diff options
Diffstat (limited to 'drivers/infiniband/core')
-rw-r--r-- | drivers/infiniband/core/cm.c | 2 | ||||
-rw-r--r-- | drivers/infiniband/core/mad.c | 24 | ||||
-rw-r--r-- | drivers/infiniband/core/sysfs.c | 2 | ||||
-rw-r--r-- | drivers/infiniband/core/ucm.c | 76 | ||||
-rw-r--r-- | drivers/infiniband/core/ud_header.c | 14 | ||||
-rw-r--r-- | drivers/infiniband/core/umem.c | 2 | ||||
-rw-r--r-- | drivers/infiniband/core/user_mad.c | 178 | ||||
-rw-r--r-- | drivers/infiniband/core/uverbs.h | 13 | ||||
-rw-r--r-- | drivers/infiniband/core/uverbs_cmd.c | 25 | ||||
-rw-r--r-- | drivers/infiniband/core/uverbs_main.c | 261 |
10 files changed, 303 insertions, 294 deletions
diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 5130fc5..764787e 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -3597,7 +3597,7 @@ static ssize_t cm_show_counter(struct kobject *obj, struct attribute *attr, atomic_long_read(&group->counter[cm_attr->index])); } -static struct sysfs_ops cm_counter_ops = { +static const struct sysfs_ops cm_counter_ops = { .show = cm_show_counter }; diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index 7522008..e351b15 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -1193,10 +1193,7 @@ static int method_in_use(struct ib_mad_mgmt_method_table **method, { int i; - for (i = find_first_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS); - i < IB_MGMT_MAX_METHODS; - i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, - 1+i)) { + for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) { if ((*method)->agent[i]) { printk(KERN_ERR PFX "Method %d already in use\n", i); return -EINVAL; @@ -1330,13 +1327,9 @@ static int add_nonoui_reg_req(struct ib_mad_reg_req *mad_reg_req, goto error3; /* Finally, add in methods being registered */ - for (i = find_first_bit(mad_reg_req->method_mask, - IB_MGMT_MAX_METHODS); - i < IB_MGMT_MAX_METHODS; - i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, - 1+i)) { + for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) (*method)->agent[i] = agent_priv; - } + return 0; error3: @@ -1429,13 +1422,9 @@ check_in_use: goto error4; /* Finally, add in methods being registered */ - for (i = find_first_bit(mad_reg_req->method_mask, - IB_MGMT_MAX_METHODS); - i < IB_MGMT_MAX_METHODS; - i = find_next_bit(mad_reg_req->method_mask, IB_MGMT_MAX_METHODS, - 1+i)) { + for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) (*method)->agent[i] = agent_priv; - } + return 0; error4: @@ -2964,6 +2953,9 @@ static void ib_mad_remove_device(struct ib_device *device) { int i, num_ports, cur_port; + if (rdma_node_get_transport(device->node_type) != RDMA_TRANSPORT_IB) + return; + if (device->node_type == RDMA_NODE_IB_SWITCH) { num_ports = 1; cur_port = 0; diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index 158a214..1558bb7 100644 --- a/drivers/infiniband/core/sysfs.c +++ b/drivers/infiniband/core/sysfs.c @@ -79,7 +79,7 @@ static ssize_t port_attr_show(struct kobject *kobj, return port_attr->show(p, port_attr, buf); } -static struct sysfs_ops port_sysfs_ops = { +static const struct sysfs_ops port_sysfs_ops = { .show = port_attr_show }; diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c index f504c9b..017d6e2 100644 --- a/drivers/infiniband/core/ucm.c +++ b/drivers/infiniband/core/ucm.c @@ -1215,15 +1215,18 @@ static void ib_ucm_release_dev(struct device *dev) ucm_dev = container_of(dev, struct ib_ucm_device, dev); cdev_del(&ucm_dev->cdev); - clear_bit(ucm_dev->devnum, dev_map); + if (ucm_dev->devnum < IB_UCM_MAX_DEVICES) + clear_bit(ucm_dev->devnum, dev_map); + else + clear_bit(ucm_dev->devnum - IB_UCM_MAX_DEVICES, dev_map); kfree(ucm_dev); } static const struct file_operations ucm_fops = { - .owner = THIS_MODULE, - .open = ib_ucm_open, + .owner = THIS_MODULE, + .open = ib_ucm_open, .release = ib_ucm_close, - .write = ib_ucm_write, + .write = ib_ucm_write, .poll = ib_ucm_poll, }; @@ -1237,8 +1240,32 @@ static ssize_t show_ibdev(struct device *dev, struct device_attribute *attr, } static DEVICE_ATTR(ibdev, S_IRUGO, show_ibdev, NULL); +static dev_t overflow_maj; +static DECLARE_BITMAP(overflow_map, IB_UCM_MAX_DEVICES); +static int find_overflow_devnum(void) +{ + int ret; + + if (!overflow_maj) { + ret = alloc_chrdev_region(&overflow_maj, 0, IB_UCM_MAX_DEVICES, + "infiniband_cm"); + if (ret) { + printk(KERN_ERR "ucm: couldn't register dynamic device number\n"); + return ret; + } + } + + ret = find_first_zero_bit(overflow_map, IB_UCM_MAX_DEVICES); + if (ret >= IB_UCM_MAX_DEVICES) + return -1; + + return ret; +} + static void ib_ucm_add_one(struct ib_device *device) { + int devnum; + dev_t