diff options
Diffstat (limited to 'src/dpdk/lib/librte_eal/linuxapp')
73 files changed, 954 insertions, 6461 deletions
diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal.c index 3fb2188f..bf6b818c 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal.c +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/eal.c @@ -69,7 +69,9 @@ #include <rte_string_fns.h> #include <rte_cpuflags.h> #include <rte_interrupts.h> +#include <rte_bus.h> #include <rte_pci.h> +#include <rte_dev.h> #include <rte_devargs.h> #include <rte_common.h> #include <rte_version.h> @@ -238,7 +240,8 @@ rte_eal_config_attach(void) mem_config = (struct rte_mem_config *) mmap(NULL, sizeof(*mem_config), PROT_READ, MAP_SHARED, mem_cfg_fd, 0); if (mem_config == MAP_FAILED) - rte_panic("Cannot mmap memory for rte_config\n"); + rte_panic("Cannot mmap memory for rte_config! error %i (%s)\n", + errno, strerror(errno)); rte_config.mem_config = mem_config; } @@ -263,9 +266,17 @@ rte_eal_config_reattach(void) mem_config = (struct rte_mem_config *) mmap(rte_mem_cfg_addr, sizeof(*mem_config), PROT_READ | PROT_WRITE, MAP_SHARED, mem_cfg_fd, 0); + if (mem_config == MAP_FAILED || mem_config != rte_mem_cfg_addr) { + if (mem_config != MAP_FAILED) + /* errno is stale, don't use */ + rte_panic("Cannot mmap memory for rte_config at [%p], got [%p]" + " - please use '--base-virtaddr' option\n", + rte_mem_cfg_addr, mem_config); + else + rte_panic("Cannot mmap memory for rte_config! error %i (%s)\n", + errno, strerror(errno)); + } close(mem_cfg_fd); - if (mem_config == MAP_FAILED || mem_config != rte_mem_cfg_addr) - rte_panic("Cannot mmap memory for rte_config\n"); rte_config.mem_config = mem_config; } @@ -740,6 +751,9 @@ rte_eal_init(int argc, char **argv) char cpuset[RTE_CPU_AFFINITY_STR_LEN]; char thread_name[RTE_MAX_THREAD_NAME_LEN]; + /* checks if the machine is adequate */ + rte_cpu_check_supported(); + if (!rte_atomic32_test_and_set(&run_once)) return -1; @@ -748,9 +762,6 @@ rte_eal_init(int argc, char **argv) thread_id = pthread_self(); - if (rte_eal_log_early_init() < 0) - rte_panic("Cannot init early logs\n"); - eal_log_level_parse(argc, argv); /* set log level as early as possible */ @@ -789,6 +800,9 @@ rte_eal_init(int argc, char **argv) rte_config_init(); + if (rte_eal_log_init(logid, internal_config.syslog_facility) < 0) + rte_panic("Cannot init logs\n"); + if (rte_eal_pci_init() < 0) rte_panic("Cannot init PCI\n"); @@ -797,11 +811,6 @@ rte_eal_init(int argc, char **argv) rte_panic("Cannot init VFIO\n"); #endif -#ifdef RTE_LIBRTE_IVSHMEM - if (rte_eal_ivshmem_init() < 0) - rte_panic("Cannot init IVSHMEM\n"); -#endif - if (rte_eal_memory_init() < 0) rte_panic("Cannot init memory\n"); @@ -814,14 +823,6 @@ rte_eal_init(int argc, char **argv) if (rte_eal_tailqs_init() < 0) rte_panic("Cannot init tail queues for objects\n"); -#ifdef RTE_LIBRTE_IVSHMEM - if (rte_eal_ivshmem_obj_init() < 0) - rte_panic("Cannot init IVSHMEM objects\n"); -#endif - - if (rte_eal_log_init(logid, internal_config.syslog_facility) < 0) - rte_panic("Cannot init logs\n"); - if (rte_eal_alarm_init() < 0) rte_panic("Cannot init interrupt-handling thread\n"); @@ -841,12 +842,12 @@ rte_eal_init(int argc, char **argv) rte_config.master_lcore, (int)thread_id, cpuset, ret == 0 ? "" : "..."); - if (rte_eal_dev_init() < 0) - rte_panic("Cannot init pmd devices\n"); - if (rte_eal_intr_init() < 0) rte_panic("Cannot init interrupt-handling thread\n"); + if (rte_bus_scan()) + rte_panic("Cannot scan the buses for devices\n"); + RTE_LCORE_FOREACH_SLAVE(i) { /* @@ -883,10 +884,17 @@ rte_eal_init(int argc, char **argv) rte_eal_mp_remote_launch(sync_func, NULL, SKIP_MASTER); rte_eal_mp_wait_lcore(); + /* Probe all the buses and devices/drivers on them */ + if (rte_bus_probe()) + rte_panic("Cannot probe devices\n"); + /* Probe & Initialize PCI devices */ if (rte_eal_pci_probe()) rte_panic("Cannot probe PCI\n"); + if (rte_eal_dev_init() < 0) + rte_panic("Cannot init pmd devices\n"); + rte_eal_mcfg_complete(); return fctret; diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_interrupts.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_interrupts.c index 54ab6253..b5b3f2bd 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_interrupts.c +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_interrupts.c @@ -73,9 +73,6 @@ static RTE_DEFINE_PER_LCORE(int, _epfd) = -1; /**< epoll fd per thread */ -// TREX_PATCH -int eal_err_read_from_file_is_error = 1; - /** * union for pipe fds. */ @@ -139,7 +136,7 @@ static pthread_t intr_thread; /* enable legacy (INTx) interrupts */ static int -vfio_enable_intx(struct rte_intr_handle *intr_handle) { +vfio_enable_intx(const struct rte_intr_handle *intr_handle) { struct vfio_irq_set *irq_set; char irq_set_buf[IRQ_SET_BUF_LEN]; int len, ret; @@ -186,7 +183,7 @@ vfio_enable_intx(struct rte_intr_handle *intr_handle) { /* disable legacy (INTx) interrupts */ static int -vfio_disable_intx(struct rte_intr_handle *intr_handle) { +vfio_disable_intx(const struct rte_intr_handle *intr_handle) { struct vfio_irq_set *irq_set; char irq_set_buf[IRQ_SET_BUF_LEN]; int len, ret; @@ -229,7 +226,7 @@ vfio_disable_intx(struct rte_intr_handle *intr_handle) { /* enable MSI interrupts */ static int -vfio_enable_msi(struct rte_intr_handle *intr_handle) { +vfio_enable_msi(const struct rte_intr_handle *intr_handle) { int len, ret; char irq_set_buf[IRQ_SET_BUF_LEN]; struct vfio_irq_set *irq_set; @@ -258,7 +255,7 @@ vfio_enable_msi(struct rte_intr_handle *intr_handle) { /* disable MSI interrupts */ static int -vfio_disable_msi(struct rte_intr_handle *intr_handle) { +vfio_disable_msi(const struct rte_intr_handle *intr_handle) { struct vfio_irq_set *irq_set; char irq_set_buf[IRQ_SET_BUF_LEN]; int len, ret; @@ -281,9 +278,30 @@ vfio_disable_msi(struct rte_intr_handle *intr_handle) { return ret; } +static int +get_max_intr(const struct rte_intr_handle *intr_handle) +{ + struct rte_intr_source *src; + + TAILQ_FOREACH(src, &intr_sources, next) { + if (src->intr_handle.fd != intr_handle->fd) + continue; + + if (!src->intr_handle.max_intr) + src->intr_handle.max_intr = 1; + else if (src->intr_handle.max_intr > RTE_MAX_RXTX_INTR_VEC_ID) + src->intr_handle.max_intr + = RTE_MAX_RXTX_INTR_VEC_ID + 1; + + return src->intr_handle.max_intr; + } + + return -1; +} + /* enable MSI-X interrupts */ static int -vfio_enable_msix(struct rte_intr_handle *intr_handle) { +vfio_enable_msix(const struct rte_intr_handle *intr_handle) { int len, ret; char irq_set_buf[MSIX_IRQ_SET_BUF_LEN]; struct vfio_irq_set *irq_set; @@ -293,12 +311,15 @@ vfio_enable_msix(struct rte_intr_handle *intr_handle) { irq_set = (struct vfio_irq_set *) irq_set_buf; irq_set->argsz = len; - if (!intr_handle->max_intr) - intr_handle->max_intr = 1; - else if (intr_handle->max_intr > RTE_MAX_RXTX_INTR_VEC_ID) - intr_handle->max_intr = RTE_MAX_RXTX_INTR_VEC_ID + 1; - irq_set->count = intr_handle->max_intr; + ret = get_max_intr(intr_handle); + if (ret < 0) { + RTE_LOG(ERR, EAL, "Invalid number of MSI-X irqs for fd %d\n", + intr_handle->fd); + return -1; + } + + irq_set->count = ret; irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD | VFIO_IRQ_SET_ACTION_TRIGGER; irq_set->index = VFIO_PCI_MSIX_IRQ_INDEX; irq_set->start = 0; @@ -321,7 +342,7 @@ vfio_enable_msix(struct rte_intr_handle *intr_handle) { /* disable MSI-X interrupts */ static int -vfio_disable_msix(struct rte_intr_handle *intr_handle) { +vfio_disable_msix(const struct rte_intr_handle *intr_handle) { struct vfio_irq_set *irq_set; char irq_set_buf[MSIX_IRQ_SET_BUF_LEN]; int len, ret; @@ -346,7 +367,7 @@ vfio_disable_msix(struct rte_intr_handle *intr_handle) { #endif static int -uio_intx_intr_disable(struct rte_intr_handle *intr_handle) +uio_intx_intr_disable(const struct rte_intr_handle *intr_handle) { unsigned char command_high; @@ -370,7 +391,7 @@ uio_intx_intr_disable(struct rte_intr_handle *intr_handle) } static int -uio_intx_intr_enable(struct rte_intr_handle *intr_handle) +uio_intx_intr_enable(const struct rte_intr_handle *intr_handle) { unsigned char command_high; @@ -394,7 +415,7 @@ uio_intx_intr_enable(struct rte_intr_handle *intr_handle) } static int -uio_intr_disable(struct rte_intr_handle *intr_handle) +uio_intr_disable(const struct rte_intr_handle *intr_handle) { const int value = 0; @@ -408,7 +429,7 @@ uio_intr_disable(struct rte_intr_handle *intr_handle) } static int -uio_intr_enable(struct rte_intr_handle *intr_handle) +uio_intr_enable(const struct rte_intr_handle *intr_handle) { const int value = 1; @@ -422,7 +443,7 @@ uio_intr_enable(struct rte_intr_handle *intr_handle) } int -rte_intr_callback_register(struct rte_intr_handle *intr_handle, +rte_intr_callback_register(const struct rte_intr_handle *intr_handle, rte_intr_callback_fn cb, void *cb_arg) { int ret, wake_thread; @@ -494,7 +515,7 @@ rte_intr_callback_register(struct rte_intr_handle *intr_handle, } int -rte_intr_callback_unregister(struct rte_intr_handle *intr_handle, +rte_intr_callback_unregister(const struct rte_intr_handle *intr_handle, rte_intr_callback_fn cb_fn, void *cb_arg) { int ret; @@ -558,7 +579,7 @@ rte_intr_callback_unregister(struct rte_intr_handle *intr_handle, } int -rte_intr_enable(struct rte_intr_handle *intr_handle) +rte_intr_enable(const struct rte_intr_handle *intr_handle) { if (!intr_handle || intr_handle->fd < 0 || intr_handle->uio_cfg_fd < 0) return -1; @@ -602,7 +623,7 @@ rte_intr_enable(struct rte_intr_handle *intr_handle) } int -rte_intr_disable(struct rte_intr_handle *intr_handle) +rte_intr_disable(const struct rte_intr_handle *intr_handle) { if (!intr_handle || intr_handle->fd < 0 || intr_handle->uio_cfg_fd < 0) return -1; @@ -712,19 +733,10 @@ eal_intr_process_interrupts(struct epoll_event *events, int nfds) if (errno == EINTR || errno == EWOULDBLOCK) continue; - // TREX_PATCH. Because of issues with e1000, we want this message to - // have lower priority only if running on e1000 card - if (eal_err_read_from_file_is_error) { - RTE_LOG(ERR, EAL, "Error reading from file " - "descriptor %d: %s\n", - events[n].data.fd, - strerror(errno)); - } else { - RTE_LOG(INFO, EAL, "Error reading from file " - "descriptor %d: %s\n", - events[n].data.fd, - strerror(errno)); - } + RTE_LOG(ERR, EAL, "Error reading from file " + "descriptor %d: %s\n", + events[n].data.fd, + strerror(errno)); } else if (bytes_read == 0) RTE_LOG(ERR, EAL, "Read nothing from file " "descriptor %d\n", events[n].data.fd); @@ -1169,7 +1181,7 @@ rte_intr_efd_enable(struct rte_intr_handle *intr_handle, uint32_t nb_efd) RTE_LOG(ERR, EAL, "can't setup eventfd, error %i (%s)\n", errno, strerror(errno)); - return -1; + return -errno; } intr_handle->efds[i] = fd; } diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_ivshmem.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_ivshmem.c deleted file mode 100644 index 67b3caf2..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_ivshmem.c +++ /dev/null @@ -1,954 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2014 Intel Corporation. All rights reserved. - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * 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 - * distribution. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * 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 - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#ifdef RTE_LIBRTE_IVSHMEM /* hide it from coverage */ - -#include <stdint.h> -#include <unistd.h> -#include <inttypes.h> -#include <sys/mman.h> -#include <sys/file.h> -#include <string.h> -#include <sys/queue.h> - -#include <rte_log.h> -#include <rte_pci.h> -#include <rte_memory.h> -#include <rte_eal.h> -#include <rte_eal_memconfig.h> -#include <rte_string_fns.h> -#include <rte_errno.h> -#include <rte_ring.h> -#include <rte_malloc.h> -#include <rte_common.h> -#include <rte_ivshmem.h> - -#include "eal_internal_cfg.h" -#include "eal_private.h" - -#define PCI_VENDOR_ID_IVSHMEM 0x1Af4 -#define PCI_DEVICE_ID_IVSHMEM 0x1110 - -#define IVSHMEM_MAGIC 0x0BADC0DE - -#define IVSHMEM_RESOURCE_PATH "/sys/bus/pci/devices/%04x:%02x:%02x.%x/resource2" -#define IVSHMEM_CONFIG_PATH "/var/run/.%s_ivshmem_config" - -#define PHYS 0x1 -#define VIRT 0x2 -#define IOREMAP 0x4 -#define FULL (PHYS|VIRT|IOREMAP) - -#define METADATA_SIZE_ALIGNED \ - (RTE_ALIGN_CEIL(sizeof(struct rte_ivshmem_metadata),pagesz)) - -#define CONTAINS(x,y)\ - (((y).addr_64 >= (x).addr_64) && ((y).addr_64 < (x).addr_64 + (x).len)) - -#define DIM(x) (sizeof(x)/sizeof(x[0])) - -struct ivshmem_pci_device { - char path[PATH_MAX]; - phys_addr_t ioremap_addr; -}; - -/* data type to store in config */ -struct ivshmem_segment { - struct rte_ivshmem_metadata_entry entry; - uint64_t align; - char path[PATH_MAX]; -}; -struct ivshmem_shared_config { - struct ivshmem_segment segment[RTE_MAX_MEMSEG]; - uint32_t segment_idx; - struct ivshmem_pci_device pci_devs[RTE_LIBRTE_IVSHMEM_MAX_PCI_DEVS]; - uint32_t pci_devs_idx; -}; -static struct ivshmem_shared_config * ivshmem_config; -static int memseg_idx; -static int pagesz; - -/* Tailq heads to add rings to */ -TAILQ_HEAD(rte_ring_list, rte_tailq_entry); - -/* - * Utility functions - */ - -static int -is_ivshmem_device(struct rte_pci_device * dev) -{ - return dev->id.vendor_id == PCI_VENDOR_ID_IVSHMEM - && dev->id.device_id == PCI_DEVICE_ID_IVSHMEM; -} - -static void * -map_metadata(int fd, uint64_t len) -{ - size_t metadata_len = sizeof(struct rte_ivshmem_metadata); - size_t aligned_len = METADATA_SIZE_ALIGNED; - - return mmap(NULL, metadata_len, PROT_READ | PROT_WRITE, - MAP_SHARED, fd, len - aligned_len); -} - -static void -unmap_metadata(void * ptr) -{ - munmap(ptr, sizeof(struct rte_ivshmem_metadata)); -} - -static int -has_ivshmem_metadata(int fd, uint64_t len) -{ - struct rte_ivshmem_metadata metadata; - void * ptr; - - ptr = map_metadata(fd, len); - - if (ptr == MAP_FAILED) - return -1; - - metadata = *(struct rte_ivshmem_metadata*) (ptr); - - unmap_metadata(ptr); - - return metadata.magic_number == IVSHMEM_MAGIC; -} - -static void -remove_segment(struct ivshmem_segment * ms, int len, int idx) -{ - int i; - - for (i = idx; i < len - 1; i++) - memcpy(&ms[i], &ms[i+1], sizeof(struct ivshmem_segment)); - memset(&ms[len-1], 0, sizeof(struct ivshmem_segment)); -} - -static int -overlap(const struct rte_memzone * mz1, const struct rte_memzone * mz2) -{ - uint64_t start1, end1, start2, end2; - uint64_t p_start1, p_end1, p_start2, p_end2; - uint64_t i_start1, i_end1, i_start2, i_end2; - int result = 0; - - /* gather virtual addresses */ - start1 = mz1->addr_64; - end1 = mz1->addr_64 + mz1->len; - start2 = mz2->addr_64; - end2 = mz2->addr_64 + mz2->len; - - /* gather physical addresses */ - p_start1 = mz1->phys_addr; - p_end1 = mz1->phys_addr + mz1->len; - p_start2 = mz2->phys_addr; - p_end2 = mz2->phys_addr + mz2->len; - - /* gather ioremap addresses */ - i_start1 = mz1->ioremap_addr; - i_end1 = mz1->ioremap_addr + mz1->len; - i_start2 = mz2->ioremap_addr; - i_end2 = mz2->ioremap_addr + mz2->len; - - /* check for overlap in virtual addresses */ - if (start1 >= start2 && start1 < end2) - result |= VIRT; - if (start2 >= start1 && start2 < end1) - result |= VIRT; - - /* check for overlap in physical addresses */ - if (p_start1 >= p_start2 && p_start1 < p_end2) - result |= PHYS; - if (p_start2 >= p_start1 && p_start2 < p_end1) - result |= PHYS; - - /* check for overlap in ioremap addresses */ - if (i_start1 >= i_start2 && i_start1 < i_end2) - result |= IOREMAP; - if (i_start2 >= i_start1 && i_start2 < i_end1) - result |= IOREMAP; - - return result; -} - -static int -adjacent(const struct rte_memzone * mz1, const struct rte_memzone * mz2) -{ - uint64_t start1, end1, start2, end2; - uint64_t p_start1, p_end1, p_start2, p_end2; - uint64_t i_start1, i_end1, i_start2, i_end2; - int result = 0; - - /* gather virtual addresses */ - start1 = mz1->addr_64; - end1 = mz1->addr_64 + mz1->len; - start2 = mz2->addr_64; - end2 = mz2->addr_64 + mz2->len; - - /* gather physical addresses */ - p_start1 = mz1->phys_addr; - p_end1 = mz1->phys_addr + mz1->len; - p_start2 = mz2->phys_addr; - p_end2 = mz2->phys_addr + mz2->len; - - /* gather ioremap addresses */ - i_start1 = mz1->ioremap_addr; - i_end1 = mz1->ioremap_addr + mz1->len; - i_start2 = mz2->ioremap_addr; - i_end2 = mz2->ioremap_addr + mz2->len; - - /* check if segments are virtually adjacent */ - if (start1 == end2) - result |= VIRT; - if (start2 == end1) - result |= VIRT; - - /* check if segments are physically adjacent */ - if (p_start1 == p_end2) - result |= PHYS; - if (p_start2 == p_end1) - result |= PHYS; - - /* check if segments are ioremap-adjacent */ - if (i_start1 == i_end2) - result |= IOREMAP; - if (i_start2 == i_end1) - result |= IOREMAP; - - return result; -} - -static int -has_adjacent_segments(struct ivshmem_segment * ms, int len) -{ - int i, j; - - for (i = 0; i < len; i++) - for (j = i + 1; j < len; j++) { - /* we're only interested in fully adjacent segments; partially - * adjacent segments can coexist. - */ - if (adjacent(&ms[i].entry.mz, &ms[j].entry.mz) == FULL) - return 1; - } - return 0; -} - -static int -has_overlapping_segments(struct ivshmem_segment * ms, int len) -{ - int i, j; - - for (i = 0; i < len; i++) - for (j = i + 1; j < len; j++) - if (overlap(&ms[i].entry.mz, &ms[j].entry.mz)) - return 1; - return 0; -} - -static int -seg_compare(const void * a, const void * b) -{ - const struct ivshmem_segment * s1 = (const struct ivshmem_segment*) a; - const struct ivshmem_segment * s2 = (const struct ivshmem_segment*) b; - - /* move unallocated zones to the end */ - if (s1->entry.mz.addr == NULL && s2->entry.mz.addr == NULL) - return 0; - if (s1->entry.mz.addr == 0) - return 1; - if (s2->entry.mz.addr == 0) - return -1; - - return s1->entry.mz.phys_addr > s2->entry.mz.phys_addr; -} - -#ifdef RTE_LIBRTE_IVSHMEM_DEBUG -static void -entry_dump(struct rte_ivshmem_metadata_entry *e) -{ - RTE_LOG(DEBUG, EAL, "\tvirt: %p-%p\n", e->mz.addr, - RTE_PTR_ADD(e->mz.addr, e->mz.len)); - RTE_LOG(DEBUG, EAL, "\tphys: 0x%" PRIx64 "-0x%" PRIx64 "\n", - e->mz.phys_addr, - e->mz.phys_addr + e->mz.len); - RTE_LOG(DEBUG, EAL, "\tio: 0x%" PRIx64 "-0x%" PRIx64 "\n", - e->mz.ioremap_addr, - e->mz.ioremap_addr + e->mz.len); - RTE_LOG(DEBUG, EAL, "\tlen: 0x%" PRIx64 "\n", e->mz.len); - RTE_LOG(DEBUG, EAL, "\toff: 0x%" PRIx64 "\n", e->offset); -} -#endif - - - -/* - * Actual useful code - */ - -/* read through metadata mapped from the IVSHMEM device */ -static int -read_metadata(char * path, int path_len, int fd, uint64_t flen) -{ - struct rte_ivshmem_metadata metadata; - struct rte_ivshmem_metadata_entry * entry; - int idx, i; - void * ptr; - - ptr = map_metadata(fd, flen); - - if (ptr == MAP_FAILED) - return -1; - - metadata = *(struct rte_ivshmem_metadata*) (ptr); - - unmap_metadata(ptr); - - RTE_LOG(DEBUG, EAL, "Parsing metadata for \"%s\"\n", metadata.name); - - idx = ivshmem_config->segment_idx; - - for (i = 0; i < RTE_LIBRTE_IVSHMEM_MAX_ENTRIES && - idx <= RTE_MAX_MEMSEG; i++) { - - if (idx == RTE_MAX_MEMSEG) { - RTE_LOG(ERR, EAL, "Not enough memory segments!\n"); - return -1; - } - - entry = &metadata.entry[i]; - - /* stop on uninitialized memzone */ - if (entry->mz.len == 0) - break; - - /* copy metadata entry */ - memcpy(&ivshmem_config->segment[idx].entry, entry, - sizeof(struct rte_ivshmem_metadata_entry)); - - /* copy path */ - snprintf(ivshmem_config->segment[idx].path, path_len, "%s", path); - - idx++; - } - ivshmem_config->segment_idx = idx; - - return 0; -} - -/* check through each segment and look for adjacent or overlapping ones. */ -static int -cleanup_segments(struct ivshmem_segment * ms, int tbl_len) -{ - struct ivshmem_segment * s, * tmp; - int i, j, concat, seg_adjacent, seg_overlapping; - uint64_t start1, start2, end1, end2, p_start1, p_start2, i_start1, i_start2; - - qsort(ms, tbl_len, sizeof(struct ivshmem_segment), - seg_compare); - - while (has_overlapping_segments(ms, tbl_len) || - has_adjacent_segments(ms, tbl_len)) { - - for (i = 0; i < tbl_len; i++) { - s = &ms[i]; - - concat = 0; - - for (j = i + 1; j < tbl_len; j++) { - tmp = &ms[j]; - - /* check if this segment is overlapping with existing segment, - * or is adjacent to existing segment */ - seg_overlapping = overlap(&s->entry.mz, &tmp->entry.mz); - seg_adjacent = adjacent(&s->entry.mz, &tmp->entry.mz); - - /* check if segments fully overlap or are fully adjacent */ - if ((seg_adjacent == FULL) || (seg_overlapping == FULL)) { - -#ifdef RTE_LIBRTE_IVSHMEM_DEBUG - RTE_LOG(DEBUG, EAL, "Concatenating segments\n"); - RTE_LOG(DEBUG, EAL, "Segment %i:\n", i); - entry_dump(&s->entry); - RTE_LOG(DEBUG, EAL, "Segment %i:\n", j); - entry_dump(&tmp->entry); -#endif - - start1 = s->entry.mz.addr_64; - start2 = tmp->entry.mz.addr_64; - p_start1 = s->entry.mz.phys_addr; - p_start2 = tmp->entry.mz.phys_addr; - i_start1 = s->entry.mz.ioremap_addr; - i_start2 = tmp->entry.mz.ioremap_addr; - end1 = s->entry.mz.addr_64 + s->entry.mz.len; - end2 = tmp->entry.mz.addr_64 + tmp->entry.mz.len; - - /* settle for minimum start address and maximum length */ - s->entry.mz.addr_64 = RTE_MIN(start1, start2); - s->entry.mz.phys_addr = RTE_MIN(p_start1, p_start2); - s->entry.mz.ioremap_addr = RTE_MIN(i_start1, i_start2); - s->entry.offset = RTE_MIN(s->entry.offset, tmp->entry.offset); - s->entry.mz.len = RTE_MAX(end1, end2) - s->entry.mz.addr_64; - concat = 1; - -#ifdef RTE_LIBRTE_IVSHMEM_DEBUG - RTE_LOG(DEBUG, EAL, "Resulting segment:\n"); - entry_dump(&s->entry); - -#endif - } - /* if segments not fully overlap, we have an error condition. - * adjacent segments can coexist. - */ - else if (seg_overlapping > 0) { - RTE_LOG(ERR, EAL, "Segments %i and %i overlap!\n", i, j); -#ifdef RTE_LIBRTE_IVSHMEM_DEBUG - RTE_LOG(DEBUG, EAL, "Segment %i:\n", i); - entry_dump(&s->entry); - RTE_LOG(DEBUG, EAL, "Segment %i:\n", j); - entry_dump(&tmp->entry); -#endif - return -1; - } - if (concat) - break; - } - /* if we concatenated, remove segment at j */ - if (concat) { - remove_segment(ms, tbl_len, j); - tbl_len--; - break; - } - } - } - - return tbl_len; -} - -static int -create_shared_config(void) -{ - char path[PATH_MAX]; - int fd; - - /* build ivshmem config file path */ - snprintf(path, sizeof(path), IVSHMEM_CONFIG_PATH, - internal_config.hugefile_prefix); - - fd = open(path, O_CREAT | O_RDWR, 0600); - - if (fd < 0) { - RTE_LOG(ERR, EAL, "Could not open %s: %s\n", path, strerror(errno)); - return -1; - } - - /* try ex-locking first - if the file is locked, we have a problem */ - if (flock(fd, LOCK_EX | LOCK_NB) == -1) { - RTE_LOG(ERR, EAL, "Locking %s failed: %s\n", path, strerror(errno)); - close(fd); - return -1; - } - - if (ftruncate(fd, sizeof(struct ivshmem_shared_config)) < 0) { - RTE_LOG(ERR, EAL, "ftruncate failed: %s\n", strerror(errno)); - return -1; - } - - ivshmem_config = mmap(NULL, sizeof(struct ivshmem_shared_config), - PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0); - - if (ivshmem_config == MAP_FAILED) - return -1; - - memset(ivshmem_config, 0, sizeof(struct ivshmem_shared_config)); - - /* change the exclusive lock we got earlier to a shared lock */ - if (flock(fd, LOCK_SH | LOCK_NB) == -1) { - RTE_LOG(ERR, EAL, "Locking %s failed: %s \n", path, strerror(errno)); - return -1; - } - - close(fd); - - return 0; -} - -/* open shared config file and, if present, map the config. - * having no config file is not an error condition, as we later check if - * ivshmem_config is NULL (if it is, that means nothing was mapped). */ -static int -open_shared_config(void) -{ - char path[PATH_MAX]; - int fd; - - /* build ivshmem config file path */ - snprintf(path, sizeof(path), IVSHMEM_CONFIG_PATH, - internal_config.hugefile_prefix); - - fd = open(path, O_RDONLY); - - /* if the file doesn't exist, just return success */ - if (fd < 0 && errno == ENOENT) - return 0; - /* else we have an error condition */ - else if (fd < 0) { - RTE_LOG(ERR, EAL, "Could not open %s: %s\n", - path, strerror(errno)); - return -1; - } - - /* try ex-locking first - if the lock *does* succeed, this means it's a - * stray config file, so it should be deleted. - */ - if (flock(fd, LOCK_EX | LOCK_NB) != -1) { - - /* if we can't remove the file, something is wrong */ - if (unlink(path) < 0) { - RTE_LOG(ERR, EAL, "Could not remove %s: %s\n", path, - strerror(errno)); - return -1; - } - - /* release the lock */ - flock(fd, LOCK_UN); - close(fd); - - /* return success as having a stray config file is equivalent to not - * having config file at all. - */ - return 0; - } - - ivshmem_config = mmap(NULL, sizeof(struct ivshmem_shared_config), - PROT_READ, MAP_SHARED, fd, 0); - - if (ivshmem_config == MAP_FAILED) - return -1; - - /* place a shared lock on config file */ - if (flock(fd, LOCK_SH | LOCK_NB) == -1) { - RTE_LOG(ERR, EAL, "Locking %s failed: %s \n", path, strerror(errno)); - return -1; - } - - close(fd); - - return 0; -} - -/* - * This function does the following: - * - * 1) Builds a table of ivshmem_segments with proper offset alignment - * 2) Cleans up that table so that we don't have any overlapping or adjacent - * memory segments - * 3) Creates memsegs from this table and maps them into memory. - */ -static inline int -map_all_segments(void) -{ - struct ivshmem_segment ms_tbl[RTE_MAX_MEMSEG]; - struct ivshmem_pci_device * pci_dev; - struct rte_mem_config * mcfg; - struct ivshmem_segment * seg; - int fd, fd_zero; - unsigned i, j; - struct rte_memzone mz; - struct rte_memseg ms; - void * base_addr; - uint64_t align, len; - phys_addr_t ioremap_addr; - - ioremap_addr = 0; - - memset(ms_tbl, 0, sizeof(ms_tbl)); - memset(&mz, 0, sizeof(struct rte_memzone)); - memset(&ms, 0, sizeof(struct rte_memseg)); - - /* first, build a table of memsegs to map, to avoid failed mmaps due to - * overlaps - */ - for (i = 0; i < ivshmem_config->segment_idx && i <= RTE_MAX_MEMSEG; i++) { - if (i == RTE_MAX_MEMSEG) { - RTE_LOG(ERR, EAL, "Too many segments requested!\n"); - return -1; - } - - seg = &ivshmem_config->segment[i]; - - /* copy segment to table */ - memcpy(&ms_tbl[i], seg, sizeof(struct ivshmem_segment)); - - /* find ioremap addr */ - for (j = 0; j < DIM(ivshmem_config->pci_devs); j++) { - pci_dev = &ivshmem_config->pci_devs[j]; - if (!strncmp(pci_dev->path, seg->path, sizeof(pci_dev->path))) { - ioremap_addr = pci_dev->ioremap_addr; - break; - } - } - if (ioremap_addr == 0) { - RTE_LOG(ERR, EAL, "Cannot find ioremap addr!\n"); - return -1; - } - - /* work out alignments */ - align = seg->entry.mz.addr_64 - - RTE_ALIGN_FLOOR(seg->entry.mz.addr_64, 0x1000); - len = RTE_ALIGN_CEIL(seg->entry.mz.len + align, 0x1000); - - /* save original alignments */ - ms_tbl[i].align = align; - - /* create a memory zone */ - mz.addr_64 = seg->entry.mz.addr_64 - align; - mz.len = len; - mz.hugepage_sz = seg->entry.mz.hugepage_sz; - mz.phys_addr = seg->entry.mz.phys_addr - align; - - /* find true physical address */ - mz.ioremap_addr = ioremap_addr + seg->entry.offset - align; - - ms_tbl[i].entry.offset = seg->entry.offset - align; - - memcpy(&ms_tbl[i].entry.mz, &mz, sizeof(struct rte_memzone)); - } - - /* clean up the segments */ - memseg_idx = cleanup_segments(ms_tbl, ivshmem_config->segment_idx); - - if (memseg_idx < 0) - return -1; - - mcfg = rte_eal_get_configuration()->mem_config; - - fd_zero = open("/dev/zero", O_RDWR); - - if (fd_zero < 0) { - RTE_LOG(ERR, EAL, "Cannot open /dev/zero: %s\n", strerror(errno)); - return -1; - } - - /* create memsegs and put them into DPDK memory */ - for (i = 0; i < (unsigned) memseg_idx; i++) { - - seg = &ms_tbl[i]; - - ms.addr_64 = seg->entry.mz.addr_64; - ms.hugepage_sz = seg->entry.mz.hugepage_sz; - ms.len = seg->entry.mz.len; - ms.nchannel = rte_memory_get_nchannel(); - ms.nrank = rte_memory_get_nrank(); - ms.phys_addr = seg->entry.mz.phys_addr; - ms.ioremap_addr = seg->entry.mz.ioremap_addr; - ms.socket_id = seg->entry.mz.socket_id; - - base_addr = mmap(ms.addr, ms.len, - PROT_READ | PROT_WRITE, MAP_PRIVATE, fd_zero, 0); - - if (base_addr == MAP_FAILED || base_addr != ms.addr) { - RTE_LOG(ERR, EAL, "Cannot map /dev/zero!\n"); - return -1; - } - - fd = open(seg->path, O_RDWR); - - if (fd < 0) { - RTE_LOG(ERR, EAL, "Cannot open %s: %s\n", seg->path, - strerror(errno)); - return -1; - } - - munmap(ms.addr, ms.len); - - base_addr = mmap(ms.addr, ms.len, - PROT_READ | PROT_WRITE, MAP_SHARED, fd, - seg->entry.offset); - - - if (base_addr == MAP_FAILED || base_addr != ms.addr) { - RTE_LOG(ERR, EAL, "Cannot map segment into memory: " - "expected %p got %p (%s)\n", ms.addr, base_addr, - strerror(errno)); - return -1; - } - - RTE_LOG(DEBUG, EAL, "Memory segment mapped: %p (len %" PRIx64 ") at " - "offset 0x%" PRIx64 "\n", - ms.addr, ms.len, seg->entry.offset); - - /* put the pointers back into their real positions using original - * alignment */ - ms.addr_64 += seg->align; - ms.phys_addr += seg->align; - ms.ioremap_addr += seg->align; - ms.len -= seg->align; - - /* at this point, the rest of DPDK memory is not initialized, so we - * expect memsegs to be empty */ - memcpy(&mcfg->memseg[i], &ms, - sizeof(struct rte_memseg)); - - close(fd); - - RTE_LOG(DEBUG, EAL, "IVSHMEM segment found, size: 0x%lx\n", - ms.len); - } - - return 0; -} - -/* this happens at a later stage, after general EAL memory initialization */ -int -rte_eal_ivshmem_obj_init(void) -{ - struct rte_ring_list* ring_list = NULL; - struct rte_mem_config * mcfg; - struct ivshmem_segment * seg; - struct rte_memzone * mz; - struct rte_ring * r; - struct rte_tailq_entry *te; - unsigned i, ms, idx; - uint64_t offset; - - /* secondary process would not need any object discovery - it'll all - * already be in shared config */ - if (rte_eal_process_type() != RTE_PROC_PRIMARY || ivshmem_config == NULL) - return 0; - - /* check that we have an initialised ring tail queue */ - ring_list = RTE_TAILQ_LOOKUP(RTE_TAILQ_RING_NAME, rte_ring_list); - if (ring_list == NULL) { - RTE_LOG(ERR, EAL, "No rte_ring tailq found!\n"); - return -1; - } - - mcfg = rte_eal_get_configuration()->mem_config; - - /* create memzones */ - for (i = 0; i < ivshmem_config->segment_idx && i <= RTE_MAX_MEMZONE; i++) { - - seg = &ivshmem_config->segment[i]; - - /* add memzone */ - if (mcfg->memzone_cnt == RTE_MAX_MEMZONE) { - RTE_LOG(ERR, EAL, "No more memory zones available!\n"); - return -1; - } - - idx = mcfg->memzone_cnt; - - RTE_LOG(DEBUG, EAL, "Found memzone: '%s' at %p (len 0x%" PRIx64 ")\n", - seg->entry.mz.name, seg->entry.mz.addr, seg->entry.mz.len); - - memcpy(&mcfg->memzone[idx], &seg->entry.mz, - sizeof(struct rte_memzone)); - - /* find ioremap address */ - for (ms = 0; ms <= RTE_MAX_MEMSEG; ms++) { - if (ms == RTE_MAX_MEMSEG) { - RTE_LOG(ERR, EAL, "Physical address of segment not found!\n"); - return -1; - } - if (CONTAINS(mcfg->memseg[ms], mcfg->memzone[idx])) { - offset = mcfg->memzone[idx].addr_64 - - mcfg->memseg[ms].addr_64; - mcfg->memzone[idx].ioremap_addr = mcfg->memseg[ms].ioremap_addr + - offset; - break; - } - } - - mcfg->memzone_cnt++; - } - - rte_rwlock_write_lock(RTE_EAL_TAILQ_RWLOCK); - - /* find rings */ - for (i = 0; i < mcfg->memzone_cnt; i++) { - mz = &mcfg->memzone[i]; - - /* check if memzone has a ring prefix */ - if (strncmp(mz->name, RTE_RING_MZ_PREFIX, - sizeof(RTE_RING_MZ_PREFIX) - 1) != 0) - continue; - - r = (struct rte_ring*) (mz->addr_64); - - te = rte_zmalloc("RING_TAILQ_ENTRY", sizeof(*te), 0); - if (te == NULL) { - RTE_LOG(ERR, EAL, "Cannot allocate ring tailq entry!