability =
cpu_to_be16(PROMISC_REQUESTED);
crq.request_capability.number = cpu_to_be64(1);
ibmvnic_send_crq(adapter, &crq);
}
} else {
crq.request_capability.capability =
cpu_to_be16(PROMISC_REQUESTED);
crq.request_capability.number = cpu_to_be64(0);
ibmvnic_send_crq(adapter, &crq);
}
kfree(allqueues);
return;
rx_failed:
kfree(adapter->tx_scrq);
adapter->tx_scrq = NULL;
tx_failed:
for (i = 0; i < registered_queues; i++)
release_sub_crq_queue(adapter, allqueues[i]);
kfree(allqueues);
allqueues_failed:
ibmvnic_remove(adapter->vdev);
}
static int pending_scrq(struct ibmvnic_adapter *adapter,
struct ibmvnic_sub_crq_queue *scrq)
{
union sub_crq *entry = &scrq->msgs[scrq->cur];
if (entry->generic.first & IBMVNIC_CRQ_CMD_RSP || adapter->closing)
return 1;
else
return 0;
}
static union sub_crq *ibmvnic_next_scrq(struct ibmvnic_adapter *adapter,
struct ibmvnic_sub_crq_queue *scrq)
{
union sub_crq *entry;
unsigned long flags;
spin_lock_irqsave(&scrq->lock, flags);
entry = &scrq->msgs[scrq->cur];
if (entry->generic.first & IBMVNIC_CRQ_CMD_RSP) {
if (++scrq->cur == scrq->size)
scrq->cur = 0;
} else {
entry = NULL;
}
spin_unlock_irqrestore(&scrq->lock, flags);
return entry;
}
static union ibmvnic_crq *ibmvnic_next_crq(struct ibmvnic_adapter *adapter)
{
struct ibmvnic_crq_queue *queue = &adapter->crq;
union ibmvnic_crq *crq;
crq = &queue->msgs[queue->cur];
if (crq->generic.first & IBMVNIC_CRQ_CMD_RSP) {
if (++queue->cur == queue->size)
queue->cur = 0;
} else {
crq = NULL;
}
return crq;
}
static int send_subcrq(struct ibmvnic_adapter *adapter, u64 remote_handle,
union sub_crq *sub_crq)
{
unsigned int ua = adapter->vdev->unit_address;
struct device *dev = &adapter->vdev->dev;
u64 *u64_crq = (u64 *)sub_crq;
int rc;
netdev_dbg(adapter->netdev,
"Sending sCRQ %016lx: %016lx %016lx %016lx %016lx\n",
(unsigned long int)cpu_to_be64(remote_handle),
(unsigned long int)cpu_to_be64(u64_crq[0]),
(unsigned long int)cpu_to_be64(u64_crq[1]),
(unsigned long int)cpu_to_be64(u64_crq[2]),
(unsigned long int)cpu_to_be64(u64_crq[3]));
/* Make sure the hypervisor sees the complete request */
mb();
rc = plpar_hcall_norets(H_SEND_SUB_CRQ, ua,
cpu_to_be64(remote_handle),
cpu_to_be64(u64_crq[0]),
cpu_to_be64(u64_crq[1]),
cpu_to_be64(u64_crq[2]),
cpu_to_be64(u64_crq[3]));
if (rc) {
if (rc == H_CLOSED)
dev_warn(dev, "CRQ Queue closed\n");
dev_err(dev, "Send error (rc=%d)\n", rc);
}
return rc;
}
static int send_subcrq_indirect(struct ibmvnic_adapter *adapter,
u64 remote_handle, u64 ioba, u64 num_entries)
{
unsigned int ua = adapter->vdev->unit_address;
struct device *dev = &adapter->vdev->dev;
int rc;
/* Make sure the hypervisor sees the complete request */
mb();
rc = plpar_hcall_norets(H_SEND_SUB_CRQ_INDIRECT, ua,
cpu_to_be64(remote_handle),
ioba, num_entries);
if (rc) {
if (rc == H_CLOSED)
dev_warn(dev, "CRQ Queue closed\n");
dev_err(dev, "Send (indirect) error (rc=%d)\n", rc);
}
return rc;
}
static int ibmvnic_send_crq(struct ibmvnic_adapter *adapter,
union ibmvnic_crq *crq)
{
unsigned int ua = adapter->vdev->unit_address;
struct device *dev = &adapter->vdev->dev;
u64 *u64_crq = (u64 *)crq;
int rc;
netdev_dbg(adapter->netdev, "Sending CRQ: %016lx %016lx\n",
(unsigned long int)cpu_to_be64(u64_crq[0]),
