diff options
Diffstat (limited to 'drivers')
69 files changed, 7201 insertions, 848 deletions
diff --git a/drivers/acpi/acpica/hwvalid.c b/drivers/acpi/acpica/hwvalid.c index bd3c937b0ac..7737afb157c 100644 --- a/drivers/acpi/acpica/hwvalid.c +++ b/drivers/acpi/acpica/hwvalid.c @@ -90,7 +90,6 @@ static const struct acpi_port_info acpi_protected_ports[] = { {"PIT2", 0x0048, 0x004B, ACPI_OSI_WIN_XP}, {"RTC", 0x0070, 0x0071, ACPI_OSI_WIN_XP}, {"CMOS", 0x0074, 0x0076, ACPI_OSI_WIN_XP}, - {"DMA1", 0x0081, 0x0083, ACPI_OSI_WIN_XP}, {"DMA1L", 0x0087, 0x0087, ACPI_OSI_WIN_XP}, {"DMA2", 0x0089, 0x008B, ACPI_OSI_WIN_XP}, {"DMA2L", 0x008F, 0x008F, ACPI_OSI_WIN_XP}, diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index b0de6312919..3c7d8942f23 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -903,7 +903,7 @@ static struct acpi_driver acpi_battery_driver = { }, }; -static void __init acpi_battery_init_async(void *unused, async_cookie_t cookie) +static void acpi_battery_init_async(void *unused, async_cookie_t cookie) { if (acpi_disabled) return; diff --git a/drivers/acpi/proc.c b/drivers/acpi/proc.c index 05dfdc96802..d0d550d22a6 100644 --- a/drivers/acpi/proc.c +++ b/drivers/acpi/proc.c @@ -343,9 +343,6 @@ acpi_system_write_alarm(struct file *file, } #endif /* HAVE_ACPI_LEGACY_ALARM */ -extern struct list_head acpi_wakeup_device_list; -extern spinlock_t acpi_device_lock; - static int acpi_system_wakeup_device_seq_show(struct seq_file *seq, void *offset) { @@ -353,7 +350,7 @@ acpi_system_wakeup_device_seq_show(struct seq_file *seq, void *offset) seq_printf(seq, "Device\tS-state\t Status Sysfs node\n"); - spin_lock(&acpi_device_lock); + mutex_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, wakeup_list); @@ -361,7 +358,6 @@ acpi_system_wakeup_device_seq_show(struct seq_file *seq, void *offset) if (!dev->wakeup.flags.valid) continue; - spin_unlock(&acpi_device_lock); ldev = acpi_get_physical_device(dev->handle); seq_printf(seq, "%s\t S%d\t%c%-8s ", @@ -376,9 +372,8 @@ acpi_system_wakeup_device_seq_show(struct seq_file *seq, void *offset) seq_printf(seq, "\n"); put_device(ldev); - spin_lock(&acpi_device_lock); } - spin_unlock(&acpi_device_lock); + mutex_unlock(&acpi_device_lock); return 0; } @@ -409,7 +404,7 @@ acpi_system_write_wakeup_device(struct file *file, strbuf[len] = '\0'; sscanf(strbuf, "%s", str); - spin_lock(&acpi_device_lock); + mutex_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, wakeup_list); @@ -446,7 +441,7 @@ acpi_system_write_wakeup_device(struct file *file, } } } - spin_unlock(&acpi_device_lock); + mutex_unlock(&acpi_device_lock); return count; } diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 4e6e758bd39..6fe121434ff 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -145,6 +145,9 @@ static void acpi_timer_check_state(int state, struct acpi_processor *pr, struct acpi_processor_power *pwr = &pr->power; u8 type = local_apic_timer_c2_ok ? ACPI_STATE_C3 : ACPI_STATE_C2; + if (cpu_has(&cpu_data(pr->id), X86_FEATURE_ARAT)) + return; + /* * Check, if one of the previous states already marked the lapic * unstable diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 20c23c04920..8ff510b91d8 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -24,7 +24,7 @@ extern struct acpi_device *acpi_root; static LIST_HEAD(acpi_device_list); static LIST_HEAD(acpi_bus_id_list); -DEFINE_SPINLOCK(acpi_device_lock); +DEFINE_MUTEX(acpi_device_lock); LIST_HEAD(acpi_wakeup_device_list); struct acpi_device_bus_id{ @@ -491,7 +491,6 @@ static int acpi_device_register(struct acpi_device *device, */ INIT_LIST_HEAD(&device->children); INIT_LIST_HEAD(&device->node); - INIT_LIST_HEAD(&device->g_list); INIT_LIST_HEAD(&device->wakeup_list); new_bus_id = kzalloc(sizeof(struct acpi_device_bus_id), GFP_KERNEL); @@ -500,7 +499,7 @@ static int acpi_device_register(struct acpi_device *device, return -ENOMEM; } - spin_lock(&acpi_device_lock); + mutex_lock(&acpi_device_lock); /* * Find suitable bus_id and instance number in acpi_bus_id_list * If failed, create one and link it into acpi_bus_id_list @@ -521,14 +520,12 @@ static int acpi_device_register(struct acpi_device *device, } dev_set_name(&device->dev, "%s:%02x", acpi_device_bus_id->bus_id, acpi_device_bus_id->instance_no); - if (device->parent) { + if (device->parent) list_add_tail(&device->node, &device->parent->children); - list_add_tail(&device->g_list, &device->parent->g_list); - } else - list_add_tail(&device->g_list, &acpi_device_list); + if (device->wakeup.flags.valid) list_add_tail(&device->wakeup_list, &acpi_wakeup_device_list); - spin_unlock(&acpi_device_lock); + mutex_unlock(&acpi_device_lock); if (device->parent) device->dev.parent = &parent->dev; @@ -549,28 +546,22 @@ static int acpi_device_register(struct acpi_device *device, device->removal_type = ACPI_BUS_REMOVAL_NORMAL; return 0; end: - spin_lock(&acpi_device_lock); - if (device->parent) { + mutex_lock(&acpi_device_lock); + if (device->parent) list_del(&device->node); - list_del(&device->g_list); - } else - list_del(&device->g_list); list_del(&device->wakeup_list); - spin_unlock(&acpi_device_lock); + mutex_unlock(&acpi_device_lock); return result; } static void acpi_device_unregister(struct acpi_device *device, int type) { - spin_lock(&acpi_device_lock); - if (device->parent) { + mutex_lock(&acpi_device_lock); + if (device->parent) list_del(&device->node); - list_del(&device->g_list); - } else - list_del(&device->g_list); list_del(&device->wakeup_list); - spin_unlock(&acpi_device_lock); + mutex_unlock(&acpi_device_lock); acpi_detach_data(device->handle, acpi_bus_data_handler); diff --git a/drivers/acpi/sleep.h b/drivers/acpi/sleep.h index cfaf8f5b0a1..8a8f3b3382a 100644 --- a/drivers/acpi/sleep.h +++ b/drivers/acpi/sleep.h @@ -5,3 +5,6 @@ extern int acpi_suspend (u32 state); extern void acpi_enable_wakeup_device_prep(u8 sleep_state); extern void acpi_enable_wakeup_device(u8 sleep_state); extern void acpi_disable_wakeup_device(u8 sleep_state); + +extern struct list_head acpi_wakeup_device_list; +extern struct mutex acpi_device_lock; diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c index e8c143caf0f..9cd15e8c893 100644 --- a/drivers/acpi/thermal.c +++ b/drivers/acpi/thermal.c @@ -98,6 +98,7 @@ MODULE_PARM_DESC(psv, "Disable or override all passive trip points."); static int acpi_thermal_add(struct acpi_device *device); static int acpi_thermal_remove(struct acpi_device *device, int type); static int acpi_thermal_resume(struct acpi_device *device); +static void acpi_thermal_notify(struct acpi_device *device, u32 event); static int acpi_thermal_state_open_fs(struct inode *inode, struct file *file); static int acpi_thermal_temp_open_fs(struct inode *inode, struct file *file); static int acpi_thermal_trip_open_fs(struct inode *inode, struct file *file); @@ -123,6 +124,7 @@ static struct acpi_driver acpi_thermal_driver = { .add = acpi_thermal_add, .remove = acpi_thermal_remove, .resume = acpi_thermal_resume, + .notify = acpi_thermal_notify, }, }; @@ -192,6 +194,7 @@ struct acpi_thermal { struct acpi_handle_list devices; struct thermal_zone_device *thermal_zone; int tz_enabled; + int kelvin_offset; struct mutex lock; }; @@ -581,7 +584,7 @@ static void acpi_thermal_check(void *data) } /* sys I/F for generic thermal sysfs support */ -#define KELVIN_TO_MILLICELSIUS(t) (t * 100 - 273200) +#define KELVIN_TO_MILLICELSIUS(t, off) (((t) - (off)) * 100) static int thermal_get_temp(struct thermal_zone_device *thermal, unsigned long *temp) @@ -596,7 +599,7 @@ static int thermal_get_temp(struct thermal_zone_device *thermal, if (result) return result; - *temp = KELVIN_TO_MILLICELSIUS(tz->temperature); + *temp = KELVIN_TO_MILLICELSIUS(tz->temperature, tz->kelvin_offset); return 0; } @@ -702,7 +705,8 @@ static int thermal_get_trip_temp(struct thermal_zone_device *thermal, if (tz->trips.critical.flags.valid) { if (!trip) { *temp = KELVIN_TO_MILLICELSIUS( - tz->trips.critical.temperature); + tz->trips.critical.temperature, + tz->kelvin_offset); return 0; } trip--; @@ -711,7 +715,8 @@ static int thermal_get_trip_temp(struct thermal_zone_device *thermal, if (tz->trips.hot.flags.valid) { if (!trip) { *temp = KELVIN_TO_MILLICELSIUS( - tz->trips.hot.temperature); + tz->trips.hot.temperature, + tz->kelvin_offset); return 0; } trip--; @@ -720,7 +725,8 @@ static int thermal_get_trip_temp(struct thermal_zone_device *thermal, if (tz->trips.passive.flags.valid) { if (!trip) { *temp = KELVIN_TO_MILLICELSIUS( - tz->trips.passive.temperature); + tz->trips.passive.temperature, + tz->kelvin_offset); return 0; } trip--; @@ -730,7 +736,8 @@ static int thermal_get_trip_temp(struct thermal_zone_device *thermal, tz->trips.active[i].flags.valid; i++) { if (!trip) { *temp = KELVIN_TO_MILLICELSIUS( - tz->trips.active[i].temperature); + tz->trips.active[i].temperature, + tz->kelvin_offset); return 0; } trip--; @@ -745,7 +752,8 @@ static int thermal_get_crit_temp(struct thermal_zone_device *thermal, if (tz->trips.critical.flags.valid) { *temperature = KELVIN_TO_MILLICELSIUS( - tz->trips.critical.temperature); + tz->trips.critical.temperature, + tz->kelvin_offset); return 0; } else return -EINVAL; @@ -1264,17 +1272,14 @@ static int acpi_thermal_remove_fs(struct acpi_device *device) Driver Interface -------------------------------------------------------------------------- */ -static void acpi_thermal_notify(acpi_handle handle, u32 event, void *data) +static void acpi_thermal_notify(struct acpi_device *device, u32 event) { - struct acpi_thermal *tz = data; - struct acpi_device *device = NULL; + struct acpi_thermal *tz = acpi_driver_data(device); if (!tz) return; - device = tz->device; - switch (event) { case ACPI_THERMAL_NOTIFY_TEMPERATURE: acpi_thermal_check(tz); @@ -1298,8 +1303,6 @@ static void acpi_thermal_notify(acpi_handle handle, u32 event, void *data) "Unsupported event [0x%x]\n", event)); break; } - - return; } static int acpi_thermal_get_info(struct acpi_thermal *tz) @@ -1334,10 +1337,28 @@ static int acpi_thermal_get_info(struct acpi_thermal *tz) return 0; } +/* + * The exact offset between Kelvin and degree Celsius is 273.15. However ACPI + * handles temperature values with a single decimal place. As a consequence, + * some implementations use an offset of 273.1 and others use an offset of + * 273.2. Try to find out which one is being used, to present the most + * accurate and visually appealing number. + * + * The heuristic below should work for all ACPI thermal zones which have a + * critical trip point with a value being a multiple of 0.5 degree Celsius. + */ +static void acpi_thermal_guess_offset(struct acpi_thermal *tz) +{ + if (tz->trips.critical.flags.valid && + (tz->trips.critical.temperature % 5) == 1) + tz->kelvin_offset = 2731; + else + tz->kelvin_offset = 2732; +} + static int acpi_thermal_add(struct acpi_device *device) { int result = 0; - acpi_status status = AE_OK; struct acpi_thermal *tz = NULL; @@ -1360,6 +1381,8 @@ static int acpi_thermal_add(struct acpi_device *device) if (result) goto free_memory; + acpi_thermal_guess_offset(tz); + result = acpi_thermal_register_thermal_zone(tz); if (result) goto free_memory; @@ -1368,21 +1391,11 @@ static int acpi_thermal_add(struct acpi_device *device) if (result) goto unregister_thermal_zone; - status = acpi_install_notify_handler(device->handle, - ACPI_DEVICE_NOTIFY, - acpi_thermal_notify, tz); - if (ACPI_FAILURE(status)) { - result = -ENODEV; - goto remove_fs; - } - printk(KERN_INFO PREFIX "%s [%s] (%ld C)\n", acpi_device_name(device), acpi_device_bid(device), KELVIN_TO_CELSIUS(tz->temperature)); goto end; -remove_fs: - acpi_thermal_remove_fs(device); unregister_thermal_zone: thermal_zone_device_unregister(tz->thermal_zone); free_memory: @@ -1393,7 +1406,6 @@ end: static int acpi_thermal_remove(struct acpi_device *device, int type) { - acpi_status status = AE_OK; struct acpi_thermal *tz = NULL; if (!device || !acpi_driver_data(device)) @@ -1401,10 +1413,6 @@ static int acpi_thermal_remove(struct acpi_device *device, int type) tz = acpi_driver_data(device); - status = acpi_remove_notify_handler(device->handle, - ACPI_DEVICE_NOTIFY, - acpi_thermal_notify); - acpi_thermal_remove_fs(device); acpi_thermal_unregister_thermal_zone(tz); mutex_destroy(&tz->lock); diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index ab06143672b..cd4fb7543a9 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -79,6 +79,7 @@ module_param(brightness_switch_enabled, bool, 0644); static int acpi_video_bus_add(struct acpi_device *device); static int acpi_video_bus_remove(struct acpi_device *device, int type); static int acpi_video_resume(struct acpi_device *device); +static void acpi_video_bus_notify(struct acpi_device *device, u32 event); static const struct acpi_device_id video_device_ids[] = { {ACPI_VIDEO_HID, 0}, @@ -94,6 +95,7 @@ static struct acpi_driver acpi_video_bus = { .add = acpi_video_bus_add, .remove = acpi_video_bus_remove, .resume = acpi_video_resume, + .notify = acpi_video_bus_notify, }, }; @@ -1986,17 +1988,15 @@ static int acpi_video_bus_stop_devices(struct acpi_video_bus *video) return acpi_video_bus_DOS(video, 0, 1); } -static void acpi_video_bus_notify(acpi_handle handle, u32 event, void *data) +static void acpi_video_bus_notify(struct acpi_device *device, u32 event) { - struct acpi_video_bus *video = data; - struct acpi_device *device = NULL; + struct acpi_video_bus *video = acpi_driver_data(device); struct input_dev *input; int keycode; if (!video) return; - device = video->device; input = video->input; switch (event) { @@ -2127,7 +2127,6 @@ static int acpi_video_resume(struct acpi_device *device) static int acpi_video_bus_add(struct acpi_device *device) { - acpi_status status; struct acpi_video_bus *video; struct input_dev *input; int error; @@ -2169,20 +2168,10 @@ static int acpi_video_bus_add(struct acpi_device *device) acpi_video_bus_get_devices(video, device); acpi_video_bus_start_devices(video); - status = acpi_install_notify_handler(device->handle, - ACPI_DEVICE_NOTIFY, - acpi_video_bus_notify, video); - if (ACPI_FAILURE(status)) { - printk(KERN_ERR PREFIX - "Error installing notify handler\n"); - error = -ENODEV; - goto err_stop_video; - } - video->input = input = input_allocate_device(); if (!input) { error = -ENOMEM; - goto err_uninstall_notify; + goto err_stop_video; } snprintf(video->phys, sizeof(video->phys), @@ -2218,9 +2207,6 @@ static int acpi_video_bus_add(struct acpi_device *device) err_free_input_dev: input_free_device(input); - err_uninstall_notify: - acpi_remove_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, - acpi_video_bus_notify); err_stop_video: acpi_video_bus_stop_devices(video); acpi_video_bus_put_devices(video); @@ -2235,7 +2221,6 @@ static int acpi_video_bus_add(struct acpi_device *device) static int acpi_video_bus_remove(struct acpi_device *device, int type) { - acpi_status status = 0; struct acpi_video_bus *video = NULL; @@ -2245,11 +2230,6 @@ static int acpi_video_bus_remove(struct acpi_device *device, int type) video = acpi_driver_data(device); acpi_video_bus_stop_devices(video); - - status = acpi_remove_notify_handler(video->device->handle, - ACPI_DEVICE_NOTIFY, - acpi_video_bus_notify); - acpi_video_bus_put_devices(video); acpi_video_bus_remove_fs(device); diff --git a/drivers/acpi/wakeup.c b/drivers/acpi/wakeup.c index 5aee8c26cc9..88725dcdf8b 100644 --- a/drivers/acpi/wakeup.c +++ b/drivers/acpi/wakeup.c @@ -12,12 +12,14 @@ #include "internal.h" #include "sleep.h" +/* + * We didn't lock acpi_device_lock in the file, because it invokes oops in + * suspend/resume and isn't really required as this is called in S-state. At + * that time, there is no device hotplug + **/ #define _COMPONENT ACPI_SYSTEM_COMPONENT ACPI_MODULE_NAME("wakeup_devices") -extern struct list_head acpi_wakeup_device_list; -extern spinlock_t acpi_device_lock; - /** * acpi_enable_wakeup_device_prep - prepare wakeup devices * @sleep_state: ACPI state @@ -29,7 +31,6 @@ void acpi_enable_wakeup_device_prep(u8 sleep_state) { struct list_head *node, *next; - spin_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, @@ -40,11 +41,8 @@ void acpi_enable_wakeup_device_prep(u8 sleep_state) (sleep_state > (u32) dev->wakeup.sleep_state)) continue; - spin_unlock(&acpi_device_lock); acpi_enable_wakeup_device_power(dev, sleep_state); - spin_lock(&acpi_device_lock); } - spin_unlock(&acpi_device_lock); } /** @@ -60,7 +58,6 @@ void acpi_enable_wakeup_device(u8 sleep_state) * Caution: this routine must be invoked when interrupt is disabled * Refer ACPI2.0: P212 */ - spin_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, wakeup_list); @@ -74,22 +71,17 @@ void acpi_enable_wakeup_device(u8 sleep_state) if ((!dev->wakeup.state.enabled && !dev->wakeup.flags.prepared) || sleep_state > (u32) dev->wakeup.sleep_state) { if (dev->wakeup.flags.run_wake) { - spin_unlock(&acpi_device_lock); /* set_gpe_type will disable GPE, leave it like that */ acpi_set_gpe_type(dev->wakeup.gpe_device, dev->wakeup.gpe_number, ACPI_GPE_TYPE_RUNTIME); - spin_lock(&acpi_device_lock); } continue; } - spin_unlock(&acpi_device_lock); if (!dev->wakeup.flags.run_wake) acpi_enable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number); - spin_lock(&acpi_device_lock); } - spin_unlock(&acpi_device_lock); } /** @@ -101,7 +93,6 @@ void acpi_disable_wakeup_device(u8 sleep_state) { struct list_head *node, *next; - spin_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, wakeup_list); @@ -112,19 +103,16 @@ void acpi_disable_wakeup_device(u8 sleep_state) if ((!dev->wakeup.state.enabled && !dev->wakeup.flags.prepared) || sleep_state > (u32) dev->wakeup.sleep_state) { if (dev->wakeup.flags.run_wake) { - spin_unlock(&acpi_device_lock); acpi_set_gpe_type(dev->wakeup.gpe_device, dev->wakeup.gpe_number, ACPI_GPE_TYPE_WAKE_RUN); /* Re-enable it, since set_gpe_type will disable it */ acpi_enable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number); - spin_lock(&acpi_device_lock); } continue; } - spin_unlock(&acpi_device_lock); acpi_disable_wakeup_device_power(dev); /* Never disable run-wake GPE */ if (!dev->wakeup.flags.run_wake) { @@ -133,16 +121,14 @@ void acpi_disable_wakeup_device(u8 sleep_state) acpi_clear_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number, ACPI_NOT_ISR); } - spin_lock(&acpi_device_lock); } - spin_unlock(&acpi_device_lock); } int __init acpi_wakeup_device_init(void) { struct list_head *node, *next; - spin_lock(&acpi_device_lock); + mutex_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, @@ -150,15 +136,13 @@ int __init acpi_wakeup_device_init(void) /* In case user doesn't load button driver */ if (!dev->wakeup.flags.run_wake || dev->wakeup.state.enabled) continue; - spin_unlock(&acpi_device_lock); acpi_set_gpe_type(dev->wakeup.gpe_device, dev->wakeup.gpe_number, ACPI_GPE_TYPE_WAKE_RUN); acpi_enable_gpe(dev->wakeup.gpe_device, dev->wakeup.gpe_number); dev->wakeup.state.enabled = 1; - spin_lock(&acpi_device_lock); } - spin_unlock(&acpi_device_lock); + mutex_unlock(&acpi_device_lock); return 0; } diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 2a2e50871b4..851de83ff45 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -297,21 +297,25 @@ static void cma_detach_from_dev(struct rdma_id_private *id_priv) id_priv->cma_dev = NULL; } -static int cma_set_qkey(struct ib_device *device, u8 port_num, - enum rdma_port_space ps, - struct rdma_dev_addr *dev_addr, u32 *qkey) +static int cma_set_qkey(struct rdma_id_private *id_priv) { struct ib_sa_mcmember_rec rec; int ret = 0; - switch (ps) { + if (id_priv->qkey) + return 0; + + switch (id_priv->id.ps) { case RDMA_PS_UDP: - *qkey = RDMA_UDP_QKEY; + id_priv->qkey = RDMA_UDP_QKEY; break; case RDMA_PS_IPOIB: - ib_addr_get_mgid(dev_addr, &rec.mgid); - ret = ib_sa_get_mcmember_rec(device, port_num, &rec.mgid, &rec); - *qkey = be32_to_cpu(rec.qkey); + ib_addr_get_mgid(&id_priv->id.route.addr.dev_addr, &rec.mgid); + ret = ib_sa_get_mcmember_rec(id_priv->id.device, + id_priv->id.port_num, &rec.mgid, + &rec); + if (!ret) + id_priv->qkey = be32_to_cpu(rec.qkey); break; default: break; @@ -341,12 +345,7 @@ static int cma_acquire_dev(struct rdma_id_private *id_priv) ret = ib_find_cached_gid(cma_dev->device, &gid, &id_priv->id.port_num, NULL); if (!ret) { - ret = cma_set_qkey(cma_dev->device, - id_priv->id.port_num, - id_priv->id.ps, dev_addr, - &id_priv->qkey); - if (!ret) - cma_attach_to_dev(id_priv, cma_dev); + cma_attach_to_dev(id_priv, cma_dev); break; } } @@ -578,6 +577,10 @@ static int cma_ib_init_qp_attr(struct rdma_id_private *id_priv, *qp_attr_mask = IB_QP_STATE | IB_QP_PKEY_INDEX | IB_QP_PORT; if (cma_is_ud_ps(id_priv->id.ps)) { + ret = cma_set_qkey(id_priv); + if (ret) + return ret; + qp_attr->qkey = id_priv->qkey; *qp_attr_mask |= IB_QP_QKEY; } else { @@ -2201,6 +2204,12 @@ static int cma_sidr_rep_handler(struct ib_cm_id *cm_id, event.status = ib_event->param.sidr_rep_rcvd.status; break; } + ret = cma_set_qkey(id_priv); + if (ret) { + event.event = RDMA_CM_EVENT_ADDR_ERROR; + event.status = -EINVAL; + break; + } if (id_priv->qkey != rep->qkey) { event.event = RDMA_CM_EVENT_UNREACHABLE; event.status = -EINVAL; @@ -2480,10 +2489,14 @@ static int cma_send_sidr_rep(struct rdma_id_private *id_priv, const void *private_data, int private_data_len) { struct ib_cm_sidr_rep_param rep; + int ret; memset(&rep, 0, sizeof rep); rep.status = status; if (status == IB_SIDR_SUCCESS) { + ret = cma_set_qkey(id_priv); + if (ret) + return ret; rep.qp_num = id_priv->qp_num; rep.qkey = id_priv->qkey; } @@ -2713,6 +2726,10 @@ static int cma_join_ib_multicast(struct rdma_id_private *id_priv, IB_SA_MCMEMBER_REC_FLOW_LABEL | IB_SA_MCMEMBER_REC_TRAFFIC_CLASS; + if (id_priv->id.ps == RDMA_PS_IPOIB) + comp_mask |= IB_SA_MCMEMBER_REC_RATE | + IB_SA_MCMEMBER_REC_RATE_SELECTOR; + mc->multicast.ib = ib_sa_join_multicast(&sa_client, id_priv->id.device, id_priv->id.port_num, &rec, comp_mask, GFP_KERNEL, diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c index a4a82bff710..8d71086f5a1 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.c +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c @@ -152,7 +152,7 @@ static int cxio_hal_clear_qp_ctx(struct cxio_rdev *rdev_p, u32 qpid) sge_cmd = qpid << 8 | 3; wqe->sge_cmd = cpu_to_be64(sge_cmd); skb->priority = CPL_PRIORITY_CONTROL; - return (cxgb3_ofld_send(rdev_p->t3cdev_p, skb)); + return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb); } int cxio_create_cq(struct cxio_rdev *rdev_p, struct t3_cq *cq) @@ -571,7 +571,7 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) (unsigned long long) rdev_p->ctrl_qp.dma_addr, rdev_p->ctrl_qp.workq, 1 << T3_CTRL_QP_SIZE_LOG2); skb->priority = CPL_PRIORITY_CONTROL; - return (cxgb3_ofld_send(rdev_p->t3cdev_p, skb)); + return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb); err: kfree_skb(skb); return err; @@ -701,7 +701,7 @@ static int __cxio_tpt_op(struct cxio_rdev *rdev_p, u32 reset_tpt_entry, u32 stag_idx; u32 wptr; - if (rdev_p->flags) + if (cxio_fatal_error(rdev_p)) return -EIO; stag_state = stag_state > 0; @@ -858,7 +858,7 @@ int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr) wqe->qp_dma_size = cpu_to_be32(attr->qp_dma_size); wqe->irs = cpu_to_be32(attr->irs); skb->priority = 0; /* 0=>ToeQ; 1=>CtrlQ */ - return (cxgb3_ofld_send(rdev_p->t3cdev_p, skb)); + return iwch_cxgb3_ofld_send(rdev_p->t3cdev_p, skb); } void cxio_register_ev_cb(cxio_hal_ev_callback_func_t ev_cb) @@ -1041,9 +1041,9 @@ void cxio_rdev_close(struct cxio_rdev *rdev_p) cxio_hal_pblpool_destroy(rdev_p); cxio_hal_rqtpool_destroy(rdev_p); list_del(&rdev_p->entry); - rdev_p->t3cdev_p->ulp = NULL; cxio_hal_destroy_ctrl_qp(rdev_p); cxio_hal_destroy_resource(rdev_p->rscp); + rdev_p->t3cdev_p->ulp = NULL; } } diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h index 094a66d1480..bfd03bf8be5 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.h +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h @@ -115,6 +115,11 @@ struct cxio_rdev { #define CXIO_ERROR_FATAL 1 }; +static inline int cxio_fatal_error(struct cxio_rdev *rdev_p) +{ + return rdev_p->flags & CXIO_ERROR_FATAL; +} + static inline int cxio_num_stags(struct cxio_rdev *rdev_p) { return min((int)T3_MAX_NUM_STAG, (int)((rdev_p->rnic_info.tpt_top - rdev_p->rnic_info.tpt_base) >> 5)); @@ -188,6 +193,7 @@ void cxio_count_scqes(struct t3_cq *cq, struct t3_wq *wq, int *count); void cxio_flush_hw_cq(struct t3_cq *cq); int cxio_poll_cq(struct t3_wq *wq, struct t3_cq *cq, struct t3_cqe *cqe, u8 *cqe_flushed, u64 *cookie, u32 *credit); +int iwch_cxgb3_ofld_send(struct t3cdev *tdev, struct sk_buff *skb); #define MOD "iw_cxgb3: " #define PDBG(fmt, args...) pr_debug(MOD fmt, ## args) diff --git a/drivers/infiniband/hw/cxgb3/iwch.c b/drivers/infiniband/hw/cxgb3/iwch.c index 37a4fc264a0..26fc0a4eaa7 100644 --- a/drivers/infiniband/hw/cxgb3/iwch.c +++ b/drivers/infiniband/hw/cxgb3/iwch.c @@ -165,12 +165,19 @@ static void close_rnic_dev(struct t3cdev *tdev) static void iwch_err_handler(struct t3cdev *tdev, u32 status, u32 error) { struct cxio_rdev *rdev = tdev->ulp; + struct iwch_dev *rnicp = rdev_to_iwch_dev(rdev); + struct ib_event event; - if (status == OFFLOAD_STATUS_DOWN) + if (status == OFFLOAD_STATUS_DOWN) { rdev->flags = CXIO_ERROR_FATAL; - return; + event.device = &rnicp->ibdev; + event.event = IB_EVENT_DEVICE_FATAL; + event.element.port_num = 0; + ib_dispatch_event(&event); + } + return; } static int __init iwch_init_module(void) diff --git a/drivers/infiniband/hw/cxgb3/iwch.h b/drivers/infiniband/hw/cxgb3/iwch.h index 3773453b2cf..84735506333 100644 --- a/drivers/infiniband/hw/cxgb3/iwch.h +++ b/drivers/infiniband/hw/cxgb3/iwch.h @@ -117,6 +117,11 @@ static inline struct iwch_dev *to_iwch_dev(struct ib_device *ibdev) return container_of(ibdev, struct iwch_dev, ibdev); } +static inline struct iwch_dev *rdev_to_iwch_dev(struct cxio_rdev *rdev) +{ + return container_of(rdev, struct iwch_dev, rdev); +} + static inline int t3b_device(const struct iwch_dev *rhp) { return rhp->rdev.t3cdev_p->type == T3B; diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.c b/drivers/infiniband/hw/cxgb3/iwch_cm.c index 8699947aaf6..fef3f1ae722 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.c @@ -139,6 +139,38 @@ static void stop_ep_timer(struct iwch_ep *ep) put_ep(&ep->com); } +int iwch_l2t_send(struct t3cdev *tdev, struct sk_buff *skb, struct l2t_entry *l2e) +{ + int error = 0; + struct cxio_rdev *rdev; + + rdev = (struct cxio_rdev *)tdev->ulp; + if (cxio_fatal_error(rdev)) { + kfree_skb(skb); + return -EIO; + } + error = l2t_send(tdev, skb, l2e); + if (error) + kfree_skb(skb); + return error; +} + +int iwch_cxgb3_ofld_send(struct t3cdev *tdev, struct sk_buff *skb) +{ + int error = 0; + struct cxio_rdev *rdev; + + rdev = (struct cxio_rdev *)tdev->ulp; + if (cxio_fatal_error(rdev)) { + kfree_skb(skb); + return -EIO; + } + error = cxgb3_ofld_send(tdev, skb); + if (error) + kfree_skb(skb); + return error; +} + static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff *skb) { struct cpl_tid_release *req; @@ -150,7 +182,7 @@ static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff *skb) req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_TID_RELEASE, hwtid)); skb->priority = CPL_PRIORITY_SETUP; - cxgb3_ofld_send(tdev, skb); + iwch_cxgb3_ofld_send(tdev, skb); return; } @@ -172,8 +204,7 @@ int iwch_quiesce_tid(struct iwch_ep *ep) req->val = cpu_to_be64(1 << S_TCB_RX_QUIESCE); skb->priority = CPL_PRIORITY_DATA; - cxgb3_ofld_send(ep->com.tdev, skb); - return 0; + return iwch_cxgb3_ofld_send(ep->com.tdev, skb); } int iwch_resume_tid(struct iwch_ep *ep) @@ -194,8 +225,7 @@ int iwch_resume_tid(struct iwch_ep *ep) req->val = 0; skb->priority = CPL_PRIORITY_DATA; - cxgb3_ofld_send(ep->com.tdev, skb); - return 0; + return iwch_cxgb3_ofld_send(ep->com.tdev, skb); } static void set_emss(struct iwch_ep *ep, u16 opt) @@ -252,18 +282,22 @@ static void *alloc_ep(int size, gfp_t gfp) void __free_ep(struct kref *kref) { - struct iwch_ep_common *epc; - epc = container_of(kref, struct iwch_ep_common, kref); - PDBG("%s ep %p state %s\n", __func__, epc, states[state_read(epc)]); - kfree(epc); + struct iwch_ep *ep; + ep = container_of(container_of(kref, struct iwch_ep_common, kref), + struct iwch_ep, com); + PDBG("%s ep %p state %s\n", __func__, ep, states[state_read(&ep->com)]); + if (ep->com.flags & RELEASE_RESOURCES) { + cxgb3_remove_tid(ep->com.tdev, (void *)ep, ep->hwtid); + dst_release(ep->dst); + l2t_release(L2DATA(ep->com.tdev), ep->l2t); + } + kfree(ep); } static void release_ep_resources(struct iwch_ep *ep) { PDBG("%s ep %p tid %d\n", __func__, ep, ep->hwtid); - cxgb3_remove_tid(ep->com.tdev, (void *)ep, ep->hwtid); - dst_release(ep->dst); - l2t_release(L2DATA(ep->com.tdev), ep->l2t); + ep->com.flags |= RELEASE_RESOURCES; put_ep(&ep->com); } @@ -382,7 +416,7 @@ static void abort_arp_failure(struct t3cdev *dev, struct sk_buff *skb) PDBG("%s t3cdev %p\n", __func__, dev); req->cmd = CPL_ABORT_NO_RST; - cxgb3_ofld_send(dev, skb); + iwch_cxgb3_ofld_send(dev, skb); } static int send_halfclose(struct iwch_ep *ep, gfp_t gfp) @@ -402,8 +436,7 @@ static int send_halfclose(struct iwch_ep *ep, gfp_t gfp) req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_OFLD_CLOSE_CON)); req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_CON_REQ, ep->hwtid)); - l2t_send(ep->com.tdev, skb, ep->l2t); - return 0; + return iwch_l2t_send(ep->com.tdev, skb, ep->l2t); } static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) @@ -424,8 +457,7 @@ static int send_abort(struct iwch_ep *ep, struct sk_buff *skb, gfp_t gfp) req->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_ABORT_REQ, ep->hwtid)); req->cmd = CPL_ABORT_SEND_RST; - l2t_send(ep->com.tdev, skb, ep->l2t); - return 0; + return iwch_l2t_send(ep->com.tdev, skb, ep->l2t); } static int send_connect(struct iwch_ep *ep) @@ -469,8 +501,7 @@ static int send_connect(struct iwch_ep *ep) req->opt0l = htonl(opt0l); req->params = 0; req->opt2 = htonl(opt2); - l2t_send(ep->com.tdev, skb, ep->l2t); - return 0; + return iwch_l2t_send(ep->com.tdev, skb, ep->l2t); } static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb) @@ -527,7 +558,7 @@ static void send_mpa_req(struct iwch_ep *ep, struct sk_buff *skb) req->sndseq = htonl(ep->snd_seq); BUG_ON(ep->mpa_skb); ep->mpa_skb = skb; - l2t_send(ep->com.tdev, skb, ep->l2t); + iwch_l2t_send(ep->com.tdev, skb, ep->l2t); start_ep_timer(ep); state_set(&ep->com, MPA_REQ_SENT); return; @@ -578,8 +609,7 @@ static int send_mpa_reject(struct iwch_ep *ep, const void *pdata, u8 plen) req->sndseq = htonl(ep->snd_seq); BUG_ON(ep->mpa_skb); ep->mpa_skb = skb; - l2t_send(ep->com.tdev, skb, ep->l2t); - return 0; + return iwch_l2t_send(ep->com.tdev, skb, ep->l2t); } static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen) @@ -630,8 +660,7 @@ static int send_mpa_reply(struct iwch_ep *ep, const void *pdata, u8 plen) req->sndseq = htonl(ep->snd_seq); ep->mpa_skb = skb; state_set(&ep->com, MPA_REP_SENT); - l2t_send(ep->com.tdev, skb, ep->l2t); - return 0; + return iwch_l2t_send(ep->com.tdev, skb, ep->l2t); } static int act_establish(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) @@ -795,7 +824,7 @@ static int update_rx_credits(struct iwch_ep *ep, u32 credits) OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_RX_DATA_ACK, ep->hwtid)); req->credit_dack = htonl(V_RX_CREDITS(credits) | V_RX_FORCE_ACK(1)); skb->priority = CPL_PRIORITY_ACK; - cxgb3_ofld_send(ep->com.tdev, skb); + iwch_cxgb3_ofld_send(ep->com.tdev, skb); return credits; } @@ -1127,8 +1156,8 @@ static int abort_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) * We get 2 abort replies from the HW. The first one must * be ignored except for scribbling that we need one more. */ - if (!(ep->flags & ABORT_REQ_IN_PROGRESS)) { - ep->flags |= ABORT_REQ_IN_PROGRESS; + if (!(ep->com.flags & ABORT_REQ_IN_PROGRESS)) { + ep->com.flags |= ABORT_REQ_IN_PROGRESS; return CPL_RET_BUF_DONE; } @@ -1203,8 +1232,7 @@ static int listen_start(struct iwch_listen_ep *ep) req->opt1 = htonl(V_CONN_POLICY(CPL_CONN_POLICY_ASK)); skb->priority = 1; - cxgb3_ofld_send(ep->com.tdev, skb); - return 0; + return iwch_cxgb3_ofld_send(ep->com.tdev, skb); } static int pass_open_rpl(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) @@ -1237,8 +1265,7 @@ static int listen_stop(struct iwch_listen_ep *ep) req->cpu_idx = 0; OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_CLOSE_LISTSRV_REQ, ep->stid)); skb->priority = 1; - cxgb3_ofld_send(ep->com.tdev, skb); - return 0; + return iwch_cxgb3_ofld_send(ep->com.tdev, skb); } static int close_listsrv_rpl(struct t3cdev *tdev, struct sk_buff *skb, @@ -1286,7 +1313,7 @@ static void accept_cr(struct iwch_ep *ep, __be32 peer_ip, struct sk_buff *skb) rpl->opt2 = htonl(opt2); rpl->rsvd = rpl->opt2; /* workaround for HW bug */ skb->priority = CPL_PRIORITY_SETUP; - l2t_send(ep->com.tdev, skb, ep->l2t); + iwch_l2t_send(ep->com.tdev, skb, ep->l2t); return; } @@ -1315,7 +1342,7 @@ static void reject_cr(struct t3cdev *tdev, u32 hwtid, __be32 peer_ip, rpl->opt0l_status = htonl(CPL_PASS_OPEN_REJECT); rpl->opt2 = 0; rpl->rsvd = rpl->opt2; - cxgb3_ofld_send(tdev, skb); + iwch_cxgb3_ofld_send(tdev, skb); } } @@ -1534,8 +1561,8 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) * We get 2 peer aborts from the HW. The first one must * be ignored except for scribbling that we need one more. */ - if (!(ep->flags & PEER_ABORT_IN_PROGRESS)) { - ep->flags |= PEER_ABORT_IN_PROGRESS; + if (!