base; struct ib_ucm_device *ucm_dev; if (!device->alloc_ucontext || @@ -1251,16 +1278,25 @@ static void ib_ucm_add_one(struct ib_device *device) ucm_dev->ib_dev = device; - ucm_dev->devnum = find_first_zero_bit(dev_map, IB_UCM_MAX_DEVICES); - if (ucm_dev->devnum >= IB_UCM_MAX_DEVICES) - goto err; - - set_bit(ucm_dev->devnum, dev_map); + devnum = find_first_zero_bit(dev_map, IB_UCM_MAX_DEVICES); + if (devnum >= IB_UCM_MAX_DEVICES) { + devnum = find_overflow_devnum(); + if (devnum < 0) + goto err; + + ucm_dev->devnum = devnum + IB_UCM_MAX_DEVICES; + base = devnum + overflow_maj; + set_bit(devnum, overflow_map); + } else { + ucm_dev->devnum = devnum; + base = devnum + IB_UCM_BASE_DEV; + set_bit(devnum, dev_map); + } cdev_init(&ucm_dev->cdev, &ucm_fops); ucm_dev->cdev.owner = THIS_MODULE; kobject_set_name(&ucm_dev->cdev.kobj, "ucm%d", ucm_dev->devnum); - if (cdev_add(&ucm_dev->cdev, IB_UCM_BASE_DEV + ucm_dev->devnum, 1)) + if (cdev_add(&ucm_dev->cdev, base, 1)) goto err; ucm_dev->dev.class = &cm_class; @@ -1281,7 +1317,10 @@ err_dev: device_unregister(&ucm_dev->dev); err_cdev: cdev_del(&ucm_dev->cdev); - clear_bit(ucm_dev->devnum, dev_map); + if (ucm_dev->devnum < IB_UCM_MAX_DEVICES) + clear_bit(devnum, dev_map); + else + clear_bit(devnum, overflow_map); err: kfree(ucm_dev); return; @@ -1297,11 +1336,8 @@ static void ib_ucm_remove_one(struct ib_device *device) device_unregister(&ucm_dev->dev); } -static ssize_t show_abi_version(struct class *class, char *buf) -{ - return sprintf(buf, "%d\n", IB_USER_CM_ABI_VERSION); -} -static CLASS_ATTR(abi_version, S_IRUGO, show_abi_version, NULL); +static CLASS_ATTR_STRING(abi_version, S_IRUGO, + __stringify(IB_USER_CM_ABI_VERSION)); static int __init ib_ucm_init(void) { @@ -1314,7 +1350,7 @@ static int __init ib_ucm_init(void) goto error1; } - ret = class_create_file(&cm_class, &class_attr_abi_version); + ret = class_create_file(&cm_class, &class_attr_abi_version.attr); if (ret) { printk(KERN_ERR "ucm: couldn't create abi_version attribute\n"); goto error2; @@ -1328,7 +1364,7 @@ static int __init ib_ucm_init(void) return 0; error3: - class_remove_file(&cm_class, &class_attr_abi_version); + class_remove_file(&cm_class, &class_attr_abi_version.attr); error2: unregister_chrdev_region(IB_UCM_BASE_DEV, IB_UCM_MAX_DEVICES); error1: @@ -1338,8 +1374,10 @@ error1: static void __exit ib_ucm_cleanup(void) { ib_unregister_client(&ucm_client); - class_remove_file(&cm_class, &class_attr_abi_version); + class_remove_file(&cm_class, &class_attr_abi_version.attr); unregister_chrdev_region(IB_UCM_BASE_DEV, IB_UCM_MAX_DEVICES); + if (overflow_maj) + unregister_chrdev_region(overflow_maj, IB_UCM_MAX_DEVICES); idr_destroy(&ctx_id_table); } diff --git a/drivers/infiniband/core/ud_header.c b/drivers/infiniband/core/ud_header.c index 8ec7876..650b501 100644 --- a/drivers/infiniband/core/ud_header.c +++ b/drivers/infiniband/core/ud_header.c @@ -181,6 +181,7 @@ static const struct ib_field deth_table[] = { * ib_ud_header_init - Initialize UD header structure * @payload_bytes:Length of packet payload * @grh_present:GRH flag (if non-zero, GRH will be included) + * @immediate_present: specify if immediate data should be used * @header:Structure to initialize * * ib_ud_header_init() initializes the lrh.link_version, lrh.link_next_header, @@ -191,21 +192,13 @@ static const struct ib_field deth_table[] = { */ void ib_ud_header_init(int payload_bytes, int grh_present, + int immediate_present, struct ib_ud_header *header) { - int header_len; u16 packet_length; memset(header, 0, sizeof *header); - header_len = - IB_LRH_BYTES + - IB_BTH_BYTES + - IB_DETH_BYTES; - if (grh_present) { - header_len += IB_GRH_BYTES; - } - header->lrh.link_version = 0; header->lrh.link_next_header = grh_present ? IB_LNH_IBA_GLOBAL : IB_LNH_IBA_LOCAL; @@ -231,7 +224,8 @@ void ib_ud_header_init(int payload_bytes, header->lrh.packet_length = cpu_to_be16(packet_length); - if (header->immediate_present) + header->immediate_present = immediate_present; + if (immediate_present) header->bth.opcode = IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE; else header->bth.opcode = IB_OPCODE_UD_SEND_ONLY; diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 6f7c096..4f906f0 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -136,7 +136,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, down_write(¤t->mm->mmap_sem); locked = npages + current->mm->locked_vm; - lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT; + lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) { ret = -ENOMEM; diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index 7de0296..04b585e 