diff options
166 files changed, 4458 insertions, 2017 deletions
diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index 0c5086db8352..80e193d82e2e 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -13,72 +13,93 @@ Mailing list: linux-ext4@vger.kernel.org 1. Quick usage instructions: =========================== - - Grab updated e2fsprogs from - ftp://ftp.kernel.org/pub/linux/kernel/people/tytso/e2fsprogs-interim/ - This is a patchset on top of e2fsprogs-1.39, which can be found at + - Compile and install the latest version of e2fsprogs (as of this + writing version 1.41) from: + + http://sourceforge.net/project/showfiles.php?group_id=2406 + + or + ftp://ftp.kernel.org/pub/linux/kernel/people/tytso/e2fsprogs/ - - It's still mke2fs -j /dev/hda1 + or grab the latest git repository from: + + git://git.kernel.org/pub/scm/fs/ext2/e2fsprogs.git + + - Create a new filesystem using the ext4dev filesystem type: + + # mke2fs -t ext4dev /dev/hda1 + + Or configure an existing ext3 filesystem to support extents and set + the test_fs flag to indicate that it's ok for an in-development + filesystem to touch this filesystem: - - mount /dev/hda1 /wherever -t ext4dev + # tune2fs -O extents -E test_fs /dev/hda1 - - To enable extents, + If the filesystem was created with 128 byte inodes, it can be + converted to use 256 byte for greater efficiency via: - mount /dev/hda1 /wherever -t ext4dev -o extents + # tune2fs -I 256 /dev/hda1 - - The filesystem is compatible with the ext3 driver until you add a file - which has extents (ie: `mount -o extents', then create a file). + (Note: we currently do not have tools to convert an ext4dev + filesystem back to ext3; so please do not do try this on production + filesystems.) - NOTE: The "extents" mount flag is temporary. It will soon go away and - extents will be enabled by the "-o extents" flag to mke2fs or tune2fs + - Mounting: + + # mount -t ext4dev /dev/hda1 /wherever - When comparing performance with other filesystems, remember that - ext3/4 by default offers higher data integrity guarantees than most. So - when comparing with a metadata-only journalling filesystem, use `mount -o - data=writeback'. And you might as well use `mount -o nobh' too along - with it. Making the journal larger than the mke2fs default often helps - performance with metadata-intensive workloads. + ext3/4 by default offers higher data integrity guarantees than most. + So when comparing with a metadata-only journalling filesystem, such + as ext3, use `mount -o data=writeback'. And you might as well use + `mount -o nobh' too along with it. Making the journal larger than + the mke2fs default often helps performance with metadata-intensive + workloads. 2. Features =========== 2.1 Currently available -* ability to use filesystems > 16TB +* ability to use filesystems > 16TB (e2fsprogs support not available yet) * extent format reduces metadata overhead (RAM, IO for access, transactions) * extent format more robust in face of on-disk corruption due to magics, * internal redunancy in tree - -2.1 Previously available, soon to be enabled by default by "mkefs.ext4": - -* dir_index and resize inode will be on by default -* large inodes will be used by default for fast EAs, nsec timestamps, etc +* improved file allocation (multi-block alloc) +* fix 32000 subdirectory limit +* nsec timestamps for mtime, atime, ctime, create time +* inode version field on disk (NFSv4, Lustre) +* reduced e2fsck time via uninit_bg feature +* journal checksumming for robustness, performance +* persistent file preallocation (e.g for streaming media, databases) +* ability to pack bitmaps and inode tables into larger virtual groups via the + flex_bg feature +* large file support +* Inode allocation using large virtual block groups via flex_bg +* delayed allocation +* large block (up to pagesize) support +* efficent new ordered mode in JBD2 and ext4(avoid using buffer head to force + the ordering) 2.2 Candidate features for future inclusion -There are several under discussion, whether they all make it in is -partly a function of how much time everyone has to work on them: +* Online defrag (patches available but not well tested) +* reduced mke2fs time via lazy itable initialization in conjuction with + the uninit_bg feature (capability to do this is available in e2fsprogs + but a kernel thread to do lazy zeroing of unused inode table blocks + after filesystem is first mounted is required for safety) -* improved file allocation (multi-block alloc, delayed alloc; basically done) -* fix 32000 subdirectory limit (patch exists, needs some e2fsck work) -* nsec timestamps for mtime, atime, ctime, create time (patch exists, - needs some e2fsck work) -* inode version field on disk (NFSv4, Lustre; prototype exists) -* reduced mke2fs/e2fsck time via uninitialized groups (prototype exists) -* journal checksumming for robustness, performance (prototype exists) -* persistent file preallocation (e.g for streaming media, databases) +There are several others under discussion, whether they all make it in is +partly a function of how much time everyone has to work on them. Features like +metadata checksumming have been discussed and planned for a bit but no patches +exist yet so I'm not sure they're in the near-term roadmap. -Features like metadata checksumming have been discussed and planned for -a bit but no patches exist yet so I'm not sure they're in the near-term -roadmap. +The big performance win will come with mballoc, delalloc and flex_bg +grouping of bitmaps and inode tables. Some test results available here: -The big performance win will come with mballoc and delalloc. CFS has -been using mballoc for a few years already with Lustre, and IBM + Bull -did a lot of benchmarking on it. The reason it isn't in the first set of -patches is partly a manageability issue, and partly because it doesn't -directly affect the on-disk format (outside of much better allocation) -so it isn't critical to get into the first round of changes. I believe -Alex is working on a new set of patches right now. + - http://www.bullopensource.org/ext4/20080530/ffsb-write-2.6.26-rc2.html + - http://www.bullopensource.org/ext4/20080530/ffsb-readwrite-2.6.26-rc2.html 3. Options ========== @@ -222,9 +243,11 @@ stripe=n Number of filesystem blocks that mballoc will try to use for allocation size and alignment. For RAID5/6 systems this should be the number of data disks * RAID chunk size in file system blocks. - +delalloc (*) Deferring block allocation until write-out time. +nodelalloc Disable delayed allocation. Blocks are allocation + when data is copied from user to page cache. Data Mode ---------- +========= There are 3 different data modes: * writeback mode @@ -236,10 +259,10 @@ typically provide the best ext4 performance. * ordered mode In data=ordered mode, ext4 only officially journals metadata, but it logically -groups metadata and data blocks into a single unit called a transaction. When -it's time to write the new metadata out to disk, the associated data blocks -are written first. In general, this mode performs slightly slower than -writeback but significantly faster than journal mode. +groups metadata information related to data changes with the data blocks into a +single unit called a transaction. When it's time to write the new metadata +out to disk, the associated data blocks are written first. In general, +this mode performs slightly slower than writeback but significantly faster than journal mode. * journal mode data=journal mode provides full data and metadata journaling. All new data is @@ -247,7 +270,8 @@ written to the journal first, and then to its final location. In the event of a crash, the journal can be replayed, bringing both data and metadata into a consistent state. This mode is the slowest except when data needs to be read from and written to disk at the same time where it -outperforms all others modes. +outperforms all others modes. Curently ext4 does not have delayed +allocation support if this data journalling mode is selected. References ========== @@ -256,7 +280,8 @@ kernel source: <file:fs/ext4/> <file:fs/jbd2/> programs: http://e2fsprogs.sourceforge.net/ - http://ext2resize.sourceforge.net useful links: http://fedoraproject.org/wiki/ext3-devel http://www.bullopensource.org/ext4/ + http://ext4.wiki.kernel.org/index.php/Main_Page + http://fedoraproject.org/wiki/Features/Ext4 diff --git a/drivers/gpu/drm/drm_memory.c b/drivers/gpu/drm/drm_memory.c index 845081b44f63..0177012845c6 100644 --- a/drivers/gpu/drm/drm_memory.c +++ b/drivers/gpu/drm/drm_memory.c @@ -167,6 +167,11 @@ void drm_core_ioremap(struct drm_map *map, struct drm_device *dev) } EXPORT_SYMBOL(drm_core_ioremap); +void drm_core_ioremap_wc(struct drm_map *map, struct drm_device *dev) +{ + map->handle = ioremap_wc(map->offset, map->size); +} +EXPORT_SYMBOL(drm_core_ioremap_wc); void drm_core_ioremapfree(struct drm_map *map, struct drm_device *dev) { if (!map->handle || !map->size) diff --git a/drivers/gpu/drm/radeon/radeon_cp.c b/drivers/gpu/drm/radeon/radeon_cp.c index e53158f0ecb5..f0de81a5689d 100644 --- a/drivers/gpu/drm/radeon/radeon_cp.c +++ b/drivers/gpu/drm/radeon/radeon_cp.c @@ -1154,7 +1154,7 @@ static int radeon_do_init_cp(struct drm_device * dev, drm_radeon_init_t * init) dev_priv->gart_info.mapping.size = dev_priv->gart_info.table_size; - drm_core_ioremap(&dev_priv->gart_info.mapping, dev); + drm_core_ioremap_wc(&dev_priv->gart_info.mapping, dev); dev_priv->gart_info.addr = dev_priv->gart_info.mapping.handle; diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index 781ea5950373..09a2bec7fd32 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -4,28 +4,33 @@ * Copyright (c) 1999-2005, Mellanox Technologies, Inc. All rights reserved. * Copyright (c) 2005 Intel Corporation. All rights reserved. * - * This Software is licensed under one of the following licenses: + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: * - * 1) under the terms of the "Common Public License 1.0" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/cpl.php. + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: * - * 2) under the terms of the "The BSD License" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/bsd-license.php. + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * 3) under the terms of the "GNU General Public License (GPL) Version 2" a - * copy of which is available from the Open Source Initiative, see - * http://www.opensource.org/licenses/gpl-license.php. + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * - * Licensee has the right to choose one of the above licenses. - * - * Redistributions of source code must retain the above copyright - * notice and one of the license notices. - * - * Redistributions in binary form must reproduce both the above copyright - * notice, one of the license notices in the documentation - * and/or other materials provided with the distribution. + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. */ #include <linux/mutex.h> @@ -100,6 +105,7 @@ int rdma_copy_addr(struct rdma_dev_addr *dev_addr, struct net_device *dev, memcpy(dev_addr->broadcast, dev->broadcast, MAX_ADDR_LEN); if (dst_dev_addr) memcpy(dev_addr->dst_dev_addr, dst_dev_addr, MAX_ADDR_LEN); + dev_addr->src_dev = dev; return 0; } EXPORT_SYMBOL(rdma_copy_addr); diff --git a/drivers/infiniband/core/agent.h b/drivers/infiniband/core/agent.h index fb9ed1489f95..6669287009c2 100644 --- a/drivers/infiniband/core/agent.h +++ b/drivers/infiniband/core/agent.h @@ -32,8 +32,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: agent.h 1389 2004-12-27 22:56:47Z roland $ */ #ifndef __AGENT_H_ diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index e85f7013de57..68883565b725 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: cache.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/module.h> diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index a47fe64e5c39..55738eead3bf 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: cm.c 4311 2005-12-05 18:42:01Z sean.hefty $ */ #include <linux/completion.h> diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 671f13738054..ae11d5cc74d0 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -4,29 +4,33 @@ * Copyright (c) 1999-2005, Mellanox Technologies, Inc. All rights reserved. * Copyright (c) 2005-2006 Intel Corporation. All rights reserved. * - * This Software is licensed under one of the following licenses: + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: * - * 1) under the terms of the "Common Public License 1.0" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/cpl.php. + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: * - * 2) under the terms of the "The BSD License" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/bsd-license.php. + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * 3) under the terms of the "GNU General Public License (GPL) Version 2" a - * copy of which is available from the Open Source Initiative, see - * http://www.opensource.org/licenses/gpl-license.php. - * - * Licensee has the right to choose one of the above licenses. - * - * Redistributions of source code must retain the above copyright - * notice and one of the license notices. - * - * Redistributions in binary form must reproduce both the above copyright - * notice, one of the license notices in the documentation - * and/or other materials provided with the distribution. + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. */ #include <linux/completion.h> @@ -126,8 +130,7 @@ struct rdma_id_private { struct completion comp; atomic_t refcount; - wait_queue_head_t wait_remove; - atomic_t dev_remove; + struct mutex handler_mutex; int backlog; int timeout_ms; @@ -351,26 +354,15 @@ static void cma_deref_id(struct rdma_id_private *id_priv) complete(&id_priv->comp); } -static int cma_disable_remove(struct rdma_id_private *id_priv, +static int cma_disable_callback(struct rdma_id_private *id_priv, enum cma_state state) { - unsigned long flags; - int ret; - - spin_lock_irqsave(&id_priv->lock, flags); - if (id_priv->state == state) { - atomic_inc(&id_priv->dev_remove); - ret = 0; - } else - ret = -EINVAL; - spin_unlock_irqrestore(&id_priv->lock, flags); - return ret; -} - -static void cma_enable_remove(struct rdma_id_private *id_priv) -{ - if (atomic_dec_and_test(&id_priv->dev_remove)) - wake_up(&id_priv->wait_remove); + mutex_lock(&id_priv->handler_mutex); + if (id_priv->state != state) { + mutex_unlock(&id_priv->handler_mutex); + return -EINVAL; + } + return 0; } static int cma_has_cm_dev(struct rdma_id_private *id_priv) @@ -395,8 +387,7 @@ struct rdma_cm_id *rdma_create_id(rdma_cm_event_handler event_handler, mutex_init(&id_priv->qp_mutex); init_completion(&id_priv->comp); atomic_set(&id_priv->refcount, 1); - init_waitqueue_head(&id_priv->wait_remove); - atomic_set(&id_priv->dev_remove, 0); + mutex_init(&id_priv->handler_mutex); INIT_LIST_HEAD(&id_priv->listen_list); INIT_LIST_HEAD(&id_priv->mc_list); get_random_bytes(&id_priv->seq_num, sizeof id_priv->seq_num); @@ -923,7 +914,7 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) struct rdma_cm_event event; int ret = 0; - if (cma_disable_remove(id_priv, CMA_CONNECT)) + if (cma_disable_callback(id_priv, CMA_CONNECT)) return 0; memset(&event, 0, sizeof event); @@ -970,7 +961,7 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) event.param.conn.private_data_len = IB_CM_REJ_PRIVATE_DATA_SIZE; break; default: - printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d", + printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d\n", ib_event->event); goto out; } @@ -980,12 +971,12 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) /* Destroy the CM ID by returning a non-zero value. */ id_priv->cm_id.ib = NULL; cma_exch(id_priv, CMA_DESTROYING); - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); rdma_destroy_id(&id_priv->id); return ret; } out: - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); return ret; } @@ -998,6 +989,7 @@ static struct rdma_id_private *cma_new_conn_id(struct rdma_cm_id *listen_id, union cma_ip_addr *src, *dst; __be16 port; u8 ip_ver; + int ret; if (cma_get_net_info(ib_event->private_data, listen_id->ps, &ip_ver, &port, &src, &dst)) @@ -1022,10 +1014,11 @@ static struct rdma_id_private *cma_new_conn_id(struct rdma_cm_id *listen_id, if (rt->num_paths == 2) rt->path_rec[1] = *ib_event->param.req_rcvd.alternate_path; - ib_addr_set_sgid(&rt->addr.dev_addr, &rt->path_rec[0].sgid); ib_addr_set_dgid(&rt->addr.dev_addr, &rt->path_rec[0].dgid); - ib_addr_set_pkey(&rt->addr.dev_addr, be16_to_cpu(rt->path_rec[0].pkey)); - rt->addr.dev_addr.dev_type = RDMA_NODE_IB_CA; + ret = rdma_translate_ip(&id->route.addr.src_addr, + &id->route.addr.dev_addr); + if (ret) + goto destroy_id; id_priv = container_of(id, struct rdma_id_private, id); id_priv->state = CMA_CONNECT; @@ -1095,7 +1088,7 @@ static int cma_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) int offset, ret; listen_id = cm_id->context; - if (cma_disable_remove(listen_id, CMA_LISTEN)) + if (cma_disable_callback(listen_id, CMA_LISTEN)) return -ECONNABORTED; memset(&event, 0, sizeof event); @@ -1116,7 +1109,7 @@ static int cma_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) goto out; } - atomic_inc(&conn_id->dev_remove); + mutex_lock_nested(&conn_id->handler_mutex, SINGLE_DEPTH_NESTING); mutex_lock(&lock); ret = cma_acquire_dev(conn_id); mutex_unlock(&lock); @@ -1138,7 +1131,7 @@ static int cma_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) !cma_is_ud_ps(conn_id->id.ps)) ib_send_cm_mra(cm_id, CMA_CM_MRA_SETTING, NULL, 0); mutex_unlock(&lock); - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); goto out; } @@ -1147,11 +1140,11 @@ static int cma_req_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) release_conn_id: cma_exch(conn_id, CMA_DESTROYING); - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); rdma_destroy_id(&conn_id->id); out: - cma_enable_remove(listen_id); + mutex_unlock(&listen_id->handler_mutex); return ret; } @@ -1217,7 +1210,7 @@ static int cma_iw_handler(struct iw_cm_id *iw_id, struct iw_cm_event *iw_event) struct sockaddr_in *sin; int ret = 0; - if (cma_disable_remove(id_priv, CMA_CONNECT)) + if (cma_disable_callback(id_priv, CMA_CONNECT)) return 0; memset(&event, 0, sizeof event); @@ -1261,12 +1254,12 @@ static int cma_iw_handler(struct iw_cm_id *iw_id, struct iw_cm_event *iw_event) /* Destroy the CM ID by returning a non-zero value. */ id_priv->cm_id.iw = NULL; cma_exch(id_priv, CMA_DESTROYING); - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); rdma_destroy_id(&id_priv->id); return ret; } - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); return ret; } @@ -1282,7 +1275,7 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id, struct ib_device_attr attr; listen_id = cm_id->context; - if (cma_disable_remove(listen_id, CMA_LISTEN)) + if (cma_disable_callback(listen_id, CMA_LISTEN)) return -ECONNABORTED; /* Create a new RDMA id for the new IW CM ID */ @@ -1294,19 +1287,19 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id, goto out; } conn_id = container_of(new_cm_id, struct rdma_id_private, id); - atomic_inc(&conn_id->dev_remove); + mutex_lock_nested(&conn_id->handler_mutex, SINGLE_DEPTH_NESTING); conn_id->state = CMA_CONNECT; dev = ip_dev_find(&init_net, iw_event->local_addr.sin_addr.s_addr); if (!dev) { ret = -EADDRNOTAVAIL; - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); rdma_destroy_id(new_cm_id); goto out; } ret = rdma_copy_addr(&conn_id->id.route.addr.dev_addr, dev, NULL); if (ret) { - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); rdma_destroy_id(new_cm_id); goto out; } @@ -1315,7 +1308,7 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id, ret = cma_acquire_dev(conn_id); mutex_unlock(&lock); if (ret) { - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); rdma_destroy_id(new_cm_id); goto out; } @@ -1331,7 +1324,7 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id, ret = ib_query_device(conn_id->id.device, &attr); if (ret) { - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); rdma_destroy_id(new_cm_id); goto out; } @@ -1347,14 +1340,17 @@ static int iw_conn_req_handler(struct iw_cm_id *cm_id, /* User wants to destroy the CM ID */ conn_id->cm_id.iw = NULL; cma_exch(conn_id, CMA_DESTROYING); - cma_enable_remove(conn_id); + mutex_unlock(&conn_id->handler_mutex); rdma_destroy_id(&conn_id->id); + goto out; } + mutex_unlock(&conn_id->handler_mutex); + out: if (dev) dev_put(dev); - cma_enable_remove(listen_id); + mutex_unlock(&listen_id->handler_mutex); return ret; } @@ -1446,7 +1442,7 @@ static void cma_listen_on_dev(struct rdma_id_private *id_priv, ret = rdma_listen(id, id_priv->backlog); if (ret) printk(KERN_WARNING "RDMA CMA: cma_listen_on_dev, error %d, " - "listening on device %s", ret, cma_dev->device->name); + "listening on device %s\n", ret, cma_dev->device->name); } static void cma_listen_on_all(struct rdma_id_private *id_priv) @@ -1586,7 +1582,7 @@ static void cma_work_handler(struct work_struct *_work) struct rdma_id_private *id_priv = work->id; int destroy = 0; - atomic_inc(&id_priv->dev_remove); + mutex_lock(&id_priv->handler_mutex); if (!cma_comp_exch(id_priv, work->old_state, work->new_state)) goto out; @@ -1595,7 +1591,7 @@ static void cma_work_handler(struct work_struct *_work) destroy = 1; } out: - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); cma_deref_id(id_priv); if (destroy) rdma_destroy_id(&id_priv->id); @@ -1758,7 +1754,7 @@ static void addr_handler(int status, struct sockaddr *src_addr, struct rdma_cm_event event; memset(&event, 0, sizeof event); - atomic_inc(&id_priv->dev_remove); + mutex_lock(&id_priv->handler_mutex); /* * Grab mutex to block rdma_destroy_id() from removing the device while @@ -1787,13 +1783,13 @@ static void addr_handler(int status, struct sockaddr *src_addr, if (id_priv->id.event_handler(&id_priv->id, &event)) { cma_exch(id_priv, CMA_DESTROYING); - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); cma_deref_id(id_priv); rdma_destroy_id(&id_priv->id); return; } out: - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); cma_deref_id(id_priv); } @@ -2120,7 +2116,7 @@ static int cma_sidr_rep_handler(struct ib_cm_id *cm_id, struct ib_cm_sidr_rep_event_param *rep = &ib_event->param.sidr_rep_rcvd; int ret = 0; - if (cma_disable_remove(id_priv, CMA_CONNECT)) + if (cma_disable_callback(id_priv, CMA_CONNECT)) return 0; memset(&event, 0, sizeof event); @@ -2151,7 +2147,7 @@ static int cma_sidr_rep_handler(struct ib_cm_id *cm_id, event.status = 0; break; default: - printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d", + printk(KERN_ERR "RDMA CMA: unexpected IB CM event: %d\n", ib_event->event); goto out; } @@ -2161,12 +2157,12 @@ static int cma_sidr_rep_handler(struct ib_cm_id *cm_id, /* Destroy the CM ID by returning a non-zero value. */ id_priv->cm_id.ib = NULL; cma_exch(id_priv, CMA_DESTROYING); - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); rdma_destroy_id(&id_priv->id); return ret; } out: - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); return ret; } @@ -2564,8 +2560,8 @@ static int cma_ib_mc_handler(int status, struct ib_sa_multicast *multicast) int ret; id_priv = mc->id_priv; - if (cma_disable_remove(id_priv, CMA_ADDR_BOUND) && - cma_disable_remove(id_priv, CMA_ADDR_RESOLVED)) + if (cma_disable_callback(id_priv, CMA_ADDR_BOUND) && + cma_disable_callback(id_priv, CMA_ADDR_RESOLVED)) return 0; mutex_lock(&id_priv->qp_mutex); @@ -2590,12 +2586,12 @@ static int cma_ib_mc_handler(int status, struct ib_sa_multicast *multicast) ret = id_priv->id.event_handler(&id_priv->id, &event); if (ret) { cma_exch(id_priv, CMA_DESTROYING); - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); rdma_destroy_id(&id_priv->id); return 0; } - cma_enable_remove(id_priv); + mutex_unlock(&id_priv->handler_mutex); return 0; } @@ -2754,6 +2750,7 @@ static int cma_remove_id_dev(struct rdma_id_private *id_priv) { struct rdma_cm_event event; enum cma_state state; + int ret = 0; /* Record that we want to remove the device */ state = cma_exch(id_priv, CMA_DEVICE_REMOVAL); @@ -2761,15 +2758,18 @@ static int cma_remove_id_dev(struct rdma_id_private *id_priv) return 0; cma_cancel_operation(id_priv, state); - wait_event(id_priv->wait_remove, !atomic_read(&id_priv->dev_remove)); + mutex_lock(&id_priv->handler_mutex); /* Check for destruction from another callback. */ if (!cma_comp(id_priv, CMA_DEVICE_REMOVAL)) - return 0; + goto out; memset(&event, 0, sizeof event); event.event = RDMA_CM_EVENT_DEVICE_REMOVAL; - return id_priv->id.event_handler(&id_priv->id, &event); + ret = id_priv->id.event_handler(&id_priv->id, &event); +out: + mutex_unlock(&id_priv->handler_mutex); + return ret; } static void cma_process_remove(struct cma_device *cma_dev) diff --git a/drivers/infiniband/core/core_priv.h b/drivers/infiniband/core/core_priv.h index 7ad47a4b166b..05ac36e6acdb 100644 --- a/drivers/infiniband/core/core_priv.h +++ b/drivers/infiniband/core/core_priv.h @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: core_priv.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef _CORE_PRIV_H diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index 5ac5ffee05cb..7913b804311e 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: device.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/module.h> diff --git a/drivers/infiniband/core/fmr_pool.c b/drivers/infiniband/core/fmr_pool.c index 1286dc1b98b2..4507043d24c8 100644 --- a/drivers/infiniband/core/fmr_pool.c +++ b/drivers/infiniband/core/fmr_pool.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: fmr_pool.c 2730 2005-06-28 16:43:03Z sean.hefty $ */ #include <linux/errno.h> diff --git a/drivers/infiniband/core/mad_priv.h b/drivers/infiniband/core/mad_priv.h index 8b75010016ec..05ce331733b0 100644 --- a/drivers/infiniband/core/mad_priv.h +++ b/drivers/infiniband/core/mad_priv.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mad_priv.h 5596 2006-03-03 01:00:07Z sean.hefty $ */ #ifndef __IB_MAD_PRIV_H__ diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c index a5e2a310f312..d0ef7d61c037 100644 --- a/drivers/infiniband/core/mad_rmpp.c +++ b/drivers/infiniband/core/mad_rmpp.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mad_rmpp.c 1921 2005-03-02 22:58:44Z sean.hefty $ */ #include "mad_priv.h" diff --git a/drivers/infiniband/core/mad_rmpp.h b/drivers/infiniband/core/mad_rmpp.h index f0616fd22494..3d336bff1148 100644 --- a/drivers/infiniband/core/mad_rmpp.h +++ b/drivers/infiniband/core/mad_rmpp.h @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mad_rmpp.h 1921 2005-02-25 22:58:44Z sean.hefty $ */ #ifndef __MAD_RMPP_H__ diff --git a/drivers/infiniband/core/packer.c b/drivers/infiniband/core/packer.c index c972d7235764..019bd4b0863e 100644 --- a/drivers/infiniband/core/packer.c +++ b/drivers/infiniband/core/packer.