(ep->com.flags & PEER_ABORT_IN_PROGRESS)) { + ep->com.flags |= PEER_ABORT_IN_PROGRESS; return CPL_RET_BUF_DONE; } @@ -1613,7 +1640,7 @@ static int peer_abort(struct t3cdev *tdev, struct sk_buff *skb, void *ctx) rpl->wr.wr_lo = htonl(V_WR_TID(ep->hwtid)); OPCODE_TID(rpl) = htonl(MK_OPCODE_TID(CPL_ABORT_RPL, ep->hwtid)); rpl->cmd = CPL_ABORT_NO_RST; - cxgb3_ofld_send(ep->com.tdev, rpl_skb); + iwch_cxgb3_ofld_send(ep->com.tdev, rpl_skb); out: if (release) release_ep_resources(ep); @@ -2017,8 +2044,11 @@ int iwch_destroy_listen(struct iw_cm_id *cm_id) ep->com.rpl_done = 0; ep->com.rpl_err = 0; err = listen_stop(ep); + if (err) + goto done; wait_event(ep->com.waitq, ep->com.rpl_done); cxgb3_free_stid(ep->com.tdev, ep->stid); +done: err = ep->com.rpl_err; cm_id->rem_ref(cm_id); put_ep(&ep->com); @@ -2030,12 +2060,22 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp) int ret=0; unsigned long flags; int close = 0; + int fatal = 0; + struct t3cdev *tdev; + struct cxio_rdev *rdev; spin_lock_irqsave(&ep->com.lock, flags); PDBG("%s ep %p state %s, abrupt %d\n", __func__, ep, states[ep->com.state], abrupt); + tdev = (struct t3cdev *)ep->com.tdev; + rdev = (struct cxio_rdev *)tdev->ulp; + if (cxio_fatal_error(rdev)) { + fatal = 1; + close_complete_upcall(ep); + ep->com.state = DEAD; + } switch (ep->com.state) { case MPA_REQ_WAIT: case MPA_REQ_SENT: @@ -2075,7 +2115,11 @@ int iwch_ep_disconnect(struct iwch_ep *ep, int abrupt, gfp_t gfp) ret = send_abort(ep, NULL, gfp); else ret = send_halfclose(ep, gfp); + if (ret) + fatal = 1; } + if (fatal) + release_ep_resources(ep); return ret; } diff --git a/drivers/infiniband/hw/cxgb3/iwch_cm.h b/drivers/infiniband/hw/cxgb3/iwch_cm.h index d7c7e09f099..43c0aea7ead 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cm.h +++ b/drivers/infiniband/hw/cxgb3/iwch_cm.h @@ -147,6 +147,7 @@ enum iwch_ep_state { enum iwch_ep_flags { PEER_ABORT_IN_PROGRESS = (1 << 0), ABORT_REQ_IN_PROGRESS = (1 << 1), + RELEASE_RESOURCES = (1 << 2), }; struct iwch_ep_common { @@ -161,6 +162,7 @@ struct iwch_ep_common { wait_queue_head_t waitq; int rpl_done; int rpl_err; + u32 flags; }; struct iwch_listen_ep { @@ -188,7 +190,6 @@ struct iwch_ep { u16 plen; u32 ird; u32 ord; - u32 flags; }; static inline struct iwch_ep *to_ep(struct iw_cm_id *cm_id) diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index c758fbd5847..2f546a62533 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -751,7 +751,7 @@ int iwch_post_zb_read(struct iwch_qp *qhp) wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid)| V_FW_RIWR_LEN(flit_cnt)); skb->priority = CPL_PRIORITY_DATA; - return cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb); + return iwch_cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb); } /* @@ -783,7 +783,7 @@ int iwch_post_terminate(struct iwch_qp *qhp, struct respQ_msg_t *rsp_msg) V_FW_RIWR_FLAGS(T3_COMPLETION_FLAG | T3_NOTIFY_FLAG)); wqe->send.wrh.gen_tid_len = cpu_to_be32(V_FW_RIWR_TID(qhp->ep->hwtid)); skb->priority = CPL_PRIORITY_DATA; - return cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb); + return iwch_cxgb3_ofld_send(qhp->rhp->rdev.t3cdev_p, skb); } /* diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 2ccb9d31771..ae3d7590346 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -394,8 +394,7 @@ static int mlx4_ib_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) PAGE_SIZE, vma->vm_page_prot)) return -EAGAIN; } else if (vma->vm_pgoff == 1 && dev->dev->caps.bf_reg_size != 0) { - /* FIXME want pgprot_writecombine() for BlueFlame pages */ - vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); if (io_remap_pfn_range(vma, vma->vm_start, to_mucontext(context)->uar.pfn + diff --git a/drivers/infiniband/hw/nes/nes.h b/drivers/infiniband/hw/nes/nes.h index 04b12ad2339..17621de54a9 100644 --- a/drivers/infiniband/hw/nes/nes.h +++ b/drivers/infiniband/hw/nes/nes.h @@ -289,8 +289,8 @@ static inline __le32 get_crc_value(struct nes_v4_quad *nes_quad) static inline void set_wqe_64bit_value(__le32 *wqe_words, u32 index, u64 value) { - wqe_words[index] = cpu_to_le32((u32) ((unsigned long)value)); - wqe_words[index + 1] = cpu_to_le32((u32)(upper_32_bits((unsigned long)value))); + wqe_words[index] = cpu_to_le32((u32) value); + wqe_words[index + 1] = cpu_to_le32(upper_32_bits(value)); } static inline void diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 52425154acd..dbd9a75474e 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -426,6 +426,7 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb, if (type == NES_TIMER_TYPE_CLOSE) { new_send->timetosend += (HZ/10); if (cm_node->recv_entry) { + kfree(new_send); WARN_ON(1); return -EINVAL; } @@ -445,8 +446,8 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb, if (ret != NETDEV_TX_OK) { nes_debug(NES_DBG_CM, "Error sending packet %p " "(jiffies = %lu)\n", new_send, jiffies); - atomic_dec(&new_send->skb->users); new_send->timetosend = jiffies; + ret = NETDEV_TX_OK; } else { cm_packets_sent++; if (!send_retrans) { @@ -630,7 +631,6 @@ static void nes_cm_timer_tick(unsigned long pass) nes_debug(NES_DBG_CM, "rexmit failed for " "node=%p\n", cm_node); cm_packets_bounced++; - atomic_dec(&send_entry->skb->users); send_entry->retrycount--; nexttimeout = jiffies + NES_SHORT_TIME; settimer = 1; @@ -666,11 +666,6 @@ static void nes_cm_timer_tick(unsigned long pass) spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); rem_ref_cm_node(cm_node->cm_core, cm_node); - if (ret != NETDEV_TX_OK) { - nes_debug(NES_DBG_CM, "rexmit failed for cm_node=%p\n", - cm_node); - break; - } } if (settimer) { @@ -1262,7 +1257,6 @@ static int rem_ref_cm_node(struct nes_cm_core *cm_core, cm_node->nesqp = NULL; } - cm_node->freed = 1; kfree(cm_node); return 0; } @@ -1999,13 +1993,17 @@ static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core, if (loopbackremotelistener == NULL) { create_event(cm_node, NES_CM_EVENT_ABORTED); } else { - atomic_inc(&cm_loopbacks); loopback_cm_info = *cm_info; loopback_cm_info.loc_port = cm_info->rem_port; loopback_cm_info.rem_port = cm_info->loc_port; loopback_cm_info.cm_id = loopbackremotelistener->cm_id; loopbackremotenode = make_cm_node(cm_core, nesvnic, &loopback_cm_info, loopbackremotelistener); + if (!loopbackremotenode) { + rem_ref_cm_node(cm_node->cm_core, cm_node); + return NULL; + } + atomic_inc(&cm_loopbacks); loopbackremotenode->loopbackpartner = cm_node; loopbackremotenode->tcp_cntxt.rcv_wscale = NES_CM_DEFAULT_RCV_WND_SCALE; @@ -2690,6 +2688,7 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) struct ib_mr *ibmr = NULL; struct ib_phys_buf ibphysbuf; struct nes_pd *nespd; + u64 tagged_offset; @@ -2755,10 +2754,11 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) ibphysbuf.addr = nesqp->ietf_frame_pbase; ibphysbuf.size = conn_param->private_data_len + sizeof(struct ietf_mpa_frame); + tagged_offset = (u64)(unsigned long)nesqp->ietf_frame; ibmr = nesibdev->ibdev.reg_phys_mr((struct ib_pd *)nespd, &ibphysbuf, 1, IB_ACCESS_LOCAL_WRITE, - (u64 *)&nesqp->ietf_frame); + &tagged_offset); if (!ibmr) { nes_debug(NES_DBG_CM, "Unable to register memory region" "for lSMM for cm_node = %p \n", @@ -2782,7 +2782,7 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) sizeof(struct ietf_mpa_frame)); set_wqe_64bit_value(wqe->wqe_words, NES_IWARP_SQ_WQE_FRAG0_LOW_IDX, - (u64)nesqp->ietf_frame); + (u64)(unsigned long)nesqp->ietf_frame); wqe->wqe_words[NES_IWARP_SQ_WQE_LENGTH0_IDX] = cpu_to_le32(conn_param->private_data_len + sizeof(struct ietf_mpa_frame)); diff --git a/drivers/infiniband/hw/nes/nes_cm.h b/drivers/infiniband/hw/nes/nes_cm.h index d5f778202eb..80bba189257 100644 --- a/drivers/infiniband/hw/nes/nes_cm.h +++ b/drivers/infiniband/hw/nes/nes_cm.h @@ -298,7 +298,6 @@ struct nes_cm_node { struct nes_vnic *nesvnic; int apbvt_set; int accept_pend; - int freed; struct list_head timer_entry; struct list_head reset_entry; struct nes_qp *nesqp; diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index 52e734042b8..d6fc9ae4406 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -46,6 +46,10 @@ static unsigned int nes_lro_max_aggr = NES_LRO_MAX_AGGR; module_param(nes_lro_max_aggr, uint, 0444); MODULE_PARM_DESC(nes_lro_max_aggr, "NIC LRO max packet aggregation"); +static int wide_ppm_offset; +module_param(wide_ppm_offset, int, 0644); +MODULE_PARM_DESC(wide_ppm_offset, "Increase CX4 interface clock ppm offset, 0=100ppm (default), 1=300ppm"); + static u32 crit_err_count; u32 int_mod_timer_init; u32 int_mod_cq_depth_256; @@ -546,8 +550,11 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { msleep(1); } if (int_cnt > 1) { + u32 sds; spin_lock_irqsave(&nesadapter->phy_lock, flags); - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1, 0x0000F088); + sds = nes_read_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1); + sds |= 0x00000040; + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1, sds); mh_detected++; reset_value = nes_read32(nesdev->regs+NES_SOFTWARE_RESET); reset_value |= 0x0000003d; @@ -736,39 +743,49 @@ static int nes_init_serdes(struct nes_device *nesdev, u8 hw_rev, u8 port_count, { int i; u32 u32temp; - u32 serdes_common_control; + u32 sds; if (hw_rev != NE020_REV) { /* init serdes 0 */ + if (wide_ppm_offset && (nesadapter->phy_type[0] == NES_PHY_TYPE_CX4)) + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_CDR_CONTROL0, 0x000FFFAA); + else + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_CDR_CONTROL0, 0x000000FF); - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_CDR_CONTROL0, 0x000000FF); if (nesadapter->phy_type[0] == NES_PHY_TYPE_PUMA_1G) { - serdes_common_control = nes_read_indexed(nesdev, - NES_IDX_ETH_SERDES_COMMON_CONTROL0); - serdes_common_control |= 0x000000100; - nes_write_indexed(nesdev, - NES_IDX_ETH_SERDES_COMMON_CONTROL0, - serdes_common_control); - } else if (!OneG_Mode) { - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_TX_HIGHZ_LANE_MODE0, 0x11110000); + sds = nes_read_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0); + sds |= 0x00000100; + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0, sds); } - if (((port_count > 1) && - (nesadapter->phy_type[0] != NES_PHY_TYPE_PUMA_1G)) || - ((port_count > 2) && - (nesadapter->phy_type[0] == NES_PHY_TYPE_PUMA_1G))) { - /* init serdes 1 */ - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_CDR_CONTROL1, 0x000000FF); - if (nesadapter->phy_type[0] == NES_PHY_TYPE_PUMA_1G) { - serdes_common_control = nes_read_indexed(nesdev, - NES_IDX_ETH_SERDES_COMMON_CONTROL1); - serdes_common_control |= 0x000000100; - nes_write_indexed(nesdev, - NES_IDX_ETH_SERDES_COMMON_CONTROL1, - serdes_common_control); - } else if (!OneG_Mode) { - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_TX_HIGHZ_LANE_MODE1, 0x11110000); - } + if (!OneG_Mode) + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_TX_HIGHZ_LANE_MODE0, 0x11110000); + + if (port_count < 2) + return 0; + + /* init serdes 1 */ + switch (nesadapter->phy_type[1]) { + case NES_PHY_TYPE_ARGUS: + case NES_PHY_TYPE_SFP_D: + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_TX_EMP0, 0x00000000); + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_TX_EMP1, 0x00000000); + break; + case NES_PHY_TYPE_CX4: + sds = nes_read_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1); + sds &= 0xFFFFFFBF; + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1, sds); + if (wide_ppm_offset) + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_CDR_CONTROL1, 0x000FFFAA); + else + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_CDR_CONTROL1, 0x000000FF); + break; + case NES_PHY_TYPE_PUMA_1G: + sds = nes_read_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1); + sds |= 0x000000100; + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL1, sds); } + if (!OneG_Mode) + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_TX_HIGHZ_LANE_MODE1, 0x11110000); } else { /* init serdes 0 */ nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0, 0x00000008); @@ -1259,203 +1276,162 @@ int nes_init_phy(struct nes_device *nesdev) { struct nes_adapter *nesadapter = nesdev->nesadapter; u32 counter = 0; - u32 sds_common_control0; + u32 sds; u32 mac_index = nesdev->mac_index; u32 tx_config = 0; u16 phy_data; u32 temp_phy_data = 0; u32 temp_phy_data2 = 0; - u32 i = 0; + u8 phy_type = nesadapter->phy_type[mac_index]; + u8 phy_index = nesadapter->phy_index[mac_index]; if ((nesadapter->OneG_Mode) && - (nesadapter->phy_type[mac_index] != NES_PHY_TYPE_PUMA_1G)) { + (phy_type != NES_PHY_TYPE_PUMA_1G)) { nes_debug(NES_DBG_PHY, "1G PHY, mac_index = %d.\n", mac_index); - if (nesadapter->phy_type[mac_index] == NES_PHY_TYPE_1G) { - printk(PFX "%s: Programming mdc config for 1G\n", __func__); + if (phy_type == NES_PHY_TYPE_1G) { tx_config = nes_read_indexed(nesdev, NES_IDX_MAC_TX_CONFIG); tx_config &= 0xFFFFFFE3; tx_config |= 0x04; nes_write_indexed(nesdev, NES_IDX_MAC_TX_CONFIG, tx_config); } - nes_read_1G_phy_reg(nesdev, 1, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 1 phy address %u = 0x%X.\n", - nesadapter->phy_index[mac_index], phy_data); - nes_write_1G_phy_reg(nesdev, 23, nesadapter->phy_index[mac_index], 0xb000); + nes_read_1G_phy_reg(nesdev, 1, phy_index, &phy_data); + nes_write_1G_phy_reg(nesdev, 23, phy_index, 0xb000); /* Reset the PHY */ - nes_write_1G_phy_reg(nesdev, 0, nesadapter->phy_index[mac_index], 0x8000); + nes_write_1G_phy_reg(nesdev, 0, phy_index, 0x8000); udelay(100); counter = 0; do { - nes_read_1G_phy_reg(nesdev, 0, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0 = 0x%X.\n", phy_data); - if (counter++ > 100) break; + nes_read_1G_phy_reg(nesdev, 0, phy_index, &phy_data); + if (counter++ > 100) + break; } while (phy_data & 0x8000); /* Setting no phy loopback */ phy_data &= 0xbfff; phy_data |= 0x1140; - nes_write_1G_phy_reg(nesdev, 0, nesadapter->phy_index[mac_index], phy_data); - nes_read_1G_phy_reg(nesdev, 0, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0 = 0x%X.\n", phy_data); - - nes_read_1G_phy_reg(nesdev, 0x17, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x17 = 0x%X.\n", phy_data); - - nes_read_1G_phy_reg(nesdev, 0x1e, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x1e = 0x%X.\n", phy_data); + nes_write_1G_phy_reg(nesdev, 0, phy_index, phy_data); + nes_read_1G_phy_reg(nesdev, 0, phy_index, &phy_data); + nes_read_1G_phy_reg(nesdev, 0x17, phy_index, &phy_data); + nes_read_1G_phy_reg(nesdev, 0x1e, phy_index, &phy_data); /* Setting the interrupt mask */ - nes_read_1G_phy_reg(nesdev, 0x19, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x19 = 0x%X.\n", phy_data); - nes_write_1G_phy_reg(nesdev, 0x19, nesadapter->phy_index[mac_index], 0xffee); - - nes_read_1G_phy_reg(nesdev, 0x19, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x19 = 0x%X.\n", phy_data); + nes_read_1G_phy_reg(nesdev, 0x19, phy_index, &phy_data); + nes_write_1G_phy_reg(nesdev, 0x19, phy_index, 0xffee); + nes_read_1G_phy_reg(nesdev, 0x19, phy_index, &phy_data); /* turning on flow control */ - nes_read_1G_phy_reg(nesdev, 4, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x4 = 0x%X.\n", phy_data); - nes_write_1G_phy_reg(nesdev, 4, nesadapter->phy_index[mac_index], - (phy_data & ~(0x03E0)) | 0xc00); - /* nes_write_1G_phy_reg(nesdev, 4, nesadapter->phy_index[mac_index], - phy_data | 0xc00); */ - nes_read_1G_phy_reg(nesdev, 4, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x4 = 0x%X.\n", phy_data); - - nes_read_1G_phy_reg(nesdev, 9, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x9 = 0x%X.\n", phy_data); - /* Clear Half duplex */ - nes_write_1G_phy_reg(nesdev, 9, nesadapter->phy_index[mac_index], - phy_data & ~(0x0100)); - nes_read_1G_phy_reg(nesdev, 9, nesadapter->phy_index[mac_index], &phy_data); - nes_debug(NES_DBG_PHY, "Phy data from register 0x9 = 0x%X.\n", phy_data); - - nes_read_1G_phy_reg(nesdev, 0, nesadapter->phy_index[mac_index], &phy_data); - nes_write_1G_phy_reg(nesdev, 0, nesadapter->phy_index[mac_index], phy_data | 0x0300); - } else { - if ((nesadapter->phy_type[mac_index] == NES_PHY_TYPE_IRIS) || - (nesadapter->phy_type[mac_index] == NES_PHY_TYPE_ARGUS)) { - /* setup 10G MDIO operation */ - tx_config = nes_read_indexed(nesdev, NES_IDX_MAC_TX_CONFIG); - tx_config &= 0xFFFFFFE3; - tx_config |= 0x15; - nes_write_indexed(nesdev, NES_IDX_MAC_TX_CONFIG, tx_config); - } - if ((nesadapter->phy_type[mac_index] == NES_PHY_TYPE_ARGUS)) { - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0xd7ee); + nes_read_1G_phy_reg(nesdev, 4, phy_index, &phy_data); + nes_write_1G_phy_reg(nesdev, 4, phy_index, (phy_data & ~(0x03E0)) | 0xc00); + nes_read_1G_phy_reg(nesdev, 4, phy_index, &phy_data); - temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); - mdelay(10); - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0xd7ee); - temp_phy_data2 = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + /* Clear Half duplex */ + nes_read_1G_phy_reg(nesdev, 9, phy_index, &phy_data); + nes_write_1G_phy_reg(nesdev, 9, phy_index, phy_data & ~(0x0100)); + nes_read_1G_phy_reg(nesdev, 9, phy_index, &phy_data); - /* - * if firmware is already running (like from a - * driver un-load/load, don't do anything. - */ - if (temp_phy_data == temp_phy_data2) { - /* configure QT2505 AMCC PHY */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0x0000, 0x8000); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc300, 0x0000); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc302, 0x0044); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc318, 0x0052); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc319, 0x0008); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc31a, 0x0098); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0x0026, 0x0E00); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0x0027, 0x0001); - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0x0028, 0xA528); + nes_read_1G_phy_reg(nesdev, 0, phy_index, &phy_data); + nes_write_1G_phy_reg(nesdev, 0, phy_index, phy_data | 0x0300); - /* - * remove micro from reset; chip boots from ROM, - * uploads EEPROM f/w image, uC executes f/w - */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc300, 0x0002); + return 0; + } - /* - * wait for heart beat to start to - * know loading is done - */ - counter = 0; - do { - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0xd7ee); - temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); - if (counter++ > 1000) { - nes_debug(NES_DBG_PHY, "AMCC PHY- breaking from heartbeat check <this is bad!!!> \n"); - break; - } - mdelay(100); - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0xd7ee); - temp_phy_data2 = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); - } while ((temp_phy_data2 == temp_phy_data)); + if ((phy_type == NES_PHY_TYPE_IRIS) || + (phy_type == NES_PHY_TYPE_ARGUS) || + (phy_type == NES_PHY_TYPE_SFP_D)) { + /* setup 10G MDIO operation */ + tx_config = nes_read_indexed(nesdev, NES_IDX_MAC_TX_CONFIG); + tx_config &= 0xFFFFFFE3; + tx_config |= 0x15; + nes_write_indexed(nesdev, NES_IDX_MAC_TX_CONFIG, tx_config); + } + if ((phy_type == NES_PHY_TYPE_ARGUS) || + (phy_type == NES_PHY_TYPE_SFP_D)) { + /* Check firmware heartbeat */ + nes_read_10G_phy_reg(nesdev, phy_index, 0x3, 0xd7ee); + temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + udelay(1500); + nes_read_10G_phy_reg(nesdev, phy_index, 0x3, 0xd7ee); + temp_phy_data2 = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + + if (temp_phy_data != temp_phy_data2) + return 0; - /* - * wait for tracking to start to know - * f/w is good to go - */ - counter = 0; - do { - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x3, 0xd7fd); - temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); - if (counter++ > 1000) { - nes_debug(NES_DBG_PHY, "AMCC PHY- breaking from status check <this is bad!!!> \n"); - break; - } - mdelay(1000); - /* - * nes_debug(NES_DBG_PHY, "AMCC PHY- phy_status not ready yet = 0x%02X\n", - * temp_phy_data); - */ - } while (((temp_phy_data & 0xff) != 0x50) && ((temp_phy_data & 0xff) != 0x70)); - - /* set LOS Control invert RXLOSB_I_PADINV */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xd003, 0x0000); - /* set LOS Control to mask of RXLOSB_I */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xc314, 0x0042); - /* set LED1 to input mode (LED1 and LED2 share same LED) */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xd006, 0x0007); - /* set LED2 to RX link_status and activity */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xd007, 0x000A); - /* set LED3 to RX link_status */ - nes_write_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 0x1, 0xd008, 0x0009); + /* no heartbeat, configure the PHY */ + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0x0000, 0x8000); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc300, 0x0000); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc316, 0x000A); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc318, 0x0052); + if (phy_type == NES_PHY_TYPE_ARGUS) { + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc302, 0x000C); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc319, 0x0008); + } else { + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc302, 0x0004); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc319, 0x0038); + } + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc31a, 0x0098); + nes_write_10G_phy_reg(nesdev, phy_index, 0x3, 0x0026, 0x0E00); + nes_write_10G_phy_reg(nesdev, phy_index, 0x3, 0x0027, 0x0001); - /* - * reset the res-calibration on t2 - * serdes; ensures it is stable after - * the amcc phy is stable - */ + /* setup LEDs */ + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xd006, 0x0007); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xd007, 0x000A); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xd008, 0x0009); - sds_common_control0 = nes_read_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0); - sds_common_control0 |= 0x1; - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0, sds_common_control0); + nes_write_10G_phy_reg(nesdev, phy_index, 0x3, 0x0028, 0xA528); - /* release the res-calibration reset */ - sds_common_control0 &= 0xfffffffe; - nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0, sds_common_control0); + /* Bring PHY out of reset */ + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc300, 0x0002); - i = 0; - while (((nes_read32(nesdev->regs + NES_SOFTWARE_RESET) & 0x00000040) != 0x00000040) - && (i++ < 5000)) { - /* mdelay(1); */ - } + /* Check for heartbeat */ + counter = 0; + mdelay(690); + nes_read_10G_phy_reg(nesdev, phy_index, 0x3, 0xd7ee); + temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + do { + if (counter++ > 150) { + nes_debug(NES_DBG_PHY, "No PHY heartbeat\n"); + break; + } + mdelay(1); + nes_read_10G_phy_reg(nesdev, phy_index, 0x3, 0xd7ee); + temp_phy_data2 = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + } while ((temp_phy_data2 == temp_phy_data)); - /* - * wait for link train done before moving on, - * or will get an interupt storm - */ - counter = 0; - do { - temp_phy_data = nes_read_indexed(nesdev, NES_IDX_PHY_PCS_CONTROL_STATUS0 + - (0x200 * (nesdev->mac_index & 1))); - if (counter++ > 1000) { - nes_debug(NES_DBG_PHY, "AMCC PHY- breaking from link train wait <this is bad, link didnt train!!!>\n"); - break; - } - mdelay(1); - } while (((temp_phy_data & 0x0f1f0000) != 0x0f0f0000)); + /* wait for tracking */ + counter = 0; + do { + nes_read_10G_phy_reg(nesdev, phy_index, 0x3, 0xd7fd); + temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + if (counter++ > 300) { + nes_debug(NES_DBG_PHY, "PHY did not track\n"); + break; } - } + mdelay(10); + } while (((temp_phy_data & 0xff) != 0x50) && ((temp_phy_data & 0xff) != 0x70)); + + /* setup signal integrity */ + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xd003, 0x0000); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xF00D, 0x00FE); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xF00E, 0x0032); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xF00F, 0x0002); + nes_write_10G_phy_reg(nesdev, phy_index, 0x1, 0xc314, 0x0063); + + /* reset serdes */ + sds = nes_read_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0 + + mac_index * 0x200); + sds |= 0x1; + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0 + + mac_index * 0x200, sds); + sds &= 0xfffffffe; + nes_write_indexed(nesdev, NES_IDX_ETH_SERDES_COMMON_CONTROL0 + + mac_index * 0x200, sds); + + counter = 0; + while (((nes_read32(nesdev->regs + NES_SOFTWARE_RESET) & 0x00000040) != 0x00000040) + && (counter++ < 5000)) + ; } return 0; } @@ -2359,6 +2335,7 @@ static void nes_process_mac_intr(struct nes_device *nesdev, u32 mac_number) u16 temp_phy_data; u32 pcs_val = 0x0f0f0000; u32 pcs_mask = 0x0f1f0000; + u32 cdr_ctrl; spin_lock_irqsave(&nesadapter->phy_lock, flags); if (nesadapter->mac_sw_state[mac_number] != NES_MAC_SW_IDLE) { @@ -2473,6 +2450,7 @@ static void nes_process_mac_intr(struct nes_device *nesdev, u32 mac_number) break; case NES_PHY_TYPE_ARGUS: + case NES_PHY_TYPE_SFP_D: /* clear the alarms */ nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 4, 0x0008); nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 4, 0xc001); @@ -2483,19 +2461,18 @@ static void nes_process_mac_intr(struct nes_device *nesdev, u32 mac_number) nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 1, 0x9004); nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 1, 0x9005); /* check link status */ - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 1, 1); + nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 1, 0x9003); temp_phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); - u32temp = 100; - do { - nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 1, 1); - phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); - if ((phy_data == temp_phy_data) || (!(--u32temp))) - break; - temp_phy_data = phy_data; - } while (1); + nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 3, 0x0021); + nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + nes_read_10G_phy_reg(nesdev, nesadapter->phy_index[mac_index], 3, 0x0021); + phy_data = (u16)nes_read_indexed(nesdev, NES_IDX_MAC_MDIO_CONTROL); + + phy_data = (!temp_phy_data && (phy_data == 0x8000)) ? 0x4 : 0x0; + nes_debug(NES_DBG_PHY, "%s: Phy data = 0x%04X, link was %s.\n", - __func__, phy_data, nesadapter->mac_link_down ? "DOWN" : "UP"); + __func__, phy_data, nesadapter->mac_link_down[mac_index] ? "DOWN" : "UP"); break; case NES_PHY_TYPE_PUMA_1G: @@ -2511,6 +2488,17 @@ static void nes_process_mac_intr(struct nes_device *nesdev, u32 mac_number) } if (phy_data & 0x0004) { + if (wide_ppm_offset && + (nesadapter->phy_type[mac_index] == NES_PHY_TYPE_CX4) && + (nesadapter->hw_rev != NE020_REV)) { + cdr_ctrl = nes_read_indexed(nesdev, + NES_IDX_ETH_SERDES_CDR_CONTROL0 + + mac_index * 0x200); + nes_write_indexed(nesdev, + NES_IDX_ETH_SERDES_CDR_CONTROL0 + + mac_index * 0x200, + cdr_ctrl | 0x000F0000); + } nesadapter->mac_link_down[mac_index] = 0; list_for_each_entry(nesvnic, &nesadapter->nesvnic_list[mac_index], list) { nes_debug(NES_DBG_PHY, "The Link is UP!!. linkup was %d\n", @@ -2525,6 +2513,17 @@ static void nes_process_mac_intr(struct nes_device *nesdev, u32 mac_number) } } } else { + if (wide_ppm_offset && + (nesadapter->phy_type[mac_index] == NES_PHY_TYPE_CX4) && + (nesadapter->hw_rev != NE020_REV)) { + cdr_ctrl = nes_read_indexed(nesdev, + NES_IDX_ETH_SERDES_CDR_CONTROL0 + + mac_index * 0x200); + nes_write_indexed(nesdev, + NES_IDX_ETH_SERDES_CDR_CONTROL0 + + mac_index * 0x200, + cdr_ctrl & 0xFFF0FFFF); + } nesadapter->mac_link_down[mac_index] = 1; list_for_each_entry(nesvnic, &nesadapter->nesvnic_list[mac_index], list) { nes_debug(NES_DBG_PHY, "The Link is Down!!. linkup was %d\n", diff --git a/drivers/infiniband/hw/nes/nes_hw.h b/drivers/infiniband/hw/nes/nes_hw.h index f41a8710d2a..c3654c6383f 100644 --- a/drivers/infiniband/hw/nes/nes_hw.h +++ b/drivers/infiniband/hw/nes/nes_hw.h @@ -35,12 +35,14 @@ #include <linux/inet_lro.h> +#define NES_PHY_TYPE_CX4 1 #define NES_PHY_TYPE_1G 2 #define NES_PHY_TYPE_IRIS 3 #define NES_PHY_TYPE_ARGUS 4 #define NES_PHY_TYPE_PUMA_1G 5 #define NES_PHY_TYPE_PUMA_10G 6 #define NES_PHY_TYPE_GLADIUS 7 +#define NES_PHY_TYPE_SFP_D 8 #define NES_MULTICAST_PF_MAX 8 diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c index ecb1f6fd627..c6e6611d301 100644 --- a/drivers/infiniband/hw/nes/nes_nic.c +++ b/drivers/infiniband/hw/nes/nes_nic.c @@ -1426,49 +1426,55 @@ static int nes_netdev_get_settings(struct net_device *netdev, struct ethtool_cmd struct nes_vnic *nesvnic = netdev_priv(netdev); struct nes_device *nesdev = nesvnic->nesdev; struct nes_adapter *nesadapter = nesdev->nesadapter; + u32 mac_index = nesdev->mac_index; + u8 phy_type = nesadapter->phy_type[mac_index]; + u8 phy_index = nesadapter->phy_index[mac_index]; u16 phy_data; et_cmd->duplex = DUPLEX_FULL; et_cmd->port = PORT_MII; + et_cmd->maxtxpkt = 511; + et_cmd->maxrxpkt = 511; if (nesadapter->OneG_Mode) { et_cmd->speed = SPEED_1000; - if (nesadapter->phy_type[nesdev->mac_index] == NES_PHY_TYPE_PUMA_1G) { + if (phy_type == NES_PHY_TYPE_PUMA_1G) { et_cmd->supported = SUPPORTED_1000baseT_Full; et_cmd->advertising = ADVERTISED_1000baseT_Full; et_cmd->autoneg = AUTONEG_DISABLE; et_cmd->transceiver = XCVR_INTERNAL; - et_cmd->phy_address = nesdev->mac_index; + et_cmd->phy_address = mac_index; } else { - et_cmd->supported = SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg; - et_cmd->advertising = ADVERTISED_1000baseT_Full | ADVERTISED_Autoneg; - nes_read_1G_phy_reg(nesdev, 0, nesadapter->phy_index[nesdev->mac_index], &phy_data); + et_cmd->supported = SUPPORTED_1000baseT_Full + | SUPPORTED_Autoneg; + et_cmd->advertising = ADVERTISED_1000baseT_Full + | ADVERTISED_Autoneg; + nes_read_1G_phy_reg(nesdev, 0, phy_index, &phy_data); if (phy_data & 0x1000) et_cmd->autoneg = AUTONEG_ENABLE; else et_cmd->autoneg = AUTONEG_DISABLE; et_cmd->transceiver = XCVR_EXTERNAL; - et_cmd->phy_address = nesadapter->phy_index[nesdev->mac_index]; + et_cmd->phy_address = phy_index; } + return 0; + } + if ((phy_type == NES_PHY_TYPE_IRIS) || + (phy_type == NES_PHY_TYPE_ARGUS) || + (phy_type == NES_PHY_TYPE_SFP_D)) { + et_cmd->transceiver = XCVR_EXTERNAL; + et_cmd->port = PORT_FIBRE; + et_cmd->supported = SUPPORTED_FIBRE; + et_cmd->advertising = ADVERTISED_FIBRE; + et_cmd->phy_address = phy_index; } else { - if ((nesadapter->phy_type[nesdev->mac_index] == NES_PHY_TYPE_IRIS) || - (nesadapter->phy_type[nesdev->mac_index] == NES_PHY_TYPE_ARGUS)) { - et_cmd->transceiver = XCVR_EXTERNAL; - et_cmd->port = PORT_FIBRE; - et_cmd->supported = SUPPORTED_FIBRE; - et_cmd->advertising = ADVERTISED_FIBRE; - et_cmd->phy_address = nesadapter->phy_index[nesdev->mac_index]; - } else { - et_cmd->transceiver = XCVR_INTERNAL; - et_cmd->supported = SUPPORTED_10000baseT_Full; - et_cmd->advertising = ADVERTISED_10000baseT_Full; - et_cmd->phy_address = nesdev->mac_index; - } - et_cmd->speed = SPEED_10000; - et_cmd->autoneg = AUTONEG_DISABLE; + et_cmd->transceiver = XCVR_INTERNAL; + et_cmd->supported = SUPPORTED_10000baseT_Full; + et_cmd->advertising = ADVERTISED_10000baseT_Full; + et_cmd->phy_address = mac_index; } - et_cmd->maxtxpkt = 511; - et_cmd->maxrxpkt = 511; + et_cmd->speed = SPEED_10000; + et_cmd->autoneg = AUTONEG_DISABLE; return 0; } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 5a76a551035..4c57f329dd5 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -70,12 +70,14 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) */ if (ppriv->pkey == pkey) { result = -ENOTUNIQ; + priv = NULL; goto err; } list_for_each_entry(priv, &ppriv->child_intfs, list) { if (priv->pkey == pkey) { result = -ENOTUNIQ; + priv = NULL; goto err; } } @@ -96,7 +98,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) result = ipoib_set_dev_features(priv, ppriv->ca); if (result) - goto device_init_failed; + goto err; priv->pkey = pkey; @@ -109,7 +111,7 @@ int ipoib_vlan_add(struct net_device *pdev, unsigned short pkey) ipoib_warn(ppriv, "failed to initialize subinterface: " "device %s, port %d", ppriv->ca->name, ppriv->port); - goto device_init_failed; + goto err; } result = register_netdevice(priv->dev); @@ -146,19 +148,19 @@ sysfs_failed: register_failed: ipoib_dev_cleanup(priv->dev); -device_init_failed: - free_netdev(priv->dev); - err: mutex_unlock(&ppriv->vlan_mutex); rtnl_unlock(); + if (priv) + free_netdev(priv->dev); + return result; } int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) { struct ipoib_dev_priv *ppriv, *priv, *tpriv; - int ret = -ENOENT; + struct net_device *dev = NULL; if (!capable(CAP_NET_ADMIN)) return -EPERM; @@ -172,14 +174,17 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) unregister_netdevice(priv->dev); ipoib_dev_cleanup(priv->dev); list_del(&priv->list); - free_netdev(priv->dev); - - ret = 0; + dev = priv->dev; break; } } mutex_unlock(&ppriv->vlan_mutex); rtnl_unlock(); - return ret; + if (dev) { + free_netdev(dev); + return 0; + } + + return -ENODEV; } diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index f01096549a9..823ceba6efa 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -1047,6 +1047,19 @@ static int populate_table(struct dm_table *table, return dm_table_complete(table); } +static int table_prealloc_integrity(struct dm_table *t, + struct mapped_device *md) +{ + struct list_head *devices = dm_table_get_devices(t); + struct dm_dev_internal *dd; + + list_for_each_entry(dd, devices, list) + if (bdev_get_integrity(dd->dm_dev.bdev)) + return blk_integrity_register(dm_disk(md), NULL); + + return 0; +} + static int table_load(struct dm_ioctl *param, size_t param_size) { int r; @@ -1068,6 +1081,14 @@ static int table_load(struct dm_ioctl *param, size_t param_size) goto out; } + r = table_prealloc_integrity(t, md); + if (r) { + DMERR("%s: could not register integrity profile.", + dm_device_name(md)); + dm_table_destroy(t); + goto out; + } + down_write(&_hash_lock); hc = dm_get_mdptr(md); if (!hc || hc->md != md) { diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index 0a225da2127..3e3fc06cb86 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -297,7 +297,8 @@ static int run_complete_job(struct kcopyd_job *job) dm_kcopyd_notify_fn fn = job->fn; struct dm_kcopyd_client *kc = job->kc; - kcopyd_put_pages(kc, job->pages); + if (job->pages) + kcopyd_put_pages(kc, job->pages); mempool_free(job, kc->job_pool); fn(read_err, write_err, context); @@ -461,6 +462,7 @@ static void segment_complete(int read_err, unsigned long write_err, sector_t progress = 0; sector_t count = 0; struct kcopyd_job *job = (struct kcopyd_job *) context; + struct dm_kcopyd_client *kc = job->kc; mutex_lock(&job->lock); @@ -490,7 +492,7 @@ static void segment_complete(int read_err, unsigned long write_err, if (count) { int i; - struct kcopyd_job *sub_job = mempool_alloc(job->kc->job_pool, + struct kcopyd_job *sub_job = mempool_alloc(kc->job_pool, GFP_NOIO); *sub_job = *job; @@ -509,13 +511,16 @@ static void segment_complete(int read_err, unsigned long write_err, } else if (atomic_dec_and_test(&job->sub_jobs)) { /* - * To avoid a race we must keep the job around - * until after the notify function has completed. - * Otherwise the client may try and stop the job - * after we've completed. + * Queue the completion callback to the kcopyd thread. + * + * Some callers assume that all the completions are called + * from a single thread and don't race with each other. + * + * We must not call the callback directly here because this + * code may not be executing in the thread. */ - job->fn(read_err, write_err, job->context); - mempool_free(job, job->kc->job_pool); + push(&kc->complete_jobs, job); + wake(kc); } } @@ -528,6 +533,8 @@ static void split_job(struct kcopyd_job *job) { int i; + atomic_inc(&job->kc->nr_jobs); + atomic_set(&job->sub_jobs, SPLIT_COUNT); for (i = 0; i < SPLIT_COUNT; i++) segment_complete(0, 0u, job); diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c index bfa107f59d9..79fb53e51c7 100644 --- a/drivers/md/dm-linear.c +++ b/drivers/md/dm-linear.c @@ -142,7 +142,6 @@ static struct target_type linear_target = { .status = linear_status, .ioctl = linear_ioctl, .merge = linear_merge, - .features = DM_TARGET_SUPPORTS_BARRIERS, }; int __init dm_linear_init(void) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index e8361b191b9..429b50b975d 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -52,8 +52,6 @@ struct dm_table { sector_t *highs; struct dm_target *targets; - unsigned barriers_supported:1; - /* * Indicates the rw permissions for the new logical * device. This should be a combination of FMODE_READ @@ -243,7 +241,6 @@ int dm_table_create(struct dm_table **result, fmode_t mode, INIT_LIST_HEAD(&t->devices); atomic_set(&t->holders, 0); - t->barriers_supported = 1; if (!num_targets) num_targets = KEYS_PER_NODE; @@ -751,10 +748,6 @@ int dm_table_add_target(struct dm_table *t, const char *type, /* FIXME: the plan is to combine high here and then have * the merge fn apply the target level restrictions. */ combine_restrictions_low(&t->limits, &tgt->limits); - - if (!