\n"); - return -1; - } - - te->data = (void *) r; - - TAILQ_INSERT_TAIL(ring_list, te, next); - - RTE_LOG(DEBUG, EAL, "Found ring: '%s' at %p\n", r->name, mz->addr); - } - rte_rwlock_write_unlock(RTE_EAL_TAILQ_RWLOCK); - -#ifdef RTE_LIBRTE_IVSHMEM_DEBUG - rte_memzone_dump(stdout); - rte_ring_list_dump(stdout); -#endif - - return 0; -} - -/* initialize ivshmem structures */ -int rte_eal_ivshmem_init(void) -{ - struct rte_pci_device * dev; - struct rte_pci_resource * res; - int fd, ret; - char path[PATH_MAX]; - - /* initialize everything to 0 */ - memset(path, 0, sizeof(path)); - ivshmem_config = NULL; - - pagesz = getpagesize(); - - RTE_LOG(DEBUG, EAL, "Searching for IVSHMEM devices...\n"); - - if (rte_eal_process_type() == RTE_PROC_SECONDARY) { - - if (open_shared_config() < 0) { - RTE_LOG(ERR, EAL, "Could not open IVSHMEM config!\n"); - return -1; - } - } - else { - - TAILQ_FOREACH(dev, &pci_device_list, next) { - - if (is_ivshmem_device(dev)) { - - /* IVSHMEM memory is always on BAR2 */ - res = &dev->mem_resource[2]; - - /* if we don't have a BAR2 */ - if (res->len == 0) - continue; - - /* construct pci device path */ - snprintf(path, sizeof(path), IVSHMEM_RESOURCE_PATH, - dev->addr.domain, dev->addr.bus, dev->addr.devid, - dev->addr.function); - - /* try to find memseg */ - fd = open(path, O_RDWR); - if (fd < 0) { - RTE_LOG(ERR, EAL, "Could not open %s\n", path); - return -1; - } - - /* check if it's a DPDK IVSHMEM device */ - ret = has_ivshmem_metadata(fd, res->len); - - /* is DPDK device */ - if (ret == 1) { - - /* config file creation is deferred until the first - * DPDK device is found. then, it has to be created - * only once. */ - if (ivshmem_config == NULL && - create_shared_config() < 0) { - RTE_LOG(ERR, EAL, "Could not create IVSHMEM config!\n"); - close(fd); - return -1; - } - - if (read_metadata(path, sizeof(path), fd, res->len) < 0) { - RTE_LOG(ERR, EAL, "Could not read metadata from" - " device %02x:%02x.%x!\n", dev->addr.bus, - dev->addr.devid, dev->addr.function); - close(fd); - return -1; - } - - if (ivshmem_config->pci_devs_idx == RTE_LIBRTE_IVSHMEM_MAX_PCI_DEVS) { - RTE_LOG(WARNING, EAL, - "IVSHMEM PCI device limit exceeded. Increase " - "CONFIG_RTE_LIBRTE_IVSHMEM_MAX_PCI_DEVS in " - "your config file.\n"); - break; - } - - RTE_LOG(INFO, EAL, "Found IVSHMEM device %02x:%02x.%x\n", - dev->addr.bus, dev->addr.devid, dev->addr.function); - - ivshmem_config->pci_devs[ivshmem_config->pci_devs_idx].ioremap_addr = res->phys_addr; - snprintf(ivshmem_config->pci_devs[ivshmem_config->pci_devs_idx].path, - sizeof(ivshmem_config->pci_devs[ivshmem_config->pci_devs_idx].path), - "%s", path); - - ivshmem_config->pci_devs_idx++; - } - /* failed to read */ - else if (ret < 0) { - RTE_LOG(ERR, EAL, "Could not read IVSHMEM device: %s\n", - strerror(errno)); - close(fd); - return -1; - } - /* not a DPDK device */ - else - RTE_LOG(DEBUG, EAL, "Skipping non-DPDK IVSHMEM device\n"); - - /* close the BAR fd */ - close(fd); - } - } - } - - /* ivshmem_config is not NULL only if config was created and/or mapped */ - if (ivshmem_config) { - if (map_all_segments() < 0) { - RTE_LOG(ERR, EAL, "Mapping IVSHMEM segments failed!\n"); - return -1; - } - } - else { - RTE_LOG(DEBUG, EAL, "No IVSHMEM configuration found! \n"); - } - - return 0; -} - -#endif diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_log.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_log.c index d3911004..e3a50aa3 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_log.c +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_log.c @@ -97,45 +97,7 @@ rte_eal_log_init(const char *id, int facility) openlog(id, LOG_NDELAY | LOG_PID, facility); - if (rte_eal_common_log_init(log_stream) < 0) - return -1; - - return 0; -} - -/* early logs */ - -/* - * early log function, used before rte_eal_log_init - */ -static ssize_t -early_log_write(__attribute__((unused)) void *c, const char *buf, size_t size) -{ - ssize_t ret; - ret = fwrite(buf, size, 1, stdout); - fflush(stdout); - if (ret == 0) - return -1; - return ret; -} - -static cookie_io_functions_t early_log_func = { - .write = early_log_write, -}; -static FILE *early_log_stream; + eal_log_set_default(log_stream); -/* - * init the log library, called by rte_eal_init() to enable early - * logs - */ -int -rte_eal_log_early_init(void) -{ - early_log_stream = fopencookie(NULL, "w+", early_log_func); - if (early_log_stream == NULL) { - printf("Cannot configure early_log_stream\n"); - return -1; - } - rte_openlog_stream(early_log_stream); return 0; } diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_memory.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_memory.c index 41e0a928..a956bb22 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_memory.c +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_memory.c @@ -376,25 +376,15 @@ map_all_hugepages(struct hugepage_file *hugepg_tbl, void *vma_addr = NULL; size_t vma_len = 0; -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - RTE_SET_USED(vma_len); -#endif - for (i = 0; i < hpi->num_pages[0]; i++) { uint64_t hugepage_sz = hpi->hugepage_sz; if (orig) { hugepg_tbl[i].file_id = i; hugepg_tbl[i].size = hugepage_sz; -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - eal_get_hugefile_temp_path(hugepg_tbl[i].filepath, - sizeof(hugepg_tbl[i].filepath), hpi->hugedir, - hugepg_tbl[i].file_id); -#else eal_get_hugefile_path(hugepg_tbl[i].filepath, sizeof(hugepg_tbl[i].filepath), hpi->hugedir, hugepg_tbl[i].file_id); -#endif hugepg_tbl[i].filepath[sizeof(hugepg_tbl[i].filepath) - 1] = '\0'; } #ifndef RTE_ARCH_64 @@ -408,8 +398,6 @@ map_all_hugepages(struct hugepage_file *hugepg_tbl, continue; } #endif - -#ifndef RTE_EAL_SINGLE_FILE_SEGMENTS else if (vma_len == 0) { unsigned j, num_pages; @@ -439,10 +427,9 @@ map_all_hugepages(struct hugepage_file *hugepg_tbl, if (vma_addr == NULL) vma_len = hugepage_sz; } -#endif /* try to create hugepage file */ - fd = open(hugepg_tbl[i].filepath, O_CREAT | O_RDWR, 0755); + fd = open(hugepg_tbl[i].filepath, O_CREAT | O_RDWR, 0600); if (fd < 0) { RTE_LOG(DEBUG, EAL, "%s(): open failed: %s\n", __func__, strerror(errno)); @@ -505,169 +492,6 @@ map_all_hugepages(struct hugepage_file *hugepg_tbl, return i; } -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - -/* - * Remaps all hugepages into single file segments - */ -static int -remap_all_hugepages(struct hugepage_file *hugepg_tbl, struct hugepage_info *hpi) -{ - int fd; - unsigned i = 0, j, num_pages, page_idx = 0; - void *vma_addr = NULL, *old_addr = NULL, *page_addr = NULL; - size_t vma_len = 0; - size_t hugepage_sz = hpi->hugepage_sz; - size_t total_size, offset; - char filepath[MAX_HUGEPAGE_PATH]; - phys_addr_t physaddr; - int socket; - - while (i < hpi->num_pages[0]) { - -#ifndef RTE_ARCH_64 - /* for 32-bit systems, don't remap 1G pages and 16G pages, - * just reuse original map address as final map address. - */ - if ((hugepage_sz == RTE_PGSIZE_1G) - || (hugepage_sz == RTE_PGSIZE_16G)) { - hugepg_tbl[i].final_va = hugepg_tbl[i].orig_va; - hugepg_tbl[i].orig_va = NULL; - i++; - continue; - } -#endif - - /* reserve a virtual area for next contiguous - * physical block: count the number of - * contiguous physical pages. */ - for (j = i+1; j < hpi->num_pages[0] ; j++) { -#ifdef RTE_ARCH_PPC_64 - /* The physical addresses are sorted in descending - * order on PPC64 */ - if (hugepg_tbl[j].physaddr != - hugepg_tbl[j-1].physaddr - hugepage_sz) - break; -#else - if (hugepg_tbl[j].physaddr != - hugepg_tbl[j-1].physaddr + hugepage_sz) - break; -#endif - } - num_pages = j - i; - vma_len = num_pages * hugepage_sz; - - socket = hugepg_tbl[i].socket_id; - - /* get the biggest virtual memory area up to - * vma_len. If it fails, vma_addr is NULL, so - * let the kernel provide the address. */ - vma_addr = get_virtual_area(&vma_len, hpi->hugepage_sz); - - /* If we can't find a big enough virtual area, work out how many pages - * we are going to get */ - if (vma_addr == NULL) - j = i + 1; - else if (vma_len != num_pages * hugepage_sz) { - num_pages = vma_len / hugepage_sz; - j = i + num_pages; - - } - - hugepg_tbl[page_idx].file_id = page_idx; - eal_get_hugefile_path(filepath, - sizeof(filepath), - hpi->hugedir, - hugepg_tbl[page_idx].file_id); - - /* try to create hugepage file */ - fd = open(filepath, O_CREAT | O_RDWR, 0755); - if (fd < 0) { - RTE_LOG(ERR, EAL, "%s(): open failed: %s\n", __func__, strerror(errno)); - return -1; - } - - total_size = 0; - for (;i < j; i++) { - - /* unmap current segment */ - if (total_size > 0) - munmap(vma_addr, total_size); - - /* unmap original page */ - munmap(hugepg_tbl[i].orig_va, hugepage_sz); - unlink(hugepg_tbl[i].filepath); - - total_size += hugepage_sz; - - old_addr = vma_addr; - - /* map new, bigger segment, and populate page tables, - * the kernel fills this segment with zeros */ - vma_addr = mmap(vma_addr, total_size, - PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, fd, 0); - - if (vma_addr == MAP_FAILED || vma_addr != old_addr) { - RTE_LOG(ERR, EAL, "%s(): mmap failed: %s\n", __func__, strerror(errno)); - close(fd); - return -1; - } - } - - /* set shared flock on the file. */ - if (flock(fd, LOCK_SH | LOCK_NB) == -1) { - RTE_LOG(ERR, EAL, "%s(): Locking file failed:%s \n", - __func__, strerror(errno)); - close(fd); - return -1; - } - - snprintf(hugepg_tbl[page_idx].filepath, MAX_HUGEPAGE_PATH, "%s", - filepath); - - physaddr = rte_mem_virt2phy(vma_addr); - - if (physaddr == RTE_BAD_PHYS_ADDR) - return -1; - - hugepg_tbl[page_idx].final_va = vma_addr; - - hugepg_tbl[page_idx].physaddr = physaddr; - - hugepg_tbl[page_idx].repeated = num_pages; - - hugepg_tbl[page_idx].socket_id = socket; - - close(fd); - - /* verify the memory segment - that is, check that every VA corresponds - * to the physical address we expect to see - */ - for (offset = 0; offset < vma_len; offset += hugepage_sz) { - uint64_t expected_physaddr; - - expected_physaddr = hugepg_tbl[page_idx].physaddr + offset; - page_addr = RTE_PTR_ADD(vma_addr, offset); - physaddr = rte_mem_virt2phy(page_addr); - - if (physaddr != expected_physaddr) { - RTE_LOG(ERR, EAL, "Segment sanity check failed: wrong physaddr " - "at %p (offset 0x%" PRIx64 ": 0x%" PRIx64 - " (expected 0x%" PRIx64 ")\n", - page_addr, offset, physaddr, expected_physaddr); - return -1; - } - } - - page_idx++; - } - - /* zero out the rest */ - memset(&hugepg_tbl[page_idx], 0, (hpi->num_pages[0] - page_idx) * sizeof(struct hugepage_file)); - return page_idx; -} -#else/* RTE_EAL_SINGLE_FILE_SEGMENTS=n */ - /* Unmap all hugepages from original mapping */ static int unmap_all_hugepages_orig(struct hugepage_file *hugepg_tbl, struct hugepage_info *hpi) @@ -681,7 +505,6 @@ unmap_all_hugepages_orig(struct hugepage_file *hugepg_tbl, struct hugepage_info } return 0; } -#endif /* RTE_EAL_SINGLE_FILE_SEGMENTS */ /* * Parse /proc/self/numa_maps to get the NUMA socket ID for each huge @@ -875,12 +698,6 @@ unmap_unneeded_hugepages(struct hugepage_file *hugepg_tbl, for (page = 0; page < nrpages; page++) { struct hugepage_file *hp = &hugepg_tbl[page]; -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - /* if this page was already cleared */ - if (hp->final_va == NULL) - continue; -#endif - /* find a page that matches the criteria */ if ((hp->size == hpi[size].hugepage_sz) && (hp->socket_id == (int) socket)) { @@ -889,11 +706,7 @@ unmap_unneeded_hugepages(struct hugepage_file *hugepg_tbl, if (pages_found == hpi[size].num_pages[socket]) { uint64_t unmap_len; -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - unmap_len = hp->size * hp->repeated; -#else unmap_len = hp->size; -#endif /* get start addr and len of the remaining segment */ munmap(hp->final_va, (size_t) unmap_len); @@ -904,50 +717,10 @@ unmap_unneeded_hugepages(struct hugepage_file *hugepg_tbl, __func__, hp->filepath, strerror(errno)); return -1; } - } -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - /* else, check how much do we need to map */ - else { - int nr_pg_left = - hpi[size].num_pages[socket] - pages_found; - - /* if we need enough memory to fit into the segment */ - if (hp->repeated <= nr_pg_left) { - pages_found += hp->repeated; - } - /* truncate the segment */ - else { - uint64_t final_size = nr_pg_left * hp->size; - uint64_t seg_size = hp->repeated * hp->size; - - void * unmap_va = RTE_PTR_ADD(hp->final_va, - final_size); - int fd; - - munmap(unmap_va, seg_size - final_size); - - fd = open(hp->filepath, O_RDWR); - if (fd < 0) { - RTE_LOG(ERR, EAL, "Cannot open %s: %s\n", - hp->filepath, strerror(errno)); - return -1; - } - if (ftruncate(fd, final_size) < 0) { - RTE_LOG(ERR, EAL, "Cannot truncate %s: %s\n", - hp->filepath, strerror(errno)); - return -1; - } - close(fd); - - pages_found += nr_pg_left; - hp->repeated = nr_pg_left; - } - } -#else - /* else, lock the page and skip */ - else + } else { + /* lock the page and skip */ pages_found++; -#endif + } } /* match page */ } /* foreach page */ @@ -1177,9 +950,6 @@ rte_eal_hugepage_init(void) int i, j, new_memseg; int nr_hugefiles, nr_hugepages = 0; void *addr; -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - int new_pages_count[MAX_HUGEPAGE_SIZES]; -#endif test_proc_pagemap_readable(); @@ -1260,13 +1030,6 @@ rte_eal_hugepage_init(void) pages_old = hpi->num_pages[0]; pages_new = map_all_hugepages(&tmp_hp[hp_offset], hpi, 1); if (pages_new < pages_old) { -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - RTE_LOG(ERR, EAL, - "%d not %d hugepages of size %u MB allocated\n", - pages_new, pages_old, - (unsigned)(hpi->hugepage_sz / 0x100000)); - goto fail; -#else RTE_LOG(DEBUG, EAL, "%d not %d hugepages of size %u MB allocated\n", pages_new, pages_old, @@ -1278,7 +1041,6 @@ rte_eal_hugepage_init(void) hpi->num_pages[0] = pages_new; if (pages_new == 0) continue; -#endif } /* find physical addresses and sockets for each hugepage */ @@ -1297,18 +1059,6 @@ rte_eal_hugepage_init(void) qsort(&tmp_hp[hp_offset], hpi->num_pages[0], sizeof(struct hugepage_file), cmp_physaddr); -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - /* remap all hugepages into single file segments */ - new_pages_count[i] = remap_all_hugepages(&tmp_hp[hp_offset], hpi); - if (new_pages_count[i] < 0){ - RTE_LOG(DEBUG, EAL, "Failed to remap %u MB pages\n", - (unsigned)(hpi->hugepage_sz / 0x100000)); - goto fail; - } - - /* we have processed a num of hugepages of this size, so inc offset */ - hp_offset += new_pages_count[i]; -#else /* remap all hugepages */ if (map_all_hugepages(&tmp_hp[hp_offset], hpi, 0) != hpi->num_pages[0]) { @@ -1323,7 +1073,6 @@ rte_eal_hugepage_init(void) /* we have processed a num of hugepages of this size, so inc offset */ hp_offset += hpi->num_pages[0]; -#endif } huge_recover_sigbus(); @@ -1331,14 +1080,7 @@ rte_eal_hugepage_init(void) if (internal_config.memory == 0 && internal_config.force_sockets == 0) internal_config.memory = eal_get_hugepage_mem_size(); -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - nr_hugefiles = 0; - for (i = 0; i < (int) internal_config.num_hugepage_sizes; i++) { - nr_hugefiles += new_pages_count[i]; - } -#else nr_hugefiles = nr_hugepages; -#endif /* clean out the numbers of pages */ @@ -1356,12 +1098,7 @@ rte_eal_hugepage_init(void) for (j = 0; j < nb_hpsizes; j++) { if (tmp_hp[i].size == internal_config.hugepage_info[j].hugepage_sz) { -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - internal_config.hugepage_info[j].num_pages[socket] += - tmp_hp[i].repeated; -#else internal_config.hugepage_info[j].num_pages[socket]++; -#endif } } } @@ -1436,15 +1173,8 @@ rte_eal_hugepage_init(void) free(tmp_hp); tmp_hp = NULL; - /* find earliest free memseg - this is needed because in case of IVSHMEM, - * segments might have already been initialized */ - for (j = 0; j < RTE_MAX_MEMSEG; j++) - if (mcfg->memseg[j].addr == NULL) { - /* move to previous segment and exit loop */ - j--; - break; - } - + /* first memseg index shall be 0 after incrementing it below */ + j = -1; for (i = 0; i < nr_hugefiles; i++) { new_memseg = 0; @@ -1482,11 +1212,7 @@ rte_eal_hugepage_init(void) mcfg->memseg[j].phys_addr = hugepage[i].physaddr; mcfg->memseg[j].addr = hugepage[i].final_va; -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - mcfg->memseg[j].len = hugepage[i].size * hugepage[i].repeated; -#else mcfg->memseg[j].len = hugepage[i].size; -#endif mcfg->memseg[j].socket_id = hugepage[i].socket_id; mcfg->memseg[j].hugepage_sz = hugepage[i].size; } @@ -1552,7 +1278,8 @@ rte_eal_hugepage_attach(void) struct hugepage_file *hp = NULL; unsigned num_hp = 0; unsigned i, s = 0; /* s used to track the segment number */ - off_t size; + unsigned max_seg = RTE_MAX_MEMSEG; + off_t size = 0; int fd, fd_zero = -1, fd_hugepage = -1; if (aslr_enabled() > 0) { @@ -1597,15 +1324,6 @@ rte_eal_hugepage_attach(void) if (mcfg->memseg[s].len == 0) break; -#ifdef RTE_LIBRTE_IVSHMEM - /* - * if segment has ioremap address set, it's an IVSHMEM segment and - * doesn't need mapping as it was already mapped earlier - */ - if (mcfg->memseg[s].ioremap_addr != 0) - continue; -#endif - /* * fdzero is mmapped to get a contiguous block of virtual * addresses of the appropriate memseg size. @@ -1615,10 +1333,21 @@ rte_eal_hugepage_attach(void) PROT_READ, MAP_PRIVATE, fd_zero, 0); if (base_addr == MAP_FAILED || base_addr != mcfg->memseg[s].addr) { - RTE_LOG(ERR, EAL, "Could not mmap %llu bytes " - "in /dev/zero to requested address [%p]: '%s'\n", - (unsigned long long)mcfg->memseg[s].len, - mcfg->memseg[s].addr, strerror(errno)); + max_seg = s; + if (base_addr != MAP_FAILED) { + /* errno is stale, don't use */ + RTE_LOG(ERR, EAL, "Could not mmap %llu bytes " + "in /dev/zero at [%p], got [%p] - " + "please use '--base-virtaddr' option\n", + (unsigned long long)mcfg->memseg[s].len, + mcfg->memseg[s].addr, base_addr); + munmap(base_addr, mcfg->memseg[s].len); + } else { + RTE_LOG(ERR, EAL, "Could not mmap %llu bytes " + "in /dev/zero at [%p]: '%s'\n", + (unsigned long long)mcfg->memseg[s].len, + mcfg->memseg[s].addr, strerror(errno)); + } if (aslr_enabled() > 0) { RTE_LOG(ERR, EAL, "It is recommended to " "disable ASLR in the kernel " @@ -1644,16 +1373,6 @@ rte_eal_hugepage_attach(void) void *addr, *base_addr; uintptr_t offset = 0; size_t mapping_size; -#ifdef RTE_LIBRTE_IVSHMEM - /* - * if segment has ioremap address set, it's an IVSHMEM segment and - * doesn't need mapping as it was already mapped earlier - */ - if (mcfg->memseg[s].ioremap_addr != 0) { - s++; - continue; - } -#endif /* * free previously mapped memory so we can map the * hugepages into the space @@ -1672,11 +1391,7 @@ rte_eal_hugepage_attach(void) hp[i].filepath); goto error; } -#ifdef RTE_EAL_SINGLE_FILE_SEGMENTS - mapping_size = hp[i].size * hp[i].repeated; -#else mapping_size = hp[i].size; -#endif addr = mmap(RTE_PTR_ADD(base_addr, offset), mapping_size, PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0); @@ -1701,11 +1416,8 @@ rte_eal_hugepage_attach(void) return 0; error: - s = 0; - while (s < RTE_MAX_MEMSEG && mcfg->memseg[s].len > 0) { - munmap(mcfg->memseg[s].addr, mcfg->memseg[s].len); - s++; - } + for (i = 0; i < max_seg && mcfg->memseg[i].len > 0; i++) + munmap(mcfg->memseg[i].addr, mcfg->memseg[i].len); if (hp != NULL && hp != MAP_FAILED) munmap(hp, size); if (fd_zero >= 0) diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci.c index cd9de7cc..e2fc219b 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci.c +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci.c @@ -54,45 +54,6 @@ * IGB_UIO driver (or doesn't initialize, if the device wasn't bound to it). */ -/* unbind kernel driver for this device */ -int -pci_unbind_kernel_driver(struct rte_pci_device *dev) -{ - int n; - FILE *f; - char filename[PATH_MAX]; - char buf[BUFSIZ]; - struct rte_pci_addr *loc = &dev->addr; - - /* open /sys/bus/pci/devices/AAAA:BB:CC.D/driver */ - snprintf(filename, sizeof(filename), - "%s/" PCI_PRI_FMT "/driver/unbind", pci_get_sysfs_path(), - loc->domain, loc->bus, loc->devid, loc->function); - - f = fopen(filename, "w"); - if (f == NULL) /* device was not bound */ - return 0; - - n = snprintf(buf, sizeof(buf), PCI_PRI_FMT "\n", - loc->domain, loc->bus, loc->devid, loc->function); - if ((n < 0) || (n >= (int)sizeof(buf))) { - RTE_LOG(ERR, EAL, "%s(): snprintf failed\n", __func__); - goto error; - } - if (fwrite(buf, n, 1, f) == 0) { - RTE_LOG(ERR, EAL, "%s(): could not write to %s\n", __func__, - filename); - goto error; - } - - fclose(f); - return 0; - -error: - fclose(f); - return -1; -} - static int pci_get_kernel_driver_by_path(const char *filename, char *dri_name) { @@ -267,8 +228,7 @@ error: /* Scan one pci sysfs entry, and fill the devices list from it. */ static int -pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, - uint8_t devid, uint8_t function) +pci_scan_one(const char *dirname, const struct rte_pci_addr *addr) { char filename[PATH_MAX]; unsigned long tmp; @@ -281,10 +241,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, return -1; memset(dev, 0, sizeof(*dev)); - dev->addr.domain = domain; - dev->addr.bus = bus; - dev->addr.devid = devid; - dev->addr.function = function; + dev->addr = *addr; /* get vendor id */ snprintf(filename, sizeof(filename), "%s/vendor", dirname); @@ -350,13 +307,13 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, dirname); if (access(filename, R_OK) != 0) { /* if no NUMA support, set default to 0 */ - dev->numa_node = 0; + dev->device.numa_node = 0; } else { if (eal_parse_sysfs_value(filename, &tmp) < 0) { free(dev); return -1; } - dev->numa_node = tmp; + dev->device.numa_node = tmp; } /* parse resources */ @@ -390,6 +347,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, /* device is valid, add in list (sorted) */ if (TAILQ_EMPTY(&pci_device_list)) { + rte_eal_device_insert(&dev->device); TAILQ_INSERT_TAIL(&pci_device_list, dev, next); } else { struct rte_pci_device *dev2; @@ -402,6 +360,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, if (ret < 0) { TAILQ_INSERT_BEFORE(dev2, dev, next); + rte_eal_device_insert(&dev->device); } else { /* already registered */ dev2->kdrv = dev->kdrv; dev2->max_vfs = dev->max_vfs; @@ -411,18 +370,30 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, } return 0; } + rte_eal_device_insert(&dev->device); TAILQ_INSERT_TAIL(&pci_device_list, dev, next); } return 0; } +int +pci_update_device(const struct rte_pci_addr *addr) +{ + char filename[PATH_MAX]; + + snprintf(filename, sizeof(filename), "%s/" PCI_PRI_FMT, + pci_get_sysfs_path(), addr->domain, addr->bus, addr->devid, + addr->function); + + return pci_scan_one(filename, addr); +} + /* * split up a pci address into its constituent parts. */ static int -parse_pci_addr_format(const char *buf, int bufsize, uint16_t *domain, - uint8_t *bus, uint8_t *devid, uint8_t *function) +parse_pci_addr_format(const char *buf, int bufsize, struct rte_pci_addr *addr) { /* first split on ':' */ union splitaddr { @@ -450,10 +421,10 @@ parse_pci_addr_format(const char *buf, int bufsize, uint16_t *domain, /* now convert to int values */ errno = 0; - *domain = (uint16_t)strtoul(splitaddr.domain, NULL, 16); - *bus = (uint8_t)strtoul(splitaddr.bus, NULL, 16); - *devid = (uint8_t)strtoul(splitaddr.devid, NULL, 16); - *function = (uint8_t)strtoul(splitaddr.function, NULL, 10); + addr->domain = (uint16_t)strtoul(splitaddr.domain, NULL, 16); + addr->bus = (uint8_t)strtoul(splitaddr.bus, NULL, 16); + addr->devid = (uint8_t)strtoul(splitaddr.devid, NULL, 16); + addr->function = (uint8_t)strtoul(splitaddr.function, NULL, 10); if (errno != 0) goto error; @@ -474,8 +445,7 @@ rte_eal_pci_scan(void) struct dirent *e; DIR *dir; char dirname[PATH_MAX]; - uint16_t domain; - uint8_t bus, devid, function; + struct rte_pci_addr addr; dir = opendir(pci_get_sysfs_path()); if (dir == NULL) { @@ -488,13 +458,12 @@ rte_eal_pci_scan(void) if (e->d_name[0] == '.') continue; - if (parse_pci_addr_format(e->d_name, sizeof(e->d_name), &domain, - &bus, &devid, &function) != 0) + if (parse_pci_addr_format(e->d_name, sizeof(e->d_name), &addr) != 0) continue; snprintf(dirname, sizeof(dirname), "%s/%s", pci_get_sysfs_path(), e->d_name); - if (pci_scan_one(dirname, domain, bus, devid, function) < 0) + if (pci_scan_one(dirname, &addr) < 0) goto error; } closedir(dir); @@ -743,9 +712,6 @@ rte_eal_pci_ioport_unmap(struct rte_pci_ioport *p) int rte_eal_pci_init(void) { - TAILQ_INIT(&pci_driver_list); - TAILQ_INIT(&pci_device_list); - /* for debug purposes, PCI can be disabled */ if (internal_config.no_pci) return 0; diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci_uio.c b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci_uio.c index 1786b754..3e4ffb57 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci_uio.c +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/eal_pci_uio.c @@ -133,7 +133,7 @@ pci_mknod_uio_dev(const char *sysfs_uio_path, unsigned uio_num) snprintf(filename, sizeof(filename), "/dev/uio%u", uio_num); dev = makedev(major, minor); ret = mknod(filename, S_IFCHR | S_IRUSR | S_IWUSR, dev); - if (f == NULL) { + if (ret != 0) { RTE_LOG(ERR, EAL, "%s(): mknod() failed %s\n", __func__, strerror(errno)); return -1; diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h b/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h index 3dacbff8..d459bf48 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h @@ -82,6 +82,7 @@ struct rte_epoll_event { /** Handle for interrupts. */ struct rte_intr_handle { + RTE_STD_C11 union { int vfio_dev_fd; /**< VFIO device file descriptor */ int uio_cfg_fd; /**< UIO config file descriptor diff --git a/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h b/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h index 2acdfd9b..09713b0c 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h +++ b/src/dpdk/lib/librte_eal/linuxapp/eal/include/exec-env/rte_kni_common.h @@ -61,6 +61,9 @@ #ifdef __KERNEL__ #include <linux/if.h> +#define RTE_STD_C11 +#else +#include <rte_common.h> #endif /** @@ -85,6 +88,7 @@ enum rte_kni_req_id { */ struct rte_kni_request { uint32_t req_id; /**< Request id */ + RTE_STD_C11 union { uint32_t new_mtu; /**< New MTU */ uint8_t if_up; /**< 1: interface up, 0: interface down */ @@ -102,7 +106,7 @@ struct rte_kni_fifo { volatile unsigned read; /**< Next position to be read */ unsigned len; /**< Circular buffer length */ unsigned elem_size; /**< Pointer size - for 32/64 bit OS */ - void * volatile buffer[0]; /**< The buffer contains mbuf pointers */ + void *volatile buffer[]; /**< The buffer contains mbuf pointers */ }; /* @@ -111,7 +115,8 @@ struct rte_kni_fifo { */ struct rte_kni_mbuf { void *buf_addr __attribute__((__aligned__(RTE_CACHE_LINE_SIZE))); - char pad0[10]; + uint64_t buf_physaddr; + char pad0[2]; uint16_t data_off; /**< Start address of data in segment buffer. */ char pad1[2]; uint8_t nb_segs; /**< Number of segments. */ @@ -159,6 +164,7 @@ struct rte_kni_device_info { uint16_t group_id; /**< Group ID */ uint32_t core_id; /**< core ID to bind for kernel thread */ + __extension__ uint8_t force_bind : 1; /**< Flag for kernel thread binding */ /* mbuf size */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/compat.h b/src/dpdk/lib/librte_eal/linuxapp/kni/compat.h index 647ba3ce..78da08e5 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/compat.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/compat.h @@ -19,13 +19,25 @@ #endif #if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 35) -#define sk_sleep(s) (s)->sk_sleep +#define sk_sleep(s) ((s)->sk_sleep) +#else +#define HAVE_SOCKET_WQ +#endif + +#if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 7, 0) +#define HAVE_STATIC_SOCK_MAP_FD +#else +#define kni_sock_map_fd(s) sock_map_fd(s, 0) #endif #if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 9, 0) #define HAVE_CHANGE_CARRIER_CB #endif +#if LINUX_VERSION_CODE < KERNEL_VERSION(3, 14, 0) +#define ether_addr_copy(dst, src) memcpy(dst, src, ETH_ALEN) +#endif + #if LINUX_VERSION_CODE >= KERNEL_VERSION(3, 19, 0) #define HAVE_IOV_ITER_MSGHDR #endif @@ -35,6 +47,23 @@ #define HAVE_REBUILD_HEADER #endif +#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 2, 0) +#define HAVE_SK_ALLOC_KERN_PARAM +#endif + #if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 7, 0) #define HAVE_TRANS_START_HELPER #endif + +/* + * KNI uses NET_NAME_UNKNOWN macro to select correct version of alloc_netdev() + * For old kernels just backported the commit that enables the macro + * (685343fc3ba6) but still uses old API, it is required to undefine macro to + * select correct version of API, this is safe since KNI doesn't use the value. + * This fix is specific to RedHat/CentOS kernels. + */ +#if (defined(RHEL_RELEASE_CODE) && \ + (RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6, 8)) && \ + (LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 34))) +#undef NET_NAME_UNKNOWN +#endif diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.c index b8c9a13f..d558af20 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.h index 1aec75ab..185ccdf1 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_82575.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.c index 6095d3b4..220c9a40 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.h index b21294ec..55c8a5f4 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_api.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_defines.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_defines.h index 63b228c5..d42c7998 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_defines.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_defines.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_hw.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_hw.h index 347cef71..35886e93 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_hw.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_hw.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.c index 1e9f3e6e..7e4c20a9 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.h index 57b2eb56..b8fa70d0 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_i210.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.c index 4ee59ba9..74319def 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.h index 6a1b0f52..3bcdd88c 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mac.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.c index a1700398..51dfae5d 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.h index c94b2185..0627f271 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_manage.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.c index 3ef0d98b..bd64429f 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.h index bbf838c8..64685d9d 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_mbx.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.c index 6188d007..1ce59154 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.h index fe62785a..17bc53c3 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_nvm.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_osdep.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_osdep.h index d1cf98e2..c1ab60c4 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_osdep.