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: packer.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/string.h> diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c index cf474ec27070..1341de793e51 100644 --- a/drivers/infiniband/core/sa_query.c +++ b/drivers/infiniband/core/sa_query.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: sa_query.c 2811 2005-07-06 18:11:43Z halr $ */ #include <linux/module.h> @@ -361,7 +359,7 @@ static void update_sm_ah(struct work_struct *work) { struct ib_sa_port *port = container_of(work, struct ib_sa_port, update_task); - struct ib_sa_sm_ah *new_ah, *old_ah; + struct ib_sa_sm_ah *new_ah; struct ib_port_attr port_attr; struct ib_ah_attr ah_attr; @@ -397,12 +395,9 @@ static void update_sm_ah(struct work_struct *work) } spin_lock_irq(&port->ah_lock); - old_ah = port->sm_ah; port->sm_ah = new_ah; spin_unlock_irq(&port->ah_lock); - if (old_ah) - kref_put(&old_ah->ref, free_sm_ah); } static void ib_sa_event(struct ib_event_handler *handler, struct ib_event *event) @@ -413,8 +408,17 @@ static void ib_sa_event(struct ib_event_handler *handler, struct ib_event *event event->event == IB_EVENT_PKEY_CHANGE || event->event == IB_EVENT_SM_CHANGE || event->event == IB_EVENT_CLIENT_REREGISTER) { - struct ib_sa_device *sa_dev; - sa_dev = container_of(handler, typeof(*sa_dev), event_handler); + unsigned long flags; + struct ib_sa_device *sa_dev = + container_of(handler, typeof(*sa_dev), event_handler); + struct ib_sa_port *port = + &sa_dev->port[event->element.port_num - sa_dev->start_port]; + + spin_lock_irqsave(&port->ah_lock, flags); + if (port->sm_ah) + kref_put(&port->sm_ah->ref, free_sm_ah); + port->sm_ah = NULL; + spin_unlock_irqrestore(&port->ah_lock, flags); schedule_work(&sa_dev->port[event->element.port_num - sa_dev->start_port].update_task); @@ -519,6 +523,10 @@ static int alloc_mad(struct ib_sa_query *query, gfp_t gfp_mask) unsigned long flags; spin_lock_irqsave(&query->port->ah_lock, flags); + if (!query->port->sm_ah) { + spin_unlock_irqrestore(&query->port->ah_lock, flags); + return -EAGAIN; + } kref_get(&query->port->sm_ah->ref); query->sm_ah = query->port->sm_ah; spin_unlock_irqrestore(&query->port->ah_lock, flags); diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c index 95756551cf7c..4d1042115598 100644 --- a/drivers/infiniband/core/sysfs.c +++ b/drivers/infiniband/core/sysfs.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: sysfs.c 1349 2004-12-16 21:09:43Z roland $ */ #include "core_priv.h" @@ -665,6 +663,120 @@ static struct class ib_class = { .dev_uevent = ib_device_uevent, }; +/* Show a given an attribute in the statistics group */ +static ssize_t show_protocol_stat(const struct device *device, + struct device_attribute *attr, char *buf, + unsigned offset) +{ + struct ib_device *dev = container_of(device, struct ib_device, dev); + union rdma_protocol_stats stats; + ssize_t ret; + + ret = dev->get_protocol_stats(dev, &stats); + if (ret) + return ret; + + return sprintf(buf, "%llu\n", + (unsigned long long) ((u64 *) &stats)[offset]); +} + +/* generate a read-only iwarp statistics attribute */ +#define IW_STATS_ENTRY(name) \ +static ssize_t show_##name(struct device *device, \ + struct device_attribute *attr, char *buf) \ +{ \ + return show_protocol_stat(device, attr, buf, \ + offsetof(struct iw_protocol_stats, name) / \ + sizeof (u64)); \ +} \ +static DEVICE_ATTR(name, S_IRUGO, show_##name, NULL) + +IW_STATS_ENTRY(ipInReceives); +IW_STATS_ENTRY(ipInHdrErrors); +IW_STATS_ENTRY(ipInTooBigErrors); +IW_STATS_ENTRY(ipInNoRoutes); +IW_STATS_ENTRY(ipInAddrErrors); +IW_STATS_ENTRY(ipInUnknownProtos); +IW_STATS_ENTRY(ipInTruncatedPkts); +IW_STATS_ENTRY(ipInDiscards); +IW_STATS_ENTRY(ipInDelivers); +IW_STATS_ENTRY(ipOutForwDatagrams); +IW_STATS_ENTRY(ipOutRequests); +IW_STATS_ENTRY(ipOutDiscards); +IW_STATS_ENTRY(ipOutNoRoutes); +IW_STATS_ENTRY(ipReasmTimeout); +IW_STATS_ENTRY(ipReasmReqds); +IW_STATS_ENTRY(ipReasmOKs); +IW_STATS_ENTRY(ipReasmFails); +IW_STATS_ENTRY(ipFragOKs); +IW_STATS_ENTRY(ipFragFails); +IW_STATS_ENTRY(ipFragCreates); +IW_STATS_ENTRY(ipInMcastPkts); +IW_STATS_ENTRY(ipOutMcastPkts); +IW_STATS_ENTRY(ipInBcastPkts); +IW_STATS_ENTRY(ipOutBcastPkts); +IW_STATS_ENTRY(tcpRtoAlgorithm); +IW_STATS_ENTRY(tcpRtoMin); +IW_STATS_ENTRY(tcpRtoMax); +IW_STATS_ENTRY(tcpMaxConn); +IW_STATS_ENTRY(tcpActiveOpens); +IW_STATS_ENTRY(tcpPassiveOpens); +IW_STATS_ENTRY(tcpAttemptFails); +IW_STATS_ENTRY(tcpEstabResets); +IW_STATS_ENTRY(tcpCurrEstab); +IW_STATS_ENTRY(tcpInSegs); +IW_STATS_ENTRY(tcpOutSegs); +IW_STATS_ENTRY(tcpRetransSegs); +IW_STATS_ENTRY(tcpInErrs); +IW_STATS_ENTRY(tcpOutRsts); + +static struct attribute *iw_proto_stats_attrs[] = { + &dev_attr_ipInReceives.attr, + &dev_attr_ipInHdrErrors.attr, + &dev_attr_ipInTooBigErrors.attr, + &dev_attr_ipInNoRoutes.attr, + &dev_attr_ipInAddrErrors.attr, + &dev_attr_ipInUnknownProtos.attr, + &dev_attr_ipInTruncatedPkts.attr, + &dev_attr_ipInDiscards.attr, + &dev_attr_ipInDelivers.attr, + &dev_attr_ipOutForwDatagrams.attr, + &dev_attr_ipOutRequests.attr, + &dev_attr_ipOutDiscards.attr, + &dev_attr_ipOutNoRoutes.attr, + &dev_attr_ipReasmTimeout.attr, + &dev_attr_ipReasmReqds.attr, + &dev_attr_ipReasmOKs.attr, + &dev_attr_ipReasmFails.attr, + &dev_attr_ipFragOKs.attr, + &dev_attr_ipFragFails.attr, + &dev_attr_ipFragCreates.attr, + &dev_attr_ipInMcastPkts.attr, + &dev_attr_ipOutMcastPkts.attr, + &dev_attr_ipInBcastPkts.attr, + &dev_attr_ipOutBcastPkts.attr, + &dev_attr_tcpRtoAlgorithm.attr, + &dev_attr_tcpRtoMin.attr, + &dev_attr_tcpRtoMax.attr, + &dev_attr_tcpMaxConn.attr, + &dev_attr_tcpActiveOpens.attr, + &dev_attr_tcpPassiveOpens.attr, + &dev_attr_tcpAttemptFails.attr, + &dev_attr_tcpEstabResets.attr, + &dev_attr_tcpCurrEstab.attr, + &dev_attr_tcpInSegs.attr, + &dev_attr_tcpOutSegs.attr, + &dev_attr_tcpRetransSegs.attr, + &dev_attr_tcpInErrs.attr, + &dev_attr_tcpOutRsts.attr, + NULL +}; + +static struct attribute_group iw_stats_group = { + .name = "proto_stats", + .attrs = iw_proto_stats_attrs, +}; + int ib_device_register_sysfs(struct ib_device *device) { struct device *class_dev = &device->dev; @@ -707,6 +819,12 @@ int ib_device_register_sysfs(struct ib_device *device) } } + if (device->node_type == RDMA_NODE_RNIC && device->get_protocol_stats) { + ret = sysfs_create_group(&class_dev->kobj, &iw_stats_group); + if (ret) + goto err_put; + } + return 0; err_put: diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c index b25675faaaf5..9494005d1c9a 100644 --- a/drivers/infiniband/core/ucm.c +++ b/drivers/infiniband/core/ucm.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ucm.c 4311 2005-12-05 18:42:01Z sean.hefty $ */ #include <linux/completion.h> diff --git a/drivers/infiniband/core/ud_header.c b/drivers/infiniband/core/ud_header.c index 997c07db6d8f..8ec7876bedcf 100644 --- a/drivers/infiniband/core/ud_header.c +++ b/drivers/infiniband/core/ud_header.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ud_header.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/errno.h> diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index a1768dbb0720..6f7c096abf13 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: uverbs_mem.c 2743 2005-06-28 22:27:59Z roland $ */ #include <linux/mm.h> diff --git a/drivers/infiniband/core/user_mad.c b/drivers/infiniband/core/user_mad.c index 208c7f34323c..268a2d23b7c9 100644 --- a/drivers/infiniband/core/user_mad.c +++ b/drivers/infiniband/core/user_mad.c @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: user_mad.c 5596 2006-03-03 01:00:07Z sean.hefty $ */ #include <linux/module.h> diff --git a/drivers/infiniband/core/uverbs.h b/drivers/infiniband/core/uverbs.h index 376a57ce1b40..b3ea9587dc80 100644 --- a/drivers/infiniband/core/uverbs.h +++ b/drivers/infiniband/core/uverbs.h @@ -32,8 +32,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: uverbs.h 2559 2005-06-06 19:43:16Z roland $ */ #ifndef UVERBS_H diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 2c3bff5fe867..56feab6c251e 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: uverbs_cmd.c 2708 2005-06-24 17:27:21Z roland $ */ #include <linux/file.h> @@ -919,7 +917,7 @@ ssize_t ib_uverbs_poll_cq(struct ib_uverbs_file *file, resp->wc[i].opcode = wc[i].opcode; resp->wc[i].vendor_err = wc[i].vendor_err; resp->wc[i].byte_len = wc[i].byte_len; - resp->wc[i].imm_data = (__u32 __force) wc[i].imm_data; + resp->wc[i].ex.imm_data = (__u32 __force) wc[i].ex.imm_data; resp->wc[i].qp_num = wc[i].qp->qp_num; resp->wc[i].src_qp = wc[i].src_qp; resp->wc[i].wc_flags = wc[i].wc_flags; diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 0f34858e31e7..aeee856c4060 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -32,8 +32,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: uverbs_main.c 2733 2005-06-28 19:14:34Z roland $ */ #include <linux/module.h> diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 05042089de6e..a7da9be43e61 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -34,8 +34,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: verbs.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/errno.h> @@ -317,7 +315,6 @@ static const struct { } qp_state_table[IB_QPS_ERR + 1][IB_QPS_ERR + 1] = { [IB_QPS_RESET] = { [IB_QPS_RESET] = { .valid = 1 }, - [IB_QPS_ERR] = { .valid = 1 }, [IB_QPS_INIT] = { .valid = 1, .req_param = { @@ -755,6 +752,52 @@ int ib_dereg_mr(struct ib_mr *mr) } EXPORT_SYMBOL(ib_dereg_mr); +struct ib_mr *ib_alloc_fast_reg_mr(struct ib_pd *pd, int max_page_list_len) +{ + struct ib_mr *mr; + + if (!pd->device->alloc_fast_reg_mr) + return ERR_PTR(-ENOSYS); + + mr = pd->device->alloc_fast_reg_mr(pd, max_page_list_len); + + if (!IS_ERR(mr)) { + mr->device = pd->device; + mr->pd = pd; + mr->uobject = NULL; + atomic_inc(&pd->usecnt); + atomic_set(&mr->usecnt, 0); + } + + return mr; +} +EXPORT_SYMBOL(ib_alloc_fast_reg_mr); + +struct ib_fast_reg_page_list *ib_alloc_fast_reg_page_list(struct ib_device *device, + int max_page_list_len) +{ + struct ib_fast_reg_page_list *page_list; + + if (!device->alloc_fast_reg_page_list) + return ERR_PTR(-ENOSYS); + + page_list = device->alloc_fast_reg_page_list(device, max_page_list_len); + + if (!IS_ERR(page_list)) { + page_list->device = device; + page_list->max_page_list_len = max_page_list_len; + } + + return page_list; +} +EXPORT_SYMBOL(ib_alloc_fast_reg_page_list); + +void ib_free_fast_reg_page_list(struct ib_fast_reg_page_list *page_list) +{ + page_list->device->free_fast_reg_page_list(page_list); +} +EXPORT_SYMBOL(ib_free_fast_reg_page_list); + /* Memory windows */ struct ib_mw *ib_alloc_mw(struct ib_pd *pd) diff --git a/drivers/infiniband/hw/amso1100/c2_rnic.c b/drivers/infiniband/hw/amso1100/c2_rnic.c index b1441aeb60c2..dd05c4835642 100644 --- a/drivers/infiniband/hw/amso1100/c2_rnic.c +++ b/drivers/infiniband/hw/amso1100/c2_rnic.c @@ -454,7 +454,7 @@ int __devinit c2_rnic_init(struct c2_dev *c2dev) (IB_DEVICE_RESIZE_MAX_WR | IB_DEVICE_CURR_QP_STATE_MOD | IB_DEVICE_SYS_IMAGE_GUID | - IB_DEVICE_ZERO_STAG | + IB_DEVICE_LOCAL_DMA_LKEY | IB_DEVICE_MEM_WINDOW); /* Allocate the qptr_array */ diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.c b/drivers/infiniband/hw/cxgb3/cxio_hal.c index 3f441fc57c17..f6d5747153a5 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.c +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.c @@ -145,7 +145,9 @@ static int cxio_hal_clear_qp_ctx(struct cxio_rdev *rdev_p, u32 qpid) } wqe = (struct t3_modify_qp_wr *) skb_put(skb, sizeof(*wqe)); memset(wqe, 0, sizeof(*wqe)); - build_fw_riwrh((struct fw_riwrh *) wqe, T3_WR_QP_MOD, 3, 0, qpid, 7); + build_fw_riwrh((struct fw_riwrh *) wqe, T3_WR_QP_MOD, + T3_COMPLETION_FLAG | T3_NOTIFY_FLAG, 0, qpid, 7, + T3_SOPEOP); wqe->flags = cpu_to_be32(MODQP_WRITE_EC); sge_cmd = qpid << 8 | 3; wqe->sge_cmd = cpu_to_be64(sge_cmd); @@ -276,7 +278,7 @@ int cxio_create_qp(struct cxio_rdev *rdev_p, u32 kernel_domain, if (!wq->qpid) return -ENOMEM; - wq->rq = kzalloc(depth * sizeof(u64), GFP_KERNEL); + wq->rq = kzalloc(depth * sizeof(struct t3_swrq), GFP_KERNEL); if (!wq->rq) goto err1; @@ -300,6 +302,7 @@ int cxio_create_qp(struct cxio_rdev *rdev_p, u32 kernel_domain, if (!kernel_domain) wq->udb = (u64)rdev_p->rnic_info.udbell_physbase + (wq->qpid << rdev_p->qpshift); + wq->rdev = rdev_p; PDBG("%s qpid 0x%x doorbell 0x%p udb 0x%llx\n", __func__, wq->qpid, wq->doorbell, (unsigned long long) wq->udb); return 0; @@ -558,7 +561,7 @@ static int cxio_hal_init_ctrl_qp(struct cxio_rdev *rdev_p) wqe = (struct t3_modify_qp_wr *) skb_put(skb, sizeof(*wqe)); memset(wqe, 0, sizeof(*wqe)); build_fw_riwrh((struct fw_riwrh *) wqe, T3_WR_QP_MOD, 0, 0, - T3_CTL_QP_TID, 7); + T3_CTL_QP_TID, 7, T3_SOPEOP); wqe->flags = cpu_to_be32(MODQP_WRITE_EC); sge_cmd = (3ULL << 56) | FW_RI_SGEEC_START << 8 | 3; wqe->sge_cmd = cpu_to_be64(sge_cmd); @@ -674,7 +677,7 @@ static int cxio_hal_ctrl_qp_write_mem(struct cxio_rdev *rdev_p, u32 addr, build_fw_riwrh((struct fw_riwrh *) wqe, T3_WR_BP, flag, Q_GENBIT(rdev_p->ctrl_qp.wptr, T3_CTRL_QP_SIZE_LOG2), T3_CTRL_QP_ID, - wr_len); + wr_len, T3_SOPEOP); if (flag == T3_COMPLETION_FLAG) ring_doorbell(rdev_p->ctrl_qp.doorbell, T3_CTRL_QP_ID); len -= 96; @@ -816,6 +819,13 @@ int cxio_deallocate_window(struct cxio_rdev *rdev_p, u32 stag) 0, 0); } +int cxio_allocate_stag(struct cxio_rdev *rdev_p, u32 *stag, u32 pdid, u32 pbl_size, u32 pbl_addr) +{ + *stag = T3_STAG_UNSET; + return __cxio_tpt_op(rdev_p, 0, stag, 0, pdid, TPT_NON_SHARED_MR, + 0, 0, 0ULL, 0, 0, pbl_size, pbl_addr); +} + int cxio_rdma_init(struct cxio_rdev *rdev_p, struct t3_rdma_init_attr *attr) { struct t3_rdma_init_wr *wqe; @@ -1257,13 +1267,16 @@ proc_cqe: wq->sq_rptr = CQE_WRID_SQ_WPTR(*hw_cqe); PDBG("%s completing sq idx %ld\n", __func__, Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)); - *cookie = (wq->sq + - Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2))->wr_id; + *cookie = wq->sq[Q_PTR2IDX(wq->sq_rptr, wq->sq_size_log2)].wr_id; wq->sq_rptr++; } else { PDBG("%s completing rq idx %ld\n", __func__, Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)); - *cookie = *(wq->rq + Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)); + *cookie = wq->rq[Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)].wr_id; + if (wq->rq[Q_PTR2IDX(wq->rq_rptr, wq->rq_size_log2)].pbl_addr) + cxio_hal_pblpool_free(wq->rdev, + wq->rq[Q_PTR2IDX(wq->rq_rptr, + wq->rq_size_log2)].pbl_addr, T3_STAG0_PBL_SIZE); wq->rq_rptr++; } diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h index 6e128f6bab05..656fe47bc84f 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.h +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h @@ -45,15 +45,17 @@ #define T3_CTRL_QP_SIZE_LOG2 8 #define T3_CTRL_CQ_ID 0 -/* TBD */ #define T3_MAX_NUM_RI (1<<15) #define T3_MAX_NUM_QP (1<<15) #define T3_MAX_NUM_CQ (1<<15) #define T3_MAX_NUM_PD (1<<15) #define T3_MAX_PBL_SIZE 256 #define T3_MAX_RQ_SIZE 1024 +#define T3_MAX_QP_DEPTH (T3_MAX_RQ_SIZE-1) +#define T3_MAX_CQ_DEPTH 8192 #define T3_MAX_NUM_STAG (1<<15) #define T3_MAX_MR_SIZE 0x100000000ULL +#define T3_PAGESIZE_MASK 0xffff000 /* 4KB-128MB */ #define T3_STAG_UNSET 0xffffffff @@ -165,6 +167,7 @@ int cxio_reregister_phys_mem(struct cxio_rdev *rdev, u32 * stag, u32 pdid, int cxio_dereg_mem(struct cxio_rdev *rdev, u32 stag, u32 pbl_size, u32 pbl_addr); int cxio_allocate_window(struct cxio_rdev *rdev, u32 * stag, u32 pdid); +int cxio_allocate_stag(struct cxio_rdev *rdev, u32 *stag, u32 pdid, u32 pbl_size, u32 pbl_addr); int cxio_deallocate_window(struct cxio_rdev *rdev, u32 stag); int cxio_rdma_init(struct cxio_rdev *rdev, struct t3_rdma_init_attr *attr); void cxio_register_ev_cb(cxio_hal_ev_callback_func_t ev_cb); diff --git a/drivers/infiniband/hw/cxgb3/cxio_wr.h b/drivers/infiniband/hw/cxgb3/cxio_wr.h index f1a25a821a45..04618f7bfbb3 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_wr.h +++ b/drivers/infiniband/hw/cxgb3/cxio_wr.h @@ -39,6 +39,9 @@ #define T3_MAX_SGE 4 #define T3_MAX_INLINE 64 +#define T3_STAG0_PBL_SIZE (2 * T3_MAX_SGE << 3) +#define T3_STAG0_MAX_PBE_LEN (128 * 1024 * 1024) +#define T3_STAG0_PAGE_SHIFT 15 #define Q_EMPTY(rptr,wptr) ((rptr)==(wptr)) #define Q_FULL(rptr,wptr,size_log2) ( (((wptr)-(rptr))>>(size_log2)) && \ @@ -72,7 +75,8 @@ enum t3_wr_opcode { T3_WR_BIND = FW_WROPCODE_RI_BIND_MW, T3_WR_RCV = FW_WROPCODE_RI_RECEIVE, T3_WR_INIT = FW_WROPCODE_RI_RDMA_INIT, - T3_WR_QP_MOD = FW_WROPCODE_RI_MODIFY_QP + T3_WR_QP_MOD = FW_WROPCODE_RI_MODIFY_QP, + T3_WR_FASTREG = FW_WROPCODE_RI_FASTREGISTER_MR } __attribute__ ((packed)); enum t3_rdma_opcode { @@ -89,7 +93,8 @@ enum t3_rdma_opcode { T3_FAST_REGISTER, T3_LOCAL_INV, T3_QP_MOD, - T3_BYPASS + T3_BYPASS, + T3_RDMA_READ_REQ_WITH_INV, } __attribute__ ((packed)); static inline enum t3_rdma_opcode wr2opcode(enum t3_wr_opcode wrop) @@ -103,6 +108,7 @@ static inline enum t3_rdma_opcode wr2opcode(enum t3_wr_opcode wrop) case T3_WR_BIND: return T3_BIND_MW; case T3_WR_INIT: return T3_RDMA_INIT; case T3_WR_QP_MOD: return T3_QP_MOD; + case T3_WR_FASTREG: return T3_FAST_REGISTER; default: break; } return -1; @@ -170,11 +176,54 @@ struct t3_send_wr { struct t3_sge sgl[T3_MAX_SGE]; /* 4+ */ }; +#define T3_MAX_FASTREG_DEPTH 24 +#define T3_MAX_FASTREG_FRAG 10 + +struct t3_fastreg_wr { + struct fw_riwrh wrh; /* 0 */ + union t3_wrid wrid; /* 1 */ + __be32 stag; /* 2 */ + __be32 len; + __be32 va_base_hi; /* 3 */ + __be32 va_base_lo_fbo; + __be32 page_type_perms; /* 4 */ + __be32 reserved1; + __be64 pbl_addrs[0]; /* 5+ */ +}; + +/* + * If a fastreg wr spans multiple wqes, then the 2nd fragment look like this. + */ +struct t3_pbl_frag { + struct fw_riwrh wrh; /* 0 */ + __be64 pbl_addrs[14]; /* 1..14 */ +}; + +#define S_FR_PAGE_COUNT 24 +#define M_FR_PAGE_COUNT 0xff +#define V_FR_PAGE_COUNT(x) ((x) << S_FR_PAGE_COUNT) +#define G_FR_PAGE_COUNT(x) ((((x) >> S_FR_PAGE_COUNT)) & M_FR_PAGE_COUNT) + +#define S_FR_PAGE_SIZE 16 +#define M_FR_PAGE_SIZE 0x1f +#define V_FR_PAGE_SIZE(x) ((x) << S_FR_PAGE_SIZE) +#define G_FR_PAGE_SIZE(x) ((((x) >> S_FR_PAGE_SIZE)) & M_FR_PAGE_SIZE) + +#define S_FR_TYPE 8 +#define M_FR_TYPE 0x1 +#define V_FR_TYPE(x) ((x) << S_FR_TYPE) +#define G_FR_TYPE(x) ((((x) >> S_FR_TYPE)) & M_FR_TYPE) + +#define S_FR_PERMS 0 +#define M_FR_PERMS 0xff +#define V_FR_PERMS(x) ((x) << S_FR_PERMS) +#define G_FR_PERMS(x) ((((x) >> S_FR_PERMS)) & M_FR_PERMS) + struct t3_local_inv_wr { struct fw_riwrh wrh; /* 0 */ union t3_wrid wrid; /* 1 */ __be32 stag; /* 2 */ - __be32 reserved3; + __be32 reserved; }; struct t3_rdma_write_wr { @@ -193,7 +242,8 @@ struct t3_rdma_read_wr { struct fw_riwrh wrh; /* 0 */ union t3_wrid wrid; /* 1 */ u8 rdmaop; /* 2 */ - u8 reserved[3]; + u8 local_inv; + u8 reserved[2]; __be32 rem_stag; __be64 rem_to; /* 3 */ __be32 local_stag; /* 4 */ @@ -201,18 +251,6 @@ struct t3_rdma_read_wr { __be64 local_to; /* 5 */ }; -enum t3_addr_type { - T3_VA_BASED_TO = 0x0, - T3_ZERO_BASED_TO = 0x1 -} __attribute__ ((packed)); - -enum t3_mem_perms { - T3_MEM_ACCESS_LOCAL_READ = 0x1, - T3_MEM_ACCESS_LOCAL_WRITE = 0x2, - T3_MEM_ACCESS_REM_READ = 0x4, - T3_MEM_ACCESS_REM_WRITE = 0x8 -} __attribute__ ((packed)); - struct t3_bind_mw_wr { struct fw_riwrh wrh; /* 0 */ union t3_wrid wrid; /* 1 */ @@ -336,6 +374,11 @@ struct t3_genbit { __be64 genbit; }; +struct t3_wq_in_err { + u64 flit[13]; + u64 err; +}; + enum rdma_init_wr_flags { MPA_INITIATOR = (1<<0), PRIV_QP = (1<<1), @@ -346,13 +389,16 @@ union t3_wr { struct t3_rdma_write_wr write; struct t3_rdma_read_wr read; struct t3_receive_wr recv; + struct t3_fastreg_wr fastreg; + struct t3_pbl_frag pbl_frag; struct t3_local_inv_wr local_inv; struct t3_bind_mw_wr bind; struct t3_bypass_wr bypass; struct t3_rdma_init_wr init; struct t3_modify_qp_wr qp_mod; struct t3_genbit genbit; - u64 flit[16]; + struct t3_wq_in_err wq_in_err; + __be64 flit[16]; }; #define T3_SQ_CQE_FLIT 13 @@ -366,12 +412,18 @@ static inline enum t3_wr_opcode fw_riwrh_opcode(struct fw_riwrh *wqe) return G_FW_RIWR_OP(be32_to_cpu(wqe->op_seop_flags)); } +enum t3_wr_hdr_bits { + T3_EOP = 1, + T3_SOP = 2, + T3_SOPEOP = T3_EOP|T3_SOP, +}; + static inline void build_fw_riwrh(struct fw_riwrh *wqe, enum t3_wr_opcode op, enum t3_wr_flags flags, u8 genbit, u32 tid, - u8 len) + u8 len, u8 sopeop) { wqe->op_seop_flags = cpu_to_be32(V_FW_RIWR_OP(op) | - V_FW_RIWR_SOPEOP(M_FW_RIWR_SOPEOP) | + V_FW_RIWR_SOPEOP(sopeop) | V_FW_RIWR_FLAGS(flags)); wmb(); wqe->gen_tid_len = cpu_to_be32(V_FW_RIWR_GEN(genbit) | @@ -404,6 +456,7 @@ enum tpt_addr_type { }; enum tpt_mem_perm { + TPT_MW_BIND = 0x10, TPT_LOCAL_READ = 0x8, TPT_LOCAL_WRITE = 0x4, TPT_REMOTE_READ = 0x2, @@ -615,6 +668,11 @@ struct t3_swsq { int signaled; }; +struct t3_swrq { + __u64 wr_id; + __u32 pbl_addr; +}; + /* * A T3 WQ implements both the SQ and RQ. */ @@ -631,14 +689,15 @@ struct t3_wq { u32 sq_wptr; /* sq_wptr - sq_rptr == count of */ u32 sq_rptr; /* pending wrs */ u32 sq_size_log2; /* sq size */ - u64 *rq; /* SW RQ (holds consumer wr_ids */ + struct t3_swrq *rq; /* SW RQ (holds consumer wr_ids */ u32 rq_wptr; /* rq_wptr - rq_rptr == count of */ u32 rq_rptr; /* pending wrs */ - u64 *rq_oldest_wr; /* oldest wr on the SW RQ */ + struct t3_swrq *rq_oldest_wr; /* oldest wr on the SW RQ */ u32 rq_size_log2; /* rq size */ u32 rq_addr; /* rq adapter address */ void __iomem *doorbell; /* kernel db */ u64 udb; /* user db if any */ + struct cxio_rdev *rdev; }; struct t3_cq { @@ -659,7 +718,7 @@ struct t3_cq { static inline void cxio_set_wq_in_error(struct t3_wq *wq) { - wq->queue->flit[13] = 1; + wq->queue->wq_in_err.err = 1; } static inline struct t3_cqe *cxio_next_hw_cqe(struct t3_cq *cq) diff --git a/drivers/infiniband/hw/cxgb3/iwch.c b/drivers/infiniband/hw/cxgb3/iwch.c index 71554eacb13c..4489c89d6710 100644 --- a/drivers/infiniband/hw/cxgb3/iwch.c +++ b/drivers/infiniband/hw/cxgb3/iwch.c @@ -71,18 +71,16 @@ static void rnic_init(struct iwch_dev *rnicp) idr_init(&rnicp->mmidr); spin_lock_init(&rnicp->lock); - rnicp->attr.vendor_id = 0x168; - rnicp->attr.vendor_part_id = 7; rnicp->attr.max_qps = T3_MAX_NUM_QP - 32; - rnicp->attr.max_wrs = (1UL << 24) - 1; + rnicp->attr.max_wrs = T3_MAX_QP_DEPTH; rnicp->attr.max_sge_per_wr = T3_MAX_SGE; rnicp->attr.max_sge_per_rdma_write_wr = T3_MAX_SGE; rnicp->attr.max_cqs = T3_MAX_NUM_CQ - 1; - rnicp->attr.max_cqes_per_cq = (1UL << 24) - 1; + rnicp->attr.max_cqes_per_cq = T3_MAX_CQ_DEPTH; rnicp->attr.max_mem_regs = cxio_num_stags(&rnicp->rdev); rnicp->attr.max_phys_buf_entries = T3_MAX_PBL_SIZE; rnicp->attr.max_pds = T3_MAX_NUM_PD - 1; - rnicp->attr.mem_pgsizes_bitmask = 0x7FFF; /* 4KB-128MB */ + rnicp->attr.mem_pgsizes_bitmask = T3_PAGESIZE_MASK; rnicp->attr.max_mr_size = T3_MAX_MR_SIZE; rnicp->attr.can_resize_wq = 0; rnicp->attr.max_rdma_reads_per_qp = 8; diff --git a/drivers/infiniband/hw/cxgb3/iwch.h b/drivers/infiniband/hw/cxgb3/iwch.h index d2409a505e8d..3773453b2cf0 100644 --- a/drivers/infiniband/hw/cxgb3/iwch.h +++ b/drivers/infiniband/hw/cxgb3/iwch.h @@ -48,8 +48,6 @@ struct iwch_qp; struct iwch_mr; struct iwch_rnic_attributes { - u32 vendor_id; - u32 vendor_part_id; u32 max_qps; u32 max_wrs; /* Max for any SQ/RQ */ u32 max_sge_per_wr; diff --git a/drivers/infiniband/hw/cxgb3/iwch_cq.c b/drivers/infiniband/hw/cxgb3/iwch_cq.c index 4ee8ccd0a9e5..cf5474ae68ff 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_cq.c +++ b/drivers/infiniband/hw/cxgb3/iwch_cq.c @@ -81,6 +81,7 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, wc->wr_id = cookie; wc->qp = &qhp->ibqp; wc->vendor_err = CQE_STATUS(cqe); + wc->wc_flags = 0; PDBG("%s qpid 0x%x type %d opcode %d status 0x%x wrid hi 0x%x " "lo 0x%x cookie 0x%llx\n", __func__, @@ -94,6 +95,11 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, else wc->byte_len = 0; wc->opcode = IB_WC_RECV; + if (CQE_OPCODE(cqe) == T3_SEND_WITH_INV || + CQE_OPCODE(cqe) == T3_SEND_WITH_SE_INV) { + wc->ex.invalidate_rkey = CQE_WRID_STAG(cqe); + wc->wc_flags |= IB_WC_WITH_INVALIDATE; + } } else { switch (CQE_OPCODE(cqe)) { case T3_RDMA_WRITE: @@ -105,17 +111,20 @@ static int iwch_poll_cq_one(struct iwch_dev *rhp, struct iwch_cq *chp, break; case T3_SEND: case T3_SEND_WITH_SE: + case T3_SEND_WITH_INV: + case T3_SEND_WITH_SE_INV: wc->opcode = IB_WC_SEND; break; case T3_BIND_MW: wc->opcode = IB_WC_BIND_MW; break; - /* these aren't supported yet */ - case T3_SEND_WITH_INV: - case T3_SEND_WITH_SE_INV: case T3_LOCAL_INV: + wc->opcode = IB_WC_LOCAL_INV; + break; case T3_FAST_REGISTER: + wc->opcode = IB_WC_FAST_REG_MR; + break; default: printk(KERN_ERR MOD "Unexpected opcode %d " "in the CQE received for QPID=0x%0x\n", diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index 95f82cfb6c54..b89640aa6e10 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -56,6 +56,7 @@ #include "iwch_provider.h" #include "iwch_cm.h" #include "iwch_user.h" +#include "common.h" static int iwch_modify_port(struct ib_device *ibdev, u8 port, int port_modify_mask, @@ -747,6 +748,7 @@ static struct ib_mw *iwch_alloc_mw(struct ib_pd *pd) mhp->attr.type = TPT_MW; mhp->attr.stag = stag; mmid = (stag) >> 8; + mhp->ibmw.rkey = stag; insert_handle(rhp, &rhp->mmidr, mhp, mmid); PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); return &(mhp->ibmw); @@ -768,6 +770,68 @@ static int iwch_dealloc_mw(struct ib_mw *mw) return 0; } +static struct ib_mr *iwch_alloc_fast_reg_mr(struct ib_pd *pd, int pbl_depth) +{ + struct iwch_dev *rhp; + struct iwch_pd *php; + struct iwch_mr *mhp; + u32 mmid; + u32 stag = 0; + int ret; + + php = to_iwch_pd(pd); + rhp = php->rhp; + mhp = kzalloc(sizeof(*mhp), GFP_KERNEL); + if (!mhp) + return ERR_PTR(-ENOMEM); + + mhp->rhp = rhp; + ret = iwch_alloc_pbl(mhp, pbl_depth); + if (ret) { + kfree(mhp); + return ERR_PTR(ret); + } + mhp->attr.pbl_size = pbl_depth; + ret = cxio_allocate_stag(&rhp->rdev, &stag, php->pdid, + mhp->attr.pbl_size, mhp->attr.pbl_addr); + if (ret) { + iwch_free_pbl(mhp); + kfree(mhp); + return ERR_PTR(ret); + } + mhp->attr.pdid = php->pdid; + mhp->attr.type = TPT_NON_SHARED_MR; + mhp->attr.stag = stag; + mhp->attr.state = 1; + mmid = (stag) >> 8; + mhp->ibmr.rkey = mhp->ibmr.lkey = stag; + insert_handle(rhp, &rhp->mmidr, mhp, mmid); + PDBG("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); + return &(mhp->ibmr); +} + +static struct ib_fast_reg_page_list *iwch_alloc_fastreg_pbl( + struct ib_device *device, + int page_list_len) +{ + struct ib_fast_reg_page_list *page_list; + + page_list = kmalloc(sizeof *page_list + page_list_len * sizeof(u64), + GFP_KERNEL); + if (!page_list) + return ERR_PTR(-ENOMEM); + + page_list->page_list = (u64 *)(page_list + 1); + page_list->max_page_list_len = page_list_len; + + return page_list; +} + +static void iwch_free_fastreg_pbl(struct ib_fast_reg_page_list *page_list) +{ + kfree(page_list); +} + static int iwch_destroy_qp(struct