(unsigned long int)cpu_to_be64(u64_crq[1]));
/* Make sure the hypervisor sees the complete request */
mb();
rc = plpar_hcall_norets(H_SEND_CRQ, ua,
cpu_to_be64(u64_crq[0]),
cpu_to_be64(u64_crq[1]));
if (rc) {
if (rc == H_CLOSED)
dev_warn(dev, "CRQ Queue closed\n");
dev_warn(dev, "Send error (rc=%d)\n", rc);
}
return rc;
}
static int ibmvnic_send_crq_init(struct ibmvnic_adapter *adapter)
{
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.generic.first = IBMVNIC_CRQ_INIT_CMD;
crq.generic.cmd = IBMVNIC_CRQ_INIT;
netdev_dbg(adapter->netdev, "Sending CRQ init\n");
return ibmvnic_send_crq(adapter, &crq);
}
static int ibmvnic_send_crq_init_complete(struct ibmvnic_adapter *adapter)
{
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.generic.first = IBMVNIC_CRQ_INIT_CMD;
crq.generic.cmd = IBMVNIC_CRQ_INIT_COMPLETE;
netdev_dbg(adapter->netdev, "Sending CRQ init complete\n");
return ibmvnic_send_crq(adapter, &crq);
}
static int send_version_xchg(struct ibmvnic_adapter *adapter)
{
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.version_exchange.first = IBMVNIC_CRQ_CMD;
crq.version_exchange.cmd = VERSION_EXCHANGE;
crq.version_exchange.version = cpu_to_be16(ibmvnic_version);
return ibmvnic_send_crq(adapter, &crq);
}
static void send_login(struct ibmvnic_adapter *adapter)
{
struct ibmvnic_login_rsp_buffer *login_rsp_buffer;
struct ibmvnic_login_buffer *login_buffer;
struct ibmvnic_inflight_cmd *inflight_cmd;
struct device *dev = &adapter->vdev->dev;
dma_addr_t rsp_buffer_token;
dma_addr_t buffer_token;
size_t rsp_buffer_size;
union ibmvnic_crq crq;
unsigned long flags;
size_t buffer_size;
__be64 *tx_list_p;
__be64 *rx_list_p;
int i;
buffer_size =
sizeof(struct ibmvnic_login_buffer) +
sizeof(u64) * (adapter->req_tx_queues + adapter->req_rx_queues);
login_buffer = kmalloc(buffer_size, GFP_ATOMIC);
if (!login_buffer)
goto buf_alloc_failed;
buffer_token = dma_map_single(dev, login_buffer, buffer_size,
DMA_TO_DEVICE);
if (dma_mapping_error(dev, buffer_token)) {
dev_err(dev, "Couldn't map login buffer\n");
goto buf_map_failed;
}
rsp_buffer_size = sizeof(struct ibmvnic_login_rsp_buffer) +
sizeof(u64) * adapter->req_tx_queues +
sizeof(u64) * adapter->req_rx_queues +
sizeof(u64) * adapter->req_rx_queues +
sizeof(u8) * IBMVNIC_TX_DESC_VERSIONS;
login_rsp_buffer = kmalloc(rsp_buffer_size, GFP_ATOMIC);
if (!login_rsp_buffer)
goto buf_rsp_alloc_failed;
rsp_buffer_token = dma_map_single(dev, login_rsp_buffer,
rsp_buffer_size, DMA_FROM_DEVICE);
if (dma_mapping_error(dev, rsp_buffer_token)) {
dev_err(dev, "Couldn't map login rsp buffer\n");
goto buf_rsp_map_failed;
}
inflight_cmd = kmalloc(sizeof(*inflight_cmd), GFP_ATOMIC);
if (!inflight_cmd) {
dev_err(dev, "Couldn't allocate inflight_cmd\n");
goto inflight_alloc_failed;
}
adapter->login_buf = login_buffer;
adapter->login_buf_token = buffer_token;
adapter->login_buf_sz = buffer_size;
adapter->login_rsp_buf = login_rsp_buffer;
adapter->login_rsp_buf_token = rsp_buffer_token;
adapter->login_rsp_buf_sz = rsp_buffer_size;
login_buffer->len = cpu_to_be32(buffer_size);
login_buffer->version = cpu_to_be32(INITIAL_VERSION_LB);
login_buffer->num_txcomp_subcrqs = cpu_to_be32(adapter->req_tx_queues);
login_buffer->off_txcomp_subcrqs =