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_osdep.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.c index 140a2a47..d8a77c45 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.h index 5387c5e7..db24fb0b 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_phy.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_regs.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_regs.h index 0e083c54..830ec991 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_regs.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/e1000_regs.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb.h index e5554ca3..d077b49e 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_debugfs.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_debugfs.c deleted file mode 100644 index c07f9f53..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_debugfs.c +++ /dev/null @@ -1,28 +0,0 @@ -/******************************************************************************* - - Intel(R) Gigabit Ethernet Linux driver - Copyright(c) 2007-2013 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - -#include "igb.h" diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ethtool.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ethtool.c index af7e68a5..d7a987d5 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ethtool.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ethtool.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_hwmon.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_hwmon.c deleted file mode 100644 index 07a1ae07..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_hwmon.c +++ /dev/null @@ -1,260 +0,0 @@ -/******************************************************************************* - - Intel(R) Gigabit Ethernet Linux driver - Copyright(c) 2007-2013 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - -#include "igb.h" -#include "e1000_82575.h" -#include "e1000_hw.h" -#ifdef IGB_HWMON -#include <linux/module.h> -#include <linux/types.h> -#include <linux/sysfs.h> -#include <linux/kobject.h> -#include <linux/device.h> -#include <linux/netdevice.h> -#include <linux/hwmon.h> -#include <linux/pci.h> - -#ifdef HAVE_I2C_SUPPORT -static struct i2c_board_info i350_sensor_info = { - I2C_BOARD_INFO("i350bb", (0Xf8 >> 1)), -}; -#endif /* HAVE_I2C_SUPPORT */ - -/* hwmon callback functions */ -static ssize_t igb_hwmon_show_location(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct hwmon_attr *igb_attr = container_of(attr, struct hwmon_attr, - dev_attr); - return sprintf(buf, "loc%u\n", - igb_attr->sensor->location); -} - -static ssize_t igb_hwmon_show_temp(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct hwmon_attr *igb_attr = container_of(attr, struct hwmon_attr, - dev_attr); - unsigned int value; - - /* reset the temp field */ - igb_attr->hw->mac.ops.get_thermal_sensor_data(igb_attr->hw); - - value = igb_attr->sensor->temp; - - /* display millidegree */ - value *= 1000; - - return sprintf(buf, "%u\n", value); -} - -static ssize_t igb_hwmon_show_cautionthresh(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct hwmon_attr *igb_attr = container_of(attr, struct hwmon_attr, - dev_attr); - unsigned int value = igb_attr->sensor->caution_thresh; - - /* display millidegree */ - value *= 1000; - - return sprintf(buf, "%u\n", value); -} - -static ssize_t igb_hwmon_show_maxopthresh(struct device *dev, - struct device_attribute *attr, - char *buf) -{ - struct hwmon_attr *igb_attr = container_of(attr, struct hwmon_attr, - dev_attr); - unsigned int value = igb_attr->sensor->max_op_thresh; - - /* display millidegree */ - value *= 1000; - - return sprintf(buf, "%u\n", value); -} - -/* igb_add_hwmon_attr - Create hwmon attr table for a hwmon sysfs file. - * @ adapter: pointer to the adapter structure - * @ offset: offset in the eeprom sensor data table - * @ type: type of sensor data to display - * - * For each file we want in hwmon's sysfs interface we need a device_attribute - * This is included in our hwmon_attr struct that contains the references to - * the data structures we need to get the data to display. - */ -static int igb_add_hwmon_attr(struct igb_adapter *adapter, - unsigned int offset, int type) { - int rc; - unsigned int n_attr; - struct hwmon_attr *igb_attr; - - n_attr = adapter->igb_hwmon_buff.n_hwmon; - igb_attr = &adapter->igb_hwmon_buff.hwmon_list[n_attr]; - - switch (type) { - case IGB_HWMON_TYPE_LOC: - igb_attr->dev_attr.show = igb_hwmon_show_location; - snprintf(igb_attr->name, sizeof(igb_attr->name), - "temp%u_label", offset); - break; - case IGB_HWMON_TYPE_TEMP: - igb_attr->dev_attr.show = igb_hwmon_show_temp; - snprintf(igb_attr->name, sizeof(igb_attr->name), - "temp%u_input", offset); - break; - case IGB_HWMON_TYPE_CAUTION: - igb_attr->dev_attr.show = igb_hwmon_show_cautionthresh; - snprintf(igb_attr->name, sizeof(igb_attr->name), - "temp%u_max", offset); - break; - case IGB_HWMON_TYPE_MAX: - igb_attr->dev_attr.show = igb_hwmon_show_maxopthresh; - snprintf(igb_attr->name, sizeof(igb_attr->name), - "temp%u_crit", offset); - break; - default: - rc = -EPERM; - return rc; - } - - /* These always the same regardless of type */ - igb_attr->sensor = - &adapter->hw.mac.thermal_sensor_data.sensor[offset]; - igb_attr->hw = &adapter->hw; - igb_attr->dev_attr.store = NULL; - igb_attr->dev_attr.attr.mode = S_IRUGO; - igb_attr->dev_attr.attr.name = igb_attr->name; - sysfs_attr_init(&igb_attr->dev_attr.attr); - rc = device_create_file(&adapter->pdev->dev, - &igb_attr->dev_attr); - if (rc == 0) - ++adapter->igb_hwmon_buff.n_hwmon; - - return rc; -} - -static void igb_sysfs_del_adapter(struct igb_adapter *adapter) -{ - int i; - - if (adapter == NULL) - return; - - for (i = 0; i < adapter->igb_hwmon_buff.n_hwmon; i++) { - device_remove_file(&adapter->pdev->dev, - &adapter->igb_hwmon_buff.hwmon_list[i].dev_attr); - } - - kfree(adapter->igb_hwmon_buff.hwmon_list); - - if (adapter->igb_hwmon_buff.device) - hwmon_device_unregister(adapter->igb_hwmon_buff.device); -} - -/* called from igb_main.c */ -void igb_sysfs_exit(struct igb_adapter *adapter) -{ - igb_sysfs_del_adapter(adapter); -} - -/* called from igb_main.c */ -int igb_sysfs_init(struct igb_adapter *adapter) -{ - struct hwmon_buff *igb_hwmon = &adapter->igb_hwmon_buff; - unsigned int i; - int n_attrs; - int rc = 0; -#ifdef HAVE_I2C_SUPPORT - struct i2c_client *client = NULL; -#endif /* HAVE_I2C_SUPPORT */ - - /* If this method isn't defined we don't support thermals */ - if (adapter->hw.mac.ops.init_thermal_sensor_thresh == NULL) - goto exit; - - /* Don't create thermal hwmon interface if no sensors present */ - rc = (adapter->hw.mac.ops.init_thermal_sensor_thresh(&adapter->hw)); - if (rc) - goto exit; -#ifdef HAVE_I2C_SUPPORT - /* init i2c_client */ - client = i2c_new_device(&adapter->i2c_adap, &i350_sensor_info); - if (client == NULL) { - dev_info(&adapter->pdev->dev, - "Failed to create new i2c device..\n"); - goto exit; - } - adapter->i2c_client = client; -#endif /* HAVE_I2C_SUPPORT */ - - /* Allocation space for max attributes - * max num sensors * values (loc, temp, max, caution) - */ - n_attrs = E1000_MAX_SENSORS * 4; - igb_hwmon->hwmon_list = kcalloc(n_attrs, sizeof(struct hwmon_attr), - GFP_KERNEL); - if (!igb_hwmon->hwmon_list) { - rc = -ENOMEM; - goto err; - } - - igb_hwmon->device = hwmon_device_register(&adapter->pdev->dev); - if (IS_ERR(igb_hwmon->device)) { - rc = PTR_ERR(igb_hwmon->device); - goto err; - } - - for (i = 0; i < E1000_MAX_SENSORS; i++) { - - /* Only create hwmon sysfs entries for sensors that have - * meaningful data. - */ - if (adapter->hw.mac.thermal_sensor_data.sensor[i].location == 0) - continue; - - /* Bail if any hwmon attr struct fails to initialize */ - rc = igb_add_hwmon_attr(adapter, i, IGB_HWMON_TYPE_CAUTION); - rc |= igb_add_hwmon_attr(adapter, i, IGB_HWMON_TYPE_LOC); - rc |= igb_add_hwmon_attr(adapter, i, IGB_HWMON_TYPE_TEMP); - rc |= igb_add_hwmon_attr(adapter, i, IGB_HWMON_TYPE_MAX); - if (rc) - goto err; - } - - goto exit; - -err: - igb_sysfs_del_adapter(adapter); -exit: - return rc; -} -#endif /* IGB_HWMON */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c index 96acec58..f4dca5a3 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> @@ -76,7 +76,7 @@ static const char igb_driver_string[] = static const char igb_copyright[] = "Copyright (c) 2007-2013 Intel Corporation."; -static DEFINE_PCI_DEVICE_TABLE(igb_pci_tbl) = { +const struct pci_device_id igb_pci_tbl[] = { { PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_BACKPLANE_1GBPS) }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_SGMII) }, { PCI_VDEVICE(INTEL, E1000_DEV_ID_I354_BACKPLANE_2_5GBPS) }, @@ -195,7 +195,11 @@ static void igb_process_mdd_event(struct igb_adapter *); #ifdef IFLA_VF_MAX static int igb_ndo_set_vf_mac( struct net_device *netdev, int vf, u8 *mac); static int igb_ndo_set_vf_vlan(struct net_device *netdev, +#ifdef HAVE_VF_VLAN_PROTO + int vf, u16 vlan, u8 qos, __be16 vlan_proto); +#else int vf, u16 vlan, u8 qos); +#endif #ifdef HAVE_VF_SPOOFCHK_CONFIGURE static int igb_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting); @@ -1558,6 +1562,7 @@ static void igb_check_swap_media(struct igb_adapter *adapter) ctrl_ext = E1000_READ_REG(hw, E1000_CTRL_EXT); connsw = E1000_READ_REG(hw, E1000_CONNSW); link = igb_has_link(adapter); + (void) link; /* need to live swap if current media is copper and we have fiber/serdes * to go to. @@ -6411,7 +6416,11 @@ static void igb_set_vmvir(struct igb_adapter *adapter, u32 vid, u32 vf) } static int igb_ndo_set_vf_vlan(struct net_device *netdev, +#ifdef HAVE_VF_VLAN_PROTO + int vf, u16 vlan, u8 qos, __be16 vlan_proto) +#else int vf, u16 vlan, u8 qos) +#endif { int err = 0; struct igb_adapter *adapter = netdev_priv(netdev); @@ -6419,6 +6428,12 @@ static int igb_ndo_set_vf_vlan(struct net_device *netdev, /* VLAN IDs accepted range 0-4094 */ if ((vf >= adapter->vfs_allocated_count) || (vlan > VLAN_VID_MASK-1) || (qos > 7)) return -EINVAL; + +#ifdef HAVE_VF_VLAN_PROTO + if (vlan_proto != htons(ETH_P_8021Q)) + return -EPROTONOSUPPORT; +#endif + if (vlan || qos) { err = igb_vlvf_set(adapter, vlan, !!vlan, vf); if (err) @@ -6579,7 +6594,12 @@ static inline void igb_vf_reset(struct igb_adapter *adapter, u32 vf) if (adapter->vf_data[vf].pf_vlan) igb_ndo_set_vf_vlan(adapter->netdev, vf, adapter->vf_data[vf].pf_vlan, +#ifdef HAVE_VF_VLAN_PROTO + adapter->vf_data[vf].pf_qos, + htons(ETH_P_8021Q)); +#else adapter->vf_data[vf].pf_qos); +#endif else igb_clear_vf_vfta(adapter, vf); #endif diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_param.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_param.c index f79ce7c1..c922ca2f 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_param.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_param.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_procfs.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_procfs.c deleted file mode 100644 index 66236d29..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_procfs.c +++ /dev/null @@ -1,363 +0,0 @@ -/******************************************************************************* - - Intel(R) Gigabit Ethernet Linux driver - Copyright(c) 2007-2013 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - -#include "igb.h" -#include "e1000_82575.h" -#include "e1000_hw.h" - -#ifdef IGB_PROCFS -#ifndef IGB_HWMON - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/proc_fs.h> -#include <linux/device.h> -#include <linux/netdevice.h> - -static struct proc_dir_entry *igb_top_dir = NULL; - - -bool igb_thermal_present(struct igb_adapter *adapter) -{ - s32 status; - struct e1000_hw *hw; - - if (adapter == NULL) - return false; - hw = &adapter->hw; - - /* - * Only set I2C bit-bang mode if an external thermal sensor is - * supported on this device. - */ - if (adapter->ets) { - status = e1000_set_i2c_bb(hw); - if (status != E1000_SUCCESS) - return false; - } - - status = hw->mac.ops.init_thermal_sensor_thresh(hw); - if (status != E1000_SUCCESS) - return false; - - return true; -} - - -static int igb_macburn(char *page, char **start, off_t off, int count, - int *eof, void *data) -{ - struct e1000_hw *hw; - struct igb_adapter *adapter = (struct igb_adapter *)data; - if (adapter == NULL) - return snprintf(page, count, "error: no adapter\n"); - - hw = &adapter->hw; - if (hw == NULL) - return snprintf(page, count, "error: no hw data\n"); - - return snprintf(page, count, "0x%02X%02X%02X%02X%02X%02X\n", - (unsigned int)hw->mac.perm_addr[0], - (unsigned int)hw->mac.perm_addr[1], - (unsigned int)hw->mac.perm_addr[2], - (unsigned int)hw->mac.perm_addr[3], - (unsigned int)hw->mac.perm_addr[4], - (unsigned int)hw->mac.perm_addr[5]); -} - -static int igb_macadmn(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - struct e1000_hw *hw; - struct igb_adapter *adapter = (struct igb_adapter *)data; - if (adapter == NULL) - return snprintf(page, count, "error: no adapter\n"); - - hw = &adapter->hw; - if (hw == NULL) - return snprintf(page, count, "error: no hw data\n"); - - return snprintf(page, count, "0x%02X%02X%02X%02X%02X%02X\n", - (unsigned int)hw->mac.addr[0], - (unsigned int)hw->mac.addr[1], - (unsigned int)hw->mac.addr[2], - (unsigned int)hw->mac.addr[3], - (unsigned int)hw->mac.addr[4], - (unsigned int)hw->mac.addr[5]); -} - -static int igb_numeports(char *page, char **start, off_t off, int count, - int *eof, void *data) -{ - struct e1000_hw *hw; - int ports; - struct igb_adapter *adapter = (struct igb_adapter *)data; - if (adapter == NULL) - return snprintf(page, count, "error: no adapter\n"); - - hw = &adapter->hw; - if (hw == NULL) - return snprintf(page, count, "error: no hw data\n"); - - ports = 4; - - return snprintf(page, count, "%d\n", ports); -} - -static int igb_porttype(char *page, char **start, off_t off, int count, - int *eof, void *data) -{ - struct igb_adapter *adapter = (struct igb_adapter *)data; - if (adapter == NULL) - return snprintf(page, count, "error: no adapter\n"); - - return snprintf(page, count, "%d\n", - test_bit(__IGB_DOWN, &adapter->state)); -} - -static int igb_therm_location(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - struct igb_therm_proc_data *therm_data = - (struct igb_therm_proc_data *)data; - - if (therm_data == NULL) - return snprintf(page, count, "error: no therm_data\n"); - - return snprintf(page, count, "%d\n", therm_data->sensor_data->location); -} - -static int igb_therm_maxopthresh(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - struct igb_therm_proc_data *therm_data = - (struct igb_therm_proc_data *)data; - - if (therm_data == NULL) - return snprintf(page, count, "error: no therm_data\n"); - - return snprintf(page, count, "%d\n", - therm_data->sensor_data->max_op_thresh); -} - -static int igb_therm_cautionthresh(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - struct igb_therm_proc_data *therm_data = - (struct igb_therm_proc_data *)data; - - if (therm_data == NULL) - return snprintf(page, count, "error: no therm_data\n"); - - return snprintf(page, count, "%d\n", - therm_data->sensor_data->caution_thresh); -} - -static int igb_therm_temp(char *page, char **start, off_t off, - int count, int *eof, void *data) -{ - s32 status; - struct igb_therm_proc_data *therm_data = - (struct igb_therm_proc_data *)data; - - if (therm_data == NULL) - return snprintf(page, count, "error: no therm_data\n"); - - status = e1000_get_thermal_sensor_data(therm_data->hw); - if (status != E1000_SUCCESS) - snprintf(page, count, "error: status %d returned\n", status); - - return snprintf(page, count, "%d\n", therm_data->sensor_data->temp); -} - -struct igb_proc_type{ - char name[32]; - int (*read)(char*, char**, off_t, int, int*, void*); -}; - -struct igb_proc_type igb_proc_entries[] = { - {"numeports", &igb_numeports}, - {"porttype", &igb_porttype}, - {"macburn", &igb_macburn}, - {"macadmn", &igb_macadmn}, - {"", NULL} -}; - -struct igb_proc_type igb_internal_entries[] = { - {"location", &igb_therm_location}, - {"temp", &igb_therm_temp}, - {"cautionthresh", &igb_therm_cautionthresh}, - {"maxopthresh", &igb_therm_maxopthresh}, - {"", NULL} -}; - -void igb_del_proc_entries(struct igb_adapter *adapter) -{ - int index, i; - char buf[16]; /* much larger than the sensor number will ever be */ - - if (igb_top_dir == NULL) - return; - - for (i = 0; i < E1000_MAX_SENSORS; i++) { - if (adapter->therm_dir[i] == NULL) - continue; - - for (index = 0; ; index++) { - if (igb_internal_entries[index].read == NULL) - break; - - remove_proc_entry(igb_internal_entries[index].name, - adapter->therm_dir[i]); - } - snprintf(buf, sizeof(buf), "sensor_%d", i); - remove_proc_entry(buf, adapter->info_dir); - } - - if (adapter->info_dir != NULL) { - for (index = 0; ; index++) { - if (igb_proc_entries[index].read == NULL) - break; - remove_proc_entry(igb_proc_entries[index].name, - adapter->info_dir); - } - remove_proc_entry("info", adapter->eth_dir); - } - - if (adapter->eth_dir != NULL) - remove_proc_entry(pci_name(adapter->pdev), igb_top_dir); -} - -/* called from igb_main.c */ -void igb_procfs_exit(struct igb_adapter *adapter) -{ - igb_del_proc_entries(adapter); -} - -int igb_procfs_topdir_init(void) -{ - igb_top_dir = proc_mkdir("driver/igb", NULL); - if (igb_top_dir == NULL) - return -ENOMEM; - - return 0; -} - -void igb_procfs_topdir_exit(void) -{ - remove_proc_entry("driver/igb", NULL); -} - -/* called from igb_main.c */ -int igb_procfs_init(struct igb_adapter *adapter) -{ - int rc = 0; - int i; - int index; - char buf[16]; /* much larger than the sensor number will ever be */ - - adapter->eth_dir = NULL; - adapter->info_dir = NULL; - for (i = 0; i < E1000_MAX_SENSORS; i++) - adapter->therm_dir[i] = NULL; - - if ( igb_top_dir == NULL ) { - rc = -ENOMEM; - goto fail; - } - - adapter->eth_dir = proc_mkdir(pci_name(adapter->pdev), igb_top_dir); - if (adapter->eth_dir == NULL) { - rc = -ENOMEM; - goto fail; - } - - adapter->info_dir = proc_mkdir("info", adapter->eth_dir); - if (adapter->info_dir == NULL) { - rc = -ENOMEM; - goto fail; - } - for (index = 0; ; index++) { - if (igb_proc_entries[index].read == NULL) { - break; - } - if (!(create_proc_read_entry(igb_proc_entries[index].name, - 0444, - adapter->info_dir, - igb_proc_entries[index].read, - adapter))) { - - rc = -ENOMEM; - goto fail; - } - } - if (igb_thermal_present(adapter) == false) - goto exit; - - for (i = 0; i < E1000_MAX_SENSORS; i++) { - - if (adapter->hw.mac.thermal_sensor_data.sensor[i].location== 0) - continue; - - snprintf(buf, sizeof(buf), "sensor_%d", i); - adapter->therm_dir[i] = proc_mkdir(buf, adapter->info_dir); - if (adapter->therm_dir[i] == NULL) { - rc = -ENOMEM; - goto fail; - } - for (index = 0; ; index++) { - if (igb_internal_entries[index].read == NULL) - break; - /* - * therm_data struct contains pointer the read func - * will be needing - */ - adapter->therm_data[i].hw = &adapter->hw; - adapter->therm_data[i].sensor_data = - &adapter->hw.mac.thermal_sensor_data.sensor[i]; - - if (!(create_proc_read_entry( - igb_internal_entries[index].name, - 0444, - adapter->therm_dir[i], - igb_internal_entries[index].read, - &adapter->therm_data[i]))) { - rc = -ENOMEM; - goto fail; - } - } - } - goto exit; - -fail: - igb_del_proc_entries(adapter); -exit: - return rc; -} - -#endif /* !IGB_HWMON */ -#endif /* IGB_PROCFS */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ptp.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ptp.c deleted file mode 100644 index 454b70ce..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_ptp.c +++ /dev/null @@ -1,944 +0,0 @@ -/******************************************************************************* - - Intel(R) Gigabit Ethernet Linux driver - Copyright(c) 2007-2013 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - -/****************************************************************************** - Copyright(c) 2011 Richard Cochran <richardcochran@gmail.com> for some of the - 82576 and 82580 code -******************************************************************************/ - -#include "igb.h" - -#include <linux/module.h> -#include <linux/device.h> -#include <linux/pci.h> -#include <linux/ptp_classify.h> - -#define INCVALUE_MASK 0x7fffffff -#define ISGN 0x80000000 - -/* - * The 82580 timesync updates the system timer every 8ns by 8ns, - * and this update value cannot be reprogrammed. - * - * Neither the 82576 nor the 82580 offer registers wide enough to hold - * nanoseconds time values for very long. For the 82580, SYSTIM always - * counts nanoseconds, but the upper 24 bits are not available. The - * frequency is adjusted by changing the 32 bit fractional nanoseconds - * register, TIMINCA. - * - * For the 82576, the SYSTIM register time unit is affect by the - * choice of the 24 bit TININCA:IV (incvalue) field. Five bits of this - * field are needed to provide the nominal 16 nanosecond period, - * leaving 19 bits for fractional nanoseconds. - * - * We scale the NIC clock cycle by a large factor so that relatively - * small clock corrections can be added or subtracted at each clock - * tick. The drawbacks of a large factor are a) that the clock - * register overflows more quickly (not such a big deal) and b) that - * the increment per tick has to fit into 24 bits. As a result we - * need to use a shift of 19 so we can fit a value of 16 into the - * TIMINCA register. - * - * - * SYSTIMH SYSTIML - * +--------------+ +---+---+------+ - * 82576 | 32 | | 8 | 5 | 19 | - * +--------------+ +---+---+------+ - * \________ 45 bits _______/ fract - * - * +----------+---+ +--------------+ - * 82580 | 24 | 8 | | 32 | - * +----------+---+ +--------------+ - * reserved \______ 40 bits _____/ - * - * - * The 45 bit 82576 SYSTIM overflows every - * 2^45 * 10^-9 / 3600 = 9.77 hours. - * - * The 40 bit 82580 SYSTIM overflows every - * 2^40 * 10^-9 / 60 = 18.3 minutes. - */ - -#define IGB_SYSTIM_OVERFLOW_PERIOD (HZ * 60 * 9) -#define IGB_PTP_TX_TIMEOUT (HZ * 15) -#define INCPERIOD_82576 (1 << E1000_TIMINCA_16NS_SHIFT) -#define INCVALUE_82576_MASK ((1 << E1000_TIMINCA_16NS_SHIFT) - 1) -#define INCVALUE_82576 (16 << IGB_82576_TSYNC_SHIFT) -#define IGB_NBITS_82580 40 - -/* - * SYSTIM read access for the 82576 - */ - -static cycle_t igb_ptp_read_82576(const struct cyclecounter *cc) -{ - struct igb_adapter *igb = container_of(cc, struct igb_adapter, cc); - struct e1000_hw *hw = &igb->hw; - u64 val; - u32 lo, hi; - - lo = E1000_READ_REG(hw, E1000_SYSTIML); - hi = E1000_READ_REG(hw, E1000_SYSTIMH); - - val = ((u64) hi) << 32; - val |= lo; - - return val; -} - -/* - * SYSTIM read access for the 82580 - */ - -static cycle_t igb_ptp_read_82580(const struct cyclecounter *cc) -{ - struct igb_adapter *igb = container_of(cc, struct igb_adapter, cc); - struct e1000_hw *hw = &igb->hw; - u64 val; - u32 lo, hi; - - /* The timestamp latches on lowest register read. For the 82580 - * the lowest register is SYSTIMR instead of SYSTIML. However we only - * need to provide nanosecond resolution, so we just ignore it. - */ - E1000_READ_REG(hw, E1000_SYSTIMR); - lo = E1000_READ_REG(hw, E1000_SYSTIML); - hi = E1000_READ_REG(hw, E1000_SYSTIMH); - - val = ((u64) hi) << 32; - val |= lo; - - return val; -} - -/* - * SYSTIM read access for I210/I211 - */ - -static void igb_ptp_read_i210(struct igb_adapter *adapter, struct timespec *ts) -{ - struct e1000_hw *hw = &adapter->hw; - u32 sec, nsec; - - /* The timestamp latches on lowest register read. For I210/I211, the - * lowest register is SYSTIMR. Since we only need to provide nanosecond - * resolution, we can ignore it. - */ - E1000_READ_REG(hw, E1000_SYSTIMR); - nsec = E1000_READ_REG(hw, E1000_SYSTIML); - sec = E1000_READ_REG(hw, E1000_SYSTIMH); - - ts->tv_sec = sec; - ts->tv_nsec = nsec; -} - -static void igb_ptp_write_i210(struct igb_adapter *adapter, - const struct timespec *ts) -{ - struct e1000_hw *hw = &adapter->hw; - - /* - * Writing the SYSTIMR register is not necessary as it only provides - * sub-nanosecond resolution. - */ - E1000_WRITE_REG(hw, E1000_SYSTIML, ts->tv_nsec); - E1000_WRITE_REG(hw, E1000_SYSTIMH, ts->tv_sec); -} - -/** - * igb_ptp_systim_to_hwtstamp - convert system time value to hw timestamp - * @adapter: board private structure - * @hwtstamps: timestamp structure to update - * @systim: unsigned 64bit system time value. - * - * We need to convert the system time value stored in the RX/TXSTMP registers - * into a hwtstamp which can be used by the upper level timestamping functions. - * - * The 'tmreg_lock' spinlock is used to protect the consistency of the - * system time value. This is needed because reading the 64 bit time - * value involves reading two (or three) 32 bit registers. The first - * read latches the value. Ditto for writing. - * - * In addition, here have extended the system time with an overflow - * counter in software. - **/ -static void igb_ptp_systim_to_hwtstamp(struct igb_adapter *adapter, - struct skb_shared_hwtstamps *hwtstamps, - u64 systim) -{ - unsigned long flags; - u64 ns; - - switch (adapter->hw.mac.type) { - case e1000_82576: - case e1000_82580: - case e1000_i350: - case e1000_i354: - spin_lock_irqsave(&adapter->tmreg_lock, flags); - - ns = timecounter_cyc2time(&adapter->tc, systim); - - spin_unlock_irqrestore(&adapter->tmreg_lock, flags); - - memset(hwtstamps, 0, sizeof(*hwtstamps)); - hwtstamps->hwtstamp = ns_to_ktime(ns); - break; - case e1000_i210: - case e1000_i211: - memset(hwtstamps, 0, sizeof(*hwtstamps)); - /* Upper 32 bits contain s, lower 32 bits contain ns. */ - hwtstamps->hwtstamp = ktime_set(systim >> 32, - systim & 0xFFFFFFFF); - break; - default: - break; - } -} - -/* - * PTP clock operations - */ - -static int igb_ptp_adjfreq_82576(struct ptp_clock_info *ptp, s32 ppb) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - struct e1000_hw *hw = &igb->hw; - int neg_adj = 0; - u64 rate; - u32 incvalue; - - if (ppb < 0) { - neg_adj = 1; - ppb = -ppb; - } - rate = ppb; - rate <<= 14; - rate = div_u64(rate, 1953125); - - incvalue = 16 << IGB_82576_TSYNC_SHIFT; - - if (neg_adj) - incvalue -= rate; - else - incvalue += rate; - - E1000_WRITE_REG(hw, E1000_TIMINCA, INCPERIOD_82576 | (incvalue & INCVALUE_82576_MASK)); - - return 0; -} - -static int igb_ptp_adjfreq_82580(struct ptp_clock_info *ptp, s32 ppb) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - struct e1000_hw *hw = &igb->hw; - int neg_adj = 0; - u64 rate; - u32 inca; - - if (ppb < 0) { - neg_adj = 1; - ppb = -ppb; - } - rate = ppb; - rate <<= 26; - rate = div_u64(rate, 1953125); - - /* At 2.5G speeds, the TIMINCA register on I354 updates the clock 2.5x - * as quickly. Account for this by dividing the adjustment by 2.5. - */ - if (hw->mac.type == e1000_i354) { - u32 status = E1000_READ_REG(hw, E1000_STATUS); - - if ((status & E1000_STATUS_2P5_SKU) && - !(status & E1000_STATUS_2P5_SKU_OVER)) { - rate <<= 1; - rate = div_u64(rate, 5); - } - } - - inca = rate & INCVALUE_MASK; - if (neg_adj) - inca |= ISGN; - - E1000_WRITE_REG(hw, E1000_TIMINCA, inca); - - return 0; -} - -static int igb_ptp_adjtime_82576(struct ptp_clock_info *ptp, s64 delta) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - unsigned long flags; - s64 now; - - spin_lock_irqsave(&igb->tmreg_lock, flags); - - now = timecounter_read(&igb->tc); - now += delta; - timecounter_init(&igb->tc, &igb->cc, now); - - spin_unlock_irqrestore(&igb->tmreg_lock, flags); - - return 0; -} - -static int igb_ptp_adjtime_i210(struct ptp_clock_info *ptp, s64 delta) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - unsigned long flags; - struct timespec now, then = ns_to_timespec(delta); - - spin_lock_irqsave(&igb->tmreg_lock, flags); - - igb_ptp_read_i210(igb, &now); - now = timespec_add(now, then); - igb_ptp_write_i210(igb, (const struct timespec *)&now); - - spin_unlock_irqrestore(&igb->tmreg_lock, flags); - - return 0; -} - -static int igb_ptp_gettime_82576(struct ptp_clock_info *ptp, - struct timespec *ts) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - unsigned long flags; - u64 ns; - u32 remainder; - - spin_lock_irqsave(&igb->tmreg_lock, flags); - - ns = timecounter_read(&igb->tc); - - spin_unlock_irqrestore(&igb->tmreg_lock, flags); - - ts->tv_sec = div_u64_rem(ns, 1000000000, &remainder); - ts->tv_nsec = remainder; - - return 0; -} - -static int igb_ptp_gettime_i210(struct ptp_clock_info *ptp, - struct timespec *ts) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - unsigned long flags; - - spin_lock_irqsave(&igb->tmreg_lock, flags); - - igb_ptp_read_i210(igb, ts); - - spin_unlock_irqrestore(&igb->tmreg_lock, flags); - - return 0; -} - -static int igb_ptp_settime_82576(struct ptp_clock_info *ptp, - const struct timespec *ts) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - unsigned long flags; - u64 ns; - - ns = ts->tv_sec * 1000000000ULL; - ns += ts->tv_nsec; - - spin_lock_irqsave(&igb->tmreg_lock, flags); - - timecounter_init(&igb->tc, &igb->cc, ns); - - spin_unlock_irqrestore(&igb->tmreg_lock, flags); - - return 0; -} - -static int igb_ptp_settime_i210(struct ptp_clock_info *ptp, - const struct timespec *ts) -{ - struct igb_adapter *igb = container_of(ptp, struct igb_adapter, - ptp_caps); - unsigned long flags; - - spin_lock_irqsave(&igb->tmreg_lock, flags); - - igb_ptp_write_i210(igb, ts); - - spin_unlock_irqrestore(&igb->tmreg_lock, flags); - - return 0; -} - -static int igb_ptp_enable(struct ptp_clock_info *ptp, - struct ptp_clock_request *rq, int on) -{ - return -EOPNOTSUPP; -} - -/** - * igb_ptp_tx_work - * @work: pointer to work struct - * - * This work function polls the TSYNCTXCTL valid bit to determine when a - * timestamp has been taken for the current stored skb. - */ -void igb_ptp_tx_work(struct work_struct *work) -{ - struct igb_adapter *adapter = container_of(work, struct igb_adapter, - ptp_tx_work); - struct e1000_hw *hw = &adapter->hw; - u32 tsynctxctl; - - if (!adapter->ptp_tx_skb) - return; - - if (time_is_before_jiffies(adapter->ptp_tx_start + - IGB_PTP_TX_TIMEOUT)) { - dev_kfree_skb_any(adapter->ptp_tx_skb); - adapter->ptp_tx_skb = NULL; - adapter->tx_hwtstamp_timeouts++; - dev_warn(&adapter->pdev->dev, "clearing Tx timestamp hang"); - return; - } - - tsynctxctl = E1000_READ_REG(hw, E1000_TSYNCTXCTL); - if (tsynctxctl & E1000_TSYNCTXCTL_VALID) - igb_ptp_tx_hwtstamp(adapter); - else - /* reschedule to check later */ - schedule_work(&adapter->ptp_tx_work); -} - -static void igb_ptp_overflow_check(struct work_struct *work) -{ - struct igb_adapter *igb = - container_of(work, struct igb_adapter, ptp_overflow_work.work); - struct timespec ts; - - igb->ptp_caps.gettime(&igb->ptp_caps, &ts); - - pr_debug("igb overflow check at %ld.%09lu\n", ts.tv_sec, ts.tv_nsec); - - schedule_delayed_work(&igb->ptp_overflow_work, - IGB_SYSTIM_OVERFLOW_PERIOD); -} - -/** - * igb_ptp_rx_hang - detect error case when Rx timestamp registers latched - * @adapter: private network adapter structure - * - * This watchdog task is scheduled to detect error case where hardware has - * dropped an Rx packet that was timestamped when the ring is full. The - * particular error is rare but leaves the device in a state unable to timestamp - * any future packets. - */ -void igb_ptp_rx_hang(struct igb_adapter *adapter) -{ - struct e1000_hw *hw = &adapter->hw; - struct igb_ring *rx_ring; - u32 tsyncrxctl = E1000_READ_REG(hw, E1000_TSYNCRXCTL); - unsigned long rx_event; - int n; - - if (hw->mac.type != e1000_82576) - return; - - /* If we don't have a valid timestamp in the registers, just update the - * timeout counter and exit - */ - if (!