ib_qp *ib_qp) { struct iwch_dev *rhp; @@ -843,6 +907,15 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, */ sqsize = roundup_pow_of_two(attrs->cap.max_send_wr); wqsize = roundup_pow_of_two(rqsize + sqsize); + + /* + * Kernel users need more wq space for fastreg WRs which can take + * 2 WR fragments. + */ + ucontext = pd->uobject ? to_iwch_ucontext(pd->uobject->context) : NULL; + if (!ucontext && wqsize < (rqsize + (2 * sqsize))) + wqsize = roundup_pow_of_two(rqsize + + roundup_pow_of_two(attrs->cap.max_send_wr * 2)); PDBG("%s wqsize %d sqsize %d rqsize %d\n", __func__, wqsize, sqsize, rqsize); qhp = kzalloc(sizeof(*qhp), GFP_KERNEL); @@ -851,7 +924,6 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, qhp->wq.size_log2 = ilog2(wqsize); qhp->wq.rq_size_log2 = ilog2(rqsize); qhp->wq.sq_size_log2 = ilog2(sqsize); - ucontext = pd->uobject ? to_iwch_ucontext(pd->uobject->context) : NULL; if (cxio_create_qp(&rhp->rdev, !udata, &qhp->wq, ucontext ? &ucontext->uctx : &rhp->rdev.uctx)) { kfree(qhp); @@ -935,10 +1007,10 @@ static struct ib_qp *iwch_create_qp(struct ib_pd *pd, qhp->ibqp.qp_num = qhp->wq.qpid; init_timer(&(qhp->timer)); PDBG("%s sq_num_entries %d, rq_num_entries %d " - "qpid 0x%0x qhp %p dma_addr 0x%llx size %d\n", + "qpid 0x%0x qhp %p dma_addr 0x%llx size %d rq_addr 0x%x\n", __func__, qhp->attr.sq_num_entries, qhp->attr.rq_num_entries, qhp->wq.qpid, qhp, (unsigned long long) qhp->wq.dma_addr, - 1 << qhp->wq.size_log2); + 1 << qhp->wq.size_log2, qhp->wq.rq_addr); return &qhp->ibqp; } @@ -1023,6 +1095,29 @@ static int iwch_query_gid(struct ib_device *ibdev, u8 port, return 0; } +static u64 fw_vers_string_to_u64(struct iwch_dev *iwch_dev) +{ + struct ethtool_drvinfo info; + struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev; + char *cp, *next; + unsigned fw_maj, fw_min, fw_mic; + + rtnl_lock(); + lldev->ethtool_ops->get_drvinfo(lldev, &info); + rtnl_unlock(); + + next = info.fw_version + 1; + cp = strsep(&next, "."); + sscanf(cp, "%i", &fw_maj); + cp = strsep(&next, "."); + sscanf(cp, "%i", &fw_min); + cp = strsep(&next, "."); + sscanf(cp, "%i", &fw_mic); + + return (((u64)fw_maj & 0xffff) << 32) | ((fw_min & 0xffff) << 16) | + (fw_mic & 0xffff); +} + static int iwch_query_device(struct ib_device *ibdev, struct ib_device_attr *props) { @@ -1033,7 +1128,10 @@ static int iwch_query_device(struct ib_device *ibdev, dev = to_iwch_dev(ibdev); memset(props, 0, sizeof *props); memcpy(&props->sys_image_guid, dev->rdev.t3cdev_p->lldev->dev_addr, 6); + props->hw_ver = dev->rdev.t3cdev_p->type; + props->fw_ver = fw_vers_string_to_u64(dev); props->device_cap_flags = dev->device_cap_flags; + props->page_size_cap = dev->attr.mem_pgsizes_bitmask; props->vendor_id = (u32)dev->rdev.rnic_info.pdev->vendor; props->vendor_part_id = (u32)dev->rdev.rnic_info.pdev->device; props->max_mr_size = dev->attr.max_mr_size; @@ -1048,6 +1146,7 @@ static int iwch_query_device(struct ib_device *ibdev, props->max_mr = dev->attr.max_mem_regs; props->max_pd = dev->attr.max_pds; props->local_ca_ack_delay = 0; + props->max_fast_reg_page_list_len = T3_MAX_FASTREG_DEPTH; return 0; } @@ -1088,6 +1187,28 @@ static ssize_t show_rev(struct device *dev, struct device_attribute *attr, return sprintf(buf, "%d\n", iwch_dev->rdev.t3cdev_p->type); } +static int fw_supports_fastreg(struct iwch_dev *iwch_dev) +{ + struct ethtool_drvinfo info; + struct net_device *lldev = iwch_dev->rdev.t3cdev_p->lldev; + char *cp, *next; + unsigned fw_maj, fw_min; + + rtnl_lock(); + lldev->ethtool_ops->get_drvinfo(lldev, &info); + rtnl_unlock(); + + next = info.fw_version+1; + cp = strsep(&next, "."); + sscanf(cp, "%i", &fw_maj); + cp = strsep(&next, "."); + sscanf(cp, "%i", &fw_min); + + PDBG("%s maj %u min %u\n", __func__, fw_maj, fw_min); + + return fw_maj > 6 || (fw_maj == 6 && fw_min > 0); +} + static ssize_t show_fw_ver(struct device *dev, struct device_attribute *attr, char *buf) { struct iwch_dev *iwch_dev = container_of(dev, struct iwch_dev, @@ -1127,6 +1248,61 @@ static ssize_t show_board(struct device *dev, struct device_attribute *attr, iwch_dev->rdev.rnic_info.pdev->device); } +static int iwch_get_mib(struct ib_device *ibdev, + union rdma_protocol_stats *stats) +{ + struct iwch_dev *dev; + struct tp_mib_stats m; + int ret; + + PDBG("%s ibdev %p\n", __func__, ibdev); + dev = to_iwch_dev(ibdev); + ret = dev->rdev.t3cdev_p->ctl(dev->rdev.t3cdev_p, RDMA_GET_MIB, &m); + if (ret) + return -ENOSYS; + + memset(stats, 0, sizeof *stats); + stats->iw.ipInReceives = ((u64) m.ipInReceive_hi << 32) + + m.ipInReceive_lo; + stats->iw.ipInHdrErrors = ((u64) m.ipInHdrErrors_hi << 32) + + m.ipInHdrErrors_lo; + stats->iw.ipInAddrErrors = ((u64) m.ipInAddrErrors_hi << 32) + + m.ipInAddrErrors_lo; + stats->iw.ipInUnknownProtos = ((u64) m.ipInUnknownProtos_hi << 32) + + m.ipInUnknownProtos_lo; + stats->iw.ipInDiscards = ((u64) m.ipInDiscards_hi << 32) + + m.ipInDiscards_lo; + stats->iw.ipInDelivers = ((u64) m.ipInDelivers_hi << 32) + + m.ipInDelivers_lo; + stats->iw.ipOutRequests = ((u64) m.ipOutRequests_hi << 32) + + m.ipOutRequests_lo; + stats->iw.ipOutDiscards = ((u64) m.ipOutDiscards_hi << 32) + + m.ipOutDiscards_lo; + stats->iw.ipOutNoRoutes = ((u64) m.ipOutNoRoutes_hi << 32) + + m.ipOutNoRoutes_lo; + stats->iw.ipReasmTimeout = (u64) m.ipReasmTimeout; + stats->iw.ipReasmReqds = (u64) m.ipReasmReqds; + stats->iw.ipReasmOKs = (u64) m.ipReasmOKs; + stats->iw.ipReasmFails = (u64) m.ipReasmFails; + stats->iw.tcpActiveOpens = (u64) m.tcpActiveOpens; + stats->iw.tcpPassiveOpens = (u64) m.tcpPassiveOpens; + stats->iw.tcpAttemptFails = (u64) m.tcpAttemptFails; + stats->iw.tcpEstabResets = (u64) m.tcpEstabResets; + stats->iw.tcpOutRsts = (u64) m.tcpOutRsts; + stats->iw.tcpCurrEstab = (u64) m.tcpCurrEstab; + stats->iw.tcpInSegs = ((u64) m.tcpInSegs_hi << 32) + + m.tcpInSegs_lo; + stats->iw.tcpOutSegs = ((u64) m.tcpOutSegs_hi << 32) + + m.tcpOutSegs_lo; + stats->iw.tcpRetransSegs = ((u64) m.tcpRetransSeg_hi << 32) + + m.tcpRetransSeg_lo; + stats->iw.tcpInErrs = ((u64) m.tcpInErrs_hi << 32) + + m.tcpInErrs_lo; + stats->iw.tcpRtoMin = (u64) m.tcpRtoMin; + stats->iw.tcpRtoMax = (u64) m.tcpRtoMax; + return 0; +} + static DEVICE_ATTR(hw_rev, S_IRUGO, show_rev, NULL); static DEVICE_ATTR(fw_ver, S_IRUGO, show_fw_ver, NULL); static DEVICE_ATTR(hca_type, S_IRUGO, show_hca, NULL); @@ -1136,7 +1312,7 @@ static struct device_attribute *iwch_class_attributes[] = { &dev_attr_hw_rev, &dev_attr_fw_ver, &dev_attr_hca_type, - &dev_attr_board_id + &dev_attr_board_id, }; int iwch_register_device(struct iwch_dev *dev) @@ -1149,8 +1325,12 @@ int iwch_register_device(struct iwch_dev *dev) memset(&dev->ibdev.node_guid, 0, sizeof(dev->ibdev.node_guid)); memcpy(&dev->ibdev.node_guid, dev->rdev.t3cdev_p->lldev->dev_addr, 6); dev->ibdev.owner = THIS_MODULE; - dev->device_cap_flags = - (IB_DEVICE_ZERO_STAG | IB_DEVICE_MEM_WINDOW); + dev->device_cap_flags = IB_DEVICE_LOCAL_DMA_LKEY | IB_DEVICE_MEM_WINDOW; + + /* cxgb3 supports STag 0. */ + dev->ibdev.local_dma_lkey = 0; + if (fw_supports_fastreg(dev)) + dev->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS; dev->ibdev.uverbs_cmd_mask = (1ull << IB_USER_VERBS_CMD_GET_CONTEXT) | @@ -1202,15 +1382,16 @@ int iwch_register_device(struct iwch_dev *dev) dev->ibdev.alloc_mw = iwch_alloc_mw; dev->ibdev.bind_mw = iwch_bind_mw; dev->ibdev.dealloc_mw = iwch_dealloc_mw; - + dev->ibdev.alloc_fast_reg_mr = iwch_alloc_fast_reg_mr; + dev->ibdev.alloc_fast_reg_page_list = iwch_alloc_fastreg_pbl; + dev->ibdev.free_fast_reg_page_list = iwch_free_fastreg_pbl; dev->ibdev.attach_mcast = iwch_multicast_attach; dev->ibdev.detach_mcast = iwch_multicast_detach; dev->ibdev.process_mad = iwch_process_mad; - dev->ibdev.req_notify_cq = iwch_arm_cq; dev->ibdev.post_send = iwch_post_send; dev->ibdev.post_recv = iwch_post_receive; - + dev->ibdev.get_protocol_stats = iwch_get_mib; dev->ibdev.iwcm = kmalloc(sizeof(struct iw_cm_verbs), GFP_KERNEL); if (!dev->ibdev.iwcm) diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.h b/drivers/infiniband/hw/cxgb3/iwch_provider.h index 836163fc5429..f5ceca05c435 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.h +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.h @@ -296,14 +296,6 @@ static inline u32 iwch_ib_to_tpt_access(int acc) TPT_LOCAL_READ; } -static inline u32 iwch_ib_to_mwbind_access(int acc) -{ - return (acc & IB_ACCESS_REMOTE_WRITE ? T3_MEM_ACCESS_REM_WRITE : 0) | - (acc & IB_ACCESS_REMOTE_READ ? T3_MEM_ACCESS_REM_READ : 0) | - (acc & IB_ACCESS_LOCAL_WRITE ? T3_MEM_ACCESS_LOCAL_WRITE : 0) | - T3_MEM_ACCESS_LOCAL_READ; -} - enum iwch_mmid_state { IWCH_STAG_STATE_VALID, IWCH_STAG_STATE_INVALID diff --git a/drivers/infiniband/hw/cxgb3/iwch_qp.c b/drivers/infiniband/hw/cxgb3/iwch_qp.c index 992613799228..9a3be3a9d5dc 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_qp.c +++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c @@ -33,10 +33,11 @@ #include "iwch.h" #include "iwch_cm.h" #include "cxio_hal.h" +#include "cxio_resource.h" #define NO_SUPPORT -1 -static int iwch_build_rdma_send(union t3_wr *wqe, struct ib_send_wr *wr, +static int build_rdma_send(union t3_wr *wqe, struct ib_send_wr *wr, u8 * flit_cnt) { int i; @@ -44,59 +45,44 @@ static int iwch_build_rdma_send(union t3_wr *wqe, struct ib_send_wr *wr, switch (wr->opcode) { case IB_WR_SEND: - case IB_WR_SEND_WITH_IMM: if (wr->send_flags & IB_SEND_SOLICITED) wqe->send.rdmaop = T3_SEND_WITH_SE; else wqe->send.rdmaop = T3_SEND; wqe->send.rem_stag = 0; break; -#if 0 /* Not currently supported */ - case TYPE_SEND_INVALIDATE: - case TYPE_SEND_INVALIDATE_IMMEDIATE: - wqe->send.rdmaop = T3_SEND_WITH_INV; - wqe->send.rem_stag = cpu_to_be32(wr->wr.rdma.rkey); - break; - case TYPE_SEND_SE_INVALIDATE: - wqe->send.rdmaop = T3_SEND_WITH_SE_INV; - wqe->send.rem_stag = cpu_to_be32(wr->wr.rdma.rkey); + case IB_WR_SEND_WITH_INV: + if (wr->send_flags & IB_SEND_SOLICITED) + wqe->send.rdmaop = T3_SEND_WITH_SE_INV; + else + wqe->send.rdmaop = T3_SEND_WITH_INV; + wqe->send.rem_stag = cpu_to_be32(wr->ex.invalidate_rkey); break; -#endif default: - break; + return -EINVAL; } if (wr->num_sge > T3_MAX_SGE) return -EINVAL; wqe->send.reserved[0] = 0; wqe->send.reserved[1] = 0; wqe->send.reserved[2] = 0; - if (wr->opcode == IB_WR_SEND_WITH_IMM) { - plen = 4; - wqe->send.sgl[0].stag = wr->ex.imm_data; - wqe->send.sgl[0].len = __constant_cpu_to_be32(0); - wqe->send.num_sgle = __constant_cpu_to_be32(0); - *flit_cnt = 5; - } else { - plen = 0; - for (i = 0; i < wr->num_sge; i++) { - if ((plen + wr->sg_list[i].length) < plen) { - return -EMSGSIZE; - } - plen += wr->sg_list[i].length; - wqe->send.sgl[i].stag = - cpu_to_be32(wr->sg_list[i].lkey); - wqe->send.sgl[i].len = - cpu_to_be32(wr->sg_list[i].length); - wqe->send.sgl[i].to = cpu_to_be64(wr->sg_list[i].addr); - } - wqe->send.num_sgle = cpu_to_be32(wr->num_sge); - *flit_cnt = 4 + ((wr->num_sge) << 1); + plen = 0; + for (i = 0; i < wr->num_sge; i++) { + if ((plen + wr->sg_list[i].length) < plen) + return -EMSGSIZE; + + plen += wr->sg_list[i].length; + wqe->send.sgl[i].stag = cpu_to_be32(wr->sg_list[i].lkey); + wqe->send.sgl[i].len = cpu_to_be32(wr->sg_list[i].length); + wqe->send.sgl[i].to = cpu_to_be64(wr->sg_list[i].addr); } + wqe->send.num_sgle = cpu_to_be32(wr->num_sge); + *flit_cnt = 4 + ((wr->num_sge) << 1); wqe->send.plen = cpu_to_be32(plen); return 0; } -static int iwch_build_rdma_write(union t3_wr *wqe, struct ib_send_wr *wr, +static int build_rdma_write(union t3_wr *wqe, struct ib_send_wr *wr, u8 *flit_cnt) { int i; @@ -137,15 +123,18 @@ static int iwch_build_rdma_write(union t3_wr *wqe, struct ib_send_wr *wr, return 0; } -static int iwch_build_rdma_read(union t3_wr *wqe, struct ib_send_wr *wr, +static int build_rdma_read(union t3_wr *wqe, struct ib_send_wr *wr, u8 *flit_cnt) { if (wr->num_sge > 1) return -EINVAL; wqe->read.rdmaop = T3_READ_REQ; + if (wr->opcode == IB_WR_RDMA_READ_WITH_INV) + wqe->read.local_inv = 1; + else + wqe->read.local_inv = 0; wqe->read.reserved[0] = 0; wqe->read.reserved[1] = 0; - wqe->read.reserved[2] = 0; wqe->read.rem_stag = cpu_to_be32(wr->wr.rdma.rkey); wqe->read.rem_to = cpu_to_be64(wr->wr.rdma.remote_addr); wqe->read.local_stag = cpu_to_be32(wr->sg_list[0].lkey); @@ -155,6 +144,57 @@ static int iwch_build_rdma_read(union t3_wr *wqe, struct ib_send_wr *wr, return 0; } +static int build_fastreg(union t3_wr *wqe, struct ib_send_wr *wr, + u8 *flit_cnt, int *wr_cnt, struct t3_wq *wq) +{ + int i; + __be64 *p; + + if (wr->wr.fast_reg.page_list_len > T3_MAX_FASTREG_DEPTH) + return -EINVAL; + *wr_cnt = 1; + wqe->fastreg.stag = cpu_to_be32(wr->wr.fast_reg.rkey); + wqe->fastreg.len = cpu_to_be32(wr->wr.fast_reg.length); + wqe->fastreg.va_base_hi = cpu_to_be32(wr->wr.fast_reg.iova_start >> 32); + wqe->fastreg.va_base_lo_fbo = + cpu_to_be32(wr->wr.fast_reg.iova_start & 0xffffffff); + wqe->fastreg.page_type_perms = cpu_to_be32( + V_FR_PAGE_COUNT(wr->wr.fast_reg.page_list_len) | + V_FR_PAGE_SIZE(wr->wr.fast_reg.page_shift-12) | + V_FR_TYPE(TPT_VATO) | + V_FR_PERMS(iwch_ib_to_tpt_access(wr->wr.fast_reg.access_flags))); + p = &wqe->fastreg.pbl_addrs[0]; + for (i = 0; i < wr->wr.fast_reg.page_list_len; i++, p++) { + + /* If we need a 2nd WR, then set it up */ + if (i == T3_MAX_FASTREG_FRAG) { + *wr_cnt = 2; + wqe = (union t3_wr *)(wq->queue + + Q_PTR2IDX((wq->wptr+1), wq->size_log2)); + build_fw_riwrh((void *)wqe, T3_WR_FASTREG, 0, + Q_GENBIT(wq->wptr + 1, wq->size_log2), + 0, 1 + wr->wr.fast_reg.page_list_len - T3_MAX_FASTREG_FRAG, + T3_EOP); + + p = &wqe->pbl_frag.pbl_addrs[0]; + } + *p = cpu_to_be64((u64)wr->wr.fast_reg.page_list->page_list[i]); + } + *flit_cnt = 5 + wr->wr.fast_reg.page_list_len; + if (*flit_cnt > 15) + *flit_cnt = 15; + return 0; +} + +static int build_inv_stag(union t3_wr *wqe, struct ib_send_wr *wr, + u8 *flit_cnt) +{ + wqe->local_inv.stag = cpu_to_be32(wr->ex.invalidate_rkey); + wqe->local_inv.reserved = 0; + *flit_cnt = sizeof(struct t3_local_inv_wr) >> 3; + return 0; +} + /* * TBD: this is going to be moved to firmware. Missing pdid/qpid check for now. */ @@ -205,23 +245,106 @@ static int iwch_sgl2pbl_map(struct iwch_dev *rhp, struct ib_sge *sg_list, return 0; } -static int iwch_build_rdma_recv(struct iwch_dev *rhp, union t3_wr *wqe, +static int build_rdma_recv(struct iwch_qp *qhp, union t3_wr *wqe, struct ib_recv_wr *wr) { - int i; - if (wr->num_sge > T3_MAX_SGE) - return -EINVAL; + int i, err = 0; + u32 pbl_addr[T3_MAX_SGE]; + u8 page_size[T3_MAX_SGE]; + + err = iwch_sgl2pbl_map(qhp->rhp, wr->sg_list, wr->num_sge, pbl_addr, + page_size); + if (err) + return err; + wqe->recv.pagesz[0] = page_size[0]; + wqe->recv.pagesz[1] = page_size[1]; + wqe->recv.pagesz[2] = page_size[2]; + wqe->recv.pagesz[3] = page_size[3]; wqe->recv.num_sgle = cpu_to_be32(wr->num_sge); for (i = 0; i < wr->num_sge; i++) { wqe->recv.sgl[i].stag = cpu_to_be32(wr->sg_list[i].lkey); wqe->recv.sgl[i].len = cpu_to_be32(wr->sg_list[i].length); + + /* to in the WQE == the offset into the page */ + wqe->recv.sgl[i].to = cpu_to_be64(((u32) wr->sg_list[i].addr) % + (1UL << (12 + page_size[i]))); + + /* pbl_addr is the adapters address in the PBL */ + wqe->recv.pbl_addr[i] = cpu_to_be32(pbl_addr[i]); + } + for (; i < T3_MAX_SGE; i++) { + wqe->recv.sgl[i].stag = 0; + wqe->recv.sgl[i].len = 0; + wqe->recv.sgl[i].to = 0; + wqe->recv.pbl_addr[i] = 0; + } + qhp->wq.rq[Q_PTR2IDX(qhp->wq.rq_wptr, + qhp->wq.rq_size_log2)].wr_id = wr->wr_id; + qhp->wq.rq[Q_PTR2IDX(qhp->wq.rq_wptr, + qhp->wq.rq_size_log2)].pbl_addr = 0; + return 0; +} + +static int build_zero_stag_recv(struct iwch_qp *qhp, union t3_wr *wqe, + struct ib_recv_wr *wr) +{ + int i; + u32 pbl_addr; + u32 pbl_offset; + + + /* + * The T3 HW requires the PBL in the HW recv descriptor to reference + * a PBL entry. So we allocate the max needed PBL memory here and pass + * it to the uP in the recv WR. The uP will build the PBL and setup + * the HW recv descriptor. + */ + pbl_addr = cxio_hal_pblpool_alloc(&qhp->rhp->rdev, T3_STAG0_PBL_SIZE); + if (!pbl_addr) + return -ENOMEM; + + /* + * Compute the 8B aligned offset. + */ + pbl_offset = (pbl_addr - qhp->rhp->rdev.rnic_info.pbl_base) >> 3; + + wqe->recv.num_sgle = cpu_to_be32(wr->num_sge); + + for (i = 0; i < wr->num_sge; i++) { + + /* + * Use a 128MB page size. This and an imposed 128MB + * sge length limit allows us to require only a 2-entry HW + * PBL for each SGE. This restriction is acceptable since + * since it is not possible to allocate 128MB of contiguous + * DMA coherent memory! + */ + if (wr->sg_list[i].length > T3_STAG0_MAX_PBE_LEN) + return -EINVAL; + wqe->recv.pagesz[i] = T3_STAG0_PAGE_SHIFT; + + /* + * T3 restricts a recv to all zero-stag or all non-zero-stag. + */ + if (wr->sg_list[i].lkey != 0) + return -EINVAL; + wqe->recv.sgl[i].stag = 0; + wqe->recv.sgl[i].len = cpu_to_be32(wr->sg_list[i].length); wqe->recv.sgl[i].to = cpu_to_be64(wr->sg_list[i].addr); + wqe->recv.pbl_addr[i] = cpu_to_be32(pbl_offset); + pbl_offset += 2; } for (; i < T3_MAX_SGE; i++) { + wqe->recv.pagesz[i] = 0; wqe->recv.sgl[i].stag = 0; wqe->recv.sgl[i].len = 0; wqe->recv.sgl[i].to = 0; + wqe->recv.pbl_addr[i] = 0; } + qhp->wq.rq[Q_PTR2IDX(qhp->wq.rq_wptr, + qhp->wq.rq_size_log2)].wr_id = wr->wr_id; + qhp->wq.rq[Q_PTR2IDX(qhp->wq.rq_wptr, + qhp->wq.rq_size_log2)].pbl_addr = pbl_addr; return 0; } @@ -238,6 +361,7 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, u32 num_wrs; unsigned long flag; struct t3_swsq *sqp; + int wr_cnt = 1; qhp = to_iwch_qp(ibqp); spin_lock_irqsave(&qhp->lock, flag); @@ -262,33 +386,45 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, t3_wr_flags = 0; if (wr->send_flags & IB_SEND_SOLICITED) t3_wr_flags |= T3_SOLICITED_EVENT_FLAG; - if (wr->send_flags & IB_SEND_FENCE) - t3_wr_flags |= T3_READ_FENCE_FLAG; if (wr->send_flags & IB_SEND_SIGNALED) t3_wr_flags |= T3_COMPLETION_FLAG; sqp = qhp->wq.sq + Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2); switch (wr->opcode) { case IB_WR_SEND: - case IB_WR_SEND_WITH_IMM: + case IB_WR_SEND_WITH_INV: + if (wr->send_flags & IB_SEND_FENCE) + t3_wr_flags |= T3_READ_FENCE_FLAG; t3_wr_opcode = T3_WR_SEND; - err = iwch_build_rdma_send(wqe, wr, &t3_wr_flit_cnt); + err = build_rdma_send(wqe, wr, &t3_wr_flit_cnt); break; case IB_WR_RDMA_WRITE: case IB_WR_RDMA_WRITE_WITH_IMM: t3_wr_opcode = T3_WR_WRITE; - err = iwch_build_rdma_write(wqe, wr, &t3_wr_flit_cnt); + err = build_rdma_write(wqe, wr, &t3_wr_flit_cnt); break; case IB_WR_RDMA_READ: + case IB_WR_RDMA_READ_WITH_INV: t3_wr_opcode = T3_WR_READ; t3_wr_flags = 0; /* T3 reads are always signaled */ - err = iwch_build_rdma_read(wqe, wr, &t3_wr_flit_cnt); + err = build_rdma_read(wqe, wr, &t3_wr_flit_cnt); if (err) break; sqp->read_len = wqe->read.local_len; if (!qhp->wq.oldest_read) qhp->wq.oldest_read = sqp; break; + case IB_WR_FAST_REG_MR: + t3_wr_opcode = T3_WR_FASTREG; + err = build_fastreg(wqe, wr, &t3_wr_flit_cnt, + &wr_cnt, &qhp->wq); + break; + case IB_WR_LOCAL_INV: + if (wr->send_flags & IB_SEND_FENCE) + t3_wr_flags |= T3_LOCAL_FENCE_FLAG; + t3_wr_opcode = T3_WR_INV_STAG; + err = build_inv_stag(wqe, wr, &t3_wr_flit_cnt); + break; default: PDBG("%s post of type=%d TBD!\n", __func__, wr->opcode); @@ -307,14 +443,15 @@ int iwch_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, build_fw_riwrh((void *) wqe, t3_wr_opcode, t3_wr_flags, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), - 0, t3_wr_flit_cnt); + 0, t3_wr_flit_cnt, + (wr_cnt == 1) ? T3_SOPEOP : T3_SOP); PDBG("%s cookie 0x%llx wq idx 0x%x swsq idx %ld opcode %d\n", __func__, (unsigned long long) wr->wr_id, idx, Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2), sqp->opcode); wr = wr->next; num_wrs--; - ++(qhp->wq.wptr); + qhp->wq.wptr += wr_cnt; ++(qhp->wq.sq_wptr); } spin_unlock_irqrestore(&qhp->lock, flag); @@ -345,21 +482,27 @@ int iwch_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr, return -EINVAL; } while (wr) { + if (wr->num_sge > T3_MAX_SGE) { + err = -EINVAL; + *bad_wr = wr; + break; + } idx = Q_PTR2IDX(qhp->wq.wptr, qhp->wq.size_log2); wqe = (union t3_wr *) (qhp->wq.queue + idx); if (num_wrs) - err = iwch_build_rdma_recv(qhp->rhp, wqe, wr); + if (wr->sg_list[0].lkey) + err = build_rdma_recv(qhp, wqe, wr); + else + err = build_zero_stag_recv(qhp, wqe, wr); else err = -ENOMEM; if (err) { *bad_wr = wr; break; } - qhp->wq.rq[Q_PTR2IDX(qhp->wq.rq_wptr, qhp->wq.rq_size_log2)] = - wr->wr_id; build_fw_riwrh((void *) wqe, T3_WR_RCV, T3_COMPLETION_FLAG, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), - 0, sizeof(struct t3_receive_wr) >> 3); + 0, sizeof(struct t3_receive_wr) >> 3, T3_SOPEOP); PDBG("%s cookie 0x%llx idx 0x%x rq_wptr 0x%x rw_rptr 0x%x " "wqe %p \n", __func__, (unsigned long long) wr->wr_id, idx, qhp->wq.rq_wptr, qhp->wq.rq_rptr, wqe); @@ -419,10 +562,10 @@ int iwch_bind_mw(struct ib_qp *qp, sgl.lkey = mw_bind->mr->lkey; sgl.length = mw_bind->length; wqe->bind.reserved = 0; - wqe->bind.type = T3_VA_BASED_TO; + wqe->bind.type = TPT_VATO; /* TBD: check perms */ - wqe->bind.perms = iwch_ib_to_mwbind_access(mw_bind->mw_access_flags); + wqe->bind.perms = iwch_ib_to_tpt_access(mw_bind->mw_access_flags); wqe->bind.mr_stag = cpu_to_be32(mw_bind->mr->lkey); wqe->bind.mw_stag = cpu_to_be32(mw->rkey); wqe->bind.mw_len = cpu_to_be32(mw_bind->length); @@ -430,7 +573,7 @@ int iwch_bind_mw(struct ib_qp *qp, err = iwch_sgl2pbl_map(rhp, &sgl, 1, &pbl_addr, &page_size); if (err) { spin_unlock_irqrestore(&qhp->lock, flag); - return err; + return err; } wqe->send.wrid.id0.hi = qhp->wq.sq_wptr; sqp = qhp->wq.sq + Q_PTR2IDX(qhp->wq.sq_wptr, qhp->wq.sq_size_log2); @@ -441,10 +584,9 @@ int iwch_bind_mw(struct ib_qp *qp, sqp->signaled = (mw_bind->send_flags & IB_SEND_SIGNALED); wqe->bind.mr_pbl_addr = cpu_to_be32(pbl_addr); wqe->bind.mr_pagesz = page_size; - wqe->flit[T3_SQ_COOKIE_FLIT] = mw_bind->wr_id; build_fw_riwrh((void *)wqe, T3_WR_BIND, t3_wr_flags, Q_GENBIT(qhp->wq.wptr, qhp->wq.size_log2), 0, - sizeof(struct t3_bind_mw_wr) >> 3); + sizeof(struct t3_bind_mw_wr) >> 3, T3_SOPEOP); ++(qhp->wq.wptr); ++(qhp->wq.sq_wptr); spin_unlock_irqrestore(&qhp->lock, flag); @@ -758,7 +900,8 @@ static int rdma_init(struct iwch_dev *rhp, struct iwch_qp *qhp, init_attr.qp_dma_size = (1UL << qhp->wq.size_log2); init_attr.rqe_count = iwch_rqes_posted(qhp); init_attr.flags = qhp->attr.mpa_attr.initiator ? MPA_INITIATOR : 0; - init_attr.flags |= capable(CAP_NET_BIND_SERVICE) ? PRIV_QP : 0; + if (!qhp->ibqp.uobject) + init_attr.flags |= PRIV_QP; if (peer2peer) { init_attr.rtr_type = RTR_READ; if (init_attr.ord == 0 && qhp->attr.mpa_attr.initiator) diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c index ce1ab0571be3..0792d930c481 100644 --- a/drivers/infiniband/hw/ehca/ehca_irq.c +++ b/drivers/infiniband/hw/ehca/ehca_irq.c @@ -531,7 +531,7 @@ void ehca_process_eq(struct ehca_shca *shca, int is_irq) { struct ehca_eq *eq = &shca->eq; struct ehca_eqe_cache_entry *eqe_cache = eq->eqe_cache; - u64 eqe_value; + u64 eqe_value, ret; unsigned long flags; int eqe_cnt, i; int eq_empty = 0; @@ -583,8 +583,13 @@ void ehca_process_eq(struct ehca_shca *shca, int is_irq) ehca_dbg(&shca->ib_device, "No eqe found for irq event"); goto unlock_irq_spinlock; - } else if (!is_irq) + } else if (!is_irq) { + ret = hipz_h_eoi(eq->ist); + if (ret != H_SUCCESS) + ehca_err(&shca->ib_device, + "bad return code EOI -rc = %ld\n", ret); ehca_dbg(&shca->ib_device, "deadman found %x eqe", eqe_cnt); + } if (unlikely(eqe_cnt == EHCA_EQE_CACHE_SIZE)) ehca_dbg(&shca->ib_device, "too many eqes for one irq event"); /* enable irq for new packets */ diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c index 482103eb6eac..598844d2edc9 100644 --- a/drivers/infiniband/hw/ehca/ehca_main.c +++ b/drivers/infiniband/hw/ehca/ehca_main.c @@ -923,6 +923,7 @@ static struct of_device_id ehca_device_table[] = }, {}, }; +MODULE_DEVICE_TABLE(of, ehca_device_table); static struct of_platform_driver ehca_driver = { .name = "ehca", diff --git a/drivers/infiniband/hw/ehca/ehca_reqs.c b/drivers/infiniband/hw/ehca/ehca_reqs.c index f093b0033daf..dd9bc68f1c7b 100644 --- a/drivers/infiniband/hw/ehca/ehca_reqs.c +++ b/drivers/infiniband/hw/ehca/ehca_reqs.c @@ -544,8 +544,16 @@ int ehca_post_recv(struct ib_qp *qp, struct ib_recv_wr *recv_wr, struct ib_recv_wr **bad_recv_wr) { - return internal_post_recv(container_of(qp, struct ehca_qp, ib_qp), - qp->device, recv_wr, bad_recv_wr); + struct ehca_qp *my_qp = container_of(qp, struct ehca_qp, ib_qp); + + /* Reject WR if QP is in RESET state */ + if (unlikely(my_qp->state == IB_QPS_RESET)) { + ehca_err(qp->device, "Invalid QP state qp_state=%d qpn=%x", + my_qp->state, qp->qp_num); + return -EINVAL; + } + + return internal_post_recv(my_qp, qp->device, recv_wr, bad_recv_wr); } int ehca_post_srq_recv(struct ib_srq *srq, @@ -681,7 +689,7 @@ poll_cq_one_read_cqe: wc->dlid_path_bits = cqe->dlid; wc->src_qp = cqe->remote_qp_number; wc->wc_flags = cqe->w_completion_flags; - wc->imm_data = cpu_to_be32(cqe->immediate_data); + wc->ex.imm_data = cpu_to_be32(cqe->immediate_data); wc->sl = cqe->service_level; poll_cq_one_exit0: diff --git a/drivers/infiniband/hw/ehca/hcp_if.c b/drivers/infiniband/hw/ehca/hcp_if.c index 5245e13c3a30..415d3a465de6 100644 --- a/drivers/infiniband/hw/ehca/hcp_if.c +++ b/drivers/infiniband/hw/ehca/hcp_if.c @@ -933,3 +933,13 @@ u64 hipz_h_error_data(const struct ipz_adapter_handle adapter_handle, r_cb, 0, 0, 0, 0); } + +u64 hipz_h_eoi(int irq) +{ + unsigned long xirr; + + iosync(); + xirr = (0xffULL << 24) | irq; + + return plpar_hcall_norets(H_EOI, xirr); +} diff --git a/drivers/infiniband/hw/ehca/hcp_if.h b/drivers/infiniband/hw/ehca/hcp_if.h index 60ce02b70663..2c3c6e0ea5c2 100644 --- a/drivers/infiniband/hw/ehca/hcp_if.h +++ b/drivers/infiniband/hw/ehca/hcp_if.h @@ -260,5 +260,6 @@ u64 hipz_h_error_data(const struct ipz_adapter_handle adapter_handle, const u64 ressource_handle, void *rblock, unsigned long *byte_count); +u64 hipz_h_eoi(int irq); #endif /* __HCP_IF_H__ */ diff --git a/drivers/infiniband/hw/ipath/ipath_cq.c b/drivers/infiniband/hw/ipath/ipath_cq.c index a03bd28d9b48..d385e4168c97 100644 --- a/drivers/infiniband/hw/ipath/ipath_cq.c +++ b/drivers/infiniband/hw/ipath/ipath_cq.c @@ -82,7 +82,7 @@ void ipath_cq_enter(struct ipath_cq *cq, struct ib_wc *entry, int solicited) wc->uqueue[head].opcode = entry->opcode; wc->uqueue[head].vendor_err = entry->vendor_err; wc->uqueue[head].byte_len = entry->byte_len; - wc->uqueue[head].imm_data = (__u32 __force)entry->imm_data; + wc->uqueue[head].ex.imm_data = (__u32 __force) entry->ex.imm_data; wc->uqueue[head].qp_num = entry->qp->qp_num; wc->uqueue[head].src_qp = entry->src_qp; wc->uqueue[head].wc_flags = entry->wc_flags; diff --git a/drivers/infiniband/hw/ipath/ipath_iba7220.c b/drivers/infiniband/hw/ipath/ipath_iba7220.c index 8eee7830f042..fb70712ac85c 100644 --- a/drivers/infiniband/hw/ipath/ipath_iba7220.c +++ b/drivers/infiniband/hw/ipath/ipath_iba7220.c @@ -2228,8 +2228,8 @@ static void ipath_autoneg_send(struct ipath_devdata *dd, int which) 0xffffffff, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x40000001, 0x1388, 0x15e, /* rest 0's */ }; - dcnt = sizeof(madpayload_start)/sizeof(madpayload_start[0]); - hcnt = sizeof(hdr)/sizeof(hdr[0]); + dcnt = ARRAY_SIZE(madpayload_start); + hcnt = ARRAY_SIZE(hdr); if (!swapped) { /* for maintainability, do it at runtime */ for (i = 0; i < hcnt; i++) { diff --git a/drivers/infiniband/hw/ipath/ipath_mad.c b/drivers/infiniband/hw/ipath/ipath_mad.c index 5f9315d77a43..be4fc9ada8e7 100644 --- a/drivers/infiniband/hw/ipath/ipath_mad.c +++ b/drivers/infiniband/hw/ipath/ipath_mad.c @@ -111,9 +111,9 @@ static int recv_subn_get_nodeinfo(struct ib_smp *smp, nip->revision = cpu_to_be32((majrev << 16) | minrev); nip->local_port_num = port; vendor = dd->ipath_vendorid; - nip->vendor_id[0] = 0; - nip->vendor_id[1] = vendor >> 8; - nip->vendor_id[2] = vendor; + nip->vendor_id[0] = IPATH_SRC_OUI_1; + nip->vendor_id[1] = IPATH_SRC_OUI_2; + nip->vendor_id[2] = IPATH_SRC_OUI_3; return reply(smp); } diff --git a/drivers/infiniband/hw/ipath/ipath_rc.c b/drivers/infiniband/hw/ipath/ipath_rc.c index 108df667d2ee..97710522624d 100644 --- a/drivers/infiniband/hw/ipath/ipath_rc.c +++ b/drivers/infiniband/hw/ipath/ipath_rc.c @@ -1703,11 +1703,11 @@ void ipath_rc_rcv(struct ipath_ibdev *dev, struct ipath_ib_header *hdr, case OP(SEND_LAST_WITH_IMMEDIATE): send_last_imm: if (header_in_data) { - wc.imm_data = *(__be32 *) data; + wc.ex.imm_data = *(__be32 *) data; data += sizeof(__be32); } else { /* Immediate data comes after BTH */ - wc.imm_data = ohdr->u.imm_data; + wc.ex.imm_data = ohdr->u.imm_data; } hdrsize += 4; wc.wc_flags = IB_WC_WITH_IMM; diff --git a/drivers/infiniband/hw/ipath/ipath_ruc.c b/drivers/infiniband/hw/ipath/ipath_ruc.c index a4b5521567fe..af051f757663 100644 --- a/drivers/infiniband/hw/ipath/ipath_ruc.c +++ b/drivers/infiniband/hw/ipath/ipath_ruc.c @@ -331,7 +331,7 @@ again: switch (wqe->wr.opcode) { case IB_WR_SEND_WITH_IMM: wc.wc_flags = IB_WC_WITH_IMM; - wc.imm_data = wqe->wr.ex.imm_data; + wc.ex.imm_data = wqe->wr.ex.imm_data; /* FALLTHROUGH */ case IB_WR_SEND: if (!ipath_get_rwqe(qp, 0)) @@ -342,7 +342,7 @@ again: if (unlikely(!