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -65,12 +65,9 @@ enum { }; /* - * Our lifetime rules for these structs are the following: each time a - * device special file is opened, we look up the corresponding struct - * ib_umad_port by minor in the umad_port[] table while holding the - * port_lock. If this lookup succeeds, we take a reference on the - * ib_umad_port's struct ib_umad_device while still holding the - * port_lock; if the lookup fails, we fail the open(). We drop these + * Our lifetime rules for these structs are the following: + * device special file is opened, we take a reference on the + * ib_umad_port's struct ib_umad_device. We drop these * references in the corresponding close(). * * In addition to references coming from open character devices, there @@ -78,19 +75,14 @@ enum { * module's reference taken when allocating the ib_umad_device in * ib_umad_add_one(). * - * When destroying an ib_umad_device, we clear all of its - * ib_umad_ports from umad_port[] while holding port_lock before - * dropping the module's reference to the ib_umad_device. This is - * always safe because any open() calls will either succeed and obtain - * a reference before we clear the umad_port[] entries, or fail after - * we clear the umad_port[] entries. + * When destroying an ib_umad_device, we drop the module's reference. */ struct ib_umad_port { - struct cdev *cdev; + struct cdev cdev; struct device *dev; - struct cdev *sm_cdev; + struct cdev sm_cdev; struct device *sm_dev; struct semaphore sm_sem; @@ -136,7 +128,6 @@ static struct class *umad_class; static const dev_t base_dev = MKDEV(IB_UMAD_MAJOR, IB_UMAD_MINOR_BASE); static DEFINE_SPINLOCK(port_lock); -static struct ib_umad_port *umad_port[IB_UMAD_MAX_PORTS]; static DECLARE_BITMAP(dev_map, IB_UMAD_MAX_PORTS); static void ib_umad_add_one(struct ib_device *device); @@ -496,8 +487,8 @@ static ssize_t ib_umad_write(struct file *filp, const char __user *buf, ah_attr.ah_flags = IB_AH_GRH; memcpy(ah_attr.grh.dgid.raw, packet->mad.hdr.gid, 16); ah_attr.grh.sgid_index = packet->mad.hdr.gid_index; - ah_attr.grh.flow_label = be32_to_cpu(packet->mad.hdr.flow_label); - ah_attr.grh.hop_limit = packet->mad.hdr.hop_limit; + ah_attr.grh.flow_label = be32_to_cpu(packet->mad.hdr.flow_label); + ah_attr.grh.hop_limit = packet->mad.hdr.hop_limit; ah_attr.grh.traffic_class = packet->mad.hdr.traffic_class; } @@ -528,9 +519,9 @@ static ssize_t ib_umad_write(struct file *filp, const char __user *buf, goto err_ah; } - packet->msg->ah = ah; + packet->msg->ah = ah; packet->msg->timeout_ms = packet->mad.hdr.timeout_ms; - packet->msg->retries = packet->mad.hdr.retries; + packet->msg->retries = packet->mad.hdr.retries; packet->msg->context[0] = packet; /* Copy MAD header. Any RMPP header is already in place. */ @@ -779,15 +770,11 @@ static long ib_umad_compat_ioctl(struct file *filp, unsigned int cmd, /* * ib_umad_open() does not need the BKL: * - * - umad_port[] accesses are protected by port_lock, the - * ib_umad_port structures are properly reference counted, and + * - the ib_umad_port structures are properly reference counted, and * everything else is purely local to the file being created, so * races against other open calls are not a problem; * - the ioctl method does not affect any global state outside of the * file structure being operated on; - * - the port is added to umad_port[] as the last part of module - * initialization so the open method will either immediately run - * -ENXIO, or all required initialization will be done. */ static int ib_umad_open(struct inode *inode, struct file *filp) { @@ -795,13 +782,10 @@ static int ib_umad_open(struct inode *inode, struct file *filp) struct ib_umad_file *file; int ret = 0; - spin_lock(&port_lock); - port = umad_port[iminor(inode) - IB_UMAD_MINOR_BASE]; + port = container_of(inode->i_cdev, struct ib_umad_port, cdev); if (port) kref_get(&port->umad_dev->ref); - spin_unlock(&port_lock); - - if (!port) + else return -ENXIO; mutex_lock(&port->file_mutex); @@ -872,16 +856,16 @@ static int ib_umad_close(struct inode *inode, struct file *filp) } static const struct file_operations umad_fops = { - .owner = THIS_MODULE, - .read = ib_umad_read, - .write = ib_umad_write, - .poll = ib_umad_poll, + .owner = THIS_MODULE, + .read = ib_umad_read, + .write = ib_umad_write, + .poll = ib_umad_poll, .unlocked_ioctl = ib_umad_ioctl, #ifdef CONFIG_COMPAT - .compat_ioctl = ib_umad_compat_ioctl, + .compat_ioctl = ib_umad_compat_ioctl, #endif - .open = ib_umad_open, - .release = ib_umad_close + .open = ib_umad_open, + .release = ib_umad_close }; static int ib_umad_sm_open(struct