(tsyncrxctl & E1000_TSYNCRXCTL_VALID)) { - adapter->last_rx_ptp_check = jiffies; - return; - } - - /* Determine the most recent watchdog or rx_timestamp event */ - rx_event = adapter->last_rx_ptp_check; - for (n = 0; n < adapter->num_rx_queues; n++) { - rx_ring = adapter->rx_ring[n]; - if (time_after(rx_ring->last_rx_timestamp, rx_event)) - rx_event = rx_ring->last_rx_timestamp; - } - - /* Only need to read the high RXSTMP register to clear the lock */ - if (time_is_before_jiffies(rx_event + 5 * HZ)) { - E1000_READ_REG(hw, E1000_RXSTMPH); - adapter->last_rx_ptp_check = jiffies; - adapter->rx_hwtstamp_cleared++; - dev_warn(&adapter->pdev->dev, "clearing Rx timestamp hang"); - } -} - -/** - * igb_ptp_tx_hwtstamp - utility function which checks for TX time stamp - * @adapter: Board private structure. - * - * If we were asked to do hardware stamping and such a time stamp is - * available, then it must have been for this skb here because we only - * allow only one such packet into the queue. - */ -void igb_ptp_tx_hwtstamp(struct igb_adapter *adapter) -{ - struct e1000_hw *hw = &adapter->hw; - struct skb_shared_hwtstamps shhwtstamps; - u64 regval; - - regval = E1000_READ_REG(hw, E1000_TXSTMPL); - regval |= (u64)E1000_READ_REG(hw, E1000_TXSTMPH) << 32; - - igb_ptp_systim_to_hwtstamp(adapter, &shhwtstamps, regval); - skb_tstamp_tx(adapter->ptp_tx_skb, &shhwtstamps); - dev_kfree_skb_any(adapter->ptp_tx_skb); - adapter->ptp_tx_skb = NULL; -} - -/** - * igb_ptp_rx_pktstamp - retrieve Rx per packet timestamp - * @q_vector: Pointer to interrupt specific structure - * @va: Pointer to address containing Rx buffer - * @skb: Buffer containing timestamp and packet - * - * This function is meant to retrieve a timestamp from the first buffer of an - * incoming frame. The value is stored in little endian format starting on - * byte 8. - */ -void igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, - unsigned char *va, - struct sk_buff *skb) -{ - __le64 *regval = (__le64 *)va; - - /* - * The timestamp is recorded in little endian format. - * DWORD: 0 1 2 3 - * Field: Reserved Reserved SYSTIML SYSTIMH - */ - igb_ptp_systim_to_hwtstamp(q_vector->adapter, skb_hwtstamps(skb), - le64_to_cpu(regval[1])); -} - -/** - * igb_ptp_rx_rgtstamp - retrieve Rx timestamp stored in register - * @q_vector: Pointer to interrupt specific structure - * @skb: Buffer containing timestamp and packet - * - * This function is meant to retrieve a timestamp from the internal registers - * of the adapter and store it in the skb. - */ -void igb_ptp_rx_rgtstamp(struct igb_q_vector *q_vector, - struct sk_buff *skb) -{ - struct igb_adapter *adapter = q_vector->adapter; - struct e1000_hw *hw = &adapter->hw; - u64 regval; - - /* - * If this bit is set, then the RX registers contain the time stamp. No - * other packet will be time stamped until we read these registers, so - * read the registers to make them available again. Because only one - * packet can be time stamped at a time, we know that the register - * values must belong to this one here and therefore we don't need to - * compare any of the additional attributes stored for it. - * - * If nothing went wrong, then it should have a shared tx_flags that we - * can turn into a skb_shared_hwtstamps. - */ - if (!(E1000_READ_REG(hw, E1000_TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID)) - return; - - regval = E1000_READ_REG(hw, E1000_RXSTMPL); - regval |= (u64)E1000_READ_REG(hw, E1000_RXSTMPH) << 32; - - igb_ptp_systim_to_hwtstamp(adapter, skb_hwtstamps(skb), regval); -} - -/** - * igb_ptp_hwtstamp_ioctl - control hardware time stamping - * @netdev: - * @ifreq: - * @cmd: - * - * Outgoing time stamping can be enabled and disabled. Play nice and - * disable it when requested, although it shouldn't case any overhead - * when no packet needs it. At most one packet in the queue may be - * marked for time stamping, otherwise it would be impossible to tell - * for sure to which packet the hardware time stamp belongs. - * - * Incoming time stamping has to be configured via the hardware - * filters. Not all combinations are supported, in particular event - * type has to be specified. Matching the kind of event packet is - * not supported, with the exception of "all V2 events regardless of - * level 2 or 4". - * - **/ -int igb_ptp_hwtstamp_ioctl(struct net_device *netdev, - struct ifreq *ifr, int cmd) -{ - struct igb_adapter *adapter = netdev_priv(netdev); - struct e1000_hw *hw = &adapter->hw; - struct hwtstamp_config config; - u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED; - u32 tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED; - u32 tsync_rx_cfg = 0; - bool is_l4 = false; - bool is_l2 = false; - u32 regval; - - if (copy_from_user(&config, ifr->ifr_data, sizeof(config))) - return -EFAULT; - - /* reserved for future extensions */ - if (config.flags) - return -EINVAL; - - switch (config.tx_type) { - case HWTSTAMP_TX_OFF: - tsync_tx_ctl = 0; - case HWTSTAMP_TX_ON: - break; - default: - return -ERANGE; - } - - switch (config.rx_filter) { - case HWTSTAMP_FILTER_NONE: - tsync_rx_ctl = 0; - break; - case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: - tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L4_V1; - tsync_rx_cfg = E1000_TSYNCRXCFG_PTP_V1_SYNC_MESSAGE; - is_l4 = true; - break; - case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: - tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L4_V1; - tsync_rx_cfg = E1000_TSYNCRXCFG_PTP_V1_DELAY_REQ_MESSAGE; - is_l4 = true; - break; - case HWTSTAMP_FILTER_PTP_V2_EVENT: - case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: - case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: - case HWTSTAMP_FILTER_PTP_V2_SYNC: - case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: - case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: - case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: - case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: - case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: - tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_EVENT_V2; - config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; - is_l2 = true; - is_l4 = true; - break; - case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: - case HWTSTAMP_FILTER_ALL: - /* - * 82576 cannot timestamp all packets, which it needs to do to - * support both V1 Sync and Delay_Req messages - */ - if (hw->mac.type != e1000_82576) { - tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_ALL; - config.rx_filter = HWTSTAMP_FILTER_ALL; - break; - } - /* fall through */ - default: - config.rx_filter = HWTSTAMP_FILTER_NONE; - return -ERANGE; - } - - if (hw->mac.type == e1000_82575) { - if (tsync_rx_ctl | tsync_tx_ctl) - return -EINVAL; - return 0; - } - - /* - * Per-packet timestamping only works if all packets are - * timestamped, so enable timestamping in all packets as - * long as one rx filter was configured. - */ - if ((hw->mac.type >= e1000_82580) && tsync_rx_ctl) { - tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED; - tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_ALL; - config.rx_filter = HWTSTAMP_FILTER_ALL; - is_l2 = true; - is_l4 = true; - - if ((hw->mac.type == e1000_i210) || - (hw->mac.type == e1000_i211)) { - regval = E1000_READ_REG(hw, E1000_RXPBS); - regval |= E1000_RXPBS_CFG_TS_EN; - E1000_WRITE_REG(hw, E1000_RXPBS, regval); - } - } - - /* enable/disable TX */ - regval = E1000_READ_REG(hw, E1000_TSYNCTXCTL); - regval &= ~E1000_TSYNCTXCTL_ENABLED; - regval |= tsync_tx_ctl; - E1000_WRITE_REG(hw, E1000_TSYNCTXCTL, regval); - - /* enable/disable RX */ - regval = E1000_READ_REG(hw, E1000_TSYNCRXCTL); - regval &= ~(E1000_TSYNCRXCTL_ENABLED | E1000_TSYNCRXCTL_TYPE_MASK); - regval |= tsync_rx_ctl; - E1000_WRITE_REG(hw, E1000_TSYNCRXCTL, regval); - - /* define which PTP packets are time stamped */ - E1000_WRITE_REG(hw, E1000_TSYNCRXCFG, tsync_rx_cfg); - - /* define ethertype filter for timestamped packets */ - if (is_l2) - E1000_WRITE_REG(hw, E1000_ETQF(3), - (E1000_ETQF_FILTER_ENABLE | /* enable filter */ - E1000_ETQF_1588 | /* enable timestamping */ - ETH_P_1588)); /* 1588 eth protocol type */ - else - E1000_WRITE_REG(hw, E1000_ETQF(3), 0); - - /* L4 Queue Filter[3]: filter by destination port and protocol */ - if (is_l4) { - u32 ftqf = (IPPROTO_UDP /* UDP */ - | E1000_FTQF_VF_BP /* VF not compared */ - | E1000_FTQF_1588_TIME_STAMP /* Enable Timestamping */ - | E1000_FTQF_MASK); /* mask all inputs */ - ftqf &= ~E1000_FTQF_MASK_PROTO_BP; /* enable protocol check */ - - E1000_WRITE_REG(hw, E1000_IMIR(3), htons(PTP_EV_PORT)); - E1000_WRITE_REG(hw, E1000_IMIREXT(3), - (E1000_IMIREXT_SIZE_BP | E1000_IMIREXT_CTRL_BP)); - if (hw->mac.type == e1000_82576) { - /* enable source port check */ - E1000_WRITE_REG(hw, E1000_SPQF(3), htons(PTP_EV_PORT)); - ftqf &= ~E1000_FTQF_MASK_SOURCE_PORT_BP; - } - E1000_WRITE_REG(hw, E1000_FTQF(3), ftqf); - } else { - E1000_WRITE_REG(hw, E1000_FTQF(3), E1000_FTQF_MASK); - } - E1000_WRITE_FLUSH(hw); - - /* clear TX/RX time stamp registers, just to be sure */ - regval = E1000_READ_REG(hw, E1000_TXSTMPL); - regval = E1000_READ_REG(hw, E1000_TXSTMPH); - regval = E1000_READ_REG(hw, E1000_RXSTMPL); - regval = E1000_READ_REG(hw, E1000_RXSTMPH); - - return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ? - -EFAULT : 0; -} - -void igb_ptp_init(struct igb_adapter *adapter) -{ - struct e1000_hw *hw = &adapter->hw; - struct net_device *netdev = adapter->netdev; - - switch (hw->mac.type) { - case e1000_82576: - snprintf(adapter->ptp_caps.name, 16, "%pm", netdev->dev_addr); - adapter->ptp_caps.owner = THIS_MODULE; - adapter->ptp_caps.max_adj = 999999881; - adapter->ptp_caps.n_ext_ts = 0; - adapter->ptp_caps.pps = 0; - adapter->ptp_caps.adjfreq = igb_ptp_adjfreq_82576; - adapter->ptp_caps.adjtime = igb_ptp_adjtime_82576; - adapter->ptp_caps.gettime = igb_ptp_gettime_82576; - adapter->ptp_caps.settime = igb_ptp_settime_82576; - adapter->ptp_caps.enable = igb_ptp_enable; - adapter->cc.read = igb_ptp_read_82576; - adapter->cc.mask = CLOCKSOURCE_MASK(64); - adapter->cc.mult = 1; - adapter->cc.shift = IGB_82576_TSYNC_SHIFT; - /* Dial the nominal frequency. */ - E1000_WRITE_REG(hw, E1000_TIMINCA, INCPERIOD_82576 | - INCVALUE_82576); - break; - case e1000_82580: - case e1000_i350: - case e1000_i354: - snprintf(adapter->ptp_caps.name, 16, "%pm", netdev->dev_addr); - adapter->ptp_caps.owner = THIS_MODULE; - adapter->ptp_caps.max_adj = 62499999; - adapter->ptp_caps.n_ext_ts = 0; - adapter->ptp_caps.pps = 0; - adapter->ptp_caps.adjfreq = igb_ptp_adjfreq_82580; - adapter->ptp_caps.adjtime = igb_ptp_adjtime_82576; - adapter->ptp_caps.gettime = igb_ptp_gettime_82576; - adapter->ptp_caps.settime = igb_ptp_settime_82576; - adapter->ptp_caps.enable = igb_ptp_enable; - adapter->cc.read = igb_ptp_read_82580; - adapter->cc.mask = CLOCKSOURCE_MASK(IGB_NBITS_82580); - adapter->cc.mult = 1; - adapter->cc.shift = 0; - /* Enable the timer functions by clearing bit 31. */ - E1000_WRITE_REG(hw, E1000_TSAUXC, 0x0); - break; - case e1000_i210: - case e1000_i211: - snprintf(adapter->ptp_caps.name, 16, "%pm", netdev->dev_addr); - adapter->ptp_caps.owner = THIS_MODULE; - adapter->ptp_caps.max_adj = 62499999; - adapter->ptp_caps.n_ext_ts = 0; - adapter->ptp_caps.pps = 0; - adapter->ptp_caps.adjfreq = igb_ptp_adjfreq_82580; - adapter->ptp_caps.adjtime = igb_ptp_adjtime_i210; - adapter->ptp_caps.gettime = igb_ptp_gettime_i210; - adapter->ptp_caps.settime = igb_ptp_settime_i210; - adapter->ptp_caps.enable = igb_ptp_enable; - /* Enable the timer functions by clearing bit 31. */ - E1000_WRITE_REG(hw, E1000_TSAUXC, 0x0); - break; - default: - adapter->ptp_clock = NULL; - return; - } - - E1000_WRITE_FLUSH(hw); - - spin_lock_init(&adapter->tmreg_lock); - INIT_WORK(&adapter->ptp_tx_work, igb_ptp_tx_work); - - /* Initialize the clock and overflow work for devices that need it. */ - if ((hw->mac.type == e1000_i210) || (hw->mac.type == e1000_i211)) { - struct timespec ts = ktime_to_timespec(ktime_get_real()); - - igb_ptp_settime_i210(&adapter->ptp_caps, &ts); - } else { - timecounter_init(&adapter->tc, &adapter->cc, - ktime_to_ns(ktime_get_real())); - - INIT_DELAYED_WORK(&adapter->ptp_overflow_work, - igb_ptp_overflow_check); - - schedule_delayed_work(&adapter->ptp_overflow_work, - IGB_SYSTIM_OVERFLOW_PERIOD); - } - - /* Initialize the time sync interrupts for devices that support it. */ - if (hw->mac.type >= e1000_82580) { - E1000_WRITE_REG(hw, E1000_TSIM, E1000_TSIM_TXTS); - E1000_WRITE_REG(hw, E1000_IMS, E1000_IMS_TS); - } - - adapter->ptp_clock = ptp_clock_register(&adapter->ptp_caps, - &adapter->pdev->dev); - if (IS_ERR(adapter->ptp_clock)) { - adapter->ptp_clock = NULL; - dev_err(&adapter->pdev->dev, "ptp_clock_register failed\n"); - } else { - dev_info(&adapter->pdev->dev, "added PHC on %s\n", - adapter->netdev->name); - adapter->flags |= IGB_FLAG_PTP; - } -} - -/** - * igb_ptp_stop - Disable PTP device and stop the overflow check. - * @adapter: Board private structure. - * - * This function stops the PTP support and cancels the delayed work. - **/ -void igb_ptp_stop(struct igb_adapter *adapter) -{ - switch (adapter->hw.mac.type) { - case e1000_82576: - case e1000_82580: - case e1000_i350: - case e1000_i354: - cancel_delayed_work_sync(&adapter->ptp_overflow_work); - break; - case e1000_i210: - case e1000_i211: - /* No delayed work to cancel. */ - break; - default: - return; - } - - cancel_work_sync(&adapter->ptp_tx_work); - if (adapter->ptp_tx_skb) { - dev_kfree_skb_any(adapter->ptp_tx_skb); - adapter->ptp_tx_skb = NULL; - } - - if (adapter->ptp_clock) { - ptp_clock_unregister(adapter->ptp_clock); - dev_info(&adapter->pdev->dev, "removed PHC on %s\n", - adapter->netdev->name); - adapter->flags &= ~IGB_FLAG_PTP; - } -} - -/** - * igb_ptp_reset - Re-enable the adapter for PTP following a reset. - * @adapter: Board private structure. - * - * This function handles the reset work required to re-enable the PTP device. - **/ -void igb_ptp_reset(struct igb_adapter *adapter) -{ - struct e1000_hw *hw = &adapter->hw; - - if (!(adapter->flags & IGB_FLAG_PTP)) - return; - - switch (adapter->hw.mac.type) { - case e1000_82576: - /* Dial the nominal frequency. */ - E1000_WRITE_REG(hw, E1000_TIMINCA, INCPERIOD_82576 | - INCVALUE_82576); - break; - case e1000_82580: - case e1000_i350: - case e1000_i354: - case e1000_i210: - case e1000_i211: - /* Enable the timer functions and interrupts. */ - E1000_WRITE_REG(hw, E1000_TSAUXC, 0x0); - E1000_WRITE_REG(hw, E1000_TSIM, E1000_TSIM_TXTS); - E1000_WRITE_REG(hw, E1000_IMS, E1000_IMS_TS); - break; - default: - /* No work to do. */ - return; - } - - /* Re-initialize the timer. */ - if ((hw->mac.type == e1000_i210) || (hw->mac.type == e1000_i211)) { - struct timespec ts = ktime_to_timespec(ktime_get_real()); - - igb_ptp_settime_i210(&adapter->ptp_caps, &ts); - } else { - timecounter_init(&adapter->tc, &adapter->cc, - ktime_to_ns(ktime_get_real())); - } -} diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_regtest.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_regtest.h index 18da64a3..9d49b45e 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_regtest.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_regtest.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.c index 015c8952..205da562 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.h index e51e7c4e..c6d4c568 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/igb_vmdq.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.c deleted file mode 100644 index bde3a83c..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.c +++ /dev/null @@ -1,1482 +0,0 @@ -/******************************************************************************* - - Intel(R) Gigabit Ethernet Linux driver - Copyright(c) 2007-2013 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - -#include "igb.h" -#include "kcompat.h" - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,4,8) ) -/* From lib/vsprintf.c */ -#include <asm/div64.h> - -static int skip_atoi(const char **s) -{ - int i=0; - - while (isdigit(**s)) - i = i*10 + *((*s)++) - '0'; - return i; -} - -#define _kc_ZEROPAD 1 /* pad with zero */ -#define _kc_SIGN 2 /* unsigned/signed long */ -#define _kc_PLUS 4 /* show plus */ -#define _kc_SPACE 8 /* space if plus */ -#define _kc_LEFT 16 /* left justified */ -#define _kc_SPECIAL 32 /* 0x */ -#define _kc_LARGE 64 /* use 'ABCDEF' instead of 'abcdef' */ - -static char * number(char * buf, char * end, long long num, int base, int size, int precision, int type) -{ - char c,sign,tmp[66]; - const char *digits; - const char small_digits[] = "0123456789abcdefghijklmnopqrstuvwxyz"; - const char large_digits[] = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ"; - int i; - - digits = (type & _kc_LARGE) ? large_digits : small_digits; - if (type & _kc_LEFT) - type &= ~_kc_ZEROPAD; - if (base < 2 || base > 36) - return 0; - c = (type & _kc_ZEROPAD) ? '0' : ' '; - sign = 0; - if (type & _kc_SIGN) { - if (num < 0) { - sign = '-'; - num = -num; - size--; - } else if (type & _kc_PLUS) { - sign = '+'; - size--; - } else if (type & _kc_SPACE) { - sign = ' '; - size--; - } - } - if (type & _kc_SPECIAL) { - if (base == 16) - size -= 2; - else if (base == 8) - size--; - } - i = 0; - if (num == 0) - tmp[i++]='0'; - else while (num != 0) - tmp[i++] = digits[do_div(num,base)]; - if (i > precision) - precision = i; - size -= precision; - if (!(type&(_kc_ZEROPAD+_kc_LEFT))) { - while(size-->0) { - if (buf <= end) - *buf = ' '; - ++buf; - } - } - if (sign) { - if (buf <= end) - *buf = sign; - ++buf; - } - if (type & _kc_SPECIAL) { - if (base==8) { - if (buf <= end) - *buf = '0'; - ++buf; - } else if (base==16) { - if (buf <= end) - *buf = '0'; - ++buf; - if (buf <= end) - *buf = digits[33]; - ++buf; - } - } - if (!(type & _kc_LEFT)) { - while (size-- > 0) { - if (buf <= end) - *buf = c; - ++buf; - } - } - while (i < precision--) { - if (buf <= end) - *buf = '0'; - ++buf; - } - while (i-- > 0) { - if (buf <= end) - *buf = tmp[i]; - ++buf; - } - while (size-- > 0) { - if (buf <= end) - *buf = ' '; - ++buf; - } - return buf; -} - -int _kc_vsnprintf(char *buf, size_t size, const char *fmt, va_list args) -{ - int len; - unsigned long long num; - int i, base; - char *str, *end, c; - const char *s; - - int flags; /* flags to number() */ - - int field_width; /* width of output field */ - int precision; /* min. # of digits for integers; max - number of chars for from string */ - int qualifier; /* 'h', 'l', or 'L' for integer fields */ - /* 'z' support added 23/7/1999 S.H. */ - /* 'z' changed to 'Z' --davidm 1/25/99 */ - - str = buf; - end = buf + size - 1; - - if (end < buf - 1) { - end = ((void *) -1); - size = end - buf + 1; - } - - for (; *fmt ; ++fmt) { - if (*fmt != '%') { - if (str <= end) - *str = *fmt; - ++str; - continue; - } - - /* process flags */ - flags = 0; - repeat: - ++fmt; /* this also skips first '%' */ - switch (*fmt) { - case '-': flags |= _kc_LEFT; goto repeat; - case '+': flags |= _kc_PLUS; goto repeat; - case ' ': flags |= _kc_SPACE; goto repeat; - case '#': flags |= _kc_SPECIAL; goto repeat; - case '0': flags |= _kc_ZEROPAD; goto repeat; - } - - /* get field width */ - field_width = -1; - if (isdigit(*fmt)) - field_width = skip_atoi(&fmt); - else if (*fmt == '*') { - ++fmt; - /* it's the next argument */ - field_width = va_arg(args, int); - if (field_width < 0) { - field_width = -field_width; - flags |= _kc_LEFT; - } - } - - /* get the precision */ - precision = -1; - if (*fmt == '.') { - ++fmt; - if (isdigit(*fmt)) - precision = skip_atoi(&fmt); - else if (*fmt == '*') { - ++fmt; - /* it's the next argument */ - precision = va_arg(args, int); - } - if (precision < 0) - precision = 0; - } - - /* get the conversion qualifier */ - qualifier = -1; - if (*fmt == 'h' || *fmt == 'l' || *fmt == 'L' || *fmt =='Z') { - qualifier = *fmt; - ++fmt; - } - - /* default base */ - base = 10; - - switch (*fmt) { - case 'c': - if (!(flags & _kc_LEFT)) { - while (--field_width > 0) { - if (str <= end) - *str = ' '; - ++str; - } - } - c = (unsigned char) va_arg(args, int); - if (str <= end) - *str = c; - ++str; - while (--field_width > 0) { - if (str <= end) - *str = ' '; - ++str; - } - continue; - - case 's': - s = va_arg(args, char *); - if (!s) - s = "<NULL>"; - - len = strnlen(s, precision); - - if (!(flags & _kc_LEFT)) { - while (len < field_width--) { - if (str <= end) - *str = ' '; - ++str; - } - } - for (i = 0; i < len; ++i) { - if (str <= end) - *str = *s; - ++str; ++s; - } - while (len < field_width--) { - if (str <= end) - *str = ' '; - ++str; - } - continue; - - case 'p': - if (field_width == -1) { - field_width = 2*sizeof(void *); - flags |= _kc_ZEROPAD; - } - str = number(str, end, - (unsigned long) va_arg(args, void *), - 16, field_width, precision, flags); - continue; - - - case 'n': - /* FIXME: - * What does C99 say about the overflow case here? */ - if (qualifier == 'l') { - long * ip = va_arg(args, long *); - *ip = (str - buf); - } else if (qualifier == 'Z') { - size_t * ip = va_arg(args, size_t *); - *ip = (str - buf); - } else { - int * ip = va_arg(args, int *); - *ip = (str - buf); - } - continue; - - case '%': - if (str <= end) - *str = '%'; - ++str; - continue; - - /* integer number formats - set up the flags and "break" */ - case 'o': - base = 8; - break; - - case 'X': - flags |= _kc_LARGE; - case 'x': - base = 16; - break; - - case 'd': - case 'i': - flags |= _kc_SIGN; - case 'u': - break; - - default: - if (str <= end) - *str = '%'; - ++str; - if (*fmt) { - if (str <= end) - *str = *fmt; - ++str; - } else { - --fmt; - } - continue; - } - if (qualifier == 'L') - num = va_arg(args, long long); - else if (qualifier == 'l') { - num = va_arg(args, unsigned long); - if (flags & _kc_SIGN) - num = (signed long) num; - } else if (qualifier == 'Z') { - num = va_arg(args, size_t); - } else if (qualifier == 'h') { - num = (unsigned short) va_arg(args, int); - if (flags & _kc_SIGN) - num = (signed short) num; - } else { - num = va_arg(args, unsigned int); - if (flags & _kc_SIGN) - num = (signed int) num; - } - str = number(str, end, num, base, - field_width, precision, flags); - } - if (str <= end) - *str = '\0'; - else if (size > 0) - /* don't write out a null byte if the buf size is zero */ - *end = '\0'; - /* the trailing null byte doesn't count towards the total - * ++str; - */ - return str-buf; -} - -int _kc_snprintf(char * buf, size_t size, const char *fmt, ...) -{ - va_list args; - int i; - - va_start(args, fmt); - i = _kc_vsnprintf(buf,size,fmt,args); - va_end(args); - return i; -} -#endif /* < 2.4.8 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,4,13) ) - -/**************************************/ -/* PCI DMA MAPPING */ - -#if defined(CONFIG_HIGHMEM) - -#ifndef PCI_DRAM_OFFSET -#define PCI_DRAM_OFFSET 0 -#endif - -u64 -_kc_pci_map_page(struct pci_dev *dev, struct page *page, unsigned long offset, - size_t size, int direction) -{ - return (((u64) (page - mem_map) << PAGE_SHIFT) + offset + - PCI_DRAM_OFFSET); -} - -#else /* CONFIG_HIGHMEM */ - -u64 -_kc_pci_map_page(struct pci_dev *dev, struct page *page, unsigned long offset, - size_t size, int direction) -{ - return pci_map_single(dev, (void *)page_address(page) + offset, size, - direction); -} - -#endif /* CONFIG_HIGHMEM */ - -void -_kc_pci_unmap_page(struct pci_dev *dev, u64 dma_addr, size_t size, - int direction) -{ - return pci_unmap_single(dev, dma_addr, size, direction); -} - -#endif /* 2.4.13 => 2.4.3 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,4,3) ) - -/**************************************/ -/* PCI DRIVER API */ - -int -_kc_pci_set_dma_mask(struct pci_dev *dev, dma_addr_t mask) -{ - if (!pci_dma_supported(dev, mask)) - return -EIO; - dev->dma_mask = mask; - return 0; -} - -int -_kc_pci_request_regions(struct pci_dev *dev, char *res_name) -{ - int i; - - for (i = 0; i < 6; i++) { - if (pci_resource_len(dev, i) == 0) - continue; - - if (pci_resource_flags(dev, i) & IORESOURCE_IO) { - if (!request_region(pci_resource_start(dev, i), pci_resource_len(dev, i), res_name)) { - pci_release_regions(dev); - return -EBUSY; - } - } else if (pci_resource_flags(dev, i) & IORESOURCE_MEM) { - if (!request_mem_region(pci_resource_start(dev, i), pci_resource_len(dev, i), res_name)) { - pci_release_regions(dev); - return -EBUSY; - } - } - } - return 0; -} - -void -_kc_pci_release_regions(struct pci_dev *dev) -{ - int i; - - for (i = 0; i < 6; i++) { - if (pci_resource_len(dev, i) == 0) - continue; - - if (pci_resource_flags(dev, i) & IORESOURCE_IO) - release_region(pci_resource_start(dev, i), pci_resource_len(dev, i)); - - else if (pci_resource_flags(dev, i) & IORESOURCE_MEM) - release_mem_region(pci_resource_start(dev, i), pci_resource_len(dev, i)); - } -} - -/**************************************/ -/* NETWORK DRIVER API */ - -struct net_device * -_kc_alloc_etherdev(int sizeof_priv) -{ - struct net_device *dev; - int alloc_size; - - alloc_size = sizeof(*dev) + sizeof_priv + IFNAMSIZ + 31; - dev = kzalloc(alloc_size, GFP_KERNEL); - if (!dev) - return NULL; - - if (sizeof_priv) - dev->priv = (void *) (((unsigned long)(dev + 1) + 31) & ~31); - dev->name[0] = '\0'; - ether_setup(dev); - - return dev; -} - -int -_kc_is_valid_ether_addr(u8 *addr) -{ - const char zaddr[6] = { 0, }; - - return !(addr[0] & 1) && memcmp(addr, zaddr, 6); -} - -#endif /* 2.4.3 => 2.4.0 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,4,6) ) - -int -_kc_pci_set_power_state(struct pci_dev *dev, int state) -{ - return 0; -} - -int -_kc_pci_enable_wake(struct pci_dev *pdev, u32 state, int enable) -{ - return 0; -} - -#endif /* 2.4.6 => 2.4.3 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0) ) -void _kc_skb_fill_page_desc(struct sk_buff *skb, int i, struct page *page, - int off, int size) -{ - skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; - frag->page = page; - frag->page_offset = off; - frag->size = size; - skb_shinfo(skb)->nr_frags = i + 1; -} - -/* - * Original Copyright: - * find_next_bit.c: fallback find next bit implementation - * - * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - */ - -/** - * find_next_bit - find the next set bit in a memory region - * @addr: The address to base the search on - * @offset: The bitnumber to start searching at - * @size: The maximum size to search - */ -unsigned long find_next_bit(const unsigned long *addr, unsigned long size, - unsigned long offset) -{ - const unsigned long *p = addr + BITOP_WORD(offset); - unsigned long result = offset & ~(BITS_PER_LONG-1); - unsigned long tmp; - - if (offset >= size) - return size; - size -= result; - offset %= BITS_PER_LONG; - if (offset) { - tmp = *(p++); - tmp &= (~0UL << offset); - if (size < BITS_PER_LONG) - goto found_first; - if (tmp) - goto found_middle; - size -= BITS_PER_LONG; - result += BITS_PER_LONG; - } - while (size & ~(BITS_PER_LONG-1)) { - if ((tmp = *(p++))) - goto found_middle; - result += BITS_PER_LONG; - size -= BITS_PER_LONG; - } - if (!size) - return result; - tmp = *p; - -found_first: - tmp &= (~0UL >> (BITS_PER_LONG - size)); - if (tmp == 0UL) /* Are any bits set? */ - return result + size; /* Nope. */ -found_middle: - return result + ffs(tmp); -} - -size_t _kc_strlcpy(char *dest, const char *src, size_t size) -{ - size_t ret = strlen(src); - - if (size) { - size_t len = (ret >= size) ? size - 1 : ret; - memcpy(dest, src, len); - dest[len] = '\0'; - } - return ret; -} - -#ifndef do_div -#if BITS_PER_LONG == 32 -uint32_t __attribute__((weak)) _kc__div64_32(uint64_t *n, uint32_t base) -{ - uint64_t rem = *n; - uint64_t b = base; - uint64_t res, d = 1; - uint32_t high = rem >> 32; - - /* Reduce the thing a bit first */ - res = 0; - if (high >= base) { - high /= base; - res = (uint64_t) high << 32; - rem -= (uint64_t) (high*base) << 32; - } - - while ((int64_t)b > 0 && b < rem) { - b = b+b; - d = d+d; - } - - do { - if (rem >= b) { - rem -= b; - res += d; - } - b >>= 1; - d >>= 1; - } while (d); - - *n = res; - return rem; -} -#endif /* BITS_PER_LONG == 32 */ -#endif /* do_div */ -#endif /* 2.6.0 => 2.4.6 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,4) ) -int _kc_scnprintf(char * buf, size_t size, const char *fmt, ...) -{ - va_list args; - int i; - - va_start(args, fmt); - i = vsnprintf(buf, size, fmt, args); - va_end(args); - return (i >= size) ? (size - 1) : i; -} -#endif /* < 2.6.4 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,10) ) -DECLARE_BITMAP(_kcompat_node_online_map, MAX_NUMNODES) = {1}; -#endif /* < 2.6.10 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,13) ) -char *_kc_kstrdup(const char *s, unsigned int gfp) -{ - size_t len; - char *buf; - - if (!s) - return NULL; - - len = strlen(s) + 1; - buf = kmalloc(len, gfp); - if (buf) - memcpy(buf, s, len); - return buf; -} -#endif /* < 2.6.13 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,14) ) -void *_kc_kzalloc(size_t size, int flags) -{ - void *ret = kmalloc(size, flags); - if (ret) - memset(ret, 0, size); - return ret; -} -#endif /* <= 2.6.13 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19) ) -int _kc_skb_pad(struct sk_buff *skb, int pad) -{ - int ntail; - - /* If the skbuff is non linear tailroom is always zero.. */ - if(!skb_cloned(skb) && skb_tailroom(skb) >= pad) { - memset(skb->data+skb->len, 0, pad); - return 0; - } - - ntail = skb->data_len + pad - (skb->end - skb->tail); - if (likely(skb_cloned(skb) || ntail > 0)) { - if (pskb_expand_head(skb, 0, ntail, GFP_ATOMIC)); - goto free_skb; - } - -#ifdef MAX_SKB_FRAGS - if (skb_is_nonlinear(skb) && - !__pskb_pull_tail(skb, skb->data_len)) - goto free_skb; - -#endif - memset(skb->data + skb->len, 0, pad); - return 0; - -free_skb: - kfree_skb(skb); - return -ENOMEM; -} - -#if (!(RHEL_RELEASE_CODE && RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(5,4))) -int _kc_pci_save_state(struct pci_dev *pdev) -{ - struct net_device *netdev = pci_get_drvdata(pdev); - struct adapter_struct *adapter = netdev_priv(netdev); - int size = PCI_CONFIG_SPACE_LEN, i; - u16 pcie_cap_offset, pcie_link_status; - -#if ( LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0) ) - /* no ->dev for 2.4 kernels */ - WARN_ON(pdev->dev.driver_data == NULL); -#endif - pcie_cap_offset = pci_find_capability(pdev, PCI_CAP_ID_EXP); - if (pcie_cap_offset) { - if (!pci_read_config_word(pdev, - pcie_cap_offset + PCIE_LINK_STATUS, - &pcie_link_status)) - size = PCIE_CONFIG_SPACE_LEN; - } - pci_config_space_ich8lan(); -#ifdef HAVE_PCI_ERS - if (adapter->config_space == NULL) -#else - WARN_ON(adapter->config_space != NULL); -#endif - adapter->config_space = kmalloc(size, GFP_KERNEL); - if (!adapter->config_space) { - printk(KERN_ERR "Out of memory in pci_save_state\n"); - return -ENOMEM; - } - for (i = 0; i < (size / 4); i++) - pci_read_config_dword(pdev, i * 4, &adapter->config_space[i]); - return 0; -} - -void _kc_pci_restore_state(struct pci_dev *pdev) -{ - struct net_device *netdev = pci_get_drvdata(pdev); - struct adapter_struct *adapter = netdev_priv(netdev); - int size = PCI_CONFIG_SPACE_LEN, i; - u16 pcie_cap_offset; - u16 pcie_link_status; - - if (adapter->config_space != NULL) { - pcie_cap_offset = pci_find_capability(pdev, PCI_CAP_ID_EXP); - if (pcie_cap_offset && - !pci_read_config_word(pdev, - pcie_cap_offset + PCIE_LINK_STATUS, - &pcie_link_status)) - size = PCIE_CONFIG_SPACE_LEN; - - pci_config_space_ich8lan(); - for (i = 0; i < (size / 4); i++) - pci_write_config_dword(pdev, i * 4, adapter->config_space[i]); -#ifndef HAVE_PCI_ERS - kfree(adapter->config_space); - adapter->config_space = NULL; -#endif - } -} -#endif /* !