(qp->qp_access_flags & IB_ACCESS_REMOTE_WRITE))) goto inv_err; wc.wc_flags = IB_WC_WITH_IMM; - wc.imm_data = wqe->wr.ex.imm_data; + wc.ex.imm_data = wqe->wr.ex.imm_data; if (!ipath_get_rwqe(qp, 1)) goto rnr_nak; /* FALLTHROUGH */ diff --git a/drivers/infiniband/hw/ipath/ipath_uc.c b/drivers/infiniband/hw/ipath/ipath_uc.c index 0596ec16fcbd..82cc588b8bf2 100644 --- a/drivers/infiniband/hw/ipath/ipath_uc.c +++ b/drivers/infiniband/hw/ipath/ipath_uc.c @@ -379,11 +379,11 @@ void ipath_uc_rcv(struct ipath_ibdev *dev, struct ipath_ib_header *hdr, case OP(SEND_LAST_WITH_IMMEDIATE): send_last_imm: if (header_in_data) { - wc.imm_data = *(__be32 *) data; + wc.ex.imm_data = *(__be32 *) data; data += sizeof(__be32); } else { /* Immediate data comes after BTH */ - wc.imm_data = ohdr->u.imm_data; + wc.ex.imm_data = ohdr->u.imm_data; } hdrsize += 4; wc.wc_flags = IB_WC_WITH_IMM; @@ -483,11 +483,11 @@ void ipath_uc_rcv(struct ipath_ibdev *dev, struct ipath_ib_header *hdr, case OP(RDMA_WRITE_LAST_WITH_IMMEDIATE): rdma_last_imm: if (header_in_data) { - wc.imm_data = *(__be32 *) data; + wc.ex.imm_data = *(__be32 *) data; data += sizeof(__be32); } else { /* Immediate data comes after BTH */ - wc.imm_data = ohdr->u.imm_data; + wc.ex.imm_data = ohdr->u.imm_data; } hdrsize += 4; wc.wc_flags = IB_WC_WITH_IMM; diff --git a/drivers/infiniband/hw/ipath/ipath_ud.c b/drivers/infiniband/hw/ipath/ipath_ud.c index 77ca8ca74e78..36aa242c487c 100644 --- a/drivers/infiniband/hw/ipath/ipath_ud.c +++ b/drivers/infiniband/hw/ipath/ipath_ud.c @@ -96,7 +96,7 @@ static void ipath_ud_loopback(struct ipath_qp *sqp, struct ipath_swqe *swqe) if (swqe->wr.opcode == IB_WR_SEND_WITH_IMM) { wc.wc_flags = IB_WC_WITH_IMM; - wc.imm_data = swqe->wr.ex.imm_data; + wc.ex.imm_data = swqe->wr.ex.imm_data; } /* @@ -492,14 +492,14 @@ void ipath_ud_rcv(struct ipath_ibdev *dev, struct ipath_ib_header *hdr, if (qp->ibqp.qp_num > 1 && opcode == IB_OPCODE_UD_SEND_ONLY_WITH_IMMEDIATE) { if (header_in_data) { - wc.imm_data = *(__be32 *) data; + wc.ex.imm_data = *(__be32 *) data; data += sizeof(__be32); } else - wc.imm_data = ohdr->u.ud.imm_data; + wc.ex.imm_data = ohdr->u.ud.imm_data; wc.wc_flags = IB_WC_WITH_IMM; hdrsize += sizeof(u32); } else if (opcode == IB_OPCODE_UD_SEND_ONLY) { - wc.imm_data = 0; + wc.ex.imm_data = 0; wc.wc_flags = 0; } else { dev->n_pkt_drops++; diff --git a/drivers/infiniband/hw/ipath/ipath_verbs.c b/drivers/infiniband/hw/ipath/ipath_verbs.c index 7779165b2c2c..9e23ab0b51a1 100644 --- a/drivers/infiniband/hw/ipath/ipath_verbs.c +++ b/drivers/infiniband/hw/ipath/ipath_verbs.c @@ -1497,7 +1497,8 @@ static int ipath_query_device(struct ib_device *ibdev, IB_DEVICE_SYS_IMAGE_GUID | IB_DEVICE_RC_RNR_NAK_GEN | IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SRQ_RESIZE; props->page_size_cap = PAGE_SIZE; - props->vendor_id = dev->dd->ipath_vendorid; + props->vendor_id = + IPATH_SRC_OUI_1 << 16 | IPATH_SRC_OUI_2 << 8 | IPATH_SRC_OUI_3; props->vendor_part_id = dev->dd->ipath_deviceid; props->hw_ver = dev->dd->ipath_pcirev; diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 4521319b1406..299f20832ab6 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -663,18 +663,18 @@ repoll: switch (cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) { case MLX4_RECV_OPCODE_RDMA_WRITE_IMM: - wc->opcode = IB_WC_RECV_RDMA_WITH_IMM; - wc->wc_flags = IB_WC_WITH_IMM; - wc->imm_data = cqe->immed_rss_invalid; + wc->opcode = IB_WC_RECV_RDMA_WITH_IMM; + wc->wc_flags = IB_WC_WITH_IMM; + wc->ex.imm_data = cqe->immed_rss_invalid; break; case MLX4_RECV_OPCODE_SEND: wc->opcode = IB_WC_RECV; wc->wc_flags = 0; break; case MLX4_RECV_OPCODE_SEND_IMM: - wc->opcode = IB_WC_RECV; - wc->wc_flags = IB_WC_WITH_IMM; - wc->imm_data = cqe->immed_rss_invalid; + wc->opcode = IB_WC_RECV; + wc->wc_flags = IB_WC_WITH_IMM; + wc->ex.imm_data = cqe->immed_rss_invalid; break; } diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c index 4c1e72fc8f57..cdca3a511e1c 100644 --- a/drivers/infiniband/hw/mlx4/mad.c +++ b/drivers/infiniband/hw/mlx4/mad.c @@ -255,7 +255,8 @@ int mlx4_ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num, return IB_MAD_RESULT_SUCCESS; } else if (in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT || in_mad->mad_hdr.mgmt_class == MLX4_IB_VENDOR_CLASS1 || - in_mad->mad_hdr.mgmt_class == MLX4_IB_VENDOR_CLASS2) { + in_mad->mad_hdr.mgmt_class == MLX4_IB_VENDOR_CLASS2 || + in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_CONG_MGMT) { if (in_mad->mad_hdr.method != IB_MGMT_METHOD_GET && in_mad->mad_hdr.method != IB_MGMT_METHOD_SET) return IB_MAD_RESULT_SUCCESS; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 4d61e32866c6..bcf50648fa18 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -90,7 +90,8 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, props->device_cap_flags = IB_DEVICE_CHANGE_PHY_PORT | IB_DEVICE_PORT_ACTIVE_EVENT | IB_DEVICE_SYS_IMAGE_GUID | - IB_DEVICE_RC_RNR_NAK_GEN; + IB_DEVICE_RC_RNR_NAK_GEN | + IB_DEVICE_BLOCK_MULTICAST_LOOPBACK; if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_BAD_PKEY_CNTR) props->device_cap_flags |= IB_DEVICE_BAD_PKEY_CNTR; if (dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_BAD_QKEY_CNTR) @@ -437,7 +438,9 @@ static int mlx4_ib_dealloc_pd(struct ib_pd *pd) static int mlx4_ib_mcg_attach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) { return mlx4_multicast_attach(to_mdev(ibqp->device)->dev, - &to_mqp(ibqp)->mqp, gid->raw); + &to_mqp(ibqp)->mqp, gid->raw, + !!(to_mqp(ibqp)->flags & + MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK)); } static int mlx4_ib_mcg_detach(struct ib_qp *ibqp, union ib_gid *gid, u16 lid) diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index 5cf994794d25..c4cf5b69eefa 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -101,7 +101,8 @@ struct mlx4_ib_wq { }; enum mlx4_ib_qp_flags { - MLX4_IB_QP_LSO = 1 << 0 + MLX4_IB_QP_LSO = 1 << 0, + MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK = 1 << 1, }; struct mlx4_ib_qp { diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index a80df22deae8..89eb6cbe592e 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -129,9 +129,10 @@ static void stamp_send_wqe(struct mlx4_ib_qp *qp, int n, int size) int ind; void *buf; __be32 stamp; + struct mlx4_wqe_ctrl_seg *ctrl; - s = roundup(size, 1U << qp->sq.wqe_shift); if (qp->sq_max_wqes_per_wr > 1) { + s = roundup(size, 1U << qp->sq.wqe_shift); for (i = 0; i < s; i += 64) { ind = (i >> qp->sq.wqe_shift) + n; stamp = ind & qp->sq.wqe_cnt ? cpu_to_be32(0x7fffffff) : @@ -141,7 +142,8 @@ static void stamp_send_wqe(struct mlx4_ib_qp *qp, int n, int size) *wqe = stamp; } } else { - buf = get_send_wqe(qp, n & (qp->sq.wqe_cnt - 1)); + ctrl = buf = get_send_wqe(qp, n & (qp->sq.wqe_cnt - 1)); + s = (ctrl->fence_size & 0x3f) << 4; for (i = 64; i < s; i += 64) { wqe = buf + i; *wqe = cpu_to_be32(0xffffffff); @@ -452,19 +454,8 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, spin_lock_init(&qp->rq.lock); qp->state = IB_QPS_RESET; - qp->atomic_rd_en = 0; - qp->resp_depth = 0; - - qp->rq.head = 0; - qp->rq.tail = 0; - qp->sq.head = 0; - qp->sq.tail = 0; - qp->sq_next_wqe = 0; - if (init_attr->sq_sig_type == IB_SIGNAL_ALL_WR) qp->sq_signal_bits = cpu_to_be32(MLX4_WQE_CTRL_CQ_UPDATE); - else - qp->sq_signal_bits = 0; err = set_rq_size(dev, &init_attr->cap, !!pd->uobject, !!init_attr->srq, qp); if (err) @@ -509,6 +500,9 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, } else { qp->sq_no_prefetch = 0; + if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) + qp->flags |= MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK; + if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO) qp->flags |= MLX4_IB_QP_LSO; @@ -682,10 +676,15 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd, struct mlx4_ib_qp *qp; int err; - /* We only support LSO, and only for kernel UD QPs. */ - if (init_attr->create_flags & ~IB_QP_CREATE_IPOIB_UD_LSO) + /* + * We only support LSO and multicast loopback blocking, and + * only for kernel UD QPs. + */ + if (init_attr->create_flags & ~(IB_QP_CREATE_IPOIB_UD_LSO | + IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK)) return ERR_PTR(-EINVAL); - if (init_attr->create_flags & IB_QP_CREATE_IPOIB_UD_LSO && + + if (init_attr->create_flags && (pd->uobject || init_attr->qp_type != IB_QPT_UD)) return ERR_PTR(-EINVAL); @@ -694,7 +693,7 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd, case IB_QPT_UC: case IB_QPT_UD: { - qp = kmalloc(sizeof *qp, GFP_KERNEL); + qp = kzalloc(sizeof *qp, GFP_KERNEL); if (!qp) return ERR_PTR(-ENOMEM); @@ -715,7 +714,7 @@ struct ib_qp *mlx4_ib_create_qp(struct ib_pd *pd, if (pd->uobject) return ERR_PTR(-EINVAL); - sqp = kmalloc(sizeof *sqp, GFP_KERNEL); + sqp = kzalloc(sizeof *sqp, GFP_KERNEL); if (!sqp) return ERR_PTR(-ENOMEM); @@ -906,7 +905,8 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, attr->path_mtu); goto out; } - context->mtu_msgmax = (attr->path_mtu << 5) | 31; + context->mtu_msgmax = (attr->path_mtu << 5) | + ilog2(dev->dev->caps.max_msg_sz); } if (qp->rq.wqe_cnt) @@ -1063,6 +1063,8 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, for (i = 0; i < qp->sq.wqe_cnt; ++i) { ctrl = get_send_wqe(qp, i); ctrl->owner_opcode = cpu_to_be32(1 << 31); + if (qp->sq_max_wqes_per_wr == 1) + ctrl->fence_size = 1 << (qp->sq.wqe_shift - 4); stamp_send_wqe(qp, i, 1 << qp->sq.wqe_shift); } @@ -1127,23 +1129,6 @@ out: return err; } -static const struct ib_qp_attr mlx4_ib_qp_attr = { .port_num = 1 }; -static const int mlx4_ib_qp_attr_mask_table[IB_QPT_UD + 1] = { - [IB_QPT_UD] = (IB_QP_PKEY_INDEX | - IB_QP_PORT | - IB_QP_QKEY), - [IB_QPT_UC] = (IB_QP_PKEY_INDEX | - IB_QP_PORT | - IB_QP_ACCESS_FLAGS), - [IB_QPT_RC] = (IB_QP_PKEY_INDEX | - IB_QP_PORT | - IB_QP_ACCESS_FLAGS), - [IB_QPT_SMI] = (IB_QP_PKEY_INDEX | - IB_QP_QKEY), - [IB_QPT_GSI] = (IB_QP_PKEY_INDEX | - IB_QP_QKEY), -}; - int mlx4_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, struct ib_udata *udata) { @@ -1186,15 +1171,6 @@ int mlx4_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, goto out; } - if (cur_state == IB_QPS_RESET && new_state == IB_QPS_ERR) { - err = __mlx4_ib_modify_qp(ibqp, &mlx4_ib_qp_attr, - mlx4_ib_qp_attr_mask_table[ibqp->qp_type], - IB_QPS_RESET, IB_QPS_INIT); - if (err) - goto out; - cur_state = IB_QPS_INIT; - } - err = __mlx4_ib_modify_qp(ibqp, attr, attr_mask, cur_state, new_state); out: @@ -1865,6 +1841,13 @@ done: qp_init_attr->cap = qp_attr->cap; + qp_init_attr->create_flags = 0; + if (qp->flags & MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK) + qp_init_attr->create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK; + + if (qp->flags & MLX4_IB_QP_LSO) + qp_init_attr->create_flags |= IB_QP_CREATE_IPOIB_UD_LSO; + out: mutex_unlock(&qp->mutex); return err; diff --git a/drivers/infiniband/hw/mthca/mthca_allocator.c b/drivers/infiniband/hw/mthca/mthca_allocator.c index a76306709618..c5ccc2daab60 100644 --- a/drivers/infiniband/hw/mthca/mthca_allocator.c +++ b/drivers/infiniband/hw/mthca/mthca_allocator.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_allocator.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/errno.h> diff --git a/drivers/infiniband/hw/mthca/mthca_av.c b/drivers/infiniband/hw/mthca/mthca_av.c index 4b111a852ff6..32f6c6315454 100644 --- a/drivers/infiniband/hw/mthca/mthca_av.c +++ b/drivers/infiniband/hw/mthca/mthca_av.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_av.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/string.h> diff --git a/drivers/infiniband/hw/mthca/mthca_catas.c b/drivers/infiniband/hw/mthca/mthca_catas.c index e948158a28d9..cc440f90000b 100644 --- a/drivers/infiniband/hw/mthca/mthca_catas.c +++ b/drivers/infiniband/hw/mthca/mthca_catas.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id$ */ #include <linux/jiffies.h> @@ -128,7 +126,6 @@ static void handle_catas(struct mthca_dev *dev) static void poll_catas(unsigned long dev_ptr) { struct mthca_dev *dev = (struct mthca_dev *) dev_ptr; - unsigned long flags; int i; for (i = 0; i < dev->catas_err.size; ++i) @@ -137,13 +134,8 @@ static void poll_catas(unsigned long dev_ptr) return; } - spin_lock_irqsave(&catas_lock, flags); - if (!dev->catas_err.stop) - mod_timer(&dev->catas_err.timer, - jiffies + MTHCA_CATAS_POLL_INTERVAL); - spin_unlock_irqrestore(&catas_lock, flags); - - return; + mod_timer(&dev->catas_err.timer, + round_jiffies(jiffies + MTHCA_CATAS_POLL_INTERVAL)); } void mthca_start_catas_poll(struct mthca_dev *dev) @@ -151,7 +143,6 @@ void mthca_start_catas_poll(struct mthca_dev *dev) unsigned long addr; init_timer(&dev->catas_err.timer); - dev->catas_err.stop = 0; dev->catas_err.map = NULL; addr = pci_resource_start(dev->pdev, 0) + @@ -182,10 +173,6 @@ void mthca_start_catas_poll(struct mthca_dev *dev) void mthca_stop_catas_poll(struct mthca_dev *dev) { - spin_lock_irq(&catas_lock); - dev->catas_err.stop = 1; - spin_unlock_irq(&catas_lock); - del_timer_sync(&dev->catas_err.timer); if (dev->catas_err.map) { diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c index 54d230ee7d63..c33e1c53c799 100644 --- a/drivers/infiniband/hw/mthca/mthca_cmd.c +++ b/drivers/infiniband/hw/mthca/mthca_cmd.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_cmd.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/completion.h> diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.h b/drivers/infiniband/hw/mthca/mthca_cmd.h index 8928ca4a9325..6efd3265f248 100644 --- a/drivers/infiniband/hw/mthca/mthca_cmd.h +++ b/drivers/infiniband/hw/mthca/mthca_cmd.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_cmd.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef MTHCA_CMD_H diff --git a/drivers/infiniband/hw/mthca/mthca_config_reg.h b/drivers/infiniband/hw/mthca/mthca_config_reg.h index afa56bfaab2e..75671f75cac4 100644 --- a/drivers/infiniband/hw/mthca/mthca_config_reg.h +++ b/drivers/infiniband/hw/mthca/mthca_config_reg.h @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_config_reg.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef MTHCA_CONFIG_REG_H diff --git a/drivers/infiniband/hw/mthca/mthca_cq.c b/drivers/infiniband/hw/mthca/mthca_cq.c index 20401d2ba6b2..d9f4735c2b37 100644 --- a/drivers/infiniband/hw/mthca/mthca_cq.c +++ b/drivers/infiniband/hw/mthca/mthca_cq.c @@ -32,8 +32,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_cq.c 1369 2004-12-20 16:17:07Z roland $ */ #include <linux/hardirq.h> @@ -622,13 +620,13 @@ static inline int mthca_poll_one(struct mthca_dev *dev, case IB_OPCODE_SEND_LAST_WITH_IMMEDIATE: case IB_OPCODE_SEND_ONLY_WITH_IMMEDIATE: entry->wc_flags = IB_WC_WITH_IMM; - entry->imm_data = cqe->imm_etype_pkey_eec; + entry->ex.imm_data = cqe->imm_etype_pkey_eec; entry->opcode = IB_WC_RECV; break; case IB_OPCODE_RDMA_WRITE_LAST_WITH_IMMEDIATE: case IB_OPCODE_RDMA_WRITE_ONLY_WITH_IMMEDIATE: entry->wc_flags = IB_WC_WITH_IMM; - entry->imm_data = cqe->imm_etype_pkey_eec; + entry->ex.imm_data = cqe->imm_etype_pkey_eec; entry->opcode = IB_WC_RECV_RDMA_WITH_IMM; break; default: diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h index 7bc32f8e377e..ee4d073c889f 100644 --- a/drivers/infiniband/hw/mthca/mthca_dev.h +++ b/drivers/infiniband/hw/mthca/mthca_dev.h @@ -32,8 +32,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_dev.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef MTHCA_DEV_H @@ -279,7 +277,6 @@ struct mthca_mcg_table { struct mthca_catas_err { u64 addr; u32 __iomem *map; - unsigned long stop; u32 size; struct timer_list timer; struct list_head list; diff --git a/drivers/infiniband/hw/mthca/mthca_doorbell.h b/drivers/infiniband/hw/mthca/mthca_doorbell.h index b374dc395be1..14f51ef97d7e 100644 --- a/drivers/infiniband/hw/mthca/mthca_doorbell.h +++ b/drivers/infiniband/hw/mthca/mthca_doorbell.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_doorbell.h 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/types.h> diff --git a/drivers/infiniband/hw/mthca/mthca_eq.c b/drivers/infiniband/hw/mthca/mthca_eq.c index 8bde7f98e58a..4e36aa7cb3d2 100644 --- a/drivers/infiniband/hw/mthca/mthca_eq.c +++ b/drivers/infiniband/hw/mthca/mthca_eq.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_eq.c 1382 2004-12-24 02:21:02Z roland $ */ #include <linux/errno.h> diff --git a/drivers/infiniband/hw/mthca/mthca_mad.c b/drivers/infiniband/hw/mthca/mthca_mad.c index 8b7e83e6e88f..640449582aba 100644 --- a/drivers/infiniband/hw/mthca/mthca_mad.c +++ b/drivers/infiniband/hw/mthca/mthca_mad.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_mad.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/string.h> diff --git a/drivers/infiniband/hw/mthca/mthca_main.c b/drivers/infiniband/hw/mthca/mthca_main.c index 200cf13fc9bb..fb9f91b60f30 100644 --- a/drivers/infiniband/hw/mthca/mthca_main.c +++ b/drivers/infiniband/hw/mthca/mthca_main.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_main.c 1396 2004-12-28 04:10:27Z roland $ */ #include <linux/module.h> diff --git a/drivers/infiniband/hw/mthca/mthca_mcg.c b/drivers/infiniband/hw/mthca/mthca_mcg.c index a8ad072be074..3f5f94879208 100644 --- a/drivers/infiniband/hw/mthca/mthca_mcg.c +++ b/drivers/infiniband/hw/mthca/mthca_mcg.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_mcg.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/string.h> diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.c b/drivers/infiniband/hw/mthca/mthca_memfree.c index d5862e5d99a0..1f7d1a29d2a8 100644 --- a/drivers/infiniband/hw/mthca/mthca_memfree.c +++ b/drivers/infiniband/hw/mthca/mthca_memfree.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id$ */ #include <linux/mm.h> diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.h b/drivers/infiniband/hw/mthca/mthca_memfree.h index a1ab06847b75..da9b8f9b884f 100644 --- a/drivers/infiniband/hw/mthca/mthca_memfree.h +++ b/drivers/infiniband/hw/mthca/mthca_memfree.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id$ */ #ifndef MTHCA_MEMFREE_H diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c index 820205dec560..8489b1e81c0f 100644 --- a/drivers/infiniband/hw/mthca/mthca_mr.c +++ b/drivers/infiniband/hw/mthca/mthca_mr.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_mr.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/slab.h> diff --git a/drivers/infiniband/hw/mthca/mthca_pd.c b/drivers/infiniband/hw/mthca/mthca_pd.c index c1e950764bd8..266f14e47406 100644 --- a/drivers/infiniband/hw/mthca/mthca_pd.c +++ b/drivers/infiniband/hw/mthca/mthca_pd.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_pd.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/errno.h> diff --git a/drivers/infiniband/hw/mthca/mthca_profile.c b/drivers/infiniband/hw/mthca/mthca_profile.c index 605a8d57fac6..d168c2540611 100644 --- a/drivers/infiniband/hw/mthca/mthca_profile.c +++ b/drivers/infiniband/hw/mthca/mthca_profile.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_profile.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/module.h> diff --git a/drivers/infiniband/hw/mthca/mthca_profile.h b/drivers/infiniband/hw/mthca/mthca_profile.h index e76cb62d8e32..62b009cc8730 100644 --- a/drivers/infiniband/hw/mthca/mthca_profile.h +++ b/drivers/infiniband/hw/mthca/mthca_profile.h @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_profile.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef MTHCA_PROFILE_H diff --git a/drivers/infiniband/hw/mthca/mthca_provider.c b/drivers/infiniband/hw/mthca/mthca_provider.c index be34f99ca625..87ad889e367b 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.c +++ b/drivers/infiniband/hw/mthca/mthca_provider.c @@ -32,8 +32,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_provider.c 4859 2006-01-09 21:55:10Z roland $ */ #include <rdma/ib_smi.h> diff --git a/drivers/infiniband/hw/mthca/mthca_provider.h b/drivers/infiniband/hw/mthca/mthca_provider.h index 934bf9544037..c621f8794b88 100644 --- a/drivers/infiniband/hw/mthca/mthca_provider.h +++ b/drivers/infiniband/hw/mthca/mthca_provider.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_provider.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef MTHCA_PROVIDER_H diff --git a/drivers/infiniband/hw/mthca/mthca_qp.c b/drivers/infiniband/hw/mthca/mthca_qp.c index 09dc3614cf2c..f5081bfde6db 100644 --- a/drivers/infiniband/hw/mthca/mthca_qp.c +++ b/drivers/infiniband/hw/mthca/mthca_qp.c @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_qp.c 1355 2004-12-17 15:23:43Z roland $ */ #include <linux/string.h> @@ -850,23 +848,6 @@ out: return err; } -static const struct ib_qp_attr dummy_init_attr = { .port_num = 1 }; -static const int dummy_init_attr_mask[] = { - [IB_QPT_UD] = (IB_QP_PKEY_INDEX | - IB_QP_PORT | - IB_QP_QKEY), - [IB_QPT_UC] = (IB_QP_PKEY_INDEX | - IB_QP_PORT | - IB_QP_ACCESS_FLAGS), - [IB_QPT_RC] = (IB_QP_PKEY_INDEX | - IB_QP_PORT | - IB_QP_ACCESS_FLAGS), - [IB_QPT_SMI] = (IB_QP_PKEY_INDEX | - IB_QP_QKEY), - [IB_QPT_GSI] = (IB_QP_PKEY_INDEX | - IB_QP_QKEY), -}; - int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, struct ib_udata *udata) { @@ -928,15 +909,6 @@ int mthca_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, goto out; } - if (cur_state == IB_QPS_RESET && new_state == IB_QPS_ERR) { - err = __mthca_modify_qp(ibqp, &dummy_init_attr, - dummy_init_attr_mask[ibqp->qp_type], - IB_QPS_RESET, IB_QPS_INIT); - if (err) - goto out; - cur_state = IB_QPS_INIT; - } - err = __mthca_modify_qp(ibqp, attr, attr_mask, cur_state, new_state); out: @@ -1277,10 +1249,10 @@ static int mthca_set_qp_size(struct mthca_dev *dev, struct ib_qp_cap *cap, return -EINVAL; /* - * For MLX transport we need 2 extra S/G entries: + * For MLX transport we need 2 extra send gather entries: * one for the header and one for the checksum at the end */ - if (qp->transport == MLX && cap->max_recv_sge + 2 > dev->limits.max_sg) + if (qp->transport == MLX && cap->max_send_sge + 2 > dev->limits.max_sg) return -EINVAL; if (mthca_is_memfree(dev)) { diff --git a/drivers/infiniband/hw/mthca/mthca_reset.c b/drivers/infiniband/hw/mthca/mthca_reset.c index 91934f2d9dba..acb6817f6060 100644 --- a/drivers/infiniband/hw/mthca/mthca_reset.c +++ b/drivers/infiniband/hw/mthca/mthca_reset.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_reset.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/init.h> diff --git a/drivers/infiniband/hw/mthca/mthca_srq.c b/drivers/infiniband/hw/mthca/mthca_srq.c index a5ffff6e1026..4fabe62aab8a 100644 --- a/drivers/infiniband/hw/mthca/mthca_srq.c +++ b/drivers/infiniband/hw/mthca/mthca_srq.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_srq.c 3047 2005-08-10 03:59:35Z roland $ */ #include <linux/slab.h> diff --git a/drivers/infiniband/hw/mthca/mthca_uar.c b/drivers/infiniband/hw/mthca/mthca_uar.c index 8b728486410d..ca5900c96fcf 100644 --- a/drivers/infiniband/hw/mthca/mthca_uar.c +++ b/drivers/infiniband/hw/mthca/mthca_uar.