(tgt->type->features & DM_TARGET_SUPPORTS_BARRIERS)) - t->barriers_supported = 0; - return 0; bad: @@ -799,12 +792,6 @@ int dm_table_complete(struct dm_table *t) check_for_valid_limits(&t->limits); - /* - * We only support barriers if there is exactly one underlying device. - */ - if (!list_is_singular(&t->devices)) - t->barriers_supported = 0; - /* how many indexes will the btree have ? */ leaf_nodes = dm_div_up(t->num_targets, KEYS_PER_NODE); t->depth = 1 + int_log(leaf_nodes, CHILDREN_PER_NODE); @@ -879,6 +866,45 @@ struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector) return &t->targets[(KEYS_PER_NODE * n) + k]; } +/* + * Set the integrity profile for this device if all devices used have + * matching profiles. + */ +static void dm_table_set_integrity(struct dm_table *t) +{ + struct list_head *devices = dm_table_get_devices(t); + struct dm_dev_internal *prev = NULL, *dd = NULL; + + if (!blk_get_integrity(dm_disk(t->md))) + return; + + list_for_each_entry(dd, devices, list) { + if (prev && + blk_integrity_compare(prev->dm_dev.bdev->bd_disk, + dd->dm_dev.bdev->bd_disk) < 0) { + DMWARN("%s: integrity not set: %s and %s mismatch", + dm_device_name(t->md), + prev->dm_dev.bdev->bd_disk->disk_name, + dd->dm_dev.bdev->bd_disk->disk_name); + goto no_integrity; + } + prev = dd; + } + + if (!prev || !bdev_get_integrity(prev->dm_dev.bdev)) + goto no_integrity; + + blk_integrity_register(dm_disk(t->md), + bdev_get_integrity(prev->dm_dev.bdev)); + + return; + +no_integrity: + blk_integrity_register(dm_disk(t->md), NULL); + + return; +} + void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q) { /* @@ -899,6 +925,7 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q) else queue_flag_set_unlocked(QUEUE_FLAG_CLUSTER, q); + dm_table_set_integrity(t); } unsigned int dm_table_get_num_targets(struct dm_table *t) @@ -1019,12 +1046,6 @@ struct mapped_device *dm_table_get_md(struct dm_table *t) return t->md; } -int dm_table_barrier_ok(struct dm_table *t) -{ - return t->barriers_supported; -} -EXPORT_SYMBOL(dm_table_barrier_ok); - EXPORT_SYMBOL(dm_vcalloc); EXPORT_SYMBOL(dm_get_device); EXPORT_SYMBOL(dm_put_device); diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 788ba96a625..8a994be035b 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -89,12 +89,13 @@ union map_info *dm_get_mapinfo(struct bio *bio) /* * Bits for the md->flags field. */ -#define DMF_BLOCK_IO 0 +#define DMF_BLOCK_IO_FOR_SUSPEND 0 #define DMF_SUSPENDED 1 #define DMF_FROZEN 2 #define DMF_FREEING 3 #define DMF_DELETING 4 #define DMF_NOFLUSH_SUSPENDING 5 +#define DMF_QUEUE_IO_TO_THREAD 6 /* * Work processed by per-device workqueue. @@ -124,6 +125,11 @@ struct mapped_device { spinlock_t deferred_lock; /* + * An error from the barrier request currently being processed. + */ + int barrier_error; + + /* * Processing queue (flush/barriers) */ struct workqueue_struct *wq; @@ -424,6 +430,10 @@ static void end_io_acct(struct dm_io *io) part_stat_add(cpu, &dm_disk(md)->part0, ticks[rw], duration); part_stat_unlock(); + /* + * After this is decremented the bio must not be touched if it is + * a barrier. + */ dm_disk(md)->part0.in_flight = pending = atomic_dec_return(&md->pending); @@ -435,21 +445,18 @@ static void end_io_acct(struct dm_io *io) /* * Add the bio to the list of deferred io. */ -static int queue_io(struct mapped_device *md, struct bio *bio) +static void queue_io(struct mapped_device *md, struct bio *bio) { down_write(&md->io_lock); - if (!test_bit(DMF_BLOCK_IO, &md->flags)) { - up_write(&md->io_lock); - return 1; - } - spin_lock_irq(&md->deferred_lock); bio_list_add(&md->deferred, bio); spin_unlock_irq(&md->deferred_lock); + if (!test_and_set_bit(DMF_QUEUE_IO_TO_THREAD, &md->flags)) + queue_work(md->wq, &md->work); + up_write(&md->io_lock); - return 0; /* deferred successfully */ } /* @@ -533,25 +540,35 @@ static void dec_pending(struct dm_io *io, int error) */ spin_lock_irqsave(&md->deferred_lock, flags); if (__noflush_suspending(md)) - bio_list_add(&md->deferred, io->bio); + bio_list_add_head(&md->deferred, io->bio); else /* noflush suspend was interrupted. */ io->error = -EIO; spin_unlock_irqrestore(&md->deferred_lock, flags); } - end_io_acct(io); - io_error = io->error; bio = io->bio; - free_io(md, io); + if (bio_barrier(bio)) { + /* + * There can be just one barrier request so we use + * a per-device variable for error reporting. + * Note that you can't touch the bio after end_io_acct + */ + md->barrier_error = io_error; + end_io_acct(io); + } else { + end_io_acct(io); - if (io_error != DM_ENDIO_REQUEUE) { - trace_block_bio_complete(md->queue, bio); + if (io_error != DM_ENDIO_REQUEUE) { + trace_block_bio_complete(md->queue, bio); - bio_endio(bio, io_error); + bio_endio(bio, io_error); + } } + + free_io(md, io); } } @@ -693,13 +710,19 @@ static struct bio *split_bvec(struct bio *bio, sector_t sector, clone->bi_sector = sector; clone->bi_bdev = bio->bi_bdev; - clone->bi_rw = bio->bi_rw; + clone->bi_rw = bio->bi_rw & ~(1 << BIO_RW_BARRIER); clone->bi_vcnt = 1; clone->bi_size = to_bytes(len); clone->bi_io_vec->bv_offset = offset; clone->bi_io_vec->bv_len = clone->bi_size; clone->bi_flags |= 1 << BIO_CLONED; + if (bio_integrity(bio)) { + bio_integrity_clone(clone, bio, GFP_NOIO); + bio_integrity_trim(clone, + bio_sector_offset(bio, idx, offset), len); + } + return clone; } @@ -714,6 +737,7 @@ static struct bio *clone_bio(struct bio *bio, sector_t sector, clone = bio_alloc_bioset(GFP_NOIO, bio->bi_max_vecs, bs); __bio_clone(clone, bio); + clone->bi_rw &= ~(1 << BIO_RW_BARRIER); clone->bi_destructor = dm_bio_destructor; clone->bi_sector = sector; clone->bi_idx = idx; @@ -721,6 +745,14 @@ static struct bio *clone_bio(struct bio *bio, sector_t sector, clone->bi_size = to_bytes(len); clone->bi_flags &= ~(1 << BIO_SEG_VALID); + if (bio_integrity(bio)) { + bio_integrity_clone(clone, bio, GFP_NOIO); + + if (idx != bio->bi_idx || clone->bi_size < bio->bi_size) + bio_integrity_trim(clone, + bio_sector_offset(bio, idx, 0), len); + } + return clone; } @@ -834,14 +866,13 @@ static void __split_and_process_bio(struct mapped_device *md, struct bio *bio) ci.map = dm_get_table(md); if (unlikely(!ci.map)) { - bio_io_error(bio); - return; - } - if (unlikely(bio_barrier(bio) && !dm_table_barrier_ok(ci.map))) { - dm_table_put(ci.map); - bio_endio(bio, -EOPNOTSUPP); + if (!bio_barrier(bio)) + bio_io_error(bio); + else + md->barrier_error = -EIO; return; } + ci.md = md; ci.bio = bio; ci.io = alloc_io(md); @@ -918,7 +949,6 @@ out: */ static int dm_request(struct request_queue *q, struct bio *bio) { - int r = -EIO; int rw = bio_data_dir(bio); struct mapped_device *md = q->queuedata; int cpu; @@ -931,34 +961,27 @@ static int dm_request(struct request_queue *q, struct bio *bio) part_stat_unlock(); /* - * If we're suspended we have to queue - * this io for later. + * If we're suspended or the thread is processing barriers + * we have to queue this io for later. */ - while (test_bit(DMF_BLOCK_IO, &md->flags)) { + if (unlikely(test_bit(DMF_QUEUE_IO_TO_THREAD, &md->flags)) || + unlikely(bio_barrier(bio))) { up_read(&md->io_lock); - if (bio_rw(bio) != READA) - r = queue_io(md, bio); + if (unlikely(test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) && + bio_rw(bio) == READA) { + bio_io_error(bio); + return 0; + } - if (r <= 0) - goto out_req; + queue_io(md, bio); - /* - * We're in a while loop, because someone could suspend - * before we get to the following read lock. - */ - down_read(&md->io_lock); + return 0; } __split_and_process_bio(md, bio); up_read(&md->io_lock); return 0; - -out_req: - if (r < 0) - bio_io_error(bio); - - return 0; } static void dm_unplug_all(struct request_queue *q) @@ -978,7 +1001,7 @@ static int dm_any_congested(void *congested_data, int bdi_bits) struct mapped_device *md = congested_data; struct dm_table *map; - if (!test_bit(DMF_BLOCK_IO, &md->flags)) { + if (!test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) { map = dm_get_table(md); if (map) { r = dm_table_any_congested(map, bdi_bits); @@ -1193,6 +1216,7 @@ static void free_dev(struct mapped_device *md) mempool_destroy(md->tio_pool); mempool_destroy(md->io_pool); bioset_free(md->bs); + blk_integrity_unregister(md->disk); del_gendisk(md->disk); free_minor(minor); @@ -1406,6 +1430,36 @@ static int dm_wait_for_completion(struct mapped_device *md, int interruptible) return r; } +static int dm_flush(struct mapped_device *md) +{ + dm_wait_for_completion(md, TASK_UNINTERRUPTIBLE); + return 0; +} + +static void process_barrier(struct mapped_device *md, struct bio *bio) +{ + int error = dm_flush(md); + + if (unlikely(error)) { + bio_endio(bio, error); + return; + } + if (bio_empty_barrier(bio)) { + bio_endio(bio, 0); + return; + } + + __split_and_process_bio(md, bio); + + error = dm_flush(md); + + if (!error && md->barrier_error) + error = md->barrier_error; + + if (md->barrier_error != DM_ENDIO_REQUEUE) + bio_endio(bio, error); +} + /* * Process the deferred bios */ @@ -1417,25 +1471,34 @@ static void dm_wq_work(struct work_struct *work) down_write(&md->io_lock); -next_bio: - spin_lock_irq(&md->deferred_lock); - c = bio_list_pop(&md->deferred); - spin_unlock_irq(&md->deferred_lock); + while (!test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) { + spin_lock_irq(&md->deferred_lock); + c = bio_list_pop(&md->deferred); + spin_unlock_irq(&md->deferred_lock); - if (c) { - __split_and_process_bio(md, c); - goto next_bio; - } + if (!c) { + clear_bit(DMF_QUEUE_IO_TO_THREAD, &md->flags); + break; + } - clear_bit(DMF_BLOCK_IO, &md->flags); + up_write(&md->io_lock); + + if (bio_barrier(c)) + process_barrier(md, c); + else + __split_and_process_bio(md, c); + + down_write(&md->io_lock); + } up_write(&md->io_lock); } static void dm_queue_flush(struct mapped_device *md) { + clear_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags); + smp_mb__after_clear_bit(); queue_work(md->wq, &md->work); - flush_workqueue(md->wq); } /* @@ -1553,20 +1616,36 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) } /* - * First we set the BLOCK_IO flag so no more ios will be mapped. + * Here we must make sure that no processes are submitting requests + * to target drivers i.e. no one may be executing + * __split_and_process_bio. This is called from dm_request and + * dm_wq_work. + * + * To get all processes out of __split_and_process_bio in dm_request, + * we take the write lock. To prevent any process from reentering + * __split_and_process_bio from dm_request, we set + * DMF_QUEUE_IO_TO_THREAD. + * + * To quiesce the thread (dm_wq_work), we set DMF_BLOCK_IO_FOR_SUSPEND + * and call flush_workqueue(md->wq). flush_workqueue will wait until + * dm_wq_work exits and DMF_BLOCK_IO_FOR_SUSPEND will prevent any + * further calls to __split_and_process_bio from dm_wq_work. */ down_write(&md->io_lock); - set_bit(DMF_BLOCK_IO, &md->flags); - + set_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags); + set_bit(DMF_QUEUE_IO_TO_THREAD, &md->flags); up_write(&md->io_lock); + flush_workqueue(md->wq); + /* - * Wait for the already-mapped ios to complete. + * At this point no more requests are entering target request routines. + * We call dm_wait_for_completion to wait for all existing requests + * to finish. */ r = dm_wait_for_completion(md, TASK_INTERRUPTIBLE); down_write(&md->io_lock); - if (noflush) clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); up_write(&md->io_lock); @@ -1579,6 +1658,12 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) goto out; /* pushback list is already flushed, so skip flush */ } + /* + * If dm_wait_for_completion returned 0, the device is completely + * quiescent now. There is no request-processing activity. All new + * requests are being added to md->deferred list. + */ + dm_table_postsuspend_targets(map); set_bit(DMF_SUSPENDED, &md->flags); diff --git a/drivers/md/dm.h b/drivers/md/dm.h index b48397c0abb..a31506d93e9 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -52,7 +52,6 @@ int dm_table_any_congested(struct dm_table *t, int bdi_bits); * To check the return value from dm_table_find_target(). */ #define dm_target_is_valid(t) ((t)->table) -int dm_table_barrier_ok(struct dm_table *t); /*----------------------------------------------------------------- * A registry of target types. diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index c232d11a7ed..06084dbf127 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -208,7 +208,7 @@ static int mmc_read_ext_csd(struct mmc_card *card) } ext_csd_struct = ext_csd[EXT_CSD_REV]; - if (ext_csd_struct > 2) { + if (ext_csd_struct > 3) { printk(KERN_ERR "%s: unrecognised EXT_CSD structure " "version %d\n", mmc_hostname(card->host), ext_csd_struct); diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c index 26fc098d77c..cd81c395e16 100644 --- a/drivers/mmc/core/sd.c +++ b/drivers/mmc/core/sd.c @@ -363,15 +363,6 @@ static int mmc_sd_init_card(struct mmc_host *host, u32 ocr, goto err; /* - * For SPI, enable CRC as appropriate. - */ - if (mmc_host_is_spi(host)) { - err = mmc_spi_set_crc(host, use_spi_crc); - if (err) - goto err; - } - - /* * Fetch CID from card. */ if (mmc_host_is_spi(host)) @@ -458,6 +449,18 @@ static int mmc_sd_init_card(struct mmc_host *host, u32 ocr, } /* + * For SPI, enable CRC as appropriate. + * This CRC enable is located AFTER the reading of the + * card registers because some SDHC cards are not able + * to provide valid CRCs for non-512-byte blocks. + */ + if (mmc_host_is_spi(host)) { + err = mmc_spi_set_crc(host, use_spi_crc); + if (err) + goto free_card; + } + + /* * Attempt to change to high-speed (if supported) */ err = mmc_switch_hs(card); diff --git a/drivers/mmc/host/imxmmc.c b/drivers/mmc/host/imxmmc.c index eb29b1d933a..e0be21a4a69 100644 --- a/drivers/mmc/host/imxmmc.c +++ b/drivers/mmc/host/imxmmc.c @@ -307,13 +307,6 @@ static void imxmci_setup_data(struct imxmci_host *host, struct mmc_data *data) wmb(); - if (host->actual_bus_width == MMC_BUS_WIDTH_4) - BLR(host->dma) = 0; /* burst 64 byte read / 64 bytes write */ - else - BLR(host->dma) = 16; /* burst 16 byte read / 16 bytes write */ - - RSSR(host->dma) = DMA_REQ_SDHC; - set_bit(IMXMCI_PEND_DMA_DATA_b, &host->pending_events); clear_bit(IMXMCI_PEND_CPU_DATA_b, &host->pending_events); @@ -818,9 +811,11 @@ static void imxmci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios) if (ios->bus_width == MMC_BUS_WIDTH_4) { host->actual_bus_width = MMC_BUS_WIDTH_4; imx_gpio_mode(PB11_PF_SD_DAT3); + BLR(host->dma) = 0; /* burst 64 byte read/write */ } else { host->actual_bus_width = MMC_BUS_WIDTH_1; imx_gpio_mode(GPIO_PORTB | GPIO_IN | GPIO_PUEN | 11); + BLR(host->dma) = 16; /* burst 16 byte read/write */ } if (host->power_mode != ios->power_mode) { @@ -938,7 +933,7 @@ static void imxmci_check_status(unsigned long data) mod_timer(&host->timer, jiffies + (HZ>>1)); } -static int imxmci_probe(struct platform_device *pdev) +static int __init imxmci_probe(struct platform_device *pdev) { struct mmc_host *mmc; struct imxmci_host *host = NULL; @@ -1034,6 +1029,7 @@ static int imxmci_probe(struct platform_device *pdev) } host->dma_allocated = 1; imx_dma_setup_handlers(host->dma, imxmci_dma_irq, NULL, host); + RSSR(host->dma) = DMA_REQ_SDHC; tasklet_init(&host->tasklet, imxmci_tasklet_fnc, (unsigned long)host); host->status_reg=0; @@ -1079,7 +1075,7 @@ out: return ret; } -static int imxmci_remove(struct platform_device *pdev) +static int __exit imxmci_remove(struct platform_device *pdev) { struct mmc_host *mmc = platform_get_drvdata(pdev); @@ -1145,8 +1141,7 @@ static int imxmci_resume(struct platform_device *dev) #endif /* CONFIG_PM */ static struct platform_driver imxmci_driver = { - .probe = imxmci_probe, - .remove = imxmci_remove, + .remove = __exit_p(imxmci_remove), .suspend = imxmci_suspend, .resume = imxmci_resume, .driver = { @@ -1157,7 +1152,7 @@ static struct platform_driver imxmci_driver = { static int __init imxmci_init(void) { - return platform_driver_register(&imxmci_driver); + return platform_driver_probe(&imxmci_driver, imxmci_probe); } static void __exit imxmci_exit(void) diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c index 72f8bde4877..f48349d18c9 100644 --- a/drivers/mmc/host/mmc_spi.c +++ b/drivers/mmc/host/mmc_spi.c @@ -24,7 +24,7 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include <linux/hrtimer.h> +#include <linux/sched.h> #include <linux/delay.h> #include <linux/bio.h> #include <linux/dma-mapping.h> @@ -95,7 +95,7 @@ * reads which takes nowhere near that long. Older cards may be able to use * shorter timeouts ... but why bother? */ -#define r1b_timeout ktime_set(3, 0) +#define r1b_timeout (HZ * 3) /****************************************************************************/ @@ -183,12 +183,11 @@ mmc_spi_readbytes(struct mmc_spi_host *host, unsigned len) return status; } -static int -mmc_spi_skip(struct mmc_spi_host *host, ktime_t timeout, unsigned n, u8 byte) +static int mmc_spi_skip(struct mmc_spi_host *host, unsigned long timeout, + unsigned n, u8 byte) { u8 *cp = host->data->status; - - timeout = ktime_add(timeout, ktime_get()); + unsigned long start = jiffies; while (1) { int status; @@ -203,22 +202,26 @@ mmc_spi_skip(struct mmc_spi_host *host, ktime_t timeout, unsigned n, u8 byte) return cp[i]; } - /* REVISIT investigate msleep() to avoid busy-wait I/O - * in at least some cases. - */ - if (ktime_to_ns(ktime_sub(ktime_get(), timeout)) > 0) + if (time_is_before_jiffies(start + timeout)) break; + + /* If we need long timeouts, we may release the CPU. + * We use jiffies here because we want to have a relation + * between elapsed time and the blocking of the scheduler. + */ + if (time_is_before_jiffies(start+1)) + schedule(); } return -ETIMEDOUT; } static inline int -mmc_spi_wait_unbusy(struct mmc_spi_host *host, ktime_t timeout) +mmc_spi_wait_unbusy(struct mmc_spi_host *host, unsigned long timeout) { return mmc_spi_skip(host, timeout, sizeof(host->data->status), 0); } -static int mmc_spi_readtoken(struct mmc_spi_host *host, ktime_t timeout) +static int mmc_spi_readtoken(struct mmc_spi_host *host, unsigned long timeout) { return mmc_spi_skip(host, timeout, 1, 0xff); } @@ -251,6 +254,10 @@ static int mmc_spi_response_get(struct mmc_spi_host *host, u8 *cp = host->data->status; u8 *end = cp + host->t.len; int value = 0; + int bitshift; + u8 leftover = 0; + unsigned short rotator; + int i; char tag[32]; snprintf(tag, sizeof(tag), " ... CMD%d response SPI_%s", @@ -268,9 +275,8 @@ static int mmc_spi_response_get(struct mmc_spi_host *host, /* Data block reads (R1 response types) may need more data... */ if (cp == end) { - unsigned i; - cp = host->data->status; + end = cp+1; /* Card sends N(CR) (== 1..8) bytes of all-ones then one * status byte ... and we already scanned 2 bytes. @@ -295,20 +301,34 @@ static int mmc_spi_response_get(struct mmc_spi_host *host, } checkstatus: - if (*cp & 0x80) { - dev_dbg(&host->spi->dev, "%s: INVALID RESPONSE, %02x\n", - tag, *cp); - value = -EBADR; - goto done; + bitshift = 0; + if (*cp & 0x80) { + /* Houston, we have an ugly card with a bit-shifted response */ + rotator = *cp++ << 8; + /* read the next byte */ + if (cp == end) { + value = mmc_spi_readbytes(host, 1); + if (value < 0) + goto done; + cp = host->data->status; + end = cp+1; + } + rotator |= *cp++; + while (rotator & 0x8000) { + bitshift++; + rotator <<= 1; + } + cmd->resp[0] = rotator >> 8; + leftover = rotator; + } else { + cmd->resp[0] = *cp++; } - - cmd->resp[0] = *cp++; cmd->error = 0; /* Status byte: the entire seven-bit R1 response. */ if (cmd->resp[0] != 0) { if ((R1_SPI_PARAMETER | R1_SPI_ADDRESS - | R1_SPI_ILLEGAL_COMMAND) + | R1_SPI_ILLEGAL_COMMAND) & cmd->resp[0]) value = -EINVAL; else if (R1_SPI_COM_CRC & cmd->resp[0]) @@ -336,12 +356,45 @@ checkstatus: * SPI R5 == R1 + data byte; IO_RW_DIRECT */ case MMC_RSP_SPI_R2: - cmd->resp[0] |= *cp << 8; + /* read the next byte */ + if (cp == end) { + value = mmc_spi_readbytes(host, 1); + if (value < 0) + goto done; + cp = host->data->status; + end = cp+1; + } + if (bitshift) { + rotator = leftover << 8; + rotator |= *cp << bitshift; + cmd->resp[0] |= (rotator & 0xFF00); + } else { + cmd->resp[0] |= *cp << 8; + } break; /* SPI R3, R4, or R7 == R1 + 4 bytes */ case MMC_RSP_SPI_R3: - cmd->resp[1] = get_unaligned_be32(cp); + rotator = leftover << 8; + cmd->resp[1] = 0; + for (i = 0; i < 4; i++) { + cmd->resp[1] <<= 8; + /* read the next byte */ + if (cp == end) { + value = mmc_spi_readbytes(host, 1); + if (value < 0) + goto done; + cp = host->data->status; + end = cp+1; + } + if (bitshift) { + rotator |= *cp++ << bitshift; + cmd->resp[1] |= (rotator >> 8); + rotator <<= 8; + } else { + cmd->resp[1] |= *cp++; + } + } break; /* SPI R1 == just one status byte */ @@ -607,7 +660,7 @@ mmc_spi_setup_data_message( */ static int mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t, - ktime_t timeout) + unsigned long timeout) { struct spi_device *spi = host->spi; int status, i; @@ -717,11 +770,13 @@ mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t, */ static int mmc_spi_readblock(struct mmc_spi_host *host, struct spi_transfer *t, - ktime_t timeout) + unsigned long timeout) { struct spi_device *spi = host->spi; int status; struct scratch *scratch = host->data; + unsigned int bitshift; + u8 leftover; /* At least one SD card sends an all-zeroes byte when N(CX) * applies, before the all-ones bytes ... just cope with that. @@ -733,38 +788,60 @@ mmc_spi_readblock(struct mmc_spi_host *host, struct spi_transfer *t, if (status == 0xff || status == 0) status = mmc_spi_readtoken(host, timeout); - if (status == SPI_TOKEN_SINGLE) { - if (host->dma_dev) { - dma_sync_single_for_device(host->dma_dev, - host->data_dma, sizeof(*scratch), - DMA_BIDIRECTIONAL); - dma_sync_single_for_device(host->dma_dev, - t->rx_dma, t->len, - DMA_FROM_DEVICE); - } + if (status < 0) { + dev_dbg(&spi->dev, "read error %02x (%d)\n", status, status); + return status; + } - status = spi_sync(spi, &host->m); + /* The token may be bit-shifted... + * the first 0-bit precedes the data stream. + */ + bitshift = 7; + while (status & 0x80) { + status <<= 1; + bitshift--; + } + leftover = status << 1; - if (host->dma_dev) { - dma_sync_single_for_cpu(host->dma_dev, - host->data_dma, sizeof(*scratch), - DMA_BIDIRECTIONAL); - dma_sync_single_for_cpu(host->dma_dev, - t->rx_dma, t->len, - DMA_FROM_DEVICE); - } + if (host->dma_dev) { + dma_sync_single_for_device(host->dma_dev, + host->data_dma, sizeof(*scratch), + DMA_BIDIRECTIONAL); + dma_sync_single_for_device(host->dma_dev, + t->rx_dma, t->len, + DMA_FROM_DEVICE); + } - } else { - dev_dbg(&spi->dev, "read error %02x (%d)\n", status, status); + status = spi_sync(spi, &host->m); - /* we've read extra garbage, timed out, etc */ - if (status < 0) - return status; + if (host->dma_dev) { + dma_sync_single_for_cpu(host->dma_dev, + host->data_dma, sizeof(*scratch), + DMA_BIDIRECTIONAL); + dma_sync_single_for_cpu(host->dma_dev, + t->rx_dma, t->len, + DMA_FROM_DEVICE); + } - /* low four bits are an R2 subset, fifth seems to be - * vendor specific ... map them all to generic error.. + if (bitshift) { + /* Walk through the data and the crc and do + * all the magic to get byte-aligned data. */ - return -EIO; + u8 *cp = t->rx_buf; + unsigned int len; + unsigned int bitright = 8 - bitshift; + u8 temp; + for (len = t->len; len; len--) { + temp = *cp; + *cp++ = leftover | (temp >> bitshift); + leftover = temp << bitright; + } + cp = (u8 *) &scratch->crc_val; + temp = *cp; + *cp++ = leftover | (temp >> bitshift); + leftover = temp << bitright; + temp = *cp; + *cp = leftover | (temp >> bitshift); } if (host->mmc->use_spi_crc) { @@ -803,7 +880,7 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd, unsigned n_sg; int multiple = (data->blocks > 1); u32 clock_rate; - ktime_t timeout; + unsigned long timeout; if (data->flags & MMC_DATA_READ) direction = DMA_FROM_DEVICE; @@ -817,8 +894,9 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd, else clock_rate = spi->max_speed_hz; - timeout = ktime_add_ns(ktime_set(0, 0), data->timeout_ns + - data->timeout_clks * 1000000 / clock_rate); + timeout = data->timeout_ns + + data->timeout_clks * 1000000 / clock_rate; + timeout = usecs_to_jiffies((unsigned int)(timeout / 1000)) + 1; /* Handle scatterlist segments one at a time, with synch for * each 512-byte block diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index d183be6f2a5..e62a22a7f00 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -298,7 +298,6 @@ mmc_omap_xfer_done(struct mmc_omap_host *host, struct mmc_data *data) struct mmc_request *mrq = host->mrq; host->mrq = NULL; - mmc_omap_fclk_lazy_disable(host); mmc_request_done(host->mmc, mrq); return; } @@ -434,6 +433,8 @@ static irqreturn_t mmc_omap_irq(int irq, void *dev_id) if (host->mrq == NULL) { OMAP_HSMMC_WRITE(host->base, STAT, OMAP_HSMMC_READ(host->base, STAT)); + /* Flush posted write */ + OMAP_HSMMC_READ(host->base, STAT); return IRQ_HANDLED; } @@ -489,8 +490,10 @@ static irqreturn_t mmc_omap_irq(int irq, void *dev_id) } OMAP_HSMMC_WRITE(host->base, STAT, status); + /* Flush posted write */ + OMAP_HSMMC_READ(host->base, STAT); - if (end_cmd || (status & CC)) + if (end_cmd || ((status & CC) && host->cmd)) mmc_omap_cmd_done(host, host->cmd); if (end_trans || (status & TC)) mmc_omap_xfer_done(host, data); diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c index c5b316e2237..cd37962ec44 100644 --- a/drivers/mmc/host/sdhci-pci.c +++ b/drivers/mmc/host/sdhci-pci.c @@ -729,6 +729,6 @@ static void __exit sdhci_drv_exit(void) module_init(sdhci_drv_init); module_exit(sdhci_drv_exit); -MODULE_AUTHOR("Pierre Ossman <drzeus@drzeus.cx>"); +MODULE_AUTHOR("Pierre Ossman <pierre@ossman.eu>"); MODULE_DESCRIPTION("Secure Digital Host Controller Interface PCI driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 30d8e3d4e6f..9234be2226e 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -1935,7 +1935,7 @@ module_exit(sdhci_drv_exit); module_param(debug_quirks, uint, 0444); -MODULE_AUTHOR("Pierre Ossman <drzeus@drzeus.cx>"); +MODULE_AUTHOR("Pierre Ossman <pierre@ossman.eu>"); MODULE_DESCRIPTION("Secure Digital Host Controller Interface core driver"); MODULE_LICENSE("GPL"); diff --git a/drivers/mmc/host/wbsd.c b/drivers/mmc/host/wbsd.c index adda3795203..89bf8cd25ca 100644 --- a/drivers/mmc/host/wbsd.c +++ b/drivers/mmc/host/wbsd.c @@ -2036,7 +2036,7 @@ module_param_named(irq, param_irq, uint, 0444); module_param_named(dma, param_dma, int, 0444); MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Pierre Ossman <drzeus@drzeus.cx>"); +MODULE_AUTHOR("Pierre Ossman <pierre@ossman.eu>"); MODULE_DESCRIPTION("Winbond W83L51xD SD/MMC card interface driver"); #ifdef CONFIG_PNP diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 9e7baec4572..9e921544ba2 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -977,6 +977,8 @@ config ETHOC depends on NET_ETHERNET && HAS_IOMEM select MII select PHYLIB + select CRC32 + select BITREVERSE help Say Y here if you want to use the OpenCores 10/100 Mbps Ethernet MAC. @@ -2056,6 +2058,27 @@ config IGB_DCA driver. DCA is a method for warming the CPU cache before data is used, with the intent of lessening the impact of cache misses. +config IGBVF + tristate "Intel(R) 82576 Virtual Function Ethernet support" + depends on PCI + ---help--- + This driver supports Intel(R) 82576 virtual functions. For more + information on how to identify your adapter, go to the Adapter & + Driver ID Guide at: + + <http://support.intel.com/support/network/adapter/pro100/21397.htm> + + For general information and support, go to the Intel support + website at: + + <http://support.intel.com> + + More specific information on configuring the driver is in + <file:Documentation/networking/e1000.txt>. + + To compile this driver as a module, choose M here. The module + will be called igbvf. + source "drivers/net/ixp2000/Kconfig" config MYRI_SBUS diff --git a/drivers/net/Makefile b/drivers/net/Makefile index edc9a0d6171..1fc4602a6ff 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -6,6 +6,7 @@ obj-$(CONFIG_E1000) += e1000/ obj-$(CONFIG_E1000E) += e1000e/ obj-$(CONFIG_IBM_NEW_EMAC) += ibm_newemac/ obj-$(CONFIG_IGB) += igb/ +obj-$(CONFIG_IGBVF) += igbvf/ obj-$(CONFIG_IXGBE) += ixgbe/ obj-$(CONFIG_IXGB) += ixgb/ obj-$(CONFIG_IP1000) += ipg.o diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 9d268be0b67..d47839184a0 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -3427,8 +3427,8 @@ static int __devinit bnx2_request_firmware(struct bnx2 *bp) { const char *mips_fw_file, *rv2p_fw_file; - const struct bnx2_mips_fw_file *mips; - const struct bnx2_rv2p_fw_file *rv2p; + const struct bnx2_mips_fw_file *mips_fw; + const struct bnx2_rv2p_fw_file *rv2p_fw; int rc; if (CHIP_NUM(bp) == CHIP_NUM_5709) { @@ -3452,21 +3452,21 @@ bnx2_request_firmware(struct bnx2 *bp) rv2p_fw_file); return rc; } - mips = (const struct bnx2_mips_fw_file *) bp->mips_firmware->data; - rv2p = (const struct bnx2_rv2p_fw_file *) bp->rv2p_firmware->data; - if (bp->mips_firmware->size < sizeof(*mips) || - check_mips_fw_entry(bp->mips_firmware, &mips->com) || - check_mips_fw_entry(bp->mips_firmware, &mips->cp) || - check_mips_fw_entry(bp->mips_firmware, &mips->rxp) || - check_mips_fw_entry(bp->mips_firmware, &mips->tpat) || - check_mips_fw_entry(bp->mips_firmware, &mips->txp)) { + mips_fw = (const struct bnx2_mips_fw_file *) bp->mips_firmware->data; + rv2p_fw = (const struct bnx2_rv2p_fw_file *) bp->rv2p_firmware->data; + if (bp->mips_firmware->size < sizeof(*mips_fw) || + check_mips_fw_entry(bp->mips_firmware, &mips_fw->com) || + check_mips_fw_entry(bp->mips_firmware, &mips_fw->cp) || + check_mips_fw_entry(bp->mips_firmware, &mips_fw->rxp) || + check_mips_fw_entry(bp->mips_firmware, &mips_fw->tpat) || + check_mips_fw_entry(bp->mips_firmware, &mips_fw->txp)) { printk(KERN_ERR PFX "Firmware file \"%s\" is invalid\n", mips_fw_file); return -EINVAL; } - if (bp->rv2p_firmware->size < sizeof(*rv2p) || - check_fw_section(bp->rv2p_firmware, &rv2p->proc1.rv2p, 8, true) || - check_fw_section(bp->rv2p_firmware, &rv2p->proc2.rv2p, 8, true)) { + if (bp->rv2p_firmware->size < sizeof(*rv2p_fw) || + check_fw_section(bp->rv2p_firmware, &rv2p_fw->proc1.rv2p, 8, true) || + check_fw_section(bp->rv2p_firmware, &rv2p_fw->proc2.rv2p, 8, true)) { printk(KERN_ERR PFX "Firmware file \"%s\" is invalid\n", rv2p_fw_file); return -EINVAL; diff --git a/drivers/net/eql.c b/drivers/net/eql.c index 51ead7941f8..5210bb1027c 100644 --- a/drivers/net/eql.c +++ b/drivers/net/eql.c @@ -542,6 +542,8 @@ static int eql_s_slave_cfg(struct net_device *dev, slave_config_t __user *scp) } spin_unlock_bh(&eql->queue.lock); + dev_put(slave_dev); + return ret; } diff --git a/drivers/net/fec.c b/drivers/net/fec.c index a515acccc61..682e7f0b558 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -1240,6 +1240,7 @@ static void __inline__ fec_phy_ack_intr(void) icrp = (volatile unsigned long *) (MCF_MBAR + MCFSIM_ICR1); *icrp = 0x0d000000; } +#endif #ifdef CONFIG_M5272 static void __inline__ fec_get_mac(struct net_device *dev) diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c index 6b0697c565b..db7274e6222 100644 --- a/drivers/net/igb/igb_main.c +++ b/drivers/net/igb/igb_main.c @@ -152,14 +152,13 @@ static struct notifier_block dca_notifier = { /* for netdump / net console */ static void igb_netpoll(struct net_device *); #endif - #ifdef CONFIG_PCI_IOV -static ssize_t igb_set_num_vfs(struct device *, struct device_attribute *, - const char *, size_t); -static ssize_t igb_show_num_vfs(struct device *, struct device_attribute *, - char *); -DEVICE_ATTR(num_vfs, S_IRUGO | S_IWUSR, igb_show_num_vfs, igb_set_num_vfs); -#endif +static unsigned int max_vfs = 0; +module_param(max_vfs, uint, 0); +MODULE_PARM_DESC(max_vfs, "Maximum number of virtual functions to allocate " + "per physical function"); +#endif /* CONFIG_PCI_IOV */ + static pci_ers_result_t igb_io_error_detected(struct pci_dev *, pci_channel_state_t); static pci_ers_result_t igb_io_slot_reset(struct pci_dev *); @@ -671,6 +670,21 @@ static void igb_set_interrupt_capability(struct igb_adapter *adapter) /* If we can't do MSI-X, try MSI */ msi_only: +#ifdef CONFIG_PCI_IOV + /* disable SR-IOV for non MSI-X configurations */ + if (adapter->vf_data) { + struct e1000_hw *hw = &adapter->hw; + /* disable iov and allow time for transactions to clear */ + pci_disable_sriov(adapter->pdev); + msleep(500); + + kfree(adapter->vf_data); + adapter->vf_data = NULL; + wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ); + msleep(100); + dev_info(&adapter->pdev->dev, "IOV Disabled\n"); + } +#endif adapter->num_rx_queues = 1; adapter->num_tx_queues = 1; if (!pci_enable_msi(adapter->pdev)) @@ -1238,6 +1252,39 @@ static int __devinit igb_probe(struct pci_dev *pdev, if (err) goto err_sw_init; +#ifdef CONFIG_PCI_IOV + /* since iov functionality isn't critical to base device function we + * can accept failure. If it fails we don't allow iov to be enabled */ + if (hw->mac.type == e1000_82576) { + /* 82576 supports a maximum of 7 VFs in addition to the PF */ + unsigned int num_vfs = (max_vfs > 7) ? 