cpu_to_be32(sizeof(struct ibmvnic_login_buffer));
login_buffer->num_rxcomp_subcrqs = cpu_to_be32(adapter->req_rx_queues);
login_buffer->off_rxcomp_subcrqs =
cpu_to_be32(sizeof(struct ibmvnic_login_buffer) +
sizeof(u64) * adapter->req_tx_queues);
login_buffer->login_rsp_ioba = cpu_to_be32(rsp_buffer_token);
login_buffer->login_rsp_len = cpu_to_be32(rsp_buffer_size);
tx_list_p = (__be64 *)((char *)login_buffer +
sizeof(struct ibmvnic_login_buffer));
rx_list_p = (__be64 *)((char *)login_buffer +
sizeof(struct ibmvnic_login_buffer) +
sizeof(u64) * adapter->req_tx_queues);
for (i = 0; i < adapter->req_tx_queues; i++) {
if (adapter->tx_scrq[i]) {
tx_list_p[i] = cpu_to_be64(adapter->tx_scrq[i]->
crq_num);
}
}
for (i = 0; i < adapter->req_rx_queues; i++) {
if (adapter->rx_scrq[i]) {
rx_list_p[i] = cpu_to_be64(adapter->rx_scrq[i]->
crq_num);
}
}
netdev_dbg(adapter->netdev, "Login Buffer:\n");
for (i = 0; i < (adapter->login_buf_sz - 1) / 8 + 1; i++) {
netdev_dbg(adapter->netdev, "%016lx\n",
((unsigned long int *)(adapter->login_buf))[i]);
}
memset(&crq, 0, sizeof(crq));
crq.login.first = IBMVNIC_CRQ_CMD;
crq.login.cmd = LOGIN;
crq.login.ioba = cpu_to_be32(buffer_token);
crq.login.len = cpu_to_be32(buffer_size);
memcpy(&inflight_cmd->crq, &crq, sizeof(crq));
spin_lock_irqsave(&adapter->inflight_lock, flags);
list_add_tail(&inflight_cmd->list, &adapter->inflight);
spin_unlock_irqrestore(&adapter->inflight_lock, flags);
ibmvnic_send_crq(adapter, &crq);
return;
inflight_alloc_failed:
dma_unmap_single(dev, rsp_buffer_token, rsp_buffer_size,
DMA_FROM_DEVICE);
buf_rsp_map_failed:
kfree(login_rsp_buffer);
buf_rsp_alloc_failed:
dma_unmap_single(dev, buffer_token, buffer_size, DMA_TO_DEVICE);
buf_map_failed:
kfree(login_buffer);
buf_alloc_failed:
return;
}
static void send_request_map(struct ibmvnic_adapter *adapter, dma_addr_t addr,
u32 len, u8 map_id)
{
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.request_map.first = IBMVNIC_CRQ_CMD;
crq.request_map.cmd = REQUEST_MAP;
crq.request_map.map_id = map_id;
crq.request_map.ioba = cpu_to_be32(addr);
crq.request_map.len = cpu_to_be32(len);
ibmvnic_send_crq(adapter, &crq);
}
static void send_request_unmap(struct ibmvnic_adapter *adapter, u8 map_id)
{
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.request_unmap.first = IBMVNIC_CRQ_CMD;
crq.request_unmap.cmd = REQUEST_UNMAP;
crq.request_unmap.map_id = map_id;
ibmvnic_send_crq(adapter, &crq);
}
static void send_map_query(struct ibmvnic_adapter *adapter)
{
union ibmvnic_crq crq;
memset(&crq, 0, sizeof(crq));
crq.query_map.first = IBMVNIC_CRQ_CMD;
crq.query_map.cmd = QUERY_MAP;
ibmvnic_send_crq(adapter, &crq);
}
/* Send a series of CRQs requesting various capabilities of the VNIC server */
static void send_cap_queries(struct ibmvnic_adapter *adapter)
{
union ibmvnic_crq crq;
atomic_set(&adapter->running_cap_queries, 0);
memset(&crq, 0, sizeof(crq));
crq.query_capability.first = IBMVNIC_CRQ_CMD;
crq.query_capability.cmd = QUERY_CAPABILITY;