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id$ */ #include <asm/page.h> /* PAGE_SHIFT */ diff --git a/drivers/infiniband/hw/mthca/mthca_user.h b/drivers/infiniband/hw/mthca/mthca_user.h index e1262c942db8..5fe56e810739 100644 --- a/drivers/infiniband/hw/mthca/mthca_user.h +++ b/drivers/infiniband/hw/mthca/mthca_user.h @@ -29,7 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * */ #ifndef MTHCA_USER_H diff --git a/drivers/infiniband/hw/mthca/mthca_wqe.h b/drivers/infiniband/hw/mthca/mthca_wqe.h index b3551a8dea1d..341a5ae881c1 100644 --- a/drivers/infiniband/hw/mthca/mthca_wqe.h +++ b/drivers/infiniband/hw/mthca/mthca_wqe.h @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: mthca_wqe.h 3047 2005-08-10 03:59:35Z roland $ */ #ifndef MTHCA_WQE_H diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c index a4e9269a29bd..d2884e778098 100644 --- a/drivers/infiniband/hw/nes/nes.c +++ b/drivers/infiniband/hw/nes/nes.c @@ -328,7 +328,7 @@ void nes_rem_ref(struct ib_qp *ibqp) set_wqe_32bit_value(cqp_wqe->wqe_words, NES_CQP_WQE_ID_IDX, nesqp->hwqp.qp_id); u64temp = (u64)nesqp->nesqp_context_pbase; set_wqe_64bit_value(cqp_wqe->wqe_words, NES_CQP_QP_WQE_CONTEXT_LOW_IDX, u64temp); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); } } diff --git a/drivers/infiniband/hw/nes/nes.h b/drivers/infiniband/hw/nes/nes.h index 61b46e9c7d2d..39bd897b40c6 100644 --- a/drivers/infiniband/hw/nes/nes.h +++ b/drivers/infiniband/hw/nes/nes.h @@ -94,9 +94,6 @@ #define MAX_DPC_ITERATIONS 128 -#define NES_CQP_REQUEST_NO_DOORBELL_RING 0 -#define NES_CQP_REQUEST_RING_DOORBELL 1 - #define NES_DRV_OPT_ENABLE_MPA_VER_0 0x00000001 #define NES_DRV_OPT_DISABLE_MPA_CRC 0x00000002 #define NES_DRV_OPT_DISABLE_FIRST_WRITE 0x00000004 @@ -538,7 +535,11 @@ void nes_read_1G_phy_reg(struct nes_device *, u8, u8, u16 *); void nes_write_10G_phy_reg(struct nes_device *, u16, u8, u16, u16); void nes_read_10G_phy_reg(struct nes_device *, u8, u8, u16); struct nes_cqp_request *nes_get_cqp_request(struct nes_device *); -void nes_post_cqp_request(struct nes_device *, struct nes_cqp_request *, int); +void nes_free_cqp_request(struct nes_device *nesdev, + struct nes_cqp_request *cqp_request); +void nes_put_cqp_request(struct nes_device *nesdev, + struct nes_cqp_request *cqp_request); +void nes_post_cqp_request(struct nes_device *, struct nes_cqp_request *); int nes_arp_table(struct nes_device *, u32, u8 *, u32); void nes_mh_fix(unsigned long); void nes_clc(unsigned long); diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 9a4b40fae40d..6aa531d5276d 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -1603,7 +1603,6 @@ static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core, return NULL; } - memset(listener, 0, sizeof(struct nes_cm_listener)); listener->loc_addr = htonl(cm_info->loc_addr); listener->loc_port = htons(cm_info->loc_port); listener->reused_node = 0; diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index d3278f111ca7..85f26d19a32b 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -398,7 +398,7 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { nesadapter->base_pd = 1; nesadapter->device_cap_flags = - IB_DEVICE_ZERO_STAG | IB_DEVICE_MEM_WINDOW; + IB_DEVICE_LOCAL_DMA_LKEY | IB_DEVICE_MEM_WINDOW; nesadapter->allocated_qps = (unsigned long *)&(((unsigned char *)nesadapter) [(sizeof(struct nes_adapter)+(sizeof(unsigned long)-1))&(~(sizeof(unsigned long)-1))]); @@ -2710,39 +2710,11 @@ static void nes_cqp_ce_handler(struct nes_device *nesdev, struct nes_hw_cq *cq) barrier(); cqp_request->request_done = 1; wake_up(&cqp_request->waitq); - if (atomic_dec_and_test(&cqp_request->refcount)) { - nes_debug(NES_DBG_CQP, "CQP request %p (opcode 0x%02X) freed.\n", - cqp_request, - le32_to_cpu(cqp_request->cqp_wqe.wqe_words[NES_CQP_WQE_OPCODE_IDX])&0x3f); - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } - } else if (cqp_request->callback) { - /* Envoke the callback routine */ - cqp_request->cqp_callback(nesdev, cqp_request); - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } + nes_put_cqp_request(nesdev, cqp_request); } else { - nes_debug(NES_DBG_CQP, "CQP request %p (opcode 0x%02X) freed.\n", - cqp_request, - le32_to_cpu(cqp_request->cqp_wqe.wqe_words[NES_CQP_WQE_OPCODE_IDX]) & 0x3f); - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } + if (cqp_request->callback) + cqp_request->cqp_callback(nesdev, cqp_request); + nes_free_cqp_request(nesdev, cqp_request); } } else { wake_up(&nesdev->cqp.waitq); @@ -3149,7 +3121,6 @@ int nes_manage_apbvt(struct nes_vnic *nesvnic, u32 accel_local_port, { struct nes_device *nesdev = nesvnic->nesdev; struct nes_hw_cqp_wqe *cqp_wqe; - unsigned long flags; struct nes_cqp_request *cqp_request; int ret = 0; u16 major_code; @@ -3176,7 +3147,7 @@ int nes_manage_apbvt(struct nes_vnic *nesvnic, u32 accel_local_port, nes_debug(NES_DBG_QP, "Waiting for CQP completion for APBVT.\n"); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); if (add_port == NES_MANAGE_APBVT_ADD) ret = wait_event_timeout(cqp_request->waitq, (cqp_request->request_done != 0), @@ -3184,15 +3155,9 @@ int nes_manage_apbvt(struct nes_vnic *nesvnic, u32 accel_local_port, nes_debug(NES_DBG_QP, "Completed, ret=%u, CQP Major:Minor codes = 0x%04X:0x%04X\n", ret, cqp_request->major_code, cqp_request->minor_code); major_code = cqp_request->major_code; - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + + nes_put_cqp_request(nesdev, cqp_request); + if (!ret) return -ETIME; else if (major_code) @@ -3252,7 +3217,7 @@ void nes_manage_arp_cache(struct net_device *netdev, unsigned char *mac_addr, nesdev->cqp.sq_head, nesdev->cqp.sq_tail); atomic_set(&cqp_request->refcount, 1); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); } @@ -3262,7 +3227,6 @@ void nes_manage_arp_cache(struct net_device *netdev, unsigned char *mac_addr, void flush_wqes(struct nes_device *nesdev, struct nes_qp *nesqp, u32 which_wq, u32 wait_completion) { - unsigned long flags; struct nes_cqp_request *cqp_request; struct nes_hw_cqp_wqe *cqp_wqe; int ret; @@ -3285,7 +3249,7 @@ void flush_wqes(struct nes_device *nesdev, struct nes_qp *nesqp, cpu_to_le32(NES_CQP_FLUSH_WQES | which_wq); cqp_wqe->wqe_words[NES_CQP_WQE_ID_IDX] = cpu_to_le32(nesqp->hwqp.qp_id); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); if (wait_completion) { /* Wait for CQP */ @@ -3294,14 +3258,6 @@ void flush_wqes(struct nes_device *nesdev, struct nes_qp *nesqp, nes_debug(NES_DBG_QP, "Flush SQ QP WQEs completed, ret=%u," " CQP Major:Minor codes = 0x%04X:0x%04X\n", ret, cqp_request->major_code, cqp_request->minor_code); - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + nes_put_cqp_request(nesdev, cqp_request); } } diff --git a/drivers/infiniband/hw/nes/nes_hw.h b/drivers/infiniband/hw/nes/nes_hw.h index 745bf94f3f07..7b81e0ae0076 100644 --- a/drivers/infiniband/hw/nes/nes_hw.h +++ b/drivers/infiniband/hw/nes/nes_hw.h @@ -1172,7 +1172,7 @@ struct nes_vnic { u32 mcrq_qp_id; struct nes_ucontext *mcrq_ucontext; struct nes_cqp_request* (*get_cqp_request)(struct nes_device *nesdev); - void (*post_cqp_request)(struct nes_device*, struct nes_cqp_request *, int); + void (*post_cqp_request)(struct nes_device*, struct nes_cqp_request *); int (*mcrq_mcast_filter)( struct nes_vnic* nesvnic, __u8* dmi_addr ); struct net_device_stats netstats; /* used to put the netdev on the adapters logical port list */ diff --git a/drivers/infiniband/hw/nes/nes_utils.c b/drivers/infiniband/hw/nes/nes_utils.c index fe83d1b2b177..fb8cbd71a2ef 100644 --- a/drivers/infiniband/hw/nes/nes_utils.c +++ b/drivers/infiniband/hw/nes/nes_utils.c @@ -567,12 +567,36 @@ struct nes_cqp_request *nes_get_cqp_request(struct nes_device *nesdev) return cqp_request; } +void nes_free_cqp_request(struct nes_device *nesdev, + struct nes_cqp_request *cqp_request) +{ + unsigned long flags; + + nes_debug(NES_DBG_CQP, "CQP request %p (opcode 0x%02X) freed.\n", + cqp_request, + le32_to_cpu(cqp_request->cqp_wqe.wqe_words[NES_CQP_WQE_OPCODE_IDX]) & 0x3f); + + if (cqp_request->dynamic) { + kfree(cqp_request); + } else { + spin_lock_irqsave(&nesdev->cqp.lock, flags); + list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); + spin_unlock_irqrestore(&nesdev->cqp.lock, flags); + } +} + +void nes_put_cqp_request(struct nes_device *nesdev, + struct nes_cqp_request *cqp_request) +{ + if (atomic_dec_and_test(&cqp_request->refcount)) + nes_free_cqp_request(nesdev, cqp_request); +} /** * nes_post_cqp_request */ void nes_post_cqp_request(struct nes_device *nesdev, - struct nes_cqp_request *cqp_request, int ring_doorbell) + struct nes_cqp_request *cqp_request) { struct nes_hw_cqp_wqe *cqp_wqe; unsigned long flags; @@ -600,10 +624,9 @@ void nes_post_cqp_request(struct nes_device *nesdev, nesdev->cqp.sq_head, nesdev->cqp.sq_tail, nesdev->cqp.sq_size, cqp_request->waiting, atomic_read(&cqp_request->refcount)); barrier(); - if (ring_doorbell) { - /* Ring doorbell (1 WQEs) */ - nes_write32(nesdev->regs+NES_WQE_ALLOC, 0x01800000 | nesdev->cqp.qp_id); - } + + /* Ring doorbell (1 WQEs) */ + nes_write32(nesdev->regs+NES_WQE_ALLOC, 0x01800000 | nesdev->cqp.qp_id); barrier(); } else { diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c index d617da9bd351..e3939d13484e 100644 --- a/drivers/infiniband/hw/nes/nes_verbs.c +++ b/drivers/infiniband/hw/nes/nes_verbs.c @@ -55,7 +55,6 @@ static void nes_unregister_ofa_device(struct nes_ib_device *nesibdev); * nes_alloc_mw */ static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd) { - unsigned long flags; struct nes_pd *nespd = to_nespd(ibpd); struct nes_vnic *nesvnic = to_nesvnic(ibpd->device); struct nes_device *nesdev = nesvnic->nesdev; @@ -119,7 +118,7 @@ static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd) { set_wqe_32bit_value(cqp_wqe->wqe_words, NES_CQP_STAG_WQE_STAG_IDX, stag); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ ret = wait_event_timeout(cqp_request->waitq, (cqp_request->request_done != 0), @@ -128,15 +127,7 @@ static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd) { " CQP Major:Minor codes = 0x%04X:0x%04X.\n", stag, ret, cqp_request->major_code, cqp_request->minor_code); if ((!ret) || (cqp_request->major_code)) { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + nes_put_cqp_request(nesdev, cqp_request); kfree(nesmr); nes_free_resource(nesadapter, nesadapter->allocated_mrs, stag_index); if (!ret) { @@ -144,17 +135,8 @@ static struct ib_mw *nes_alloc_mw(struct ib_pd *ibpd) { } else { return ERR_PTR(-ENOMEM); } - } else { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } } + nes_put_cqp_request(nesdev, cqp_request); nesmr->ibmw.rkey = stag; nesmr->mode = IWNES_MEMREG_TYPE_MW; @@ -178,7 +160,6 @@ static int nes_dealloc_mw(struct ib_mw *ibmw) struct nes_hw_cqp_wqe *cqp_wqe; struct nes_cqp_request *cqp_request; int err = 0; - unsigned long flags; int ret; /* Deallocate the window with the adapter */ @@ -194,7 +175,7 @@ static int nes_dealloc_mw(struct ib_mw *ibmw) set_wqe_32bit_value(cqp_wqe->wqe_words, NES_CQP_STAG_WQE_STAG_IDX, ibmw->rkey); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ nes_debug(NES_DBG_MR, "Waiting for deallocate STag 0x%08X to complete.\n", @@ -204,32 +185,12 @@ static int nes_dealloc_mw(struct ib_mw *ibmw) nes_debug(NES_DBG_MR, "Deallocate STag completed, wait_event_timeout ret = %u," " CQP Major:Minor codes = 0x%04X:0x%04X.\n", ret, cqp_request->major_code, cqp_request->minor_code); - if ((!ret) || (cqp_request->major_code)) { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } - if (!ret) { - err = -ETIME; - } else { - err = -EIO; - } - } else { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } - } + if (!ret) + err = -ETIME; + else if (cqp_request->major_code) + err = -EIO; + + nes_put_cqp_request(nesdev, cqp_request); nes_free_resource(nesadapter, nesadapter->allocated_mrs, (ibmw->rkey & 0x0fffff00) >> 8); @@ -516,7 +477,7 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd, (nesfmr->nesmr.pbls_used-1) : nesfmr->nesmr.pbls_used); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ ret = wait_event_timeout(cqp_request->waitq, (cqp_request->request_done != 0), @@ -526,29 +487,11 @@ static struct ib_fmr *nes_alloc_fmr(struct ib_pd *ibpd, stag, ret, cqp_request->major_code, cqp_request->minor_code); if ((!ret) || (cqp_request->major_code)) { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + nes_put_cqp_request(nesdev, cqp_request); ret = (!ret) ? -ETIME : -EIO; goto failed_leaf_vpbl_pages_alloc; - } else { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } } - + nes_put_cqp_request(nesdev, cqp_request); nesfmr->nesmr.ibfmr.lkey = stag; nesfmr->nesmr.ibfmr.rkey = stag; nesfmr->attr = *ibfmr_attr; @@ -1474,7 +1417,7 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, set_wqe_64bit_value(cqp_wqe->wqe_words, NES_CQP_QP_WQE_CONTEXT_LOW_IDX, u64temp); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ nes_debug(NES_DBG_QP, "Waiting for create iWARP QP%u to complete.\n", @@ -1487,15 +1430,7 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, nesqp->hwqp.qp_id, ret, nesdev->cqp.sq_head, nesdev->cqp.sq_tail, cqp_request->major_code, cqp_request->minor_code); if ((!ret) || (cqp_request->major_code)) { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + nes_put_cqp_request(nesdev, cqp_request); nes_free_resource(nesadapter, nesadapter->allocated_qps, qp_num); nes_free_qp_mem(nesdev, nesqp,virt_wqs); kfree(nesqp->allocated_buffer); @@ -1504,18 +1439,10 @@ static struct ib_qp *nes_create_qp(struct ib_pd *ibpd, } else { return ERR_PTR(-EIO); } - } else { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } } + nes_put_cqp_request(nesdev, cqp_request); + if (ibpd->uobject) { uresp.mmap_sq_db_index = nesqp->mmap_sq_db_index; uresp.actual_sq_size = sq_size; @@ -1817,7 +1744,7 @@ static struct ib_cq *nes_create_cq(struct ib_device *ibdev, int entries, cpu_to_le32(((u32)((u64temp) >> 33)) & 0x7FFFFFFF); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ nes_debug(NES_DBG_CQ, "Waiting for create iWARP CQ%u to complete.\n", @@ -1827,32 +1754,15 @@ static struct ib_cq *nes_create_cq(struct ib_device *ibdev, int entries, nes_debug(NES_DBG_CQ, "Create iWARP CQ%u completed, wait_event_timeout ret = %d.\n", nescq->hw_cq.cq_number, ret); if ((!ret) || (cqp_request->major_code)) { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + nes_put_cqp_request(nesdev, cqp_request); if (!context) pci_free_consistent(nesdev->pcidev, nescq->cq_mem_size, mem, nescq->hw_cq.cq_pbase); nes_free_resource(nesadapter, nesadapter->allocated_cqs, cq_num); kfree(nescq); return ERR_PTR(-EIO); - } else { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } } + nes_put_cqp_request(nesdev, cqp_request); if (context) { /* free the nespbl */ @@ -1931,7 +1841,7 @@ static int nes_destroy_cq(struct ib_cq *ib_cq) (nescq->hw_cq.cq_number | ((u32)PCI_FUNC(nesdev->pcidev->devfn) << 16))); nes_free_resource(nesadapter, nesadapter->allocated_cqs, nescq->hw_cq.cq_number); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ nes_debug(NES_DBG_CQ, "Waiting for destroy iWARP CQ%u to complete.\n", @@ -1942,37 +1852,18 @@ static int nes_destroy_cq(struct ib_cq *ib_cq) " CQP Major:Minor codes = 0x%04X:0x%04X.\n", nescq->hw_cq.cq_number, ret, cqp_request->major_code, cqp_request->minor_code); - if ((!ret) || (cqp_request->major_code)) { - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } - if (!ret) { - nes_debug(NES_DBG_CQ, "iWARP CQ%u destroy timeout expired\n", + if (!ret) { + nes_debug(NES_DBG_CQ, "iWARP CQ%u destroy timeout expired\n", nescq->hw_cq.cq_number); - ret = -ETIME; - } else { - nes_debug(NES_DBG_CQ, "iWARP CQ%u destroy failed\n", + ret = -ETIME; + } else if (cqp_request->major_code) { + nes_debug(NES_DBG_CQ, "iWARP CQ%u destroy failed\n", nescq->hw_cq.cq_number); - ret = -EIO; - } + ret = -EIO; } else { ret = 0; - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } } + nes_put_cqp_request(nesdev, cqp_request); if (nescq->cq_mem_size) pci_free_consistent(nesdev->pcidev, nescq->cq_mem_size, @@ -2096,7 +1987,7 @@ static int nes_reg_mr(struct nes_device *nesdev, struct nes_pd *nespd, barrier(); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ ret = wait_event_timeout(cqp_request->waitq, (0 != cqp_request->request_done), @@ -2105,15 +1996,8 @@ static int nes_reg_mr(struct nes_device *nesdev, struct nes_pd *nespd, " CQP Major:Minor codes = 0x%04X:0x%04X.\n", stag, ret, cqp_request->major_code, cqp_request->minor_code); major_code = cqp_request->major_code; - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + nes_put_cqp_request(nesdev, cqp_request); + if (!ret) return -ETIME; else if (major_code) @@ -2754,7 +2638,7 @@ static int nes_dereg_mr(struct ib_mr *ib_mr) set_wqe_32bit_value(cqp_wqe->wqe_words, NES_CQP_STAG_WQE_STAG_IDX, ib_mr->rkey); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ nes_debug(NES_DBG_MR, "Waiting for deallocate STag 0x%08X completed\n", ib_mr->rkey); @@ -2771,15 +2655,9 @@ static int nes_dereg_mr(struct ib_mr *ib_mr) major_code = cqp_request->major_code; minor_code = cqp_request->minor_code; - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + + nes_put_cqp_request(nesdev, cqp_request); + if (!ret) { nes_debug(NES_DBG_MR, "Timeout waiting to destroy STag," " ib_mr=%p, rkey = 0x%08X\n", @@ -2904,7 +2782,6 @@ int nes_hw_modify_qp(struct nes_device *nesdev, struct nes_qp *nesqp, /* struct iw_cm_id *cm_id = nesqp->cm_id; */ /* struct iw_cm_event cm_event; */ struct nes_cqp_request *cqp_request; - unsigned long flags; int ret; u16 major_code; @@ -2932,7 +2809,7 @@ int nes_hw_modify_qp(struct nes_device *nesdev, struct nes_qp *nesqp, set_wqe_64bit_value(cqp_wqe->wqe_words, NES_CQP_QP_WQE_CONTEXT_LOW_IDX, (u64)nesqp->nesqp_context_pbase); atomic_set(&cqp_request->refcount, 2); - nes_post_cqp_request(nesdev, cqp_request, NES_CQP_REQUEST_RING_DOORBELL); + nes_post_cqp_request(nesdev, cqp_request); /* Wait for CQP */ if (wait_completion) { @@ -2950,15 +2827,9 @@ int nes_hw_modify_qp(struct nes_device *nesdev, struct nes_qp *nesqp, nesqp->hwqp.qp_id, cqp_request->major_code, cqp_request->minor_code, next_iwarp_state); } - if (atomic_dec_and_test(&cqp_request->refcount)) { - if (cqp_request->dynamic) { - kfree(cqp_request); - } else { - spin_lock_irqsave(&nesdev->cqp.lock, flags); - list_add_tail(&cqp_request->list, &nesdev->cqp_avail_reqs); - spin_unlock_irqrestore(&nesdev->cqp.lock, flags); - } - } + + nes_put_cqp_request(nesdev, cqp_request); + if (!ret) return -ETIME; else if (major_code) diff --git a/drivers/infiniband/ulp/ipoib/Kconfig b/drivers/infiniband/ulp/ipoib/Kconfig index 1f76bad020f3..691525cf394a 100644 --- a/drivers/infiniband/ulp/ipoib/Kconfig +++ b/drivers/infiniband/ulp/ipoib/Kconfig @@ -1,6 +1,7 @@ config INFINIBAND_IPOIB tristate "IP-over-InfiniBand" depends on NETDEVICES && INET && (IPV6 || IPV6=n) + select INET_LRO ---help--- Support for the IP-over-InfiniBand protocol (IPoIB). This transports IP packets over InfiniBand so you can use your IB diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index ca126fc2b853..b0ffc9abe8c0 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib.h 1358 2004-12-17 22:00:11Z roland $ */ #ifndef _IPOIB_H @@ -52,9 +50,16 @@ #include <rdma/ib_verbs.h> #include <rdma/ib_pack.h> #include <rdma/ib_sa.h> +#include <linux/inet_lro.h> /* constants */ +enum ipoib_flush_level { + IPOIB_FLUSH_LIGHT, + IPOIB_FLUSH_NORMAL, + IPOIB_FLUSH_HEAVY +}; + enum { IPOIB_ENCAP_LEN = 4, @@ -65,8 +70,8 @@ enum { IPOIB_CM_BUF_SIZE = IPOIB_CM_MTU + IPOIB_ENCAP_LEN, IPOIB_CM_HEAD_SIZE = IPOIB_CM_BUF_SIZE % PAGE_SIZE, IPOIB_CM_RX_SG = ALIGN(IPOIB_CM_BUF_SIZE, PAGE_SIZE) / PAGE_SIZE, - IPOIB_RX_RING_SIZE = 128, - IPOIB_TX_RING_SIZE = 64, + IPOIB_RX_RING_SIZE = 256, + IPOIB_TX_RING_SIZE = 128, IPOIB_MAX_QUEUE_SIZE = 8192, IPOIB_MIN_QUEUE_SIZE = 2, IPOIB_CM_MAX_CONN_QP = 4096, @@ -84,7 +89,6 @@ enum { IPOIB_FLAG_SUBINTERFACE = 5, IPOIB_MCAST_RUN = 6, IPOIB_STOP_REAPER = 7, - IPOIB_MCAST_STARTED = 8, IPOIB_FLAG_ADMIN_CM = 9, IPOIB_FLAG_UMCAST = 10, IPOIB_FLAG_CSUM = 11, @@ -96,7 +100,11 @@ enum { IPOIB_MCAST_FLAG_BUSY = 2, /* joining or already joined */ IPOIB_MCAST_FLAG_ATTACHED = 3, + IPOIB_MAX_LRO_DESCRIPTORS = 8, + IPOIB_LRO_MAX_AGGR = 64, + MAX_SEND_CQE = 16, + IPOIB_CM_COPYBREAK = 256, }; #define IPOIB_OP_RECV (1ul << 31) @@ -149,6 +157,11 @@ struct ipoib_tx_buf { u64 mapping[MAX_SKB_FRAGS + 1]; }; +struct ipoib_cm_tx_buf { + struct sk_buff *skb; + u64 mapping; +}; + struct ib_cm_id; struct ipoib_cm_data { @@ -207,7 +220,7 @@ struct ipoib_cm_tx { struct net_device *dev; struct ipoib_neigh *neigh; struct ipoib_path *path; - struct ipoib_tx_buf *tx_ring; + struct ipoib_cm_tx_buf *tx_ring; unsigned tx_head; unsigned tx_tail; unsigned long flags; @@ -249,6 +262,11 @@ struct ipoib_ethtool_st { u16 max_coalesced_frames; }; +struct ipoib_lro { + struct net_lro_mgr lro_mgr; + struct net_lro_desc lro_desc[IPOIB_MAX_LRO_DESCRIPTORS]; +}; + /* * Device private locking: tx_lock protects members used in TX fast * path (and we use LLTX so upper layers don't do extra locking). @@ -264,7 +282,6 @@ struct ipoib_dev_priv { unsigned long flags; - struct mutex mcast_mutex; struct mutex vlan_mutex; struct rb_root path_tree; @@ -276,10 +293,11 @@ struct ipoib_dev_priv { struct delayed_work pkey_poll_task; struct delayed_work mcast_task; - struct work_struct flush_task; + struct work_struct flush_light; + struct work_struct flush_normal; + struct work_struct flush_heavy; struct work_struct restart_task; struct delayed_work ah_reap_task; - struct work_struct pkey_event_task; struct ib_device *ca; u8 port; @@ -335,6 +353,8 @@ struct ipoib_dev_priv { int hca_caps; struct ipoib_ethtool_st ethtool; struct timer_list poll_timer; + + struct ipoib_lro lro; }; struct ipoib_ah { @@ -359,6 +379,7 @@ struct ipoib_path { struct rb_node rb_node; struct list_head list; + int valid; }; struct ipoib_neigh { @@ -423,11 +444,14 @@ void ipoib_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_ah *address, u32 qpn); void ipoib_reap_ah(struct work_struct *work); +void ipoib_mark_paths_invalid(struct net_device *dev); void ipoib_flush_paths(struct net_device *dev); struct ipoib_dev_priv *ipoib_intf_alloc(const char *format); int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port); -void ipoib_ib_dev_flush(struct work_struct *work); +void ipoib_ib_dev_flush_light(struct work_struct *work); +void ipoib_ib_dev_flush_normal(struct work_struct *work); +void ipoib_ib_dev_flush_heavy(struct work_struct *work); void ipoib_pkey_event(struct work_struct *work); void ipoib_ib_dev_cleanup(struct net_device *dev); @@ -466,9 +490,7 @@ void ipoib_path_iter_read(struct ipoib_path_iter *iter, #endif int ipoib_mcast_attach(struct net_device *dev, u16 mlid, - union ib_gid *mgid); -int ipoib_mcast_detach(struct net_device *dev, u16 mlid, - union ib_gid *mgid); + union ib_gid *mgid, int set_qkey); int ipoib_init_qp(struct net_device *dev); int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 97e67d36378f..0f2d3045061a 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id$ */ #include <rdma/ib_cm.h> @@ -113,18 +111,20 @@ static int ipoib_cm_post_receive_srq(struct net_device *dev, int id) } static int ipoib_cm_post_receive_nonsrq(struct net_device *dev, - struct ipoib_cm_rx *rx, int id) + struct ipoib_cm_rx *rx, + struct ib_recv_wr *wr, + struct ib_sge *sge, int id) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct ib_recv_wr *bad_wr; int i, ret; - priv->cm.rx_wr.wr_id = id | IPOIB_OP_CM | IPOIB_OP_RECV; + wr->wr_id = id | IPOIB_OP_CM | IPOIB_OP_RECV; for (i = 0; i < IPOIB_CM_RX_SG; ++i) - priv->cm.rx_sge[i].addr = rx->rx_ring[id].mapping[i]; + sge[i].addr = rx->rx_ring[id].mapping[i]; - ret = ib_post_recv(rx->qp, &priv->cm.rx_wr, &bad_wr); + ret = ib_post_recv(rx->qp, wr, &bad_wr); if (unlikely(ret)) { ipoib_warn(priv, "post recv failed for buf %d (%d)\n", id, ret); ipoib_cm_dma_unmap_rx(priv, IPOIB_CM_RX_SG - 1, @@ -322,10 +322,33 @@ static int ipoib_cm_modify_rx_qp(struct net_device *dev, return 0; } +static void ipoib_cm_init_rx_wr(struct net_device *dev, + struct ib_recv_wr *wr, + struct ib_sge *sge) +{ + struct ipoib_dev_priv *priv = netdev_priv(dev); + int i; + + for (i = 0; i < priv->cm.num_frags; ++i) + sge[i].lkey = priv->mr->lkey; + + sge[0].length = IPOIB_CM_HEAD_SIZE; + for (i = 1; i < priv->cm.num_frags; ++i) + sge[i].length = PAGE_SIZE; + + wr->next = NULL; + wr->sg_list = priv->cm.rx_sge; + wr->num_sge = priv->cm.num_frags; +} + static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_id, struct ipoib_cm_rx *rx) { struct ipoib_dev_priv *priv = netdev_priv(dev); + struct { + struct ib_recv_wr wr; + struct ib_sge sge[IPOIB_CM_RX_SG]; + } *t; int ret; int i; @@ -333,6 +356,14 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i if (!rx->rx_ring) return -ENOMEM; + t = kmalloc(sizeof *t, GFP_KERNEL); + if (!t) { + ret = -ENOMEM; + goto err_free; + } + + ipoib_cm_init_rx_wr(dev, &t->wr, t->sge); + spin_lock_irq(&priv->lock); if (priv->cm.nonsrq_conn_qp >= ipoib_max_conn_qp) { @@ -351,8 +382,8 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i ipoib_warn(priv, "failed to allocate receive buffer %d\n", i); ret = -ENOMEM; goto err_count; - } - ret = ipoib_cm_post_receive_nonsrq(dev, rx, i); + } + ret = ipoib_cm_post_receive_nonsrq(dev, rx, &t->wr, t->sge, i); if (ret) { ipoib_warn(priv, "ipoib_cm_post_receive_nonsrq " "failed for buf %d\n", i); @@ -363,6 +394,8 @@ static int ipoib_cm_nonsrq_init_rx(struct net_device *dev, struct ib_cm_id *cm_i rx->recv_count = ipoib_recvq_size; + kfree(t); + return 0; err_count: @@ -371,6 +404,7 @@ err_count: spin_unlock_irq(&priv->lock); err_free: + kfree(t); ipoib_cm_free_rx_ring(dev, rx->rx_ring); return ret; @@ -525,6 +559,7 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) u64 mapping[IPOIB_CM_RX_SG]; int frags; int has_srq; + struct sk_buff *small_skb; ipoib_dbg_data(priv, "cm recv completion: id %d, status: %d\n", wr_id, wc->status); @@ -579,6 +614,23 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) } } + if (wc->byte_len < IPOIB_CM_COPYBREAK) { + int dlen = wc->byte_len; + + small_skb = dev_alloc_skb(dlen + 12); + if (small_skb) { + skb_reserve(small_skb, 12); + ib_dma_sync_single_for_cpu(priv->ca, rx_ring[wr_id].mapping[0], + dlen, DMA_FROM_DEVICE); + skb_copy_from_linear_data(skb, small_skb->data, dlen); + ib_dma_sync_single_for_device(priv->ca, rx_ring[wr_id].mapping[0], + dlen, DMA_FROM_DEVICE); + skb_put(small_skb, dlen); + skb = small_skb; + goto copied; + } + } + frags = PAGE_ALIGN(wc->byte_len - min(wc->byte_len, (unsigned)IPOIB_CM_HEAD_SIZE)) / PAGE_SIZE; @@ -601,6 +653,7 @@ void ipoib_cm_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) skb_put_frags(skb, IPOIB_CM_HEAD_SIZE, wc->byte_len, newskb); +copied: skb->protocol = ((struct ipoib_header *) skb->data)->proto; skb_reset_mac_header(skb); skb_pull(skb, IPOIB_ENCAP_LEN); @@ -620,7 +673,10 @@ repost: ipoib_warn(priv, "ipoib_cm_post_receive_srq failed " "for buf %d\n", wr_id); } else { - if (unlikely(ipoib_cm_post_receive_nonsrq(dev, p, wr_id))) { + if (unlikely(ipoib_cm_post_receive_nonsrq(dev, p, + &priv->cm.rx_wr, + priv->cm.rx_sge, + wr_id))) { --p->recv_count; ipoib_warn(priv, "ipoib_cm_post_receive_nonsrq failed " "for buf %d\n", wr_id); @@ -647,7 +703,7 @@ static inline int post_send(struct ipoib_dev_priv *priv, void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_tx *tx) { struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ipoib_tx_buf *tx_req; + struct ipoib_cm_tx_buf *tx_req; u64 addr; if (unlikely(skb->len > tx->mtu)) { @@ -678,7 +734,7 @@ void ipoib_cm_send(struct net_device *dev, struct sk_buff *skb, struct ipoib_cm_ return; } - tx_req->mapping[0] = addr; + tx_req->mapping = addr; if (unlikely(post_send(priv, tx, tx->tx_head & (ipoib_sendq_size - 1), addr, skb->len))) { @@ -703,7 +759,7 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) struct ipoib_dev_priv *priv = netdev_priv(dev); struct ipoib_cm_tx *tx = wc->qp->qp_context; unsigned int wr_id = wc->wr_id & ~IPOIB_OP_CM; - struct ipoib_tx_buf *tx_req; + struct ipoib_cm_tx_buf *tx_req; unsigned long flags; ipoib_dbg_data(priv, "cm send completion: id %d, status: %d\n", @@ -717,7 +773,7 @@ void ipoib_cm_handle_tx_wc(struct net_device *dev, struct ib_wc *wc) tx_req = &tx->tx_ring[wr_id]; - ib_dma_unmap_single(priv->ca, tx_req->mapping[0], tx_req->skb->len, DMA_TO_DEVICE); + ib_dma_unmap_single(priv->ca, tx_req->mapping, tx_req->skb->len, DMA_TO_DEVICE); /* FIXME: is this right? Shouldn't we only increment on success? */ ++dev->stats.tx_packets; @@ -1087,7 +1143,7 @@ err_tx: static void ipoib_cm_tx_destroy(struct ipoib_cm_tx *p) { struct ipoib_dev_priv *priv = netdev_priv(p->dev); - struct ipoib_tx_buf *tx_req; + struct ipoib_cm_tx_buf *tx_req; unsigned long flags; unsigned long begin; @@ -1115,7 +1171,7 @@ timeout: while ((int) p->tx_tail - (int) p->tx_head < 0) { tx_req = &p->tx_ring[p->tx_tail & (ipoib_sendq_size - 1)]; - ib_dma_unmap_single(priv->ca, tx_req->mapping[0], tx_req->skb->len, + ib_dma_unmap_single(priv->ca, tx_req->mapping, tx_req->skb->len, DMA_TO_DEVICE); dev_kfree_skb_any(tx_req->skb); ++p->tx_tail; @@ -1384,7 +1440,9 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr, ipoib_warn(priv, "enabling connected mode " "will cause multicast packet drops\n"); + rtnl_lock(); dev->features &= ~(NETIF_F_IP_CSUM | NETIF_F_SG | NETIF_F_TSO); + rtnl_unlock(); priv->tx_wr.send_flags &= ~IB_SEND_IP_CSUM; ipoib_flush_paths(dev); @@ -1393,14 +1451,16 @@ static ssize_t set_mode(struct device *d, struct device_attribute *attr, if (!strcmp(buf, "datagram\n")) { clear_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); - dev->mtu = min(priv->mcast_mtu, dev->mtu); - ipoib_flush_paths(dev); + rtnl_lock(); if (test_bit(IPOIB_FLAG_CSUM, &priv->flags)) { dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG; if (priv->hca_caps & IB_DEVICE_UD_TSO) dev->features |= NETIF_F_TSO; } + dev_set_mtu(dev, min(priv->mcast_mtu, dev->mtu)); + rtnl_unlock(); + ipoib_flush_paths(dev); return count; } @@ -1485,15 +1545,7 @@ int ipoib_cm_dev_init(struct net_device *dev) priv->cm.num_frags = IPOIB_CM_RX_SG; } - for (i = 0; i < priv->cm.num_frags; ++i) - priv->cm.rx_sge[i].lkey = priv->mr->lkey; - - priv->cm.rx_sge[0].length = IPOIB_CM_HEAD_SIZE; - for (i = 1; i < priv->cm.num_frags; ++i) - priv->cm.rx_sge[i].length = PAGE_SIZE; - priv->cm.rx_wr.next = NULL; - priv->cm.rx_wr.sg_list = priv->cm.rx_sge; - priv->cm.rx_wr.num_sge = priv->cm.num_frags; + ipoib_cm_init_rx_wr(dev, &priv->cm.rx_wr, priv->cm.rx_sge); if (ipoib_cm_has_srq(dev)) { for (i = 0; i < ipoib_recvq_size; ++i) { diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c index 10279b79c44d..66af5c1a76e5 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c @@ -86,11 +86,57 @@ static int ipoib_set_coalesce(struct net_device *dev, return 0; } +static const char ipoib_stats_keys[][ETH_GSTRING_LEN] = { + "LRO aggregated", "LRO flushed", + "LRO avg aggr", "LRO no desc" +}; + +static void ipoib_get_strings(struct net_device *netdev, u32 stringset, u8 *data) +{ + switch (stringset) { + case ETH_SS_STATS: + memcpy(data, *ipoib_stats_keys, sizeof(ipoib_stats_keys)); + break; + } +} + +static int ipoib_get_sset_count(struct net_device *dev, int sset) +{ + switch (sset) { + case ETH_SS_STATS: + return ARRAY_SIZE(ipoib_stats_keys); + default: + return -EOPNOTSUPP; + } +} + +static void ipoib_get_ethtool_stats(struct net_device *dev, + struct ethtool_stats *stats, uint64_t *data) +{ + struct ipoib_dev_priv *priv = netdev_priv(dev); + int index = 0; + + /* Get LRO statistics */ + data[index++] = priv->lro.lro_mgr.stats.aggregated; + data[index++] = priv->lro.lro_mgr.stats.flushed; + if (priv->lro.lro_mgr.stats.flushed) + data[index++] = priv->lro.lro_mgr.stats.aggregated / + priv->lro.lro_mgr.stats.flushed; + else + data[index++] = 0; + data[index++] = priv->lro.lro_mgr.stats.no_desc; +} + static const struct ethtool_ops ipoib_ethtool_ops = { .get_drvinfo = ipoib_get_drvinfo, .get_tso = ethtool_op_get_tso, .get_coalesce = ipoib_get_coalesce, .set_coalesce = ipoib_set_coalesce, + .get_flags = ethtool_op_get_flags, + .set_flags = ethtool_op_set_flags, + .get_strings = ipoib_get_strings, + .get_sset_count = ipoib_get_sset_count, + .get_ethtool_stats = ipoib_get_ethtool_stats, }; void ipoib_set_ethtool_ops(struct net_device *dev) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c index 8b882bbd1d05..961c585da216 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib_fs.c 1389 2004-12-27 22:56:47Z roland $ */ #include <linux/err.h> diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index f429bce24c20..66cafa20c246 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib_ib.c 1386 2004-12-27 16:23:17Z roland $ */ #include <linux/delay.h> @@ -290,7 +288,10 @@ static void ipoib_ib_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) if (test_bit(IPOIB_FLAG_CSUM, &priv->flags) && likely(wc->csum_ok)) skb->ip_summed = CHECKSUM_UNNECESSARY; - netif_receive_skb(skb); + if (dev->features & NETIF_F_LRO) + lro_receive_skb(&priv->lro.lro_mgr, skb, NULL); + else + netif_receive_skb(skb); repost: if (unlikely(ipoib_ib_post_receive(dev, wr_id))) @@ -442,6 +443,9 @@ poll_more: } if (done < budget) { + if (dev->features & NETIF_F_LRO) + lro_flush_all(&priv->lro.lro_mgr); + netif_rx_complete(dev, napi); if (unlikely(ib_req_notify_cq(priv->recv_cq, IB_CQ_NEXT_COMP | @@ -898,7 +902,8 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port) return 0; } -static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event) +static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, + enum ipoib_flush_level level) { struct ipoib_dev_priv *cpriv; struct net_device *dev = priv->dev; @@ -911,7 +916,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event) * the parent is down. */ list_for_each_entry(cpriv, &priv->child_intfs, list) - __ipoib_ib_dev_flush(cpriv, pkey_event); + __ipoib_ib_dev_flush(cpriv, level); mutex_unlock(&priv->vlan_mutex); @@ -925,7 +930,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event) return; } - if (pkey_event) { + if (level == IPOIB_FLUSH_HEAVY) { if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) { clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); ipoib_ib_dev_down(dev, 0); @@ -943,11 +948,15 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event) priv->pkey_index = new_index; } - ipoib_dbg(priv, "flushing\n"); + if (level == IPOIB_FLUSH_LIGHT) { + ipoib_mark_paths_invalid(dev); + ipoib_mcast_dev_flush(dev); + } - ipoib_ib_dev_down(dev, 0); + if (level >= IPOIB_FLUSH_NORMAL) + ipoib_ib_dev_down(dev, 0); - if (pkey_event) { + if (level == IPOIB_FLUSH_HEAVY) { ipoib_ib_dev_stop(dev, 0); ipoib_ib_dev_open(dev); } @@ -957,27 +966,34 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event) * we get here, don't bring it back up if it's not configured up */ if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) { - ipoib_ib_dev_up(dev); + if (level >= IPOIB_FLUSH_NORMAL) + ipoib_ib_dev_up(dev); ipoib_mcast_restart_task(&priv->restart_task); } } -void ipoib_ib_dev_flush(struct work_struct *work) +void ipoib_ib_dev_flush_light(struct work_struct *work) +{ + struct ipoib_dev_priv *priv = + container_of(work, struct ipoib_dev_priv, flush_light); + + __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_LIGHT); +} + +void ipoib_ib_dev_flush_normal(struct work_struct *work) { struct ipoib_dev_priv *priv = - container_of(work, struct ipoib_dev_priv, flush_task); + container_of(work, struct ipoib_dev_priv, flush_normal); - ipoib_dbg(priv, "Flushing %s\n", priv->dev->name); - __ipoib_ib_dev_flush(priv, 0); + __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_NORMAL); } -void ipoib_pkey_event(struct work_struct *work) +void ipoib_ib_dev_flush_heavy(struct work_struct *work) { struct ipoib_dev_priv *priv = - container_of(work, struct ipoib_dev_priv, pkey_event_task); + container_of(work, struct ipoib_dev_priv, flush_heavy); - ipoib_dbg(priv, "Flushing %s and restarting its QP\n", priv->dev->name); - __ipoib_ib_dev_flush(priv, 1); + __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_HEAVY); } void ipoib_ib_dev_cleanup(struct net_device *dev) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 2442090ac8d1..8be9ea0436e6 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib_main.c 1377 2004-12-23 19:57:12Z roland $ */ #include "ipoib.h" @@ -62,6 +60,15 @@ MODULE_PARM_DESC(send_queue_size, "Number of descriptors in send queue"); module_param_named(recv_queue_size, ipoib_recvq_size, int, 0444); MODULE_PARM_DESC(recv_queue_size, "Number of descriptors in receive queue"); +static int lro; +module_param(lro, bool, 0444); +MODULE_PARM_DESC(lro, "Enable LRO (Large Receive Offload)"); + +static int lro_max_aggr = IPOIB_LRO_MAX_AGGR; +module_param(lro_max_aggr, int, 0644); +MODULE_PARM_DESC(lro_max_aggr, "LRO: Max packets to be aggregated " + "(default = 64)"); + #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG int ipoib_debug_level; @@ -350,6 +357,23 @@ void ipoib_path_iter_read(struct ipoib_path_iter *iter, #endif /* CONFIG_INFINIBAND_IPOIB_DEBUG */ +void ipoib_mark_paths_invalid(struct net_device *dev) +{ + struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_path *path, *tp; + + spin_lock_irq(&priv->lock); + + list_for_each_entry_safe(path, tp, &priv->path_list, list) { + ipoib_dbg(priv, "mark path LID 0x%04x GID " IPOIB_GID_FMT " invalid\n", + be16_to_cpu(path->pathrec.dlid), + IPOIB_GID_ARG(path->pathrec.dgid)); + path->valid = 0; + } + + spin_unlock_irq(&priv->lock); +} + void ipoib_flush_paths(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -386,6 +410,7 @@ static void path_rec_completion(int status, struct net_device *dev = path->dev; struct ipoib_dev_priv *priv = netdev_priv(dev); struct ipoib_ah *ah = NULL; + struct ipoib_ah *old_ah; struct ipoib_neigh *neigh, *tn; struct sk_buff_head skqueue; struct sk_buff *skb; @@ -409,6 +434,7 @@ static void path_rec_completion(int status, spin_lock_irqsave(&priv->lock, flags); + old_ah = path->ah; path->ah = ah; if (ah) { @@ -421,6 +447,17 @@ static void path_rec_completion(int status, __skb_queue_tail(&skqueue, skb); list_for_each_entry_safe(neigh, tn, &path->neigh_list, list) { + if (neigh->ah) { + WARN_ON(neigh->ah != old_ah); + /* + * Dropping the ah reference inside + * priv->lock is safe here, because we + * will hold one more reference from + * the original value of path->ah (ie + * old_ah). + */ + ipoib_put_ah(neigh->ah); + } kref_get(&path->ah->ref); neigh->ah = path->ah; memcpy(&neigh->dgid.raw, &path->pathrec.dgid.raw, @@ -443,6 +480,7 @@ static void path_rec_completion(int status, while ((skb = __skb_dequeue(&neigh->queue))) __skb_queue_tail(&skqueue, skb); } + path->valid = 1; } path->query = NULL; @@ -450,6 +488,9 @@ static void path_rec_completion(int status, spin_unlock_irqrestore(&priv->lock, flags); + if (old_ah) + ipoib_put_ah(old_ah); + while ((skb = __skb_dequeue(&skqueue))) { skb->dev = dev; if (dev_queue_xmit(skb)) @@ -623,8 +664,9 @@ static void unicast_arp_send(struct sk_buff *skb, struct net_device *dev, spin_lock(&priv->lock); path = __path_find(dev, phdr->hwaddr + 4); - if (!path) { - path = path_rec_create(dev, phdr->hwaddr + 4); + if (!path || !path->valid) { + if (!path) + path = path_rec_create(dev, phdr->hwaddr + 4); if (path) { /* put pseudoheader back on for next time */ skb_push(skb, sizeof *phdr); @@ -938,6 +980,54 @@ static const struct header_ops ipoib_header_ops = { .create = ipoib_hard_header, }; +static int get_skb_hdr(struct sk_buff *skb, void **iphdr, + void **tcph, u64 *hdr_flags, void *priv) +{ + unsigned int ip_len; + struct iphdr *iph; + + if (unlikely(skb->protocol != htons(ETH_P_IP))) + return -1; + + /* + * In the future we may add an else clause that verifies the + * checksum and allows devices which do not calculate checksum + * to use LRO. + */ + if (unlikely(skb->ip_summed != CHECKSUM_UNNECESSARY)) + return -1; + + /* Check for non-TCP packet */ + skb_reset_network_header(skb); + iph = ip_hdr(skb); + if (iph->protocol != IPPROTO_TCP) + return -1; + + ip_len = ip_hdrlen(skb); + skb_set_transport_header(skb, ip_len); + *tcph = tcp_hdr(skb); + + /* check if IP header and TCP header are complete */ + if (ntohs(iph->tot_len) < ip_len + tcp_hdrlen(skb)) + return -1; + + *hdr_flags = LRO_IPV4 | LRO_TCP; + *iphdr = iph; + + return 0; +} + +static void ipoib_lro_setup(struct ipoib_dev_priv *priv) +{ + priv->lro.lro_mgr.max_aggr = lro_max_aggr; + priv->lro.lro_mgr.max_desc = IPOIB_MAX_LRO_DESCRIPTORS; + priv->lro.lro_mgr.lro_arr = priv->lro.lro_desc; + priv->lro.lro_mgr.get_skb_header = get_skb_hdr; + priv->lro.lro_mgr.features = LRO_F_NAPI; + priv->lro.lro_mgr.dev = priv->dev; + priv->lro.lro_mgr.ip_summed_aggr = CHECKSUM_UNNECESSARY; +} + static void ipoib_setup(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -977,10 +1067,11 @@ static void ipoib_setup(struct net_device *dev) priv->dev = dev; + ipoib_lro_setup(priv); + spin_lock_init(&priv->lock); spin_lock_init(&priv->tx_lock); - mutex_init(&priv->mcast_mutex); mutex_init(&priv->vlan_mutex); INIT_LIST_HEAD(&priv->path_list); @@ -989,9 +1080,10 @@ static void ipoib_setup(struct net_device *dev) INIT_LIST_HEAD(&priv->multicast_list); INIT_DELAYED_WORK(&priv->pkey_poll_task, ipoib_pkey_poll); - INIT_WORK(&priv->pkey_event_task, ipoib_pkey_event); INIT_DELAYED_WORK(&priv->mcast_task, ipoib_mcast_join_task); - INIT_WORK(&priv->flush_task, ipoib_ib_dev_flush); + INIT_WORK(&priv->flush_light, ipoib_ib_dev_flush_light); + INIT_WORK(&priv->flush_normal, ipoib_ib_dev_flush_normal); + INIT_WORK(&priv->flush_heavy, ipoib_ib_dev_flush_heavy); INIT_WORK(&priv->restart_task, ipoib_mcast_restart_task); INIT_DELAYED_WORK(&priv->ah_reap_task, ipoib_reap_ah); } @@ -1154,6 +1246,9 @@ static struct net_device *ipoib_add_port(const char *format, priv->dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM; } + if (lro) + priv->dev->features |= NETIF_F_LRO; + /* * Set the full membership bit, so that we join the right * broadcast group, etc. @@ -1304,6 +1399,12 @@ static int __init ipoib_init_module(void) ipoib_max_conn_qp = min(ipoib_max_conn_qp, IPOIB_CM_MAX_CONN_QP); #endif + /* + * When copying small received packets, we only copy from the + * linear data part of the SKB, so we rely on this condition. + */ + BUILD_BUG_ON(IPOIB_CM_COPYBREAK > IPOIB_CM_HEAD_SIZE); + ret = ipoib_register_debugfs(); if (ret) return ret; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 3f663fb852c1..1fcc9a898d81 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib_multicast.c 1362 2004-12-18 15:56:29Z roland $ */ #include <linux/skbuff.h> @@ -188,6 +186,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, struct ipoib_dev_priv *priv = netdev_priv(dev); struct ipoib_ah *ah; int ret; + int set_qkey = 0; mcast->mcmember = *mcmember; @@ -202,6 +201,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, priv->qkey = be32_to_cpu(priv->broadcast->mcmember.qkey); spin_unlock_irq(&priv->lock); priv->tx_wr.wr.ud.remote_qkey = priv->qkey; + set_qkey = 1; } if (!test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) { @@ -214,7 +214,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, } ret = ipoib_mcast_attach(dev, be16_to_cpu(mcast->mcmember.mlid), - &mcast->mcmember.mgid); + &mcast->mcmember.mgid, set_qkey); if (ret < 0) { ipoib_warn(priv, "couldn't attach QP to multicast group " IPOIB_GID_FMT "\n", @@ -575,8 +575,11 @@ void ipoib_mcast_join_task(struct work_struct *work) priv->mcast_mtu = IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu)); - if (!ipoib_cm_admin_enabled(dev)) - dev->mtu = min(priv->mcast_mtu, priv->admin_mtu); + if (!ipoib_cm_admin_enabled(dev)) { + rtnl_lock(); + dev_set_mtu(dev, min(priv->mcast_mtu, priv->admin_mtu)); + rtnl_unlock(); + } ipoib_dbg_mcast(priv, "successfully joined all multicast groups\n"); @@ -594,10 +597,6 @@ int ipoib_mcast_start_thread(struct net_device *dev) queue_delayed_work(ipoib_workqueue, &priv->mcast_task, 0); mutex_unlock(&mcast_mutex); - spin_lock_irq(&priv->lock); - set_bit(IPOIB_MCAST_STARTED, &priv->flags); - spin_unlock_irq(&priv->lock); - return 0; } @@ -607,10 +606,6 @@ int ipoib_mcast_stop_thread(struct net_device *dev, int flush) ipoib_dbg_mcast(priv, "stopping multicast thread\n"); - spin_lock_irq(&priv->lock); - clear_bit(IPOIB_MCAST_STARTED, &priv->flags); - spin_unlock_irq(&priv->lock); - mutex_lock(&mcast_mutex); clear_bit(IPOIB_MCAST_RUN, &priv->flags); cancel_delayed_work(&priv->mcast_task); @@ -635,10 +630,10 @@ static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) IPOIB_GID_ARG(mcast->mcmember.mgid)); /* Remove ourselves from the multicast group */ - ret = ipoib_mcast_detach(dev, be16_to_cpu(mcast->mcmember.mlid), - &mcast->mcmember.mgid); + ret = ib_detach_mcast(priv->qp, &mcast->mcmember.mgid, + be16_to_cpu(mcast->mcmember.mlid)); if (ret) - ipoib_warn(priv, "ipoib_mcast_detach failed (result = %d)\n", ret); + ipoib_warn(priv, "ib_detach_mcast failed (result = %d)\n", ret); } return 0; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c index 8766d29ce3b7..68325119f740 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c @@ -29,24 +29,17 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib_verbs.c 1349 2004-12-16 21:09:43Z roland $ */ #include "ipoib.h" -int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid) +int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid, int set_qkey) { struct ipoib_dev_priv *priv = netdev_priv(dev); - struct ib_qp_attr *qp_attr; + struct ib_qp_attr *qp_attr = NULL; int ret; u16 pkey_index; - ret = -ENOMEM; - qp_attr = kmalloc(sizeof *qp_attr, GFP_KERNEL); - if (!qp_attr) - goto out; - if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &pkey_index)) { clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); ret = -ENXIO; @@ -54,18 +47,23 @@ int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid) } set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags); - /* set correct QKey for QP */ - qp_attr->qkey = priv->qkey; - ret = ib_modify_qp(priv->qp, qp_attr, IB_QP_QKEY); - if (ret) { - ipoib_warn(priv, "failed to modify QP, ret = %d\n", ret); - goto out; + if (set_qkey) { + ret = -ENOMEM; + qp_attr = kmalloc(sizeof *qp_attr, GFP_KERNEL); + if (!qp_attr) + goto out; + + /* set correct QKey for QP */ + qp_attr->qkey = priv->qkey; + ret = ib_modify_qp(priv->qp, qp_attr, IB_QP_QKEY); + if (ret) { + ipoib_warn(priv, "failed to modify QP, ret = %d\n", ret); + goto out; + } } /* attach QP to multicast group */ - mutex_lock(&priv->mcast_mutex); ret = ib_attach_mcast(priv->qp, mgid, mlid); - mutex_unlock(&priv->mcast_mutex); if (ret) ipoib_warn(priv, "failed to attach to multicast group, ret = %d\n", ret); @@ -74,20 +72,6 @@ out: return ret; } -int ipoib_mcast_detach(struct net_device *dev, u16 mlid, union ib_gid *mgid) -{ - struct ipoib_dev_priv *priv = netdev_priv(dev); - int ret; - - mutex_lock(&priv->mcast_mutex); - ret = ib_detach_mcast(priv->qp, mgid, mlid); - mutex_unlock(&priv->mcast_mutex); - if (ret) - ipoib_warn(priv, "ib_detach_mcast failed (result = %d)\n", ret); - - return ret; -} - int ipoib_init_qp(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); @@ -201,7 +185,10 @@ int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca) init_attr.recv_cq = priv->recv_cq; if (priv->hca_caps & IB_DEVICE_UD_TSO) - init_attr.create_flags = IB_QP_CREATE_IPOIB_UD_LSO; + init_attr.create_flags |= IB_QP_CREATE_IPOIB_UD_LSO; + + if (priv->hca_caps & IB_DEVICE_BLOCK_MULTICAST_LOOPBACK) + init_attr.create_flags |= IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK; if (dev->features & NETIF_F_SG) init_attr.cap.max_send_sge = MAX_SKB_FRAGS + 1; @@ -289,15 +276,17 @@ void ipoib_event(struct ib_event_handler *handler, if (record->element.port_num != priv->port) return; - if (record->event == IB_EVENT_PORT_ERR || - record->event == IB_EVENT_PORT_ACTIVE || - record->event == IB_EVENT_LID_CHANGE || - record->event == IB_EVENT_SM_CHANGE || + ipoib_dbg(priv, "Event %d on device %s port %d\n", record->event, + record->device->name, record->element.port_num); + + if (record->event == IB_EVENT_SM_CHANGE || record->event == IB_EVENT_CLIENT_REREGISTER) { - ipoib_dbg(priv, "Port state change event\n"); - queue_work(ipoib_workqueue, &priv->flush_task); + queue_work(ipoib_workqueue, &priv->flush_light); + } else if (record->event == IB_EVENT_PORT_ERR || + record->event == IB_EVENT_PORT_ACTIVE || + record->event == IB_EVENT_LID_CHANGE) { + queue_work(ipoib_workqueue, &priv->flush_normal); } else if (record->event == IB_EVENT_PKEY_CHANGE) { - ipoib_dbg(priv, "P_Key change event on port:%d\n", priv->port); - queue_work(ipoib_workqueue, &priv->pkey_event_task); + queue_work(ipoib_workqueue, &priv->flush_heavy); } } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 1cdb5cfb0ff1..b08eb56196d3 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ipoib_vlan.c 1349 2004-12-16 21:09:43Z roland $ */ #include <linux/module.h> diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index aeb58cae9a3f..356fac6d105a 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -42,9 +42,6 @@ * Zhenyu Wang * Modified by: * Erez Zilber - * - * - * $Id: iscsi_iser.c 6965 2006-05-07 11:36:20Z ogerlitz $ */ #include <linux/types.h> diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h index a8c1b300e34d..0e10703cf59e 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.h +++ b/drivers/infiniband/ulp/iser/iscsi_iser.h @@ -36,8 +36,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: iscsi_iser.h 7051 2006-05-10 12:29:11Z ogerlitz $ */ #ifndef __ISCSI_ISER_H__ #define __ISCSI_ISER_H__ diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c index 08dc81c46f41..31ad498bdc51 100644 --- a/drivers/infiniband/ulp/iser/iser_initiator.c +++ b/drivers/infiniband/ulp/iser/iser_initiator.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: iser_initiator.c 6964 2006-05-07 11:11:43Z ogerlitz $ */ #include <linux/kernel.h> #include <linux/slab.h> diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index cac50c4dc159..81e49cb10ed3 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: iser_memory.c 6964 2006-05-07 11:11:43Z ogerlitz $ */ #include <linux/module.h> #include <linux/kernel.h> diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index d19cfe605ebb..77cabee7cc08 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: iser_verbs.c 7051 2006-05-10 12:29:11Z ogerlitz $ */ #include <linux/kernel.h> #include <linux/module.h> diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index 435145709dd6..ed7c5f72cb8b 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_srp.c 3932 2005-11-01 17:19:29Z roland $ */ #include <linux/module.h> @@ -49,8 +47,6 @@ #include <scsi/srp.h> #include <scsi/scsi_transport_srp.h> -#include <rdma/ib_cache.h> - #include "ib_srp.h" #define DRV_NAME "ib_srp" @@ -183,10 +179,10 @@ static int srp_init_qp(struct srp_target_port *target, if (!attr) return -ENOMEM; - ret = ib_find_cached_pkey(target->srp_host->srp_dev->dev, - target->srp_host->port, - be16_to_cpu(target->path.pkey), - &attr->pkey_index); + ret = ib_find_pkey(target->srp_host->srp_dev->dev, + target->srp_host->port, + be16_to_cpu(target->path.pkey), + &attr->pkey_index); if (ret) goto out; @@ -1883,8 +1879,7 @@ static ssize_t srp_create_target(struct device *dev, if (ret) goto err; - ib_get_cached_gid(host->srp_dev->dev, host->port, 0, - &target->path.sgid); + ib_query_gid(host->srp_dev->dev, host->port, 0, &target->path.sgid); shost_printk(KERN_DEBUG, target->scsi_host, PFX "new target: id_ext %016llx ioc_guid %016llx pkey %04x " diff --git a/drivers/infiniband/ulp/srp/ib_srp.h b/drivers/infiniband/ulp/srp/ib_srp.h index 63d2ae724061..e185b907fc12 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.h +++ b/drivers/infiniband/ulp/srp/ib_srp.h @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_srp.h 3932 2005-11-01 17:19:29Z roland $ */ #ifndef IB_SRP_H diff --git a/drivers/net/cxgb3/cxgb3_ctl_defs.h b/drivers/net/cxgb3/cxgb3_ctl_defs.h index 6c4f32066919..ed0ecd9679cb 100644 --- a/drivers/net/cxgb3/cxgb3_ctl_defs.h +++ b/drivers/net/cxgb3/cxgb3_ctl_defs.h @@ -54,6 +54,7 @@ enum { RDMA_CQ_DISABLE = 16, RDMA_CTRL_QP_SETUP = 17, RDMA_GET_MEM = 18, + RDMA_GET_MIB = 19, GET_RX_PAGE_INFO = 50, }; diff --git a/drivers/net/cxgb3/cxgb3_offload.c b/drivers/net/cxgb3/cxgb3_offload.c index ff9c013ce535..cf2696873796 100644 --- a/drivers/net/cxgb3/cxgb3_offload.c +++ b/drivers/net/cxgb3/cxgb3_offload.c @@ -303,6 +303,12 @@ static int cxgb_rdma_ctl(struct adapter *adapter, unsigned int req, void *data) spin_unlock_irq(&adapter->sge.reg_lock); break; } + case RDMA_GET_MIB: { + spin_lock(&adapter->stats_lock); + t3_tp_get_mib_stats(adapter, (struct tp_mib_stats *)data); + spin_unlock(&adapter->stats_lock); + break; + } default: ret = -EOPNOTSUPP; } @@ -381,6 +387,7 @@ static int cxgb_offload_ctl(struct t3cdev *tdev, unsigned int req, void *data) case RDMA_CQ_DISABLE: case RDMA_CTRL_QP_SETUP: case RDMA_GET_MEM: + case RDMA_GET_MIB: if (!offload_running(adapter)) return -EAGAIN; return cxgb_rdma_ctl(adapter, req, data); diff --git a/drivers/net/cxgb3/version.h b/drivers/net/cxgb3/version.h index a0177fc55e28..29db711303b9 100644 --- a/drivers/net/cxgb3/version.h +++ b/drivers/net/cxgb3/version.h @@ -38,7 +38,7 @@ #define DRV_VERSION "1.0-ko" /* Firmware version */ -#define FW_VERSION_MAJOR 6 +#define FW_VERSION_MAJOR 7 #define FW_VERSION_MINOR 0 #define FW_VERSION_MICRO 0 #endif /* __CHELSIO_VERSION_H */ diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c index d82f2751d2c7..2b5006b9be67 100644 --- a/drivers/net/mlx4/fw.c +++ b/drivers/net/mlx4/fw.c @@ -101,6 +101,34 @@ static void dump_dev_cap_flags(struct mlx4_dev *dev, u32 flags) mlx4_dbg(dev, " %s\n", fname[i]); } +int mlx4_MOD_STAT_CFG(struct mlx4_dev *dev, struct mlx4_mod_stat_cfg *cfg) +{ + struct mlx4_cmd_mailbox *mailbox; + u32 *inbox; + int err = 0; + +#define MOD_STAT_CFG_IN_SIZE 0x100 + +#define MOD_STAT_CFG_PG_SZ_M_OFFSET 0x002 +#define MOD_STAT_CFG_PG_SZ_OFFSET 0x003 + + mailbox = mlx4_alloc_cmd_mailbox(dev); + if (IS_ERR(mailbox)) + return PTR_ERR(mailbox); + inbox = mailbox->buf; + + memset(inbox, 0, MOD_STAT_CFG_IN_SIZE); + + MLX4_PUT(inbox, cfg->log_pg_sz, MOD_STAT_CFG_PG_SZ_OFFSET); + MLX4_PUT(inbox, cfg->log_pg_sz_m, MOD_STAT_CFG_PG_SZ_M_OFFSET); + + err = mlx4_cmd(dev, mailbox->dma, 0, 0, MLX4_CMD_MOD_STAT_CFG, + MLX4_CMD_TIME_CLASS_A); + + mlx4_free_cmd_mailbox(dev, mailbox); + return err; +} + int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) { struct mlx4_cmd_mailbox *mailbox; diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h index 306cb9b0242d..a0e046c149b7 100644 --- a/drivers/net/mlx4/fw.h +++ b/drivers/net/mlx4/fw.h @@ -38,6 +38,11 @@ #include "mlx4.h" #include "icm.h" +struct mlx4_mod_stat_cfg { + u8 log_pg_sz; + u8 log_pg_sz_m; +}; + struct mlx4_dev_cap { int max_srq_sz; int max_qp_sz; @@ -162,5 +167,6 @@ int mlx4_SET_ICM_SIZE(struct mlx4_dev *dev, u64 icm_size, u64 *aux_pages); int mlx4_MAP_ICM_AUX(struct mlx4_dev *dev, struct mlx4_icm *icm); int mlx4_UNMAP_ICM_AUX(struct mlx4_dev *dev); int mlx4_NOP(struct mlx4_dev *dev); +int mlx4_MOD_STAT_CFG(struct mlx4_dev *dev, struct mlx4_mod_stat_cfg *cfg); #endif /* MLX4_FW_H */ diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index