(RHEL_RELEASE_CODE >= RHEL 5.4) */ - -#ifdef HAVE_PCI_ERS -void _kc_free_netdev(struct net_device *netdev) -{ - struct adapter_struct *adapter = netdev_priv(netdev); - - if (adapter->config_space != NULL) - kfree(adapter->config_space); -#ifdef CONFIG_SYSFS - if (netdev->reg_state == NETREG_UNINITIALIZED) { - kfree((char *)netdev - netdev->padded); - } else { - BUG_ON(netdev->reg_state != NETREG_UNREGISTERED); - netdev->reg_state = NETREG_RELEASED; - class_device_put(&netdev->class_dev); - } -#else - kfree((char *)netdev - netdev->padded); -#endif -} -#endif - -void *_kc_kmemdup(const void *src, size_t len, unsigned gfp) -{ - void *p; - - p = kzalloc(len, gfp); - if (p) - memcpy(p, src, len); - return p; -} -#endif /* <= 2.6.19 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,21) ) -struct pci_dev *_kc_netdev_to_pdev(struct net_device *netdev) -{ - return ((struct adapter_struct *)netdev_priv(netdev))->pdev; -} -#endif /* < 2.6.21 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,22) ) -/* hexdump code taken from lib/hexdump.c */ -static void _kc_hex_dump_to_buffer(const void *buf, size_t len, int rowsize, - int groupsize, unsigned char *linebuf, - size_t linebuflen, bool ascii) -{ - const u8 *ptr = buf; - u8 ch; - int j, lx = 0; - int ascii_column; - - if (rowsize != 16 && rowsize != 32) - rowsize = 16; - - if (!len) - goto nil; - if (len > rowsize) /* limit to one line at a time */ - len = rowsize; - if ((len % groupsize) != 0) /* no mixed size output */ - groupsize = 1; - - switch (groupsize) { - case 8: { - const u64 *ptr8 = buf; - int ngroups = len / groupsize; - - for (j = 0; j < ngroups; j++) - lx += scnprintf((char *)(linebuf + lx), linebuflen - lx, - "%s%16.16llx", j ? " " : "", - (unsigned long long)*(ptr8 + j)); - ascii_column = 17 * ngroups + 2; - break; - } - - case 4: { - const u32 *ptr4 = buf; - int ngroups = len / groupsize; - - for (j = 0; j < ngroups; j++) - lx += scnprintf((char *)(linebuf + lx), linebuflen - lx, - "%s%8.8x", j ? " " : "", *(ptr4 + j)); - ascii_column = 9 * ngroups + 2; - break; - } - - case 2: { - const u16 *ptr2 = buf; - int ngroups = len / groupsize; - - for (j = 0; j < ngroups; j++) - lx += scnprintf((char *)(linebuf + lx), linebuflen - lx, - "%s%4.4x", j ? " " : "", *(ptr2 + j)); - ascii_column = 5 * ngroups + 2; - break; - } - - default: - for (j = 0; (j < len) && (lx + 3) <= linebuflen; j++) { - ch = ptr[j]; - linebuf[lx++] = hex_asc(ch >> 4); - linebuf[lx++] = hex_asc(ch & 0x0f); - linebuf[lx++] = ' '; - } - if (j) - lx--; - - ascii_column = 3 * rowsize + 2; - break; - } - if (!ascii) - goto nil; - - while (lx < (linebuflen - 1) && lx < (ascii_column - 1)) - linebuf[lx++] = ' '; - for (j = 0; (j < len) && (lx + 2) < linebuflen; j++) - linebuf[lx++] = (isascii(ptr[j]) && isprint(ptr[j])) ? ptr[j] - : '.'; -nil: - linebuf[lx++] = '\0'; -} - -void _kc_print_hex_dump(const char *level, - const char *prefix_str, int prefix_type, - int rowsize, int groupsize, - const void *buf, size_t len, bool ascii) -{ - const u8 *ptr = buf; - int i, linelen, remaining = len; - unsigned char linebuf[200]; - - if (rowsize != 16 && rowsize != 32) - rowsize = 16; - - for (i = 0; i < len; i += rowsize) { - linelen = min(remaining, rowsize); - remaining -= rowsize; - _kc_hex_dump_to_buffer(ptr + i, linelen, rowsize, groupsize, - linebuf, sizeof(linebuf), ascii); - - switch (prefix_type) { - case DUMP_PREFIX_ADDRESS: - printk("%s%s%*p: %s\n", level, prefix_str, - (int)(2 * sizeof(void *)), ptr + i, linebuf); - break; - case DUMP_PREFIX_OFFSET: - printk("%s%s%.8x: %s\n", level, prefix_str, i, linebuf); - break; - default: - printk("%s%s%s\n", level, prefix_str, linebuf); - break; - } - } -} - -#ifdef HAVE_I2C_SUPPORT -struct i2c_client * -_kc_i2c_new_device(struct i2c_adapter *adap, struct i2c_board_info const *info) -{ - struct i2c_client *client; - int status; - - client = kzalloc(sizeof *client, GFP_KERNEL); - if (!client) - return NULL; - - client->adapter = adap; - - client->dev.platform_data = info->platform_data; - - client->flags = info->flags; - client->addr = info->addr; - - strlcpy(client->name, info->type, sizeof(client->name)); - - /* Check for address business */ - status = i2c_check_addr(adap, client->addr); - if (status) - goto out_err; - - client->dev.parent = &client->adapter->dev; - client->dev.bus = &i2c_bus_type; - - status = i2c_attach_client(client); - if (status) - goto out_err; - - dev_dbg(&adap->dev, "client [%s] registered with bus id %s\n", - client->name, dev_name(&client->dev)); - - return client; - -out_err: - dev_err(&adap->dev, "Failed to register i2c client %s at 0x%02x " - "(%d)\n", client->name, client->addr, status); - kfree(client); - return NULL; -} -#endif /* HAVE_I2C_SUPPORT */ -#endif /* < 2.6.22 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,24) ) -#ifdef NAPI -struct net_device *napi_to_poll_dev(const struct napi_struct *napi) -{ - struct adapter_q_vector *q_vector = container_of(napi, - struct adapter_q_vector, - napi); - return &q_vector->poll_dev; -} - -int __kc_adapter_clean(struct net_device *netdev, int *budget) -{ - int work_done; - int work_to_do = min(*budget, netdev->quota); - /* kcompat.h netif_napi_add puts napi struct in "fake netdev->priv" */ - struct napi_struct *napi = netdev->priv; - work_done = napi->poll(napi, work_to_do); - *budget -= work_done; - netdev->quota -= work_done; - return (work_done >= work_to_do) ? 1 : 0; -} -#endif /* NAPI */ -#endif /* <= 2.6.24 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,26) ) -void _kc_pci_disable_link_state(struct pci_dev *pdev, int state) -{ - struct pci_dev *parent = pdev->bus->self; - u16 link_state; - int pos; - - if (!parent) - return; - - pos = pci_find_capability(parent, PCI_CAP_ID_EXP); - if (pos) { - pci_read_config_word(parent, pos + PCI_EXP_LNKCTL, &link_state); - link_state &= ~state; - pci_write_config_word(parent, pos + PCI_EXP_LNKCTL, link_state); - } -} -#endif /* < 2.6.26 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,27) ) -#ifdef HAVE_TX_MQ -void _kc_netif_tx_stop_all_queues(struct net_device *netdev) -{ - struct adapter_struct *adapter = netdev_priv(netdev); - int i; - - netif_stop_queue(netdev); - if (netif_is_multiqueue(netdev)) - for (i = 0; i < adapter->num_tx_queues; i++) - netif_stop_subqueue(netdev, i); -} -void _kc_netif_tx_wake_all_queues(struct net_device *netdev) -{ - struct adapter_struct *adapter = netdev_priv(netdev); - int i; - - netif_wake_queue(netdev); - if (netif_is_multiqueue(netdev)) - for (i = 0; i < adapter->num_tx_queues; i++) - netif_wake_subqueue(netdev, i); -} -void _kc_netif_tx_start_all_queues(struct net_device *netdev) -{ - struct adapter_struct *adapter = netdev_priv(netdev); - int i; - - netif_start_queue(netdev); - if (netif_is_multiqueue(netdev)) - for (i = 0; i < adapter->num_tx_queues; i++) - netif_start_subqueue(netdev, i); -} -#endif /* HAVE_TX_MQ */ - -#ifndef __WARN_printf -void __kc_warn_slowpath(const char *file, int line, const char *fmt, ...) -{ - va_list args; - - printk(KERN_WARNING "------------[ cut here ]------------\n"); - printk(KERN_WARNING "WARNING: at %s:%d %s()\n", file, line); - va_start(args, fmt); - vprintk(fmt, args); - va_end(args); - - dump_stack(); -} -#endif /* __WARN_printf */ -#endif /* < 2.6.27 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,28) ) - -int -_kc_pci_prepare_to_sleep(struct pci_dev *dev) -{ - pci_power_t target_state; - int error; - - target_state = pci_choose_state(dev, PMSG_SUSPEND); - - pci_enable_wake(dev, target_state, true); - - error = pci_set_power_state(dev, target_state); - - if (error) - pci_enable_wake(dev, target_state, false); - - return error; -} - -int -_kc_pci_wake_from_d3(struct pci_dev *dev, bool enable) -{ - int err; - - err = pci_enable_wake(dev, PCI_D3cold, enable); - if (err) - goto out; - - err = pci_enable_wake(dev, PCI_D3hot, enable); - -out: - return err; -} -#endif /* < 2.6.28 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,29) ) -static void __kc_pci_set_master(struct pci_dev *pdev, bool enable) -{ - u16 old_cmd, cmd; - - pci_read_config_word(pdev, PCI_COMMAND, &old_cmd); - if (enable) - cmd = old_cmd | PCI_COMMAND_MASTER; - else - cmd = old_cmd & ~PCI_COMMAND_MASTER; - if (cmd != old_cmd) { - dev_dbg(pci_dev_to_dev(pdev), "%s bus mastering\n", - enable ? "enabling" : "disabling"); - pci_write_config_word(pdev, PCI_COMMAND, cmd); - } -#if ( LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,7) ) - pdev->is_busmaster = enable; -#endif -} - -void _kc_pci_clear_master(struct pci_dev *dev) -{ - __kc_pci_set_master(dev, false); -} -#endif /* < 2.6.29 */ - -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,34) ) -#if (RHEL_RELEASE_CODE < RHEL_RELEASE_VERSION(6,0)) -int _kc_pci_num_vf(struct pci_dev *dev) -{ - int num_vf = 0; -#ifdef CONFIG_PCI_IOV - struct pci_dev *vfdev; - - /* loop through all ethernet devices starting at PF dev */ - vfdev = pci_get_class(PCI_CLASS_NETWORK_ETHERNET << 8, NULL); - while (vfdev) { - if (vfdev->is_virtfn && vfdev->physfn == dev) - num_vf++; - - vfdev = pci_get_class(PCI_CLASS_NETWORK_ETHERNET << 8, vfdev); - } - -#endif - return num_vf; -} -#endif /* RHEL_RELEASE_CODE */ -#endif /* < 2.6.34 */ - -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,35) ) -#ifdef HAVE_TX_MQ -#if (!(RHEL_RELEASE_CODE && RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6,0))) -#ifndef CONFIG_NETDEVICES_MULTIQUEUE -void _kc_netif_set_real_num_tx_queues(struct net_device *dev, unsigned int txq) -{ - unsigned int real_num = dev->real_num_tx_queues; - struct Qdisc *qdisc; - int i; - - if (unlikely(txq > dev->num_tx_queues)) - ; - else if (txq > real_num) - dev->real_num_tx_queues = txq; - else if ( txq < real_num) { - dev->real_num_tx_queues = txq; - for (i = txq; i < dev->num_tx_queues; i++) { - qdisc = netdev_get_tx_queue(dev, i)->qdisc; - if (qdisc) { - spin_lock_bh(qdisc_lock(qdisc)); - qdisc_reset(qdisc); - spin_unlock_bh(qdisc_lock(qdisc)); - } - } - } -} -#endif /* CONFIG_NETDEVICES_MULTIQUEUE */ -#endif /* !(RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6,0)) */ -#endif /* HAVE_TX_MQ */ - -ssize_t _kc_simple_write_to_buffer(void *to, size_t available, loff_t *ppos, - const void __user *from, size_t count) -{ - loff_t pos = *ppos; - size_t res; - - if (pos < 0) - return -EINVAL; - if (pos >= available || !count) - return 0; - if (count > available - pos) - count = available - pos; - res = copy_from_user(to + pos, from, count); - if (res == count) - return -EFAULT; - count -= res; - *ppos = pos + count; - return count; -} - -#endif /* < 2.6.35 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,36) ) -static const u32 _kc_flags_dup_features = - (ETH_FLAG_LRO | ETH_FLAG_NTUPLE | ETH_FLAG_RXHASH); - -u32 _kc_ethtool_op_get_flags(struct net_device *dev) -{ - return dev->features & _kc_flags_dup_features; -} - -int _kc_ethtool_op_set_flags(struct net_device *dev, u32 data, u32 supported) -{ - if (data & ~supported) - return -EINVAL; - - dev->features = ((dev->features & ~_kc_flags_dup_features) | - (data & _kc_flags_dup_features)); - return 0; -} -#endif /* < 2.6.36 */ - -/******************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(2,6,39) ) -#if (!(RHEL_RELEASE_CODE && RHEL_RELEASE_CODE > RHEL_RELEASE_VERSION(6,0))) - - - -#endif /* !(RHEL_RELEASE_CODE > RHEL_RELEASE_VERSION(6,0)) */ -#endif /* < 2.6.39 */ - -/******************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(3,4,0) ) -void _kc_skb_add_rx_frag(struct sk_buff *skb, int i, struct page *page, - int off, int size, unsigned int truesize) -{ - skb_fill_page_desc(skb, i, page, off, size); - skb->len += size; - skb->data_len += size; - skb->truesize += truesize; -} - -int _kc_simple_open(struct inode *inode, struct file *file) -{ - if (inode->i_private) - file->private_data = inode->i_private; - - return 0; -} - -#endif /* < 3.4.0 */ - -/******************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(3,7,0) ) -#if !(SLE_VERSION_CODE && SLE_VERSION_CODE >= SLE_VERSION(11,3,0)) && \ - !(RHEL_RELEASE_CODE && RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6,5)) -static inline int __kc_pcie_cap_version(struct pci_dev *dev) -{ - int pos; - u16 reg16; - - pos = pci_find_capability(dev, PCI_CAP_ID_EXP); - if (!pos) - return 0; - pci_read_config_word(dev, pos + PCI_EXP_FLAGS, ®16); - return reg16 & PCI_EXP_FLAGS_VERS; -} - -static inline bool __kc_pcie_cap_has_devctl(const struct pci_dev __always_unused *dev) -{ - return true; -} - -static inline bool __kc_pcie_cap_has_lnkctl(struct pci_dev *dev) -{ - int type = pci_pcie_type(dev); - - return __kc_pcie_cap_version(dev) > 1 || - type == PCI_EXP_TYPE_ROOT_PORT || - type == PCI_EXP_TYPE_ENDPOINT || - type == PCI_EXP_TYPE_LEG_END; -} - -static inline bool __kc_pcie_cap_has_sltctl(struct pci_dev *dev) -{ - int type = pci_pcie_type(dev); - int pos; - u16 pcie_flags_reg; - - pos = pci_find_capability(dev, PCI_CAP_ID_EXP); - if (!pos) - return 0; - pci_read_config_word(dev, pos + PCI_EXP_FLAGS, &pcie_flags_reg); - - return __kc_pcie_cap_version(dev) > 1 || - type == PCI_EXP_TYPE_ROOT_PORT || - (type == PCI_EXP_TYPE_DOWNSTREAM && - pcie_flags_reg & PCI_EXP_FLAGS_SLOT); -} - -static inline bool __kc_pcie_cap_has_rtctl(struct pci_dev *dev) -{ - int type = pci_pcie_type(dev); - - return __kc_pcie_cap_version(dev) > 1 || - type == PCI_EXP_TYPE_ROOT_PORT || - type == PCI_EXP_TYPE_RC_EC; -} - -static bool __kc_pcie_capability_reg_implemented(struct pci_dev *dev, int pos) -{ - if (!pci_is_pcie(dev)) - return false; - - switch (pos) { - case PCI_EXP_FLAGS_TYPE: - return true; - case PCI_EXP_DEVCAP: - case PCI_EXP_DEVCTL: - case PCI_EXP_DEVSTA: - return __kc_pcie_cap_has_devctl(dev); - case PCI_EXP_LNKCAP: - case PCI_EXP_LNKCTL: - case PCI_EXP_LNKSTA: - return __kc_pcie_cap_has_lnkctl(dev); - case PCI_EXP_SLTCAP: - case PCI_EXP_SLTCTL: - case PCI_EXP_SLTSTA: - return __kc_pcie_cap_has_sltctl(dev); - case PCI_EXP_RTCTL: - case PCI_EXP_RTCAP: - case PCI_EXP_RTSTA: - return __kc_pcie_cap_has_rtctl(dev); - case PCI_EXP_DEVCAP2: - case PCI_EXP_DEVCTL2: - case PCI_EXP_LNKCAP2: - case PCI_EXP_LNKCTL2: - case PCI_EXP_LNKSTA2: - return __kc_pcie_cap_version(dev) > 1; - default: - return false; - } -} - -/* - * Note that these accessor functions are only for the "PCI Express - * Capability" (see PCIe spec r3.0, sec 7.8). They do not apply to the - * other "PCI Express Extended Capabilities" (AER, VC, ACS, MFVC, etc.) - */ -int __kc_pcie_capability_read_word(struct pci_dev *dev, int pos, u16 *val) -{ - int ret; - - *val = 0; - if (pos & 1) - return -EINVAL; - - if (__kc_pcie_capability_reg_implemented(dev, pos)) { - ret = pci_read_config_word(dev, pci_pcie_cap(dev) + pos, val); - /* - * Reset *val to 0 if pci_read_config_word() fails, it may - * have been written as 0xFFFF if hardware error happens - * during pci_read_config_word(). - */ - if (ret) - *val = 0; - return ret; - } - - /* - * For Functions that do not implement the Slot Capabilities, - * Slot Status, and Slot Control registers, these spaces must - * be hardwired to 0b, with the exception of the Presence Detect - * State bit in the Slot Status register of Downstream Ports, - * which must be hardwired to 1b. (PCIe Base Spec 3.0, sec 7.8) - */ - if (pci_is_pcie(dev) && pos == PCI_EXP_SLTSTA && - pci_pcie_type(dev) == PCI_EXP_TYPE_DOWNSTREAM) { - *val = PCI_EXP_SLTSTA_PDS; - } - - return 0; -} - -int __kc_pcie_capability_write_word(struct pci_dev *dev, int pos, u16 val) -{ - if (pos & 1) - return -EINVAL; - - if (!__kc_pcie_capability_reg_implemented(dev, pos)) - return 0; - - return pci_write_config_word(dev, pci_pcie_cap(dev) + pos, val); -} - -int __kc_pcie_capability_clear_and_set_word(struct pci_dev *dev, int pos, - u16 clear, u16 set) -{ - int ret; - u16 val; - - ret = __kc_pcie_capability_read_word(dev, pos, &val); - if (!ret) { - val &= ~clear; - val |= set; - ret = __kc_pcie_capability_write_word(dev, pos, val); - } - - return ret; -} -#endif /* !(SLE_VERSION_CODE && SLE_VERSION_CODE >= SLE_VERSION(11,3,0)) && \ - !(RHEL_RELEASE_CODE && RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6,5)) */ -#endif /* < 3.7.0 */ - -/******************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(3,9,0) ) -#endif /* 3.9.0 */ - -/*****************************************************************************/ -#if ( LINUX_VERSION_CODE < KERNEL_VERSION(3,10,0) ) -#ifdef CONFIG_PCI_IOV -int __kc_pci_vfs_assigned(struct pci_dev *dev) -{ - unsigned int vfs_assigned = 0; -#ifdef HAVE_PCI_DEV_FLAGS_ASSIGNED - int pos; - struct pci_dev *vfdev; - unsigned short dev_id; - - /* only search if we are a PF */ - if (!dev->is_physfn) - return 0; - - /* find SR-IOV capability */ - pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_SRIOV); - if (!pos) - return 0; - - /* - * determine the device ID for the VFs, the vendor ID will be the - * same as the PF so there is no need to check for that one - */ - pci_read_config_word(dev, pos + PCI_SRIOV_VF_DID, &dev_id); - - /* loop through all the VFs to see if we own any that are assigned */ - vfdev = pci_get_device(dev->vendor, dev_id, NULL); - while (vfdev) { - /* - * It is considered assigned if it is a virtual function with - * our dev as the physical function and the assigned bit is set - */ - if (vfdev->is_virtfn && (vfdev->physfn == dev) && - (vfdev->dev_flags & PCI_DEV_FLAGS_ASSIGNED)) - vfs_assigned++; - - vfdev = pci_get_device(dev->vendor, dev_id, vfdev); - } - -#endif /* HAVE_PCI_DEV_FLAGS_ASSIGNED */ - return vfs_assigned; -} - -#endif /* CONFIG_PCI_IOV */ -#endif /* 3.10.0 */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h index e2cf71e0..84826b26 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> @@ -3891,7 +3891,7 @@ skb_set_hash(struct sk_buff *skb, __u32 hash, __always_unused int type) #if (( LINUX_VERSION_CODE >= KERNEL_VERSION(3,19,0) ) \ || ( RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(7,2) )) #define HAVE_NDO_DFLT_BRIDGE_ADD_MASK -#if (!( RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(7,2) )) +#if ( RHEL_RELEASE_CODE != RHEL_RELEASE_VERSION(7,2) ) #define HAVE_NDO_FDB_ADD_VID #endif /* !RHEL 7.2 */ #endif /* >= 3.19.0 */ @@ -3901,12 +3901,13 @@ skb_set_hash(struct sk_buff *skb, __u32 hash, __always_unused int type) /* vlan_tx_xx functions got renamed to skb_vlan */ #define vlan_tx_tag_get skb_vlan_tag_get #define vlan_tx_tag_present skb_vlan_tag_present -#if (!( RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(7,2) )) +#if ( RHEL_RELEASE_CODE != RHEL_RELEASE_VERSION(7,2) ) #define HAVE_NDO_BRIDGE_SET_DEL_LINK_FLAGS #endif /* !RHEL 7.2 */ #endif /* 4.0.0 */ -#if ( LINUX_VERSION_CODE >= KERNEL_VERSION(4,1,0) ) +#if (( LINUX_VERSION_CODE >= KERNEL_VERSION(4,1,0) ) \ + || ( RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(7,3) )) /* ndo_bridge_getlink adds new nlflags parameter */ #define HAVE_NDO_BRIDGE_GETLINK_NLFLAGS #endif /* >= 4.1.0 */ @@ -3915,4 +3916,21 @@ skb_set_hash(struct sk_buff *skb, __u32 hash, __always_unused int type) /* ndo_bridge_getlink adds new filter_mask and vlan_fill parameters */ #define HAVE_NDO_BRIDGE_GETLINK_FILTER_MASK_VLAN_FILL #endif /* >= 4.2.0 */ + +/* + * vlan_tx_tag_* macros renamed to skb_vlan_tag_* (Linux commit: df8a39defad4) + * For older kernels backported this commit, need to use renamed functions. + * This fix is specific to RedHat/CentOS kernels. + */ +#if (defined(RHEL_RELEASE_CODE) && \ + (RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6, 8)) && \ + (LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 34))) +#define vlan_tx_tag_get skb_vlan_tag_get +#define vlan_tx_tag_present skb_vlan_tag_present +#endif + +#if ( LINUX_VERSION_CODE >= KERNEL_VERSION(4,9,0) ) +#define HAVE_VF_VLAN_PROTO +#endif /* >= 4.9.0 */ + #endif /* _KCOMPAT_H_ */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat_ethtool.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat_ethtool.c deleted file mode 100644 index e1a89388..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/igb/kcompat_ethtool.c +++ /dev/null @@ -1,1171 +0,0 @@ -/******************************************************************************* - - Intel(R) Gigabit Ethernet Linux driver - Copyright(c) 2007-2013 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - -/* - * net/core/ethtool.c - Ethtool ioctl handler - * Copyright (c) 2003 Matthew Wilcox <matthew@wil.cx> - * - * This file is where we call all the ethtool_ops commands to get - * the information ethtool needs. We fall back to calling do_ioctl() - * for drivers which haven't been converted to ethtool_ops yet. - * - * It's GPL, stupid. - * - * Modification by sfeldma@pobox.com to work as backward compat - * solution for pre-ethtool_ops kernels. - * - copied struct ethtool_ops from ethtool.h - * - defined SET_ETHTOOL_OPS - * - put in some #ifndef NETIF_F_xxx wrappers - * - changes refs to dev->ethtool_ops to ethtool_ops - * - changed dev_ethtool to ethtool_ioctl - * - remove EXPORT_SYMBOL()s - * - added _kc_ prefix in built-in ethtool_op_xxx ops. - */ - -#include <linux/module.h> -#include <linux/types.h> -#include <linux/errno.h> -#include <linux/mii.h> -#include <linux/ethtool.h> -#include <linux/netdevice.h> -#include <asm/uaccess.h> - -#include "kcompat.h" - -#undef SUPPORTED_10000baseT_Full -#define SUPPORTED_10000baseT_Full (1 << 12) -#undef ADVERTISED_10000baseT_Full -#define ADVERTISED_10000baseT_Full (1 << 12) -#undef SPEED_10000 -#define SPEED_10000 10000 - -#undef ethtool_ops -#define ethtool_ops _kc_ethtool_ops - -struct _kc_ethtool_ops { - int (*get_settings)(struct net_device *, struct ethtool_cmd *); - int (*set_settings)(struct net_device *, struct ethtool_cmd *); - void (*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *); - int (*get_regs_len)(struct net_device *); - void (*get_regs)(struct net_device *, struct ethtool_regs *, void *); - void (*get_wol)(struct net_device *, struct ethtool_wolinfo *); - int (*set_wol)(struct net_device *, struct ethtool_wolinfo *); - u32 (*get_msglevel)(struct net_device *); - void (*set_msglevel)(struct net_device *, u32); - int (*nway_reset)(struct net_device *); - u32 (*get_link)(struct net_device *); - int (*get_eeprom_len)(struct net_device *); - int (*get_eeprom)(struct net_device *, struct ethtool_eeprom *, u8 *); - int (*set_eeprom)(struct net_device *, struct ethtool_eeprom *, u8 *); - int (*get_coalesce)(struct net_device *, struct ethtool_coalesce *); - int (*set_coalesce)(struct net_device *, struct ethtool_coalesce *); - void (*get_ringparam)(struct net_device *, struct ethtool_ringparam *); - int (*set_ringparam)(struct net_device *, struct ethtool_ringparam *); - void (*get_pauseparam)(struct net_device *, - struct ethtool_pauseparam*); - int (*set_pauseparam)(struct net_device *, - struct ethtool_pauseparam*); - u32 (*get_rx_csum)(struct net_device *); - int (*set_rx_csum)(struct net_device *, u32); - u32 (*get_tx_csum)(struct net_device *); - int (*set_tx_csum)(struct net_device *, u32); - u32 (*get_sg)(struct net_device *); - int (*set_sg)(struct net_device *, u32); - u32 (*get_tso)(struct net_device *); - int (*set_tso)(struct net_device *, u32); - int (*self_test_count)(struct net_device *); - void (*self_test)(struct net_device *, struct ethtool_test *, u64 *); - void (*get_strings)(struct net_device *, u32 stringset, u8 *); - int (*phys_id)(struct net_device *, u32); - int (*get_stats_count)(struct net_device *); - void (*get_ethtool_stats)(struct net_device *, struct ethtool_stats *, - u64 *); -} *ethtool_ops = NULL; - -#undef SET_ETHTOOL_OPS -#define SET_ETHTOOL_OPS(netdev, ops) (ethtool_ops = (ops)) - -/* - * Some useful ethtool_ops methods that are device independent. If we find that - * all drivers want to do the same thing here, we can turn these into dev_() - * function calls. - */ - -#undef ethtool_op_get_link -#define ethtool_op_get_link _kc_ethtool_op_get_link -u32 _kc_ethtool_op_get_link(struct net_device *dev) -{ - return netif_carrier_ok(dev) ? 1 : 0; -} - -#undef ethtool_op_get_tx_csum -#define ethtool_op_get_tx_csum _kc_ethtool_op_get_tx_csum -u32 _kc_ethtool_op_get_tx_csum(struct net_device *dev) -{ -#ifdef NETIF_F_IP_CSUM - return (dev->features & NETIF_F_IP_CSUM) != 0; -#else - return 0; -#endif -} - -#undef ethtool_op_set_tx_csum -#define ethtool_op_set_tx_csum _kc_ethtool_op_set_tx_csum -int _kc_ethtool_op_set_tx_csum(struct net_device *dev, u32 data) -{ -#ifdef NETIF_F_IP_CSUM - if (data) -#ifdef NETIF_F_IPV6_CSUM - dev->features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM); - else - dev->features &= ~(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM); -#else - dev->features |= NETIF_F_IP_CSUM; - else - dev->features &= ~NETIF_F_IP_CSUM; -#endif -#endif - - return 0; -} - -#undef ethtool_op_get_sg -#define ethtool_op_get_sg _kc_ethtool_op_get_sg -u32 _kc_ethtool_op_get_sg(struct net_device *dev) -{ -#ifdef NETIF_F_SG - return (dev->features & NETIF_F_SG) != 0; -#else - return 0; -#endif -} - -#undef ethtool_op_set_sg -#define ethtool_op_set_sg _kc_ethtool_op_set_sg -int _kc_ethtool_op_set_sg(struct net_device *dev, u32 data) -{ -#ifdef NETIF_F_SG - if (data) - dev->features |= NETIF_F_SG; - else - dev->features &= ~NETIF_F_SG; -#endif - - return 0; -} - -#undef ethtool_op_get_tso -#define ethtool_op_get_tso _kc_ethtool_op_get_tso -u32 _kc_ethtool_op_get_tso(struct net_device *dev) -{ -#ifdef NETIF_F_TSO - return (dev->features & NETIF_F_TSO) != 0; -#else - return 0; -#endif -} - -#undef ethtool_op_set_tso -#define ethtool_op_set_tso _kc_ethtool_op_set_tso -int _kc_ethtool_op_set_tso(struct net_device *dev, u32 data) -{ -#ifdef NETIF_F_TSO - if (data) - dev->features |= NETIF_F_TSO; - else - dev->features &= ~NETIF_F_TSO; -#endif - - return 0; -} - -/* Handlers for each ethtool command */ - -static int ethtool_get_settings(struct net_device *dev, void *useraddr) -{ - struct ethtool_cmd cmd = { ETHTOOL_GSET }; - int err; - - if (!ethtool_ops->get_settings) - return -EOPNOTSUPP; - - err = ethtool_ops->get_settings(dev, &cmd); - if (err < 0) - return err; - - if (copy_to_user(useraddr, &cmd, sizeof(cmd))) - return -EFAULT; - return 0; -} - -static int ethtool_set_settings(struct net_device *dev, void *useraddr) -{ - struct ethtool_cmd cmd; - - if (!ethtool_ops->set_settings) - return -EOPNOTSUPP; - - if (copy_from_user(&cmd, useraddr, sizeof(cmd))) - return -EFAULT; - - return ethtool_ops->set_settings(dev, &cmd); -} - -static int ethtool_get_drvinfo(struct net_device *dev, void *useraddr) -{ - struct ethtool_drvinfo info; - struct ethtool_ops *ops = ethtool_ops; - - if (!ops->get_drvinfo) - return -EOPNOTSUPP; - - memset(&info, 0, sizeof(info)); - info.cmd = ETHTOOL_GDRVINFO; - ops->get_drvinfo(dev, &info); - - if (ops->self_test_count) - info.testinfo_len = ops->self_test_count(dev); - if (ops->get_stats_count) - info.n_stats = ops->get_stats_count(dev); - if (ops->get_regs_len) - info.regdump_len = ops->get_regs_len(dev); - if (ops->get_eeprom_len) - info.eedump_len = ops->get_eeprom_len(dev); - - if (copy_to_user(useraddr, &info, sizeof(info))) - return -EFAULT; - return 0; -} - -static int ethtool_get_regs(struct net_device *dev, char *useraddr) -{ - struct ethtool_regs regs; - struct ethtool_ops *ops = ethtool_ops; - void *regbuf; - int reglen, ret; - - if (!ops->get_regs || !ops->get_regs_len) - return -EOPNOTSUPP; - - if (copy_from_user(®s, useraddr, sizeof(regs))) - return -EFAULT; - - reglen = ops->get_regs_len(dev); - if (regs.len > reglen) - regs.len = reglen; - - regbuf = kmalloc(reglen, GFP_USER); - if (!regbuf) - return -ENOMEM; - - ops->get_regs(dev, ®s, regbuf); - - ret = -EFAULT; - if (copy_to_user(useraddr, ®s, sizeof(regs))) - goto out; - useraddr += offsetof(struct ethtool_regs, data); - if (copy_to_user(useraddr, regbuf, reglen)) - goto out; - ret = 0; - -out: - kfree(regbuf); - return ret; -} - -static int ethtool_get_wol(struct net_device *dev, char *useraddr) -{ - struct ethtool_wolinfo wol = { ETHTOOL_GWOL }; - - if (!ethtool_ops->get_wol) - return -EOPNOTSUPP; - - ethtool_ops->get_wol(dev, &wol); - - if (copy_to_user(useraddr, &wol, sizeof(wol))) - return -EFAULT; - return 0; -} - -static int ethtool_set_wol(struct net_device *dev, char *useraddr) -{ - struct ethtool_wolinfo wol; - - if (!ethtool_ops->set_wol) - return -EOPNOTSUPP; - - if (copy_from_user(&wol, useraddr, sizeof(wol))) - return -EFAULT; - - return ethtool_ops->set_wol(dev, &wol); -} - -static int ethtool_get_msglevel(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata = { ETHTOOL_GMSGLVL }; - - if (!ethtool_ops->get_msglevel) - return -EOPNOTSUPP; - - edata.data = ethtool_ops->get_msglevel(dev); - - if (copy_to_user(useraddr, &edata, sizeof(edata))) - return -EFAULT; - return 0; -} - -static int ethtool_set_msglevel(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata; - - if (!ethtool_ops->set_msglevel) - return -EOPNOTSUPP; - - if (copy_from_user(&edata, useraddr, sizeof(edata))) - return -EFAULT; - - ethtool_ops->set_msglevel(dev, edata.data); - return 0; -} - -static int ethtool_nway_reset(struct net_device *dev) -{ - if (!ethtool_ops->nway_reset) - return -EOPNOTSUPP; - - return ethtool_ops->nway_reset(dev); -} - -static int ethtool_get_link(struct net_device *dev, void *useraddr) -{ - struct ethtool_value edata = { ETHTOOL_GLINK }; - - if (!ethtool_ops->get_link) - return -EOPNOTSUPP; - - edata.data = ethtool_ops->get_link(dev); - - if (copy_to_user(useraddr, &edata, sizeof(edata))) - return -EFAULT; - return 0; -} - -static int ethtool_get_eeprom(struct net_device *dev, void *useraddr) -{ - struct ethtool_eeprom eeprom; - struct ethtool_ops *ops = ethtool_ops; - u8 *data; - int ret; - - if (!ops->get_eeprom || !ops->get_eeprom_len) - return -EOPNOTSUPP; - - if (copy_from_user(&eeprom, useraddr, sizeof(eeprom))) - return -EFAULT; - - /* Check for wrap and zero */ - if (eeprom.offset + eeprom.len <= eeprom.offset) - return -EINVAL; - - /* Check for exceeding total eeprom len */ - if (eeprom.offset + eeprom.len > ops->get_eeprom_len(dev)) - return -EINVAL; - - data = kmalloc(eeprom.len, GFP_USER); - if (!data) - return -ENOMEM; - - ret = -EFAULT; - if (copy_from_user(data, useraddr + sizeof(eeprom), eeprom.len)) - goto out; - - ret = ops->get_eeprom(dev, &eeprom, data); - if (ret) - goto out; - - ret = -EFAULT; - if (copy_to_user(useraddr, &eeprom, sizeof(eeprom))) - goto out; - if (copy_to_user(useraddr + sizeof(eeprom), data, eeprom.len)) - goto out; - ret = 0; - -out: - kfree(data); - return ret; -} - -static int ethtool_set_eeprom(struct net_device *dev, void *useraddr) -{ - struct ethtool_eeprom eeprom; - struct ethtool_ops *ops = ethtool_ops; - u8 *data; - int ret; - - if (!ops->set_eeprom || !ops->get_eeprom_len) - return -EOPNOTSUPP; - - if (copy_from_user(&eeprom, useraddr, sizeof(eeprom))) - return -EFAULT; - - /* Check for wrap and zero */ - if (eeprom.offset + eeprom.len <= eeprom.offset) - return -EINVAL; - - /* Check for exceeding total eeprom len */ - if (eeprom.offset + eeprom.len > ops->get_eeprom_len(dev)) - return -EINVAL; - - data = kmalloc(eeprom.len, GFP_USER); - if (!data) - return -ENOMEM; - - ret = -EFAULT; - if (copy_from_user(data, useraddr + sizeof(eeprom), eeprom.len)) - goto out; - - ret = ops->set_eeprom(dev, &eeprom, data); - if (ret) - goto out; - - if (copy_to_user(useraddr + sizeof(eeprom), data, eeprom.len)) - ret = -EFAULT; - -out: - kfree(data); - return ret; -} - -static int ethtool_get_coalesce(struct net_device *dev, void *useraddr) -{ - struct ethtool_coalesce coalesce = { ETHTOOL_GCOALESCE }; - - if (!ethtool_ops->get_coalesce) - return -EOPNOTSUPP; - - ethtool_ops->get_coalesce(dev, &coalesce); - - if (copy_to_user(useraddr, &coalesce, sizeof(coalesce))) - return -EFAULT; - return 0; -} - -static int ethtool_set_coalesce(struct net_device *dev, void *useraddr) -{ - struct ethtool_coalesce coalesce; - - if (!ethtool_ops->get_coalesce) - return -EOPNOTSUPP; - - if (copy_from_user(&coalesce, useraddr, sizeof(coalesce))) - return -EFAULT; - - return ethtool_ops->set_coalesce(dev, &coalesce); -} - -static int ethtool_get_ringparam(struct net_device *dev, void *useraddr) -{ - struct ethtool_ringparam ringparam = { ETHTOOL_GRINGPARAM }; - - if (!ethtool_ops->get_ringparam) - return -EOPNOTSUPP; - - ethtool_ops->get_ringparam(dev, &ringparam); - - if (copy_to_user(useraddr, &ringparam, sizeof(ringparam))) - return -EFAULT; - return 0; -} - -static int ethtool_set_ringparam(struct net_device *dev, void *useraddr) -{ - struct ethtool_ringparam ringparam; - - if (!ethtool_ops->get_ringparam) - return -EOPNOTSUPP; - - if (copy_from_user(&ringparam, useraddr, sizeof(ringparam))) - return -EFAULT; - - return ethtool_ops->set_ringparam(dev, &ringparam); -} - -static int ethtool_get_pauseparam(struct net_device *dev, void *useraddr) -{ - struct ethtool_pauseparam pauseparam = { ETHTOOL_GPAUSEPARAM }; - - if (!ethtool_ops->get_pauseparam) - return -EOPNOTSUPP; - - ethtool_ops->get_pauseparam(dev, &pauseparam); - - if (copy_to_user(useraddr, &pauseparam, sizeof(pauseparam))) - return -EFAULT; - return 0; -} - -static int ethtool_set_pauseparam(struct net_device *dev, void *useraddr) -{ - struct ethtool_pauseparam pauseparam; - - if (!ethtool_ops->get_pauseparam) - return -EOPNOTSUPP; - - if (copy_from_user(&pauseparam, useraddr, sizeof(pauseparam))) - return -EFAULT; - - return ethtool_ops->set_pauseparam(dev, &pauseparam); -} - -static int ethtool_get_rx_csum(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata = { ETHTOOL_GRXCSUM }; - - if (!ethtool_ops->get_rx_csum) - return -EOPNOTSUPP; - - edata.data = ethtool_ops->get_rx_csum(dev); - - if (copy_to_user(useraddr, &edata, sizeof(edata))) - return -EFAULT; - return 0; -} - -static int ethtool_set_rx_csum(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata; - - if (!ethtool_ops->set_rx_csum) - return -EOPNOTSUPP; - - if (copy_from_user(&edata, useraddr, sizeof(edata))) - return -EFAULT; - - ethtool_ops->set_rx_csum(dev, edata.data); - return 0; -} - -static int ethtool_get_tx_csum(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata = { ETHTOOL_GTXCSUM }; - - if (!ethtool_ops->get_tx_csum) - return -EOPNOTSUPP; - - edata.data = ethtool_ops->get_tx_csum(dev); - - if (copy_to_user(useraddr, &edata, sizeof(edata))) - return -EFAULT; - return 0; -} - -static int ethtool_set_tx_csum(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata; - - if (!ethtool_ops->set_tx_csum) - return -EOPNOTSUPP; - - if (copy_from_user(&edata, useraddr, sizeof(edata))) - return -EFAULT; - - return ethtool_ops->set_tx_csum(dev, edata.data); -} - -static int ethtool_get_sg(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata = { ETHTOOL_GSG }; - - if (!ethtool_ops->get_sg) - return -EOPNOTSUPP; - - edata.data = ethtool_ops->get_sg(dev); - - if (copy_to_user(useraddr, &edata, sizeof(edata))) - return -EFAULT; - return 0; -} - -static int ethtool_set_sg(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata; - - if (!ethtool_ops->set_sg) - return -EOPNOTSUPP; - - if (copy_from_user(&edata, useraddr, sizeof(edata))) - return -EFAULT; - - return ethtool_ops->set_sg(dev, edata.data); -} - -static int ethtool_get_tso(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata = { ETHTOOL_GTSO }; - - if (!ethtool_ops->get_tso) - return -EOPNOTSUPP; - - edata.data = ethtool_ops->get_tso(dev); - - if (copy_to_user(useraddr, &edata, sizeof(edata))) - return -EFAULT; - return 0; -} - -static int ethtool_set_tso(struct net_device *dev, char *useraddr) -{ - struct ethtool_value edata; - - if (!ethtool_ops->set_tso) - return -EOPNOTSUPP; - - if (copy_from_user(&edata, useraddr, sizeof(edata))) - return -EFAULT; - - return ethtool_ops->set_tso(dev, edata.data); -} - -static int ethtool_self_test(struct net_device *dev, char *useraddr) -{ - struct ethtool_test test; - struct ethtool_ops *ops = ethtool_ops; - u64 *data; - int ret; - - if (!ops->self_test || !ops->self_test_count) - return -EOPNOTSUPP; - - if (copy_from_user(&test, useraddr, sizeof(test))) - return -EFAULT; - - test.len = ops->self_test_count(dev); - data = kmalloc(test.len * sizeof(u64), GFP_USER); - if (!data) - return -ENOMEM; - - ops->self_test(dev, &test, data); - - ret = -EFAULT; - if (copy_to_user(useraddr, &test, sizeof(test))) - goto out; - useraddr += sizeof(test); - if (copy_to_user(useraddr, data, test.len * sizeof(u64))) - goto out; - ret = 0; - -out: - kfree(data); - return ret; -} - -static int ethtool_get_strings(struct net_device *dev, void *useraddr) -{ - struct ethtool_gstrings gstrings; - struct ethtool_ops *ops = ethtool_ops; - u8 *data; - int ret; - - if (!ops->get_strings) - return -EOPNOTSUPP; - - if (copy_from_user(&gstrings, useraddr, sizeof(gstrings))) - return -EFAULT; - - switch (gstrings.string_set) { - case ETH_SS_TEST: - if (!ops->self_test_count) - return -EOPNOTSUPP; - gstrings.len = ops->self_test_count(dev); - break; - case ETH_SS_STATS: - if (!ops->get_stats_count) - return -EOPNOTSUPP; - gstrings.len = ops->get_stats_count(dev); - break; - default: - return -EINVAL; - } - - data = kmalloc(gstrings.len * ETH_GSTRING_LEN, GFP_USER); - if (!data) - return -ENOMEM; - - ops->get_strings(dev, gstrings.string_set, data); - - ret = -EFAULT; - if (copy_to_user(useraddr, &gstrings, sizeof(gstrings))) - goto out; - useraddr += sizeof(gstrings); - if (copy_to_user(useraddr, data, gstrings.len * ETH_GSTRING_LEN)) - goto out; - ret = 0; - -out: - kfree(data); - return ret; -} - -static int ethtool_phys_id(struct net_device *dev, void *useraddr) -{ - struct ethtool_value id; - - if (!ethtool_ops->phys_id) - return -EOPNOTSUPP; - - if (copy_from_user(&id, useraddr, sizeof(id))) - return -EFAULT; - - return ethtool_ops->phys_id(dev, id.data); -} - -static int ethtool_get_stats(struct net_device *dev, void *useraddr) -{ - struct ethtool_stats stats; - struct ethtool_ops *ops = ethtool_ops; - u64 *data; - int ret; - - if (!ops->get_ethtool_stats || !ops->get_stats_count) - return -EOPNOTSUPP; - - if (copy_from_user(&stats, useraddr, sizeof(stats))) - return -EFAULT; - - stats.n_stats = ops->get_stats_count(dev); - data = kmalloc(stats.n_stats * sizeof(u64), GFP_USER); - if (!data) - return -ENOMEM; - - ops->get_ethtool_stats(dev, &stats, data); - - ret = -EFAULT; - if (copy_to_user(useraddr, &stats, sizeof(stats))) - goto out; - useraddr += sizeof(stats); - if (copy_to_user(useraddr, data, stats.n_stats * sizeof(u64))) - goto out; - ret = 0; - -out: - kfree(data); - return ret; -} - -/* The main entry point in this file. Called from net/core/dev.c */ - -#define ETHTOOL_OPS_COMPAT -int ethtool_ioctl(struct ifreq *ifr) -{ - struct net_device *dev = __dev_get_by_name(ifr->ifr_name); - void *useraddr = (void *) ifr->ifr_data; - u32 ethcmd; - - /* - * XXX: This can be pushed down into the ethtool_* handlers that - * need it. Keep existing behavior for the moment. - */ - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (!dev || !netif_device_present(dev)) - return -ENODEV; - - if (copy_from_user(ðcmd, useraddr, sizeof (ethcmd))) - return -EFAULT; - - switch (ethcmd) { - case ETHTOOL_GSET: - return ethtool_get_settings(dev, useraddr); - case ETHTOOL_SSET: - return ethtool_set_settings(dev, useraddr); - case ETHTOOL_GDRVINFO: - return ethtool_get_drvinfo(dev, useraddr); - case ETHTOOL_GREGS: - return ethtool_get_regs(dev, useraddr); - case ETHTOOL_GWOL: - return ethtool_get_wol(dev, useraddr); - case ETHTOOL_SWOL: - return ethtool_set_wol(dev, useraddr); - case ETHTOOL_GMSGLVL: - return ethtool_get_msglevel(dev, useraddr); - case ETHTOOL_SMSGLVL: - return ethtool_set_msglevel(dev, useraddr); - case ETHTOOL_NWAY_RST: - return ethtool_nway_reset(dev); - case ETHTOOL_GLINK: - return ethtool_get_link(dev, useraddr); - case ETHTOOL_GEEPROM: - return ethtool_get_eeprom(dev, useraddr); - case ETHTOOL_SEEPROM: - return ethtool_set_eeprom(dev, useraddr); - case ETHTOOL_GCOALESCE: - return ethtool_get_coalesce(dev, useraddr); - case ETHTOOL_SCOALESCE: - return ethtool_set_coalesce(dev, useraddr); - case ETHTOOL_GRINGPARAM: - return ethtool_get_ringparam(dev, useraddr); - case ETHTOOL_SRINGPARAM: - return ethtool_set_ringparam(dev, useraddr); - case ETHTOOL_GPAUSEPARAM: - return ethtool_get_pauseparam(dev, useraddr); - case ETHTOOL_SPAUSEPARAM: - return ethtool_set_pauseparam(dev, useraddr); - case ETHTOOL_GRXCSUM: - return ethtool_get_rx_csum(dev, useraddr); - case ETHTOOL_SRXCSUM: - return ethtool_set_rx_csum(dev, useraddr); - case ETHTOOL_GTXCSUM: - return ethtool_get_tx_csum(dev, useraddr); - case ETHTOOL_STXCSUM: - return ethtool_set_tx_csum(dev, useraddr); - case ETHTOOL_GSG: - return ethtool_get_sg(dev, useraddr); - case ETHTOOL_SSG: - return ethtool_set_sg(dev, useraddr); - case ETHTOOL_GTSO: - return ethtool_get_tso(dev, useraddr); - case ETHTOOL_STSO: - return ethtool_set_tso(dev, useraddr); - case ETHTOOL_TEST: - return ethtool_self_test(dev, useraddr); - case ETHTOOL_GSTRINGS: - return ethtool_get_strings(dev, useraddr); - case ETHTOOL_PHYS_ID: - return ethtool_phys_id(dev, useraddr); - case ETHTOOL_GSTATS: - return ethtool_get_stats(dev, useraddr); - default: - return -EOPNOTSUPP; - } - - return -EOPNOTSUPP; -} - -#define mii_if_info _kc_mii_if_info -struct _kc_mii_if_info { - int phy_id; - int advertising; - int phy_id_mask; - int reg_num_mask; - - unsigned int full_duplex : 1; /* is full duplex? */ - unsigned int force_media : 1; /* is autoneg. disabled? */ - - struct net_device *dev; - int (*mdio_read) (struct net_device *dev, int phy_id, int location); - void (*mdio_write) (struct net_device *dev, int phy_id, int location, int val); -}; - -struct ethtool_cmd; -struct mii_ioctl_data; - -#undef mii_link_ok -#define mii_link_ok _kc_mii_link_ok -#undef mii_nway_restart -#define mii_nway_restart _kc_mii_nway_restart -#undef mii_ethtool_gset -#define mii_ethtool_gset _kc_mii_ethtool_gset -#undef mii_ethtool_sset -#define mii_ethtool_sset _kc_mii_ethtool_sset -#undef mii_check_link -#define mii_check_link _kc_mii_check_link -extern int _kc_mii_link_ok (struct mii_if_info *mii); -extern int _kc_mii_nway_restart (struct mii_if_info *mii); -extern int _kc_mii_ethtool_gset(struct mii_if_info *mii, - struct ethtool_cmd *ecmd); -extern int _kc_mii_ethtool_sset(struct mii_if_info *mii, - struct ethtool_cmd *ecmd); -extern void _kc_mii_check_link (struct mii_if_info *mii); -#if ( LINUX_VERSION_CODE > KERNEL_VERSION(2,4,6) ) -#undef generic_mii_ioctl -#define generic_mii_ioctl _kc_generic_mii_ioctl -extern int _kc_generic_mii_ioctl(struct mii_if_info *mii_if, - struct mii_ioctl_data *mii_data, int cmd, - unsigned int *duplex_changed); -#endif /* > 2.4.6 */ - - -struct _kc_pci_dev_ext { - struct pci_dev *dev; - void *pci_drvdata; - struct pci_driver *driver; -}; - -struct _kc_net_dev_ext { - struct net_device *dev; - unsigned int carrier; -}; - - -/**************************************/ -/* mii support */ - -int _kc_mii_ethtool_gset(struct mii_if_info *mii, struct ethtool_cmd *ecmd) -{ - struct net_device *dev = mii->dev; - u32 advert, bmcr, lpa, nego; - - ecmd->supported = - (SUPPORTED_10baseT_Half | SUPPORTED_10baseT_Full | - SUPPORTED_100baseT_Half | SUPPORTED_100baseT_Full | - SUPPORTED_Autoneg | SUPPORTED_TP | SUPPORTED_MII); - - /* only supports twisted-pair */ - ecmd->port = PORT_MII; - - /* only supports internal transceiver */ - ecmd->transceiver = XCVR_INTERNAL; - - /* this isn't fully supported at higher layers */ - ecmd->phy_address = mii->phy_id; - - ecmd->advertising = ADVERTISED_TP | ADVERTISED_MII; - advert = mii->mdio_read(dev, mii->phy_id, MII_ADVERTISE); - if (advert & ADVERTISE_10HALF) - ecmd->advertising |= ADVERTISED_10baseT_Half; - if (advert & ADVERTISE_10FULL) - ecmd->advertising |= ADVERTISED_10baseT_Full; - if (advert & ADVERTISE_100HALF) - ecmd->advertising |= ADVERTISED_100baseT_Half; - if (advert & ADVERTISE_100FULL) - ecmd->advertising |= ADVERTISED_100baseT_Full; - - bmcr = mii->mdio_read(dev, mii->phy_id, MII_BMCR); - lpa = mii->mdio_read(dev, mii->phy_id, MII_LPA); - if (bmcr & BMCR_ANENABLE) { - ecmd->advertising |= ADVERTISED_Autoneg; - ecmd->autoneg = AUTONEG_ENABLE; - - nego = mii_nway_result(advert & lpa); - if (nego == LPA_100FULL || nego == LPA_100HALF) - ecmd->speed = SPEED_100; - else - ecmd->speed = SPEED_10; - if (nego == LPA_100FULL || nego == LPA_10FULL) { - ecmd->duplex = DUPLEX_FULL; - mii->full_duplex = 1; - } else { - ecmd->duplex = DUPLEX_HALF; - mii->full_duplex = 0; - } - } else { - ecmd->autoneg = AUTONEG_DISABLE; - - ecmd->speed = (bmcr & BMCR_SPEED100) ? SPEED_100 : SPEED_10; - ecmd->duplex = (bmcr & BMCR_FULLDPLX) ? DUPLEX_FULL : DUPLEX_HALF; - } - - /* ignore maxtxpkt, maxrxpkt for now */ - - return 0; -} - -int _kc_mii_ethtool_sset(struct mii_if_info *mii, struct ethtool_cmd *ecmd) -{ - struct net_device *dev = mii->dev; - - if (ecmd->speed != SPEED_10 && ecmd->speed != SPEED_100) - return -EINVAL; - if (ecmd->duplex != DUPLEX_HALF && ecmd->duplex != DUPLEX_FULL) - return -EINVAL; - if (ecmd->port != PORT_MII) - return -EINVAL; - if (ecmd->transceiver != XCVR_INTERNAL) - return -EINVAL; - if (ecmd->phy_address != mii->phy_id) - return -EINVAL; - if (ecmd->autoneg != AUTONEG_DISABLE && ecmd->autoneg != AUTONEG_ENABLE) - return -EINVAL; - - /* ignore supported, maxtxpkt, maxrxpkt */ - - if (ecmd->autoneg == AUTONEG_ENABLE) { - u32 bmcr, advert, tmp; - - if ((ecmd->advertising & (ADVERTISED_10baseT_Half | - ADVERTISED_10baseT_Full | - ADVERTISED_100baseT_Half | - ADVERTISED_100baseT_Full)) == 0) - return -EINVAL; - - /* advertise only what has been requested */ - advert = mii->mdio_read(dev, mii->phy_id, MII_ADVERTISE); - tmp = advert & ~(ADVERTISE_ALL | ADVERTISE_100BASE4); - if (ADVERTISED_10baseT_Half) - tmp |= ADVERTISE_10HALF; - if (ADVERTISED_10baseT_Full) - tmp |= ADVERTISE_10FULL; - if (ADVERTISED_100baseT_Half) - tmp |= ADVERTISE_100HALF; - if (ADVERTISED_100baseT_Full) - tmp |= ADVERTISE_100FULL; - if (advert != tmp) { - mii->mdio_write(dev, mii->phy_id, MII_ADVERTISE, tmp); - mii->advertising = tmp; - } - - /* turn on autonegotiation, and force a renegotiate */ - bmcr = mii->mdio_read(dev, mii->phy_id, MII_BMCR); - bmcr |= (BMCR_ANENABLE | BMCR_ANRESTART); - mii->mdio_write(dev, mii->phy_id, MII_BMCR, bmcr); - - mii->force_media = 0; - } else { - u32 bmcr, tmp; - - /* turn off auto negotiation, set speed and duplexity */ - bmcr = mii->mdio_read(dev, mii->phy_id, MII_BMCR); - tmp = bmcr & ~(BMCR_ANENABLE | BMCR_SPEED100 | BMCR_FULLDPLX); - if (ecmd->speed == SPEED_100) - tmp |= BMCR_SPEED100; - if (ecmd->duplex == DUPLEX_FULL) { - tmp |= BMCR_FULLDPLX; - mii->full_duplex = 1; - } else - mii->full_duplex = 0; - if (bmcr != tmp) - mii->mdio_write(dev, mii->phy_id, MII_BMCR, tmp); - - mii->force_media = 1; - } - return 0; -} - -int _kc_mii_link_ok (struct mii_if_info *mii) -{ - /* first, a dummy read, needed to latch some MII phys */ - mii->mdio_read(mii->dev, mii->phy_id, MII_BMSR); - if (mii->mdio_read(mii->dev, mii->phy_id, MII_BMSR) & BMSR_LSTATUS) - return 1; - return 0; -} - -int _kc_mii_nway_restart (struct mii_if_info *mii) -{ - int bmcr; - int r = -EINVAL; - - /* if autoneg is off, it's an error */ - bmcr = mii->mdio_read(mii->dev, mii->phy_id, MII_BMCR); - - if (bmcr & BMCR_ANENABLE) { - bmcr |= BMCR_ANRESTART; - mii->mdio_write(mii->dev, mii->phy_id, MII_BMCR, bmcr); - r = 0; - } - - return r; -} - -void _kc_mii_check_link (struct mii_if_info *mii) -{ - int cur_link = mii_link_ok(mii); - int prev_link = netif_carrier_ok(mii->dev); - - if (cur_link && !prev_link) - netif_carrier_on(mii->dev); - else if (prev_link && !cur_link) - netif_carrier_off(mii->dev); -} - -#if ( LINUX_VERSION_CODE > KERNEL_VERSION(2,4,6) ) -int _kc_generic_mii_ioctl(struct mii_if_info *mii_if, - struct mii_ioctl_data *mii_data, int cmd, - unsigned int *duplex_chg_out) -{ - int rc = 0; - unsigned int duplex_changed = 0; - - if (duplex_chg_out) - *duplex_chg_out = 0; - - mii_data->phy_id &= mii_if->phy_id_mask; - mii_data->reg_num &= mii_if->reg_num_mask; - - switch(cmd) { - case SIOCDEVPRIVATE: /* binary compat, remove in 2.5 */ - case SIOCGMIIPHY: - mii_data->phy_id = mii_if->phy_id; - /* fall through */ - - case SIOCDEVPRIVATE + 1:/* binary compat, remove in 2.5 */ - case SIOCGMIIREG: - mii_data->val_out = - mii_if->mdio_read(mii_if->dev, mii_data->phy_id, - mii_data->reg_num); - break; - - case SIOCDEVPRIVATE + 2:/* binary compat, remove in 2.5 */ - case SIOCSMIIREG: { - u16 val = mii_data->val_in; - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - if (mii_data->phy_id == mii_if->phy_id) { - switch(mii_data->reg_num) { - case MII_BMCR: { - unsigned int new_duplex = 0; - if (val & (BMCR_RESET|BMCR_ANENABLE)) - mii_if->force_media = 0; - else - mii_if->force_media = 1; - if (mii_if->force_media && - (val & BMCR_FULLDPLX)) - new_duplex = 1; - if (mii_if->full_duplex != new_duplex) { - duplex_changed = 1; - mii_if->full_duplex = new_duplex; - } - break; - } - case MII_ADVERTISE: - mii_if->advertising = val; - break; - default: - /* do nothing */ - break; - } - } - - mii_if->mdio_write(mii_if->dev, mii_data->phy_id, - mii_data->reg_num, val); - break; - } - - default: - rc = -EOPNOTSUPP; - break; - } - - if ((rc == 0) && (duplex_chg_out) && (duplex_changed)) - *duplex_chg_out = 1; - - return rc; -} -#endif /* > 2.4.6 */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe.h index 222c2c71..59415469 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.c index 24015844..e17b7f18 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.h index c6abb020..00a584f4 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82598.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.c index c6f4130d..30de47eb 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.h index 02be92ab..41024400 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_82599.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.c index ef7ce629..f00fe796 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.h index a6ab30d2..98b74000 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_api.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.c index 93659ca0..88b33fa0 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.h index 9bd6f534..6ae5926f 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_common.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_dcb.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_dcb.h index a6690451..5e6f9ac9 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_dcb.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_dcb.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_ethtool.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_ethtool.c index 11472bd3..bc3cb2f4 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_ethtool.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_ethtool.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_fcoe.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_fcoe.h index cad28622..48f7dcfc 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_fcoe.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_fcoe.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_main.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_main.c index 92fc9fc7..d26016c9 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_main.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_main.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> @@ -86,7 +86,7 @@ const char ixgbe_driver_version[] = DRV_VERSION; * { Vendor ID, Device ID, SubVendor ID, SubDevice ID, * Class, Class Mask, private data (not used) } */ -DEFINE_PCI_DEVICE_TABLE(ixgbe_pci_tbl) = { +const struct pci_device_id ixgbe_pci_tbl[] = { {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82598)}, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82598AF_DUAL_PORT)}, {PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82598AF_SINGLE_PORT)}, diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_mbx.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_mbx.h index 124f00de..5ced84f8 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_mbx.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_mbx.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_osdep.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_osdep.h index d161600b..c6f8e21f 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_osdep.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_osdep.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.c index e3f5275e..234fa632 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.h index bbe5a9e3..5ae171ac 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_phy.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_sriov.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_sriov.h deleted file mode 100644 index 5e3559fd..00000000 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_sriov.h +++ /dev/null @@ -1,73 +0,0 @@ -/******************************************************************************* - - Intel 10 Gigabit PCI Express Linux driver - Copyright(c) 1999 - 2012 Intel Corporation. - - This program is free software; you can redistribute it and/or modify it - under the terms and conditions of the GNU General Public License, - version 2, as published by the Free Software Foundation. - - This program is distributed in the hope 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; if not, write to the Free Software Foundation, Inc., - 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - - The full GNU General Public License is included in this distribution in - the file called "COPYING". - - Contact Information: - e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> - Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 - -*******************************************************************************/ - - -#ifndef _IXGBE_SRIOV_H_ -#define _IXGBE_SRIOV_H_ - -int ixgbe_set_vf_multicasts(struct ixgbe_adapter *adapter, - int entries, u16 *hash_list, u32 vf); -void ixgbe_restore_vf_multicasts(struct ixgbe_adapter *adapter); -int ixgbe_set_vf_vlan(struct ixgbe_adapter *adapter, int add, int vid, u32 vf); -void ixgbe_set_vmolr(struct ixgbe_hw *hw, u32 vf, bool aupe); -void ixgbe_vf_reset_event(struct ixgbe_adapter *adapter, u32 vf); -void ixgbe_vf_reset_msg(struct ixgbe_adapter *adapter, u32 vf); -void ixgbe_msg_task(struct ixgbe_adapter *adapter); -int ixgbe_set_vf_mac(struct ixgbe_adapter *adapter, - int vf, unsigned char *mac_addr); -void ixgbe_disable_tx_rx(struct ixgbe_adapter *adapter); -void ixgbe_ping_all_vfs(struct ixgbe_adapter *adapter); -#ifdef IFLA_VF_MAX -int ixgbe_ndo_set_vf_mac(struct net_device *netdev, int queue, u8 *mac); -int ixgbe_ndo_set_vf_vlan(struct net_device *netdev, int queue, u16 vlan, - u8 qos); -int ixgbe_ndo_set_vf_bw(struct net_device *netdev, int vf, int tx_rate); -#ifdef HAVE_VF_SPOOFCHK_CONFIGURE -int ixgbe_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting); -#endif -int ixgbe_ndo_get_vf_config(struct net_device *netdev, - int vf, struct ifla_vf_info *ivi); -#endif -void ixgbe_disable_sriov(struct ixgbe_adapter *adapter); -#ifdef CONFIG_PCI_IOV -int ixgbe_vf_configuration(struct pci_dev *pdev, unsigned int event_mask); -void ixgbe_enable_sriov(struct ixgbe_adapter *adapter); -#endif -int ixgbe_check_vf_assignment(struct ixgbe_adapter *adapter); -#ifdef IFLA_VF_MAX -void ixgbe_check_vf_rate_limit(struct ixgbe_adapter *adapter); -#endif /* IFLA_VF_MAX */ -void ixgbe_dump_registers(struct ixgbe_adapter *adapter); - -/* - * These are defined in ixgbe_type.h on behalf of the VF driver - * but we need them here unwrapped for the PF driver. - */ -#define IXGBE_DEV_ID_82599_VF 0x10ED -#define IXGBE_DEV_ID_X540_VF 0x1515 - -#endif /* _IXGBE_SRIOV_H_ */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_type.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_type.h index 6b21c879..bda61fa4 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_type.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_type.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.c index b99d9e84..2affe242 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.h index 77e8952d..38bcc87b 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/ixgbe_x540.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.c b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.c index 5f2523ed..d84c7ccb 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.c @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.h b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.h index bf27579b..4c7a6408 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/ethtool/ixgbe/kcompat.h @@ -17,7 +17,7 @@ 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. The full GNU General Public License is included in this distribution in - the file called "COPYING". + the file called "LICENSE.GPL". Contact Information: e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> @@ -3140,4 +3140,16 @@ static inline int __kc_pci_vfs_assigned(struct pci_dev *dev) #define SET_ETHTOOL_OPS(netdev, ops) ((netdev)->ethtool_ops = (ops)) #endif /* >= 3.16.0 */ +/* + * vlan_tx_tag_* macros renamed to skb_vlan_tag_* (Linux commit: df8a39defad4) + * For older kernels backported this commit, need to use renamed functions. + * This fix is specific to RedHat/CentOS kernels. + */ +#if (defined(RHEL_RELEASE_CODE) && \ + RHEL_RELEASE_CODE >= RHEL_RELEASE_VERSION(6, 8) && \ + LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 34)) +#define vlan_tx_tag_get skb_vlan_tag_get +#define vlan_tx_tag_present skb_vlan_tag_present +#endif + #endif /* _KCOMPAT_H_ */ diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_dev.h b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_dev.h index a0e5cb6b..58cbadd3 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_dev.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_dev.h @@ -25,6 +25,11 @@ #ifndef _KNI_DEV_H_ #define _KNI_DEV_H_ +#ifdef pr_fmt +#undef pr_fmt +#endif +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/if.h> #include <linux/wait.h> #include <linux/sched.h> @@ -39,10 +44,11 @@ #include <exec-env/rte_kni_common.h> #define KNI_KTHREAD_RESCHEDULE_INTERVAL 5 /* us */ +#define MBUF_BURST_SZ 32 + /** * A structure describing the private information for a kni device. */ - struct kni_dev { /* kni list */ struct list_head list; @@ -50,7 +56,7 @@ struct kni_dev { struct net_device_stats stats; int status; uint16_t group_id; /* Group ID of a group of KNI devices */ - unsigned core_id; /* Core ID to bind */ + uint32_t core_id; /* Core ID to bind */ char name[RTE_KNI_NAMESIZE]; /* Network device name */ struct task_struct *pthread; @@ -84,38 +90,36 @@ struct kni_dev { /* response queue */ void *resp_q; - void * sync_kva; + void *sync_kva; void *sync_va; void *mbuf_kva; void *mbuf_va; /* mbuf size */ - unsigned mbuf_size; + uint32_t mbuf_size; /* synchro for request processing */ unsigned long synchro; #ifdef RTE_KNI_VHOST - struct kni_vhost_queue* vhost_queue; + struct kni_vhost_queue *vhost_queue; + volatile enum { BE_STOP = 0x1, BE_START = 0x2, BE_FINISH = 0x4, - }vq_status; + } vq_status; #endif + /* buffers */ + void *pa[MBUF_BURST_SZ]; + void *va[MBUF_BURST_SZ]; + void *alloc_pa[MBUF_BURST_SZ]; + void *alloc_va[MBUF_BURST_SZ]; }; -#define KNI_ERR(args...) printk(KERN_DEBUG "KNI: Error: " args) -#define KNI_PRINT(args...) printk(KERN_DEBUG "KNI: " args) -#ifdef RTE_KNI_KO_DEBUG - #define KNI_DBG(args...) printk(KERN_DEBUG "KNI: " args) -#else - #define KNI_DBG(args...) -#endif - #ifdef RTE_KNI_VHOST -unsigned int +uint32_t kni_poll(struct file *file, struct socket *sock, poll_table * wait); int kni_chk_vhost_rx(struct kni_dev *kni); int kni_vhost_init(struct kni_dev *kni); @@ -127,23 +131,22 @@ struct kni_vhost_queue { int vnet_hdr_sz; struct kni_dev *kni; int sockfd; - unsigned int flags; - struct sk_buff* cache; - struct rte_kni_fifo* fifo; + uint32_t flags; + struct sk_buff *cache; + struct rte_kni_fifo *fifo; }; #endif -#ifdef RTE_KNI_VHOST_DEBUG_RX - #define KNI_DBG_RX(args...) printk(KERN_DEBUG "KNI RX: " args) -#else - #define KNI_DBG_RX(args...) -#endif +void kni_net_rx(struct kni_dev *kni); +void kni_net_init(struct net_device *dev); +void kni_net_config_lo_mode(char *lo_str); +void kni_net_poll_resp(struct kni_dev *kni); +void kni_set_ethtool_ops(struct net_device *netdev); -#ifdef RTE_KNI_VHOST_DEBUG_TX - #define KNI_DBG_TX(args...) printk(KERN_DEBUG "KNI TX: " args) -#else - #define KNI_DBG_TX(args...) -#endif +int ixgbe_kni_probe(struct pci_dev *pdev, struct net_device **lad_dev); +void ixgbe_kni_remove(struct pci_dev *pdev); +int igb_kni_probe(struct pci_dev *pdev, struct net_device **lad_dev); +void igb_kni_remove(struct pci_dev *pdev); #endif diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_ethtool.c b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_ethtool.c index 06b6d463..0c88589c 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_ethtool.