crq.query_capability.capability = cpu_to_be16(MIN_TX_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MIN_RX_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MIN_RX_ADD_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MAX_TX_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MAX_RX_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MAX_RX_ADD_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(MIN_TX_ENTRIES_PER_SUBCRQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(MIN_RX_ADD_ENTRIES_PER_SUBCRQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(MAX_TX_ENTRIES_PER_SUBCRQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(MAX_RX_ADD_ENTRIES_PER_SUBCRQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(TCP_IP_OFFLOAD);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(PROMISC_SUPPORTED);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MIN_MTU);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MAX_MTU);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MAX_MULTICAST_FILTERS);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(VLAN_HEADER_INSERTION);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(MAX_TX_SG_ENTRIES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(RX_SG_SUPPORTED);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(OPT_TX_COMP_SUB_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(OPT_RX_COMP_QUEUES);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(OPT_RX_BUFADD_Q_PER_RX_COMP_Q);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(OPT_TX_ENTRIES_PER_SUBCRQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability =
cpu_to_be16(OPT_RXBA_ENTRIES_PER_SUBCRQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
crq.query_capability.capability = cpu_to_be16(TX_RX_DESC_REQ);
atomic_inc(&adapter->running_cap_queries);
ibmvnic_send_crq(adapter, &crq);
}
static void handle_query_ip_offload_rsp(struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_query_ip_offload_buffer *buf = &adapter->ip_offload_buf;
union ibmvnic_crq crq;
int i;
dma_unmap_single(dev, adapter->ip_offload_tok,
sizeof(adapter->ip_offload_buf), DMA_FROM_DEVICE);
netdev_dbg(adapter->netdev, "Query IP Offload Buffer:\n");
for (i = 0; i < (sizeof(adapter->ip_offload_buf) - 1) / 8 + 1; i++)
netdev_dbg(adapter->netdev, "%016lx\n",
((unsigned long int *)(buf))[i]);
netdev_dbg(adapter->netdev, "ipv4_chksum = %d\n", buf->ipv4_chksum);
netdev_dbg(adapter->netdev, "ipv6_chksum = %d\n", buf->ipv6_chksum);
netdev_dbg(adapter->netdev, "tcp_ipv4_chksum = %d\n",
buf->tcp_ipv4_chksum);
netdev_dbg(adapter->netdev, "tcp_ipv6_chksum = %d\n",
buf->tcp_ipv6_chksum);
netdev_dbg(adapter->netdev, "udp_ipv4_chksum = %d\n",
buf->udp_ipv4_chksum);
netdev_dbg(adapter->netdev, "udp_ipv6_chksum = %d\n",
buf->udp_ipv6_chksum);
netdev_dbg(adapter->netdev, "large_tx_ipv4 = %d\n",
buf->large_tx_ipv4);
netdev_dbg(adapter->netdev, "large_tx_ipv6 = %d\n",
buf->large_tx_ipv6);
netdev_dbg(adapter->netdev, "large_rx_ipv4 = %d\n",
buf->large_rx_ipv4);
netdev_dbg(adapter->netdev, "large_rx_ipv6 = %d\n",
buf->large_rx_ipv6);
netdev_dbg(adapter->netdev, "max_ipv4_hdr_sz = %d\n",
buf->max_ipv4_header_size);
netdev_dbg(adapter->netdev, "max_ipv6_hdr_sz = %d\n",
buf->max_ipv6_header_size);
netdev_dbg(adapter->netdev, "max_tcp_hdr_size = %d\n",
buf->max_tcp_header_size);
netdev_dbg(adapter->netdev, "max_udp_hdr_size = %d\n",
buf->max_udp_header_size);