7 : max_vfs; + int i; + unsigned char mac_addr[ETH_ALEN]; + + if (num_vfs) + adapter->vf_data = kcalloc(num_vfs, + sizeof(struct vf_data_storage), + GFP_KERNEL); + if (!adapter->vf_data) { + dev_err(&pdev->dev, "Could not allocate VF private " + "data - IOV enable failed\n"); + } else { + err = pci_enable_sriov(pdev, num_vfs); + if (!err) { + adapter->vfs_allocated_count = num_vfs; + dev_info(&pdev->dev, "%d vfs allocated\n", num_vfs); + for (i = 0; i < adapter->vfs_allocated_count; i++) { + random_ether_addr(mac_addr); + igb_set_vf_mac(adapter, i, mac_addr); + } + } else { + kfree(adapter->vf_data); + adapter->vf_data = NULL; + } + } + } + +#endif /* setup the private structure */ err = igb_sw_init(adapter); if (err) @@ -1397,19 +1444,6 @@ static int __devinit igb_probe(struct pci_dev *pdev, if (err) goto err_register; -#ifdef CONFIG_PCI_IOV - /* since iov functionality isn't critical to base device function we - * can accept failure. If it fails we don't allow iov to be enabled */ - if (hw->mac.type == e1000_82576) { - err = pci_enable_sriov(pdev, 0); - if (!err) - err = device_create_file(&netdev->dev, - &dev_attr_num_vfs); - if (err) - dev_err(&pdev->dev, "Failed to initialize IOV\n"); - } - -#endif #ifdef CONFIG_IGB_DCA if (dca_add_requester(&pdev->dev) == 0) { adapter->flags |= IGB_FLAG_DCA_ENABLED; @@ -5422,89 +5456,4 @@ static void igb_vmm_control(struct igb_adapter *adapter) igb_vmdq_set_replication_pf(hw, true); } -#ifdef CONFIG_PCI_IOV -static ssize_t igb_show_num_vfs(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct igb_adapter *adapter = netdev_priv(to_net_dev(dev)); - - return sprintf(buf, "%d\n", adapter->vfs_allocated_count); -} - -static ssize_t igb_set_num_vfs(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct net_device *netdev = to_net_dev(dev); - struct igb_adapter *adapter = netdev_priv(netdev); - struct e1000_hw *hw = &adapter->hw; - struct pci_dev *pdev = adapter->pdev; - unsigned int num_vfs, i; - unsigned char mac_addr[ETH_ALEN]; - int err; - - sscanf(buf, "%u", &num_vfs); - - if (num_vfs > 7) - num_vfs = 7; - - /* value unchanged do nothing */ - if (num_vfs == adapter->vfs_allocated_count) - return count; - - if (netdev->flags & IFF_UP) - igb_close(netdev); - - igb_reset_interrupt_capability(adapter); - igb_free_queues(adapter); - adapter->tx_ring = NULL; - adapter->rx_ring = NULL; - adapter->vfs_allocated_count = 0; - - /* reclaim resources allocated to VFs since we are changing count */ - if (adapter->vf_data) { - /* disable iov and allow time for transactions to clear */ - pci_disable_sriov(pdev); - msleep(500); - - kfree(adapter->vf_data); - adapter->vf_data = NULL; - wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ); - msleep(100); - dev_info(&pdev->dev, "IOV Disabled\n"); - } - - if (num_vfs) { - adapter->vf_data = kcalloc(num_vfs, - sizeof(struct vf_data_storage), - GFP_KERNEL); - if (!adapter->vf_data) { - dev_err(&pdev->dev, "Could not allocate VF private " - "data - IOV enable failed\n"); - } else { - err = pci_enable_sriov(pdev, num_vfs); - if (!err) { - adapter->vfs_allocated_count = num_vfs; - dev_info(&pdev->dev, "%d vfs allocated\n", num_vfs); - for (i = 0; i < adapter->vfs_allocated_count; i++) { - random_ether_addr(mac_addr); - igb_set_vf_mac(adapter, i, mac_addr); - } - } else { - kfree(adapter->vf_data); - adapter->vf_data = NULL; - } - } - } - - igb_set_interrupt_capability(adapter); - igb_alloc_queues(adapter); - igb_reset(adapter); - - if (netdev->flags & IFF_UP) - igb_open(netdev); - - return count; -} -#endif /* CONFIG_PCI_IOV */ /* igb_main.c */ diff --git a/drivers/net/igbvf/Makefile b/drivers/net/igbvf/Makefile new file mode 100644 index 00000000000..c2f150d8f2d --- /dev/null +++ b/drivers/net/igbvf/Makefile @@ -0,0 +1,38 @@ +################################################################################ +# +# Intel(R) 82576 Virtual Function Linux driver +# Copyright(c) 2009 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 +# +################################################################################ + +# +# Makefile for the Intel(R) 82576 VF ethernet driver +# + +obj-$(CONFIG_IGBVF) += igbvf.o + +igbvf-objs := vf.o \ + mbx.o \ + ethtool.o \ + netdev.o + diff --git a/drivers/net/igbvf/defines.h b/drivers/net/igbvf/defines.h new file mode 100644 index 00000000000..88a47537518 --- /dev/null +++ b/drivers/net/igbvf/defines.h @@ -0,0 +1,125 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 1999 - 2009 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 _E1000_DEFINES_H_ +#define _E1000_DEFINES_H_ + +/* Number of Transmit and Receive Descriptors must be a multiple of 8 */ +#define REQ_TX_DESCRIPTOR_MULTIPLE 8 +#define REQ_RX_DESCRIPTOR_MULTIPLE 8 + +/* IVAR valid bit */ +#define E1000_IVAR_VALID 0x80 + +/* Receive Descriptor bit definitions */ +#define E1000_RXD_STAT_DD 0x01 /* Descriptor Done */ +#define E1000_RXD_STAT_EOP 0x02 /* End of Packet */ +#define E1000_RXD_STAT_IXSM 0x04 /* Ignore checksum */ +#define E1000_RXD_STAT_VP 0x08 /* IEEE VLAN Packet */ +#define E1000_RXD_STAT_UDPCS 0x10 /* UDP xsum calculated */ +#define E1000_RXD_STAT_TCPCS 0x20 /* TCP xsum calculated */ +#define E1000_RXD_STAT_IPCS 0x40 /* IP xsum calculated */ +#define E1000_RXD_ERR_SE 0x02 /* Symbol Error */ +#define E1000_RXD_SPC_VLAN_MASK 0x0FFF /* VLAN ID is in lower 12 bits */ + +#define E1000_RXDEXT_STATERR_CE 0x01000000 +#define E1000_RXDEXT_STATERR_SE 0x02000000 +#define E1000_RXDEXT_STATERR_SEQ 0x04000000 +#define E1000_RXDEXT_STATERR_CXE 0x10000000 +#define E1000_RXDEXT_STATERR_TCPE 0x20000000 +#define E1000_RXDEXT_STATERR_IPE 0x40000000 +#define E1000_RXDEXT_STATERR_RXE 0x80000000 + + +/* Same mask, but for extended and packet split descriptors */ +#define E1000_RXDEXT_ERR_FRAME_ERR_MASK ( \ + E1000_RXDEXT_STATERR_CE | \ + E1000_RXDEXT_STATERR_SE | \ + E1000_RXDEXT_STATERR_SEQ | \ + E1000_RXDEXT_STATERR_CXE | \ + E1000_RXDEXT_STATERR_RXE) + +/* Device Control */ +#define E1000_CTRL_RST 0x04000000 /* Global reset */ + +/* Device Status */ +#define E1000_STATUS_FD 0x00000001 /* Full duplex.0=half,1=full */ +#define E1000_STATUS_LU 0x00000002 /* Link up.0=no,1=link */ +#define E1000_STATUS_TXOFF 0x00000010 /* transmission paused */ +#define E1000_STATUS_SPEED_10 0x00000000 /* Speed 10Mb/s */ +#define E1000_STATUS_SPEED_100 0x00000040 /* Speed 100Mb/s */ +#define E1000_STATUS_SPEED_1000 0x00000080 /* Speed 1000Mb/s */ + +#define SPEED_10 10 +#define SPEED_100 100 +#define SPEED_1000 1000 +#define HALF_DUPLEX 1 +#define FULL_DUPLEX 2 + +/* Transmit Descriptor bit definitions */ +#define E1000_TXD_POPTS_IXSM 0x01 /* Insert IP checksum */ +#define E1000_TXD_POPTS_TXSM 0x02 /* Insert TCP/UDP checksum */ +#define E1000_TXD_CMD_DEXT 0x20000000 /* Descriptor extension (0 = legacy) */ +#define E1000_TXD_STAT_DD 0x00000001 /* Descriptor Done */ + +#define MAX_JUMBO_FRAME_SIZE 0x3F00 + +/* 802.1q VLAN Packet Size */ +#define VLAN_TAG_SIZE 4 /* 802.3ac tag (not DMA'd) */ + +/* Error Codes */ +#define E1000_SUCCESS 0 +#define E1000_ERR_CONFIG 3 +#define E1000_ERR_MAC_INIT 5 +#define E1000_ERR_MBX 15 + +#ifndef ETH_ADDR_LEN +#define ETH_ADDR_LEN 6 +#endif + +/* SRRCTL bit definitions */ +#define E1000_SRRCTL_BSIZEPKT_SHIFT 10 /* Shift _right_ */ +#define E1000_SRRCTL_BSIZEHDRSIZE_MASK 0x00000F00 +#define E1000_SRRCTL_BSIZEHDRSIZE_SHIFT 2 /* Shift _left_ */ +#define E1000_SRRCTL_DESCTYPE_ADV_ONEBUF 0x02000000 +#define E1000_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS 0x0A000000 +#define E1000_SRRCTL_DESCTYPE_MASK 0x0E000000 +#define E1000_SRRCTL_DROP_EN 0x80000000 + +#define E1000_SRRCTL_BSIZEPKT_MASK 0x0000007F +#define E1000_SRRCTL_BSIZEHDR_MASK 0x00003F00 + +/* Additional Descriptor Control definitions */ +#define E1000_TXDCTL_QUEUE_ENABLE 0x02000000 /* Enable specific Tx Queue */ +#define E1000_RXDCTL_QUEUE_ENABLE 0x02000000 /* Enable specific Rx Queue */ + +/* Direct Cache Access (DCA) definitions */ +#define E1000_DCA_TXCTRL_TX_WB_RO_EN (1 << 11) /* Tx Desc writeback RO bit */ + +#define E1000_VF_INIT_TIMEOUT 200 /* Number of retries to clear RSTI */ + +#endif /* _E1000_DEFINES_H_ */ diff --git a/drivers/net/igbvf/ethtool.c b/drivers/net/igbvf/ethtool.c new file mode 100644 index 00000000000..1dcaa690531 --- /dev/null +++ b/drivers/net/igbvf/ethtool.c @@ -0,0 +1,540 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 + +*******************************************************************************/ + +/* ethtool support for igbvf */ + +#include <linux/netdevice.h> +#include <linux/ethtool.h> +#include <linux/pci.h> +#include <linux/vmalloc.h> +#include <linux/delay.h> + +#include "igbvf.h" +#include <linux/if_vlan.h> + + +struct igbvf_stats { + char stat_string[ETH_GSTRING_LEN]; + int sizeof_stat; + int stat_offset; + int base_stat_offset; +}; + +#define IGBVF_STAT(current, base) \ + sizeof(((struct igbvf_adapter *)0)->current), \ + offsetof(struct igbvf_adapter, current), \ + offsetof(struct igbvf_adapter, base) + +static const struct igbvf_stats igbvf_gstrings_stats[] = { + { "rx_packets", IGBVF_STAT(stats.gprc, stats.base_gprc) }, + { "tx_packets", IGBVF_STAT(stats.gptc, stats.base_gptc) }, + { "rx_bytes", IGBVF_STAT(stats.gorc, stats.base_gorc) }, + { "tx_bytes", IGBVF_STAT(stats.gotc, stats.base_gotc) }, + { "multicast", IGBVF_STAT(stats.mprc, stats.base_mprc) }, + { "lbrx_bytes", IGBVF_STAT(stats.gorlbc, stats.base_gorlbc) }, + { "lbrx_packets", IGBVF_STAT(stats.gprlbc, stats.base_gprlbc) }, + { "tx_restart_queue", IGBVF_STAT(restart_queue, zero_base) }, + { "rx_long_byte_count", IGBVF_STAT(stats.gorc, stats.base_gorc) }, + { "rx_csum_offload_good", IGBVF_STAT(hw_csum_good, zero_base) }, + { "rx_csum_offload_errors", IGBVF_STAT(hw_csum_err, zero_base) }, + { "rx_header_split", IGBVF_STAT(rx_hdr_split, zero_base) }, + { "alloc_rx_buff_failed", IGBVF_STAT(alloc_rx_buff_failed, zero_base) }, +}; + +#define IGBVF_GLOBAL_STATS_LEN ARRAY_SIZE(igbvf_gstrings_stats) + +static const char igbvf_gstrings_test[][ETH_GSTRING_LEN] = { + "Link test (on/offline)" +}; + +#define IGBVF_TEST_LEN ARRAY_SIZE(igbvf_gstrings_test) + +static int igbvf_get_settings(struct net_device *netdev, + struct ethtool_cmd *ecmd) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + u32 status; + + ecmd->supported = SUPPORTED_1000baseT_Full; + + ecmd->advertising = ADVERTISED_1000baseT_Full; + + ecmd->port = -1; + ecmd->transceiver = XCVR_DUMMY1; + + status = er32(STATUS); + if (status & E1000_STATUS_LU) { + if (status & E1000_STATUS_SPEED_1000) + ecmd->speed = 1000; + else if (status & E1000_STATUS_SPEED_100) + ecmd->speed = 100; + else + ecmd->speed = 10; + + if (status & E1000_STATUS_FD) + ecmd->duplex = DUPLEX_FULL; + else + ecmd->duplex = DUPLEX_HALF; + } else { + ecmd->speed = -1; + ecmd->duplex = -1; + } + + ecmd->autoneg = AUTONEG_DISABLE; + + return 0; +} + +static u32 igbvf_get_link(struct net_device *netdev) +{ + return netif_carrier_ok(netdev); +} + +static int igbvf_set_settings(struct net_device *netdev, + struct ethtool_cmd *ecmd) +{ + return -EOPNOTSUPP; +} + +static void igbvf_get_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ + return; +} + +static int igbvf_set_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ + return -EOPNOTSUPP; +} + +static u32 igbvf_get_tx_csum(struct net_device *netdev) +{ + return ((netdev->features & NETIF_F_IP_CSUM) != 0); +} + +static int igbvf_set_tx_csum(struct net_device *netdev, u32 data) +{ + if (data) + netdev->features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM); + else + netdev->features &= ~(NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM); + return 0; +} + +static int igbvf_set_tso(struct net_device *netdev, u32 data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + int i; + struct net_device *v_netdev; + + if (data) { + netdev->features |= NETIF_F_TSO; + netdev->features |= NETIF_F_TSO6; + } else { + netdev->features &= ~NETIF_F_TSO; + netdev->features &= ~NETIF_F_TSO6; + /* disable TSO on all VLANs if they're present */ + if (!adapter->vlgrp) + goto tso_out; + for (i = 0; i < VLAN_GROUP_ARRAY_LEN; i++) { + v_netdev = vlan_group_get_device(adapter->vlgrp, i); + if (!v_netdev) + continue; + + v_netdev->features &= ~NETIF_F_TSO; + v_netdev->features &= ~NETIF_F_TSO6; + vlan_group_set_device(adapter->vlgrp, i, v_netdev); + } + } + +tso_out: + dev_info(&adapter->pdev->dev, "TSO is %s\n", + data ? "Enabled" : "Disabled"); + adapter->flags |= FLAG_TSO_FORCE; + return 0; +} + +static u32 igbvf_get_msglevel(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + return adapter->msg_enable; +} + +static void igbvf_set_msglevel(struct net_device *netdev, u32 data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + adapter->msg_enable = data; +} + +static int igbvf_get_regs_len(struct net_device *netdev) +{ +#define IGBVF_REGS_LEN 8 + return IGBVF_REGS_LEN * sizeof(u32); +} + +static void igbvf_get_regs(struct net_device *netdev, + struct ethtool_regs *regs, void *p) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + u32 *regs_buff = p; + u8 revision_id; + + memset(p, 0, IGBVF_REGS_LEN * sizeof(u32)); + + pci_read_config_byte(adapter->pdev, PCI_REVISION_ID, &revision_id); + + regs->version = (1 << 24) | (revision_id << 16) | adapter->pdev->device; + + regs_buff[0] = er32(CTRL); + regs_buff[1] = er32(STATUS); + + regs_buff[2] = er32(RDLEN(0)); + regs_buff[3] = er32(RDH(0)); + regs_buff[4] = er32(RDT(0)); + + regs_buff[5] = er32(TDLEN(0)); + regs_buff[6] = er32(TDH(0)); + regs_buff[7] = er32(TDT(0)); +} + +static int igbvf_get_eeprom_len(struct net_device *netdev) +{ + return 0; +} + +static int igbvf_get_eeprom(struct net_device *netdev, + struct ethtool_eeprom *eeprom, u8 *bytes) +{ + return -EOPNOTSUPP; +} + +static int igbvf_set_eeprom(struct net_device *netdev, + struct ethtool_eeprom *eeprom, u8 *bytes) +{ + return -EOPNOTSUPP; +} + +static void igbvf_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *drvinfo) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + char firmware_version[32] = "N/A"; + + strncpy(drvinfo->driver, igbvf_driver_name, 32); + strncpy(drvinfo->version, igbvf_driver_version, 32); + strncpy(drvinfo->fw_version, firmware_version, 32); + strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + drvinfo->regdump_len = igbvf_get_regs_len(netdev); + drvinfo->eedump_len = igbvf_get_eeprom_len(netdev); +} + +static void igbvf_get_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct igbvf_ring *rx_ring = adapter->rx_ring; + + ring->rx_max_pending = IGBVF_MAX_RXD; + ring->tx_max_pending = IGBVF_MAX_TXD; + ring->rx_mini_max_pending = 0; + ring->rx_jumbo_max_pending = 0; + ring->rx_pending = rx_ring->count; + ring->tx_pending = tx_ring->count; + ring->rx_mini_pending = 0; + ring->rx_jumbo_pending = 0; +} + +static int igbvf_set_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct igbvf_ring *temp_ring; + int err; + u32 new_rx_count, new_tx_count; + + if ((ring->rx_mini_pending) || (ring->rx_jumbo_pending)) + return -EINVAL; + + new_rx_count = max(ring->rx_pending, (u32)IGBVF_MIN_RXD); + new_rx_count = min(new_rx_count, (u32)IGBVF_MAX_RXD); + new_rx_count = ALIGN(new_rx_count, REQ_RX_DESCRIPTOR_MULTIPLE); + + new_tx_count = max(ring->tx_pending, (u32)IGBVF_MIN_TXD); + new_tx_count = min(new_tx_count, (u32)IGBVF_MAX_TXD); + new_tx_count = ALIGN(new_tx_count, REQ_TX_DESCRIPTOR_MULTIPLE); + + if ((new_tx_count == adapter->tx_ring->count) && + (new_rx_count == adapter->rx_ring->count)) { + /* nothing to do */ + return 0; + } + + temp_ring = vmalloc(sizeof(struct igbvf_ring)); + if (!temp_ring) + return -ENOMEM; + + while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state)) + msleep(1); + + if (netif_running(adapter->netdev)) + igbvf_down(adapter); + + /* + * We can't just free everything and then setup again, + * because the ISRs in MSI-X mode get passed pointers + * to the tx and rx ring structs. + */ + if (new_tx_count != adapter->tx_ring->count) { + memcpy(temp_ring, adapter->tx_ring, sizeof(struct igbvf_ring)); + + temp_ring->count = new_tx_count; + err = igbvf_setup_tx_resources(adapter, temp_ring); + if (err) + goto err_setup; + + igbvf_free_tx_resources(adapter->tx_ring); + + memcpy(adapter->tx_ring, temp_ring, sizeof(struct igbvf_ring)); + } + + if (new_rx_count != adapter->rx_ring->count) { + memcpy(temp_ring, adapter->rx_ring, sizeof(struct igbvf_ring)); + + temp_ring->count = new_rx_count; + err = igbvf_setup_rx_resources(adapter, temp_ring); + if (err) + goto err_setup; + + igbvf_free_rx_resources(adapter->rx_ring); + + memcpy(adapter->rx_ring, temp_ring,sizeof(struct igbvf_ring)); + } + + err = 0; +err_setup: + if (netif_running(adapter->netdev)) + igbvf_up(adapter); + + clear_bit(__IGBVF_RESETTING, &adapter->state); + vfree(temp_ring); + return err; +} + +static int igbvf_link_test(struct igbvf_adapter *adapter, u64 *data) +{ + struct e1000_hw *hw = &adapter->hw; + *data = 0; + + hw->mac.ops.check_for_link(hw); + + if (!(er32(STATUS) & E1000_STATUS_LU)) + *data = 1; + + return *data; +} + +static int igbvf_get_self_test_count(struct net_device *netdev) +{ + return IGBVF_TEST_LEN; +} + +static int igbvf_get_stats_count(struct net_device *netdev) +{ + return IGBVF_GLOBAL_STATS_LEN; +} + +static void igbvf_diag_test(struct net_device *netdev, + struct ethtool_test *eth_test, u64 *data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + set_bit(__IGBVF_TESTING, &adapter->state); + + /* + * Link test performed before hardware reset so autoneg doesn't + * interfere with test result + */ + if (igbvf_link_test(adapter, &data[0])) + eth_test->flags |= ETH_TEST_FL_FAILED; + + clear_bit(__IGBVF_TESTING, &adapter->state); + msleep_interruptible(4 * 1000); +} + +static void igbvf_get_wol(struct net_device *netdev, + struct ethtool_wolinfo *wol) +{ + wol->supported = 0; + wol->wolopts = 0; + + return; +} + +static int igbvf_set_wol(struct net_device *netdev, + struct ethtool_wolinfo *wol) +{ + return -EOPNOTSUPP; +} + +static int igbvf_phys_id(struct net_device *netdev, u32 data) +{ + return 0; +} + +static int igbvf_get_coalesce(struct net_device *netdev, + struct ethtool_coalesce *ec) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (adapter->itr_setting <= 3) + ec->rx_coalesce_usecs = adapter->itr_setting; + else + ec->rx_coalesce_usecs = adapter->itr_setting >> 2; + + return 0; +} + +static int igbvf_set_coalesce(struct net_device *netdev, + struct ethtool_coalesce *ec) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + if ((ec->rx_coalesce_usecs > IGBVF_MAX_ITR_USECS) || + ((ec->rx_coalesce_usecs > 3) && + (ec->rx_coalesce_usecs < IGBVF_MIN_ITR_USECS)) || + (ec->rx_coalesce_usecs == 2)) + return -EINVAL; + + /* convert to rate of irq's per second */ + if (ec->rx_coalesce_usecs && ec->rx_coalesce_usecs <= 3) { + adapter->itr = IGBVF_START_ITR; + adapter->itr_setting = ec->rx_coalesce_usecs; + } else { + adapter->itr = ec->rx_coalesce_usecs << 2; + adapter->itr_setting = adapter->itr; + } + + writel(adapter->itr, + hw->hw_addr + adapter->rx_ring[0].itr_register); + + return 0; +} + +static int igbvf_nway_reset(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + if (netif_running(netdev)) + igbvf_reinit_locked(adapter); + return 0; +} + + +static void igbvf_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, + u64 *data) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + int i; + + igbvf_update_stats(adapter); + for (i = 0; i < IGBVF_GLOBAL_STATS_LEN; i++) { + char *p = (char *)adapter + + igbvf_gstrings_stats[i].stat_offset; + char *b = (char *)adapter + + igbvf_gstrings_stats[i].base_stat_offset; + data[i] = ((igbvf_gstrings_stats[i].sizeof_stat == + sizeof(u64)) ? (*(u64 *)p - *(u64 *)b) : + (*(u32 *)p - *(u32 *)b)); + } + +} + +static void igbvf_get_strings(struct net_device *netdev, u32 stringset, + u8 *data) +{ + u8 *p = data; + int i; + + switch (stringset) { + case ETH_SS_TEST: + memcpy(data, *igbvf_gstrings_test, sizeof(igbvf_gstrings_test)); + break; + case ETH_SS_STATS: + for (i = 0; i < IGBVF_GLOBAL_STATS_LEN; i++) { + memcpy(p, igbvf_gstrings_stats[i].stat_string, + ETH_GSTRING_LEN); + p += ETH_GSTRING_LEN; + } + break; + } +} + +static const struct ethtool_ops igbvf_ethtool_ops = { + .get_settings = igbvf_get_settings, + .set_settings = igbvf_set_settings, + .get_drvinfo = igbvf_get_drvinfo, + .get_regs_len = igbvf_get_regs_len, + .get_regs = igbvf_get_regs, + .get_wol = igbvf_get_wol, + .set_wol = igbvf_set_wol, + .get_msglevel = igbvf_get_msglevel, + .set_msglevel = igbvf_set_msglevel, + .nway_reset = igbvf_nway_reset, + .get_link = igbvf_get_link, + .get_eeprom_len = igbvf_get_eeprom_len, + .get_eeprom = igbvf_get_eeprom, + .set_eeprom = igbvf_set_eeprom, + .get_ringparam = igbvf_get_ringparam, + .set_ringparam = igbvf_set_ringparam, + .get_pauseparam = igbvf_get_pauseparam, + .set_pauseparam = igbvf_set_pauseparam, + .get_tx_csum = igbvf_get_tx_csum, + .set_tx_csum = igbvf_set_tx_csum, + .get_sg = ethtool_op_get_sg, + .set_sg = ethtool_op_set_sg, + .get_tso = ethtool_op_get_tso, + .set_tso = igbvf_set_tso, + .self_test = igbvf_diag_test, + .get_strings = igbvf_get_strings, + .phys_id = igbvf_phys_id, + .get_ethtool_stats = igbvf_get_ethtool_stats, + .self_test_count = igbvf_get_self_test_count, + .get_stats_count = igbvf_get_stats_count, + .get_coalesce = igbvf_get_coalesce, + .set_coalesce = igbvf_set_coalesce, +}; + +void igbvf_set_ethtool_ops(struct net_device *netdev) +{ + /* have to "undeclare" const on this struct to remove warnings */ + SET_ETHTOOL_OPS(netdev, (struct ethtool_ops *)&igbvf_ethtool_ops); +} diff --git a/drivers/net/igbvf/igbvf.h b/drivers/net/igbvf/igbvf.h new file mode 100644 index 00000000000..936ed2a9435 --- /dev/null +++ b/drivers/net/igbvf/igbvf.h @@ -0,0 +1,335 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 + +*******************************************************************************/ + +/* Linux PRO/1000 Ethernet Driver main header file */ + +#ifndef _IGBVF_H_ +#define _IGBVF_H_ + +#include <linux/types.h> +#include <linux/timer.h> +#include <linux/io.h> +#include <linux/netdevice.h> + + +#include "vf.h" + +/* Forward declarations */ +struct igbvf_info; +struct igbvf_adapter; + +/* Interrupt defines */ +#define IGBVF_START_ITR 648 /* ~6000 ints/sec */ + +/* Interrupt modes, as used by the IntMode paramter */ +#define IGBVF_INT_MODE_LEGACY 0 +#define IGBVF_INT_MODE_MSI 1 +#define IGBVF_INT_MODE_MSIX 2 + +/* Tx/Rx descriptor defines */ +#define IGBVF_DEFAULT_TXD 256 +#define IGBVF_MAX_TXD 4096 +#define IGBVF_MIN_TXD 80 + +#define IGBVF_DEFAULT_RXD 256 +#define IGBVF_MAX_RXD 4096 +#define IGBVF_MIN_RXD 80 + +#define IGBVF_MIN_ITR_USECS 10 /* 100000 irq/sec */ +#define IGBVF_MAX_ITR_USECS 10000 /* 100 irq/sec */ + +/* RX descriptor control thresholds. + * PTHRESH - MAC will consider prefetch if it has fewer than this number of + * descriptors available in its onboard memory. + * Setting this to 0 disables RX descriptor prefetch. + * HTHRESH - MAC will only prefetch if there are at least this many descriptors + * available in host memory. + * If PTHRESH is 0, this should also be 0. + * WTHRESH - RX descriptor writeback threshold - MAC will delay writing back + * descriptors until either it has this many to write back, or the + * ITR timer expires. + */ +#define IGBVF_RX_PTHRESH 16 +#define IGBVF_RX_HTHRESH 8 +#define IGBVF_RX_WTHRESH 1 + +/* this is the size past which hardware will drop packets when setting LPE=0 */ +#define MAXIMUM_ETHERNET_VLAN_SIZE 1522 + +#define IGBVF_FC_PAUSE_TIME 0x0680 /* 858 usec */ + +/* How many Tx Descriptors do we need to call netif_wake_queue ? */ +#define IGBVF_TX_QUEUE_WAKE 32 +/* How many Rx Buffers do we bundle into one write to the hardware ? */ +#define IGBVF_RX_BUFFER_WRITE 16 /* Must be power of 2 */ + +#define AUTO_ALL_MODES 0 +#define IGBVF_EEPROM_APME 0x0400 + +#define IGBVF_MNG_VLAN_NONE (-1) + +/* Number of packet split data buffers (not including the header buffer) */ +#define PS_PAGE_BUFFERS (MAX_PS_BUFFERS - 1) + +enum igbvf_boards { + board_vf, +}; + +struct igbvf_queue_stats { + u64 packets; + u64 bytes; +}; + +/* + * wrappers around a pointer to a socket buffer, + * so a DMA handle can be stored along with the buffer + */ +struct igbvf_buffer { + dma_addr_t dma; + struct sk_buff *skb; + union { + /* Tx */ + struct { + unsigned long time_stamp; + u16 length; + u16 next_to_watch; + }; + /* Rx */ + struct { + struct page *page; + u64 page_dma; + unsigned int page_offset; + }; + }; + struct page *page; +}; + +union igbvf_desc { + union e1000_adv_rx_desc rx_desc; + union e1000_adv_tx_desc tx_desc; + struct e1000_adv_tx_context_desc tx_context_desc; +}; + +struct igbvf_ring { + struct igbvf_adapter *adapter; /* backlink */ + union igbvf_desc *desc; /* pointer to ring memory */ + dma_addr_t dma; /* phys address of ring */ + unsigned int size; /* length of ring in bytes */ + unsigned int count; /* number of desc. in ring */ + + u16 next_to_use; + u16 next_to_clean; + + u16 head; + u16 tail; + + /* array of buffer information structs */ + struct igbvf_buffer *buffer_info; + struct napi_struct napi; + + char name[IFNAMSIZ + 5]; + u32 eims_value; + u32 itr_val; + u16 itr_register; + int set_itr; + + struct sk_buff *rx_skb_top; + + struct igbvf_queue_stats stats; +}; + +/* board specific private data structure */ +struct igbvf_adapter { + struct timer_list watchdog_timer; + struct timer_list blink_timer; + + struct work_struct reset_task; + struct work_struct watchdog_task; + + const struct igbvf_info *ei; + + struct vlan_group *vlgrp; + u32 bd_number; + u32 rx_buffer_len; + u32 polling_interval; + u16 mng_vlan_id; + u16 link_speed; + u16 link_duplex; + + spinlock_t tx_queue_lock; /* prevent concurrent tail updates */ + + /* track device up/down/testing state */ + unsigned long state; + + /* Interrupt Throttle Rate */ + u32 itr; + u32 itr_setting; + u16 tx_itr; + u16 rx_itr; + + /* + * Tx + */ + struct igbvf_ring *tx_ring /* One per active queue */ + ____cacheline_aligned_in_smp; + + unsigned long tx_queue_len; + unsigned int restart_queue; + u32 txd_cmd; + + bool detect_tx_hung; + u8 tx_timeout_factor; + + u32 tx_int_delay; + u32 tx_abs_int_delay; + + unsigned int total_tx_bytes; + unsigned int total_tx_packets; + unsigned int total_rx_bytes; + unsigned int total_rx_packets; + + /* Tx stats */ + u32 tx_timeout_count; + u32 tx_fifo_head; + u32 tx_head_addr; + u32 tx_fifo_size; + u32 tx_dma_failed; + + /* + * Rx + */ + struct igbvf_ring *rx_ring; + + u32 rx_int_delay; + u32 rx_abs_int_delay; + + /* Rx stats */ + u64 hw_csum_err; + u64 hw_csum_good; + u64 rx_hdr_split; + u32 alloc_rx_buff_failed; + u32 rx_dma_failed; + + unsigned int rx_ps_hdr_size; + u32 max_frame_size; + u32 min_frame_size; + + /* OS defined structs */ + struct net_device *netdev; + struct pci_dev *pdev; + struct net_device_stats net_stats; + spinlock_t stats_lock; /* prevent concurrent stats updates */ + + /* structs defined in e1000_hw.h */ + struct e1000_hw hw; + + /* The VF counters don't clear on read so we have to get a base + * count on driver start up and always subtract that base on + * on the first update, thus the flag.. + */ + struct e1000_vf_stats stats; + u64 zero_base; + + struct igbvf_ring test_tx_ring; + struct igbvf_ring test_rx_ring; + u32 test_icr; + + u32 msg_enable; + struct msix_entry *msix_entries; + int int_mode; + u32 eims_enable_mask; + u32 eims_other; + u32 int_counter0; + u32 int_counter1; + + u32 eeprom_wol; + u32 wol; + u32 pba; + + bool fc_autoneg; + + unsigned long led_status; + + unsigned int flags; +}; + +struct igbvf_info { + enum e1000_mac_type mac; + unsigned int flags; + u32 pba; + void (*init_ops)(struct e1000_hw *); + s32 (*get_variants)(struct igbvf_adapter *); +}; + +/* hardware capability, feature, and workaround flags */ +#define FLAG_HAS_HW_VLAN_FILTER (1 << 0) +#define FLAG_HAS_JUMBO_FRAMES (1 << 1) +#define FLAG_MSI_ENABLED (1 << 2) +#define FLAG_RX_CSUM_ENABLED (1 << 3) +#define FLAG_TSO_FORCE (1 << 4) + +#define IGBVF_RX_DESC_ADV(R, i) \ + (&((((R).desc))[i].rx_desc)) +#define IGBVF_TX_DESC_ADV(R, i) \ + (&((((R).desc))[i].tx_desc)) +#define IGBVF_TX_CTXTDESC_ADV(R, i) \ + (&((((R).desc))[i].tx_context_desc)) + +enum igbvf_state_t { + __IGBVF_TESTING, + __IGBVF_RESETTING, + __IGBVF_DOWN +}; + +enum latency_range { + lowest_latency = 0, + low_latency = 1, + bulk_latency = 2, + latency_invalid = 255 +}; + +extern char igbvf_driver_name[]; +extern const char igbvf_driver_version[]; + +extern void igbvf_check_options(struct igbvf_adapter *); +extern void igbvf_set_ethtool_ops(struct net_device *); + +extern int igbvf_up(struct igbvf_adapter *); +extern void igbvf_down(struct igbvf_adapter *); +extern void igbvf_reinit_locked(struct igbvf_adapter *); +extern void igbvf_reset(struct igbvf_adapter *); +extern int igbvf_setup_rx_resources(struct igbvf_adapter *, struct igbvf_ring *); +extern int igbvf_setup_tx_resources(struct igbvf_adapter *, struct igbvf_ring *); +extern void igbvf_free_rx_resources(struct igbvf_ring *); +extern void igbvf_free_tx_resources(struct igbvf_ring *); +extern void igbvf_update_stats(struct igbvf_adapter *); +extern void igbvf_set_interrupt_capability(struct igbvf_adapter *); +extern void igbvf_reset_interrupt_capability(struct igbvf_adapter *); + +extern unsigned int copybreak; + +#endif /* _IGBVF_H_ */ diff --git a/drivers/net/igbvf/mbx.c b/drivers/net/igbvf/mbx.c new file mode 100644 index 00000000000..819a8ec901d --- /dev/null +++ b/drivers/net/igbvf/mbx.c @@ -0,0 +1,350 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 "mbx.h" + +/** + * e1000_poll_for_msg - Wait for message notification + * @hw: pointer to the HW structure + * + * returns SUCCESS if it successfully received a message notification + **/ +static s32 e1000_poll_for_msg(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; + + if (!mbx->ops.check_for_msg) + goto out; + + while (countdown && mbx->ops.check_for_msg(hw)) { + countdown--; + udelay(mbx->usec_delay); + } + + /* if we failed, all future posted messages fail until reset */ + if (!countdown) + mbx->timeout = 0; +out: + return countdown ? E1000_SUCCESS : -E1000_ERR_MBX; +} + +/** + * e1000_poll_for_ack - Wait for message acknowledgement + * @hw: pointer to the HW structure + * + * returns SUCCESS if it successfully received a message acknowledgement + **/ +static s32 e1000_poll_for_ack(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + int countdown = mbx->timeout; + + if (!mbx->ops.check_for_ack) + goto out; + + while (countdown && mbx->ops.check_for_ack(hw)) { + countdown--; + udelay(mbx->usec_delay); + } + + /* if we failed, all future posted messages fail until reset */ + if (!countdown) + mbx->timeout = 0; +out: + return countdown ? E1000_SUCCESS : -E1000_ERR_MBX; +} + +/** + * e1000_read_posted_mbx - Wait for message notification and receive message + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully received a message notification and + * copied it into the receive buffer. + **/ +static s32 e1000_read_posted_mbx(struct e1000_hw *hw, u32 *msg, u16 size) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + s32 ret_val = -E1000_ERR_MBX; + + if (!mbx->ops.read) + goto out; + + ret_val = e1000_poll_for_msg(hw); + + /* if ack received read message, otherwise we timed out */ + if (!ret_val) + ret_val = mbx->ops.read(hw, msg, size); +out: + return ret_val; +} + +/** + * e1000_write_posted_mbx - Write a message to the mailbox, wait for ack + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully copied message into the buffer and + * received an ack to that message within delay * timeout period + **/ +static s32 e1000_write_posted_mbx(struct e1000_hw *hw, u32 *msg, u16 size) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + s32 ret_val = -E1000_ERR_MBX; + + /* exit if we either can't write or there isn't a defined timeout */ + if (!mbx->ops.write || !mbx->timeout) + goto out; + + /* send msg*/ + ret_val = mbx->ops.write(hw, msg, size); + + /* if msg sent wait until we receive an ack */ + if (!ret_val) + ret_val = e1000_poll_for_ack(hw); +out: + return ret_val; +} + +/** + * e1000_read_v2p_mailbox - read v2p mailbox + * @hw: pointer to the HW structure + * + * This function is used to read the v2p mailbox without losing the read to + * clear status bits. + **/ +static u32 e1000_read_v2p_mailbox(struct e1000_hw *hw) +{ + u32 v2p_mailbox = er32(V2PMAILBOX(0)); + + v2p_mailbox |= hw->dev_spec.vf.v2p_mailbox; + hw->dev_spec.vf.v2p_mailbox |= v2p_mailbox & E1000_V2PMAILBOX_R2C_BITS; + + return v2p_mailbox; +} + +/** + * e1000_check_for_bit_vf - Determine if a status bit was set + * @hw: pointer to the HW structure + * @mask: bitmask for bits to be tested and cleared + * + * This function is used to check for the read to clear bits within + * the V2P mailbox. + **/ +static s32 e1000_check_for_bit_vf(struct e1000_hw *hw, u32 mask) +{ + u32 v2p_mailbox = e1000_read_v2p_mailbox(hw); + s32 ret_val = -E1000_ERR_MBX; + + if (v2p_mailbox & mask) + ret_val = E1000_SUCCESS; + + hw->dev_spec.vf.v2p_mailbox &= ~mask; + + return ret_val; +} + +/** + * e1000_check_for_msg_vf - checks to see if the PF has sent mail + * @hw: pointer to the HW structure + * + * returns SUCCESS if the PF has set the Status bit or else ERR_MBX + **/ +static s32 e1000_check_for_msg_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + if (!e1000_check_for_bit_vf(hw, E1000_V2PMAILBOX_PFSTS)) { + ret_val = E1000_SUCCESS; + hw->mbx.stats.reqs++; + } + + return ret_val; +} + +/** + * e1000_check_for_ack_vf - checks to see if the PF has ACK'd + * @hw: pointer to the HW structure + * + * returns SUCCESS if the PF has set the ACK bit or else ERR_MBX + **/ +static s32 e1000_check_for_ack_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + if (!e1000_check_for_bit_vf(hw, E1000_V2PMAILBOX_PFACK)) { + ret_val = E1000_SUCCESS; + hw->mbx.stats.acks++; + } + + return ret_val; +} + +/** + * e1000_check_for_rst_vf - checks to see if the PF has reset + * @hw: pointer to the HW structure + * + * returns true if the PF has set the reset done bit or else false + **/ +static s32 e1000_check_for_rst_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + if (!e1000_check_for_bit_vf(hw, (E1000_V2PMAILBOX_RSTD | + E1000_V2PMAILBOX_RSTI))) { + ret_val = E1000_SUCCESS; + hw->mbx.stats.rsts++; + } + + return ret_val; +} + +/** + * e1000_obtain_mbx_lock_vf - obtain mailbox lock + * @hw: pointer to the HW structure + * + * return SUCCESS if we obtained the mailbox lock + **/ +static s32 e1000_obtain_mbx_lock_vf(struct e1000_hw *hw) +{ + s32 ret_val = -E1000_ERR_MBX; + + /* Take ownership of the buffer */ + ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_VFU); + + /* reserve mailbox for vf use */ + if (e1000_read_v2p_mailbox(hw) & E1000_V2PMAILBOX_VFU) + ret_val = E1000_SUCCESS; + + return ret_val; +} + +/** + * e1000_write_mbx_vf - Write a message to the mailbox + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfully copied message into the buffer + **/ +static s32 e1000_write_mbx_vf(struct e1000_hw *hw, u32 *msg, u16 size) +{ + s32 err; + u16 i; + + /* lock the mailbox to prevent pf/vf race condition */ + err = e1000_obtain_mbx_lock_vf(hw); + if (err) + goto out_no_write; + + /* flush any ack or msg as we are going to overwrite mailbox */ + e1000_check_for_ack_vf(hw); + e1000_check_for_msg_vf(hw); + + /* copy the caller specified message to the mailbox memory buffer */ + for (i = 0; i < size; i++) + array_ew32(VMBMEM(0), i, msg[i]); + + /* update stats */ + hw->mbx.stats.msgs_tx++; + + /* Drop VFU and interrupt the PF to tell it a message has been sent */ + ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_REQ); + +out_no_write: + return err; +} + +/** + * e1000_read_mbx_vf - Reads a message from the inbox intended for vf + * @hw: pointer to the HW structure + * @msg: The message buffer + * @size: Length of buffer + * + * returns SUCCESS if it successfuly read message from buffer + **/ +static s32 e1000_read_mbx_vf(struct e1000_hw *hw, u32 *msg, u16 size) +{ + s32 err; + u16 i; + + /* lock the mailbox to prevent pf/vf race condition */ + err = e1000_obtain_mbx_lock_vf(hw); + if (err) + goto out_no_read; + + /* copy the message from the mailbox memory buffer */ + for (i = 0; i < size; i++) + msg[i] = array_er32(VMBMEM(0), i); + + /* Acknowledge receipt and release mailbox, then we're done */ + ew32(V2PMAILBOX(0), E1000_V2PMAILBOX_ACK); + + /* update stats */ + hw->mbx.stats.msgs_rx++; + +out_no_read: + return err; +} + +/** + * e1000_init_mbx_params_vf - set initial values for vf mailbox + * @hw: pointer to the HW structure + * + * Initializes the hw->mbx struct to correct values for vf mailbox + */ +s32 e1000_init_mbx_params_vf(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + + /* start mailbox as timed out and let the reset_hw call set the timeout + * value to being communications */ + mbx->timeout = 0; + mbx->usec_delay = E1000_VF_MBX_INIT_DELAY; + + mbx->size = E1000_VFMAILBOX_SIZE; + + mbx->ops.read = e1000_read_mbx_vf; + mbx->ops.write = e1000_write_mbx_vf; + mbx->ops.read_posted = e1000_read_posted_mbx; + mbx->ops.write_posted = e1000_write_posted_mbx; + mbx->ops.check_for_msg = e1000_check_for_msg_vf; + mbx->ops.check_for_ack = e1000_check_for_ack_vf; + mbx->ops.check_for_rst = e1000_check_for_rst_vf; + + mbx->stats.msgs_tx = 0; + mbx->stats.msgs_rx = 0; + mbx->stats.reqs = 0; + mbx->stats.acks = 0; + mbx->stats.rsts = 0; + + return E1000_SUCCESS; +} + diff --git a/drivers/net/igbvf/mbx.h b/drivers/net/igbvf/mbx.h new file mode 100644 index 00000000000..4938609dbfb --- /dev/null +++ b/drivers/net/igbvf/mbx.h @@ -0,0 +1,75 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 1999 - 2009 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 _E1000_MBX_H_ +#define _E1000_MBX_H_ + +#include "vf.h" + +#define E1000_V2PMAILBOX_REQ 0x00000001 /* Request for PF Ready bit */ +#define E1000_V2PMAILBOX_ACK 0x00000002 /* Ack PF message received */ +#define E1000_V2PMAILBOX_VFU 0x00000004 /* VF owns the mailbox buffer */ +#define E1000_V2PMAILBOX_PFU 0x00000008 /* PF owns the mailbox buffer */ +#define E1000_V2PMAILBOX_PFSTS 0x00000010 /* PF wrote a message in the MB */ +#define E1000_V2PMAILBOX_PFACK 0x00000020 /* PF ack the previous VF msg */ +#define E1000_V2PMAILBOX_RSTI 0x00000040 /* PF has reset indication */ +#define E1000_V2PMAILBOX_RSTD 0x00000080 /* PF has indicated reset done */ +#define E1000_V2PMAILBOX_R2C_BITS 0x000000B0 /* All read to clear bits */ + +#define E1000_VFMAILBOX_SIZE 16 /* 16 32 bit words - 64 bytes */ + +/* If it's a E1000_VF_* msg then it originates in the VF and is sent to the + * PF. The reverse is true if it is E1000_PF_*. + * Message ACK's are the value or'd with 0xF0000000 + */ +#define E1000_VT_MSGTYPE_ACK 0x80000000 /* Messages below or'd with + * this are the ACK */ +#define E1000_VT_MSGTYPE_NACK 0x40000000 /* Messages below or'd with + * this are the NACK */ +#define E1000_VT_MSGTYPE_CTS 0x20000000 /* Indicates that VF is still + clear to send requests */ + +/* We have a total wait time of 1s for vf mailbox posted messages */ +#define E1000_VF_MBX_INIT_TIMEOUT 2000 /* retry count for mailbox timeout */ +#define E1000_VF_MBX_INIT_DELAY 500 /* usec delay between retries */ + +#define E1000_VT_MSGINFO_SHIFT 16 +/* bits 23:16 are used for exra info for certain messages */ +#define E1000_VT_MSGINFO_MASK (0xFF << E1000_VT_MSGINFO_SHIFT) + +#define E1000_VF_RESET 0x01 /* VF requests reset */ +#define E1000_VF_SET_MAC_ADDR 0x02 /* VF requests PF to set MAC addr */ +#define E1000_VF_SET_MULTICAST 0x03 /* VF requests PF to set MC addr */ +#define E1000_VF_SET_VLAN 0x04 /* VF requests PF to set VLAN */ +#define E1000_VF_SET_LPE 0x05 /* VF requests PF to set VMOLR.LPE */ + +#define E1000_PF_CONTROL_MSG 0x0100 /* PF control message */ + +void e1000_init_mbx_ops_generic(struct e1000_hw *hw); +s32 e1000_init_mbx_params_vf(struct e1000_hw *); + +#endif /* _E1000_MBX_H_ */ diff --git a/drivers/net/igbvf/netdev.c b/drivers/net/igbvf/netdev.c new file mode 100644 index 00000000000..c5648420ded --- /dev/null +++ b/drivers/net/igbvf/netdev.c @@ -0,0 +1,2919 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 <linux/module.