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_ethtool.c @@ -31,6 +31,7 @@ static int kni_check_if_running(struct net_device *dev) { struct kni_dev *priv = netdev_priv(dev); + if (priv->lad_dev) return 0; else @@ -41,6 +42,7 @@ static void kni_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_drvinfo(priv->lad_dev, info); } @@ -48,6 +50,7 @@ static int kni_get_settings(struct net_device *dev, struct ethtool_cmd *ecmd) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->get_settings(priv->lad_dev, ecmd); } @@ -55,6 +58,7 @@ static int kni_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->set_settings(priv->lad_dev, ecmd); } @@ -62,6 +66,7 @@ static void kni_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_wol(priv->lad_dev, wol); } @@ -69,6 +74,7 @@ static int kni_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->set_wol(priv->lad_dev, wol); } @@ -76,6 +82,7 @@ static int kni_nway_reset(struct net_device *dev) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->nway_reset(priv->lad_dev); } @@ -83,6 +90,7 @@ static int kni_get_eeprom_len(struct net_device *dev) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->get_eeprom_len(priv->lad_dev); } @@ -91,6 +99,7 @@ kni_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *bytes) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->get_eeprom(priv->lad_dev, eeprom, bytes); } @@ -100,6 +109,7 @@ kni_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *bytes) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->set_eeprom(priv->lad_dev, eeprom, bytes); } @@ -108,6 +118,7 @@ static void kni_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ring) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_ringparam(priv->lad_dev, ring); } @@ -115,6 +126,7 @@ static int kni_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ring) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->set_ringparam(priv->lad_dev, ring); } @@ -122,6 +134,7 @@ static void kni_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pause) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_pauseparam(priv->lad_dev, pause); } @@ -129,6 +142,7 @@ static int kni_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *pause) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->set_pauseparam(priv->lad_dev, pause); } @@ -137,6 +151,7 @@ static u32 kni_get_msglevel(struct net_device *dev) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->get_msglevel(priv->lad_dev); } @@ -144,6 +159,7 @@ static void kni_set_msglevel(struct net_device *dev, u32 data) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->set_msglevel(priv->lad_dev, data); } @@ -151,6 +167,7 @@ static int kni_get_regs_len(struct net_device *dev) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->get_regs_len(priv->lad_dev); } @@ -158,6 +175,7 @@ static void kni_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *p) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_regs(priv->lad_dev, regs, p); } @@ -165,6 +183,7 @@ static void kni_get_strings(struct net_device *dev, u32 stringset, u8 *data) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_strings(priv->lad_dev, stringset, data); } @@ -173,6 +192,7 @@ static int kni_get_sset_count(struct net_device *dev, int sset) { struct kni_dev *priv = netdev_priv(dev); + return priv->lad_dev->ethtool_ops->get_sset_count(priv->lad_dev, sset); } @@ -181,24 +201,25 @@ kni_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *data) { struct kni_dev *priv = netdev_priv(dev); + priv->lad_dev->ethtool_ops->get_ethtool_stats(priv->lad_dev, stats, data); } struct ethtool_ops kni_ethtool_ops = { - .begin = kni_check_if_running, + .begin = kni_check_if_running, .get_drvinfo = kni_get_drvinfo, .get_settings = kni_get_settings, .set_settings = kni_set_settings, .get_regs_len = kni_get_regs_len, - .get_regs = kni_get_regs, - .get_wol = kni_get_wol, - .set_wol = kni_set_wol, - .nway_reset = kni_nway_reset, - .get_link = ethtool_op_get_link, + .get_regs = kni_get_regs, + .get_wol = kni_get_wol, + .set_wol = kni_set_wol, + .nway_reset = kni_nway_reset, + .get_link = ethtool_op_get_link, .get_eeprom_len = kni_get_eeprom_len, - .get_eeprom = kni_get_eeprom, - .set_eeprom = kni_set_eeprom, + .get_eeprom = kni_get_eeprom, + .set_eeprom = kni_set_eeprom, .get_ringparam = kni_get_ringparam, .set_ringparam = kni_set_ringparam, .get_pauseparam = kni_get_pauseparam, @@ -207,7 +228,7 @@ struct ethtool_ops kni_ethtool_ops = { .set_msglevel = kni_set_msglevel, .get_strings = kni_get_strings, .get_sset_count = kni_get_sset_count, - .get_ethtool_stats = kni_get_ethtool_stats, + .get_ethtool_stats = kni_get_ethtool_stats, }; void diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_fifo.h b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_fifo.h index 3ea750e2..025ec1c9 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_fifo.h +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_fifo.h @@ -30,13 +30,13 @@ /** * Adds num elements into the fifo. Return the number actually written */ -static inline unsigned -kni_fifo_put(struct rte_kni_fifo *fifo, void **data, unsigned num) +static inline uint32_t +kni_fifo_put(struct rte_kni_fifo *fifo, void **data, uint32_t num) { - unsigned i = 0; - unsigned fifo_write = fifo->write; - unsigned fifo_read = fifo->read; - unsigned new_write = fifo_write; + uint32_t i = 0; + uint32_t fifo_write = fifo->write; + uint32_t fifo_read = fifo->read; + uint32_t new_write = fifo_write; for (i = 0; i < num; i++) { new_write = (new_write + 1) & (fifo->len - 1); @@ -54,12 +54,12 @@ kni_fifo_put(struct rte_kni_fifo *fifo, void **data, unsigned num) /** * Get up to num elements from the fifo. Return the number actully read */ -static inline unsigned -kni_fifo_get(struct rte_kni_fifo *fifo, void **data, unsigned num) +static inline uint32_t +kni_fifo_get(struct rte_kni_fifo *fifo, void **data, uint32_t num) { - unsigned i = 0; - unsigned new_read = fifo->read; - unsigned fifo_write = fifo->write; + uint32_t i = 0; + uint32_t new_read = fifo->read; + uint32_t fifo_write = fifo->write; for (i = 0; i < num; i++) { if (new_read == fifo_write) @@ -76,16 +76,16 @@ kni_fifo_get(struct rte_kni_fifo *fifo, void **data, unsigned num) /** * Get the num of elements in the fifo */ -static inline unsigned +static inline uint32_t kni_fifo_count(struct rte_kni_fifo *fifo) { - return (fifo->len + fifo->write - fifo->read) & ( fifo->len - 1); + return (fifo->len + fifo->write - fifo->read) & (fifo->len - 1); } /** * Get the num of available elements in the fifo */ -static inline unsigned +static inline uint32_t kni_fifo_free_count(struct rte_kni_fifo *fifo) { return (fifo->read - fifo->write - 1) & (fifo->len - 1); @@ -96,7 +96,7 @@ kni_fifo_free_count(struct rte_kni_fifo *fifo) * Initializes the kni fifo structure */ static inline void -kni_fifo_init(struct rte_kni_fifo *fifo, unsigned size) +kni_fifo_init(struct rte_kni_fifo *fifo, uint32_t size) { fifo->write = 0; fifo->read = 0; diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_misc.c b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_misc.c index 59d15ca6..33b61f2a 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_misc.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_misc.c @@ -30,6 +30,7 @@ #include <linux/pci.h> #include <linux/kthread.h> #include <linux/rwsem.h> +#include <linux/mutex.h> #include <linux/nsproxy.h> #include <net/net_namespace.h> #include <net/netns/generic.h> @@ -47,52 +48,15 @@ MODULE_DESCRIPTION("Kernel Module for managing kni devices"); #define KNI_MAX_DEVICES 32 -extern void kni_net_rx(struct kni_dev *kni); -extern void kni_net_init(struct net_device *dev); -extern void kni_net_config_lo_mode(char *lo_str); -extern void kni_net_poll_resp(struct kni_dev *kni); -extern void kni_set_ethtool_ops(struct net_device *netdev); - -extern int ixgbe_kni_probe(struct pci_dev *pdev, struct net_device **lad_dev); -extern void ixgbe_kni_remove(struct pci_dev *pdev); -extern int igb_kni_probe(struct pci_dev *pdev, struct net_device **lad_dev); -extern void igb_kni_remove(struct pci_dev *pdev); - -static int kni_open(struct inode *inode, struct file *file); -static int kni_release(struct inode *inode, struct file *file); -static int kni_ioctl(struct inode *inode, unsigned int ioctl_num, - unsigned long ioctl_param); -static int kni_compat_ioctl(struct inode *inode, unsigned int ioctl_num, - unsigned long ioctl_param); -static int kni_dev_remove(struct kni_dev *dev); - -static int __init kni_parse_kthread_mode(void); - -/* KNI processing for single kernel thread mode */ -static int kni_thread_single(void *unused); -/* KNI processing for multiple kernel thread mode */ -static int kni_thread_multiple(void *param); - -static struct file_operations kni_fops = { - .owner = THIS_MODULE, - .open = kni_open, - .release = kni_release, - .unlocked_ioctl = (void *)kni_ioctl, - .compat_ioctl = (void *)kni_compat_ioctl, -}; - -static struct miscdevice kni_misc = { - .minor = MISC_DYNAMIC_MINOR, - .name = KNI_DEVICE, - .fops = &kni_fops, -}; +extern const struct pci_device_id ixgbe_pci_tbl[]; +extern const struct pci_device_id igb_pci_tbl[]; /* loopback mode */ -static char *lo_mode = NULL; +static char *lo_mode; /* Kernel thread mode */ -static char *kthread_mode = NULL; -static unsigned multiple_kthread_on = 0; +static char *kthread_mode; +static uint32_t multiple_kthread_on; #define KNI_DEV_IN_USE_BIT_NUM 0 /* Bit number for device in use */ @@ -100,20 +64,24 @@ static int kni_net_id; struct kni_net { unsigned long device_in_use; /* device in use flag */ + struct mutex kni_kthread_lock; struct task_struct *kni_kthread; struct rw_semaphore kni_list_lock; struct list_head kni_list_head; }; -static int __net_init kni_init_net(struct net *net) +static int __net_init +kni_init_net(struct net *net) { #ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS struct kni_net *knet = net_generic(net, kni_net_id); + + memset(knet, 0, sizeof(*knet)); #else struct kni_net *knet; int ret; - knet = kmalloc(sizeof(struct kni_net), GFP_KERNEL); + knet = kzalloc(sizeof(struct kni_net), GFP_KERNEL); if (!knet) { ret = -ENOMEM; return ret; @@ -123,6 +91,8 @@ static int __net_init kni_init_net(struct net *net) /* Clear the bit of device in use */ clear_bit(KNI_DEV_IN_USE_BIT_NUM, &knet->device_in_use); + mutex_init(&knet->kni_kthread_lock); + init_rwsem(&knet->kni_list_lock); INIT_LIST_HEAD(&knet->kni_list_head); @@ -137,11 +107,15 @@ static int __net_init kni_init_net(struct net *net) #endif } -static void __net_exit kni_exit_net(struct net *net) +static void __net_exit +kni_exit_net(struct net *net) { -#ifndef HAVE_SIMPLIFIED_PERNET_OPERATIONS - struct kni_net *knet = net_generic(net, kni_net_id); + struct kni_net *knet __maybe_unused; + + knet = net_generic(net, kni_net_id); + mutex_destroy(&knet->kni_kthread_lock); +#ifndef HAVE_SIMPLIFIED_PERNET_OPERATIONS kfree(knet); #endif } @@ -155,72 +129,56 @@ static struct pernet_operations kni_net_ops = { #endif }; -static int __init -kni_init(void) +static int +kni_thread_single(void *data) { - int rc; - - KNI_PRINT("######## DPDK kni module loading ########\n"); - - if (kni_parse_kthread_mode() < 0) { - KNI_ERR("Invalid parameter for kthread_mode\n"); - return -EINVAL; - } + struct kni_net *knet = data; + int j; + struct kni_dev *dev; -#ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS - rc = register_pernet_subsys(&kni_net_ops); + while (!kthread_should_stop()) { + down_read(&knet->kni_list_lock); + for (j = 0; j < KNI_RX_LOOP_NUM; j++) { + list_for_each_entry(dev, &knet->kni_list_head, list) { +#ifdef RTE_KNI_VHOST + kni_chk_vhost_rx(dev); #else - rc = register_pernet_gen_subsys(&kni_net_id, &kni_net_ops); + kni_net_rx(dev); +#endif + kni_net_poll_resp(dev); + } + } + up_read(&knet->kni_list_lock); +#ifdef RTE_KNI_PREEMPT_DEFAULT + /* reschedule out for a while */ + schedule_timeout_interruptible( + usecs_to_jiffies(KNI_KTHREAD_RESCHEDULE_INTERVAL)); #endif - if (rc) - return -EPERM; - - rc = misc_register(&kni_misc); - if (rc != 0) { - KNI_ERR("Misc registration failed\n"); - goto out; } - /* Configure the lo mode according to the input parameter */ - kni_net_config_lo_mode(lo_mode); - - KNI_PRINT("######## DPDK kni module loaded ########\n"); - return 0; - -out: -#ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS - unregister_pernet_subsys(&kni_net_ops); -#else - register_pernet_gen_subsys(&kni_net_id, &kni_net_ops); -#endif - return rc; } -static void __exit -kni_exit(void) +static int +kni_thread_multiple(void *param) { - misc_deregister(&kni_misc); -#ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS - unregister_pernet_subsys(&kni_net_ops); + int j; + struct kni_dev *dev = (struct kni_dev *)param; + + while (!kthread_should_stop()) { + for (j = 0; j < KNI_RX_LOOP_NUM; j++) { +#ifdef RTE_KNI_VHOST + kni_chk_vhost_rx(dev); #else - register_pernet_gen_subsys(&kni_net_id, &kni_net_ops); + kni_net_rx(dev); #endif - KNI_PRINT("####### DPDK kni module unloaded #######\n"); -} - -static int __init -kni_parse_kthread_mode(void) -{ - if (!kthread_mode) - return 0; - - if (strcmp(kthread_mode, "single") == 0) - return 0; - else if (strcmp(kthread_mode, "multiple") == 0) - multiple_kthread_on = 1; - else - return -1; + kni_net_poll_resp(dev); + } +#ifdef RTE_KNI_PREEMPT_DEFAULT + schedule_timeout_interruptible( + usecs_to_jiffies(KNI_KTHREAD_RESCHEDULE_INTERVAL)); +#endif + } return 0; } @@ -235,21 +193,31 @@ kni_open(struct inode *inode, struct file *file) if (test_and_set_bit(KNI_DEV_IN_USE_BIT_NUM, &knet->device_in_use)) return -EBUSY; - /* Create kernel thread for single mode */ - if (multiple_kthread_on == 0) { - KNI_PRINT("Single kernel thread for all KNI devices\n"); - /* Create kernel thread for RX */ - knet->kni_kthread = kthread_run(kni_thread_single, (void *)knet, - "kni_single"); - if (IS_ERR(knet->kni_kthread)) { - KNI_ERR("Unable to create kernel threaed\n"); - return PTR_ERR(knet->kni_kthread); - } - } else - KNI_PRINT("Multiple kernel thread mode enabled\n"); - file->private_data = get_net(net); - KNI_PRINT("/dev/kni opened\n"); + pr_debug("/dev/kni opened\n"); + + return 0; +} + +static int +kni_dev_remove(struct kni_dev *dev) +{ + if (!dev) + return -ENODEV; + +#ifdef CONFIG_RTE_KNI_KMOD_ETHTOOL + if (dev->pci_dev) { + if (pci_match_id(ixgbe_pci_tbl, dev->pci_dev)) + ixgbe_kni_remove(dev->pci_dev); + else if (pci_match_id(igb_pci_tbl, dev->pci_dev)) + igb_kni_remove(dev->pci_dev); + } +#endif + + if (dev->net_dev) { + unregister_netdev(dev->net_dev); + free_netdev(dev->net_dev); + } return 0; } @@ -263,9 +231,13 @@ kni_release(struct inode *inode, struct file *file) /* Stop kernel thread for single mode */ if (multiple_kthread_on == 0) { + mutex_lock(&knet->kni_kthread_lock); /* Stop kernel thread */ - kthread_stop(knet->kni_kthread); - knet->kni_kthread = NULL; + if (knet->kni_kthread != NULL) { + kthread_stop(knet->kni_kthread); + knet->kni_kthread = NULL; + } + mutex_unlock(&knet->kni_kthread_lock); } down_write(&knet->kni_list_lock); @@ -288,121 +260,78 @@ kni_release(struct inode *inode, struct file *file) clear_bit(KNI_DEV_IN_USE_BIT_NUM, &knet->device_in_use); put_net(net); - KNI_PRINT("/dev/kni closed\n"); + pr_debug("/dev/kni closed\n"); return 0; } static int -kni_thread_single(void *data) +kni_check_param(struct kni_dev *kni, struct rte_kni_device_info *dev) { - struct kni_net *knet = data; - int j; - struct kni_dev *dev; + if (!kni || !dev) + return -1; - while (!kthread_should_stop()) { - down_read(&knet->kni_list_lock); - for (j = 0; j < KNI_RX_LOOP_NUM; j++) { - list_for_each_entry(dev, &knet->kni_list_head, list) { -#ifdef RTE_KNI_VHOST - kni_chk_vhost_rx(dev); -#else - kni_net_rx(dev); -#endif - kni_net_poll_resp(dev); - } - } - up_read(&knet->kni_list_lock); -#ifdef RTE_KNI_PREEMPT_DEFAULT - /* reschedule out for a while */ - schedule_timeout_interruptible(usecs_to_jiffies( \ - KNI_KTHREAD_RESCHEDULE_INTERVAL)); -#endif + /* Check if network name has been used */ + if (!strncmp(kni->name, dev->name, RTE_KNI_NAMESIZE)) { + pr_err("KNI name %s duplicated\n", dev->name); + return -1; } return 0; } static int -kni_thread_multiple(void *param) +kni_run_thread(struct kni_net *knet, struct kni_dev *kni, uint8_t force_bind) { - int j; - struct kni_dev *dev = (struct kni_dev *)param; - - while (!kthread_should_stop()) { - for (j = 0; j < KNI_RX_LOOP_NUM; j++) { -#ifdef RTE_KNI_VHOST - kni_chk_vhost_rx(dev); -#else - kni_net_rx(dev); -#endif - kni_net_poll_resp(dev); + /** + * Create a new kernel thread for multiple mode, set its core affinity, + * and finally wake it up. + */ + if (multiple_kthread_on) { + kni->pthread = kthread_create(kni_thread_multiple, + (void *)kni, "kni_%s", kni->name); + if (IS_ERR(kni->pthread)) { + kni_dev_remove(kni); + return -ECANCELED; } -#ifdef RTE_KNI_PREEMPT_DEFAULT - schedule_timeout_interruptible(usecs_to_jiffies( \ - KNI_KTHREAD_RESCHEDULE_INTERVAL)); -#endif - } - - return 0; -} - -static int -kni_dev_remove(struct kni_dev *dev) -{ - if (!dev) - return -ENODEV; - - switch (dev->device_id) { - #define RTE_PCI_DEV_ID_DECL_IGB(vend, dev) case (dev): - #include <rte_pci_dev_ids.h> - igb_kni_remove(dev->pci_dev); - break; - #define RTE_PCI_DEV_ID_DECL_IXGBE(vend, dev) case (dev): - #include <rte_pci_dev_ids.h> - ixgbe_kni_remove(dev->pci_dev); - break; - default: - break; - } - - if (dev->net_dev) { - unregister_netdev(dev->net_dev); - free_netdev(dev->net_dev); - } - return 0; -} + if (force_bind) + kthread_bind(kni->pthread, kni->core_id); + wake_up_process(kni->pthread); + } else { + mutex_lock(&knet->kni_kthread_lock); + + if (knet->kni_kthread == NULL) { + knet->kni_kthread = kthread_create(kni_thread_single, + (void *)knet, "kni_single"); + if (IS_ERR(knet->kni_kthread)) { + mutex_unlock(&knet->kni_kthread_lock); + kni_dev_remove(kni); + return -ECANCELED; + } -static int -kni_check_param(struct kni_dev *kni, struct rte_kni_device_info *dev) -{ - if (!kni || !dev) - return -1; + if (force_bind) + kthread_bind(knet->kni_kthread, kni->core_id); + wake_up_process(knet->kni_kthread); + } - /* Check if network name has been used */ - if (!strncmp(kni->name, dev->name, RTE_KNI_NAMESIZE)) { - KNI_ERR("KNI name %s duplicated\n", dev->name); - return -1; + mutex_unlock(&knet->kni_kthread_lock); } return 0; } static int -kni_ioctl_create(struct net *net, - unsigned int ioctl_num, unsigned long ioctl_param) +kni_ioctl_create(struct net *net, uint32_t ioctl_num, + unsigned long ioctl_param) { struct kni_net *knet = net_generic(net, kni_net_id); int ret; struct rte_kni_device_info dev_info; - struct pci_dev *pci = NULL; - struct pci_dev *found_pci = NULL; struct net_device *net_dev = NULL; - struct net_device *lad_dev = NULL; struct kni_dev *kni, *dev, *n; - printk(KERN_INFO "KNI: Creating kni...\n"); + pr_info("Creating kni...\n"); /* Check the buffer size, to avoid warning */ if (_IOC_SIZE(ioctl_num) > sizeof(dev_info)) return -EINVAL; @@ -410,17 +339,21 @@ kni_ioctl_create(struct net *net, /* Copy kni info from user space */ ret = copy_from_user(&dev_info, (void *)ioctl_param, sizeof(dev_info)); if (ret) { - KNI_ERR("copy_from_user in kni_ioctl_create"); + pr_err("copy_from_user in kni_ioctl_create"); return -EIO; } + /* Check if name is zero-ended */ + if (strnlen(dev_info.name, sizeof(dev_info.name)) == sizeof(dev_info.name)) { + pr_err("kni.name not zero-terminated"); + return -EINVAL; + } + /** - * Check if the cpu core id is valid for binding, - * for multiple kernel thread mode. + * Check if the cpu core id is valid for binding. */ - if (multiple_kthread_on && dev_info.force_bind && - !cpu_online(dev_info.core_id)) { - KNI_ERR("cpu %u is not online\n", dev_info.core_id); + if (dev_info.force_bind && !cpu_online(dev_info.core_id)) { + pr_err("cpu %u is not online\n", dev_info.core_id); return -EINVAL; } @@ -435,12 +368,12 @@ kni_ioctl_create(struct net *net, up_read(&knet->kni_list_lock); net_dev = alloc_netdev(sizeof(struct kni_dev), dev_info.name, -#ifdef NET_NAME_UNKNOWN - NET_NAME_UNKNOWN, +#ifdef NET_NAME_USER + NET_NAME_USER, #endif kni_net_init); if (net_dev == NULL) { - KNI_ERR("error allocating device \"%s\"\n", dev_info.name); + pr_err("error allocating device \"%s\"\n", dev_info.name); return -EBUSY; } @@ -464,44 +397,43 @@ kni_ioctl_create(struct net *net, kni->sync_va = dev_info.sync_va; kni->sync_kva = phys_to_virt(dev_info.sync_phys); - kni->mbuf_kva = phys_to_virt(dev_info.mbuf_phys); - kni->mbuf_va = dev_info.mbuf_va; - #ifdef RTE_KNI_VHOST kni->vhost_queue = NULL; kni->vq_status = BE_STOP; #endif kni->mbuf_size = dev_info.mbuf_size; - KNI_PRINT("tx_phys: 0x%016llx, tx_q addr: 0x%p\n", + pr_debug("tx_phys: 0x%016llx, tx_q addr: 0x%p\n", (unsigned long long) dev_info.tx_phys, kni->tx_q); - KNI_PRINT("rx_phys: 0x%016llx, rx_q addr: 0x%p\n", + pr_debug("rx_phys: 0x%016llx, rx_q addr: 0x%p\n", (unsigned long long) dev_info.rx_phys, kni->rx_q); - KNI_PRINT("alloc_phys: 0x%016llx, alloc_q addr: 0x%p\n", + pr_debug("alloc_phys: 0x%016llx, alloc_q addr: 0x%p\n", (unsigned long long) dev_info.alloc_phys, kni->alloc_q); - KNI_PRINT("free_phys: 0x%016llx, free_q addr: 0x%p\n", + pr_debug("free_phys: 0x%016llx, free_q addr: 0x%p\n", (unsigned long long) dev_info.free_phys, kni->free_q); - KNI_PRINT("req_phys: 0x%016llx, req_q addr: 0x%p\n", + pr_debug("req_phys: 0x%016llx, req_q addr: 0x%p\n", (unsigned long long) dev_info.req_phys, kni->req_q); - KNI_PRINT("resp_phys: 0x%016llx, resp_q addr: 0x%p\n", + pr_debug("resp_phys: 0x%016llx, resp_q addr: 0x%p\n", (unsigned long long) dev_info.resp_phys, kni->resp_q); - KNI_PRINT("mbuf_phys: 0x%016llx, mbuf_kva: 0x%p\n", - (unsigned long long) dev_info.mbuf_phys, kni->mbuf_kva); - KNI_PRINT("mbuf_va: 0x%p\n", dev_info.mbuf_va); - KNI_PRINT("mbuf_size: %u\n", kni->mbuf_size); + pr_debug("mbuf_size: %u\n", kni->mbuf_size); - KNI_DBG("PCI: %02x:%02x.%02x %04x:%04x\n", + pr_debug("PCI: %02x:%02x.%02x %04x:%04x\n", dev_info.bus, dev_info.devid, dev_info.function, dev_info.vendor_id, dev_info.device_id); +#ifdef CONFIG_RTE_KNI_KMOD_ETHTOOL + struct pci_dev *found_pci = NULL; + struct net_device *lad_dev = NULL; + struct pci_dev *pci = NULL; + pci = pci_get_device(dev_info.vendor_id, dev_info.device_id, NULL); /* Support Ethtool */ while (pci) { - KNI_PRINT("pci_bus: %02x:%02x:%02x \n", + pr_debug("pci_bus: %02x:%02x:%02x\n", pci->bus->number, PCI_SLOT(pci->devfn), PCI_FUNC(pci->devfn)); @@ -510,28 +442,21 @@ kni_ioctl_create(struct net *net, (PCI_SLOT(pci->devfn) == dev_info.devid) && (PCI_FUNC(pci->devfn) == dev_info.function)) { found_pci = pci; - switch (dev_info.device_id) { - #define RTE_PCI_DEV_ID_DECL_IGB(vend, dev) case (dev): - #include <rte_pci_dev_ids.h> - ret = igb_kni_probe(found_pci, &lad_dev); - break; - #define RTE_PCI_DEV_ID_DECL_IXGBE(vend, dev) \ - case (dev): - #include <rte_pci_dev_ids.h> + + if (pci_match_id(ixgbe_pci_tbl, found_pci)) ret = ixgbe_kni_probe(found_pci, &lad_dev); - break; - default: + else if (pci_match_id(igb_pci_tbl, found_pci)) + ret = igb_kni_probe(found_pci, &lad_dev); + else ret = -1; - break; - } - KNI_DBG("PCI found: pci=0x%p, lad_dev=0x%p\n", + pr_debug("PCI found: pci=0x%p, lad_dev=0x%p\n", pci, lad_dev); if (ret == 0) { kni->lad_dev = lad_dev; kni_set_ethtool_ops(kni->net_dev); } else { - KNI_ERR("Device not supported by ethtool"); + pr_err("Device not supported by ethtool"); kni->lad_dev = NULL; } @@ -544,9 +469,10 @@ kni_ioctl_create(struct net *net, } if (pci) pci_dev_put(pci); +#endif if (kni->lad_dev) - memcpy(net_dev->dev_addr, kni->lad_dev->dev_addr, ETH_ALEN); + ether_addr_copy(net_dev->dev_addr, kni->lad_dev->dev_addr); else /* * Generate random mac address. eth_random_addr() is the newer @@ -556,9 +482,11 @@ kni_ioctl_create(struct net *net, ret = register_netdev(net_dev); if (ret) { - KNI_ERR("error %i registering device \"%s\"\n", + pr_err("error %i registering device \"%s\"\n", ret, dev_info.name); + kni->net_dev = NULL; kni_dev_remove(kni); + free_netdev(net_dev); return -ENODEV; } @@ -566,22 +494,9 @@ kni_ioctl_create(struct net *net, kni_vhost_init(kni); #endif - /** - * Create a new kernel thread for multiple mode, set its core affinity, - * and finally wake it up. - */ - if (multiple_kthread_on) { - kni->pthread = kthread_create(kni_thread_multiple, - (void *)kni, - "kni_%s", kni->name); - if (IS_ERR(kni->pthread)) { - kni_dev_remove(kni); - return -ECANCELED; - } - if (dev_info.force_bind) - kthread_bind(kni->pthread, kni->core_id); - wake_up_process(kni->pthread); - } + ret = kni_run_thread(knet, kni, dev_info.force_bind); + if (ret != 0) + return ret; down_write(&knet->kni_list_lock); list_add(&kni->list, &knet->kni_list_head); @@ -591,8 +506,8 @@ kni_ioctl_create(struct net *net, } static int -kni_ioctl_release(struct net *net, - unsigned int ioctl_num, unsigned long ioctl_param) +kni_ioctl_release(struct net *net, uint32_t ioctl_num, + unsigned long ioctl_param) { struct kni_net *knet = net_generic(net, kni_net_id); int ret = -EINVAL; @@ -600,11 +515,11 @@ kni_ioctl_release(struct net *net, struct rte_kni_device_info dev_info; if (_IOC_SIZE(ioctl_num) > sizeof(dev_info)) - return -EINVAL; + return -EINVAL; ret = copy_from_user(&dev_info, (void *)ioctl_param, sizeof(dev_info)); if (ret) { - KNI_ERR("copy_from_user in kni_ioctl_release"); + pr_err("copy_from_user in kni_ioctl_release"); return -EIO; } @@ -631,21 +546,19 @@ kni_ioctl_release(struct net *net, break; } up_write(&knet->kni_list_lock); - printk(KERN_INFO "KNI: %s release kni named %s\n", + pr_info("%s release kni named %s\n", (ret == 0 ? "Successfully" : "Unsuccessfully"), dev_info.name); return ret; } static int -kni_ioctl(struct inode *inode, - unsigned int ioctl_num, - unsigned long ioctl_param) +kni_ioctl(struct inode *inode, uint32_t ioctl_num, unsigned long ioctl_param) { int ret = -EINVAL; struct net *net = current->nsproxy->net_ns; - KNI_DBG("IOCTL num=0x%0x param=0x%0lx\n", ioctl_num, ioctl_param); + pr_debug("IOCTL num=0x%0x param=0x%0lx\n", ioctl_num, ioctl_param); /* * Switch according to the ioctl called @@ -661,7 +574,7 @@ kni_ioctl(struct inode *inode, ret = kni_ioctl_release(net, ioctl_num, ioctl_param); break; default: - KNI_DBG("IOCTL default\n"); + pr_debug("IOCTL default\n"); break; } @@ -669,16 +582,99 @@ kni_ioctl(struct inode *inode, } static int -kni_compat_ioctl(struct inode *inode, - unsigned int ioctl_num, +kni_compat_ioctl(struct inode *inode, uint32_t ioctl_num, unsigned long ioctl_param) { /* 32 bits app on 64 bits OS to be supported later */ - KNI_PRINT("Not implemented.\n"); + pr_debug("Not implemented.\n"); return -EINVAL; } +static const struct file_operations kni_fops = { + .owner = THIS_MODULE, + .open = kni_open, + .release = kni_release, + .unlocked_ioctl = (void *)kni_ioctl, + .compat_ioctl = (void *)kni_compat_ioctl, +}; + +static struct miscdevice kni_misc = { + .minor = MISC_DYNAMIC_MINOR, + .name = KNI_DEVICE, + .fops = &kni_fops, +}; + +static int __init +kni_parse_kthread_mode(void) +{ + if (!kthread_mode) + return 0; + + if (strcmp(kthread_mode, "single") == 0) + return 0; + else if (strcmp(kthread_mode, "multiple") == 0) + multiple_kthread_on = 1; + else + return -1; + + return 0; +} + +static int __init +kni_init(void) +{ + int rc; + + if (kni_parse_kthread_mode() < 0) { + pr_err("Invalid parameter for kthread_mode\n"); + return -EINVAL; + } + + if (multiple_kthread_on == 0) + pr_debug("Single kernel thread for all KNI devices\n"); + else + pr_debug("Multiple kernel thread mode enabled\n"); + +#ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS + rc = register_pernet_subsys(&kni_net_ops); +#else + rc = register_pernet_gen_subsys(&kni_net_id, &kni_net_ops); +#endif + if (rc) + return -EPERM; + + rc = misc_register(&kni_misc); + if (rc != 0) { + pr_err("Misc registration failed\n"); + goto out; + } + + /* Configure the lo mode according to the input parameter */ + kni_net_config_lo_mode(lo_mode); + + return 0; + +out: +#ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS + unregister_pernet_subsys(&kni_net_ops); +#else + unregister_pernet_gen_subsys(kni_net_id, &kni_net_ops); +#endif + return rc; +} + +static void __exit +kni_exit(void) +{ + misc_deregister(&kni_misc); +#ifdef HAVE_SIMPLIFIED_PERNET_OPERATIONS + unregister_pernet_subsys(&kni_net_ops); +#else + unregister_pernet_gen_subsys(kni_net_id, &kni_net_ops); +#endif +} + module_init(kni_init); module_exit(kni_exit); diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_net.c b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_net.c index fc82193a..4ac99cfe 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_net.