netdev_dbg(adapter->netdev, "max_large_tx_size = %d\n",
buf->max_large_tx_size);
netdev_dbg(adapter->netdev, "max_large_rx_size = %d\n",
buf->max_large_rx_size);
netdev_dbg(adapter->netdev, "ipv6_ext_hdr = %d\n",
buf->ipv6_extension_header);
netdev_dbg(adapter->netdev, "tcp_pseudosum_req = %d\n",
buf->tcp_pseudosum_req);
netdev_dbg(adapter->netdev, "num_ipv6_ext_hd = %d\n",
buf->num_ipv6_ext_headers);
netdev_dbg(adapter->netdev, "off_ipv6_ext_hd = %d\n",
buf->off_ipv6_ext_headers);
adapter->ip_offload_ctrl_tok =
dma_map_single(dev, &adapter->ip_offload_ctrl,
sizeof(adapter->ip_offload_ctrl), DMA_TO_DEVICE);
if (dma_mapping_error(dev, adapter->ip_offload_ctrl_tok)) {
dev_err(dev, "Couldn't map ip offload control buffer\n");
return;
}
adapter->ip_offload_ctrl.version = cpu_to_be32(INITIAL_VERSION_IOB);
adapter->ip_offload_ctrl.tcp_ipv4_chksum = buf->tcp_ipv4_chksum;
adapter->ip_offload_ctrl.udp_ipv4_chksum = buf->udp_ipv4_chksum;
adapter->ip_offload_ctrl.tcp_ipv6_chksum = buf->tcp_ipv6_chksum;
adapter->ip_offload_ctrl.udp_ipv6_chksum = buf->udp_ipv6_chksum;
/* large_tx/rx disabled for now, additional features needed */
adapter->ip_offload_ctrl.large_tx_ipv4 = 0;
adapter->ip_offload_ctrl.large_tx_ipv6 = 0;
adapter->ip_offload_ctrl.large_rx_ipv4 = 0;
adapter->ip_offload_ctrl.large_rx_ipv6 = 0;
adapter->netdev->features = NETIF_F_GSO;
if (buf->tcp_ipv4_chksum || buf->udp_ipv4_chksum)
adapter->netdev->features |= NETIF_F_IP_CSUM;
if (buf->tcp_ipv6_chksum || buf->udp_ipv6_chksum)
adapter->netdev->features |= NETIF_F_IPV6_CSUM;
if ((adapter->netdev->features &
(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)))
adapter->netdev->features |= NETIF_F_RXCSUM;
memset(&crq, 0, sizeof(crq));
crq.control_ip_offload.first = IBMVNIC_CRQ_CMD;
crq.control_ip_offload.cmd = CONTROL_IP_OFFLOAD;
crq.control_ip_offload.len =
cpu_to_be32(sizeof(adapter->ip_offload_ctrl));
crq.control_ip_offload.ioba = cpu_to_be32(adapter->ip_offload_ctrl_tok);
ibmvnic_send_crq(adapter, &crq);
}
static void handle_error_info_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_error_buff *error_buff, *tmp;
unsigned long flags;
bool found = false;
int i;
if (!crq->request_error_rsp.rc.code) {
dev_info(dev, "Request Error Rsp returned with rc=%x\n",
crq->request_error_rsp.rc.code);
return;
}
spin_lock_irqsave(&adapter->error_list_lock, flags);
list_for_each_entry_safe(error_buff, tmp, &adapter->errors, list)
if (error_buff->error_id == crq->request_error_rsp.error_id) {
found = true;
list_del(&error_buff->list);
break;
}
spin_unlock_irqrestore(&adapter->error_list_lock, flags);
if (!found) {
dev_err(dev, "Couldn't find error id %x\n",
crq->request_error_rsp.error_id);
return;
}
dev_err(dev, "Detailed info for error id %x:",
crq->request_error_rsp.error_id);
for (i = 0; i < error_buff->len; i++) {
pr_cont("%02x", (int)error_buff->buff[i]);
if (i % 8 == 7)
pr_cont(" ");
}
pr_cont("\n");
dma_unmap_single(dev, error_buff->dma, error_buff->len,
DMA_FROM_DEVICE);
kfree(error_buff->buff);
kfree(error_buff);
}
static void handle_dump_size_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
int len = be32_to_cpu(crq->request_dump_size_rsp.len);
struct ibmvnic_inflight_cmd *inflight_cmd;
struct device *dev = &adapter->vdev->dev;