a6aa49fc1d68..d3736013fe9b 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -485,6 +485,7 @@ static int mlx4_init_hca(struct mlx4_dev *dev) struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_adapter adapter; struct mlx4_dev_cap dev_cap; + struct mlx4_mod_stat_cfg mlx4_cfg; struct mlx4_profile profile; struct mlx4_init_hca_param init_hca; u64 icm_size; @@ -502,6 +503,12 @@ static int mlx4_init_hca(struct mlx4_dev *dev) return err; } + mlx4_cfg.log_pg_sz_m = 1; + mlx4_cfg.log_pg_sz = 0; + err = mlx4_MOD_STAT_CFG(dev, &mlx4_cfg); + if (err) + mlx4_warn(dev, "Failed to override log_pg_sz parameter\n"); + err = mlx4_dev_cap(dev, &dev_cap); if (err) { mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n"); diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c index 57f7f1f0d4ec..b4b57870ddfd 100644 --- a/drivers/net/mlx4/mcg.c +++ b/drivers/net/mlx4/mcg.c @@ -38,6 +38,9 @@ #include "mlx4.h" +#define MGM_QPN_MASK 0x00FFFFFF +#define MGM_BLCK_LB_BIT 30 + struct mlx4_mgm { __be32 next_gid_index; __be32 members_count; @@ -153,7 +156,8 @@ static int find_mgm(struct mlx4_dev *dev, return err; } -int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) +int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], + int block_mcast_loopback) { struct mlx4_priv *priv = mlx4_priv(dev); struct mlx4_cmd_mailbox *mailbox; @@ -202,13 +206,18 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) } for (i = 0; i < members_count; ++i) - if (mgm->qp[i] == cpu_to_be32(qp->qpn)) { + if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn) { mlx4_dbg(dev, "QP %06x already a member of MGM\n", qp->qpn); err = 0; goto out; } - mgm->qp[members_count++] = cpu_to_be32(qp->qpn); + if (block_mcast_loopback) + mgm->qp[members_count++] = cpu_to_be32((qp->qpn & MGM_QPN_MASK) | + (1 << MGM_BLCK_LB_BIT)); + else + mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK); + mgm->members_count = cpu_to_be32(members_count); err = mlx4_WRITE_MCG(dev, index, mailbox); @@ -283,7 +292,7 @@ int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]) members_count = be32_to_cpu(mgm->members_count); for (loc = -1, i = 0; i < members_count; ++i) - if (mgm->qp[i] == cpu_to_be32(qp->qpn)) + if ((be32_to_cpu(mgm->qp[i]) & MGM_QPN_MASK) == qp->qpn) loc = i; if (loc == -1) { diff --git a/firmware/Makefile b/firmware/Makefile index 809a52624bda..e4f2fb3d1917 100644 --- a/firmware/Makefile +++ b/firmware/Makefile @@ -34,8 +34,6 @@ fw-shipped-$(CONFIG_SND_SB16_CSP) += sb16/mulaw_main.csp sb16/alaw_main.csp \ sb16/ima_adpcm_capture.csp fw-shipped-$(CONFIG_SND_YMFPCI) += yamaha/ds1_ctrl.fw yamaha/ds1_dsp.fw \ yamaha/ds1e_ctrl.fw -fw-shipped-$(CONFIG_TIGON3) += tigon/tg3.bin tigon/tg3_tso.bin \ - tigon/tg3_tso5.bin fw-shipped-$(CONFIG_USB_DABUSB) += dabusb/firmware.fw dabusb/bitstream.bin fw-shipped-$(CONFIG_USB_EMI26) += emi26/loader.fw emi26/firmware.fw \ emi26/bitstream.fw diff --git a/fs/buffer.c b/fs/buffer.c index 0f51c0f7c266..5fa1512cd9a2 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1691,11 +1691,13 @@ static int __block_write_full_page(struct inode *inode, struct page *page, */ clear_buffer_dirty(bh); set_buffer_uptodate(bh); - } else if (!buffer_mapped(bh) && buffer_dirty(bh)) { + } else if ((!buffer_mapped(bh) || buffer_delay(bh)) && + buffer_dirty(bh)) { WARN_ON(bh->b_size != blocksize); err = get_block(inode, block, bh, 1); if (err) goto recover; + clear_buffer_delay(bh); if (buffer_new(bh)) { /* blockdev mappings never come here */ clear_buffer_new(bh); @@ -1774,7 +1776,8 @@ recover: bh = head; /* Recovery: lock and submit the mapped buffers */ do { - if (buffer_mapped(bh) && buffer_dirty(bh)) { + if (buffer_mapped(bh) && buffer_dirty(bh) && + !buffer_delay(bh)) { lock_buffer(bh); mark_buffer_async_write(bh); } else { @@ -2061,6 +2064,7 @@ int generic_write_end(struct file *file, struct address_space *mapping, struct page *page, void *fsdata) { struct inode *inode = mapping->host; + int i_size_changed = 0; copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -2073,12 +2077,21 @@ int generic_write_end(struct file *file, struct address_space *mapping, */ if (pos+copied > inode->i_size) { i_size_write(inode, pos+copied); - mark_inode_dirty(inode); + i_size_changed = 1; } unlock_page(page); page_cache_release(page); + /* + * Don't mark the inode dirty under page lock. First, it unnecessarily + * makes the holding time of page lock longer. Second, it forces lock + * ordering of page lock and transaction start for journaling + * filesystems. + */ + if (i_size_changed) + mark_inode_dirty(inode); + return copied; } EXPORT_SYMBOL(generic_write_end); diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 9cc80b9cc8d8..495ab21b9832 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -47,7 +47,7 @@ static int ext4_block_in_group(struct super_block *sb, ext4_fsblk_t block, ext4_group_t block_group) { ext4_group_t actual_group; - ext4_get_group_no_and_offset(sb, block, &actual_group, 0); + ext4_get_group_no_and_offset(sb, block, &actual_group, NULL); if (actual_group == block_group) return 1; return 0; @@ -121,12 +121,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks); } } else { /* For META_BG_BLOCK_GROUPS */ - int group_rel = (block_group - - le32_to_cpu(sbi->s_es->s_first_meta_bg)) % - EXT4_DESC_PER_BLOCK(sb); - if (group_rel == 0 || group_rel == 1 || - (group_rel == EXT4_DESC_PER_BLOCK(sb) - 1)) - bit_max += 1; + bit_max += ext4_bg_num_gdb(sb, block_group); } if (block_group == sbi->s_groups_count - 1) { @@ -295,7 +290,7 @@ err_out: return 0; } /** - * read_block_bitmap() + * ext4_read_block_bitmap() * @sb: super block * @block_group: given block group * @@ -305,7 +300,7 @@ err_out: * Return buffer_head on success or NULL in case of failure. */ struct buffer_head * -read_block_bitmap(struct super_block *sb, ext4_group_t block_group) +ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group) { struct ext4_group_desc * desc; struct buffer_head * bh = NULL; @@ -409,8 +404,7 @@ restart: prev = rsv; } printk("Window map complete.\n"); - if (bad) - BUG(); + BUG_ON(bad); } #define rsv_window_dump(root, verbose) \ __rsv_window_dump((root), (verbose), __func__) @@ -694,7 +688,7 @@ do_more: count -= overflow; } brelse(bitmap_bh); - bitmap_bh = read_block_bitmap(sb, block_group); + bitmap_bh = ext4_read_block_bitmap(sb, block_group); if (!bitmap_bh) goto error_return; desc = ext4_get_group_desc (sb, block_group, &gd_bh); @@ -810,6 +804,13 @@ do_more: spin_unlock(sb_bgl_lock(sbi, block_group)); percpu_counter_add(&sbi->s_freeblocks_counter, count); + if (sbi->s_log_groups_per_flex) { + ext4_group_t flex_group = ext4_flex_group(sbi, block_group); + spin_lock(sb_bgl_lock(sbi, flex_group)); + sbi->s_flex_groups[flex_group].free_blocks += count; + spin_unlock(sb_bgl_lock(sbi, flex_group)); + } + /* We dirtied the bitmap block */ BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); err = ext4_journal_dirty_metadata(handle, bitmap_bh); @@ -1598,23 +1599,35 @@ out: /** * ext4_has_free_blocks() - * @sbi: in-core super block structure. + * @sbi: in-core super block structure. + * @nblocks: number of neeed blocks * - * Check if filesystem has at least 1 free block available for allocation. + * Check if filesystem has free blocks available for allocation. + * Return the number of blocks avaible for allocation for this request + * On success, return nblocks */ -static int ext4_has_free_blocks(struct ext4_sb_info *sbi) +ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi, + ext4_fsblk_t nblocks) { - ext4_fsblk_t free_blocks, root_blocks; + ext4_fsblk_t free_blocks; + ext4_fsblk_t root_blocks = 0; free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); - root_blocks = ext4_r_blocks_count(sbi->s_es); - if (free_blocks < root_blocks + 1 && !capable(CAP_SYS_RESOURCE) && + + if (!capable(CAP_SYS_RESOURCE) && sbi->s_resuid != current->fsuid && - (sbi->s_resgid == 0 || !in_group_p (sbi->s_resgid))) { - return 0; - } - return 1; -} + (sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid))) + root_blocks = ext4_r_blocks_count(sbi->s_es); +#ifdef CONFIG_SMP + if (free_blocks - root_blocks < FBC_BATCH) + free_blocks = + percpu_counter_sum_and_set(&sbi->s_freeblocks_counter); +#endif + if (free_blocks - root_blocks < nblocks) + return free_blocks - root_blocks; + return nblocks; + } + /** * ext4_should_retry_alloc() @@ -1630,7 +1643,7 @@ static int ext4_has_free_blocks(struct ext4_sb_info *sbi) */ int ext4_should_retry_alloc(struct super_block *sb, int *retries) { - if (!ext4_has_free_blocks(EXT4_SB(sb)) || (*retries)++ > 3) + if (!ext4_has_free_blocks(EXT4_SB(sb), 1) || (*retries)++ > 3) return 0; jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); @@ -1639,20 +1652,24 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries) } /** - * ext4_new_blocks_old() -- core block(s) allocation function + * ext4_old_new_blocks() -- core block bitmap based block allocation function + * * @handle: handle to this transaction * @inode: file inode * @goal: given target block(filesystem wide) * @count: target number of blocks to allocate * @errp: error code * - * ext4_new_blocks uses a goal block to assist allocation. It tries to - * allocate block(s) from the block group contains the goal block first. If that - * fails, it will try to allocate block(s) from other block groups without - * any specific goal block. + * ext4_old_new_blocks uses a goal block to assist allocation and look up + * the block bitmap directly to do block allocation. It tries to + * allocate block(s) from the block group contains the goal block first. If + * that fails, it will try to allocate block(s) from other block groups + * without any specific goal block. + * + * This function is called when -o nomballoc mount option is enabled * */ -ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, +ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp) { struct buffer_head *bitmap_bh = NULL; @@ -1676,13 +1693,26 @@ ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, ext4_group_t ngroups; unsigned long num = *count; - *errp = -ENOSPC; sb = inode->i_sb; if (!sb) { + *errp = -ENODEV; printk("ext4_new_block: nonexistent device"); return 0; } + sbi = EXT4_SB(sb); + if (!EXT4_I(inode)->i_delalloc_reserved_flag) { + /* + * With delalloc we already reserved the blocks + */ + *count = ext4_has_free_blocks(sbi, *count); + } + if (*count == 0) { + *errp = -ENOSPC; + return 0; /*return with ENOSPC error */ + } + num = *count; + /* * Check quota for allocation of this block. */ @@ -1706,11 +1736,6 @@ ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0)) my_rsv = &block_i->rsv_window_node; - if (!ext4_has_free_blocks(sbi)) { - *errp = -ENOSPC; - goto out; - } - /* * First, test whether the goal block is free. */ @@ -1734,7 +1759,7 @@ retry_alloc: my_rsv = NULL; if (free_blocks > 0) { - bitmap_bh = read_block_bitmap(sb, group_no); + bitmap_bh = ext4_read_block_bitmap(sb, group_no); if (!bitmap_bh) goto io_error; grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, @@ -1770,7 +1795,7 @@ retry_alloc: continue; brelse(bitmap_bh); - bitmap_bh = read_block_bitmap(sb, group_no); + bitmap_bh = ext4_read_block_bitmap(sb, group_no); if (!bitmap_bh) goto io_error; /* @@ -1882,7 +1907,15 @@ allocated: le16_add_cpu(&gdp->bg_free_blocks_count, -num); gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp); spin_unlock(sb_bgl_lock(sbi, group_no)); - percpu_counter_sub(&sbi->s_freeblocks_counter, num); + if (!EXT4_I(inode)->i_delalloc_reserved_flag) + percpu_counter_sub(&sbi->s_freeblocks_counter, num); + + if (sbi->s_log_groups_per_flex) { + ext4_group_t flex_group = ext4_flex_group(sbi, group_no); + spin_lock(sb_bgl_lock(sbi, flex_group)); + sbi->s_flex_groups[flex_group].free_blocks -= num; + spin_unlock(sb_bgl_lock(sbi, flex_group)); + } BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor"); err = ext4_journal_dirty_metadata(handle, gdp_bh); @@ -1915,46 +1948,104 @@ out: return 0; } -ext4_fsblk_t ext4_new_block(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, int *errp) +#define EXT4_META_BLOCK 0x1 + +static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode, + ext4_lblk_t iblock, ext4_fsblk_t goal, + unsigned long *count, int *errp, int flags) { struct ext4_allocation_request ar; ext4_fsblk_t ret; if (!test_opt(inode->i_sb, MBALLOC)) { - unsigned long count = 1; - ret = ext4_new_blocks_old(handle, inode, goal, &count, errp); - return ret; + return ext4_old_new_blocks(handle, inode, goal, count, errp); } memset(&ar, 0, sizeof(ar)); + /* Fill with neighbour allocated blocks */ + ar.inode = inode; ar.goal = goal; - ar.len = 1; + ar.len = *count; + ar.logical = iblock; + + if (S_ISREG(inode->i_mode) && !(flags & EXT4_META_BLOCK)) + /* enable in-core preallocation for data block allocation */ + ar.flags = EXT4_MB_HINT_DATA; + else + /* disable in-core preallocation for non-regular files */ + ar.flags = 0; + ret = ext4_mb_new_blocks(handle, &ar, errp); + *count = ar.len; return ret; } -ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, +/* + * ext4_new_meta_blocks() -- allocate block for meta data (indexing) blocks + * + * @handle: handle to this transaction + * @inode: file inode + * @goal: given target block(filesystem wide) + * @count: total number of blocks need + * @errp: error code + * + * Return 1st allocated block numberon success, *count stores total account + * error stores in errp pointer + */ +ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp) { - struct ext4_allocation_request ar; ext4_fsblk_t ret; - - if (!test_opt(inode->i_sb, MBALLOC)) { - ret = ext4_new_blocks_old(handle, inode, goal, count, errp); - return ret; + ret = do_blk_alloc(handle, inode, 0, goal, + count, errp, EXT4_META_BLOCK); + /* + * Account for the allocated meta blocks + */ + if (!(*errp)) { + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + EXT4_I(inode)->i_allocated_meta_blocks += *count; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); } - - memset(&ar, 0, sizeof(ar)); - ar.inode = inode; - ar.goal = goal; - ar.len = *count; - ret = ext4_mb_new_blocks(handle, &ar, errp); - *count = ar.len; return ret; } +/* + * ext4_new_meta_block() -- allocate block for meta data (indexing) blocks + * + * @handle: handle to this transaction + * @inode: file inode + * @goal: given target block(filesystem wide) + * @errp: error code + * + * Return allocated block number on success + */ +ext4_fsblk_t ext4_new_meta_block(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int *errp) +{ + unsigned long count = 1; + return ext4_new_meta_blocks(handle, inode, goal, &count, errp); +} + +/* + * ext4_new_blocks() -- allocate data blocks + * + * @handle: handle to this transaction + * @inode: file inode + * @goal: given target block(filesystem wide) + * @count: total number of blocks need + * @errp: error code + * + * Return 1st allocated block numberon success, *count stores total account + * error stores in errp pointer + */ + +ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, + ext4_lblk_t iblock, ext4_fsblk_t goal, + unsigned long *count, int *errp) +{ + return do_blk_alloc(handle, inode, iblock, goal, count, errp, 0); +} /** * ext4_count_free_blocks() -- count filesystem free blocks @@ -1986,7 +2077,7 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) continue; desc_count += le16_to_cpu(gdp->bg_free_blocks_count); brelse(bitmap_bh); - bitmap_bh = read_block_bitmap(sb, i); + bitmap_bh = ext4_read_block_bitmap(sb, i); if (bitmap_bh == NULL) continue; diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 2bf0331ea194..d3d23d73c08b 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -129,7 +129,8 @@ static int ext4_readdir(struct file * filp, struct buffer_head *bh = NULL; map_bh.b_state = 0; - err = ext4_get_blocks_wrap(NULL, inode, blk, 1, &map_bh, 0, 0); + err = ext4_get_blocks_wrap(NULL, inode, blk, 1, &map_bh, + 0, 0, 0); if (err > 0) { pgoff_t index = map_bh.b_blocknr >> (PAGE_CACHE_SHIFT - inode->i_blkbits); @@ -272,7 +273,7 @@ static void free_rb_tree_fname(struct rb_root *root) while (n) { /* Do the node's children first */ - if ((n)->rb_left) { + if (n->rb_left) { n = n->rb_left; continue; } @@ -301,24 +302,18 @@ static void free_rb_tree_fname(struct rb_root *root) parent->rb_right = NULL; n = parent; } - root->rb_node = NULL; } -static struct dir_private_info *create_dir_info(loff_t pos) +static struct dir_private_info *ext4_htree_create_dir_info(loff_t pos) { struct dir_private_info *p; - p = kmalloc(sizeof(struct dir_private_info), GFP_KERNEL); + p = kzalloc(sizeof(struct dir_private_info), GFP_KERNEL); if (!p) return NULL; - p->root.rb_node = NULL; - p->curr_node = NULL; - p->extra_fname = NULL; - p->last_pos = 0; p->curr_hash = pos2maj_hash(pos); p->curr_minor_hash = pos2min_hash(pos); - p->next_hash = 0; return p; } @@ -433,7 +428,7 @@ static int ext4_dx_readdir(struct file * filp, int ret; if (!info) { - info = create_dir_info(filp->f_pos); + info = ext4_htree_create_dir_info(filp->f_pos); if (!info) return -ENOMEM; filp->private_data = info; diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 8158083f7ac0..303e41cf7b14 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -22,7 +22,7 @@ #include "ext4_i.h" /* - * The second extended filesystem constants/structures + * The fourth extended filesystem constants/structures */ /* @@ -45,7 +45,7 @@ #define ext4_debug(f, a...) \ do { \ printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:", \ - __FILE__, __LINE__, __FUNCTION__); \ + __FILE__, __LINE__, __func__); \ printk (KERN_DEBUG f, ## a); \ } while (0) #else @@ -74,6 +74,9 @@ #define EXT4_MB_HINT_GOAL_ONLY 256 /* goal is meaningful */ #define EXT4_MB_HINT_TRY_GOAL 512 +/* blocks already pre-reserved by delayed allocation */ +#define EXT4_MB_DELALLOC_RESERVED 1024 + struct ext4_allocation_request { /* target inode for block we're allocating */ @@ -170,6 +173,15 @@ struct ext4_group_desc __u32 bg_reserved2[3]; }; +/* + * Structure of a flex block group info + */ + +struct flex_groups { + __u32 free_inodes; + __u32 free_blocks; +}; + #define EXT4_BG_INODE_UNINIT 0x0001 /* Inode table/bitmap not in use */ #define EXT4_BG_BLOCK_UNINIT 0x0002 /* Block bitmap not in use */ #define EXT4_BG_INODE_ZEROED 0x0004 /* On-disk itable initialized to zero */ @@ -527,6 +539,7 @@ do { \ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ #define EXT4_MOUNT_MBALLOC 0x4000000 /* Buddy allocation support */ +#define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */ /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt @@ -647,7 +660,10 @@ struct ext4_super_block { __le16 s_mmp_interval; /* # seconds to wait in MMP checking */ __le64 s_mmp_block; /* Block for multi-mount protection */ __le32 s_raid_stripe_width; /* blocks on all data disks (N*stride)*/ - __u32 s_reserved[163]; /* Padding to the end of the block */ + __u8 s_log_groups_per_flex; /* FLEX_BG group size */ + __u8 s_reserved_char_pad2; + __le16 s_reserved_pad; + __u32 s_reserved[162]; /* Padding to the end of the block */ }; #ifdef __KERNEL__ @@ -958,12 +974,17 @@ extern ext4_grpblk_t ext4_block_group_offset(struct super_block *sb, extern int ext4_bg_has_super(struct super_block *sb, ext4_group_t group); extern unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group); -extern ext4_fsblk_t ext4_new_block (handle_t *handle, struct inode *inode, +extern ext4_fsblk_t ext4_new_meta_block(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, int *errp); -extern ext4_fsblk_t ext4_new_blocks (handle_t *handle, struct inode *inode, +extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp); -extern ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode, +extern ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, + ext4_lblk_t iblock, ext4_fsblk_t goal, + unsigned long *count, int *errp); +extern ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode, ext4_fsblk_t goal, unsigned long *count, int *errp); +extern ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi, + ext4_fsblk_t nblocks); extern void ext4_free_blocks (handle_t *handle, struct inode *inode, ext4_fsblk_t block, unsigned long count, int metadata); extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb, @@ -1016,9 +1037,14 @@ extern int __init init_ext4_mballoc(void); extern void exit_ext4_mballoc(void); extern void ext4_mb_free_blocks(handle_t *, struct inode *, unsigned long, unsigned long, int, unsigned long *); +extern int ext4_mb_add_more_groupinfo(struct super_block *sb, + ext4_group_t i, struct ext4_group_desc *desc); +extern void ext4_mb_update_group_info(struct ext4_group_info *grp, + ext4_grpblk_t add); /* inode.c */ +void ext4_da_release_space(struct inode *inode, int used, int to_free); int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, struct buffer_head *bh, ext4_fsblk_t blocknr); struct buffer_head *ext4_getblk(handle_t *, struct inode *, @@ -1033,19 +1059,23 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, extern struct inode *ext4_iget(struct super_block *, unsigned long); extern int ext4_write_inode (struct inode *, int); extern int ext4_setattr (struct dentry *, struct iattr *); +extern int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat); extern void ext4_delete_inode (struct inode *); extern int ext4_sync_inode (handle_t *, struct inode *); extern void ext4_discard_reservation (struct inode *); extern void ext4_dirty_inode(struct inode *); extern int ext4_change_inode_journal_flag(struct inode *, int); extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); +extern int ext4_can_truncate(struct inode *inode); extern void ext4_truncate (struct inode *); extern void ext4_set_inode_flags(struct inode *); extern void ext4_get_inode_flags(struct ext4_inode_info *); extern void ext4_set_aops(struct inode *inode); extern int ext4_writepage_trans_blocks(struct inode *); -extern int ext4_block_truncate_page(handle_t *handle, struct page *page, +extern int ext4_block_truncate_page(handle_t *handle, struct address_space *mapping, loff_t from); +extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page); /* ioctl.c */ extern long ext4_ioctl(struct file *, unsigned int, unsigned long); @@ -1159,10 +1189,21 @@ struct ext4_group_info *ext4_get_group_info(struct super_block *sb, } +static inline ext4_group_t ext4_flex_group(struct ext4_sb_info *sbi, + ext4_group_t block_group) +{ + return block_group >> sbi->s_log_groups_per_flex; +} + +static inline unsigned int ext4_flex_bg_size(struct ext4_sb_info *sbi) +{ + return 1 << sbi->s_log_groups_per_flex; +} + #define ext4_std_error(sb, errno) \ do { \ if ((errno)) \ - __ext4_std_error((sb), __FUNCTION__, (errno)); \ + __ext4_std_error((sb), __func__, (errno)); \ } while (0) /* @@ -1191,7 +1232,7 @@ extern int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, unsigned long max_blocks, struct buffer_head *bh_result, int create, int extend_disksize); -extern void ext4_ext_truncate(struct inode *, struct page *); +extern void ext4_ext_truncate(struct inode *); extern void ext4_ext_init(struct super_block *); extern void ext4_ext_release(struct super_block *); extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset, @@ -1199,7 +1240,7 @@ extern long ext4_fallocate(struct inode *inode, int mode, loff_t offset, extern int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, unsigned long max_blocks, struct buffer_head *bh, int create, - int extend_disksize); + int extend_disksize, int flag); #endif /* __KERNEL__ */ #endif /* _EXT4_H */ diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index 75333b595fab..6c166c0a54b7 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -212,6 +212,7 @@ static inline int ext4_ext_get_actual_len(struct ext4_extent *ext) (le16_to_cpu(ext->ee_len) - EXT_INIT_MAX_LEN)); } +extern int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks); extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *); extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t); extern int ext4_extent_tree_init(handle_t *, struct inode *); diff --git a/fs/ext4/ext4_i.h b/fs/ext4/ext4_i.h index 26a4ae255d79..ef7409f0e7e4 100644 --- a/fs/ext4/ext4_i.h +++ b/fs/ext4/ext4_i.h @@ -79,7 +79,7 @@ struct ext4_ext_cache { }; /* - * third extended file system inode data in memory + * fourth extended file system inode data in memory */ struct ext4_inode_info { __le32 i_data[15]; /* unconverted */ @@ -150,6 +150,7 @@ struct ext4_inode_info { */ struct rw_semaphore i_data_sem; struct inode vfs_inode; + struct jbd2_inode jinode; unsigned long i_ext_generation; struct ext4_ext_cache i_cached_extent; @@ -162,6 +163,13 @@ struct ext4_inode_info { /* mballoc */ struct list_head i_prealloc_list; spinlock_t i_prealloc_lock; + + /* allocation reservation info for delalloc */ + unsigned long i_reserved_data_blocks; + unsigned long i_reserved_meta_blocks; + unsigned long i_allocated_meta_blocks; + unsigned short i_delalloc_reserved_flag; + spinlock_t i_block_reservation_lock; }; #endif /* _EXT4_I */ diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index 9255a7d28b24..eb8bc3afe6e9 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -142,19 +142,17 @@ int __ext4_journal_dirty_metadata(const char *where, handle_t *handle, struct buffer_head *bh); #define ext4_journal_get_undo_access(handle, bh) \ - __ext4_journal_get_undo_access(__FUNCTION__, (handle), (bh)) + __ext4_journal_get_undo_access(__func__, (handle), (bh)) #define ext4_journal_get_write_access(handle, bh) \ - __ext4_journal_get_write_access(__FUNCTION__, (handle), (bh)) + __ext4_journal_get_write_access(__func__, (handle), (bh)) #define ext4_journal_revoke(handle, blocknr, bh) \ - __ext4_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh)) + __ext4_journal_revoke(__func__, (handle), (blocknr), (bh)) #define ext4_journal_get_create_access(handle, bh) \ - __ext4_journal_get_create_access(__FUNCTION__, (handle), (bh)) + __ext4_journal_get_create_access(__func__, (handle), (bh)) #define ext4_journal_dirty_metadata(handle, bh) \ - __ext4_journal_dirty_metadata(__FUNCTION__, (handle), (bh)) + __ext4_journal_dirty_metadata(__func__, (handle), (bh)) #define ext4_journal_forget(handle, bh) \ - __ext4_journal_forget(__FUNCTION__, (handle), (bh)) - -int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh); + __ext4_journal_forget(__func__, (handle), (bh)) handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks); int __ext4_journal_stop(const char *where, handle_t *handle); @@ -165,7 +163,7 @@ static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks) } #define ext4_journal_stop(handle) \ - __ext4_journal_stop(__FUNCTION__, (handle)) + __ext4_journal_stop(__func__, (handle)) static inline handle_t *ext4_journal_current_handle(void) { @@ -192,6 +190,11 @@ static inline int ext4_journal_force_commit(journal_t *journal) return jbd2_journal_force_commit(journal); } +static