inode *inode, struct file *filp) @@ -892,13 +876,10 @@ static int ib_umad_sm_open(struct inode *inode, struct file *filp) }; int ret; - spin_lock(&port_lock); - port = umad_port[iminor(inode) - IB_UMAD_MINOR_BASE - IB_UMAD_MAX_PORTS]; + port = container_of(inode->i_cdev, struct ib_umad_port, sm_cdev); if (port) kref_get(&port->umad_dev->ref); - spin_unlock(&port_lock); - - if (!port) + else return -ENXIO; if (filp->f_flags & O_NONBLOCK) { @@ -949,8 +930,8 @@ static int ib_umad_sm_close(struct inode *inode, struct file *filp) } static const struct file_operations umad_sm_fops = { - .owner = THIS_MODULE, - .open = ib_umad_sm_open, + .owner = THIS_MODULE, + .open = ib_umad_sm_open, .release = ib_umad_sm_close }; @@ -984,22 +965,54 @@ static ssize_t show_port(struct device *dev, struct device_attribute *attr, } static DEVICE_ATTR(port, S_IRUGO, show_port, NULL); -static ssize_t show_abi_version(struct class *class, char *buf) +static CLASS_ATTR_STRING(abi_version, S_IRUGO, + __stringify(IB_USER_MAD_ABI_VERSION)); + +static dev_t overflow_maj; +static DECLARE_BITMAP(overflow_map, IB_UMAD_MAX_PORTS); +static int find_overflow_devnum(void) { - return sprintf(buf, "%d\n", IB_USER_MAD_ABI_VERSION); + int ret; + + if (!overflow_maj) { + ret = alloc_chrdev_region(&overflow_maj, 0, IB_UMAD_MAX_PORTS * 2, + "infiniband_mad"); + if (ret) { + printk(KERN_ERR "user_mad: couldn't register dynamic device number\n"); + return ret; + } + } + + ret = find_first_zero_bit(overflow_map, IB_UMAD_MAX_PORTS); + if (ret >= IB_UMAD_MAX_PORTS) + return -1; + + return ret; } -static CLASS_ATTR(abi_version, S_IRUGO, show_abi_version, NULL); static int ib_umad_init_port(struct ib_device *device, int port_num, struct ib_umad_port *port) { + int devnum; + dev_t base; + spin_lock(&port_lock); - port->dev_num = find_first_zero_bit(dev_map, IB_UMAD_MAX_PORTS); - if (port->dev_num >= IB_UMAD_MAX_PORTS) { + devnum = find_first_zero_bit(dev_map, IB_UMAD_MAX_PORTS); + if (devnum >= IB_UMAD_MAX_PORTS) { spin_unlock(&port_lock); - return -1; + devnum = find_overflow_devnum(); + if (devnum < 0) + return -1; + + spin_lock(&port_lock); + port->dev_num = devnum + IB_UMAD_MAX_PORTS; + base = devnum + overflow_maj; + set_bit(devnum, overflow_map); + } else { + port->dev_num = devnum; + base = devnum + base_dev; + set_bit(devnum, dev_map); } - set_bit(port->dev_num, dev_map); spin_unlock(&port_lock); port->ib_dev = device; @@ -1008,17 +1021,14 @@ static int ib_umad_init_port(struct ib_device *device, int port_num, mutex_init(&port->file_mutex); INIT_LIST_HEAD(&port->file_list); - port->cdev = cdev_alloc(); - if (!port->cdev) - return -1; - port->cdev->owner = THIS_MODULE; - port->cdev->ops = &umad_fops; - kobject_set_name(&port->cdev->kobj, "umad%d", port->dev_num); - if (cdev_add(port->cdev, base_dev + port->dev_num, 1)) + cdev_init(&port->cdev, &umad_fops); + port->cdev.owner = THIS_MODULE; + kobject_set_name(&port->cdev.kobj, "umad%d", port->dev_num); + if (cdev_add(&port->cdev, base, 1)) goto err_cdev; port->dev = device_create(umad_class, device->dma_device, - port->cdev->dev, port, + port->cdev.dev, port, "umad%d", port->dev_num); if (IS_ERR(port->dev)) goto err_cdev; @@ -1028,17 +1038,15 @@ static int ib_umad_init_port(struct ib_device *device, int port_num, if (device_create_file(port->dev, &dev_attr_port)) goto err_dev; - port->sm_cdev = cdev_alloc(); - if (!port->sm_cdev) - goto err_dev; - port->sm_cdev->owner = THIS_MODULE; - port->sm_cdev->ops = &umad_sm_fops; - kobject_set_name(&port->sm_cdev->kobj, "issm%d", port->dev_num); - if (cdev_add(port->sm_cdev, base_dev + port->dev_num + IB_UMAD_MAX_PORTS, 1)) + base += IB_UMAD_MAX_PORTS; + cdev_init(&port->sm_cdev, &umad_sm_fops); + port->sm_cdev.owner = THIS_MODULE; + kobject_set_name(&port->sm_cdev.kobj, "issm%d", port->dev_num); + if (cdev_add(&port->sm_cdev, base, 1)) goto err_sm_cdev; port->sm_dev = device_create(umad_class, device->dma_device, - port->sm_cdev->dev, port, + port->sm_cdev.dev, port, "issm%d", port->dev_num); if (IS_ERR(port->sm_dev)) goto err_sm_cdev; @@ -1048,24 +1056,23 @@ static int ib_umad_init_port(struct ib_device *device, int port_num, if (device_create_file(port->sm_dev, &dev_attr_port)) goto err_sm_dev; - spin_lock(&port_lock); - umad_port[port->dev_num] = port; - spin_unlock(&port_lock); - return 0; err_sm_dev: - device_destroy(umad_class, port->sm_cdev->dev); + device_destroy(umad_class, port->sm_cdev.dev); err_sm_cdev: - cdev_del(port->sm_cdev); + cdev_del(&port->sm_cdev); err_dev: - device_destroy(umad_class, port->cdev->dev); + device_destroy(umad_class, port->cdev.dev); err_cdev: - cdev_del(port->cdev); - clear_bit(port->dev_num, dev_map); + cdev_del(&port->cdev); + if (port->dev_num < IB_UMAD_MAX_PORTS) + clear_bit(devnum, dev_map); + else + clear_bit(devnum, overflow_map); return -1; } @@ -1079,15 +1086,11 @@ static void ib_umad_kill_port(struct ib_umad_port *port) dev_set_drvdata(port->dev, NULL); dev_set_drvdata(port->sm_dev, NULL); - device_destroy(umad_class, port->cdev->dev); - device_destroy(umad_class, port->sm_cdev->dev); + device_destroy(umad_class, port->cdev.dev); + device_destroy(umad_class, port->sm_cdev.dev); - cdev_del(port->cdev); - cdev_del(port->sm_cdev); - - spin_lock(&port_lock); - umad_port[port->dev_num] = NULL; - spin_unlock(&port_lock); + cdev_del(&port->cdev); + cdev_del(&port->sm_cdev); mutex_lock(&port->file_mutex); @@ -1106,7 +1109,10 @@ static void ib_umad_kill_port(struct ib_umad_port *port) mutex_unlock(&port->file_mutex); - clear_bit(port->dev_num, dev_map); + if (port->dev_num < IB_UMAD_MAX_PORTS) + clear_bit(port->dev_num, dev_map); + else + clear_bit(port->dev_num - IB_UMAD_MAX_PORTS, overflow_map); } static void ib_umad_add_one(struct ib_device *device) @@ -1185,7 +1191,7 @@ static int __init ib_umad_init(void) goto out_chrdev; } - ret = class_create_file(umad_class, &class_attr_abi_version); + ret = class_create_file(umad_class, &class_attr_abi_version.attr); if (ret) { printk(KERN_ERR "user_mad: couldn't create abi_version attribute\n"); goto out_class; @@ -1214,6 +1220,8 @@ static void __exit ib_umad_cleanup(void) ib_unregister_client(&umad_client); class_destroy(umad_class); unregister_chrdev_region(base_dev, IB_UMAD_MAX_PORTS * 2); + if (overflow_maj) + unregister_chrdev_region(overflow_maj, IB_UMAD_MAX_PORTS * 2); } module_init(ib_umad_init); diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index b3ea958..a078e56 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -41,6 +41,7 @@ #include <linux/idr.h> #include <linux/mutex.h> #include <linux/completion.h> +#include <linux/cdev.h> #include <rdma/ib_verbs.h> #include <rdma/ib_umem.h> @@ -69,23 +70,23 @@ struct ib_uverbs_device { struct kref ref; + int num_comp_vectors; struct completion comp; - int devnum; - struct cdev *cdev; struct device *dev; struct ib_device *ib_dev; - int num_comp_vectors; + int devnum; + struct cdev cdev; }; struct ib_uverbs_event_file { struct kref ref; + int is_async; struct ib_uverbs_file *uverbs_file; spinlock_t lock; + int is_closed; wait_queue_head_t poll_wait; struct fasync_struct *async_queue; struct list_head event_list; - int is_async; - int is_closed; }; struct ib_uverbs_file { @@ -145,7 +146,7 @@ extern struct idr ib_uverbs_srq_idr; void idr_remove_uobj(struct idr *idp, struct ib_uobject *uobj); struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, - int is_async, int *fd); + int is_async); struct ib_uverbs_event_file *ib_uverbs_lookup_comp_file(int fd); void ib_uverbs_release_ucq(struct ib_uverbs_file *file, diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 112d397..f71cf13 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -301,10 +301,15 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, resp.num_comp_vectors = file->device->num_comp_vectors; - filp = ib_uverbs_alloc_event_file(file, 1, &resp.async_fd); + ret = get_unused_fd(); + if (ret < 0) + goto err_free; + resp.async_fd = ret; + + filp = ib_uverbs_alloc_event_file(file, 1); if (IS_ERR(filp)) { ret = PTR_ERR(filp); - goto err_free; + goto err_fd; } if (copy_to_user((void __user *) (unsigned long) cmd.response, @@ -332,9 +337,11 @@ ssize_t ib_uverbs_get_context(struct ib_uverbs_file *file, return in_len; err_file: - put_unused_fd(resp.async_fd); fput(filp); +err_fd: + put_unused_fd(resp.async_fd); + err_free: ibdev->dealloc_ucontext(ucontext); @@ -715,6 +722,7 @@ ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file, struct ib_uverbs_create_comp_channel cmd; struct ib_uverbs_create_comp_channel_resp resp; struct file *filp; + int ret; if (out_len < sizeof resp) return -ENOSPC; @@ -722,9 +730,16 @@ ssize_t ib_uverbs_create_comp_channel(struct ib_uverbs_file *file, if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; - filp = ib_uverbs_alloc_event_file(file, 0, &resp.fd); - if (IS_ERR(filp)) + ret = get_unused_fd(); + if (ret < 0) + return ret; + resp.fd = ret; + + filp = ib_uverbs_alloc_event_file(file, 0); + if (IS_ERR(filp)) { + put_unused_fd(resp.fd); return PTR_ERR(filp); + } if (copy_to_user((void __user *) (unsigned long) cmd.response, &resp, sizeof resp)) { diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 5f284ff..d805cf3 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -42,8 +42,8 @@ #include <linux/poll.h> #include <linux/sched.h> #include <linux/file.h> -#include <linux/mount.h> #include <linux/cdev.h> +#include <linux/anon_inodes.h> #include <asm/uaccess.h> @@ -53,8 +53,6 @@ MODULE_AUTHOR("Roland Dreier"); MODULE_DESCRIPTION("InfiniBand userspace verbs access"); MODULE_LICENSE("Dual BSD/GPL"); -#define INFINIBANDEVENTFS_MAGIC 0x49426576 /* "IBev" */ - enum { IB_UVERBS_MAJOR = 231, IB_UVERBS_BASE_MINOR = 192, @@ -75,44 +73,41 @@ DEFINE_IDR(ib_uverbs_qp_idr); DEFINE_IDR(ib_uverbs_srq_idr); static DEFINE_SPINLOCK(map_lock); -static struct ib_uverbs_device *dev_table[IB_UVERBS_MAX_DEVICES]; static DECLARE_BITMAP(dev_map, IB_UVERBS_MAX_DEVICES); static ssize_t (*uverbs_cmd_table[])(struct ib_uverbs_file *file, const char __user *buf, int in_len, int out_len) = { - [IB_USER_VERBS_CMD_GET_CONTEXT] = ib_uverbs_get_context, - [IB_USER_VERBS_CMD_QUERY_DEVICE] = ib_uverbs_query_device, - [IB_USER_VERBS_CMD_QUERY_PORT] = ib_uverbs_query_port, - [IB_USER_VERBS_CMD_ALLOC_PD] = ib_uverbs_alloc_pd, - [IB_USER_VERBS_CMD_DEALLOC_PD] = ib_uverbs_dealloc_pd, - [IB_USER_VERBS_CMD_REG_MR] = ib_uverbs_reg_mr, - [IB_USER_VERBS_CMD_DEREG_MR] = ib_uverbs_dereg_mr, + [IB_USER_VERBS_CMD_GET_CONTEXT] = ib_uverbs_get_context, + [IB_USER_VERBS_CMD_QUERY_DEVICE] = ib_uverbs_query_device, + [IB_USER_VERBS_CMD_QUERY_PORT] = ib_uverbs_query_port, + [IB_USER_VERBS_CMD_ALLOC_PD] = ib_uverbs_alloc_pd, + [IB_USER_VERBS_CMD_DEALLOC_PD] = ib_uverbs_dealloc_pd, + [IB_USER_VERBS_CMD_REG_MR] = ib_uverbs_reg_mr, + [IB_USER_VERBS_CMD_DEREG_MR] = ib_uverbs_dereg_mr, [IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL] = ib_uverbs_create_comp_channel, - [IB_USER_VERBS_CMD_CREATE_CQ] = ib_uverbs_create_cq, - [IB_USER_VERBS_CMD_RESIZE_CQ] = ib_uverbs_resize_cq, - [IB_USER_VERBS_CMD_POLL_CQ] = ib_uverbs_poll_cq, - [IB_USER_VERBS_CMD_REQ_NOTIFY_CQ] = ib_uverbs_req_notify_cq, - [IB_USER_VERBS_CMD_DESTROY_CQ] = ib_uverbs_destroy_cq, - [IB_USER_VERBS_CMD_CREATE_QP] = ib_uverbs_create_qp, - [IB_USER_VERBS_CMD_QUERY_QP] = ib_uverbs_query_qp, - [IB_USER_VERBS_CMD_MODIFY_QP] = ib_uverbs_modify_qp, - [IB_USER_VERBS_CMD_DESTROY_QP] = ib_uverbs_destroy_qp, - [IB_USER_VERBS_CMD_POST_SEND] = ib_uverbs_post_send, - [IB_USER_VERBS_CMD_POST_RECV] = ib_uverbs_post_recv, - [IB_USER_VERBS_CMD_POST_SRQ_RECV] = ib_uverbs_post_srq_recv, - [IB_USER_VERBS_CMD_CREATE_AH] = ib_uverbs_create_ah, - [IB_USER_VERBS_CMD_DESTROY_AH] = ib_uverbs_destroy_ah, - [IB_USER_VERBS_CMD_ATTACH_MCAST] = ib_uverbs_attach_mcast, - [IB_USER_VERBS_CMD_DETACH_MCAST] = ib_uverbs_detach_mcast, - [IB_USER_VERBS_CMD_CREATE_SRQ] = ib_uverbs_create_srq, - [IB_USER_VERBS_CMD_MODIFY_SRQ] = ib_uverbs_modify_srq, - [IB_USER_VERBS_CMD_QUERY_SRQ] = ib_uverbs_query_srq, - [IB_USER_VERBS_CMD_DESTROY_SRQ] = ib_uverbs_destroy_srq, + [IB_USER_VERBS_CMD_CREATE_CQ] = ib_uverbs_create_cq, + [IB_USER_VERBS_CMD_RESIZE_CQ] = ib_uverbs_resize_cq, + [IB_USER_VERBS_CMD_POLL_CQ] = ib_uverbs_poll_cq, + [IB_USER_VERBS_CMD_REQ_NOTIFY_CQ] = ib_uverbs_req_notify_cq, + [IB_USER_VERBS_CMD_DESTROY_CQ] = ib_uverbs_destroy_cq, + [IB_USER_VERBS_CMD_CREATE_QP] = ib_uverbs_create_qp, + [IB_USER_VERBS_CMD_QUERY_QP] = ib_uverbs_query_qp, + [IB_USER_VERBS_CMD_MODIFY_QP] = ib_uverbs_modify_qp, + [IB_USER_VERBS_CMD_DESTROY_QP] = ib_uverbs_destroy_qp, + [IB_USER_VERBS_CMD_POST_SEND] = ib_uverbs_post_send, + [IB_USER_VERBS_CMD_POST_RECV] = ib_uverbs_post_recv, + [IB_USER_VERBS_CMD_POST_SRQ_RECV] = ib_uverbs_post_srq_recv, + [IB_USER_VERBS_CMD_CREATE_AH] = ib_uverbs_create_ah, + [IB_USER_VERBS_CMD_DESTROY_AH] = ib_uverbs_destroy_ah, + [IB_USER_VERBS_CMD_ATTACH_MCAST] = ib_uverbs_attach_mcast, + [IB_USER_VERBS_CMD_DETACH_MCAST] = ib_uverbs_detach_mcast, + [IB_USER_VERBS_CMD_CREATE_SRQ] = ib_uverbs_create_srq, + [IB_USER_VERBS_CMD_MODIFY_SRQ] = ib_uverbs_modify_srq, + [IB_USER_VERBS_CMD_QUERY_SRQ] = ib_uverbs_query_srq, + [IB_USER_VERBS_CMD_DESTROY_SRQ] = ib_uverbs_destroy_srq, }; -static struct vfsmount *uverbs_event_mnt; - static void ib_uverbs_add_one(struct ib_device *device); static void ib_uverbs_remove_one(struct ib_device *device); @@ -370,7 +365,7 @@ static int ib_uverbs_event_close(struct inode *inode, struct file *filp) static const struct file_operations uverbs_event_fops = { .owner = THIS_MODULE, - .read = ib_uverbs_event_read, + .read = ib_uverbs_event_read, .poll = ib_uverbs_event_poll, .release = ib_uverbs_event_close, .fasync = ib_uverbs_event_fasync @@ -489,12 +484,10 @@ void ib_uverbs_event_handler(struct ib_event_handler *handler, } struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, - int is_async, int *fd) + int is_async) { struct