union ibmvnic_crq newcrq;
unsigned long flags;
/* allocate and map buffer */
adapter->dump_data = kmalloc(len, GFP_KERNEL);
if (!adapter->dump_data) {
complete(&adapter->fw_done);
return;
}
adapter->dump_data_token = dma_map_single(dev, adapter->dump_data, len,
DMA_FROM_DEVICE);
if (dma_mapping_error(dev, adapter->dump_data_token)) {
if (!firmware_has_feature(FW_FEATURE_CMO))
dev_err(dev, "Couldn't map dump data\n");
kfree(adapter->dump_data);
complete(&adapter->fw_done);
return;
}
inflight_cmd = kmalloc(sizeof(*inflight_cmd), GFP_ATOMIC);
if (!inflight_cmd) {
dma_unmap_single(dev, adapter->dump_data_token, len,
DMA_FROM_DEVICE);
kfree(adapter->dump_data);
complete(&adapter->fw_done);
return;
}
memset(&newcrq, 0, sizeof(newcrq));
newcrq.request_dump.first = IBMVNIC_CRQ_CMD;
newcrq.request_dump.cmd = REQUEST_DUMP;
newcrq.request_dump.ioba = cpu_to_be32(adapter->dump_data_token);
newcrq.request_dump.len = cpu_to_be32(adapter->dump_data_size);
memcpy(&inflight_cmd->crq, &newcrq, sizeof(newcrq));
spin_lock_irqsave(&adapter->inflight_lock, flags);
list_add_tail(&inflight_cmd->list, &adapter->inflight);
spin_unlock_irqrestore(&adapter->inflight_lock, flags);
ibmvnic_send_crq(adapter, &newcrq);
}
static void handle_error_indication(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
int detail_len = be32_to_cpu(crq->error_indication.detail_error_sz);
struct ibmvnic_inflight_cmd *inflight_cmd;
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_error_buff *error_buff;
union ibmvnic_crq new_crq;
unsigned long flags;
dev_err(dev, "Firmware reports %serror id %x, cause %d\n",
crq->error_indication.
flags & IBMVNIC_FATAL_ERROR ? "FATAL " : "",
crq->error_indication.error_id,
crq->error_indication.error_cause);
error_buff = kmalloc(sizeof(*error_buff), GFP_ATOMIC);
if (!error_buff)
return;
error_buff->buff = kmalloc(detail_len, GFP_ATOMIC);
if (!error_buff->buff) {
kfree(error_buff);
return;
}
error_buff->dma = dma_map_single(dev, error_buff->buff, detail_len,
DMA_FROM_DEVICE);
if (dma_mapping_error(dev, error_buff->dma)) {
if (!firmware_has_feature(FW_FEATURE_CMO))
dev_err(dev, "Couldn't map error buffer\n");
kfree(error_buff->buff);
kfree(error_buff);
return;
}
inflight_cmd = kmalloc(sizeof(*inflight_cmd), GFP_ATOMIC);
if (!inflight_cmd) {
dma_unmap_single(dev, error_buff->dma, detail_len,
DMA_FROM_DEVICE);
kfree(error_buff->buff);
kfree(error_buff);
return;
}
error_buff->len = detail_len;
error_buff->error_id = crq->error_indication.error_id;
spin_lock_irqsave(&adapter->error_list_lock, flags);
list_add_tail(&error_buff->list, &adapter->errors);
spin_unlock_irqrestore(&adapter->error_list_lock, flags);
memset(&new_crq, 0, sizeof(new_crq));
new_crq.request_error_info.first = IBMVNIC_CRQ_CMD;
new_crq.request_error_info.cmd = REQUEST_ERROR_INFO;
new_crq.request_error_info.ioba = cpu_to_be32(error_buff->dma);
new_crq.request_error_info.len = cpu_to_be32(detail_len);
new_crq.request_error_info.error_id = crq->error_indication.error_id;
memcpy(&inflight_cmd->crq, &crq, sizeof(crq));
spin_lock_irqsave(&adapter->inflight_lock, flags);
list_add_tail(&inflight_cmd->list, &adapter->inflight);
spin_unlock_irqrestore(&adapter->inflight_lock, flags);