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_net.c @@ -44,23 +44,103 @@ #define WD_TIMEOUT 5 /*jiffies */ -#define MBUF_BURST_SZ 32 - #define KNI_WAIT_RESPONSE_TIMEOUT 300 /* 3 seconds */ /* typedef for rx function */ typedef void (*kni_net_rx_t)(struct kni_dev *kni); -static int kni_net_tx(struct sk_buff *skb, struct net_device *dev); static void kni_net_rx_normal(struct kni_dev *kni); -static void kni_net_rx_lo_fifo(struct kni_dev *kni); -static void kni_net_rx_lo_fifo_skb(struct kni_dev *kni); -static int kni_net_process_request(struct kni_dev *kni, - struct rte_kni_request *req); /* kni rx function pointer, with default to normal rx */ static kni_net_rx_t kni_net_rx_func = kni_net_rx_normal; +/* physical address to kernel virtual address */ +static void * +pa2kva(void *pa) +{ + return phys_to_virt((unsigned long)pa); +} + +/* physical address to virtual address */ +static void * +pa2va(void *pa, struct rte_kni_mbuf *m) +{ + void *va; + + va = (void *)((unsigned long)pa + + (unsigned long)m->buf_addr - + (unsigned long)m->buf_physaddr); + return va; +} + +/* mbuf data kernel virtual address from mbuf kernel virtual address */ +static void * +kva2data_kva(struct rte_kni_mbuf *m) +{ + return phys_to_virt(m->buf_physaddr + m->data_off); +} + +/* virtual address to physical address */ +static void * +va2pa(void *va, struct rte_kni_mbuf *m) +{ + void *pa; + + pa = (void *)((unsigned long)va - + ((unsigned long)m->buf_addr - + (unsigned long)m->buf_physaddr)); + return pa; +} + +/* + * It can be called to process the request. + */ +static int +kni_net_process_request(struct kni_dev *kni, struct rte_kni_request *req) +{ + int ret = -1; + void *resp_va; + uint32_t num; + int ret_val; + + if (!kni || !req) { + pr_err("No kni instance or request\n"); + return -EINVAL; + } + + mutex_lock(&kni->sync_lock); + + /* Construct data */ + memcpy(kni->sync_kva, req, sizeof(struct rte_kni_request)); + num = kni_fifo_put(kni->req_q, &kni->sync_va, 1); + if (num < 1) { + pr_err("Cannot send to req_q\n"); + ret = -EBUSY; + goto fail; + } + + ret_val = wait_event_interruptible_timeout(kni->wq, + kni_fifo_count(kni->resp_q), 3 * HZ); + if (signal_pending(current) || ret_val <= 0) { + ret = -ETIME; + goto fail; + } + num = kni_fifo_get(kni->resp_q, (void **)&resp_va, 1); + if (num != 1 || resp_va != kni->sync_va) { + /* This should never happen */ + pr_err("No data in resp_q\n"); + ret = -ENODATA; + goto fail; + } + + memcpy(req, kni->sync_kva, sizeof(struct rte_kni_request)); + ret = 0; + +fail: + mutex_unlock(&kni->sync_lock); + return ret; +} + /* * Open and close */ @@ -116,18 +196,112 @@ kni_net_config(struct net_device *dev, struct ifmap *map) } /* + * Transmit a packet (called by the kernel) + */ +#ifdef RTE_KNI_VHOST +static int +kni_net_tx(struct sk_buff *skb, struct net_device *dev) +{ + struct kni_dev *kni = netdev_priv(dev); + + dev_kfree_skb(skb); + kni->stats.tx_dropped++; + + return NETDEV_TX_OK; +} +#else +static int +kni_net_tx(struct sk_buff *skb, struct net_device *dev) +{ + int len = 0; + uint32_t ret; + struct kni_dev *kni = netdev_priv(dev); + struct rte_kni_mbuf *pkt_kva = NULL; + void *pkt_pa = NULL; + void *pkt_va = NULL; + + /* save the timestamp */ +#ifdef HAVE_TRANS_START_HELPER + netif_trans_update(dev); +#else + dev->trans_start = jiffies; +#endif + + /* Check if the length of skb is less than mbuf size */ + if (skb->len > kni->mbuf_size) + goto drop; + + /** + * Check if it has at least one free entry in tx_q and + * one entry in alloc_q. + */ + if (kni_fifo_free_count(kni->tx_q) == 0 || + kni_fifo_count(kni->alloc_q) == 0) { + /** + * If no free entry in tx_q or no entry in alloc_q, + * drops skb and goes out. + */ + goto drop; + } + + /* dequeue a mbuf from alloc_q */ + ret = kni_fifo_get(kni->alloc_q, &pkt_pa, 1); + if (likely(ret == 1)) { + void *data_kva; + + pkt_kva = pa2kva(pkt_pa); + data_kva = kva2data_kva(pkt_kva); + pkt_va = pa2va(pkt_pa, pkt_kva); + + len = skb->len; + memcpy(data_kva, skb->data, len); + if (unlikely(len < ETH_ZLEN)) { + memset(data_kva + len, 0, ETH_ZLEN - len); + len = ETH_ZLEN; + } + pkt_kva->pkt_len = len; + pkt_kva->data_len = len; + + /* enqueue mbuf into tx_q */ + ret = kni_fifo_put(kni->tx_q, &pkt_va, 1); + if (unlikely(ret != 1)) { + /* Failing should not happen */ + pr_err("Fail to enqueue mbuf into tx_q\n"); + goto drop; + } + } else { + /* Failing should not happen */ + pr_err("Fail to dequeue mbuf from alloc_q\n"); + goto drop; + } + + /* Free skb and update statistics */ + dev_kfree_skb(skb); + kni->stats.tx_bytes += len; + kni->stats.tx_packets++; + + return NETDEV_TX_OK; + +drop: + /* Free skb and update statistics */ + dev_kfree_skb(skb); + kni->stats.tx_dropped++; + + return NETDEV_TX_OK; +} +#endif + +/* * RX: normal working mode */ static void kni_net_rx_normal(struct kni_dev *kni) { - unsigned ret; + uint32_t ret; uint32_t len; - unsigned i, num_rx, num_fq; + uint32_t i, num_rx, num_fq; struct rte_kni_mbuf *kva; - struct rte_kni_mbuf *va[MBUF_BURST_SZ]; - void * data_kva; - + void *data_kva; struct sk_buff *skb; struct net_device *dev = kni->net_dev; @@ -139,24 +313,22 @@ kni_net_rx_normal(struct kni_dev *kni) } /* Calculate the number of entries to dequeue from rx_q */ - num_rx = min(num_fq, (unsigned)MBUF_BURST_SZ); + num_rx = min_t(uint32_t, num_fq, MBUF_BURST_SZ); /* Burst dequeue from rx_q */ - num_rx = kni_fifo_get(kni->rx_q, (void **)va, num_rx); + num_rx = kni_fifo_get(kni->rx_q, kni->pa, num_rx); if (num_rx == 0) return; /* Transfer received packets to netif */ for (i = 0; i < num_rx; i++) { - kva = (void *)va[i] - kni->mbuf_va + kni->mbuf_kva; + kva = pa2kva(kni->pa[i]); len = kva->pkt_len; - - data_kva = kva->buf_addr + kva->data_off - kni->mbuf_va - + kni->mbuf_kva; + data_kva = kva2data_kva(kva); + kni->va[i] = pa2va(kni->pa[i], kva); skb = dev_alloc_skb(len + 2); if (!skb) { - KNI_ERR("Out of mem, dropping pkts\n"); /* Update statistics */ kni->stats.rx_dropped++; continue; @@ -178,9 +350,8 @@ kni_net_rx_normal(struct kni_dev *kni) if (!kva->next) break; - kva = kva->next - kni->mbuf_va + kni->mbuf_kva; - data_kva = kva->buf_addr + kva->data_off - - kni->mbuf_va + kni->mbuf_kva; + kva = pa2kva(va2pa(kva->next, kva)); + data_kva = kva2data_kva(kva); } } @@ -197,10 +368,10 @@ kni_net_rx_normal(struct kni_dev *kni) } /* Burst enqueue mbufs into free_q */ - ret = kni_fifo_put(kni->free_q, (void **)va, num_rx); + ret = kni_fifo_put(kni->free_q, kni->va, num_rx); if (ret != num_rx) /* Failing should not happen */ - KNI_ERR("Fail to enqueue entries into free_q\n"); + pr_err("Fail to enqueue entries into free_q\n"); } /* @@ -209,15 +380,12 @@ kni_net_rx_normal(struct kni_dev *kni) static void kni_net_rx_lo_fifo(struct kni_dev *kni) { - unsigned ret; + uint32_t ret; uint32_t len; - unsigned i, num, num_rq, num_tq, num_aq, num_fq; + uint32_t i, num, num_rq, num_tq, num_aq, num_fq; struct rte_kni_mbuf *kva; - struct rte_kni_mbuf *va[MBUF_BURST_SZ]; - void * data_kva; - + void *data_kva; struct rte_kni_mbuf *alloc_kva; - struct rte_kni_mbuf *alloc_va[MBUF_BURST_SZ]; void *alloc_data_kva; /* Get the number of entries in rx_q */ @@ -236,33 +404,32 @@ kni_net_rx_lo_fifo(struct kni_dev *kni) num = min(num_rq, num_tq); num = min(num, num_aq); num = min(num, num_fq); - num = min(num, (unsigned)MBUF_BURST_SZ); + num = min_t(uint32_t, num, MBUF_BURST_SZ); /* Return if no entry to dequeue from rx_q */ if (num == 0) return; /* Burst dequeue from rx_q */ - ret = kni_fifo_get(kni->rx_q, (void **)va, num); + ret = kni_fifo_get(kni->rx_q, kni->pa, num); if (ret == 0) return; /* Failing should not happen */ /* Dequeue entries from alloc_q */ - ret = kni_fifo_get(kni->alloc_q, (void **)alloc_va, num); + ret = kni_fifo_get(kni->alloc_q, kni->alloc_pa, num); if (ret) { num = ret; /* Copy mbufs */ for (i = 0; i < num; i++) { - kva = (void *)va[i] - kni->mbuf_va + kni->mbuf_kva; + kva = pa2kva(kni->pa[i]); len = kva->pkt_len; - data_kva = kva->buf_addr + kva->data_off - - kni->mbuf_va + kni->mbuf_kva; - - alloc_kva = (void *)alloc_va[i] - kni->mbuf_va + - kni->mbuf_kva; - alloc_data_kva = alloc_kva->buf_addr + - alloc_kva->data_off - kni->mbuf_va + - kni->mbuf_kva; + data_kva = kva2data_kva(kva); + kni->va[i] = pa2va(kni->pa[i], kva); + + alloc_kva = pa2kva(kni->alloc_pa[i]); + alloc_data_kva = kva2data_kva(alloc_kva); + kni->alloc_va[i] = pa2va(kni->alloc_pa[i], alloc_kva); + memcpy(alloc_data_kva, data_kva, len); alloc_kva->pkt_len = len; alloc_kva->data_len = len; @@ -272,17 +439,17 @@ kni_net_rx_lo_fifo(struct kni_dev *kni) } /* Burst enqueue mbufs into tx_q */ - ret = kni_fifo_put(kni->tx_q, (void **)alloc_va, num); + ret = kni_fifo_put(kni->tx_q, kni->alloc_va, num); if (ret != num) /* Failing should not happen */ - KNI_ERR("Fail to enqueue mbufs into tx_q\n"); + pr_err("Fail to enqueue mbufs into tx_q\n"); } /* Burst enqueue mbufs into free_q */ - ret = kni_fifo_put(kni->free_q, (void **)va, num); + ret = kni_fifo_put(kni->free_q, kni->va, num); if (ret != num) /* Failing should not happen */ - KNI_ERR("Fail to enqueue mbufs into free_q\n"); + pr_err("Fail to enqueue mbufs into free_q\n"); /** * Update statistic, and enqueue/dequeue failure is impossible, @@ -298,13 +465,11 @@ kni_net_rx_lo_fifo(struct kni_dev *kni) static void kni_net_rx_lo_fifo_skb(struct kni_dev *kni) { - unsigned ret; + uint32_t ret; uint32_t len; - unsigned i, num_rq, num_fq, num; + uint32_t i, num_rq, num_fq, num; struct rte_kni_mbuf *kva; - struct rte_kni_mbuf *va[MBUF_BURST_SZ]; - void * data_kva; - + void *data_kva; struct sk_buff *skb; struct net_device *dev = kni->net_dev; @@ -316,28 +481,26 @@ kni_net_rx_lo_fifo_skb(struct kni_dev *kni) /* Calculate the number of entries to dequeue from rx_q */ num = min(num_rq, num_fq); - num = min(num, (unsigned)MBUF_BURST_SZ); + num = min_t(uint32_t, num, MBUF_BURST_SZ); /* Return if no entry to dequeue from rx_q */ if (num == 0) return; /* Burst dequeue mbufs from rx_q */ - ret = kni_fifo_get(kni->rx_q, (void **)va, num); + ret = kni_fifo_get(kni->rx_q, kni->pa, num); if (ret == 0) return; /* Copy mbufs to sk buffer and then call tx interface */ for (i = 0; i < num; i++) { - kva = (void *)va[i] - kni->mbuf_va + kni->mbuf_kva; + kva = pa2kva(kni->pa[i]); len = kva->pkt_len; - data_kva = kva->buf_addr + kva->data_off - kni->mbuf_va + - kni->mbuf_kva; + data_kva = kva2data_kva(kva); + kni->va[i] = pa2va(kni->pa[i], kva); skb = dev_alloc_skb(len + 2); - if (skb == NULL) - KNI_ERR("Out of mem, dropping pkts\n"); - else { + if (skb) { /* Align IP on 16B boundary */ skb_reserve(skb, 2); memcpy(skb_put(skb, len), data_kva, len); @@ -349,7 +512,6 @@ kni_net_rx_lo_fifo_skb(struct kni_dev *kni) /* Simulate real usage, allocate/copy skb twice */ skb = dev_alloc_skb(len + 2); if (skb == NULL) { - KNI_ERR("Out of mem, dropping pkts\n"); kni->stats.rx_dropped++; continue; } @@ -370,9 +532,8 @@ kni_net_rx_lo_fifo_skb(struct kni_dev *kni) if (!kva->next) break; - kva = kva->next - kni->mbuf_va + kni->mbuf_kva; - data_kva = kva->buf_addr + kva->data_off - - kni->mbuf_va + kni->mbuf_kva; + kva = pa2kva(va2pa(kva->next, kva)); + data_kva = kva2data_kva(kva); } } @@ -387,10 +548,10 @@ kni_net_rx_lo_fifo_skb(struct kni_dev *kni) } /* enqueue all the mbufs from rx_q into free_q */ - ret = kni_fifo_put(kni->free_q, (void **)&va, num); + ret = kni_fifo_put(kni->free_q, kni->va, num); if (ret != num) /* Failing should not happen */ - KNI_ERR("Fail to enqueue mbufs into free_q\n"); + pr_err("Fail to enqueue mbufs into free_q\n"); } /* rx interface */ @@ -405,114 +566,18 @@ kni_net_rx(struct kni_dev *kni) } /* - * Transmit a packet (called by the kernel) - */ -#ifdef RTE_KNI_VHOST -static int -kni_net_tx(struct sk_buff *skb, struct net_device *dev) -{ - struct kni_dev *kni = netdev_priv(dev); - - dev_kfree_skb(skb); - kni->stats.tx_dropped++; - - return NETDEV_TX_OK; -} -#else -static int -kni_net_tx(struct sk_buff *skb, struct net_device *dev) -{ - int len = 0; - unsigned ret; - struct kni_dev *kni = netdev_priv(dev); - struct rte_kni_mbuf *pkt_kva = NULL; - struct rte_kni_mbuf *pkt_va = NULL; - - /* save the timestamp */ -#ifdef HAVE_TRANS_START_HELPER - netif_trans_update(dev); -#else - dev->trans_start = jiffies; -#endif - - /* Check if the length of skb is less than mbuf size */ - if (skb->len > kni->mbuf_size) - goto drop; - - /** - * Check if it has at least one free entry in tx_q and - * one entry in alloc_q. - */ - if (kni_fifo_free_count(kni->tx_q) == 0 || - kni_fifo_count(kni->alloc_q) == 0) { - /** - * If no free entry in tx_q or no entry in alloc_q, - * drops skb and goes out. - */ - goto drop; - } - - /* dequeue a mbuf from alloc_q */ - ret = kni_fifo_get(kni->alloc_q, (void **)&pkt_va, 1); - if (likely(ret == 1)) { - void *data_kva; - - pkt_kva = (void *)pkt_va - kni->mbuf_va + kni->mbuf_kva; - data_kva = pkt_kva->buf_addr + pkt_kva->data_off - kni->mbuf_va - + kni->mbuf_kva; - - len = skb->len; - memcpy(data_kva, skb->data, len); - if (unlikely(len < ETH_ZLEN)) { - memset(data_kva + len, 0, ETH_ZLEN - len); - len = ETH_ZLEN; - } - pkt_kva->pkt_len = len; - pkt_kva->data_len = len; - - /* enqueue mbuf into tx_q */ - ret = kni_fifo_put(kni->tx_q, (void **)&pkt_va, 1); - if (unlikely(ret != 1)) { - /* Failing should not happen */ - KNI_ERR("Fail to enqueue mbuf into tx_q\n"); - goto drop; - } - } else { - /* Failing should not happen */ - KNI_ERR("Fail to dequeue mbuf from alloc_q\n"); - goto drop; - } - - /* Free skb and update statistics */ - dev_kfree_skb(skb); - kni->stats.tx_bytes += len; - kni->stats.tx_packets++; - - return NETDEV_TX_OK; - -drop: - /* Free skb and update statistics */ - dev_kfree_skb(skb); - kni->stats.tx_dropped++; - - return NETDEV_TX_OK; -} -#endif - -/* * Deal with a transmit timeout. */ static void -kni_net_tx_timeout (struct net_device *dev) +kni_net_tx_timeout(struct net_device *dev) { struct kni_dev *kni = netdev_priv(dev); - KNI_DBG("Transmit timeout at %ld, latency %ld\n", jiffies, - jiffies - dev->trans_start); + pr_debug("Transmit timeout at %ld, latency %ld\n", jiffies, + jiffies - dev_trans_start(dev)); kni->stats.tx_errors++; netif_wake_queue(dev); - return; } /* @@ -521,8 +586,8 @@ kni_net_tx_timeout (struct net_device *dev) static int kni_net_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { - KNI_DBG("kni_net_ioctl %d\n", - ((struct kni_dev *)netdev_priv(dev))->group_id); + pr_debug("kni_net_ioctl group:%d cmd:%d\n", + ((struct kni_dev *)netdev_priv(dev))->group_id, cmd); return 0; } @@ -539,7 +604,7 @@ kni_net_change_mtu(struct net_device *dev, int new_mtu) struct rte_kni_request req; struct kni_dev *kni = netdev_priv(dev); - KNI_DBG("kni_net_change_mtu new mtu %d to be set\n", new_mtu); + pr_debug("kni_net_change_mtu new mtu %d to be set\n", new_mtu); memset(&req, 0, sizeof(req)); req.req_id = RTE_KNI_REQ_CHANGE_MTU; @@ -562,61 +627,13 @@ kni_net_poll_resp(struct kni_dev *kni) } /* - * It can be called to process the request. - */ -static int -kni_net_process_request(struct kni_dev *kni, struct rte_kni_request *req) -{ - int ret = -1; - void *resp_va; - unsigned num; - int ret_val; - - if (!kni || !req) { - KNI_ERR("No kni instance or request\n"); - return -EINVAL; - } - - mutex_lock(&kni->sync_lock); - - /* Construct data */ - memcpy(kni->sync_kva, req, sizeof(struct rte_kni_request)); - num = kni_fifo_put(kni->req_q, &kni->sync_va, 1); - if (num < 1) { - KNI_ERR("Cannot send to req_q\n"); - ret = -EBUSY; - goto fail; - } - - ret_val = wait_event_interruptible_timeout(kni->wq, - kni_fifo_count(kni->resp_q), 3 * HZ); - if (signal_pending(current) || ret_val <= 0) { - ret = -ETIME; - goto fail; - } - num = kni_fifo_get(kni->resp_q, (void **)&resp_va, 1); - if (num != 1 || resp_va != kni->sync_va) { - /* This should never happen */ - KNI_ERR("No data in resp_q\n"); - ret = -ENODATA; - goto fail; - } - - memcpy(req, kni->sync_kva, sizeof(struct rte_kni_request)); - ret = 0; - -fail: - mutex_unlock(&kni->sync_lock); - return ret; -} - -/* * Return statistics to the caller */ static struct net_device_stats * kni_net_stats(struct net_device *dev) { struct kni_dev *kni = netdev_priv(dev); + return &kni->stats; } @@ -626,7 +643,7 @@ kni_net_stats(struct net_device *dev) static int kni_net_header(struct sk_buff *skb, struct net_device *dev, unsigned short type, const void *daddr, - const void *saddr, unsigned int len) + const void *saddr, uint32_t len) { struct ethhdr *eth = (struct ethhdr *) skb_push(skb, ETH_HLEN); @@ -637,7 +654,6 @@ kni_net_header(struct sk_buff *skb, struct net_device *dev, return dev->hard_header_len; } - /* * Re-fill the eth header */ @@ -662,9 +678,11 @@ kni_net_rebuild_header(struct sk_buff *skb) * * Returns 0 on success, negative on failure **/ -static int kni_net_set_mac(struct net_device *netdev, void *p) +static int +kni_net_set_mac(struct net_device *netdev, void *p) { struct sockaddr *addr = p; + if (!is_valid_ether_addr((unsigned char *)(addr->sa_data))) return -EADDRNOTAVAIL; memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); @@ -672,7 +690,8 @@ static int kni_net_set_mac(struct net_device *netdev, void *p) } #ifdef HAVE_CHANGE_CARRIER_CB -static int kni_net_change_carrier(struct net_device *dev, bool new_carrier) +static int +kni_net_change_carrier(struct net_device *dev, bool new_carrier) { if (new_carrier) netif_carrier_on(dev); @@ -711,8 +730,6 @@ kni_net_init(struct net_device *dev) { struct kni_dev *kni = netdev_priv(dev); - KNI_DBG("kni_net_init\n"); - init_waitqueue_head(&kni->wq); mutex_init(&kni->sync_lock); @@ -726,18 +743,18 @@ void kni_net_config_lo_mode(char *lo_str) { if (!lo_str) { - KNI_PRINT("loopback disabled"); + pr_debug("loopback disabled"); return; } if (!strcmp(lo_str, "lo_mode_none")) - KNI_PRINT("loopback disabled"); + pr_debug("loopback disabled"); else if (!strcmp(lo_str, "lo_mode_fifo")) { - KNI_PRINT("loopback mode=lo_mode_fifo enabled"); + pr_debug("loopback mode=lo_mode_fifo enabled"); kni_net_rx_func = kni_net_rx_lo_fifo; } else if (!strcmp(lo_str, "lo_mode_fifo_skb")) { - KNI_PRINT("loopback mode=lo_mode_fifo_skb enabled"); + pr_debug("loopback mode=lo_mode_fifo_skb enabled"); kni_net_rx_func = kni_net_rx_lo_fifo_skb; } else - KNI_PRINT("Incognizant parameter, loopback disabled"); + pr_debug("Incognizant parameter, loopback disabled"); } diff --git a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_vhost.c b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_vhost.c index a3ca8499..f54c34b1 100644 --- a/src/dpdk/lib/librte_eal/linuxapp/kni/kni_vhost.c +++ b/src/dpdk/lib/librte_eal/linuxapp/kni/kni_vhost.c @@ -32,6 +32,7 @@ #include <linux/sched.h> #include <linux/if_tun.h> #include <linux/version.h> +#include <linux/file.h> #include "compat.h" #include "kni_dev.h" @@ -39,21 +40,12 @@ #define RX_BURST_SZ 4 -extern void put_unused_fd(unsigned int fd); - -#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,7,0) -extern struct file* -sock_alloc_file(struct socket *sock, - int flags, const char *dname); - -extern int get_unused_fd_flags(unsigned flags); - -extern void fd_install(unsigned int fd, struct file *file); - +#ifdef HAVE_STATIC_SOCK_MAP_FD static int kni_sock_map_fd(struct socket *sock) { struct file *file; int fd = get_unused_fd_flags(0); + if (fd < 0) return fd; @@ -65,8 +57,6 @@ static int kni_sock_map_fd(struct socket *sock) fd_install(fd, file); return fd; } -#else -#define kni_sock_map_fd(s) sock_map_fd(s, 0) #endif static struct proto kni_raw_proto = { @@ -77,13 +67,13 @@ static struct proto kni_raw_proto = { static inline int kni_vhost_net_tx(struct kni_dev *kni, struct msghdr *m, - unsigned offset, unsigned len) + uint32_t offset, uint32_t len) { struct rte_kni_mbuf *pkt_kva = NULL; struct rte_kni_mbuf *pkt_va = NULL; int ret; - KNI_DBG_TX("tx offset=%d, len=%d, iovlen=%d\n", + pr_debug("tx offset=%d, len=%d, iovlen=%d\n", #ifdef HAVE_IOV_ITER_MSGHDR offset, len, (int)m->msg_iter.iov->iov_len); #else @@ -110,7 +100,7 @@ kni_vhost_net_tx(struct kni_dev *kni, struct msghdr *m, pkt_kva = (void *)pkt_va - kni->mbuf_va + kni->mbuf_kva; data_kva = pkt_kva->buf_addr + pkt_kva->data_off - - kni->mbuf_va + kni->mbuf_kva; + - kni->mbuf_va + kni->mbuf_kva; #ifdef HAVE_IOV_ITER_MSGHDR copy_from_iter(data_kva, len, &m->msg_iter); @@ -129,12 +119,12 @@ kni_vhost_net_tx(struct kni_dev *kni, struct msghdr *m, ret = kni_fifo_put(kni->tx_q, (void **)&pkt_va, 1); if (unlikely(ret != 1)) { /* Failing should not happen */ - KNI_ERR("Fail to enqueue mbuf into tx_q\n"); + pr_err("Fail to enqueue mbuf into tx_q\n"); goto drop; } } else { /* Failing should not happen */ - KNI_ERR("Fail to dequeue mbuf from alloc_q\n"); + pr_err("Fail to dequeue mbuf from alloc_q\n"); goto drop; } @@ -153,12 +143,12 @@ drop: static inline int kni_vhost_net_rx(struct kni_dev *kni, struct msghdr *m, - unsigned offset, unsigned len) + uint32_t offset, uint32_t len) { uint32_t pkt_len; struct rte_kni_mbuf *kva; struct rte_kni_mbuf *va; - void * data_kva; + void *data_kva; struct sk_buff *skb; struct kni_vhost_queue *q = kni->vhost_queue; @@ -173,19 +163,19 @@ kni_vhost_net_rx(struct kni_dev *kni, struct msghdr *m, if (unlikely(skb == NULL)) return 0; - kva = (struct rte_kni_mbuf*)skb->data; + kva = (struct rte_kni_mbuf *)skb->data; /* free skb to cache */ skb->data = NULL; - if (unlikely(1 != kni_fifo_put(q->fifo, (void **)&skb, 1))) + if (unlikely(kni_fifo_put(q->fifo, (void **)&skb, 1) != 1)) /* Failing should not happen */ - KNI_ERR("Fail to enqueue entries into rx cache fifo\n"); + pr_err("Fail to enqueue entries into rx cache fifo\n"); pkt_len = kva->data_len; if (unlikely(pkt_len > len)) goto drop; - KNI_DBG_RX("rx offset=%d, len=%d, pkt_len=%d, iovlen=%d\n", + pr_debug("rx offset=%d, len=%d, pkt_len=%d, iovlen=%d\n", #ifdef HAVE_IOV_ITER_MSGHDR offset, len, pkt_len, (int)m->msg_iter.iov->iov_len); #else @@ -205,12 +195,12 @@ kni_vhost_net_rx(struct kni_dev *kni, struct msghdr *m, kni->stats.rx_packets++; /* enqueue mbufs into free_q */ - va = (void*)kva - kni->mbuf_kva + kni->mbuf_va; - if (unlikely(1 != kni_fifo_put(kni->free_q, (void **)&va, 1))) + va = (void *)kva - kni->mbuf_kva + kni->mbuf_va; + if (unlikely(kni_fifo_put(kni->free_q, (void **)&va, 1) != 1)) /* Failing should not happen */ - KNI_ERR("Fail to enqueue entries into free_q\n"); + pr_err("Fail to enqueue entries into free_q\n"); - KNI_DBG_RX("receive done %d\n", pkt_len); + pr_debug("receive done %d\n", pkt_len); return pkt_len; @@ -221,29 +211,25 @@ drop: return 0; } -static unsigned int -kni_sock_poll(struct file *file, struct socket *sock, poll_table * wait) +static uint32_t +kni_sock_poll(struct file *file, struct socket *sock, poll_table *wait) { struct kni_vhost_queue *q = container_of(sock->sk, struct kni_vhost_queue, sk); struct kni_dev *kni; - unsigned int mask = 0; + uint32_t mask = 0; if (unlikely(q == NULL || q->kni == NULL)) return POLLERR; kni = q->kni; -#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,35) - KNI_DBG("start kni_poll on group %d, wq 0x%16llx\n", +#ifdef HAVE_SOCKET_WQ + pr_debug("start kni_poll on group %d, wq 0x%16llx\n", kni->group_id, (uint64_t)sock->wq); -#else - KNI_DBG("start kni_poll on group %d, wait at 0x%16llx\n", - kni->group_id, (uint64_t)&sock->wait); -#endif - -#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,35) poll_wait(file, &sock->wq->wait, wait); #else + pr_debug("start kni_poll on group %d, wait at 0x%16llx\n", + kni->group_id, (uint64_t)&sock->wait); poll_wait(file, &sock->wait, wait); #endif @@ -252,11 +238,12 @@ kni_sock_poll(struct file *file, struct socket *sock, poll_table * wait) if (sock_writeable(&q->sk) || #ifdef SOCKWQ_ASYNC_NOSPACE - (!test_and_set_bit(SOCKWQ_ASYNC_NOSPACE, &q->sock->flags) && + (!test_and_set_bit(SOCKWQ_ASYNC_NOSPACE, &q->sock->flags) && + sock_writeable(&q->sk))) #else - (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &q->sock->flags) && + (!test_and_set_bit(SOCK_ASYNC_NOSPACE, &q->sock->flags) && + sock_writeable(&q->sk))) #endif - sock_writeable(&q->sk))) mask |= POLLOUT | POLLWRNORM; return mask; @@ -269,7 +256,7 @@ kni_vhost_enqueue(struct kni_dev *kni, struct kni_vhost_queue *q, struct rte_kni_mbuf *kva; kva = (void *)(va) - kni->mbuf_va + kni->mbuf_kva; - (skb)->data = (unsigned char*)kva; + (skb)->data = (unsigned char *)kva; (skb)->len = kva->data_len; skb_queue_tail(&q->sk.sk_receive_queue, skb); } @@ -279,6 +266,7 @@ kni_vhost_enqueue_burst(struct kni_dev *kni, struct kni_vhost_queue *q, struct sk_buff **skb, struct rte_kni_mbuf **va) { int i; + for (i = 0; i < RX_BURST_SZ; skb++, va++, i++) kni_vhost_enqueue(kni, q, *skb, *va); } @@ -287,9 +275,9 @@ int kni_chk_vhost_rx(struct kni_dev *kni) { struct kni_vhost_queue *q = kni->vhost_queue; - unsigned nb_in, nb_mbuf, nb_skb; - const unsigned BURST_MASK = RX_BURST_SZ - 1; - unsigned nb_burst, nb_backlog, i; + uint32_t nb_in, nb_mbuf, nb_skb; + const uint32_t BURST_MASK = RX_BURST_SZ - 1; + uint32_t nb_burst, nb_backlog, i; struct sk_buff *skb[RX_BURST_SZ]; struct rte_kni_mbuf *va[RX_BURST_SZ]; @@ -305,20 +293,18 @@ kni_chk_vhost_rx(struct kni_dev *kni) nb_mbuf = kni_fifo_count(kni->rx_q); nb_in = min(nb_mbuf, nb_skb); - nb_in = min(nb_in, (unsigned)RX_BURST_SZ); + nb_in = min_t(uint32_t, nb_in, RX_BURST_SZ); nb_burst = (nb_in & ~BURST_MASK); nb_backlog = (nb_in & BURST_MASK); /* enqueue skb_queue per BURST_SIZE bulk */ - if (0 != nb_burst) { - if (unlikely(RX_BURST_SZ != kni_fifo_get( - kni->rx_q, (void **)&va, - RX_BURST_SZ))) + if (nb_burst != 0) { + if (unlikely(kni_fifo_get(kni->rx_q, (void **)&va, RX_BURST_SZ) + != RX_BURST_SZ)) goto except; - if (unlikely(RX_BURST_SZ != kni_fifo_get( - q->fifo, (void **)&skb, - RX_BURST_SZ))) + if (unlikely(kni_fifo_get(q->fifo, (void **)&skb, RX_BURST_SZ) + != RX_BURST_SZ)) goto except; kni_vhost_enqueue_burst(kni, q, skb, va); @@ -326,12 +312,10 @@ kni_chk_vhost_rx(struct kni_dev *kni) /* all leftover, do one by one */ for (i = 0; i < nb_backlog; ++i) { - if (unlikely(1 != kni_fifo_get( - kni->rx_q,(void **)&va, 1))) + if (unlikely(kni_fifo_get(kni->rx_q, (void **)&va, 1) != 1)) goto except; - if (unlikely(1 != kni_fifo_get( - q->fifo, (void **)&skb, 1))) + if (unlikely(kni_fifo_get(q->fifo, (void **)&skb, 1) != 1)) goto except; kni_vhost_enqueue(kni, q, *skb, *va); @@ -342,7 +326,7 @@ kni_chk_vhost_rx(struct kni_dev *kni) ((nb_mbuf < RX_BURST_SZ) && (nb_mbuf != 0))) { wake_up_interruptible_poll(sk_sleep(&q->sk), POLLIN | POLLRDNORM | POLLRDBAND); - KNI_DBG_RX("RX CHK KICK nb_mbuf %d, nb_skb %d, nb_in %d\n", + pr_debug("RX CHK KICK nb_mbuf %d, nb_skb %d, nb_in %d\n", nb_mbuf, nb_skb, nb_in); } @@ -350,7 +334,7 @@ kni_chk_vhost_rx(struct kni_dev *kni) except: /* Failing should not happen */ - KNI_ERR("Fail to enqueue fifo, it shouldn't happen \n"); + pr_err("Fail to enqueue fifo, it shouldn't happen\n"); BUG_ON(1); return 0; @@ -373,7 +357,7 @@ kni_sock_sndmsg(struct socket *sock, if (unlikely(q == NULL || q->kni == NULL)) return 0; - KNI_DBG_TX("kni_sndmsg len %ld, flags 0x%08x, nb_iov %d\n", + pr_debug("kni_sndmsg len %ld, flags 0x%08x, nb_iov %d\n", #ifdef HAVE_IOV_ITER_MSGHDR len, q->flags, (int)m->msg_iter.iov->iov_len); #else @@ -420,13 +404,14 @@ kni_sock_rcvmsg(struct socket *sock, #ifdef RTE_KNI_VHOST_VNET_HDR_EN if (likely(q->flags & IFF_VNET_HDR)) { vnet_hdr_len = q->vnet_hdr_sz; - if ((len -= vnet_hdr_len) < 0) + len -= vnet_hdr_len; + if (len < 0) return -EINVAL; } #endif - if (unlikely(0 == (pkt_len = kni_vhost_net_rx(q->kni, - m, vnet_hdr_len, len)))) + pkt_len = kni_vhost_net_rx(q->kni, m, vnet_hdr_len, len); + if (unlikely(pkt_len == 0)) return 0; #ifdef RTE_KNI_VHOST_VNET_HDR_EN @@ -440,7 +425,7 @@ kni_sock_rcvmsg(struct socket *sock, #endif /* HAVE_IOV_ITER_MSGHDR */ return -EFAULT; #endif /* RTE_KNI_VHOST_VNET_HDR_EN */ - KNI_DBG_RX("kni_rcvmsg expect_len %ld, flags 0x%08x, pkt_len %d\n", + pr_debug("kni_rcvmsg expect_len %ld, flags 0x%08x, pkt_len %d\n", (unsigned long)len, q->flags, pkt_len); return pkt_len + vnet_hdr_len; @@ -448,25 +433,24 @@ kni_sock_rcvmsg(struct socket *sock, /* dummy tap like ioctl */ static int -kni_sock_ioctl(struct socket *sock, unsigned int cmd, - unsigned long arg) +kni_sock_ioctl(struct socket *sock, uint32_t cmd, unsigned long arg) { void __user *argp = (void __user *)arg; struct ifreq __user *ifr = argp; - unsigned int __user *up = argp; + uint32_t __user *up = argp; struct kni_vhost_queue *q = container_of(sock->sk, struct kni_vhost_queue, sk); struct kni_dev *kni; - unsigned int u; + uint32_t u; int __user *sp = argp; int s; int ret; - KNI_DBG("tap ioctl cmd 0x%08x\n", cmd); + pr_debug("tap ioctl cmd 0x%08x\n", cmd); switch (cmd) { case TUNSETIFF: - KNI_DBG("TUNSETIFF\n"); + pr_debug("TUNSETIFF\n"); /* ignore the name, just look at flags */ if (get_user(u, &ifr->ifr_flags)) return -EFAULT; @@ -480,7 +464,7 @@ kni_sock_ioctl(struct socket *sock, unsigned int cmd, return ret; case TUNGETIFF: - KNI_DBG("TUNGETIFF\n"); + pr_debug("TUNGETIFF\n"); rcu_read_lock_bh(); kni = rcu_dereference_bh(q->kni); if (kni) @@ -491,14 +475,14 @@ kni_sock_ioctl(struct socket *sock, unsigned int cmd, return -ENOLINK; ret = 0; - if (copy_to_user(&ifr->ifr_name, kni->net_dev->name, IFNAMSIZ) || - put_user(q->flags, &ifr->ifr_flags)) + if (copy_to_user(&ifr->ifr_name, kni->net_dev->name, IFNAMSIZ) + || put_user(q->flags, &ifr->ifr_flags)) ret = -EFAULT; dev_put(kni->net_dev); return ret; case TUNGETFEATURES: - KNI_DBG("TUNGETFEATURES\n"); + pr_debug("TUNGETFEATURES\n"); u = IFF_TAP | IFF_NO_PI; #ifdef RTE_KNI_VHOST_VNET_HDR_EN u |= IFF_VNET_HDR; @@ -508,7 +492,7 @@ kni_sock_ioctl(struct socket *sock, unsigned int cmd, return 0; case TUNSETSNDBUF: - KNI_DBG("TUNSETSNDBUF\n"); + pr_debug("TUNSETSNDBUF\n"); if (get_user(u, up)) return -EFAULT; @@ -519,7 +503,7 @@ kni_sock_ioctl(struct socket *sock, unsigned int cmd, s = q->vnet_hdr_sz; if (put_user(s, sp)) return -EFAULT; - KNI_DBG("TUNGETVNETHDRSZ %d\n", s); + pr_debug("TUNGETVNETHDRSZ %d\n", s); return 0; case TUNSETVNETHDRSZ: @@ -528,12 +512,12 @@ kni_sock_ioctl(struct socket *sock, unsigned int cmd, if (s < (int)sizeof(struct virtio_net_hdr)) return -EINVAL; - KNI_DBG("TUNSETVNETHDRSZ %d\n", s); + pr_debug("TUNSETVNETHDRSZ %d\n", s); q->vnet_hdr_sz = s; return 0; case TUNSETOFFLOAD: - KNI_DBG("TUNSETOFFLOAD %lx\n", arg); + pr_debug("TUNSETOFFLOAD %lx\n", arg); #ifdef RTE_KNI_VHOST_VNET_HDR_EN /* not support any offload yet */ if (!(q->flags & IFF_VNET_HDR)) @@ -545,26 +529,26 @@ kni_sock_ioctl(struct socket *sock, unsigned int cmd, #endif default: - KNI_DBG("NOT SUPPORT\n"); + pr_debug("NOT SUPPORT\n"); return -EINVAL; } } static int -kni_sock_compat_ioctl(struct socket *sock, unsigned int cmd, +kni_sock_compat_ioctl(struct socket *sock, uint32_t cmd, unsigned long arg) { /* 32 bits app on 64 bits OS to be supported later */ - KNI_PRINT("Not implemented.\n"); + pr_debug("Not implemented.\n"); return -EINVAL; } #define KNI_VHOST_WAIT_WQ_SAFE() \ -do { \ +do { \ while ((BE_FINISH | BE_STOP) == kni->vq_status) \ - msleep(1); \ -}while(0) \ + msleep(1); \ +} while (0) \ static int @@ -577,7 +561,8 @@ kni_sock_release(struct socket *sock) if (q == NULL) return 0; - if (NULL != (kni = q->kni)) { + kni = q->kni; + if (kni != NULL) { kni->vq_status = BE_STOP; KNI_VHOST_WAIT_WQ_SAFE(); kni->vhost_queue = NULL; @@ -592,18 +577,17 @@ kni_sock_release(struct socket *sock) sock_put(&q->sk); - KNI_DBG("dummy sock release done\n"); + pr_debug("dummy sock release done\n"); return 0; } int -kni_sock_getname (struct socket *sock, - struct sockaddr *addr, - int *sockaddr_len, int peer) +kni_sock_getname(struct socket *sock, struct sockaddr *addr, + int *sockaddr_len, int peer) { - KNI_DBG("dummy sock getname\n"); - ((struct sockaddr_ll*)addr)->sll_family = AF_PACKET; + pr_debug("dummy sock getname\n"); + ((struct sockaddr_ll *)addr)->sll_family = AF_PACKET; return 0; } @@ -646,7 +630,7 @@ kni_sk_destruct(struct sock *sk) /* make sure there's no packet in buffer */ while (skb_dequeue(&sk->sk_receive_queue) != NULL) - ; + ; mb(); @@ -673,7 +657,7 @@ kni_vhost_backend_init(struct kni_dev *kni) if (kni->vhost_queue != NULL) return -1; -#if LINUX_VERSION_CODE >= KERNEL_VERSION(4, 2, 0) +#ifdef HAVE_SK_ALLOC_KERN_PARAM q = (struct kni_vhost_queue *)sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &kni_raw_proto, 0); #else @@ -694,8 +678,9 @@ kni_vhost_backend_init(struct kni_dev *kni) } /* cache init */ - q->cache = kzalloc(RTE_KNI_VHOST_MAX_CACHE_SIZE * sizeof(struct sk_buff), - GFP_KERNEL); + q->cache = kzalloc( + RTE_KNI_VHOST_MAX_CACHE_SIZE * sizeof(struct sk_buff), + GFP_KERNEL); if (!q->cache) goto free_fd; @@ -708,7 +693,7 @@ kni_vhost_backend_init(struct kni_dev *kni) for (i = 0; i < RTE_KNI_VHOST_MAX_CACHE_SIZE; i++) { elem = &q->cache[i]; - kni_fifo_put(fifo, (void**)&elem, 1); + kni_fifo_put(fifo, (void **)&elem, 1); } q->fifo = fifo; @@ -738,14 +723,12 @@ kni_vhost_backend_init(struct kni_dev *kni) kni->vq_status = BE_START; -#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,35) - KNI_DBG("backend init sockfd=%d, sock->wq=0x%16llx," - "sk->sk_wq=0x%16llx", +#ifdef HAVE_SOCKET_WQ + pr_debug("backend init sockfd=%d, sock->wq=0x%16llx,sk->sk_wq=0x%16llx", q->sockfd, (uint64_t)q->sock->wq, (uint64_t)q->sk.sk_wq); #else - KNI_DBG("backend init sockfd=%d, sock->wait at 0x%16llx," - "sk->sk_sleep=0x%16llx", + pr_debug("backend init sockfd=%d, sock->wait at 0x%16llx,sk->sk_sleep=0x%16llx", q->sockfd, (uint64_t)&q->sock->wait, (uint64_t)q->sk.sk_sleep); #endif @@ -768,7 +751,7 @@ free_sock: q->sock = NULL; free_sk: - sk_free((struct sock*)q); + sk_free((struct sock *)q); return err; } @@ -781,6 +764,7 @@ show_sock_fd(struct device *dev, struct device_attribute *attr, struct net_device *net_dev = container_of(dev, struct net_device, dev); struct kni_dev *kni = netdev_priv(net_dev); int sockfd = -1; + if (kni->vhost_queue != NULL) sockfd = kni->vhost_queue->sockfd; return snprintf(buf, 10, "%d\n", sockfd); @@ -792,6 +776,7 @@ show_sock_en(struct device *dev, struct device_attribute *attr, { struct net_device *net_dev = container_of(dev, struct net_device, dev); struct kni_dev *kni = netdev_priv(net_dev); + return snprintf(buf, 10, "%u\n", (kni->vhost_queue == NULL ? 0 : 1)); } @@ -804,7 +789,7 @@ set_sock_en(struct device *dev, struct device_attribute *attr, unsigned long en; int err = 0; - if (0 != kstrtoul(buf, 0, &en)) + if (kstrtoul(buf, 0, &en) != 0) return -EINVAL; if (en) @@ -818,7 +803,7 @@ static DEVICE_ATTR(sock_en, S_IRUGO | S_IWUSR, show_sock_en, set_sock_en); static struct attribute *dev_attrs[] = { &dev_attr_sock_fd.attr, &dev_attr_sock_en.attr, - NULL, + NULL, }; static const struct attribute_group dev_attr_grp = { @@ -836,7 +821,7 @@ kni_vhost_backend_release(struct kni_dev *kni) /* dettach from kni */ q->kni = NULL; - KNI_DBG("release backend done\n"); + pr_debug("release backend done\n"); return 0; } @@ -851,7 +836,7 @@ kni_vhost_init(struct kni_dev *kni) kni->vq_status = BE_STOP; - KNI_DBG("kni_vhost_init done\n"); + pr_debug("kni_vhost_init done\n"); return 0; } |