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/vmalloc.h> +#include <linux/pagemap.h> +#include <linux/delay.h> +#include <linux/netdevice.h> +#include <linux/tcp.h> +#include <linux/ipv6.h> +#include <net/checksum.h> +#include <net/ip6_checksum.h> +#include <linux/mii.h> +#include <linux/ethtool.h> +#include <linux/if_vlan.h> +#include <linux/pm_qos_params.h> + +#include "igbvf.h" + +#define DRV_VERSION "1.0.0-k0" +char igbvf_driver_name[] = "igbvf"; +const char igbvf_driver_version[] = DRV_VERSION; +static const char igbvf_driver_string[] = + "Intel(R) Virtual Function Network Driver"; +static const char igbvf_copyright[] = "Copyright (c) 2009 Intel Corporation."; + +static int igbvf_poll(struct napi_struct *napi, int budget); + +static struct igbvf_info igbvf_vf_info = { + .mac = e1000_vfadapt, + .flags = FLAG_HAS_JUMBO_FRAMES + | FLAG_RX_CSUM_ENABLED, + .pba = 10, + .init_ops = e1000_init_function_pointers_vf, +}; + +static const struct igbvf_info *igbvf_info_tbl[] = { + [board_vf] = &igbvf_vf_info, +}; + +/** + * igbvf_desc_unused - calculate if we have unused descriptors + **/ +static int igbvf_desc_unused(struct igbvf_ring *ring) +{ + if (ring->next_to_clean > ring->next_to_use) + return ring->next_to_clean - ring->next_to_use - 1; + + return ring->count + ring->next_to_clean - ring->next_to_use - 1; +} + +/** + * igbvf_receive_skb - helper function to handle Rx indications + * @adapter: board private structure + * @status: descriptor status field as written by hardware + * @vlan: descriptor vlan field as written by hardware (no le/be conversion) + * @skb: pointer to sk_buff to be indicated to stack + **/ +static void igbvf_receive_skb(struct igbvf_adapter *adapter, + struct net_device *netdev, + struct sk_buff *skb, + u32 status, u16 vlan) +{ + if (adapter->vlgrp && (status & E1000_RXD_STAT_VP)) + vlan_hwaccel_receive_skb(skb, adapter->vlgrp, + le16_to_cpu(vlan) & + E1000_RXD_SPC_VLAN_MASK); + else + netif_receive_skb(skb); + + netdev->last_rx = jiffies; +} + +static inline void igbvf_rx_checksum_adv(struct igbvf_adapter *adapter, + u32 status_err, struct sk_buff *skb) +{ + skb->ip_summed = CHECKSUM_NONE; + + /* Ignore Checksum bit is set or checksum is disabled through ethtool */ + if ((status_err & E1000_RXD_STAT_IXSM)) + return; + /* TCP/UDP checksum error bit is set */ + if (status_err & + (E1000_RXDEXT_STATERR_TCPE | E1000_RXDEXT_STATERR_IPE)) { + /* let the stack verify checksum errors */ + adapter->hw_csum_err++; + return; + } + /* It must be a TCP or UDP packet with a valid checksum */ + if (status_err & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + adapter->hw_csum_good++; +} + +/** + * igbvf_alloc_rx_buffers - Replace used receive buffers; packet split + * @rx_ring: address of ring structure to repopulate + * @cleaned_count: number of buffers to repopulate + **/ +static void igbvf_alloc_rx_buffers(struct igbvf_ring *rx_ring, + int cleaned_count) +{ + struct igbvf_adapter *adapter = rx_ring->adapter; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + union e1000_adv_rx_desc *rx_desc; + struct igbvf_buffer *buffer_info; + struct sk_buff *skb; + unsigned int i; + int bufsz; + + i = rx_ring->next_to_use; + buffer_info = &rx_ring->buffer_info[i]; + + if (adapter->rx_ps_hdr_size) + bufsz = adapter->rx_ps_hdr_size; + else + bufsz = adapter->rx_buffer_len; + bufsz += NET_IP_ALIGN; + + while (cleaned_count--) { + rx_desc = IGBVF_RX_DESC_ADV(*rx_ring, i); + + if (adapter->rx_ps_hdr_size && !buffer_info->page_dma) { + if (!buffer_info->page) { + buffer_info->page = alloc_page(GFP_ATOMIC); + if (!buffer_info->page) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + buffer_info->page_offset = 0; + } else { + buffer_info->page_offset ^= PAGE_SIZE / 2; + } + buffer_info->page_dma = + pci_map_page(pdev, buffer_info->page, + buffer_info->page_offset, + PAGE_SIZE / 2, + PCI_DMA_FROMDEVICE); + } + + if (!buffer_info->skb) { + skb = netdev_alloc_skb(netdev, bufsz); + if (!skb) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + + /* Make buffer alignment 2 beyond a 16 byte boundary + * this will result in a 16 byte aligned IP header after + * the 14 byte MAC header is removed + */ + skb_reserve(skb, NET_IP_ALIGN); + + buffer_info->skb = skb; + buffer_info->dma = pci_map_single(pdev, skb->data, + bufsz, + PCI_DMA_FROMDEVICE); + } + /* Refresh the desc even if buffer_addrs didn't change because + * each write-back erases this info. */ + if (adapter->rx_ps_hdr_size) { + rx_desc->read.pkt_addr = + cpu_to_le64(buffer_info->page_dma); + rx_desc->read.hdr_addr = cpu_to_le64(buffer_info->dma); + } else { + rx_desc->read.pkt_addr = + cpu_to_le64(buffer_info->dma); + rx_desc->read.hdr_addr = 0; + } + + i++; + if (i == rx_ring->count) + i = 0; + buffer_info = &rx_ring->buffer_info[i]; + } + +no_buffers: + if (rx_ring->next_to_use != i) { + rx_ring->next_to_use = i; + if (i == 0) + i = (rx_ring->count - 1); + else + i--; + + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). */ + wmb(); + writel(i, adapter->hw.hw_addr + rx_ring->tail); + } +} + +/** + * igbvf_clean_rx_irq - Send received data up the network stack; legacy + * @adapter: board private structure + * + * the return value indicates whether actual cleaning was done, there + * is no guarantee that everything was cleaned + **/ +static bool igbvf_clean_rx_irq(struct igbvf_adapter *adapter, + int *work_done, int work_to_do) +{ + struct igbvf_ring *rx_ring = adapter->rx_ring; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + union e1000_adv_rx_desc *rx_desc, *next_rxd; + struct igbvf_buffer *buffer_info, *next_buffer; + struct sk_buff *skb; + bool cleaned = false; + int cleaned_count = 0; + unsigned int total_bytes = 0, total_packets = 0; + unsigned int i; + u32 length, hlen, staterr; + + i = rx_ring->next_to_clean; + rx_desc = IGBVF_RX_DESC_ADV(*rx_ring, i); + staterr = le32_to_cpu(rx_desc->wb.upper.status_error); + + while (staterr & E1000_RXD_STAT_DD) { + if (*work_done >= work_to_do) + break; + (*work_done)++; + + buffer_info = &rx_ring->buffer_info[i]; + + /* HW will not DMA in data larger than the given buffer, even + * if it parses the (NFS, of course) header to be larger. In + * that case, it fills the header buffer and spills the rest + * into the page. + */ + hlen = (le16_to_cpu(rx_desc->wb.lower.lo_dword.hs_rss.hdr_info) & + E1000_RXDADV_HDRBUFLEN_MASK) >> E1000_RXDADV_HDRBUFLEN_SHIFT; + if (hlen > adapter->rx_ps_hdr_size) + hlen = adapter->rx_ps_hdr_size; + + length = le16_to_cpu(rx_desc->wb.upper.length); + cleaned = true; + cleaned_count++; + + skb = buffer_info->skb; + prefetch(skb->data - NET_IP_ALIGN); + buffer_info->skb = NULL; + if (!adapter->rx_ps_hdr_size) { + pci_unmap_single(pdev, buffer_info->dma, + adapter->rx_buffer_len, + PCI_DMA_FROMDEVICE); + buffer_info->dma = 0; + skb_put(skb, length); + goto send_up; + } + + if (!skb_shinfo(skb)->nr_frags) { + pci_unmap_single(pdev, buffer_info->dma, + adapter->rx_ps_hdr_size + NET_IP_ALIGN, + PCI_DMA_FROMDEVICE); + skb_put(skb, hlen); + } + + if (length) { + pci_unmap_page(pdev, buffer_info->page_dma, + PAGE_SIZE / 2, + PCI_DMA_FROMDEVICE); + buffer_info->page_dma = 0; + + skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags++, + buffer_info->page, + buffer_info->page_offset, + length); + + if ((adapter->rx_buffer_len > (PAGE_SIZE / 2)) || + (page_count(buffer_info->page) != 1)) + buffer_info->page = NULL; + else + get_page(buffer_info->page); + + skb->len += length; + skb->data_len += length; + skb->truesize += length; + } +send_up: + i++; + if (i == rx_ring->count) + i = 0; + next_rxd = IGBVF_RX_DESC_ADV(*rx_ring, i); + prefetch(next_rxd); + next_buffer = &rx_ring->buffer_info[i]; + + if (!(staterr & E1000_RXD_STAT_EOP)) { + buffer_info->skb = next_buffer->skb; + buffer_info->dma = next_buffer->dma; + next_buffer->skb = skb; + next_buffer->dma = 0; + goto next_desc; + } + + if (staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) { + dev_kfree_skb_irq(skb); + goto next_desc; + } + + total_bytes += skb->len; + total_packets++; + + igbvf_rx_checksum_adv(adapter, staterr, skb); + + skb->protocol = eth_type_trans(skb, netdev); + + igbvf_receive_skb(adapter, netdev, skb, staterr, + rx_desc->wb.upper.vlan); + + netdev->last_rx = jiffies; + +next_desc: + rx_desc->wb.upper.status_error = 0; + + /* return some buffers to hardware, one at a time is too slow */ + if (cleaned_count >= IGBVF_RX_BUFFER_WRITE) { + igbvf_alloc_rx_buffers(rx_ring, cleaned_count); + cleaned_count = 0; + } + + /* use prefetched values */ + rx_desc = next_rxd; + buffer_info = next_buffer; + + staterr = le32_to_cpu(rx_desc->wb.upper.status_error); + } + + rx_ring->next_to_clean = i; + cleaned_count = igbvf_desc_unused(rx_ring); + + if (cleaned_count) + igbvf_alloc_rx_buffers(rx_ring, cleaned_count); + + adapter->total_rx_packets += total_packets; + adapter->total_rx_bytes += total_bytes; + adapter->net_stats.rx_bytes += total_bytes; + adapter->net_stats.rx_packets += total_packets; + return cleaned; +} + +static void igbvf_put_txbuf(struct igbvf_adapter *adapter, + struct igbvf_buffer *buffer_info) +{ + buffer_info->dma = 0; + if (buffer_info->skb) { + skb_dma_unmap(&adapter->pdev->dev, buffer_info->skb, + DMA_TO_DEVICE); + dev_kfree_skb_any(buffer_info->skb); + buffer_info->skb = NULL; + } + buffer_info->time_stamp = 0; +} + +static void igbvf_print_tx_hang(struct igbvf_adapter *adapter) +{ + struct igbvf_ring *tx_ring = adapter->tx_ring; + unsigned int i = tx_ring->next_to_clean; + unsigned int eop = tx_ring->buffer_info[i].next_to_watch; + union e1000_adv_tx_desc *eop_desc = IGBVF_TX_DESC_ADV(*tx_ring, eop); + + /* detected Tx unit hang */ + dev_err(&adapter->pdev->dev, + "Detected Tx Unit Hang:\n" + " TDH <%x>\n" + " TDT <%x>\n" + " next_to_use <%x>\n" + " next_to_clean <%x>\n" + "buffer_info[next_to_clean]:\n" + " time_stamp <%lx>\n" + " next_to_watch <%x>\n" + " jiffies <%lx>\n" + " next_to_watch.status <%x>\n", + readl(adapter->hw.hw_addr + tx_ring->head), + readl(adapter->hw.hw_addr + tx_ring->tail), + tx_ring->next_to_use, + tx_ring->next_to_clean, + tx_ring->buffer_info[eop].time_stamp, + eop, + jiffies, + eop_desc->wb.status); +} + +/** + * igbvf_setup_tx_resources - allocate Tx resources (Descriptors) + * @adapter: board private structure + * + * Return 0 on success, negative on failure + **/ +int igbvf_setup_tx_resources(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring) +{ + struct pci_dev *pdev = adapter->pdev; + int size; + + size = sizeof(struct igbvf_buffer) * tx_ring->count; + tx_ring->buffer_info = vmalloc(size); + if (!tx_ring->buffer_info) + goto err; + memset(tx_ring->buffer_info, 0, size); + + /* round up to nearest 4K */ + tx_ring->size = tx_ring->count * sizeof(union e1000_adv_tx_desc); + tx_ring->size = ALIGN(tx_ring->size, 4096); + + tx_ring->desc = pci_alloc_consistent(pdev, tx_ring->size, + &tx_ring->dma); + + if (!tx_ring->desc) + goto err; + + tx_ring->adapter = adapter; + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + + return 0; +err: + vfree(tx_ring->buffer_info); + dev_err(&adapter->pdev->dev, + "Unable to allocate memory for the transmit descriptor ring\n"); + return -ENOMEM; +} + +/** + * igbvf_setup_rx_resources - allocate Rx resources (Descriptors) + * @adapter: board private structure + * + * Returns 0 on success, negative on failure + **/ +int igbvf_setup_rx_resources(struct igbvf_adapter *adapter, + struct igbvf_ring *rx_ring) +{ + struct pci_dev *pdev = adapter->pdev; + int size, desc_len; + + size = sizeof(struct igbvf_buffer) * rx_ring->count; + rx_ring->buffer_info = vmalloc(size); + if (!rx_ring->buffer_info) + goto err; + memset(rx_ring->buffer_info, 0, size); + + desc_len = sizeof(union e1000_adv_rx_desc); + + /* Round up to nearest 4K */ + rx_ring->size = rx_ring->count * desc_len; + rx_ring->size = ALIGN(rx_ring->size, 4096); + + rx_ring->desc = pci_alloc_consistent(pdev, rx_ring->size, + &rx_ring->dma); + + if (!rx_ring->desc) + goto err; + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + + rx_ring->adapter = adapter; + + return 0; + +err: + vfree(rx_ring->buffer_info); + rx_ring->buffer_info = NULL; + dev_err(&adapter->pdev->dev, + "Unable to allocate memory for the receive descriptor ring\n"); + return -ENOMEM; +} + +/** + * igbvf_clean_tx_ring - Free Tx Buffers + * @tx_ring: ring to be cleaned + **/ +static void igbvf_clean_tx_ring(struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = tx_ring->adapter; + struct igbvf_buffer *buffer_info; + unsigned long size; + unsigned int i; + + if (!tx_ring->buffer_info) + return; + + /* Free all the Tx ring sk_buffs */ + for (i = 0; i < tx_ring->count; i++) { + buffer_info = &tx_ring->buffer_info[i]; + igbvf_put_txbuf(adapter, buffer_info); + } + + size = sizeof(struct igbvf_buffer) * tx_ring->count; + memset(tx_ring->buffer_info, 0, size); + + /* Zero out the descriptor ring */ + memset(tx_ring->desc, 0, tx_ring->size); + + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + + writel(0, adapter->hw.hw_addr + tx_ring->head); + writel(0, adapter->hw.hw_addr + tx_ring->tail); +} + +/** + * igbvf_free_tx_resources - Free Tx Resources per Queue + * @tx_ring: ring to free resources from + * + * Free all transmit software resources + **/ +void igbvf_free_tx_resources(struct igbvf_ring *tx_ring) +{ + struct pci_dev *pdev = tx_ring->adapter->pdev; + + igbvf_clean_tx_ring(tx_ring); + + vfree(tx_ring->buffer_info); + tx_ring->buffer_info = NULL; + + pci_free_consistent(pdev, tx_ring->size, tx_ring->desc, tx_ring->dma); + + tx_ring->desc = NULL; +} + +/** + * igbvf_clean_rx_ring - Free Rx Buffers per Queue + * @adapter: board private structure + **/ +static void igbvf_clean_rx_ring(struct igbvf_ring *rx_ring) +{ + struct igbvf_adapter *adapter = rx_ring->adapter; + struct igbvf_buffer *buffer_info; + struct pci_dev *pdev = adapter->pdev; + unsigned long size; + unsigned int i; + + if (!rx_ring->buffer_info) + return; + + /* Free all the Rx ring sk_buffs */ + for (i = 0; i < rx_ring->count; i++) { + buffer_info = &rx_ring->buffer_info[i]; + if (buffer_info->dma) { + if (adapter->rx_ps_hdr_size){ + pci_unmap_single(pdev, buffer_info->dma, + adapter->rx_ps_hdr_size, + PCI_DMA_FROMDEVICE); + } else { + pci_unmap_single(pdev, buffer_info->dma, + adapter->rx_buffer_len, + PCI_DMA_FROMDEVICE); + } + buffer_info->dma = 0; + } + + if (buffer_info->skb) { + dev_kfree_skb(buffer_info->skb); + buffer_info->skb = NULL; + } + + if (buffer_info->page) { + if (buffer_info->page_dma) + pci_unmap_page(pdev, buffer_info->page_dma, + PAGE_SIZE / 2, + PCI_DMA_FROMDEVICE); + put_page(buffer_info->page); + buffer_info->page = NULL; + buffer_info->page_dma = 0; + buffer_info->page_offset = 0; + } + } + + size = sizeof(struct igbvf_buffer) * rx_ring->count; + memset(rx_ring->buffer_info, 0, size); + + /* Zero out the descriptor ring */ + memset(rx_ring->desc, 0, rx_ring->size); + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + + writel(0, adapter->hw.hw_addr + rx_ring->head); + writel(0, adapter->hw.hw_addr + rx_ring->tail); +} + +/** + * igbvf_free_rx_resources - Free Rx Resources + * @rx_ring: ring to clean the resources from + * + * Free all receive software resources + **/ + +void igbvf_free_rx_resources(struct igbvf_ring *rx_ring) +{ + struct pci_dev *pdev = rx_ring->adapter->pdev; + + igbvf_clean_rx_ring(rx_ring); + + vfree(rx_ring->buffer_info); + rx_ring->buffer_info = NULL; + + dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc, + rx_ring->dma); + rx_ring->desc = NULL; +} + +/** + * igbvf_update_itr - update the dynamic ITR value based on statistics + * @adapter: pointer to adapter + * @itr_setting: current adapter->itr + * @packets: the number of packets during this measurement interval + * @bytes: the number of bytes during this measurement interval + * + * Stores a new ITR value based on packets and byte + * counts during the last interrupt. The advantage of per interrupt + * computation is faster updates and more accurate ITR for the current + * traffic pattern. Constants in this function were computed + * based on theoretical maximum wire speed and thresholds were set based + * on testing data as well as attempting to minimize response time + * while increasing bulk throughput. This functionality is controlled + * by the InterruptThrottleRate module parameter. + **/ +static unsigned int igbvf_update_itr(struct igbvf_adapter *adapter, + u16 itr_setting, int packets, + int bytes) +{ + unsigned int retval = itr_setting; + + if (packets == 0) + goto update_itr_done; + + switch (itr_setting) { + case lowest_latency: + /* handle TSO and jumbo frames */ + if (bytes/packets > 8000) + retval = bulk_latency; + else if ((packets < 5) && (bytes > 512)) + retval = low_latency; + break; + case low_latency: /* 50 usec aka 20000 ints/s */ + if (bytes > 10000) { + /* this if handles the TSO accounting */ + if (bytes/packets > 8000) + retval = bulk_latency; + else if ((packets < 10) || ((bytes/packets) > 1200)) + retval = bulk_latency; + else if ((packets > 35)) + retval = lowest_latency; + } else if (bytes/packets > 2000) { + retval = bulk_latency; + } else if (packets <= 2 && bytes < 512) { + retval = lowest_latency; + } + break; + case bulk_latency: /* 250 usec aka 4000 ints/s */ + if (bytes > 25000) { + if (packets > 35) + retval = low_latency; + } else if (bytes < 6000) { + retval = low_latency; + } + break; + } + +update_itr_done: + return retval; +} + +static void igbvf_set_itr(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + u16 current_itr; + u32 new_itr = adapter->itr; + + adapter->tx_itr = igbvf_update_itr(adapter, adapter->tx_itr, + adapter->total_tx_packets, + adapter->total_tx_bytes); + /* conservative mode (itr 3) eliminates the lowest_latency setting */ + if (adapter->itr_setting == 3 && adapter->tx_itr == lowest_latency) + adapter->tx_itr = low_latency; + + adapter->rx_itr = igbvf_update_itr(adapter, adapter->rx_itr, + adapter->total_rx_packets, + adapter->total_rx_bytes); + /* conservative mode (itr 3) eliminates the lowest_latency setting */ + if (adapter->itr_setting == 3 && adapter->rx_itr == lowest_latency) + adapter->rx_itr = low_latency; + + current_itr = max(adapter->rx_itr, adapter->tx_itr); + + switch (current_itr) { + /* counts and packets in update_itr are dependent on these numbers */ + case lowest_latency: + new_itr = 70000; + break; + case low_latency: + new_itr = 20000; /* aka hwitr = ~200 */ + break; + case bulk_latency: + new_itr = 4000; + break; + default: + break; + } + + if (new_itr != adapter->itr) { + /* + * this attempts to bias the interrupt rate towards Bulk + * by adding intermediate steps when interrupt rate is + * increasing + */ + new_itr = new_itr > adapter->itr ? + min(adapter->itr + (new_itr >> 2), new_itr) : + new_itr; + adapter->itr = new_itr; + adapter->rx_ring->itr_val = 1952; + + if (adapter->msix_entries) + adapter->rx_ring->set_itr = 1; + else + ew32(ITR, 1952); + } +} + +/** + * igbvf_clean_tx_irq - Reclaim resources after transmit completes + * @adapter: board private structure + * returns true if ring is completely cleaned + **/ +static bool igbvf_clean_tx_irq(struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = tx_ring->adapter; + struct e1000_hw *hw = &adapter->hw; + struct net_device *netdev = adapter->netdev; + struct igbvf_buffer *buffer_info; + struct sk_buff *skb; + union e1000_adv_tx_desc *tx_desc, *eop_desc; + unsigned int total_bytes = 0, total_packets = 0; + unsigned int i, eop, count = 0; + bool cleaned = false; + + i = tx_ring->next_to_clean; + eop = tx_ring->buffer_info[i].next_to_watch; + eop_desc = IGBVF_TX_DESC_ADV(*tx_ring, eop); + + while ((eop_desc->wb.status & cpu_to_le32(E1000_TXD_STAT_DD)) && + (count < tx_ring->count)) { + for (cleaned = false; !cleaned; count++) { + tx_desc = IGBVF_TX_DESC_ADV(*tx_ring, i); + buffer_info = &tx_ring->buffer_info[i]; + cleaned = (i == eop); + skb = buffer_info->skb; + + if (skb) { + unsigned int segs, bytecount; + + /* gso_segs is currently only valid for tcp */ + segs = skb_shinfo(skb)->gso_segs ?: 1; + /* multiply data chunks by size of headers */ + bytecount = ((segs - 1) * skb_headlen(skb)) + + skb->len; + total_packets += segs; + total_bytes += bytecount; + } + + igbvf_put_txbuf(adapter, buffer_info); + tx_desc->wb.status = 0; + + i++; + if (i == tx_ring->count) + i = 0; + } + eop = tx_ring->buffer_info[i].next_to_watch; + eop_desc = IGBVF_TX_DESC_ADV(*tx_ring, eop); + } + + tx_ring->next_to_clean = i; + + if (unlikely(count && + netif_carrier_ok(netdev) && + igbvf_desc_unused(tx_ring) >= IGBVF_TX_QUEUE_WAKE)) { + /* Make sure that anybody stopping the queue after this + * sees the new next_to_clean. + */ + smp_mb(); + if (netif_queue_stopped(netdev) && + !(test_bit(__IGBVF_DOWN, &adapter->state))) { + netif_wake_queue(netdev); + ++adapter->restart_queue; + } + } + + if (adapter->detect_tx_hung) { + /* Detect a transmit hang in hardware, this serializes the + * check with the clearing of time_stamp and movement of i */ + adapter->detect_tx_hung = false; + if (tx_ring->buffer_info[i].time_stamp && + time_after(jiffies, tx_ring->buffer_info[i].time_stamp + + (adapter->tx_timeout_factor * HZ)) + && !(er32(STATUS) & E1000_STATUS_TXOFF)) { + + tx_desc = IGBVF_TX_DESC_ADV(*tx_ring, i); + /* detected Tx unit hang */ + igbvf_print_tx_hang(adapter); + + netif_stop_queue(netdev); + } + } + adapter->net_stats.tx_bytes += total_bytes; + adapter->net_stats.tx_packets += total_packets; + return (count < tx_ring->count); +} + +static irqreturn_t igbvf_msix_other(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + adapter->int_counter1++; + + netif_carrier_off(netdev); + hw->mac.get_link_status = 1; + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + ew32(EIMS, adapter->eims_other); + + return IRQ_HANDLED; +} + +static irqreturn_t igbvf_intr_msix_tx(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + + + adapter->total_tx_bytes = 0; + adapter->total_tx_packets = 0; + + /* auto mask will automatically reenable the interrupt when we write + * EICS */ + if (!igbvf_clean_tx_irq(tx_ring)) + /* Ring was not completely cleaned, so fire another interrupt */ + ew32(EICS, tx_ring->eims_value); + else + ew32(EIMS, tx_ring->eims_value); + + return IRQ_HANDLED; +} + +static irqreturn_t igbvf_intr_msix_rx(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + + adapter->int_counter0++; + + /* Write the ITR value calculated at the end of the + * previous interrupt. + */ + if (adapter->rx_ring->set_itr) { + writel(adapter->rx_ring->itr_val, + adapter->hw.hw_addr + adapter->rx_ring->itr_register); + adapter->rx_ring->set_itr = 0; + } + + if (napi_schedule_prep(&adapter->rx_ring->napi)) { + adapter->total_rx_bytes = 0; + adapter->total_rx_packets = 0; + __napi_schedule(&adapter->rx_ring->napi); + } + + return IRQ_HANDLED; +} + +#define IGBVF_NO_QUEUE -1 + +static void igbvf_assign_vector(struct igbvf_adapter *adapter, int rx_queue, + int tx_queue, int msix_vector) +{ + struct e1000_hw *hw = &adapter->hw; + u32 ivar, index; + + /* 82576 uses a table-based method for assigning vectors. + Each queue has a single entry in the table to which we write + a vector number along with a "valid" bit. Sadly, the layout + of the table is somewhat counterintuitive. */ + if (rx_queue > IGBVF_NO_QUEUE) { + index = (rx_queue >> 1); + ivar = array_er32(IVAR0, index); + if (rx_queue & 0x1) { + /* vector goes into third byte of register */ + ivar = ivar & 0xFF00FFFF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 16; + } else { + /* vector goes into low byte of register */ + ivar = ivar & 0xFFFFFF00; + ivar |= msix_vector | E1000_IVAR_VALID; + } + adapter->rx_ring[rx_queue].eims_value = 1 << msix_vector; + array_ew32(IVAR0, index, ivar); + } + if (tx_queue > IGBVF_NO_QUEUE) { + index = (tx_queue >> 1); + ivar = array_er32(IVAR0, index); + if (tx_queue & 0x1) { + /* vector goes into high byte of register */ + ivar = ivar & 0x00FFFFFF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 24; + } else { + /* vector goes into second byte of register */ + ivar = ivar & 0xFFFF00FF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 8; + } + adapter->tx_ring[tx_queue].eims_value = 1 << msix_vector; + array_ew32(IVAR0, index, ivar); + } +} + +/** + * igbvf_configure_msix - Configure MSI-X hardware + * + * igbvf_configure_msix sets up the hardware to properly + * generate MSI-X interrupts. + **/ +static void igbvf_configure_msix(struct igbvf_adapter *adapter) +{ + u32 tmp; + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct igbvf_ring *rx_ring = adapter->rx_ring; + int vector = 0; + + adapter->eims_enable_mask = 0; + + igbvf_assign_vector(adapter, IGBVF_NO_QUEUE, 0, vector++); + adapter->eims_enable_mask |= tx_ring->eims_value; + if (tx_ring->itr_val) + writel(tx_ring->itr_val, + hw->hw_addr + tx_ring->itr_register); + else + writel(1952, hw->hw_addr + tx_ring->itr_register); + + igbvf_assign_vector(adapter, 0, IGBVF_NO_QUEUE, vector++); + adapter->eims_enable_mask |= rx_ring->eims_value; + if (rx_ring->itr_val) + writel(rx_ring->itr_val, + hw->hw_addr + rx_ring->itr_register); + else + writel(1952, hw->hw_addr + rx_ring->itr_register); + + /* set vector for other causes, i.e. link changes */ + + tmp = (vector++ | E1000_IVAR_VALID); + + ew32(IVAR_MISC, tmp); + + adapter->eims_enable_mask = (1 << (vector)) - 1; + adapter->eims_other = 1 << (vector - 1); + e1e_flush(); +} + +void igbvf_reset_interrupt_capability(struct igbvf_adapter *adapter) +{ + if (adapter->msix_entries) { + pci_disable_msix(adapter->pdev); + kfree(adapter->msix_entries); + adapter->msix_entries = NULL; + } +} + +/** + * igbvf_set_interrupt_capability - set MSI or MSI-X if supported + * + * Attempt to configure interrupts using the best available + * capabilities of the hardware and kernel. + **/ +void igbvf_set_interrupt_capability(struct igbvf_adapter *adapter) +{ + int err = -ENOMEM; + int i; + + /* we allocate 3 vectors, 1 for tx, 1 for rx, one for pf messages */ + adapter->msix_entries = kcalloc(3, sizeof(struct msix_entry), + GFP_KERNEL); + if (adapter->msix_entries) { + for (i = 0; i < 3; i++) + adapter->msix_entries[i].entry = i; + + err = pci_enable_msix(adapter->pdev, + adapter->msix_entries, 3); + } + + if (err) { + /* MSI-X failed */ + dev_err(&adapter->pdev->dev, + "Failed to initialize MSI-X interrupts.\n"); + igbvf_reset_interrupt_capability(adapter); + } +} + +/** + * igbvf_request_msix - Initialize MSI-X interrupts + * + * igbvf_request_msix allocates MSI-X vectors and requests interrupts from the + * kernel. + **/ +static int igbvf_request_msix(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int err = 0, vector = 0; + + if (strlen(netdev->name) < (IFNAMSIZ - 5)) { + sprintf(adapter->tx_ring->name, "%s-tx-0", netdev->name); + sprintf(adapter->rx_ring->name, "%s-rx-0", netdev->name); + } else { + memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ); + memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ); + } + + err = request_irq(adapter->msix_entries[vector].vector, + &igbvf_intr_msix_tx, 0, adapter->tx_ring->name, + netdev); + if (err) + goto out; + + adapter->tx_ring->itr_register = E1000_EITR(vector); + adapter->tx_ring->itr_val = 1952; + vector++; + + err = request_irq(adapter->msix_entries[vector].vector, + &igbvf_intr_msix_rx, 0, adapter->rx_ring->name, + netdev); + if (err) + goto out; + + adapter->rx_ring->itr_register = E1000_EITR(vector); + adapter->rx_ring->itr_val = 1952; + vector++; + + err = request_irq(adapter->msix_entries[vector].vector, + &igbvf_msix_other, 0, netdev->name, netdev); + if (err) + goto out; + + igbvf_configure_msix(adapter); + return 0; +out: + return err; +} + +/** + * igbvf_alloc_queues - Allocate memory for all rings + * @adapter: board private structure to initialize + **/ +static int __devinit igbvf_alloc_queues(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + + adapter->tx_ring = kzalloc(sizeof(struct igbvf_ring), GFP_KERNEL); + if (!adapter->tx_ring) + return -ENOMEM; + + adapter->rx_ring = kzalloc(sizeof(struct igbvf_ring), GFP_KERNEL); + if (!adapter->rx_ring) { + kfree(adapter->tx_ring); + return -ENOMEM; + } + + netif_napi_add(netdev, &adapter->rx_ring->napi, igbvf_poll, 64); + + return 0; +} + +/** + * igbvf_request_irq - initialize interrupts + * + * Attempts to configure interrupts using the best available + * capabilities of the hardware and kernel. + **/ +static int igbvf_request_irq(struct igbvf_adapter *adapter) +{ + int err = -1; + + /* igbvf supports msi-x only */ + if (adapter->msix_entries) + err = igbvf_request_msix(adapter); + + if (!err) + return err; + + dev_err(&adapter->pdev->dev, + "Unable to allocate interrupt, Error: %d\n", err); + + return err; +} + +static void igbvf_free_irq(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int vector; + + if (adapter->msix_entries) { + for (vector = 0; vector < 3; vector++) + free_irq(adapter->msix_entries[vector].vector, netdev); + } +} + +/** + * igbvf_irq_disable - Mask off interrupt generation on the NIC + **/ +static void igbvf_irq_disable(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + ew32(EIMC, ~0); + + if (adapter->msix_entries) + ew32(EIAC, 0); +} + +/** + * igbvf_irq_enable - Enable default interrupt generation settings + **/ +static void igbvf_irq_enable(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + ew32(EIAC, adapter->eims_enable_mask); + ew32(EIAM, adapter->eims_enable_mask); + ew32(EIMS, adapter->eims_enable_mask); +} + +/** + * igbvf_poll - NAPI Rx polling callback + * @napi: struct associated with this polling callback + * @budget: amount of packets driver is allowed to process this poll + **/ +static int igbvf_poll(struct napi_struct *napi, int budget) +{ + struct igbvf_ring *rx_ring = container_of(napi, struct igbvf_ring, napi); + struct igbvf_adapter *adapter = rx_ring->adapter; + struct e1000_hw *hw = &adapter->hw; + int work_done = 0; + + igbvf_clean_rx_irq(adapter, &work_done, budget); + + /* If not enough Rx work done, exit the polling mode */ + if (work_done < budget) { + napi_complete(napi); + + if (adapter->itr_setting & 3) + igbvf_set_itr(adapter); + + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + ew32(EIMS, adapter->rx_ring->eims_value); + } + + return work_done; +} + +/** + * igbvf_set_rlpml - set receive large packet maximum length + * @adapter: board private structure + * + * Configure the maximum size of packets that will be received + */ +static void igbvf_set_rlpml(struct igbvf_adapter *adapter) +{ + int max_frame_size = adapter->max_frame_size; + struct e1000_hw *hw = &adapter->hw; + + if (adapter->vlgrp) + max_frame_size += VLAN_TAG_SIZE; + + e1000_rlpml_set_vf(hw, max_frame_size); +} + +static void igbvf_vlan_rx_add_vid(struct net_device *netdev, u16 vid) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + if (hw->mac.ops.set_vfta(hw, vid, true)) + dev_err(&adapter->pdev->dev, "Failed to add vlan id %d\n", vid); +} + +static void igbvf_vlan_rx_kill_vid(struct net_device *netdev, u16 vid) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + igbvf_irq_disable(adapter); + vlan_group_set_device(adapter->vlgrp, vid, NULL); + + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + igbvf_irq_enable(adapter); + + if (hw->mac.ops.set_vfta(hw, vid, false)) + dev_err(&adapter->pdev->dev, + "Failed to remove vlan id %d\n", vid); +} + +static void igbvf_vlan_rx_register(struct net_device *netdev, + struct vlan_group *grp) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + adapter->vlgrp = grp; +} + +static void igbvf_restore_vlan(struct igbvf_adapter *adapter) +{ + u16 vid; + + if (!adapter->vlgrp) + return; + + for (vid = 0; vid < VLAN_GROUP_ARRAY_LEN; vid++) { + if (!vlan_group_get_device(adapter->vlgrp, vid)) + continue; + igbvf_vlan_rx_add_vid(adapter->netdev, vid); + } + + igbvf_set_rlpml(adapter); +} + +/** + * igbvf_configure_tx - Configure Transmit Unit after Reset + * @adapter: board private structure + * + * Configure the Tx unit of the MAC after a reset. + **/ +static void igbvf_configure_tx(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + u64 tdba; + u32 txdctl, dca_txctrl; + + /* disable transmits */ + txdctl = er32(TXDCTL(0)); + ew32(TXDCTL(0), txdctl & ~E1000_TXDCTL_QUEUE_ENABLE); + msleep(10); + + /* Setup the HW Tx Head and Tail descriptor pointers */ + ew32(TDLEN(0), tx_ring->count * sizeof(union e1000_adv_tx_desc)); + tdba = tx_ring->dma; + ew32(TDBAL(0), (tdba & DMA_32BIT_MASK)); + ew32(TDBAH(0), (tdba >> 32)); + ew32(TDH(0), 0); + ew32(TDT(0), 0); + tx_ring->head = E1000_TDH(0); + tx_ring->tail = E1000_TDT(0); + + /* Turn off Relaxed Ordering on head write-backs. The writebacks + * MUST be delivered in order or it will completely screw up + * our bookeeping. + */ + dca_txctrl = er32(DCA_TXCTRL(0)); + dca_txctrl &= ~E1000_DCA_TXCTRL_TX_WB_RO_EN; + ew32(DCA_TXCTRL(0), dca_txctrl); + + /* enable transmits */ + txdctl |= E1000_TXDCTL_QUEUE_ENABLE; + ew32(TXDCTL(0), txdctl); + + /* Setup Transmit Descriptor Settings for eop descriptor */ + adapter->txd_cmd = E1000_ADVTXD_DCMD_EOP | E1000_ADVTXD_DCMD_IFCS; + + /* enable Report Status bit */ + adapter->txd_cmd |= E1000_ADVTXD_DCMD_RS; + + adapter->tx_queue_len = adapter->netdev->tx_queue_len; +} + +/** + * igbvf_setup_srrctl - configure the receive control registers + * @adapter: Board private structure + **/ +static void igbvf_setup_srrctl(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + u32 srrctl = 0; + + srrctl &= ~(E1000_SRRCTL_DESCTYPE_MASK | + E1000_SRRCTL_BSIZEHDR_MASK | + E1000_SRRCTL_BSIZEPKT_MASK); + + /* Enable queue drop to avoid head of line blocking */ + srrctl |= E1000_SRRCTL_DROP_EN; + + /* Setup buffer sizes */ + srrctl |= ALIGN(adapter->rx_buffer_len, 1024) >> + E1000_SRRCTL_BSIZEPKT_SHIFT; + + if (adapter->rx_buffer_len < 2048) { + adapter->rx_ps_hdr_size = 0; + srrctl |= E1000_SRRCTL_DESCTYPE_ADV_ONEBUF; + } else { + adapter->rx_ps_hdr_size = 128; + srrctl |= adapter->rx_ps_hdr_size << + E1000_SRRCTL_BSIZEHDRSIZE_SHIFT; + srrctl |= E1000_SRRCTL_DESCTYPE_HDR_SPLIT_ALWAYS; + } + + ew32(SRRCTL(0), srrctl); +} + +/** + * igbvf_configure_rx - Configure Receive Unit after Reset + * @adapter: board private structure + * + * Configure the Rx unit of the MAC after a reset. + **/ +static void igbvf_configure_rx(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *rx_ring = adapter->rx_ring; + u64 rdba; + u32 rdlen, rxdctl; + + /* disable receives */ + rxdctl = er32(RXDCTL(0)); + ew32(RXDCTL(0), rxdctl & ~E1000_RXDCTL_QUEUE_ENABLE); + msleep(10); + + rdlen = rx_ring->count * sizeof(union e1000_adv_rx_desc); + + /* + * Setup the HW Rx Head and Tail Descriptor Pointers and + * the Base and Length of the Rx Descriptor Ring + */ + rdba = rx_ring->dma; + ew32(RDBAL(0), (rdba & DMA_32BIT_MASK)); + ew32(RDBAH(0), (rdba >> 32)); + ew32(RDLEN(0), rx_ring->count * sizeof(union e1000_adv_rx_desc)); + rx_ring->head = E1000_RDH(0); + rx_ring->tail = E1000_RDT(0); + ew32(RDH(0), 0); + ew32(RDT(0), 0); + + rxdctl |= E1000_RXDCTL_QUEUE_ENABLE; + rxdctl &= 0xFFF00000; + rxdctl |= IGBVF_RX_PTHRESH; + rxdctl |= IGBVF_RX_HTHRESH << 8; + rxdctl |= IGBVF_RX_WTHRESH << 16; + + igbvf_set_rlpml(adapter); + + /* enable receives */ + ew32(RXDCTL(0), rxdctl); +} + +/** + * igbvf_set_multi - Multicast and Promiscuous mode set + * @netdev: network interface device structure + * + * The set_multi entry point is called whenever the multicast address + * list or the network