ib_uverbs_event_file *ev_file; - struct path path; struct file *filp; - int ret; ev_file = kmalloc(sizeof *ev_file, GFP_KERNEL); if (!ev_file) @@ -509,38 +502,12 @@ struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, ev_file->is_async = is_async; ev_file->is_closed = 0; - *fd = get_unused_fd(); - if (*fd < 0) { - ret = *fd; - goto err; - } - - /* - * fops_get() can't fail here, because we're coming from a - * system call on a uverbs file, which will already have a - * module reference. - */ - path.mnt = uverbs_event_mnt; - path.dentry = uverbs_event_mnt->mnt_root; - path_get(&path); - filp = alloc_file(&path, FMODE_READ, fops_get(&uverbs_event_fops)); - if (!filp) { - ret = -ENFILE; - goto err_fd; - } - - filp->private_data = ev_file; + filp = anon_inode_getfile("[infinibandevent]", &uverbs_event_fops, + ev_file, O_RDONLY); + if (IS_ERR(filp)) + kfree(ev_file); return filp; - -err_fd: - fops_put(&uverbs_event_fops); - path_put(&path); - put_unused_fd(*fd); - -err: - kfree(ev_file); - return ERR_PTR(ret); } /* @@ -617,14 +584,12 @@ static int ib_uverbs_mmap(struct file *filp, struct vm_area_struct *vma) /* * ib_uverbs_open() does not need the BKL: * - * - dev_table[] accesses are protected by map_lock, the - * ib_uverbs_device structures are properly reference counted, and + * - the ib_uverbs_device structures are properly reference counted and * everything else is purely local to the file being created, so * races against other open calls are not a problem; * - there is no ioctl method to race against; - * - the device is added to dev_table[] as the last part of module - * initialization, the open method will either immediately run - * -ENXIO, or all required initialization will be done. + * - the open method will either immediately run -ENXIO, or all + * required initialization will be done. */ static int ib_uverbs_open(struct inode *inode, struct file *filp) { @@ -632,13 +597,10 @@ static int ib_uverbs_open(struct inode *inode, struct file *filp) struct ib_uverbs_file *file; int ret; - spin_lock(&map_lock); - dev = dev_table[iminor(inode) - IB_UVERBS_BASE_MINOR]; + dev = container_of(inode->i_cdev, struct ib_uverbs_device, cdev); if (dev) kref_get(&dev->ref); - spin_unlock(&map_lock); - - if (!dev) + else return -ENXIO; if (!try_module_get(dev->ib_dev->owner)) { @@ -685,17 +647,17 @@ static int ib_uverbs_close(struct inode *inode, struct file *filp) } static const struct file_operations uverbs_fops = { - .owner = THIS_MODULE, - .write = ib_uverbs_write, - .open = ib_uverbs_open, + .owner = THIS_MODULE, + .write = ib_uverbs_write, + .open = ib_uverbs_open, .release = ib_uverbs_close }; static const struct file_operations uverbs_mmap_fops = { - .owner = THIS_MODULE, - .write = ib_uverbs_write, + .owner = THIS_MODULE, + .write = ib_uverbs_write, .mmap = ib_uverbs_mmap, - .open = ib_uverbs_open, + .open = ib_uverbs_open, .release = ib_uverbs_close }; @@ -729,14 +691,41 @@ static ssize_t show_dev_abi_version(struct device *device, } static DEVICE_ATTR(abi_version, S_IRUGO, show_dev_abi_version, NULL); -static ssize_t show_abi_version(struct class *class, char *buf) +static CLASS_ATTR_STRING(abi_version, S_IRUGO, + __stringify(IB_USER_VERBS_ABI_VERSION)); + +static dev_t overflow_maj; +static DECLARE_BITMAP(overflow_map, IB_UVERBS_MAX_DEVICES); + +/* + * If we have more than IB_UVERBS_MAX_DEVICES, dynamically overflow by + * requesting a new major number and doubling the number of max devices we + * support. It's stupid, but simple. + */ +static int find_overflow_devnum(void) { - return sprintf(buf, "%d\n", IB_USER_VERBS_ABI_VERSION); + int ret; + + if (!overflow_maj) { + ret = alloc_chrdev_region(&overflow_maj, 0, IB_UVERBS_MAX_DEVICES, + "infiniband_verbs"); + if (ret) { + printk(KERN_ERR "user_verbs: couldn't register dynamic device number\n"); + return ret; + } + } + + ret = find_first_zero_bit(overflow_map, IB_UVERBS_MAX_DEVICES); + if (ret >= IB_UVERBS_MAX_DEVICES) + return -1; + + return ret; } -static CLASS_ATTR(abi_version, S_IRUGO, show_abi_version, NULL); static void ib_uverbs_add_one(struct ib_device *device) { + int devnum; + dev_t base; struct ib_uverbs_device *uverbs_dev; if (!device->alloc_ucontext) @@ -750,28 +739,36 @@ static void ib_uverbs_add_one(struct ib_device *device) init_completion(&uverbs_dev->comp); spin_lock(&map_lock); - uverbs_dev->devnum = find_first_zero_bit(dev_map, IB_UVERBS_MAX_DEVICES); - if (uverbs_dev->devnum >= IB_UVERBS_MAX_DEVICES) { + devnum = find_first_zero_bit(dev_map, IB_UVERBS_MAX_DEVICES); + if (devnum >= IB_UVERBS_MAX_DEVICES) { spin_unlock(&map_lock); - goto err; + devnum = find_overflow_devnum(); + if (devnum < 0) + goto err; + + spin_lock(&map_lock); + uverbs_dev->devnum = devnum + IB_UVERBS_MAX_DEVICES; + base = devnum + overflow_maj; + set_bit(devnum, overflow_map); + } else { + uverbs_dev->devnum = devnum; + base = devnum + IB_UVERBS_BASE_DEV; + set_bit(devnum, dev_map); } - set_bit(uverbs_dev->devnum, dev_map); spin_unlock(&map_lock); uverbs_dev->ib_dev = device; uverbs_dev->num_comp_vectors = device->num_comp_vectors; - uverbs_dev->cdev = cdev_alloc(); - if (!uverbs_dev->cdev) - goto err; - uverbs_dev->cdev->owner = THIS_MODULE; - uverbs_dev->cdev->ops = device->mmap ? &uverbs_mmap_fops : &uverbs_fops; - kobject_set_name(&uverbs_dev->cdev->kobj, "uverbs%d", uverbs_dev->devnum); - if (cdev_add(uverbs_dev->cdev, IB_UVERBS_BASE_DEV + uverbs_dev->devnum, 1)) + cdev_init(&uverbs_dev->cdev, NULL); + uverbs_dev->cdev.owner = THIS_MODULE; + uverbs_dev->cdev.ops = device->mmap ? &uverbs_mmap_fops : &uverbs_fops; + kobject_set_name(&uverbs_dev->cdev.kobj, "uverbs%d", uverbs_dev->devnum); + if (cdev_add(&uverbs_dev->cdev, base, 1)) goto err_cdev; uverbs_dev->dev = device_create(uverbs_class, device->dma_device, - uverbs_dev->cdev->dev, uverbs_dev, + uverbs_dev->cdev.dev, uverbs_dev, "uverbs%d", uverbs_dev->devnum); if (IS_ERR(uverbs_dev->dev)) goto err_cdev; @@ -781,20 +778,19 @@ static void ib_uverbs_add_one(struct ib_device *device) if (device_create_file(uverbs_dev->dev, &dev_attr_abi_version)) goto err_class; - spin_lock(&map_lock); - dev_table[uverbs_dev->devnum] = uverbs_dev; - spin_unlock(&map_lock); - ib_set_client_data(device, &uverbs_client, uverbs_dev); return; err_class: - device_destroy(uverbs_class, uverbs_dev->cdev->dev); + device_destroy(uverbs_class, uverbs_dev->cdev.dev); err_cdev: - cdev_del(uverbs_dev->cdev); - clear_bit(uverbs_dev->devnum, dev_map); + cdev_del(&uverbs_dev->cdev); + if (uverbs_dev->devnum < IB_UVERBS_MAX_DEVICES) + clear_bit(devnum, dev_map); + else + clear_bit(devnum, overflow_map); err: kref_put(&uverbs_dev->ref, ib_uverbs_release_dev); @@ -811,35 +807,19 @@ static void ib_uverbs_remove_one(struct ib_device *device) return; dev_set_drvdata(uverbs_dev->dev, NULL); - device_destroy(uverbs_class, uverbs_dev->cdev->dev); - cdev_del(uverbs_dev->cdev); + device_destroy(uverbs_class, uverbs_dev->cdev.dev); + cdev_del(&uverbs_dev->cdev); - spin_lock(&map_lock); - dev_table[uverbs_dev->devnum] = NULL; - spin_unlock(&map_lock); - - clear_bit(uverbs_dev->devnum, dev_map); + if (uverbs_dev->devnum < IB_UVERBS_MAX_DEVICES) + clear_bit(uverbs_dev->devnum, dev_map); + else + clear_bit(uverbs_dev->devnum - IB_UVERBS_MAX_DEVICES, overflow_map); kref_put(&uverbs_dev->ref, ib_uverbs_release_dev); wait_for_completion(&uverbs_dev->comp); kfree(uverbs_dev); } -static int uverbs_event_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, - struct vfsmount *mnt) -{ - return get_sb_pseudo(fs_type, "infinibandevent:", NULL, - INFINIBANDEVENTFS_MAGIC, mnt); -} - -static struct file_system_type uverbs_event_fs = { - /* No owner field so module can be unloaded */ - .name = "infinibandeventfs", - .get_sb = uverbs_event_get_sb, - .kill_sb = kill_litter_super -}; - static int __init ib_uverbs_init(void) { int ret; @@ -858,39 +838,20 @@ static int __init ib_uverbs_init(void) goto out_chrdev; } - ret = class_create_file(uverbs_class, &class_attr_abi_version); + ret = class_create_file(uverbs_class, &class_attr_abi_version.attr); if (ret) { printk(KERN_ERR "user_verbs: couldn't create abi_version attribute\n"); goto out_class; } - ret = register_filesystem(&uverbs_event_fs); - if (ret) { - printk(KERN_ERR "user_verbs: couldn't register infinibandeventfs\n"); - goto out_class; - } - - uverbs_event_mnt = kern_mount(&uverbs_event_fs); - if (IS_ERR(uverbs_event_mnt)) { - ret = PTR_ERR(uverbs_event_mnt); - printk(KERN_ERR "user_verbs: couldn't mount infinibandeventfs\n"); - goto out_fs; - } - ret = ib_register_client(&uverbs_client); if (ret) { printk(KERN_ERR "user_verbs: couldn't register client\n"); - goto out_mnt; + goto out_class; } return 0; -out_mnt: - mntput(uverbs_event_mnt); - -out_fs: - unregister_filesystem(&uverbs_event_fs); - out_class: class_destroy(uverbs_class); @@ -904,10 +865,10 @@ out: static void __exit ib_uverbs_cleanup(void) { ib_unregister_client(&uverbs_client); - mntput(uverbs_event_mnt); - unregister_filesystem(&uverbs_event_fs); class_destroy(uverbs_class); unregister_chrdev_region(IB_UVERBS_BASE_DEV, IB_UVERBS_MAX_DEVICES); + if (overflow_maj) + unregister_chrdev_region(overflow_maj, IB_UVERBS_MAX_DEVICES); idr_destroy(&ib_uverbs_pd_idr); idr_destroy(&ib_uverbs_mr_idr); idr_destroy(&ib_uverbs_mw_idr); |