ibmvnic_send_crq(adapter, &new_crq);
}
static void handle_change_mac_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
struct device *dev = &adapter->vdev->dev;
long rc;
rc = crq->change_mac_addr_rsp.rc.code;
if (rc) {
dev_err(dev, "Error %ld in CHANGE_MAC_ADDR_RSP\n", rc);
return;
}
memcpy(netdev->dev_addr, &crq->change_mac_addr_rsp.mac_addr[0],
ETH_ALEN);
}
static void handle_request_cap_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
u64 *req_value;
char *name;
switch (be16_to_cpu(crq->request_capability_rsp.capability)) {
case REQ_TX_QUEUES:
req_value = &adapter->req_tx_queues;
name = "tx";
break;
case REQ_RX_QUEUES:
req_value = &adapter->req_rx_queues;
name = "rx";
break;
case REQ_RX_ADD_QUEUES:
req_value = &adapter->req_rx_add_queues;
name = "rx_add";
break;
case REQ_TX_ENTRIES_PER_SUBCRQ:
req_value = &adapter->req_tx_entries_per_subcrq;
name = "tx_entries_per_subcrq";
break;
case REQ_RX_ADD_ENTRIES_PER_SUBCRQ:
req_value = &adapter->req_rx_add_entries_per_subcrq;
name = "rx_add_entries_per_subcrq";
break;
case REQ_MTU:
req_value = &adapter->req_mtu;
name = "mtu";
break;
case PROMISC_REQUESTED:
req_value = &adapter->promisc;
name = "promisc";
break;
default:
dev_err(dev, "Got invalid cap request rsp %d\n",
crq->request_capability.capability);
return;
}
switch (crq->request_capability_rsp.rc.code) {
case SUCCESS:
break;
case PARTIALSUCCESS:
dev_info(dev, "req=%lld, rsp=%ld in %s queue, retrying.\n",
*req_value,
(long int)be32_to_cpu(crq->request_capability_rsp.
number), name);
release_sub_crqs_no_irqs(adapter);
*req_value = be32_to_cpu(crq->request_capability_rsp.number);
init_sub_crqs(adapter, 1);
return;
default:
dev_err(dev, "Error %d in request cap rsp\n",
crq->request_capability_rsp.rc.code);
return;
}
/* Done receiving requested capabilities, query IP offload support */
if (++adapter->requested_caps == 7) {
union ibmvnic_crq newcrq;
int buf_sz = sizeof(struct ibmvnic_query_ip_offload_buffer);
struct ibmvnic_query_ip_offload_buffer *ip_offload_buf =
&adapter->ip_offload_buf;
adapter->ip_offload_tok = dma_map_single(dev, ip_offload_buf,
buf_sz,
DMA_FROM_DEVICE);
if (dma_mapping_error(dev, adapter->ip_offload_tok)) {
if (!firmware_has_feature(FW_FEATURE_CMO))
dev_err(dev, "Couldn't map offload buffer\n");
return;
}
memset(&newcrq, 0, sizeof(newcrq));
newcrq.query_ip_offload.first = IBMVNIC_CRQ_CMD;
newcrq.query_ip_offload.cmd = QUERY_IP_OFFLOAD;
newcrq.query_ip_offload.len = cpu_to_be32(buf_sz);
newcrq.query_ip_offload.ioba =
cpu_to_be32(adapter->ip_offload_tok);
ibmvnic_send_crq(adapter, &newcrq);
}
}
static int handle_login_rsp(union ibmvnic_crq *login_rsp_crq,
struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
struct ibmvnic_login_rsp_buffer *login_rsp = adapter->login_rsp_buf;
struct ibmvnic_login_buffer *login = adapter->login_buf;
union ibmvnic_crq crq;
int i;
dma_unmap_single(dev, adapter->login_buf_token, adapter->login_buf_sz,
DMA_BIDIRECTIONAL);
dma_unmap_single(dev, adapter->login_rsp_buf_token,
adapter->login_rsp_buf_sz, DMA_BIDIRECTIONAL);
/* If the number of queues requested can't be allocated by the
* server, the login response will return with code 1. We will need
* to resend the login buffer with fewer queues requested.