interface flags are updated. This routine is + * responsible for configuring the hardware for proper multicast, + * promiscuous mode, and all-multi behavior. + **/ +static void igbvf_set_multi(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct dev_mc_list *mc_ptr; + u8 *mta_list = NULL; + int i; + + if (netdev->mc_count) { + mta_list = kmalloc(netdev->mc_count * 6, GFP_ATOMIC); + if (!mta_list) { + dev_err(&adapter->pdev->dev, + "failed to allocate multicast filter list\n"); + return; + } + } + + /* prepare a packed array of only addresses. */ + mc_ptr = netdev->mc_list; + + for (i = 0; i < netdev->mc_count; i++) { + if (!mc_ptr) + break; + memcpy(mta_list + (i*ETH_ALEN), mc_ptr->dmi_addr, + ETH_ALEN); + mc_ptr = mc_ptr->next; + } + + hw->mac.ops.update_mc_addr_list(hw, mta_list, i, 0, 0); + kfree(mta_list); +} + +/** + * igbvf_configure - configure the hardware for Rx and Tx + * @adapter: private board structure + **/ +static void igbvf_configure(struct igbvf_adapter *adapter) +{ + igbvf_set_multi(adapter->netdev); + + igbvf_restore_vlan(adapter); + + igbvf_configure_tx(adapter); + igbvf_setup_srrctl(adapter); + igbvf_configure_rx(adapter); + igbvf_alloc_rx_buffers(adapter->rx_ring, + igbvf_desc_unused(adapter->rx_ring)); +} + +/* igbvf_reset - bring the hardware into a known good state + * + * This function boots the hardware and enables some settings that + * require a configuration cycle of the hardware - those cannot be + * set/changed during runtime. After reset the device needs to be + * properly configured for Rx, Tx etc. + */ +void igbvf_reset(struct igbvf_adapter *adapter) +{ + struct e1000_mac_info *mac = &adapter->hw.mac; + struct net_device *netdev = adapter->netdev; + struct e1000_hw *hw = &adapter->hw; + + /* Allow time for pending master requests to run */ + if (mac->ops.reset_hw(hw)) + dev_err(&adapter->pdev->dev, "PF still resetting\n"); + + mac->ops.init_hw(hw); + + if (is_valid_ether_addr(adapter->hw.mac.addr)) { + memcpy(netdev->dev_addr, adapter->hw.mac.addr, + netdev->addr_len); + memcpy(netdev->perm_addr, adapter->hw.mac.addr, + netdev->addr_len); + } +} + +int igbvf_up(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + /* hardware has been reset, we need to reload some things */ + igbvf_configure(adapter); + + clear_bit(__IGBVF_DOWN, &adapter->state); + + napi_enable(&adapter->rx_ring->napi); + if (adapter->msix_entries) + igbvf_configure_msix(adapter); + + /* Clear any pending interrupts. */ + er32(EICR); + igbvf_irq_enable(adapter); + + /* start the watchdog */ + hw->mac.get_link_status = 1; + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + + return 0; +} + +void igbvf_down(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + struct e1000_hw *hw = &adapter->hw; + u32 rxdctl, txdctl; + + /* + * signal that we're down so the interrupt handler does not + * reschedule our watchdog timer + */ + set_bit(__IGBVF_DOWN, &adapter->state); + + /* disable receives in the hardware */ + rxdctl = er32(RXDCTL(0)); + ew32(RXDCTL(0), rxdctl & ~E1000_RXDCTL_QUEUE_ENABLE); + + netif_stop_queue(netdev); + + /* disable transmits in the hardware */ + txdctl = er32(TXDCTL(0)); + ew32(TXDCTL(0), txdctl & ~E1000_TXDCTL_QUEUE_ENABLE); + + /* flush both disables and wait for them to finish */ + e1e_flush(); + msleep(10); + + napi_disable(&adapter->rx_ring->napi); + + igbvf_irq_disable(adapter); + + del_timer_sync(&adapter->watchdog_timer); + + netdev->tx_queue_len = adapter->tx_queue_len; + netif_carrier_off(netdev); + + /* record the stats before reset*/ + igbvf_update_stats(adapter); + + adapter->link_speed = 0; + adapter->link_duplex = 0; + + igbvf_reset(adapter); + igbvf_clean_tx_ring(adapter->tx_ring); + igbvf_clean_rx_ring(adapter->rx_ring); +} + +void igbvf_reinit_locked(struct igbvf_adapter *adapter) +{ + might_sleep(); + while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state)) + msleep(1); + igbvf_down(adapter); + igbvf_up(adapter); + clear_bit(__IGBVF_RESETTING, &adapter->state); +} + +/** + * igbvf_sw_init - Initialize general software structures (struct igbvf_adapter) + * @adapter: board private structure to initialize + * + * igbvf_sw_init initializes the Adapter private data structure. + * Fields are initialized based on PCI device information and + * OS network device settings (MTU size). + **/ +static int __devinit igbvf_sw_init(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + s32 rc; + + adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN; + adapter->rx_ps_hdr_size = 0; + adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN; + adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN; + + adapter->tx_int_delay = 8; + adapter->tx_abs_int_delay = 32; + adapter->rx_int_delay = 0; + adapter->rx_abs_int_delay = 8; + adapter->itr_setting = 3; + adapter->itr = 20000; + + /* Set various function pointers */ + adapter->ei->init_ops(&adapter->hw); + + rc = adapter->hw.mac.ops.init_params(&adapter->hw); + if (rc) + return rc; + + rc = adapter->hw.mbx.ops.init_params(&adapter->hw); + if (rc) + return rc; + + igbvf_set_interrupt_capability(adapter); + + if (igbvf_alloc_queues(adapter)) + return -ENOMEM; + + spin_lock_init(&adapter->tx_queue_lock); + + /* Explicitly disable IRQ since the NIC can be in any state. */ + igbvf_irq_disable(adapter); + + spin_lock_init(&adapter->stats_lock); + + set_bit(__IGBVF_DOWN, &adapter->state); + return 0; +} + +static void igbvf_initialize_last_counter_stats(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + adapter->stats.last_gprc = er32(VFGPRC); + adapter->stats.last_gorc = er32(VFGORC); + adapter->stats.last_gptc = er32(VFGPTC); + adapter->stats.last_gotc = er32(VFGOTC); + adapter->stats.last_mprc = er32(VFMPRC); + adapter->stats.last_gotlbc = er32(VFGOTLBC); + adapter->stats.last_gptlbc = er32(VFGPTLBC); + adapter->stats.last_gorlbc = er32(VFGORLBC); + adapter->stats.last_gprlbc = er32(VFGPRLBC); + + adapter->stats.base_gprc = er32(VFGPRC); + adapter->stats.base_gorc = er32(VFGORC); + adapter->stats.base_gptc = er32(VFGPTC); + adapter->stats.base_gotc = er32(VFGOTC); + adapter->stats.base_mprc = er32(VFMPRC); + adapter->stats.base_gotlbc = er32(VFGOTLBC); + adapter->stats.base_gptlbc = er32(VFGPTLBC); + adapter->stats.base_gorlbc = er32(VFGORLBC); + adapter->stats.base_gprlbc = er32(VFGPRLBC); +} + +/** + * igbvf_open - Called when a network interface is made active + * @netdev: network interface device structure + * + * Returns 0 on success, negative value on failure + * + * The open entry point is called when a network interface is made + * active by the system (IFF_UP). At this point all resources needed + * for transmit and receive operations are allocated, the interrupt + * handler is registered with the OS, the watchdog timer is started, + * and the stack is notified that the interface is ready. + **/ +static int igbvf_open(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + int err; + + /* disallow open during test */ + if (test_bit(__IGBVF_TESTING, &adapter->state)) + return -EBUSY; + + /* allocate transmit descriptors */ + err = igbvf_setup_tx_resources(adapter, adapter->tx_ring); + if (err) + goto err_setup_tx; + + /* allocate receive descriptors */ + err = igbvf_setup_rx_resources(adapter, adapter->rx_ring); + if (err) + goto err_setup_rx; + + /* + * before we allocate an interrupt, we must be ready to handle it. + * Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt + * as soon as we call pci_request_irq, so we have to setup our + * clean_rx handler before we do so. + */ + igbvf_configure(adapter); + + err = igbvf_request_irq(adapter); + if (err) + goto err_req_irq; + + /* From here on the code is the same as igbvf_up() */ + clear_bit(__IGBVF_DOWN, &adapter->state); + + napi_enable(&adapter->rx_ring->napi); + + /* clear any pending interrupts */ + er32(EICR); + + igbvf_irq_enable(adapter); + + /* start the watchdog */ + hw->mac.get_link_status = 1; + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + return 0; + +err_req_irq: + igbvf_free_rx_resources(adapter->rx_ring); +err_setup_rx: + igbvf_free_tx_resources(adapter->tx_ring); +err_setup_tx: + igbvf_reset(adapter); + + return err; +} + +/** + * igbvf_close - Disables a network interface + * @netdev: network interface device structure + * + * Returns 0, this is not allowed to fail + * + * The close entry point is called when an interface is de-activated + * by the OS. The hardware is still under the drivers control, but + * needs to be disabled. A global MAC reset is issued to stop the + * hardware, and all transmit and receive resources are freed. + **/ +static int igbvf_close(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + WARN_ON(test_bit(__IGBVF_RESETTING, &adapter->state)); + igbvf_down(adapter); + + igbvf_free_irq(adapter); + + igbvf_free_tx_resources(adapter->tx_ring); + igbvf_free_rx_resources(adapter->rx_ring); + + return 0; +} +/** + * igbvf_set_mac - Change the Ethernet Address of the NIC + * @netdev: network interface device structure + * @p: pointer to an address structure + * + * Returns 0 on success, negative on failure + **/ +static int igbvf_set_mac(struct net_device *netdev, void *p) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct sockaddr *addr = p; + + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(hw->mac.addr, addr->sa_data, netdev->addr_len); + + hw->mac.ops.rar_set(hw, hw->mac.addr, 0); + + if (memcmp(addr->sa_data, hw->mac.addr, 6)) + return -EADDRNOTAVAIL; + + memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); + + return 0; +} + +#define UPDATE_VF_COUNTER(reg, name) \ + { \ + u32 current_counter = er32(reg); \ + if (current_counter < adapter->stats.last_##name) \ + adapter->stats.name += 0x100000000LL; \ + adapter->stats.last_##name = current_counter; \ + adapter->stats.name &= 0xFFFFFFFF00000000LL; \ + adapter->stats.name |= current_counter; \ + } + +/** + * igbvf_update_stats - Update the board statistics counters + * @adapter: board private structure +**/ +void igbvf_update_stats(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct pci_dev *pdev = adapter->pdev; + + /* + * Prevent stats update while adapter is being reset, link is down + * or if the pci connection is down. + */ + if (adapter->link_speed == 0) + return; + + if (test_bit(__IGBVF_RESETTING, &adapter->state)) + return; + + if (pci_channel_offline(pdev)) + return; + + UPDATE_VF_COUNTER(VFGPRC, gprc); + UPDATE_VF_COUNTER(VFGORC, gorc); + UPDATE_VF_COUNTER(VFGPTC, gptc); + UPDATE_VF_COUNTER(VFGOTC, gotc); + UPDATE_VF_COUNTER(VFMPRC, mprc); + UPDATE_VF_COUNTER(VFGOTLBC, gotlbc); + UPDATE_VF_COUNTER(VFGPTLBC, gptlbc); + UPDATE_VF_COUNTER(VFGORLBC, gorlbc); + UPDATE_VF_COUNTER(VFGPRLBC, gprlbc); + + /* Fill out the OS statistics structure */ + adapter->net_stats.multicast = adapter->stats.mprc; +} + +static void igbvf_print_link_info(struct igbvf_adapter *adapter) +{ + dev_info(&adapter->pdev->dev, "Link is Up %d Mbps %s\n", + adapter->link_speed, + ((adapter->link_duplex == FULL_DUPLEX) ? + "Full Duplex" : "Half Duplex")); +} + +static bool igbvf_has_link(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + s32 ret_val = E1000_SUCCESS; + bool link_active; + + ret_val = hw->mac.ops.check_for_link(hw); + link_active = !hw->mac.get_link_status; + + /* if check for link returns error we will need to reset */ + if (ret_val) + schedule_work(&adapter->reset_task); + + return link_active; +} + +/** + * igbvf_watchdog - Timer Call-back + * @data: pointer to adapter cast into an unsigned long + **/ +static void igbvf_watchdog(unsigned long data) +{ + struct igbvf_adapter *adapter = (struct igbvf_adapter *) data; + + /* Do the rest outside of interrupt context */ + schedule_work(&adapter->watchdog_task); +} + +static void igbvf_watchdog_task(struct work_struct *work) +{ + struct igbvf_adapter *adapter = container_of(work, + struct igbvf_adapter, + watchdog_task); + struct net_device *netdev = adapter->netdev; + struct e1000_mac_info *mac = &adapter->hw.mac; + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct e1000_hw *hw = &adapter->hw; + u32 link; + int tx_pending = 0; + + link = igbvf_has_link(adapter); + + if (link) { + if (!netif_carrier_ok(netdev)) { + bool txb2b = 1; + + mac->ops.get_link_up_info(&adapter->hw, + &adapter->link_speed, + &adapter->link_duplex); + igbvf_print_link_info(adapter); + + /* + * tweak tx_queue_len according to speed/duplex + * and adjust the timeout factor + */ + netdev->tx_queue_len = adapter->tx_queue_len; + adapter->tx_timeout_factor = 1; + switch (adapter->link_speed) { + case SPEED_10: + txb2b = 0; + netdev->tx_queue_len = 10; + adapter->tx_timeout_factor = 16; + break; + case SPEED_100: + txb2b = 0; + netdev->tx_queue_len = 100; + /* maybe add some timeout factor ? */ + break; + } + + netif_carrier_on(netdev); + netif_wake_queue(netdev); + } + } else { + if (netif_carrier_ok(netdev)) { + adapter->link_speed = 0; + adapter->link_duplex = 0; + dev_info(&adapter->pdev->dev, "Link is Down\n"); + netif_carrier_off(netdev); + netif_stop_queue(netdev); + } + } + + if (netif_carrier_ok(netdev)) { + igbvf_update_stats(adapter); + } else { + tx_pending = (igbvf_desc_unused(tx_ring) + 1 < + tx_ring->count); + if (tx_pending) { + /* + * We've lost link, so the controller stops DMA, + * but we've got queued Tx work that's never going + * to get done, so reset controller to flush Tx. + * (Do the reset outside of interrupt context). + */ + adapter->tx_timeout_count++; + schedule_work(&adapter->reset_task); + } + } + + /* Cause software interrupt to ensure Rx ring is cleaned */ + ew32(EICS, adapter->rx_ring->eims_value); + + /* Force detection of hung controller every watchdog period */ + adapter->detect_tx_hung = 1; + + /* Reset the timer */ + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + mod_timer(&adapter->watchdog_timer, + round_jiffies(jiffies + (2 * HZ))); +} + +#define IGBVF_TX_FLAGS_CSUM 0x00000001 +#define IGBVF_TX_FLAGS_VLAN 0x00000002 +#define IGBVF_TX_FLAGS_TSO 0x00000004 +#define IGBVF_TX_FLAGS_IPV4 0x00000008 +#define IGBVF_TX_FLAGS_VLAN_MASK 0xffff0000 +#define IGBVF_TX_FLAGS_VLAN_SHIFT 16 + +static int igbvf_tso(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring, + struct sk_buff *skb, u32 tx_flags, u8 *hdr_len) +{ + struct e1000_adv_tx_context_desc *context_desc; + unsigned int i; + int err; + struct igbvf_buffer *buffer_info; + u32 info = 0, tu_cmd = 0; + u32 mss_l4len_idx, l4len; + *hdr_len = 0; + + if (skb_header_cloned(skb)) { + err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC); + if (err) { + dev_err(&adapter->pdev->dev, + "igbvf_tso returning an error\n"); + return err; + } + } + + l4len = tcp_hdrlen(skb); + *hdr_len += l4len; + + if (skb->protocol == htons(ETH_P_IP)) { + struct iphdr *iph = ip_hdr(skb); + iph->tot_len = 0; + iph->check = 0; + tcp_hdr(skb)->check = ~csum_tcpudp_magic(iph->saddr, + iph->daddr, 0, + IPPROTO_TCP, + 0); + } else if (skb_shinfo(skb)->gso_type == SKB_GSO_TCPV6) { + ipv6_hdr(skb)->payload_len = 0; + tcp_hdr(skb)->check = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + 0, IPPROTO_TCP, 0); + } + + i = tx_ring->next_to_use; + + buffer_info = &tx_ring->buffer_info[i]; + context_desc = IGBVF_TX_CTXTDESC_ADV(*tx_ring, i); + /* VLAN MACLEN IPLEN */ + if (tx_flags & IGBVF_TX_FLAGS_VLAN) + info |= (tx_flags & IGBVF_TX_FLAGS_VLAN_MASK); + info |= (skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT); + *hdr_len += skb_network_offset(skb); + info |= (skb_transport_header(skb) - skb_network_header(skb)); + *hdr_len += (skb_transport_header(skb) - skb_network_header(skb)); + context_desc->vlan_macip_lens = cpu_to_le32(info); + + /* ADV DTYP TUCMD MKRLOC/ISCSIHEDLEN */ + tu_cmd |= (E1000_TXD_CMD_DEXT | E1000_ADVTXD_DTYP_CTXT); + + if (skb->protocol == htons(ETH_P_IP)) + tu_cmd |= E1000_ADVTXD_TUCMD_IPV4; + tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP; + + context_desc->type_tucmd_mlhl = cpu_to_le32(tu_cmd); + + /* MSS L4LEN IDX */ + mss_l4len_idx = (skb_shinfo(skb)->gso_size << E1000_ADVTXD_MSS_SHIFT); + mss_l4len_idx |= (l4len << E1000_ADVTXD_L4LEN_SHIFT); + + context_desc->mss_l4len_idx = cpu_to_le32(mss_l4len_idx); + context_desc->seqnum_seed = 0; + + buffer_info->time_stamp = jiffies; + buffer_info->next_to_watch = i; + buffer_info->dma = 0; + i++; + if (i == tx_ring->count) + i = 0; + + tx_ring->next_to_use = i; + + return true; +} + +static inline bool igbvf_tx_csum(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring, + struct sk_buff *skb, u32 tx_flags) +{ + struct e1000_adv_tx_context_desc *context_desc; + unsigned int i; + struct igbvf_buffer *buffer_info; + u32 info = 0, tu_cmd = 0; + + if ((skb->ip_summed == CHECKSUM_PARTIAL) || + (tx_flags & IGBVF_TX_FLAGS_VLAN)) { + i = tx_ring->next_to_use; + buffer_info = &tx_ring->buffer_info[i]; + context_desc = IGBVF_TX_CTXTDESC_ADV(*tx_ring, i); + + if (tx_flags & IGBVF_TX_FLAGS_VLAN) + info |= (tx_flags & IGBVF_TX_FLAGS_VLAN_MASK); + + info |= (skb_network_offset(skb) << E1000_ADVTXD_MACLEN_SHIFT); + if (skb->ip_summed == CHECKSUM_PARTIAL) + info |= (skb_transport_header(skb) - + skb_network_header(skb)); + + + context_desc->vlan_macip_lens = cpu_to_le32(info); + + tu_cmd |= (E1000_TXD_CMD_DEXT | E1000_ADVTXD_DTYP_CTXT); + + if (skb->ip_summed == CHECKSUM_PARTIAL) { + switch (skb->protocol) { + case __constant_htons(ETH_P_IP): + tu_cmd |= E1000_ADVTXD_TUCMD_IPV4; + if (ip_hdr(skb)->protocol == IPPROTO_TCP) + tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP; + break; + case __constant_htons(ETH_P_IPV6): + if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP) + tu_cmd |= E1000_ADVTXD_TUCMD_L4T_TCP; + break; + default: + break; + } + } + + context_desc->type_tucmd_mlhl = cpu_to_le32(tu_cmd); + context_desc->seqnum_seed = 0; + context_desc->mss_l4len_idx = 0; + + buffer_info->time_stamp = jiffies; + buffer_info->next_to_watch = i; + buffer_info->dma = 0; + i++; + if (i == tx_ring->count) + i = 0; + tx_ring->next_to_use = i; + + return true; + } + + return false; +} + +static int igbvf_maybe_stop_tx(struct net_device *netdev, int size) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + /* there is enough descriptors then we don't need to worry */ + if (igbvf_desc_unused(adapter->tx_ring) >= size) + return 0; + + netif_stop_queue(netdev); + + smp_mb(); + + /* We need to check again just in case room has been made available */ + if (igbvf_desc_unused(adapter->tx_ring) < size) + return -EBUSY; + + netif_wake_queue(netdev); + + ++adapter->restart_queue; + return 0; +} + +#define IGBVF_MAX_TXD_PWR 16 +#define IGBVF_MAX_DATA_PER_TXD (1 << IGBVF_MAX_TXD_PWR) + +static inline int igbvf_tx_map_adv(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring, + struct sk_buff *skb, + unsigned int first) +{ + struct igbvf_buffer *buffer_info; + unsigned int len = skb_headlen(skb); + unsigned int count = 0, i; + unsigned int f; + dma_addr_t *map; + + i = tx_ring->next_to_use; + + if (skb_dma_map(&adapter->pdev->dev, skb, DMA_TO_DEVICE)) { + dev_err(&adapter->pdev->dev, "TX DMA map failed\n"); + return 0; + } + + map = skb_shinfo(skb)->dma_maps; + + buffer_info = &tx_ring->buffer_info[i]; + BUG_ON(len >= IGBVF_MAX_DATA_PER_TXD); + buffer_info->length = len; + /* set time_stamp *before* dma to help avoid a possible race */ + buffer_info->time_stamp = jiffies; + buffer_info->next_to_watch = i; + buffer_info->dma = map[count]; + count++; + + for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) { + struct skb_frag_struct *frag; + + i++; + if (i == tx_ring->count) + i = 0; + + frag = &skb_shinfo(skb)->frags[f]; + len = frag->size; + + buffer_info = &tx_ring->buffer_info[i]; + BUG_ON(len >= IGBVF_MAX_DATA_PER_TXD); + buffer_info->length = len; + buffer_info->time_stamp = jiffies; + buffer_info->next_to_watch = i; + buffer_info->dma = map[count]; + count++; + } + + tx_ring->buffer_info[i].skb = skb; + tx_ring->buffer_info[first].next_to_watch = i; + + return count; +} + +static inline void igbvf_tx_queue_adv(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring, + int tx_flags, int count, u32 paylen, + u8 hdr_len) +{ + union e1000_adv_tx_desc *tx_desc = NULL; + struct igbvf_buffer *buffer_info; + u32 olinfo_status = 0, cmd_type_len; + unsigned int i; + + cmd_type_len = (E1000_ADVTXD_DTYP_DATA | E1000_ADVTXD_DCMD_IFCS | + E1000_ADVTXD_DCMD_DEXT); + + if (tx_flags & IGBVF_TX_FLAGS_VLAN) + cmd_type_len |= E1000_ADVTXD_DCMD_VLE; + + if (tx_flags & IGBVF_TX_FLAGS_TSO) { + cmd_type_len |= E1000_ADVTXD_DCMD_TSE; + + /* insert tcp checksum */ + olinfo_status |= E1000_TXD_POPTS_TXSM << 8; + + /* insert ip checksum */ + if (tx_flags & IGBVF_TX_FLAGS_IPV4) + olinfo_status |= E1000_TXD_POPTS_IXSM << 8; + + } else if (tx_flags & IGBVF_TX_FLAGS_CSUM) { + olinfo_status |= E1000_TXD_POPTS_TXSM << 8; + } + + olinfo_status |= ((paylen - hdr_len) << E1000_ADVTXD_PAYLEN_SHIFT); + + i = tx_ring->next_to_use; + while (count--) { + buffer_info = &tx_ring->buffer_info[i]; + tx_desc = IGBVF_TX_DESC_ADV(*tx_ring, i); + tx_desc->read.buffer_addr = cpu_to_le64(buffer_info->dma); + tx_desc->read.cmd_type_len = + cpu_to_le32(cmd_type_len | buffer_info->length); + tx_desc->read.olinfo_status = cpu_to_le32(olinfo_status); + i++; + if (i == tx_ring->count) + i = 0; + } + + tx_desc->read.cmd_type_len |= cpu_to_le32(adapter->txd_cmd); + /* Force memory writes to complete before letting h/w + * know there are new descriptors to fetch. (Only + * applicable for weak-ordered memory model archs, + * such as IA-64). */ + wmb(); + + tx_ring->next_to_use = i; + writel(i, adapter->hw.hw_addr + tx_ring->tail); + /* we need this if more than one processor can write to our tail + * at a time, it syncronizes IO on IA64/Altix systems */ + mmiowb(); +} + +static int igbvf_xmit_frame_ring_adv(struct sk_buff *skb, + struct net_device *netdev, + struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + unsigned int first, tx_flags = 0; + u8 hdr_len = 0; + int count = 0; + int tso = 0; + + if (test_bit(__IGBVF_DOWN, &adapter->state)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + if (skb->len <= 0) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + /* + * need: count + 4 desc gap to keep tail from touching + * + 2 desc gap to keep tail from touching head, + * + 1 desc for skb->data, + * + 1 desc for context descriptor, + * head, otherwise try next time + */ + if (igbvf_maybe_stop_tx(netdev, skb_shinfo(skb)->nr_frags + 4)) { + /* this is a hard error */ + return NETDEV_TX_BUSY; + } + + if (adapter->vlgrp && vlan_tx_tag_present(skb)) { + tx_flags |= IGBVF_TX_FLAGS_VLAN; + tx_flags |= (vlan_tx_tag_get(skb) << IGBVF_TX_FLAGS_VLAN_SHIFT); + } + + if (skb->protocol == htons(ETH_P_IP)) + tx_flags |= IGBVF_TX_FLAGS_IPV4; + + first = tx_ring->next_to_use; + + tso = skb_is_gso(skb) ? + igbvf_tso(adapter, tx_ring, skb, tx_flags, &hdr_len) : 0; + if (unlikely(tso < 0)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + if (tso) + tx_flags |= IGBVF_TX_FLAGS_TSO; + else if (igbvf_tx_csum(adapter, tx_ring, skb, tx_flags) && + (skb->ip_summed == CHECKSUM_PARTIAL)) + tx_flags |= IGBVF_TX_FLAGS_CSUM; + + /* + * count reflects descriptors mapped, if 0 then mapping error + * has occured and we need to rewind the descriptor queue + */ + count = igbvf_tx_map_adv(adapter, tx_ring, skb, first); + + if (count) { + igbvf_tx_queue_adv(adapter, tx_ring, tx_flags, count, + skb->len, hdr_len); + netdev->trans_start = jiffies; + /* Make sure there is space in the ring for the next send. */ + igbvf_maybe_stop_tx(netdev, MAX_SKB_FRAGS + 4); + } else { + dev_kfree_skb_any(skb); + tx_ring->buffer_info[first].time_stamp = 0; + tx_ring->next_to_use = first; + } + + return NETDEV_TX_OK; +} + +static int igbvf_xmit_frame(struct sk_buff *skb, struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct igbvf_ring *tx_ring; + int retval; + + if (test_bit(__IGBVF_DOWN, &adapter->state)) { + dev_kfree_skb_any(skb); + return NETDEV_TX_OK; + } + + tx_ring = &adapter->tx_ring[0]; + + retval = igbvf_xmit_frame_ring_adv(skb, netdev, tx_ring); + + return retval; +} + +/** + * igbvf_tx_timeout - Respond to a Tx Hang + * @netdev: network interface device structure + **/ +static void igbvf_tx_timeout(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + /* Do the reset outside of interrupt context */ + adapter->tx_timeout_count++; + schedule_work(&adapter->reset_task); +} + +static void igbvf_reset_task(struct work_struct *work) +{ + struct igbvf_adapter *adapter; + adapter = container_of(work, struct igbvf_adapter, reset_task); + + igbvf_reinit_locked(adapter); +} + +/** + * igbvf_get_stats - Get System Network Statistics + * @netdev: network interface device structure + * + * Returns the address of the device statistics structure. + * The statistics are actually updated from the timer callback. + **/ +static struct net_device_stats *igbvf_get_stats(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + /* only return the current stats */ + return &adapter->net_stats; +} + +/** + * igbvf_change_mtu - Change the Maximum Transfer Unit + * @netdev: network interface device structure + * @new_mtu: new value for maximum frame size + * + * Returns 0 on success, negative on failure + **/ +static int igbvf_change_mtu(struct net_device *netdev, int new_mtu) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + int max_frame = new_mtu + ETH_HLEN + ETH_FCS_LEN; + + if ((new_mtu < 68) || (max_frame > MAX_JUMBO_FRAME_SIZE)) { + dev_err(&adapter->pdev->dev, "Invalid MTU setting\n"); + return -EINVAL; + } + + /* Jumbo frame size limits */ + if (max_frame > ETH_FRAME_LEN + ETH_FCS_LEN) { + if (!(adapter->flags & FLAG_HAS_JUMBO_FRAMES)) { + dev_err(&adapter->pdev->dev, + "Jumbo Frames not supported.\n"); + return -EINVAL; + } + } + +#define MAX_STD_JUMBO_FRAME_SIZE 9234 + if (max_frame > MAX_STD_JUMBO_FRAME_SIZE) { + dev_err(&adapter->pdev->dev, "MTU > 9216 not supported.\n"); + return -EINVAL; + } + + while (test_and_set_bit(__IGBVF_RESETTING, &adapter->state)) + msleep(1); + /* igbvf_down has a dependency on max_frame_size */ + adapter->max_frame_size = max_frame; + if (netif_running(netdev)) + igbvf_down(adapter); + + /* + * NOTE: netdev_alloc_skb reserves 16 bytes, and typically NET_IP_ALIGN + * means we reserve 2 more, this pushes us to allocate from the next + * larger slab size. + * i.e. RXBUFFER_2048 --> size-4096 slab + * However with the new *_jumbo_rx* routines, jumbo receives will use + * fragmented skbs + */ + + if (max_frame <= 1024) + adapter->rx_buffer_len = 1024; + else if (max_frame <= 2048) + adapter->rx_buffer_len = 2048; + else +#if (PAGE_SIZE / 2) > 16384 + adapter->rx_buffer_len = 16384; +#else + adapter->rx_buffer_len = PAGE_SIZE / 2; +#endif + + + /* adjust allocation if LPE protects us, and we aren't using SBP */ + if ((max_frame == ETH_FRAME_LEN + ETH_FCS_LEN) || + (max_frame == ETH_FRAME_LEN + VLAN_HLEN + ETH_FCS_LEN)) + adapter->rx_buffer_len = ETH_FRAME_LEN + VLAN_HLEN + + ETH_FCS_LEN; + + dev_info(&adapter->pdev->dev, "changing MTU from %d to %d\n", + netdev->mtu, new_mtu); + netdev->mtu = new_mtu; + + if (netif_running(netdev)) + igbvf_up(adapter); + else + igbvf_reset(adapter); + + clear_bit(__IGBVF_RESETTING, &adapter->state); + + return 0; +} + +static int igbvf_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) +{ + switch (cmd) { + default: + return -EOPNOTSUPP; + } +} + +static int igbvf_suspend(struct pci_dev *pdev, pm_message_t state) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); +#ifdef CONFIG_PM + int retval = 0; +#endif + + netif_device_detach(netdev); + + if (netif_running(netdev)) { + WARN_ON(test_bit(__IGBVF_RESETTING, &adapter->state)); + igbvf_down(adapter); + igbvf_free_irq(adapter); + } + +#ifdef CONFIG_PM + retval = pci_save_state(pdev); + if (retval) + return retval; +#endif + + pci_disable_device(pdev); + + return 0; +} + +#ifdef CONFIG_PM +static int igbvf_resume(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + u32 err; + + pci_restore_state(pdev); + err = pci_enable_device_mem(pdev); + if (err) { + dev_err(&pdev->dev, "Cannot enable PCI device from suspend\n"); + return err; + } + + pci_set_master(pdev); + + if (netif_running(netdev)) { + err = igbvf_request_irq(adapter); + if (err) + return err; + } + + igbvf_reset(adapter); + + if (netif_running(netdev)) + igbvf_up(adapter); + + netif_device_attach(netdev); + + return 0; +} +#endif + +static void igbvf_shutdown(struct pci_dev *pdev) +{ + igbvf_suspend(pdev, PMSG_SUSPEND); +} + +#ifdef CONFIG_NET_POLL_CONTROLLER +/* + * Polling 'interrupt' - used by things like netconsole to send skbs + * without having to re-enable interrupts. It's not called while + * the interrupt routine is executing. + */ +static void igbvf_netpoll(struct net_device *netdev) +{ + struct igbvf_adapter *adapter = netdev_priv(netdev); + + disable_irq(adapter->pdev->irq); + + igbvf_clean_tx_irq(adapter->tx_ring); + + enable_irq(adapter->pdev->irq); +} +#endif + +/** + * igbvf_io_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current pci connection state + * + * This function is called after a PCI bus error affecting + * this device has been detected. + */ +static pci_ers_result_t igbvf_io_error_detected(struct pci_dev *pdev, + pci_channel_state_t state) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + netif_device_detach(netdev); + + if (netif_running(netdev)) + igbvf_down(adapter); + pci_disable_device(pdev); + + /* Request a slot slot reset. */ + return PCI_ERS_RESULT_NEED_RESET; +} + +/** + * igbvf_io_slot_reset - called after the pci bus has been reset. + * @pdev: Pointer to PCI device + * + * Restart the card from scratch, as if from a cold-boot. Implementation + * resembles the first-half of the igbvf_resume routine. + */ +static pci_ers_result_t igbvf_io_slot_reset(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (pci_enable_device_mem(pdev)) { + dev_err(&pdev->dev, + "Cannot re-enable PCI device after reset.\n"); + return PCI_ERS_RESULT_DISCONNECT; + } + pci_set_master(pdev); + + igbvf_reset(adapter); + + return PCI_ERS_RESULT_RECOVERED; +} + +/** + * igbvf_io_resume - called when traffic can start flowing again. + * @pdev: Pointer to PCI device + * + * This callback is called when the error recovery driver tells us that + * its OK to resume normal operation. Implementation resembles the + * second-half of the igbvf_resume routine. + */ +static void igbvf_io_resume(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + + if (netif_running(netdev)) { + if (igbvf_up(adapter)) { + dev_err(&pdev->dev, + "can't bring device back up after reset\n"); + return; + } + } + + netif_device_attach(netdev); +} + +static void igbvf_print_device_info(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + + dev_info(&pdev->dev, "Intel(R) 82576 Virtual Function\n"); + dev_info(&pdev->dev, "Address: %02x:%02x:%02x:%02x:%02x:%02x\n", + /* MAC address */ + netdev->dev_addr[0], netdev->dev_addr[1], + netdev->dev_addr[2], netdev->dev_addr[3], + netdev->dev_addr[4], netdev->dev_addr[5]); + dev_info(&pdev->dev, "MAC: %d\n", hw->mac.type); +} + +static const struct net_device_ops igbvf_netdev_ops = { + .ndo_open = igbvf_open, + .ndo_stop = igbvf_close, + .ndo_start_xmit = igbvf_xmit_frame, + .ndo_get_stats = igbvf_get_stats, + .ndo_set_multicast_list = igbvf_set_multi, + .ndo_set_mac_address = igbvf_set_mac, + .ndo_change_mtu = igbvf_change_mtu, + .ndo_do_ioctl = igbvf_ioctl, + .ndo_tx_timeout = igbvf_tx_timeout, + .ndo_vlan_rx_register = igbvf_vlan_rx_register, + .ndo_vlan_rx_add_vid = igbvf_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = igbvf_vlan_rx_kill_vid, +#ifdef CONFIG_NET_POLL_CONTROLLER + .ndo_poll_controller = igbvf_netpoll, +#endif +}; + +/** + * igbvf_probe - Device Initialization Routine + * @pdev: PCI device information struct + * @ent: entry in igbvf_pci_tbl + * + * Returns 0 on success, negative on failure + * + * igbvf_probe initializes an adapter identified by a pci_dev structure. + * The OS initialization, configuring of the adapter private structure, + * and a hardware reset occur. + **/ +static int __devinit igbvf_probe(struct pci_dev *pdev, + const struct pci_device_id *ent) +{ + struct net_device *netdev; + struct igbvf_adapter *adapter; + struct e1000_hw *hw; + const struct igbvf_info *ei = igbvf_info_tbl[ent->driver_data]; + + static int cards_found; + int err, pci_using_dac; + + err = pci_enable_device_mem(pdev); + if (err) + return err; + + pci_using_dac = 0; + err = pci_set_dma_mask(pdev, DMA_64BIT_MASK); + if (!err) { + err = pci_set_consistent_dma_mask(pdev, DMA_64BIT_MASK); + if (!err) + pci_using_dac = 1; + } else { + err = pci_set_dma_mask(pdev, DMA_32BIT_MASK); + if (err) { + err = pci_set_consistent_dma_mask(pdev, DMA_32BIT_MASK); + if (err) { + dev_err(&pdev->dev, "No usable DMA " + "configuration, aborting\n"); + goto err_dma; + } + } + } + + err = pci_request_regions(pdev, igbvf_driver_name); + if (err) + goto err_pci_reg; + + pci_set_master(pdev); + + err = -ENOMEM; + netdev = alloc_etherdev(sizeof(struct igbvf_adapter)); + if (!netdev) + goto err_alloc_etherdev; + + SET_NETDEV_DEV(netdev, &pdev->dev); + + pci_set_drvdata(pdev, netdev); + adapter = netdev_priv(netdev); + hw = &adapter->hw; + adapter->netdev = netdev; + adapter->pdev = pdev; + adapter->ei = ei; + adapter->pba = ei->pba; + adapter->flags = ei->flags; + adapter->hw.back = adapter; + adapter->hw.mac.type = ei->mac; + adapter->msg_enable = (1 << NETIF_MSG_DRV | NETIF_MSG_PROBE) - 1; + + /* PCI config space info */ + + hw->vendor_id = pdev->vendor; + hw->device_id = pdev->device; + hw->subsystem_vendor_id = pdev->subsystem_vendor; + hw->subsystem_device_id = pdev->subsystem_device; + + pci_read_config_byte(pdev, PCI_REVISION_ID, &hw->revision_id); + + err = -EIO; + adapter->hw.hw_addr = ioremap(pci_resource_start(pdev, 0), + pci_resource_len(pdev, 0)); + + if (!adapter->hw.hw_addr) + goto err_ioremap; + + if (ei->get_variants) { + err = ei->get_variants(adapter); + if (err) + goto err_ioremap; + } + + /* setup adapter struct */ + err = igbvf_sw_init(adapter); + if (err) + goto err_sw_init; + + /* construct the net_device struct */ + netdev->netdev_ops = &igbvf_netdev_ops; + + igbvf_set_ethtool_ops(netdev); + netdev->watchdog_timeo = 5 * HZ; + strncpy(netdev->name, pci_name(pdev), sizeof(netdev->name) - 1); + + adapter->bd_number = cards_found++; + + netdev->features = NETIF_F_SG | + NETIF_F_IP_CSUM | + NETIF_F_HW_VLAN_TX | + NETIF_F_HW_VLAN_RX | + NETIF_F_HW_VLAN_FILTER; + + netdev->features |= NETIF_F_IPV6_CSUM; + netdev->features |= NETIF_F_TSO; + netdev->features |= NETIF_F_TSO6; + + if (pci_using_dac) + netdev->features |= NETIF_F_HIGHDMA; + + netdev->vlan_features |= NETIF_F_TSO; + netdev->vlan_features |= NETIF_F_TSO6; + netdev->vlan_features |= NETIF_F_IP_CSUM; + netdev->vlan_features |= NETIF_F_IPV6_CSUM; + netdev->vlan_features |= NETIF_F_SG; + + /*reset the controller to put the device in a known good state */ + err = hw->mac.ops.reset_hw(hw); + if (err) { + dev_info(&pdev->dev, + "PF still in reset state, assigning new address\n"); + random_ether_addr(hw->mac.addr); + } else { + err = hw->mac.ops.read_mac_addr(hw); + if (err) { + dev_err(&pdev->dev, "Error reading MAC address\n"); + goto err_hw_init; + } + } + + memcpy(netdev->dev_addr, adapter->hw.mac.addr, netdev->addr_len); + memcpy(netdev->perm_addr, adapter->hw.mac.addr, netdev->addr_len); + + if (!is_valid_ether_addr(netdev->perm_addr)) { + dev_err(&pdev->dev, "Invalid MAC Address: " + "%02x:%02x:%02x:%02x:%02x:%02x\n", + netdev->dev_addr[0], netdev->dev_addr[1], + netdev->dev_addr[2], netdev->dev_addr[3], + netdev->dev_addr[4], netdev->dev_addr[5]); + err = -EIO; + goto err_hw_init; + } + + setup_timer(&adapter->watchdog_timer, &igbvf_watchdog, + (unsigned long) adapter); + + INIT_WORK(&adapter->reset_task, igbvf_reset_task); + INIT_WORK(&adapter->watchdog_task, igbvf_watchdog_task); + + /* ring size defaults */ + adapter->rx_ring->count = 1024; + adapter->tx_ring->count = 1024; + + /* reset the hardware with the new settings */ + igbvf_reset(adapter); + + /* tell the stack to leave us alone until igbvf_open() is called */ + netif_carrier_off(netdev); + netif_stop_queue(netdev); + + strcpy(netdev->name, "eth%d"); + err = register_netdev(netdev); + if (err) + goto err_hw_init; + + igbvf_print_device_info(adapter); + + igbvf_initialize_last_counter_stats(adapter); + + return 0; + +err_hw_init: + kfree(adapter->tx_ring); + kfree(adapter->rx_ring); +err_sw_init: + igbvf_reset_interrupt_capability(adapter); + iounmap(adapter->hw.hw_addr); +err_ioremap: + free_netdev(netdev); +err_alloc_etherdev: + pci_release_regions(pdev); +err_pci_reg: +err_dma: + pci_disable_device(pdev); + return err; +} + +/** + * igbvf_remove - Device Removal Routine + * @pdev: PCI device information struct + * + * igbvf_remove is called by the PCI subsystem to alert the driver + * that it should release a PCI device. The could be caused by a + * Hot-Plug event, or because the driver is going to be removed from + * memory. + **/ +static void __devexit igbvf_remove(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + /* + * flush_scheduled work may reschedule our watchdog task, so + * explicitly disable watchdog tasks from being rescheduled + */ + set_bit(__IGBVF_DOWN, &adapter->state); + del_timer_sync(&adapter->watchdog_timer); + + flush_scheduled_work(); + + unregister_netdev(netdev); + + igbvf_reset_interrupt_capability(adapter); + + /* + * it is important to delete the napi struct prior to freeing the + * rx ring so that you do not end up with null pointer refs + */ + netif_napi_del(&adapter->rx_ring->napi); + kfree(adapter->tx_ring); + kfree(adapter->rx_ring); + + iounmap(hw->hw_addr); + if (hw->flash_address) + iounmap(hw->flash_address); + pci_release_regions(pdev); + + free_netdev(netdev); + + pci_disable_device(pdev); +} + +/* PCI Error Recovery (ERS) */ +static struct pci_error_handlers igbvf_err_handler = { + .error_detected = igbvf_io_error_detected, + .slot_reset = igbvf_io_slot_reset, + .resume = igbvf_io_resume, +}; + +static struct pci_device_id igbvf_pci_tbl[] = { + { PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_VF), board_vf }, + { } /* terminate list */ +}; +MODULE_DEVICE_TABLE(pci, igbvf_pci_tbl); + +/* PCI Device API Driver */ +static struct pci_driver igbvf_driver = { + .name = igbvf_driver_name, + .id_table = igbvf_pci_tbl, + .probe = igbvf_probe, + .remove = __devexit_p(igbvf_remove), +#ifdef CONFIG_PM + /* Power Management Hooks */ + .suspend = igbvf_suspend, + .resume = igbvf_resume, +#endif + .shutdown = igbvf_shutdown, + .err_handler = &igbvf_err_handler +}; + +/** + * igbvf_init_module - Driver Registration Routine + * + * igbvf_init_module is the first routine called when the driver is + * loaded. All it does is register with the PCI subsystem. + **/ +static int __init igbvf_init_module(void) +{ + int ret; + printk(KERN_INFO "%s - version %s\n", + igbvf_driver_string, igbvf_driver_version); + printk(KERN_INFO "%s\n", igbvf_copyright); + + ret = pci_register_driver(&igbvf_driver); + pm_qos_add_requirement(PM_QOS_CPU_DMA_LATENCY, igbvf_driver_name, + PM_QOS_DEFAULT_VALUE); + + return ret; +} +module_init(igbvf_init_module); + +/** + * igbvf_exit_module - Driver Exit Cleanup Routine + * + * igbvf_exit_module is called just before the driver is removed + * from memory. + **/ +static void __exit igbvf_exit_module(void) +{ + pci_unregister_driver(&igbvf_driver); + pm_qos_remove_requirement(PM_QOS_CPU_DMA_LATENCY, igbvf_driver_name); +} +module_exit(igbvf_exit_module); + + +MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>"); +MODULE_DESCRIPTION("Intel(R) 82576 Virtual Function Network Driver"); +MODULE_LICENSE("GPL"); +MODULE_VERSION(DRV_VERSION); + +/* netdev.c */ diff --git a/drivers/net/igbvf/regs.h b/drivers/net/igbvf/regs.h new file mode 100644 index 00000000000..b9e24ed70d0 --- /dev/null +++ b/drivers/net/igbvf/regs.h @@ -0,0 +1,108 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 _E1000_REGS_H_ +#define _E1000_REGS_H_ + +#define E1000_CTRL 0x00000 /* Device Control - RW */ +#define E1000_STATUS 0x00008 /* Device Status - RO */ +#define E1000_ITR 0x000C4 /* Interrupt Throttling Rate - RW */ +#define E1000_EICR 0x01580 /* Ext. Interrupt Cause Read - R/clr */ +#define E1000_EITR(_n) (0x01680 + (0x4 * (_n))) +#define E1000_EICS 0x01520 /* Ext. Interrupt Cause Set - W0 */ +#define E1000_EIMS 0x01524 /* Ext. Interrupt Mask Set/Read - RW */ +#define E1000_EIMC 0x01528 /* Ext. Interrupt Mask Clear - WO */ +#define E1000_EIAC 0x0152C /* Ext. Interrupt Auto Clear - RW */ +#define E1000_EIAM 0x01530 /* Ext. Interrupt Ack Auto Clear Mask - RW */ +#define E1000_IVAR0 0x01700 /* Interrupt Vector Allocation (array) - RW */ +#define E1000_IVAR_MISC 0x01740 /* IVAR for "other" causes - RW */ +/* + * Convenience macros + * + * Note: "_n" is the queue number of the register to be written to. + * + * Example usage: + * E1000_RDBAL_REG(current_rx_queue) + */ +#define E1000_RDBAL(_n) ((_n) < 4 ? (0x02800 + ((_n) * 0x100)) : \ + (0x0C000 + ((_n) * 0x40))) +#define E1000_RDBAH(_n) ((_n) < 4 ? (0x02804 + ((_n) * 0x100)) : \ + (0x0C004 + ((_n) * 0x40))) +#define E1000_RDLEN(_n) ((_n) < 4 ? (0x02808 + ((_n) * 0x100)) : \ + (0x0C008 + ((_n) * 0x40))) +#define E1000_SRRCTL(_n) ((_n) < 4 ? (0x0280C + ((_n) * 0x100)) : \ + (0x0C00C + ((_n) * 0x40))) +#define E1000_RDH(_n) ((_n) < 4 ? (0x02810 + ((_n) * 0x100)) : \ + (0x0C010 + ((_n) * 0x40))) +#define E1000_RDT(_n) ((_n) < 4 ? (0x02818 + ((_n) * 0x100)) : \ + (0x0C018 + ((_n) * 0x40))) +#define E1000_RXDCTL(_n) ((_n) < 4 ? (0x02828 + ((_n) * 0x100)) : \ + (0x0C028 + ((_n) * 0x40))) +#define E1000_TDBAL(_n) ((_n) < 4 ? (0x03800 + ((_n) * 0x100)) : \ + (0x0E000 + ((_n) * 0x40))) +#define E1000_TDBAH(_n) ((_n) < 4 ? (0x03804 + ((_n) * 0x100)) : \ + (0x0E004 + ((_n) * 0x40))) +#define E1000_TDLEN(_n) ((_n) < 4 ? (0x03808 + ((_n) * 0x100)) : \ + (0x0E008 + ((_n) * 0x40))) +#define E1000_TDH(_n) ((_n) < 4 ? (0x03810 + ((_n) * 0x100)) : \ + (0x0E010 + ((_n) * 0x40))) +#define E1000_TDT(_n) ((_n) < 4 ? (0x03818 + ((_n) * 0x100)) : \ + (0x0E018 + ((_n) * 0x40))) +#define E1000_TXDCTL(_n) ((_n) < 4 ? (0x03828 + ((_n) * 0x100)) : \ + (0x0E028 + ((_n) * 0x40))) +#define E1000_DCA_TXCTRL(_n) (0x03814 + (_n << 8)) +#define E1000_DCA_RXCTRL(_n) (0x02814 + (_n << 8)) +#define E1000_RAL(_i) (((_i) <= 15) ? (0x05400 + ((_i) * 8)) : \ + (0x054E0 + ((_i - 16) * 8))) +#define E1000_RAH(_i) (((_i) <= 15) ? (0x05404 + ((_i) * 8)) : \ + (0x054E4 + ((_i - 16) * 8))) + +/* Statistics registers */ +#define E1000_VFGPRC 0x00F10 +#define E1000_VFGORC 0x00F18 +#define E1000_VFMPRC 0x00F3C +#define E1000_VFGPTC 0x00F14 +#define E1000_VFGOTC 0x00F34 +#define E1000_VFGOTLBC 0x00F50 +#define E1000_VFGPTLBC 0x00F44 +#define E1000_VFGORLBC 0x00F48 +#define E1000_VFGPRLBC 0x00F40 + +/* These act per VF so an array friendly macro is used */ +#define E1000_V2PMAILBOX(_n) (0x00C40 + (4 * (_n))) +#define E1000_VMBMEM(_n) (0x00800 + (64 * (_n))) + +/* Define macros for handling registers */ +#define er32(reg) readl(hw->hw_addr + E1000_##reg) +#define ew32(reg, val) writel((val), hw->hw_addr + E1000_##reg) +#define array_er32(reg, offset) \ + readl(hw->hw_addr + E1000_##reg + (offset << 2)) +#define array_ew32(reg, offset, val) \ + writel((val), hw->hw_addr + E1000_##reg + (offset << 2)) +#define e1e_flush() er32(STATUS) + +#endif diff --git a/drivers/net/igbvf/vf.c b/drivers/net/igbvf/vf.c new file mode 100644 index 00000000000..aa246c93279 --- /dev/null +++ b/drivers/net/igbvf/vf.c @@ -0,0 +1,398 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 "vf.h" + +static s32 e1000_check_for_link_vf(struct e1000_hw *hw); +static s32 e1000_get_link_up_info_vf(struct e1000_hw *hw, u16 *speed, + u16 *duplex); +static s32 e1000_init_hw_vf(struct e1000_hw *hw); +static s32 e1000_reset_hw_vf(struct e1000_hw *hw); + +static void e1000_update_mc_addr_list_vf(struct e1000_hw *hw, u8 *, + u32, u32, u32); +static void e1000_rar_set_vf(struct e1000_hw *, u8 *, u32); +static s32 e1000_read_mac_addr_vf(struct e1000_hw *); +static s32 e1000_set_vfta_vf(struct e1000_hw *, u16, bool); + +/** + * e1000_init_mac_params_vf - Inits MAC params + * @hw: pointer to the HW structure + **/ +s32 e1000_init_mac_params_vf(struct e1000_hw *hw) +{ + struct e1000_mac_info *mac = &hw->mac; + + /* VF's have no MTA Registers - PF feature only */ + mac->mta_reg_count = 128; + /* VF's have no access to RAR entries */ + mac->rar_entry_count = 1; + + /* Function pointers */ + /* reset */ + mac->ops.reset_hw = e1000_reset_hw_vf; + /* hw initialization */ + mac->ops.init_hw = e1000_init_hw_vf; + /* check for link */ + mac->ops.check_for_link = e1000_check_for_link_vf; + /* link info */ + mac->ops.get_link_up_info = e1000_get_link_up_info_vf; + /* multicast address update */ + mac->ops.update_mc_addr_list = e1000_update_mc_addr_list_vf; + /* set mac address */ + mac->ops.rar_set = e1000_rar_set_vf; + /* read mac address */ + mac->ops.read_mac_addr = e1000_read_mac_addr_vf; + /* set vlan filter table array */ + mac->ops.set_vfta = e1000_set_vfta_vf; + + return E1000_SUCCESS; +} + +/** + * e1000_init_function_pointers_vf - Inits function pointers + * @hw: pointer to the HW structure + **/ +void e1000_init_function_pointers_vf(struct e1000_hw *hw) +{ + hw->mac.ops.init_params = e1000_init_mac_params_vf; + hw->mbx.ops.init_params = e1000_init_mbx_params_vf; +} + +/** + * e1000_get_link_up_info_vf - Gets link info. + * @hw: pointer to the HW structure + * @speed: pointer to 16 bit value to store link speed. + * @duplex: pointer to 16 bit value to store duplex. + * + * Since we cannot read the PHY and get accurate link info, we must rely upon + * the status register's data which is often stale and inaccurate. + **/ +static s32 e1000_get_link_up_info_vf(struct e1000_hw *hw, u16 *speed, + u16 *duplex) +{ + s32 status; + + status = er32(STATUS); + if (status & E1000_STATUS_SPEED_1000) + *speed = SPEED_1000; + else if (status & E1000_STATUS_SPEED_100) + *speed = SPEED_100; + else + *speed = SPEED_10; + + if (status & E1000_STATUS_FD) + *duplex = FULL_DUPLEX; + else + *duplex = HALF_DUPLEX; + + return E1000_SUCCESS; +} + +/** + * e1000_reset_hw_vf - Resets the HW + * @hw: pointer to the HW structure + * + * VF's provide a function level reset. This is done using bit 26 of ctrl_reg. + * This is all the reset we can perform on a VF. + **/ +static s32 e1000_reset_hw_vf(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 timeout = E1000_VF_INIT_TIMEOUT; + u32 ret_val = -E1000_ERR_MAC_INIT; + u32 msgbuf[3]; + u8 *addr = (u8 *)(&msgbuf[1]); + u32 ctrl; + + /* assert vf queue/interrupt reset */ + ctrl = er32(CTRL); + ew32(CTRL, ctrl | E1000_CTRL_RST); + + /* we cannot initialize while the RSTI / RSTD bits are asserted */ + while (!mbx->ops.check_for_rst(hw) && timeout) { + timeout--; + udelay(5); + } + + if (timeout) { + /* mailbox timeout can now become active */ + mbx->timeout = E1000_VF_MBX_INIT_TIMEOUT; + + /* notify pf of vf reset completion */ + msgbuf[0] = E1000_VF_RESET; + mbx->ops.write_posted(hw, msgbuf, 1); + + msleep(10); + + /* set our "perm_addr" based on info provided by PF */ + ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + if (!ret_val) { + if (msgbuf[0] == (E1000_VF_RESET | E1000_VT_MSGTYPE_ACK)) + memcpy(hw->mac.perm_addr, addr, 6); + else + ret_val = -E1000_ERR_MAC_INIT; + } + } + + return ret_val; +} + +/** + * e1000_init_hw_vf - Inits the HW + * @hw: pointer to the HW structure + * + * Not much to do here except clear the PF Reset indication if there is one. + **/ +static s32 e1000_init_hw_vf(struct e1000_hw *hw) +{ + /* attempt to set and restore our mac address */ + e1000_rar_set_vf(hw, hw->mac.addr, 0); + + return E1000_SUCCESS; +} + +/** + * e1000_hash_mc_addr_vf - Generate a multicast hash value + * @hw: pointer to the HW structure + * @mc_addr: pointer to a multicast address + * + * Generates a multicast address hash value which is used to determine + * the multicast filter table array address and new table value. See + * e1000_mta_set_generic() + **/ +static u32 e1000_hash_mc_addr_vf(struct e1000_hw *hw, u8 *mc_addr) +{ + u32 hash_value, hash_mask; + u8 bit_shift = 0; + + /* Register count multiplied by bits per register */ + hash_mask = (hw->mac.mta_reg_count * 32) - 1; + + /* + * The bit_shift is the number of left-shifts + * where 0xFF would still fall within the hash mask. + */ + while (hash_mask >> bit_shift != 0xFF) + bit_shift++; + + hash_value = hash_mask & (((mc_addr[4] >> (8 - bit_shift)) | + (((u16) mc_addr[5]) << bit_shift))); + + return hash_value; +} + +/** + * e1000_update_mc_addr_list_vf - Update Multicast addresses + * @hw: pointer to the HW structure + * @mc_addr_list: array of multicast addresses to program + * @mc_addr_count: number of multicast addresses to program + * @rar_used_count: the first RAR register free to program + * @rar_count: total number of supported Receive Address Registers + * + * Updates the Receive Address Registers and Multicast Table Array. + * The caller must have a packed mc_addr_list of multicast addresses. + * The parameter rar_count will usually be hw->mac.rar_entry_count + * unless there are workarounds that change this. + **/ +void e1000_update_mc_addr_list_vf(struct e1000_hw *hw, + u8 *mc_addr_list, u32 mc_addr_count, + u32 rar_used_count, u32 rar_count) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[E1000_VFMAILBOX_SIZE]; + u16 *hash_list = (u16 *)&msgbuf[1]; + u32 hash_value; + u32 cnt, i; + + /* Each entry in the list uses 1 16 bit word. We have 30 + * 16 bit words available in our HW msg buffer (minus 1 for the + * msg type). That's 30 hash values if we pack 'em right. If + * there are more than 30 MC addresses to add then punt the + * extras for now and then add code to handle more than 30 later. + * It would be unusual for a server to request that many multi-cast + * addresses except for in large enterprise network environments. + */ + + cnt = (mc_addr_count > 30) ? 30 : mc_addr_count; + msgbuf[0] = E1000_VF_SET_MULTICAST; + msgbuf[0] |= cnt << E1000_VT_MSGINFO_SHIFT; + + for (i = 0; i < cnt; i++) { + hash_value = e1000_hash_mc_addr_vf(hw, mc_addr_list); + hash_list[i] = hash_value & 0x0FFFF; + mc_addr_list += ETH_ADDR_LEN; + } + + mbx->ops.write_posted(hw, msgbuf, E1000_VFMAILBOX_SIZE); +} + +/** + * e1000_set_vfta_vf - Set/Unset vlan filter table address + * @hw: pointer to the HW structure + * @vid: determines the vfta register and bit to set/unset + * @set: if true then set bit, else clear bit + **/ +static s32 e1000_set_vfta_vf(struct e1000_hw *hw, u16 vid, bool set) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[2]; + s32 err; + + msgbuf[0] = E1000_VF_SET_VLAN; + msgbuf[1] = vid; + /* Setting the 8 bit field MSG INFO to true indicates "add" */ + if (set) + msgbuf[0] |= 1 << E1000_VT_MSGINFO_SHIFT; + + mbx->ops.write_posted(hw, msgbuf, 2); + + err = mbx->ops.read_posted(hw, msgbuf, 2); + + /* if nacked the vlan was rejected */ + if (!err && (msgbuf[0] == (E1000_VF_SET_VLAN | E1000_VT_MSGTYPE_NACK))) + err = -E1000_ERR_MAC_INIT; + + return err; +} + +/** e1000_rlpml_set_vf - Set the maximum receive packet length + * @hw: pointer to the HW structure + * @max_size: value to assign to max frame size + **/ +void e1000_rlpml_set_vf(struct e1000_hw *hw, u16 max_size) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[2]; + + msgbuf[0] = E1000_VF_SET_LPE; + msgbuf[1] = max_size; + + mbx->ops.write_posted(hw, msgbuf, 2); +} + +/** + * e1000_rar_set_vf - set device MAC address + * @hw: pointer to the HW structure + * @addr: pointer to the receive address + * @index receive address array register + **/ +static void e1000_rar_set_vf(struct e1000_hw *hw, u8 * addr, u32 index) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + u32 msgbuf[3]; + u8 *msg_addr = (u8 *)(&msgbuf[1]); + s32 ret_val; + + memset(msgbuf, 0, 12); + msgbuf[0] = E1000_VF_SET_MAC_ADDR; + memcpy(msg_addr, addr, 6); + ret_val = mbx->ops.write_posted(hw, msgbuf, 3); + + if (!ret_val) + ret_val = mbx->ops.read_posted(hw, msgbuf, 3); + + /* if nacked the address was rejected, use "perm_addr" */ + if (!ret_val && + (msgbuf[0] == (E1000_VF_SET_MAC_ADDR | E1000_VT_MSGTYPE_NACK))) + e1000_read_mac_addr_vf(hw); +} + +/** + * e1000_read_mac_addr_vf - Read device MAC address + * @hw: pointer to the HW structure + **/ +static s32 e1000_read_mac_addr_vf(struct e1000_hw *hw) +{ + int i; + + for (i = 0; i < ETH_ADDR_LEN; i++) + hw->mac.addr[i] = hw->mac.perm_addr[i]; + + return E1000_SUCCESS; +} + +/** + * e1000_check_for_link_vf - Check for link for a virtual interface + * @hw: pointer to the HW structure + * + * Checks to see if the underlying PF is still talking to the VF and + * if it is then it reports the link state to the hardware, otherwise + * it reports link down and returns an error. + **/ +static s32 e1000_check_for_link_vf(struct e1000_hw *hw) +{ + struct e1000_mbx_info *mbx = &hw->mbx; + struct e1000_mac_info *mac = &hw->mac; + s32 ret_val = E1000_SUCCESS; + u32 in_msg = 0; + + /* + * We only want to run this if there has been a rst asserted. + * in this case that could mean a link change, device reset, + * or a virtual function reset + */ + + /* If we were hit with a reset drop the link */ + if (!mbx->ops.check_for_rst(hw)) + mac->get_link_status = true; + + if (!mac->get_link_status) + goto out; + + /* if link status is down no point in checking to see if pf is up */ + if (!(er32(STATUS) & E1000_STATUS_LU)) + goto out; + + /* if the read failed it could just be a mailbox collision, best wait + * until we are called again and don't report an error */ + if (mbx->ops.read(hw, &in_msg, 1)) + goto out; + + /* if incoming message isn't clear to send we are waiting on response */ + if (!(in_msg & E1000_VT_MSGTYPE_CTS)) { + /* message is not CTS and is NACK we must have lost CTS status */ + if (in_msg & E1000_VT_MSGTYPE_NACK) + ret_val = -E1000_ERR_MAC_INIT; + goto out; + } + + /* the pf is talking, if we timed out in the past we reinit */ + if (!mbx->timeout) { + ret_val = -E1000_ERR_MAC_INIT; + goto out; + } + + /* if we passed all the tests above then the link is up and we no + * longer need to check for link */ + mac->get_link_status = false; + +out: + return ret_val; +} + diff --git a/drivers/net/igbvf/vf.h b/drivers/net/igbvf/vf.h new file mode 100644 index 00000000000..ec07228f947 --- /dev/null +++ b/drivers/net/igbvf/vf.h @@ -0,0 +1,265 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 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 _E1000_VF_H_ +#define _E1000_VF_H_ + +#include <linux/pci.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/if_ether.h> + +#include "regs.h" +#include "defines.h" + +struct e1000_hw; + +#define E1000_DEV_ID_82576_VF 0x10CA +#define E1000_REVISION_0 0 +#define E1000_REVISION_1 1 +#define E1000_REVISION_2 2 +#define E1000_REVISION_3 3 +#define E1000_REVISION_4 4 + +#define E1000_FUNC_0 0 +#define E1000_FUNC_1 1 + +/* + * Receive Address Register Count + * Number of high/low register pairs in the RAR. The RAR (Receive Address + * Registers) holds the directed and multicast addresses that we monitor. + * These entries are also used for MAC-based filtering. + */ +#define E1000_RAR_ENTRIES_VF 1 + +/* Receive Descriptor - Advanced */ +union e1000_adv_rx_desc { + struct { + u64 pkt_addr; /* Packet buffer address */ + u64 hdr_addr; /* Header buffer address */ + } read; + struct { + struct { + union { + u32 data; + struct { + u16 pkt_info; /* RSS/Packet type */ + u16 hdr_info; /* Split Header, + * hdr buffer length */ + } hs_rss; + } lo_dword; + union { + u32 rss; /* RSS Hash */ + struct { + u16 ip_id; /* IP id */ + u16 csum; /* Packet Checksum */ + } csum_ip; + } hi_dword; + } lower; + struct { + u32 status_error; /* ext status/error */ + u16 length; /* Packet length */ + u16 vlan; /* VLAN tag */ + } upper; + } wb; /* writeback */ +}; + +#define E1000_RXDADV_HDRBUFLEN_MASK 0x7FE0 +#define E1000_RXDADV_HDRBUFLEN_SHIFT 5 + +/* Transmit Descriptor - Advanced */ +union e1000_adv_tx_desc { + struct { + u64 buffer_addr; /* Address of descriptor's data buf */ + u32 cmd_type_len; + u32 olinfo_status; + } read; + struct { + u64 rsvd; /* Reserved */ + u32 nxtseq_seed; + u32 status; + } wb; +}; + +/* Adv Transmit Descriptor Config Masks */ +#define E1000_ADVTXD_DTYP_CTXT 0x00200000 /* Advanced Context Descriptor */ +#define E1000_ADVTXD_DTYP_DATA 0x00300000 /* Advanced Data Descriptor */ +#define E1000_ADVTXD_DCMD_EOP 0x01000000 /* End of Packet */ +#define E1000_ADVTXD_DCMD_IFCS 0x02000000 /* Insert FCS (Ethernet CRC) */ +#define E1000_ADVTXD_DCMD_RS 0x08000000 /* Report Status */ +#define E1000_ADVTXD_DCMD_DEXT 0x20000000 /* Descriptor extension (1=Adv) */ +#define E1000_ADVTXD_DCMD_VLE 0x40000000 /* VLAN pkt enable */ +#define E1000_ADVTXD_DCMD_TSE 0x80000000 /* TCP Seg enable */ +#define E1000_ADVTXD_PAYLEN_SHIFT 14 /* Adv desc PAYLEN shift */ + +/* Context descriptors */ +struct e1000_adv_tx_context_desc { + u32 vlan_macip_lens; + u32 seqnum_seed; + u32 type_tucmd_mlhl; + u32 mss_l4len_idx; +}; + +#define E1000_ADVTXD_MACLEN_SHIFT 9 /* Adv ctxt desc mac len shift */ +#define E1000_ADVTXD_TUCMD_IPV4 0x00000400 /* IP Packet Type: 1=IPv4 */ +#define E1000_ADVTXD_TUCMD_L4T_TCP 0x00000800 /* L4 Packet TYPE of TCP */ +#define E1000_ADVTXD_L4LEN_SHIFT 8 /* Adv ctxt L4LEN shift */ +#define E1000_ADVTXD_MSS_SHIFT 16 /* Adv ctxt MSS shift */ + +enum e1000_mac_type { + e1000_undefined = 0, + e1000_vfadapt, + e1000_num_macs /* List is 1-based, so subtract 1 for true count. */ +}; + +struct e1000_vf_stats { + u64 base_gprc; + u64 base_gptc; + u64 base_gorc; + u64 base_gotc; + u64 base_mprc; + u64 base_gotlbc; + u64 base_gptlbc; + u64 base_gorlbc; + u64 base_gprlbc; + + u32 last_gprc; + u32 last_gptc; + u32 last_gorc; + u32 last_gotc; + u32 last_mprc; + u32 last_gotlbc; + u32 last_gptlbc; + u32 last_gorlbc; + u32 last_gprlbc; + + u64 gprc; + u64 gptc; + u64 gorc; + u64 gotc; + u64 mprc; + u64 gotlbc; + u64 gptlbc; + u64 gorlbc; + u64 gprlbc; +}; + +#include "mbx.h" + +struct e1000_mac_operations { + /* Function pointers for the MAC. */ + s32 (*init_params)(struct e1000_hw *); + s32 (*check_for_link)(struct e1000_hw *); + void (*clear_vfta)(struct e1000_hw *); + s32 (*get_bus_info)(struct e1000_hw *); + s32 (*get_link_up_info)(struct e1000_hw *, u16 *, u16 *); + void (*update_mc_addr_list)(struct e1000_hw *, u8 *, u32, u32, u32); + s32 (*reset_hw)(struct e1000_hw *); + s32 (*init_hw)(struct e1000_hw *); + s32 (*setup_link)(struct e1000_hw *); + void (*write_vfta)(struct e1000_hw *, u32, u32); + void (*mta_set)(struct e1000_hw *, u32); + void (*rar_set)(struct e1000_hw *, u8*, u32); + s32 (*read_mac_addr)(struct e1000_hw *); + s32 (*set_vfta)(struct e1000_hw *, u16, bool); +}; + +struct e1000_mac_info { + struct e1000_mac_operations ops; + u8 addr[6]; + u8 perm_addr[6]; + + enum e1000_mac_type type; + + u16 mta_reg_count; + u16 rar_entry_count; + + bool get_link_status; +}; + +struct e1000_mbx_operations { + s32 (*init_params)(struct e1000_hw *hw); + s32 (*read)(struct e1000_hw *, u32 *, u16); + s32 (*write)(struct e1000_hw *, u32 *, u16); + s32 (*read_posted)(struct e1000_hw *, u32 *, u16); + s32 (*write_posted)(struct e1000_hw *, u32 *, u16); + s32 (*check_for_msg)(struct e1000_hw *); + s32 (*check_for_ack)(struct e1000_hw *); + s32 (*check_for_rst)(struct e1000_hw *); +}; + +struct e1000_mbx_stats { + u32 msgs_tx; + u32 msgs_rx; + + u32 acks; + u32 reqs; + u32 rsts; +}; + +struct e1000_mbx_info { + struct e1000_mbx_operations ops; + struct e1000_mbx_stats stats; + u32 timeout; + u32 usec_delay; + u16 size; +}; + +struct e1000_dev_spec_vf { + u32 vf_number; + u32 v2p_mailbox; +}; + +struct e1000_hw { + void *back; + + u8 __iomem *hw_addr; + u8 __iomem *flash_address; + unsigned long io_base; + + struct e1000_mac_info mac; + struct e1000_mbx_info mbx; + + union { + struct e1000_dev_spec_vf vf; + } dev_spec; + + u16 device_id; + u16 subsystem_vendor_id; + u16 subsystem_device_id; + u16 vendor_id; + + u8 revision_id; +}; + +/* These functions must be implemented by drivers */ +void e1000_rlpml_set_vf(struct e1000_hw *, u16); +void e1000_init_function_pointers_vf(struct e1000_hw *hw); +s32 e1000_init_mac_params_vf(struct e1000_hw *hw); + + +#endif /* _E1000_VF_H_ */ diff --git a/drivers/net/mlx4/port.c b/drivers/net/mlx4/port.c index 7cce3342ef8..606aa58afde 100644 --- a/drivers/net/mlx4/port.c +++ b/drivers/net/mlx4/port.c @@ -299,13 +299,14 @@ int mlx4_SET_PORT(struct mlx4_dev *dev, u8 port) struct mlx4_cmd_mailbox *mailbox; int err; + if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) + return 0; + mailbox = mlx4_alloc_cmd_mailbox(dev); if (IS_ERR(mailbox)) return PTR_ERR(mailbox); memset(mailbox->buf, 0, 256); - if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH) - return 0; ((__be32 *) mailbox->buf)[1] = dev->caps.ib_port_def_cap[port]; err = mlx4_cmd(dev, mailbox->dma, port, 0, MLX4_CMD_SET_PORT, diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index a56d9d2df73..b3185bf2c15 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -2274,8 +2274,6 @@ static void port_start(struct mv643xx_eth_private *mp) pscr |= FORCE_LINK_PASS; wrlp(mp, PORT_SERIAL_CONTROL, pscr); - wrlp(mp, SDMA_CONFIG, PORT_SDMA_CONFIG_DEFAULT_VALUE); - /* * Configure TX path and queues. */ @@ -2957,6 +2955,8 @@ static int mv643xx_eth_probe(struct platform_device *pdev) netif_carrier_off(dev); + wrlp(mp, SDMA_CONFIG, PORT_SDMA_CONFIG_DEFAULT_VALUE); + set_rx_coal(mp, 250); set_tx_coal(mp, 0); diff --git a/drivers/net/niu.c b/drivers/net/niu.c index 73cac6c78cb..2b1745328cf 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -4834,6 +4834,7 @@ static int niu_compute_rbr_cfig_b(struct rx_ring_info *rp, u64 *ret) { u64 val = 0; + *ret = 0; switch (rp->rbr_block_size) { case 4 * 1024: val |= (RBR_BLKSIZE_4K << RBR_CFIG_B_BLKSIZE_SHIFT); @@ -9542,7 +9543,7 @@ static struct niu_parent * __devinit niu_new_parent(struct niu *np, plat_dev = platform_device_register_simple("niu", niu_parent_index, NULL, 0); - if (!plat_dev) + if (IS_ERR(plat_dev)) return NULL; for (i = 0; attr_name(niu_parent_attributes[i]); i++) { diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c index 5e8540b6ffa..6f97b47d74a 100644 --- a/drivers/net/r6040.c +++ b/drivers/net/r6040.c @@ -160,6 +160,7 @@ MODULE_AUTHOR("Sten Wang <sten.wang@rdc.com.tw>," "Florian Fainelli <florian@openwrt.org>"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("RDC R6040 NAPI PCI FastEthernet driver"); +MODULE_VERSION(DRV_VERSION " " DRV_RELDATE); /* RX and TX interrupts that we handle */ #define RX_INTS (RX_FIFO_FULL | RX_NO_DESC | RX_FINISH) diff --git a/drivers/net/smsc911x.c b/drivers/net/smsc911x.c index 6da67812982..eb7db032a78 100644 --- a/drivers/net/smsc911x.c +++ b/drivers/net/smsc911x.c @@ -317,7 +317,7 @@ static int smsc911x_mii_read(struct mii_bus *bus, int phyaddr, int regidx) goto out; } - SMSC_WARNING(HW, "Timed out waiting for MII write to finish"); + SMSC_WARNING(HW, "Timed out waiting for MII read to finish"); reg = -EIO; out: diff --git a/drivers/platform/x86/fujitsu-laptop.c b/drivers/platform/x86/fujitsu-laptop.c index 45940f31fe9..218b9a16ac3 100644 --- a/drivers/platform/x86/fujitsu-laptop.c +++ b/drivers/platform/x86/fujitsu-laptop.c @@ -174,8 +174,7 @@ struct fujitsu_hotkey_t { static struct fujitsu_hotkey_t *fujitsu_hotkey; -static void acpi_fujitsu_hotkey_notify(acpi_handle handle, u32 event, - void *data); +static void acpi_fujitsu_hotkey_notify(struct acpi_device *device, u32 event); #ifdef CONFIG_LEDS_CLASS static enum led_brightness logolamp_get(struct led_classdev *cdev); @@ -203,7 +202,7 @@ struct led_classdev kblamps_led = { static u32 dbg_level = 0x03; #endif -static void acpi_fujitsu_notify(acpi_handle handle, u32 event, void *data); +static void acpi_fujitsu_notify(struct acpi_device *device, u32 event); /* Fujitsu ACPI interface function */ @@ -658,7 +657,6 @@ static struct dmi_system_id fujitsu_dmi_table[] = { static int acpi_fujitsu_add(struct acpi_device *device) { - acpi_status status; acpi_handle handle; int result = 0; int state = 0; @@ -673,20 +671,10 @@ static int acpi_fujitsu_add(struct acpi_device *device) sprintf(acpi_device_class(device), "%s", ACPI_FUJITSU_CLASS); device->driver_data = fujitsu; - status = acpi_install_notify_handler(device->handle, - ACPI_DEVICE_NOTIFY, - acpi_fujitsu_notify, fujitsu); - - if (ACPI_FAILURE(status)) { - printk(KERN_ERR "Error installing notify handler\n"); - error = -ENODEV; - goto err_stop; - } - fujitsu->input = input = input_allocate_device(); if (!input) { error = -ENOMEM; - goto err_uninstall_notify; + goto err_stop; } snprintf(fujitsu->phys, sizeof(fujitsu->phys), @@ -743,9 +731,6 @@ static int acpi_fujitsu_add(struct acpi_device *device) end: err_free_input_dev: input_free_device(input); -err_uninstall_notify: - acpi_remove_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, - acpi_fujitsu_notify); err_stop: return result; @@ -753,7 +738,6 @@ err_stop: static int acpi_fujitsu_remove(struct acpi_device *device, int type) { - acpi_status status; struct fujitsu_t *fujitsu = NULL; if (!device || !acpi_driver_data(device)) @@ -761,10 +745,6 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type) fujitsu = acpi_driver_data(device); - status = acpi_remove_notify_handler(fujitsu->acpi_handle, - ACPI_DEVICE_NOTIFY, - acpi_fujitsu_notify); - if (!device || !acpi_driver_data(device)) return -EINVAL; @@ -775,7 +755,7 @@ static int acpi_fujitsu_remove(struct acpi_device *device, int type) /* Brightness notify */ -static void acpi_fujitsu_notify(acpi_handle handle, u32 event, void *data) +static void acpi_fujitsu_notify(struct acpi_device *device, u32 event) { struct input_dev *input; int keycode; @@ -829,15 +809,12 @@ static void acpi_fujitsu_notify(acpi_handle handle, u32 event, void *data) input_report_key(input, keycode, 0); input_sync(input); } - - return; } /* ACPI device for hotkey handling */ static int acpi_fujitsu_hotkey_add(struct acpi_device *device) { - acpi_status status; acpi_handle handle; int result = 0; int state = 0; @@ -854,17 +831,6 @@ static int acpi_fujitsu_hotkey_add(struct acpi_device *device) sprintf(acpi_device_class(device), "%s", ACPI_FUJITSU_CLASS); device->driver_data = fujitsu_hotkey; - status = acpi_install_notify_handler(device->handle, - ACPI_DEVICE_NOTIFY, - acpi_fujitsu_hotkey_notify, - fujitsu_hotkey); - - if (ACPI_FAILURE(status)) { - printk(KERN_ERR "Error installing notify handler\n"); - error = -ENODEV; - goto err_stop; - } - /* kfifo */ spin_lock_init(&fujitsu_hotkey->fifo_lock); fujitsu_hotkey->fifo = @@ -879,7 +845,7 @@ static int acpi_fujitsu_hotkey_add(struct acpi_device *device) fujitsu_hotkey->input = input = input_allocate_device(); if (!input) { error = -ENOMEM; - goto err_uninstall_notify; + goto err_free_fifo; } snprintf(fujitsu_hotkey->phys, sizeof(fujitsu_hotkey->phys), @@ -975,9 +941,7 @@ static int acpi_fujitsu_hotkey_add(struct acpi_device *device) end: err_free_input_dev: input_free_device(input); -err_uninstall_notify: - acpi_remove_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, - acpi_fujitsu_hotkey_notify); +err_free_fifo: kfifo_free(fujitsu_hotkey->fifo); err_stop: @@ -986,7 +950,6 @@ err_stop: static int acpi_fujitsu_hotkey_remove(struct acpi_device *device, int type) { - acpi_status status; struct fujitsu_hotkey_t *fujitsu_hotkey = NULL; if (!device || !acpi_driver_data(device)) @@ -994,10 +957,6 @@ static int acpi_fujitsu_hotkey_remove(struct acpi_device *device, int type) fujitsu_hotkey = acpi_driver_data(device); - status = acpi_remove_notify_handler(fujitsu_hotkey->acpi_handle, - ACPI_DEVICE_NOTIFY, - acpi_fujitsu_hotkey_notify); - fujitsu_hotkey->acpi_handle = NULL; kfifo_free(fujitsu_hotkey->fifo); @@ -1005,8 +964,7 @@ static int acpi_fujitsu_hotkey_remove(struct acpi_device *device, int type) return 0; } -static void acpi_fujitsu_hotkey_notify(acpi_handle handle, u32 event, - void *data) +static void acpi_fujitsu_hotkey_notify(struct acpi_device *device, u32 event) { struct input_dev *input; int keycode, keycode_r; @@ -1089,8 +1047,6 @@ static void acpi_fujitsu_hotkey_notify(acpi_handle handle, u32 event, input_sync(input); break; } - - return; } /* Initialization */ @@ -1107,6 +1063,7 @@ static struct acpi_driver acpi_fujitsu_driver = { .ops = { .add = acpi_fujitsu_add, .remove = acpi_fujitsu_remove, + .notify = acpi_fujitsu_notify, }, }; @@ -1122,6 +1079,7 @@ static struct acpi_driver acpi_fujitsu_hotkey_driver = { .ops = { .add = acpi_fujitsu_hotkey_add, .remove = acpi_fujitsu_hotkey_remove, + .notify = acpi_fujitsu_hotkey_notify, }, }; diff --git a/drivers/platform/x86/panasonic-laptop.c b/drivers/platform/x86/panasonic-laptop.c index a5ce4bc202e..fe7cf0188ac 100644 --- a/drivers/platform/x86/panasonic-laptop.c +++ b/drivers/platform/x86/panasonic-laptop.c @@ -176,6 +176,7 @@ enum SINF_BITS { SINF_NUM_BATTERIES = 0, static int acpi_pcc_hotkey_add(struct acpi_device *device); static int acpi_pcc_hotkey_remove(struct acpi_device *device, int type); static int acpi_pcc_hotkey_resume(struct acpi_device *device); +static void acpi_pcc_hotkey_notify(struct acpi_device *device, u32 event); static const struct acpi_device_id pcc_device_ids[] = { { "MAT0012", 0}, @@ -194,6 +195,7 @@ static struct acpi_driver acpi_pcc_driver = { .add = acpi_pcc_hotkey_add, .remove = acpi_pcc_hotkey_remove, .resume = acpi_pcc_hotkey_resume, + .notify = acpi_pcc_hotkey_notify, }, }; @@ -271,7 +273,7 @@ static int acpi_pcc_retrieve_biosdata(struct pcc_acpi *pcc, u32 *sinf) union acpi_object *hkey = NULL; int i; - status = acpi_evaluate_object(pcc->handle, METHOD_HKEY_SINF, 0, + status = acpi_evaluate_object(pcc->handle, METHOD_HKEY_SINF, NULL, &buffer); if (ACPI_FAILURE(status)) { ACPI_DEBUG_PRINT((ACPI_DB_ERROR, @@ -527,9 +529,9 @@ static void acpi_pcc_generate_keyinput(struct pcc_acpi *pcc) return; } -static void acpi_pcc_hotkey_notify(acpi_handle handle, u32 event, void *data) +static void acpi_pcc_hotkey_notify(struct acpi_device *device, u32 event) { - struct pcc_acpi *pcc = (struct pcc_acpi *) data; + struct pcc_acpi *pcc = acpi_driver_data(device); switch (event) { case HKEY_NOTIFY: @@ -599,7 +601,6 @@ static int acpi_pcc_hotkey_resume(struct acpi_device *device) static int acpi_pcc_hotkey_add(struct acpi_device *device) { - acpi_status status; struct pcc_acpi *pcc; int num_sifr, result; @@ -640,22 +641,11 @@ static int acpi_pcc_hotkey_add(struct acpi_device *device) goto out_sinf; } - /* initialize hotkey input device */ - status = acpi_install_notify_handler(pcc->handle, ACPI_DEVICE_NOTIFY, - acpi_pcc_hotkey_notify, pcc); - - if (ACPI_FAILURE(status)) { - ACPI_DEBUG_PRINT((ACPI_DB_ERROR, - "Error installing notify handler\n")); - result = -ENODEV; - goto out_input; - } - /* initialize backlight */ pcc->backlight = backlight_device_register("panasonic", NULL, pcc, &pcc_backlight_ops); if (IS_ERR(pcc->backlight)) - goto out_notify; + goto out_input; if (!acpi_pcc_retrieve_biosdata(pcc, pcc->sinf)) { ACPI_DEBUG_PRINT((ACPI_DB_ERROR, @@ -680,9 +670,6 @@ static int acpi_pcc_hotkey_add(struct acpi_device *device) out_backlight: backlight_device_unregister(pcc->backlight); -out_notify: - acpi_remove_notify_handler(pcc->handle, ACPI_DEVICE_NOTIFY, - acpi_pcc_hotkey_notify); out_input: input_unregister_device(pcc->input_dev); /* no need to input_free_device() since core input API refcount and @@ -723,9 +710,6 @@ static int acpi_pcc_hotkey_remove(struct acpi_device *device, int type) backlight_device_unregister(pcc->backlight); - acpi_remove_notify_handler(pcc->handle, ACPI_DEVICE_NOTIFY, - acpi_pcc_hotkey_notify); - input_unregister_device(pcc->input_dev); /* no need to input_free_device() since core input API refcount and * free()s the device */ diff --git a/drivers/platform/x86/sony-laptop.c b/drivers/platform/x86/sony-laptop.c index a90ec5cb2f2..d3c92d777bd 100644 --- a/drivers/platform/x86/sony-laptop.c +++ b/drivers/platform/x86/sony-laptop.c @@ -914,7 +914,7 @@ static struct sony_nc_event sony_127_events[] = { /* * ACPI callbacks */ -static void sony_acpi_notify(acpi_handle handle, u32 event, void *data) +static void sony_nc_notify(struct acpi_device *device, u32 event) { u32 ev = event; @@ -933,7 +933,7 @@ static void sony_acpi_notify(acpi_handle handle, u32 event, void *data) struct sony_nc_event *key_event; if (sony_call_snc_handle(key_handle, 0x200, &result)) { - dprintk("sony_acpi_notify, unable to decode" + dprintk("sony_nc_notify, unable to decode" " event 0x%.2x 0x%.2x\n", key_handle, ev); /* restore the original event */ @@ -968,7 +968,7 @@ static void sony_acpi_notify(acpi_handle handle, u32 event, void *data) } else sony_laptop_report_input_event(ev); - dprintk("sony_acpi_notify, event: 0x%.2x\n", ev); + dprintk("sony_nc_notify, event: 0x%.2x\n", ev); acpi_bus_generate_proc_event(sony_nc_acpi_device, 1, ev); } @@ -1276,15 +1276,6 @@ static int sony_nc_add(struct acpi_device *device) goto outwalk; } - status = acpi_install_notify_handler(sony_nc_acpi_handle, - ACPI_DEVICE_NOTIFY, - sony_acpi_notify, NULL); - if (ACPI_FAILURE(status)) { - printk(KERN_WARNING DRV_PFX "unable to install notify handler (%u)\n", status); - result = -ENODEV; - goto outinput; - } - if (acpi_video_backlight_support()) { printk(KERN_INFO DRV_PFX "brightness ignored, must be " "controlled by ACPI video driver\n"); @@ -1362,13 +1353,6 @@ static int sony_nc_add(struct acpi_device *device) if (sony_backlight_device) backlight_device_unregister(sony_backlight_device); - status = acpi_remove_notify_handler(sony_nc_acpi_handle, - ACPI_DEVICE_NOTIFY, - sony_acpi_notify); - if (ACPI_FAILURE(status)) - printk(KERN_WARNING DRV_PFX "unable to remove notify handler\n"); - - outinput: sony_laptop_remove_input(); outwalk: @@ -1378,7 +1362,6 @@ static int sony_nc_add(struct acpi_device *device) static int sony_nc_remove(struct acpi_device *device, int type) { - acpi_status status; struct sony_nc_value *item; if (sony_backlight_device) @@ -1386,12 +1369,6 @@ static int sony_nc_remove(struct acpi_device *device, int type) sony_nc_acpi_device = NULL; - status = acpi_remove_notify_handler(sony_nc_acpi_handle, - ACPI_DEVICE_NOTIFY, - sony_acpi_notify); - if (ACPI_FAILURE(status)) - printk(KERN_WARNING DRV_PFX "unable to remove notify handler\n"); - for (item = sony_nc_values; item->name; ++item) { device_remove_file(&sony_pf_device->dev, &item->devattr); } @@ -1425,6 +1402,7 @@ static struct acpi_driver sony_nc_driver = { .add = sony_nc_add, .remove = sony_nc_remove, .resume = sony_nc_resume, + .notify = sony_nc_notify, }, }; diff --git a/drivers/platform/x86/wmi.c b/drivers/platform/x86/wmi.c index 2f269e117b8..043b208d971 100644 --- a/drivers/platform/x86/wmi.c +++ b/drivers/platform/x86/wmi.c @@ -81,6 +81,7 @@ static struct wmi_block wmi_blocks; static int acpi_wmi_remove(struct acpi_device *device, int type); static int acpi_wmi_add(struct acpi_device *device); +static void acpi_wmi_notify(struct acpi_device *device, u32 event); static const struct acpi_device_id wmi_device_ids[] = { {"PNP0C14", 0}, @@ -96,6 +97,7 @@ static struct acpi_driver acpi_wmi_driver = { .ops = { .add = acpi_wmi_add, .remove = acpi_wmi_remove, + .notify = acpi_wmi_notify, }, }; @@ -643,12 +645,11 @@ acpi_wmi_ec_space_handler(u32 function, acpi_physical_address address, } } -static void acpi_wmi_notify(acpi_handle handle, u32 event, void *data) +static void acpi_wmi_notify(struct acpi_device *device, u32 event) { struct guid_block *block; struct wmi_block *wblock; struct list_head *p; - struct acpi_device *device = data; list_for_each(p, &wmi_blocks.list) { wblock = list_entry(p, struct wmi_block, list); @@ -669,9 +670,6 @@ static void acpi_wmi_notify(acpi_handle handle, u32 event, void *data) static int acpi_wmi_remove(struct acpi_device *device, int type) { - acpi_remove_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, - acpi_wmi_notify); - acpi_remove_address_space_handler(device->handle, ACPI_ADR_SPACE_EC, &acpi_wmi_ec_space_handler); @@ -683,13 +681,6 @@ static int __init acpi_wmi_add(struct acpi_device *device) acpi_status status; int result = 0; - status = acpi_install_notify_handler(device->handle, ACPI_DEVICE_NOTIFY, - acpi_wmi_notify, device); - if (ACPI_FAILURE(status)) { - printk(KERN_ERR PREFIX "Error installing notify handler\n"); - return -ENODEV; - } - status = acpi_install_address_space_handler(device->handle, ACPI_ADR_SPACE_EC, &acpi_wmi_ec_space_handler, diff --git a/drivers/power/pcf50633-charger.c b/drivers/power/pcf50633-charger.c index 41aec2acbb9..e8b278f7178 100644 --- a/drivers/power/pcf50633-charger.c +++ b/drivers/power/pcf50633-charger.c @@ -36,6 +36,8 @@ struct pcf50633_mbc { struct power_supply usb; struct power_supply adapter; + + struct delayed_work charging_restart_work; }; int pcf50633_mbc_usb_curlim_set(struct pcf50633 *pcf, int ma) @@ -43,6 +45,8 @@ int pcf50633_mbc_usb_curlim_set(struct pcf50633 *pcf, int ma) struct pcf50633_mbc *mbc = platform_get_drvdata(pcf->mbc_pdev); int ret = 0; u8 bits; + int charging_start = 1; + u8 mbcs2, chgmod; if (ma >= 1000) bits = PCF50633_MBCC7_USB_1000mA; @@ -50,8 +54,10 @@ int pcf50633_mbc_usb_curlim_set(struct pcf50633 *pcf, int ma) bits = PCF50633_MBCC7_USB_500mA; else if (ma >= 100) bits = PCF50633_MBCC7_USB_100mA; - else + else { bits = PCF50633_MBCC7_USB_SUSPEND; + charging_start = 0; + } ret = pcf50633_reg_set_bit_mask(pcf, PCF50633_REG_MBCC7, PCF50633_MBCC7_USB_MASK, bits); @@ -60,6 +66,22 @@ int pcf50633_mbc_usb_curlim_set(struct pcf50633 *pcf, int ma) else dev_info(pcf->dev, "usb curlim to %d mA\n", ma); + /* Manual charging start */ + mbcs2 = pcf50633_reg_read(pcf, PCF50633_REG_MBCS2); + chgmod = (mbcs2 & PCF50633_MBCS2_MBC_MASK); + + /* If chgmod == BATFULL, setting chgena has no effect. + * We need to set resume instead. + */ + if (chgmod != PCF50633_MBCS2_MBC_BAT_FULL) + pcf50633_reg_set_bit_mask(pcf, PCF50633_REG_MBCC1, + PCF50633_MBCC1_CHGENA, PCF50633_MBCC1_CHGENA); + else + pcf50633_reg_set_bit_mask(pcf, PCF50633_REG_MBCC1, + PCF50633_MBCC1_RESUME, PCF50633_MBCC1_RESUME); + + mbc->usb_active = charging_start; + power_supply_changed(&mbc->usb); return ret; @@ -84,21 +106,6 @@ int pcf50633_mbc_get_status(struct pcf50633 *pcf) } EXPORT_SYMBOL_GPL(pcf50633_mbc_get_status); -void pcf50633_mbc_set_status(struct pcf50633 *pcf, int what, int status) -{ - struct pcf50633_mbc *mbc = platform_get_drvdata(pcf->mbc_pdev); - - if (what & PCF50633_MBC_USB_ONLINE) - mbc->usb_online = !!status; - if (what & PCF50633_MBC_USB_ACTIVE) - mbc->usb_active = !!status; - if (what & PCF50633_MBC_ADAPTER_ONLINE) - mbc->adapter_online = !!status; - if (what & PCF50633_MBC_ADAPTER_ACTIVE) - mbc->adapter_active = !!status; -} -EXPORT_SYMBOL_GPL(pcf50633_mbc_set_status); - static ssize_t show_chgmode(struct device *dev, struct device_attribute *attr, char *buf) { @@ -160,10 +167,44 @@ static struct attribute_group mbc_attr_group = { .attrs = pcf50633_mbc_sysfs_entries, }; +/* MBC state machine switches into charging mode when the battery voltage + * falls below 96% of a battery float voltage. But the voltage drop in Li-ion + * batteries is marginal(1~2 %) till about 80% of its capacity - which means, + * after a BATFULL, charging won't be restarted until 80%. + * + * This work_struct function restarts charging at regular intervals to make + * sure we don't discharge too much + */ + +static void pcf50633_mbc_charging_restart(struct work_struct *work) +{ + struct pcf50633_mbc *mbc; + u8 mbcs2, chgmod; + + mbc = container_of(work, struct pcf50633_mbc, + charging_restart_work.work); + + mbcs2 = pcf50633_reg_read(mbc->pcf, PCF50633_REG_MBCS2); + chgmod = (mbcs2 & PCF50633_MBCS2_MBC_MASK); + + if (chgmod != PCF50633_MBCS2_MBC_BAT_FULL) + return; + + /* Restart charging */ + pcf50633_reg_set_bit_mask(mbc->pcf, PCF50633_REG_MBCC1, + PCF50633_MBCC1_RESUME, PCF50633_MBCC1_RESUME); + mbc->usb_active = 1; + power_supply_changed(&mbc->usb); + + dev_info(mbc->pcf->dev, "Charging restarted\n"); +} + static void pcf50633_mbc_irq_handler(int irq, void *data) { struct pcf50633_mbc *mbc = data; + int chg_restart_interval = + mbc->pcf->pdata->charging_restart_interval; /* USB */ if (irq == PCF50633_IRQ_USBINS) { @@ -172,6 +213,7 @@ pcf50633_mbc_irq_handler(int irq, void *data) mbc->usb_online = 0; mbc->usb_active = 0; pcf50633_mbc_usb_curlim_set(mbc->pcf, 0); + cancel_delayed_work_sync(&mbc->charging_restart_work); } /* Adapter */ @@ -186,7 +228,14 @@ pcf50633_mbc_irq_handler(int irq, void *data) if (irq == PCF50633_IRQ_BATFULL) { mbc->usb_active = 0; mbc->adapter_active = 0; - } + + if (chg_restart_interval > 0) + schedule_delayed_work(&mbc->charging_restart_work, + chg_restart_interval); + } else if (irq == PCF50633_IRQ_USBLIMON) + mbc->usb_active = 0; + else if (irq == PCF50633_IRQ_USBLIMOFF) + mbc->usb_active = 1; power_supply_changed(&mbc->usb); power_supply_changed(&mbc->adapter); @@ -303,6 +352,9 @@ static int __devinit pcf50633_mbc_probe(struct platform_device *pdev) return ret; } + INIT_DELAYED_WORK(&mbc->charging_restart_work, + pcf50633_mbc_charging_restart); + ret = sysfs_create_group(&pdev->dev.kobj, &mbc_attr_group); if (ret) dev_err(mbc->pcf->dev, "failed to create sysfs entries\n"); @@ -328,6 +380,8 @@ static int __devexit pcf50633_mbc_remove(struct platform_device *pdev) power_supply_unregister(&mbc->usb); power_supply_unregister(&mbc->adapter); + cancel_delayed_work_sync(&mbc->charging_restart_work); + kfree(mbc); return 0; diff --git a/drivers/power/pda_power.c b/drivers/power/pda_power.c index b56a704409d..a232de6a570 100644 --- a/drivers/power/pda_power.c +++ b/drivers/power/pda_power.c @@ -12,11 +12,14 @@ #include <linux/module.h> #include <linux/platform_device.h> +#include <linux/err.h> #include <linux/interrupt.h> #include <linux/power_supply.h> #include <linux/pda_power.h> +#include <linux/regulator/consumer.h> #include <linux/timer.h> #include <linux/jiffies.h> +#include <linux/usb/otg.h> static inline unsigned int get_irq_flags(struct resource *res) { @@ -35,6 +38,11 @@ static struct timer_list supply_timer; static struct timer_list polling_timer; static int polling; +#ifdef CONFIG_USB_OTG_UTILS +static struct otg_transceiver *transceiver; +#endif +static struct regulator *ac_draw; + enum { PDA_PSY_OFFLINE = 0, PDA_PSY_ONLINE = 1, @@ -104,18 +112,35 @@ static void update_status(void) static void update_charger(void) { - if (!pdata->set_charge) - return; - - if (new_ac_status > 0) { - dev_dbg(dev, "charger on (AC)\n"); - pdata->set_charge(PDA_POWER_CHARGE_AC); - } else if (new_usb_status > 0) { - dev_dbg(dev, "charger on (USB)\n"); - pdata->set_charge(PDA_POWER_CHARGE_USB); - } else { - dev_dbg(dev, "charger off\n"); - pdata->set_charge(0); + static int regulator_enabled; + int max_uA = pdata->ac_max_uA; + + if (pdata->set_charge) { + if (new_ac_status > 0) { + dev_dbg(dev, "charger on (AC)\n"); + pdata->set_charge(PDA_POWER_CHARGE_AC); + } else if (new_usb_status > 0) { + dev_dbg(dev, "charger on (USB)\n"); + pdata->set_charge(PDA_POWER_CHARGE_USB); + } else { + dev_dbg(dev, "charger off\n"); + pdata->set_charge(0); + } + } else if (ac_draw) { + if (new_ac_status > 0) { + regulator_set_current_limit(ac_draw, max_uA, max_uA); + if (!regulator_enabled) { + dev_dbg(dev, "charger on (AC)\n"); + regulator_enable(ac_draw); + regulator_enabled = 1; + } + } else { + if (regulator_enabled) { + dev_dbg(dev, "charger off\n"); + regulator_disable(ac_draw); + regulator_enabled = 0; + } + } } } @@ -194,6 +219,13 @@ static void polling_timer_func(unsigned long unused) jiffies + msecs_to_jiffies(pdata->polling_interval)); } +#ifdef CONFIG_USB_OTG_UTILS +static int otg_is_usb_online(void) +{ + return (transceiver->state == OTG_STATE_B_PERIPHERAL); +} +#endif + static int pda_power_probe(struct platform_device *pdev) { int ret = 0; @@ -227,6 +259,9 @@ static int pda_power_probe(struct platform_device *pdev) if (!pdata->polling_interval) pdata->polling_interval = 2000; + if (!pdata->ac_max_uA) + pdata->ac_max_uA = 500000; + setup_timer(&charger_timer, charger_timer_func, 0); setup_timer(&supply_timer, supply_timer_func, 0); @@ -240,6 +275,13 @@ static int pda_power_probe(struct platform_device *pdev) pda_psy_usb.num_supplicants = pdata->num_supplicants; } + ac_draw = regulator_get(dev, "ac_draw"); + if (IS_ERR(ac_draw)) { + dev_dbg(dev, "couldn't get ac_draw regulator\n"); + ac_draw = NULL; + ret = PTR_ERR(ac_draw); + } + if (pdata->is_ac_online) { ret = power_supply_register(&pdev->dev, &pda_psy_ac); if (ret) { @@ -261,6 +303,13 @@ static int pda_power_probe(struct platform_device *pdev) } } +#ifdef CONFIG_USB_OTG_UTILS + transceiver = otg_get_transceiver(); + if (transceiver && !pdata->is_usb_online) { + pdata->is_usb_online = otg_is_usb_online; + } +#endif + if (pdata->is_usb_online) { ret = power_supply_register(&pdev->dev, &pda_psy_usb); if (ret) { @@ -300,10 +349,18 @@ usb_irq_failed: usb_supply_failed: if (pdata->is_ac_online && ac_irq) free_irq(ac_irq->start, &pda_psy_ac); +#ifdef CONFIG_USB_OTG_UTILS + if (transceiver) + otg_put_transceiver(transceiver); +#endif ac_irq_failed: if (pdata->is_ac_online) power_supply_unregister(&pda_psy_ac); ac_supply_failed: + if (ac_draw) { + regulator_put(ac_draw); + ac_draw = NULL; + } if (pdata->exit) pdata->exit(dev); init_failed: @@ -327,6 +384,14 @@ static int pda_power_remove(struct platform_device *pdev) power_supply_unregister(&pda_psy_usb); if (pdata->is_ac_online) power_supply_unregister(&pda_psy_ac); +#ifdef CONFIG_USB_OTG_UTILS + if (transceiver) + otg_put_transceiver(transceiver); +#endif + if (ac_draw) { + regulator_put(ac_draw); + ac_draw = NULL; + } if (pdata->exit) pdata->exit(dev); diff --git a/drivers/serial/max3100.c b/drivers/serial/max3100.c new file mode 100644 index 00000000000..9fd33e5622b --- /dev/null +++ b/drivers/serial/max3100.c @@ -0,0 +1,927 @@ +/* + * + * Copyright (C) 2008 Christian Pellegrin <chripell@evolware.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * + * Notes: the MAX3100 doesn't provide an interrupt on CTS so we have + * to use polling for flow control. TX empty IRQ is unusable, since + * writing conf clears FIFO buffer and we cannot have this interrupt + * always asking us for attention. + * + * Example platform data: + + static struct plat_max3100 max3100_plat_data = { + .loopback = 0, + .crystal = 0, + .poll_time = 100, + }; + + static struct spi_board_info spi_board_info[] = { + { + .modalias = "max3100", + .platform_data = &max3100_plat_data, + .irq = IRQ_EINT12, + .max_speed_hz = 5*1000*1000, + .chip_select = 0, + }, + }; + + * The initial minor number is 209 in the low-density serial port: + * mknod /dev/ttyMAX0 c 204 209 + */ + +#define MAX3100_MAJOR 204 +#define MAX3100_MINOR 209 +/* 4 MAX3100s should be enough for everyone */ +#define MAX_MAX3100 4 + +#include <linux/delay.h> +#include <linux/device.h> +#include <linux/serial_core.h> +#include <linux/serial.h> +#include <linux/spi/spi.h> +#include <linux/freezer.h> + +#include <linux/serial_max3100.h> + +#define MAX3100_C (1<<14) +#define MAX3100_D (0<<14) +#define MAX3100_W (1<<15) +#define MAX3100_RX (0<<15) + +#define MAX3100_WC (MAX3100_W | MAX3100_C) +#define MAX3100_RC (MAX3100_RX | MAX3100_C) +#define MAX3100_WD (MAX3100_W | MAX3100_D) +#define MAX3100_RD (MAX3100_RX | MAX3100_D) +#define MAX3100_CMD (3 << 14) + +#define MAX3100_T (1<<14) +#define MAX3100_R (1<<15) + +#define MAX3100_FEN (1<<13) +#define MAX3100_SHDN (1<<12) +#define MAX3100_TM (1<<11) +#define MAX3100_RM (1<<10) +#define MAX3100_PM (1<<9) +#define MAX3100_RAM (1<<8) +#define MAX3100_IR (1<<7) +#define MAX3100_ST (1<<6) +#define MAX3100_PE (1<<5) +#define MAX3100_L (1<<4) +#define MAX3100_BAUD (0xf) + +#define MAX3100_TE (1<<10) +#define MAX3100_RAFE (1<<10) +#define MAX3100_RTS (1<<9) +#define MAX3100_CTS (1<<9) +#define MAX3100_PT (1<<8) +#define MAX3100_DATA (0xff) + +#define MAX3100_RT (MAX3100_R | MAX3100_T) +#define MAX3100_RTC (MAX3100_RT | MAX3100_CTS | MAX3100_RAFE) + +/* the following simulate a status reg for ignore_status_mask */ +#define MAX3100_STATUS_PE 1 +#define MAX3100_STATUS_FE 2 +#define MAX3100_STATUS_OE 4 + +struct max3100_port { + struct uart_port port; + struct spi_device *spi; + + int cts; /* last CTS received for flow ctrl */ + int tx_empty; /* last TX empty bit */ + + spinlock_t conf_lock; /* shared data */ + int conf_commit; /* need to make changes */ + int conf; /* configuration for the MAX31000 + * (bits 0-7, bits 8-11 are irqs) */ + int rts_commit; /* need to change rts */ + int rts; /* rts status */ + int baud; /* current baud rate */ + + int parity; /* keeps track if we should send parity */ +#define MAX3100_PARITY_ON 1 +#define MAX3100_PARITY_ODD 2 +#define MAX3100_7BIT 4 + int rx_enabled; /* if we should rx chars */ + + int irq; /* irq assigned to the max3100 */ + + int minor; /* minor number */ + int crystal; /* 1 if 3.6864Mhz crystal 0 for 1.8432 */ + int loopback; /* 1 if we are in loopback mode */ + + /* for handling irqs: need workqueue since we do spi_sync */ + struct workqueue_struct *workqueue; + struct work_struct work; + /* set to 1 to make the workhandler exit as soon as possible */ + int force_end_work; + /* need to know we are suspending to avoid deadlock on workqueue */ + int suspending; + + /* hook for suspending MAX3100 via dedicated pin */ + void (*max3100_hw_suspend) (int suspend); + + /* poll time (in ms) for ctrl lines */ + int poll_time; + /* and its timer */ + struct timer_list timer; +}; + +static struct max3100_port *max3100s[MAX_MAX3100]; /* the chips */ +static DEFINE_MUTEX(max3100s_lock); /* race on probe */ + +static int max3100_do_parity(struct max3100_port *s, u16 c) +{ + int parity; + + if (s->parity & MAX3100_PARITY_ODD) + parity = 1; + else + parity = 0; + + if (s->parity & MAX3100_7BIT) + c &= 0x7f; + else + c &= 0xff; + + parity = parity ^ (hweight8(c) & 1); + return parity; +} + +static int max3100_check_parity(struct max3100_port *s, u16 c) +{ + return max3100_do_parity(s, c) == ((c >> 8) & 1); +} + +static void max3100_calc_parity(struct max3100_port *s, u16 *c) +{ + if (s->parity & MAX3100_7BIT) + *c &= 0x7f; + else + *c &= 0xff; + + if (s->parity & MAX3100_PARITY_ON) + *c |= max3100_do_parity(s, *c) << 8; +} + +static void max3100_work(struct work_struct *w); + +static void max3100_dowork(struct max3100_port *s) +{ + if (!s->force_end_work && !work_pending(&s->work) && + !freezing(current) && !s->suspending) + queue_work(s->workqueue, &s->work); +} + +static void max3100_timeout(unsigned long data) +{ + struct max3100_port *s = (struct max3100_port *)data; + + if (s->port.info) { + max3100_dowork(s); + mod_timer(&s->timer, jiffies + s->poll_time); + } +} + +static int max3100_sr(struct max3100_port *s, u16 tx, u16 *rx) +{ + struct spi_message message; + u16 etx, erx; + int status; + struct spi_transfer tran = { + .tx_buf = &etx, + .rx_buf = &erx, + .len = 2, + }; + + etx = cpu_to_be16(tx); + spi_message_init(&message); + spi_message_add_tail(&tran, &message); + status = spi_sync(s->spi, &message); + if (status) { + dev_warn(&s->spi->dev, "error while calling spi_sync\n"); + return -EIO; + } + *rx = be16_to_cpu(erx); + s->tx_empty = (*rx & MAX3100_T) > 0; + dev_dbg(&s->spi->dev, "%04x - %04x\n", tx, *rx); + return 0; +} + +static int max3100_handlerx(struct max3100_port *s, u16 rx) +{ + unsigned int ch, flg, status = 0; + int ret = 0, cts; + + if (rx & MAX3100_R && s->rx_enabled) { + dev_dbg(&s->spi->dev, "%s\n", __func__); + ch = rx & (s->parity & MAX3100_7BIT ? 0x7f : 0xff); + if (rx & MAX3100_RAFE) { + s->port.icount.frame++; + flg = TTY_FRAME; + status |= MAX3100_STATUS_FE; + } else { + if (s->parity & MAX3100_PARITY_ON) { + if (max3100_check_parity(s, rx)) { + s->port.icount.rx++; + flg = TTY_NORMAL; + } else { + s->port.icount.parity++; + flg = TTY_PARITY; + status |= MAX3100_STATUS_PE; + } + } else { + s->port.icount.rx++; + flg = TTY_NORMAL; + } + } + uart_insert_char(&s->port, status, MAX3100_STATUS_OE, ch, flg); + ret = 1; + } + + cts = (rx & MAX3100_CTS) > 0; + if (s->cts != cts) { + s->cts = cts; + uart_handle_cts_change(&s->port, cts ? TIOCM_CTS : 0); + } + + return ret; +} + +static void max3100_work(struct work_struct *w) +{ + struct max3100_port *s = container_of(w, struct max3100_port, work); + int rxchars; + u16 tx, rx; + int conf, cconf, rts, crts; + struct circ_buf *xmit = &s->port.info->xmit; + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + rxchars = 0; + do { + spin_lock(&s->conf_lock); + conf = s->conf; + cconf = s->conf_commit; + s->conf_commit = 0; + rts = s->rts; + crts = s->rts_commit; + s->rts_commit = 0; + spin_unlock(&s->conf_lock); + if (cconf) + max3100_sr(s, MAX3100_WC | conf, &rx); + if (crts) { + max3100_sr(s, MAX3100_WD | MAX3100_TE | + (s->rts ? MAX3100_RTS : 0), &rx); + rxchars += max3100_handlerx(s, rx); + } + + max3100_sr(s, MAX3100_RD, &rx); + rxchars += max3100_handlerx(s, rx); + + if (rx & MAX3100_T) { + tx = 0xffff; + if (s->port.x_char) { + tx = s->port.x_char; + s->port.icount.tx++; + s->port.x_char = 0; + } else if (!uart_circ_empty(xmit) && + !uart_tx_stopped(&s->port)) { + tx = xmit->buf[xmit->tail]; + xmit->tail = (xmit->tail + 1) & + (UART_XMIT_SIZE - 1); + s->port.icount.tx++; + } + if (tx != 0xffff) { + max3100_calc_parity(s, &tx); + tx |= MAX3100_WD | (s->rts ? MAX3100_RTS : 0); + max3100_sr(s, tx, &rx); + rxchars += max3100_handlerx(s, rx); + } + } + + if (rxchars > 16 && s->port.info->port.tty != NULL) { + tty_flip_buffer_push(s->port.info->port.tty); + rxchars = 0; + } + if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS) + uart_write_wakeup(&s->port); + + } while (!s->force_end_work && + !freezing(current) && + ((rx & MAX3100_R) || + (!uart_circ_empty(xmit) && + !uart_tx_stopped(&s->port)))); + + if (rxchars > 0 && s->port.info->port.tty != NULL) + tty_flip_buffer_push(s->port.info->port.tty); +} + +static irqreturn_t max3100_irq(int irqno, void *dev_id) +{ + struct max3100_port *s = dev_id; + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + max3100_dowork(s); + return IRQ_HANDLED; +} + +static void max3100_enable_ms(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + if (s->poll_time > 0) + mod_timer(&s->timer, jiffies); + dev_dbg(&s->spi->dev, "%s\n", __func__); +} + +static void max3100_start_tx(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + max3100_dowork(s); +} + +static void max3100_stop_rx(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + s->rx_enabled = 0; + spin_lock(&s->conf_lock); + s->conf &= ~MAX3100_RM; + s->conf_commit = 1; + spin_unlock(&s->conf_lock); + max3100_dowork(s); +} + +static unsigned int max3100_tx_empty(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + /* may not be truly up-to-date */ + max3100_dowork(s); + return s->tx_empty; +} + +static unsigned int max3100_get_mctrl(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + /* may not be truly up-to-date */ + max3100_dowork(s); + /* always assert DCD and DSR since these lines are not wired */ + return (s->cts ? TIOCM_CTS : 0) | TIOCM_DSR | TIOCM_CAR; +} + +static void max3100_set_mctrl(struct uart_port *port, unsigned int mctrl) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + int rts; + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + rts = (mctrl & TIOCM_RTS) > 0; + + spin_lock(&s->conf_lock); + if (s->rts != rts) { + s->rts = rts; + s->rts_commit = 1; + max3100_dowork(s); + } + spin_unlock(&s->conf_lock); +} + +static void +max3100_set_termios(struct uart_port *port, struct ktermios *termios, + struct ktermios *old) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + int baud = 0; + unsigned cflag; + u32 param_new, param_mask, parity = 0; + struct tty_struct *tty = s->port.info->port.tty; + + dev_dbg(&s->spi->dev, "%s\n", __func__); + if (!tty) + return; + + cflag = termios->c_cflag; + param_new = 0; + param_mask = 0; + + baud = tty_get_baud_rate(tty); + param_new = s->conf & MAX3100_BAUD; + switch (baud) { + case 300: + if (s->crystal) + baud = s->baud; + else + param_new = 15; + break; + case 600: + param_new = 14 + s->crystal; + break; + case 1200: + param_new = 13 + s->crystal; + break; + case 2400: + param_new = 12 + s->crystal; + break; + case 4800: + param_new = 11 + s->crystal; + break; + case 9600: + param_new = 10 + s->crystal; + break; + case 19200: + param_new = 9 + s->crystal; + break; + case 38400: + param_new = 8 + s->crystal; + break; + case 57600: + param_new = 1 + s->crystal; + break; + case 115200: + param_new = 0 + s->crystal; + break; + case 230400: + if (s->crystal) + param_new = 0; + else + baud = s->baud; + break; + default: + baud = s->baud; + } + tty_encode_baud_rate(tty, baud, baud); + s->baud = baud; + param_mask |= MAX3100_BAUD; + + if ((cflag & CSIZE) == CS8) { + param_new &= ~MAX3100_L; + parity &= ~MAX3100_7BIT; + } else { + param_new |= MAX3100_L; + parity |= MAX3100_7BIT; + cflag = (cflag & ~CSIZE) | CS7; + } + param_mask |= MAX3100_L; + + if (cflag & CSTOPB) + param_new |= MAX3100_ST; + else + param_new &= ~MAX3100_ST; + param_mask |= MAX3100_ST; + + if (cflag & PARENB) { + param_new |= MAX3100_PE; + parity |= MAX3100_PARITY_ON; + } else { + param_new &= ~MAX3100_PE; + parity &= ~MAX3100_PARITY_ON; + } + param_mask |= MAX3100_PE; + + if (cflag & PARODD) + parity |= MAX3100_PARITY_ODD; + else + parity &= ~MAX3100_PARITY_ODD; + + /* mask termios capabilities we don't support */ + cflag &= ~CMSPAR; + termios->c_cflag = cflag; + + s->port.ignore_status_mask = 0; + if (termios->c_iflag & IGNPAR) + s->port.ignore_status_mask |= + MAX3100_STATUS_PE | MAX3100_STATUS_FE | + MAX3100_STATUS_OE; + + /* we are sending char from a workqueue so enable */ + s->port.info->port.tty->low_latency = 1; + + if (s->poll_time > 0) + del_timer_sync(&s->timer); + + uart_update_timeout(port, termios->c_cflag, baud); + + spin_lock(&s->conf_lock); + s->conf = (s->conf & ~param_mask) | (param_new & param_mask); + s->conf_commit = 1; + s->parity = parity; + spin_unlock(&s->conf_lock); + max3100_dowork(s); + + if (UART_ENABLE_MS(&s->port, termios->c_cflag)) + max3100_enable_ms(&s->port); +} + +static void max3100_shutdown(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + if (s->suspending) + return; + + s->force_end_work = 1; + + if (s->poll_time > 0) + del_timer_sync(&s->timer); + + if (s->workqueue) { + flush_workqueue(s->workqueue); + destroy_workqueue(s->workqueue); + s->workqueue = NULL; + } + if (s->irq) + free_irq(s->irq, s); + + /* set shutdown mode to save power */ + if (s->max3100_hw_suspend) + s->max3100_hw_suspend(1); + else { + u16 tx, rx; + + tx = MAX3100_WC | MAX3100_SHDN; + max3100_sr(s, tx, &rx); + } +} + +static int max3100_startup(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + char b[12]; + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + s->conf = MAX3100_RM; + s->baud = s->crystal ? 230400 : 115200; + s->rx_enabled = 1; + + if (s->suspending) + return 0; + + s->force_end_work = 0; + s->parity = 0; + s->rts = 0; + + sprintf(b, "max3100-%d", s->minor); + s->workqueue = create_freezeable_workqueue(b); + if (!s->workqueue) { + dev_warn(&s->spi->dev, "cannot create workqueue\n"); + return -EBUSY; + } + INIT_WORK(&s->work, max3100_work); + + if (request_irq(s->irq, max3100_irq, + IRQF_TRIGGER_FALLING, "max3100", s) < 0) { + dev_warn(&s->spi->dev, "cannot allocate irq %d\n", s->irq); + s->irq = 0; + destroy_workqueue(s->workqueue); + s->workqueue = NULL; + return -EBUSY; + } + + if (s->loopback) { + u16 tx, rx; + tx = 0x4001; + max3100_sr(s, tx, &rx); + } + + if (s->max3100_hw_suspend) + s->max3100_hw_suspend(0); + s->conf_commit = 1; + max3100_dowork(s); + /* wait for clock to settle */ + msleep(50); + + max3100_enable_ms(&s->port); + + return 0; +} + +static const char *max3100_type(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + return s->port.type == PORT_MAX3100 ? "MAX3100" : NULL; +} + +static void max3100_release_port(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); +} + +static void max3100_config_port(struct uart_port *port, int flags) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + if (flags & UART_CONFIG_TYPE) + s->port.type = PORT_MAX3100; +} + +static int max3100_verify_port(struct uart_port *port, + struct serial_struct *ser) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + int ret = -EINVAL; + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + if (ser->type == PORT_UNKNOWN || ser->type == PORT_MAX3100) + ret = 0; + return ret; +} + +static void max3100_stop_tx(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); +} + +static int max3100_request_port(struct uart_port *port) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + return 0; +} + +static void max3100_break_ctl(struct uart_port *port, int break_state) +{ + struct max3100_port *s = container_of(port, + struct max3100_port, + port); + + dev_dbg(&s->spi->dev, "%s\n", __func__); +} + +static struct uart_ops max3100_ops = { + .tx_empty = max3100_tx_empty, + .set_mctrl = max3100_set_mctrl, + .get_mctrl = max3100_get_mctrl, + .stop_tx = max3100_stop_tx, + .start_tx = max3100_start_tx, + .stop_rx = max3100_stop_rx, + .enable_ms = max3100_enable_ms, + .break_ctl = max3100_break_ctl, + .startup = max3100_startup, + .shutdown = max3100_shutdown, + .set_termios = max3100_set_termios, + .type = max3100_type, + .release_port = max3100_release_port, + .request_port = max3100_request_port, + .config_port = max3100_config_port, + .verify_port = max3100_verify_port, +}; + +static struct uart_driver max3100_uart_driver = { + .owner = THIS_MODULE, + .driver_name = "ttyMAX", + .dev_name = "ttyMAX", + .major = MAX3100_MAJOR, + .minor = MAX3100_MINOR, + .nr = MAX_MAX3100, +}; +static int uart_driver_registered; + +static int __devinit max3100_probe(struct spi_device *spi) +{ + int i, retval; + struct plat_max3100 *pdata; + u16 tx, rx; + + mutex_lock(&max3100s_lock); + + if (!uart_driver_registered) { + uart_driver_registered = 1; + retval = uart_register_driver(&max3100_uart_driver); + if (retval) { + printk(KERN_ERR "Couldn't register max3100 uart driver\n"); + mutex_unlock(&max3100s_lock); + return retval; + } + } + + for (i = 0; i < MAX_MAX3100; i++) + if (!max3100s[i]) + break; + if (i == MAX_MAX3100) { + dev_warn(&spi->dev, "too many MAX3100 chips\n"); + mutex_unlock(&max3100s_lock); + return -ENOMEM; + } + + max3100s[i] = kzalloc(sizeof(struct max3100_port), GFP_KERNEL); + if (!max3100s[i]) { + dev_warn(&spi->dev, + "kmalloc for max3100 structure %d failed!\n", i); + mutex_unlock(&max3100s_lock); + return -ENOMEM; + } + max3100s[i]->spi = spi; + max3100s[i]->irq = spi->irq; + spin_lock_init(&max3100s[i]->conf_lock); + dev_set_drvdata(&spi->dev, max3100s[i]); + pdata = spi->dev.platform_data; + max3100s[i]->crystal = pdata->crystal; + max3100s[i]->loopback = pdata->loopback; + max3100s[i]->poll_time = pdata->poll_time * HZ / 1000; + if (pdata->poll_time > 0 && max3100s[i]->poll_time == 0) + max3100s[i]->poll_time = 1; + max3100s[i]->max3100_hw_suspend = pdata->max3100_hw_suspend; + max3100s[i]->minor = i; + init_timer(&max3100s[i]->timer); + max3100s[i]->timer.function = max3100_timeout; + max3100s[i]->timer.data = (unsigned long) max3100s[i]; + + dev_dbg(&spi->dev, "%s: adding port %d\n", __func__, i); + max3100s[i]->port.irq = max3100s[i]->irq; + max3100s[i]->port.uartclk = max3100s[i]->crystal ? 3686400 : 1843200; + max3100s[i]->port.fifosize = 16; + max3100s[i]->port.ops = &max3100_ops; + max3100s[i]->port.flags = UPF_SKIP_TEST | UPF_BOOT_AUTOCONF; + max3100s[i]->port.line = i; + max3100s[i]->port.type = PORT_MAX3100; + max3100s[i]->port.dev = &spi->dev; + retval = uart_add_one_port(&max3100_uart_driver, &max3100s[i]->port); + if (retval < 0) + dev_warn(&spi->dev, + "uart_add_one_port failed for line %d with error %d\n", + i, retval); + + /* set shutdown mode to save power. Will be woken-up on open */ + if (max3100s[i]->max3100_hw_suspend) + max3100s[i]->max3100_hw_suspend(1); + else { + tx = MAX3100_WC | MAX3100_SHDN; + max3100_sr(max3100s[i], tx, &rx); + } + mutex_unlock(&max3100s_lock); + return 0; +} + +static int __devexit max3100_remove(struct spi_device *spi) +{ + struct max3100_port *s = dev_get_drvdata(&spi->dev); + int i; + + mutex_lock(&max3100s_lock); + + /* find out the index for the chip we are removing */ + for (i = 0; i < MAX_MAX3100; i++) + if (max3100s[i] == s) + break; + + dev_dbg(&spi->dev, "%s: removing port %d\n", __func__, i); + uart_remove_one_port(&max3100_uart_driver, &max3100s[i]->port); + kfree(max3100s[i]); + max3100s[i] = NULL; + + /* check if this is the last chip we have */ + for (i = 0; i < MAX_MAX3100; i++) + if (max3100s[i]) { + mutex_unlock(&max3100s_lock); + return 0; + } + pr_debug("removing max3100 driver\n"); + uart_unregister_driver(&max3100_uart_driver); + + mutex_unlock(&max3100s_lock); + return 0; +} + +#ifdef CONFIG_PM + +static int max3100_suspend(struct spi_device *spi, pm_message_t state) +{ + struct max3100_port *s = dev_get_drvdata(&spi->dev); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + disable_irq(s->irq); + + s->suspending = 1; + uart_suspend_port(&max3100_uart_driver, &s->port); + + if (s->max3100_hw_suspend) + s->max3100_hw_suspend(1); + else { + /* no HW suspend, so do SW one */ + u16 tx, rx; + + tx = MAX3100_WC | MAX3100_SHDN; + max3100_sr(s, tx, &rx); + } + return 0; +} + +static int max3100_resume(struct spi_device *spi) +{ + struct max3100_port *s = dev_get_drvdata(&spi->dev); + + dev_dbg(&s->spi->dev, "%s\n", __func__); + + if (s->max3100_hw_suspend) + s->max3100_hw_suspend(0); + uart_resume_port(&max3100_uart_driver, &s->port); + s->suspending = 0; + + enable_irq(s->irq); + + s->conf_commit = 1; + if (s->workqueue) + max3100_dowork(s); + + return 0; +} + +#else +#define max3100_suspend NULL +#define max3100_resume NULL +#endif + +static struct spi_driver max3100_driver = { + .driver = { + .name = "max3100", + .bus = &spi_bus_type, + .owner = THIS_MODULE, + }, + + .probe = max3100_probe, + .remove = __devexit_p(max3100_remove), + .suspend = max3100_suspend, + .resume = max3100_resume, +}; + +static int __init max3100_init(void) +{ + return spi_register_driver(&max3100_driver); +} +module_init(max3100_init); + +static void __exit max3100_exit(void) +{ + spi_unregister_driver(&max3100_driver); +} +module_exit(max3100_exit); + +MODULE_DESCRIPTION("MAX3100 driver"); +MODULE_AUTHOR("Christian Pellegrin <chripell@evolware.org>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c index a4dc79b1d7a..47c6837850b 100644 --- a/drivers/serial/sunsu.c +++ b/drivers/serial/sunsu.c @@ -1178,7 +1178,7 @@ static struct uart_driver sunsu_reg = { .major = TTY_MAJOR, }; -static int __init sunsu_kbd_ms_init(struct uart_sunsu_port *up) +static int __devinit sunsu_kbd_ms_init(struct uart_sunsu_port *up) { int quot, baud; #ifdef CONFIG_SERIO diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c index 4ed228a8994..bb5e6f67157 100644 --- a/drivers/usb/host/ohci-at91.c +++ b/drivers/usb/host/ohci-at91.c @@ -280,7 +280,7 @@ static int ohci_hcd_at91_drv_probe(struct platform_device *pdev) * are always powered while this driver is active, and use * active-low power switches. */ - for (i = 0; i < pdata->ports; i++) { + for (i = 0; i < ARRAY_SIZE(pdata->vbus_pin); i++) { if (pdata->vbus_pin[i] <= 0) continue; gpio_request(pdata->vbus_pin[i], "ohci_vbus"); @@ -298,7 +298,7 @@ static int ohci_hcd_at91_drv_remove(struct platform_device *pdev) int i; if (pdata) { - for (i = 0; i < pdata->ports; i++) { + for (i = 0; i < ARRAY_SIZE(pdata->vbus_pin); i++) { if (pdata->vbus_pin[i] <= 0) continue; gpio_direction_output(pdata->vbus_pin[i], 1); |