*/
if (login_rsp_crq->generic.rc.code) {
adapter->renegotiate = true;
complete(&adapter->init_done);
return 0;
}
netdev_dbg(adapter->netdev, "Login Response Buffer:\n");
for (i = 0; i < (adapter->login_rsp_buf_sz - 1) / 8 + 1; i++) {
netdev_dbg(adapter->netdev, "%016lx\n",
((unsigned long int *)(adapter->login_rsp_buf))[i]);
}
/* Sanity checks */
if (login->num_txcomp_subcrqs != login_rsp->num_txsubm_subcrqs ||
(be32_to_cpu(login->num_rxcomp_subcrqs) *
adapter->req_rx_add_queues !=
be32_to_cpu(login_rsp->num_rxadd_subcrqs))) {
dev_err(dev, "FATAL: Inconsistent login and login rsp\n");
ibmvnic_remove(adapter->vdev);
return -EIO;
}
complete(&adapter->init_done);
memset(&crq, 0, sizeof(crq));
crq.request_ras_comp_num.first = IBMVNIC_CRQ_CMD;
crq.request_ras_comp_num.cmd = REQUEST_RAS_COMP_NUM;
ibmvnic_send_crq(adapter, &crq);
return 0;
}
static void handle_request_map_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
u8 map_id = crq->request_map_rsp.map_id;
int tx_subcrqs;
int rx_subcrqs;
long rc;
int i;
tx_subcrqs = be32_to_cpu(adapter->login_rsp_buf->num_txsubm_subcrqs);
rx_subcrqs = be32_to_cpu(adapter->login_rsp_buf->num_rxadd_subcrqs);
rc = crq->request_map_rsp.rc.code;
if (rc) {
dev_err(dev, "Error %ld in REQUEST_MAP_RSP\n", rc);
adapter->map_id--;
/* need to find and zero tx/rx_pool map_id */
for (i = 0; i < tx_subcrqs; i++) {
if (adapter->tx_pool[i].long_term_buff.map_id == map_id)
adapter->tx_pool[i].long_term_buff.map_id = 0;
}
for (i = 0; i < rx_subcrqs; i++) {
if (adapter->rx_pool[i].long_term_buff.map_id == map_id)
adapter->rx_pool[i].long_term_buff.map_id = 0;
}
}
complete(&adapter->fw_done);
}
static void handle_request_unmap_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct device *dev = &adapter->vdev->dev;
long rc;
rc = crq->request_unmap_rsp.rc.code;
if (rc)
dev_err(dev, "Error %ld in REQUEST_UNMAP_RSP\n", rc);
}
static void handle_query_map_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
struct device *dev = &adapter->vdev->dev;
long rc;
rc = crq->query_map_rsp.rc.code;
if (rc) {
dev_err(dev, "Error %ld in QUERY_MAP_RSP\n", rc);
return;
}
netdev_dbg(netdev, "page_size = %d\ntot_pages = %d\nfree_pages = %d\n",
crq->query_map_rsp.page_size, crq->query_map_rsp.tot_pages,
crq->query_map_rsp.free_pages);
}
static void handle_query_cap_rsp(union ibmvnic_crq *crq,
struct ibmvnic_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
struct device *dev = &adapter->vdev->dev;
long rc;
atomic_dec(&adapter->running_cap_queries);
netdev_dbg(netdev, "Outstanding queries: %d\n",
atomic_read(&adapter->running_cap_queries));
rc = crq->query_capability.rc.code;
if (rc) {
dev_err(dev, "Error %ld in QUERY_CAP_RSP\n", rc);
goto out;
}
switch (be16_to_cpu(crq->query_capability.capability)) {
case MIN_TX_QUEUES:
adapter->min_tx_queues =
be64_to_cpu(crq->query_capability.number);
netdev_dbg(netdev, "min_tx_queues = %lld\n",
adapter->min_tx_queues);
break;
case MIN_RX_QUEUES:
adapter->min_rx_queues =
be64_to_cpu(crq->query_capability.number);
netdev_dbg(netdev, "min_rx_queues = %lld\n",
adapter->min_rx_queues);
break;
case MIN_RX_ADD_QUEUES:
adapter->min_rx_add_queues =
be64_to_cpu(crq->query_capability.number);
netdev_dbg(netdev, "min_rx_add_queues = %lld\n",