inline int ext4_jbd2_file_inode(handle_t *handle, struct inode *inode) +{ + return jbd2_journal_file_inode(handle, &EXT4_I(inode)->jinode); +} + /* super.c */ int ext4_force_commit(struct super_block *sb); diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h index 5802e69f2191..6300226d5531 100644 --- a/fs/ext4/ext4_sb.h +++ b/fs/ext4/ext4_sb.h @@ -25,7 +25,7 @@ #include <linux/rbtree.h> /* - * third extended-fs super-block data in memory + * fourth extended-fs super-block data in memory */ struct ext4_sb_info { unsigned long s_desc_size; /* Size of a group descriptor in bytes */ @@ -143,6 +143,9 @@ struct ext4_sb_info { /* locality groups */ struct ext4_locality_group *s_locality_groups; + + unsigned int s_log_groups_per_flex; + struct flex_groups *s_flex_groups; }; #endif /* _EXT4_SB */ diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 47929c4e3dae..42c4c0c892ed 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -92,17 +92,16 @@ static void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); } -static handle_t *ext4_ext_journal_restart(handle_t *handle, int needed) +static int ext4_ext_journal_restart(handle_t *handle, int needed) { int err; if (handle->h_buffer_credits > needed) - return handle; - if (!ext4_journal_extend(handle, needed)) - return handle; - err = ext4_journal_restart(handle, needed); - - return handle; + return 0; + err = ext4_journal_extend(handle, needed); + if (err) + return err; + return ext4_journal_restart(handle, needed); } /* @@ -180,15 +179,18 @@ static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, return bg_start + colour + block; } +/* + * Allocation for a meta data block + */ static ext4_fsblk_t -ext4_ext_new_block(handle_t *handle, struct inode *inode, +ext4_ext_new_meta_block(handle_t *handle, struct inode *inode, struct ext4_ext_path *path, struct ext4_extent *ex, int *err) { ext4_fsblk_t goal, newblock; goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block)); - newblock = ext4_new_block(handle, inode, goal, err); + newblock = ext4_new_meta_block(handle, inode, goal, err); return newblock; } @@ -246,6 +248,36 @@ static int ext4_ext_space_root_idx(struct inode *inode) return size; } +/* + * Calculate the number of metadata blocks needed + * to allocate @blocks + * Worse case is one block per extent + */ +int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks) +{ + int lcap, icap, rcap, leafs, idxs, num; + int newextents = blocks; + + rcap = ext4_ext_space_root_idx(inode); + lcap = ext4_ext_space_block(inode); + icap = ext4_ext_space_block_idx(inode); + + /* number of new leaf blocks needed */ + num = leafs = (newextents + lcap - 1) / lcap; + + /* + * Worse case, we need separate index block(s) + * to link all new leaf blocks + */ + idxs = (leafs + icap - 1) / icap; + do { + num += idxs; + idxs = (idxs + icap - 1) / icap; + } while (idxs > rcap); + + return num; +} + static int ext4_ext_max_entries(struct inode *inode, int depth) { @@ -524,6 +556,7 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, alloc = 1; } path[0].p_hdr = eh; + path[0].p_bh = NULL; i = depth; /* walk through the tree */ @@ -552,12 +585,14 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block, } path[ppos].p_depth = i; - path[ppos].p_hdr = eh; path[ppos].p_ext = NULL; path[ppos].p_idx = NULL; /* find extent */ ext4_ext_binsearch(inode, path + ppos, block); + /* if not an empty leaf */ + if (path[ppos].p_ext) + path[ppos].p_block = ext_pblock(path[ppos].p_ext); ext4_ext_show_path(inode, path); @@ -688,7 +723,8 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, /* allocate all needed blocks */ ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); for (a = 0; a < depth - at; a++) { - newblock = ext4_ext_new_block(handle, inode, path, newext, &err); + newblock = ext4_ext_new_meta_block(handle, inode, path, + newext, &err); if (newblock == 0) goto cleanup; ablocks[a] = newblock; @@ -884,7 +920,7 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, ext4_fsblk_t newblock; int err = 0; - newblock = ext4_ext_new_block(handle, inode, path, newext, &err); + newblock = ext4_ext_new_meta_block(handle, inode, path, newext, &err); if (newblock == 0) return err; @@ -981,6 +1017,8 @@ repeat: /* if we found index with free entry, then use that * entry: create all needed subtree and add new leaf */ err = ext4_ext_split(handle, inode, path, newext, i); + if (err) + goto out; /* refill path */ ext4_ext_drop_refs(path); @@ -1883,11 +1921,9 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); #endif - handle = ext4_ext_journal_restart(handle, credits); - if (IS_ERR(handle)) { - err = PTR_ERR(handle); + err = ext4_ext_journal_restart(handle, credits); + if (err) goto out; - } err = ext4_ext_get_access(handle, inode, path + depth); if (err) @@ -2529,6 +2565,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, int err = 0, depth, ret; unsigned long allocated = 0; struct ext4_allocation_request ar; + loff_t disksize; __clear_bit(BH_New, &bh_result->b_state); ext_debug("blocks %u/%lu requested for inode %u\n", @@ -2616,8 +2653,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, */ if (allocated > max_blocks) allocated = max_blocks; - /* mark the buffer unwritten */ - __set_bit(BH_Unwritten, &bh_result->b_state); + set_buffer_unwritten(bh_result); goto out2; } @@ -2716,14 +2752,19 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, goto out2; } - if (extend_disksize && inode->i_size > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = inode->i_size; - /* previous routine could use block we allocated */ newblock = ext_pblock(&newex); allocated = ext4_ext_get_actual_len(&newex); outnew: - __set_bit(BH_New, &bh_result->b_state); + if (extend_disksize) { + disksize = ((loff_t) iblock + ar.len) << inode->i_blkbits; + if (disksize > i_size_read(inode)) + disksize = i_size_read(inode); + if (disksize > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = disksize; + } + + set_buffer_new(bh_result); /* Cache only when it is _not_ an uninitialized extent */ if (create != EXT4_CREATE_UNINITIALIZED_EXT) @@ -2733,7 +2774,7 @@ out: if (allocated > max_blocks) allocated = max_blocks; ext4_ext_show_leaf(inode, path); - __set_bit(BH_Mapped, &bh_result->b_state); + set_buffer_mapped(bh_result); bh_result->b_bdev = inode->i_sb->s_bdev; bh_result->b_blocknr = newblock; out2: @@ -2744,7 +2785,7 @@ out2: return err ? err : allocated; } -void ext4_ext_truncate(struct inode * inode, struct page *page) +void ext4_ext_truncate(struct inode *inode) { struct address_space *mapping = inode->i_mapping; struct super_block *sb = inode->i_sb; @@ -2757,18 +2798,14 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) */ err = ext4_writepage_trans_blocks(inode) + 3; handle = ext4_journal_start(inode, err); - if (IS_ERR(handle)) { - if (page) { - clear_highpage(page); - flush_dcache_page(page); - unlock_page(page); - page_cache_release(page); - } + if (IS_ERR(handle)) return; - } - if (page) - ext4_block_truncate_page(handle, page, mapping, inode->i_size); + if (inode->i_size & (sb->s_blocksize - 1)) + ext4_block_truncate_page(handle, mapping, inode->i_size); + + if (ext4_orphan_add(handle, inode)) + goto out_stop; down_write(&EXT4_I(inode)->i_data_sem); ext4_ext_invalidate_cache(inode); @@ -2780,8 +2817,6 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) * Probably we need not scan at all, * because page truncation is enough. */ - if (ext4_orphan_add(handle, inode)) - goto out_stop; /* we have to know where to truncate from in crash case */ EXT4_I(inode)->i_disksize = inode->i_size; @@ -2798,6 +2833,7 @@ void ext4_ext_truncate(struct inode * inode, struct page *page) handle->h_sync = 1; out_stop: + up_write(&EXT4_I(inode)->i_data_sem); /* * If this was a simple ftruncate() and the file will remain alive, * then we need to clear up the orphan record which we created above. @@ -2808,7 +2844,6 @@ out_stop: if (inode->i_nlink) ext4_orphan_del(handle, inode); - up_write(&EXT4_I(inode)->i_data_sem); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); @@ -2911,7 +2946,7 @@ retry: } ret = ext4_get_blocks_wrap(handle, inode, block, max_blocks, &map_bh, - EXT4_CREATE_UNINITIALIZED_EXT, 0); + EXT4_CREATE_UNINITIALIZED_EXT, 0, 0); if (ret <= 0) { #ifdef EXT4FS_DEBUG WARN_ON(ret <= 0); diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 4159be6366ab..430eb7978db4 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -123,6 +123,23 @@ force_commit: return ret; } +static struct vm_operations_struct ext4_file_vm_ops = { + .fault = filemap_fault, + .page_mkwrite = ext4_page_mkwrite, +}; + +static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) +{ + struct address_space *mapping = file->f_mapping; + + if (!mapping->a_ops->readpage) + return -ENOEXEC; + file_accessed(file); + vma->vm_ops = &ext4_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; + return 0; +} + const struct file_operations ext4_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -133,7 +150,7 @@ const struct file_operations ext4_file_operations = { #ifdef CONFIG_COMPAT .compat_ioctl = ext4_compat_ioctl, #endif - .mmap = generic_file_mmap, + .mmap = ext4_file_mmap, .open = generic_file_open, .release = ext4_release_file, .fsync = ext4_sync_file, @@ -144,6 +161,7 @@ const struct file_operations ext4_file_operations = { const struct inode_operations ext4_file_inode_operations = { .truncate = ext4_truncate, .setattr = ext4_setattr, + .getattr = ext4_getattr, #ifdef CONFIG_EXT4DEV_FS_XATTR .setxattr = generic_setxattr, .getxattr = generic_getxattr, diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 1c8ba48d4f8d..a45c3737ad31 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -27,6 +27,7 @@ #include <linux/sched.h> #include <linux/writeback.h> #include <linux/jbd2.h> +#include <linux/blkdev.h> #include "ext4.h" #include "ext4_jbd2.h" @@ -45,6 +46,7 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) { struct inode *inode = dentry->d_inode; + journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; int ret = 0; J_ASSERT(ext4_journal_current_handle() == NULL); @@ -85,6 +87,8 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) .nr_to_write = 0, /* sys_fsync did this */ }; ret = sync_inode(inode, &wbc); + if (journal && (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(inode->i_sb->s_bdev, NULL); } out: return ret; diff --git a/fs/ext4/group.h b/fs/ext4/group.h index 7eb0604e7eea..c2c0a8d06d0e 100644 --- a/fs/ext4/group.h +++ b/fs/ext4/group.h @@ -13,7 +13,7 @@ extern __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 group, struct ext4_group_desc *gdp); extern int ext4_group_desc_csum_verify(struct ext4_sb_info *sbi, __u32 group, struct ext4_group_desc *gdp); -struct buffer_head *read_block_bitmap(struct super_block *sb, +struct buffer_head *ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group); extern unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh, diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index c6efbab0c801..a92eb305344f 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -157,6 +157,7 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) struct ext4_super_block * es; struct ext4_sb_info *sbi; int fatal = 0, err; + ext4_group_t flex_group; if (atomic_read(&inode->i_count) > 1) { printk ("ext4_free_inode: inode has count=%d\n", @@ -232,6 +233,12 @@ void ext4_free_inode (handle_t *handle, struct inode * inode) if (is_directory) percpu_counter_dec(&sbi->s_dirs_counter); + if (sbi->s_log_groups_per_flex) { + flex_group = ext4_flex_group(sbi, block_group); + spin_lock(sb_bgl_lock(sbi, flex_group)); + sbi->s_flex_groups[flex_group].free_inodes++; + spin_unlock(sb_bgl_lock(sbi, flex_group)); + } } BUFFER_TRACE(bh2, "call ext4_journal_dirty_metadata"); err = ext4_journal_dirty_metadata(handle, bh2); @@ -286,6 +293,80 @@ static int find_group_dir(struct super_block *sb, struct inode *parent, return ret; } +#define free_block_ratio 10 + +static int find_group_flex(struct super_block *sb, struct inode *parent, + ext4_group_t *best_group) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_group_desc *desc; + struct buffer_head *bh; + struct flex_groups *flex_group = sbi->s_flex_groups; + ext4_group_t parent_group = EXT4_I(parent)->i_block_group; + ext4_group_t parent_fbg_group = ext4_flex_group(sbi, parent_group); + ext4_group_t ngroups = sbi->s_groups_count; + int flex_size = ext4_flex_bg_size(sbi); + ext4_group_t best_flex = parent_fbg_group; + int blocks_per_flex = sbi->s_blocks_per_group * flex_size; + int flexbg_free_blocks; + int flex_freeb_ratio; + ext4_group_t n_fbg_groups; + ext4_group_t i; + + n_fbg_groups = (sbi->s_groups_count + flex_size - 1) >> + sbi->s_log_groups_per_flex; + +find_close_to_parent: + flexbg_free_blocks = flex_group[best_flex].free_blocks; + flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex; + if (flex_group[best_flex].free_inodes && + flex_freeb_ratio > free_block_ratio) + goto found_flexbg; + + if (best_flex && best_flex == parent_fbg_group) { + best_flex--; + goto find_close_to_parent; + } + + for (i = 0; i < n_fbg_groups; i++) { + if (i == parent_fbg_group || i == parent_fbg_group - 1) + continue; + + flexbg_free_blocks = flex_group[i].free_blocks; + flex_freeb_ratio = flexbg_free_blocks * 100 / blocks_per_flex; + + if (flex_freeb_ratio > free_block_ratio && + flex_group[i].free_inodes) { + best_flex = i; + goto found_flexbg; + } + + if (best_flex < 0 || + (flex_group[i].free_blocks > + flex_group[best_flex].free_blocks && + flex_group[i].free_inodes)) + best_flex = i; + } + + if (!flex_group[best_flex].free_inodes || + !flex_group[best_flex].free_blocks) + return -1; + +found_flexbg: + for (i = best_flex * flex_size; i < ngroups && + i < (best_flex + 1) * flex_size; i++) { + desc = ext4_get_group_desc(sb, i, &bh); + if (le16_to_cpu(desc->bg_free_inodes_count)) { + *best_group = i; + goto out; + } + } + + return -1; +out: + return 0; +} + /* * Orlov's allocator for directories. * @@ -501,6 +582,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) struct inode *ret; ext4_group_t i; int free = 0; + ext4_group_t flex_group; /* Cannot create files in a deleted directory */ if (!dir || !dir->i_nlink) @@ -514,6 +596,12 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) sbi = EXT4_SB(sb); es = sbi->s_es; + + if (sbi->s_log_groups_per_flex) { + ret2 = find_group_flex(sb, dir, &group); + goto got_group; + } + if (S_ISDIR(mode)) { if (test_opt (sb, OLDALLOC)) ret2 = find_group_dir(sb, dir, &group); @@ -522,6 +610,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) } else ret2 = find_group_other(sb, dir, &group); +got_group: err = -ENOSPC; if (ret2 == -1) goto out; @@ -600,7 +689,7 @@ got: /* We may have to initialize the block bitmap if it isn't already */ if (EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_GDT_CSUM) && gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { - struct buffer_head *block_bh = read_block_bitmap(sb, group); + struct buffer_head *block_bh = ext4_read_block_bitmap(sb, group); BUFFER_TRACE(block_bh, "get block bitmap access"); err = ext4_journal_get_write_access(handle, block_bh); @@ -676,6 +765,13 @@ got: percpu_counter_inc(&sbi->s_dirs_counter); sb->s_dirt = 1; + if (sbi->s_log_groups_per_flex) { + flex_group = ext4_flex_group(sbi, group); + spin_lock(sb_bgl_lock(sbi, flex_group)); + sbi->s_flex_groups[flex_group].free_inodes--; + spin_unlock(sb_bgl_lock(sbi, flex_group)); + } + inode->i_uid = current->fsuid; if (test_opt (sb, GRPID)) inode->i_gid = dir->i_gid; @@ -740,14 +836,10 @@ got: goto fail_free_drop; if (test_opt(sb, EXTENTS)) { - /* set extent flag only for diretory, file and normal symlink*/ + /* set extent flag only for directory, file and normal symlink*/ if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) { EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL; ext4_ext_tree_init(handle, inode); - err = ext4_update_incompat_feature(handle, sb, - EXT4_FEATURE_INCOMPAT_EXTENTS); - if (err) - goto fail_free_drop; } } @@ -817,6 +909,14 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) if (IS_ERR(inode)) goto iget_failed; + /* + * If the orphans has i_nlinks > 0 then it should be able to be + * truncated, otherwise it won't be removed from the orphan list + * during processing and an infinite loop will result. + */ + if (inode->i_nlink && !ext4_can_truncate(inode)) + goto bad_orphan; + if (NEXT_ORPHAN(inode) > max_ino) goto bad_orphan; brelse(bitmap_bh); @@ -838,6 +938,7 @@ bad_orphan: printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", NEXT_ORPHAN(inode)); printk(KERN_NOTICE "max_ino=%lu\n", max_ino); + printk(KERN_NOTICE "i_nlink=%u\n", inode->i_nlink); /* Avoid freeing blocks if we got a bad deleted inode */ if (inode->i_nlink == 0) inode->i_blocks = 0; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 8d9707746413..8ca2763df091 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -32,12 +32,23 @@ #include <linux/string.h> #include <linux/buffer_head.h> #include <linux/writeback.h> +#include <linux/pagevec.h> #include <linux/mpage.h> #include <linux/uio.h> #include <linux/bio.h> #include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" +#include "ext4_extents.h" + +static inline int ext4_begin_ordered_truncate(struct inode *inode, + loff_t new_size) +{ + return jbd2_journal_begin_ordered_truncate(&EXT4_I(inode)->jinode, + new_size); +} + +static void ext4_invalidatepage(struct page *page, unsigned long offset); /* * Test whether an inode is a fast symlink. @@ -181,6 +192,8 @@ void ext4_delete_inode (struct inode * inode) { handle_t *handle; + if (ext4_should_order_data(inode)) + ext4_begin_ordered_truncate(inode, 0); truncate_inode_pages(&inode->i_data, 0); if (is_bad_inode(inode)) @@ -508,11 +521,12 @@ static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned long blks, * direct blocks */ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, - ext4_fsblk_t goal, int indirect_blks, int blks, - ext4_fsblk_t new_blocks[4], int *err) + ext4_lblk_t iblock, ext4_fsblk_t goal, + int indirect_blks, int blks, + ext4_fsblk_t new_blocks[4], int *err) { int target, i; - unsigned long count = 0; + unsigned long count = 0, blk_allocated = 0; int index = 0; ext4_fsblk_t current_block = 0; int ret = 0; @@ -525,12 +539,13 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, * the first direct block of this branch. That's the * minimum number of blocks need to allocate(required) */ - target = blks + indirect_blks; - - while (1) { + /* first we try to allocate the indirect blocks */ + target = indirect_blks; + while (target > 0) { count = target; /* allocating blocks for indirect blocks and direct blocks */ - current_block = ext4_new_blocks(handle,inode,goal,&count,err); + current_block = ext4_new_meta_blocks(handle, inode, + goal, &count, err); if (*err) goto failed_out; @@ -540,16 +555,48 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, new_blocks[index++] = current_block++; count--; } - - if (count > 0) + if (count > 0) { + /* + * save the new block number + * for the first direct block + */ + new_blocks[index] = current_block; + printk(KERN_INFO "%s returned more blocks than " + "requested\n", __func__); + WARN_ON(1); break; + } } - /* save the new block number for the first direct block */ - new_blocks[index] = current_block; - + target = blks - count ; + blk_allocated = count; + if (!target) + goto allocated; + /* Now allocate data blocks */ + count = target; + /* allocating blocks for data blocks */ + current_block = ext4_new_blocks(handle, inode, iblock, + goal, &count, err); + if (*err && (target == blks)) { + /* + * if the allocation failed and we didn't allocate + * any blocks before + */ + goto failed_out; + } + if (!*err) { + if (target == blks) { + /* + * save the new block number + * for the first direct block + */ + new_blocks[index] = current_block; + } + blk_allocated += count; + } +allocated: /* total number of blocks allocated for direct blocks */ - ret = count; + ret = blk_allocated; *err = 0; return ret; failed_out: @@ -584,8 +631,9 @@ failed_out: * as described above and return 0. */ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, - int indirect_blks, int *blks, ext4_fsblk_t goal, - ext4_lblk_t *offsets, Indirect *branch) + ext4_lblk_t iblock, int indirect_blks, + int *blks, ext4_fsblk_t goal, + ext4_lblk_t *offsets, Indirect *branch) { int blocksize = inode->i_sb->s_blocksize; int i, n = 0; @@ -595,7 +643,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode, ext4_fsblk_t new_blocks[4]; ext4_fsblk_t current_block; - num = ext4_alloc_blocks(handle, inode, goal, indirect_blks, + num = ext4_alloc_blocks(handle, inode, iblock, goal, indirect_blks, *blks, new_blocks, &err); if (err) return err; @@ -799,6 +847,7 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, struct ext4_inode_info *ei = EXT4_I(inode); int count = 0; ext4_fsblk_t first_block = 0; + loff_t disksize; J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)); @@ -855,8 +904,9 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, /* * Block out ext4_truncate while we alter the tree */ - err = ext4_alloc_branch(handle, inode, indirect_blks, &count, goal, - offsets + (partial - chain), partial); + err = ext4_alloc_branch(handle, inode, iblock, indirect_blks, + &count, goal, + offsets + (partial - chain), partial); /* * The ext4_splice_branch call will free and forget any buffers @@ -873,8 +923,13 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, * protect it if you're about to implement concurrent * ext4_get_block() -bzzz */ - if (!err && extend_disksize && inode->i_size > ei->i_disksize) - ei->i_disksize = inode->i_size; + if (!err && extend_disksize) { + disksize = ((loff_t) iblock + count) << inode->i_blkbits; + if (disksize > i_size_read(inode)) + disksize = i_size_read(inode); + if (disksize > ei->i_disksize) + ei->i_disksize = disksize; + } if (err) goto cleanup; @@ -934,7 +989,7 @@ out: */ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, unsigned long max_blocks, struct buffer_head *bh, - int create, int extend_disksize) + int create, int extend_disksize, int flag) { int retval; @@ -975,6 +1030,15 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, * with create == 1 flag. */ down_write((&EXT4_I(inode)->i_data_sem)); + + /* + * if the caller is from delayed allocation writeout path + * we have already reserved fs blocks for allocation + * let the underlying get_block() function know to + * avoid double accounting + */ + if (flag) + EXT4_I(inode)->i_delalloc_reserved_flag = 1; /* * We need to check for EXT4 here because migrate * could have changed the inode type in between @@ -996,6 +1060,18 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block, ~EXT4_EXT_MIGRATE; } } + + if (flag) { + EXT4_I(inode)->i_delalloc_reserved_flag = 0; + /* + * Update reserved blocks/metadata blocks + * after successful block allocation + * which were deferred till now + */ + if ((retval > 0) && buffer_delay(bh)) + ext4_da_release_space(inode, retval, 0); + } + up_write((&EXT4_I(inode)->i_data_sem)); return retval; } @@ -1021,7 +1097,7 @@ static int ext4_get_block(struct inode *inode, sector_t iblock, } ret = ext4_get_blocks_wrap(handle, inode, iblock, - max_blocks, bh_result, create, 0); + max_blocks, bh_result, create, 0, 0); if (ret > 0) { bh_result->b_size = (ret << inode->i_blkbits); ret = 0; @@ -1047,7 +1123,7 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, dummy.b_blocknr = -1000; buffer_trace_init(&dummy.b_history); err = ext4_get_blocks_wrap(handle, inode, block, 1, - &dummy, create, 1); + &dummy, create, 1, 0); /* * ext4_get_blocks_handle() returns number of blocks * mapped. 0 in case of a HOLE. @@ -1203,19 +1279,20 @@ static int ext4_write_begin(struct file *file, struct address_space *mapping, to = from + len; retry: - page = __grab_cache_page(mapping, index); - if (!page) - return -ENOMEM; - *pagep = page; - handle = ext4_journal_start(inode, needed_blocks); if (IS_ERR(handle)) { - unlock_page(page); - page_cache_release(page); ret = PTR_ERR(handle); goto out; } + page = __grab_cache_page(mapping, index); + if (!page) { + ext4_journal_stop(handle); + ret = -ENOMEM; + goto out; + } + *pagep = page; + ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, ext4_get_block); @@ -1225,8 +1302,8 @@ retry: } if (ret) { - ext4_journal_stop(handle); unlock_page(page); + ext4_journal_stop(handle); page_cache_release(page); } @@ -1236,15 +1313,6 @@ out: return ret; } -int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh) -{ - int err = jbd2_journal_dirty_data(handle, bh); - if (err) - ext4_journal_abort_handle(__func__, __func__, - bh, handle, err); - return err; -} - /* For write_end() in data=journal mode */ static int write_end_fn(handle_t *handle, struct buffer_head *bh) { @@ -1255,29 +1323,6 @@ static int write_end_fn(handle_t *handle, struct buffer_head *bh) } /* - * Generic write_end handler for ordered and writeback ext4 journal modes. - * We can't use generic_write_end, because that unlocks the page and we need to - * unlock the page after ext4_journal_stop, but ext4_journal_stop must run - * after block_write_end. - */ -static int ext4_generic_write_end(struct file *file, - struct address_space *mapping, - loff_t pos, unsigned len, unsigned copied, - struct page *page, void *fsdata) -{ - struct inode *inode = file->f_mapping->host; - - copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); - - if (pos+copied > inode->i_size) { - i_size_write(inode, pos+copied); - mark_inode_dirty(inode); - } - - return copied; -} - -/* * We need to pick up the new inode size which generic_commit_write gave us * `file' can be NULL - eg, when called from page_symlink(). * @@ -1290,15 +1335,14 @@ static int ext4_ordered_write_end(struct file *file, struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); - struct inode *inode = file->f_mapping->host; + struct inode *inode = mapping->host; unsigned from, to; int ret = 0, ret2; from = pos & (PAGE_CACHE_SIZE - 1); to = from + len; - ret = walk_page_buffers(handle, page_buffers(page), - from, to, NULL, ext4_journal_dirty_data); + ret = ext4_jbd2_file_inode(handle, inode); if (ret == 0) { /* @@ -1311,7 +1355,7 @@ static int ext4_ordered_write_end(struct file *file, new_i_size = pos + copied; if (new_i_size > EXT4_I(inode)->i_disksize) EXT4_I(inode)->i_disksize = new_i_size; - ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, + ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; if (ret2 < 0) @@ -1320,8 +1364,6 @@ static int ext4_ordered_write_end(struct file *file, ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - unlock_page(page); - page_cache_release(page); return ret ? ret : copied; } @@ -1332,7 +1374,7 @@ static int ext4_writeback_write_end(struct file *file, struct page *page, void *fsdata) { handle_t *handle = ext4_journal_current_handle(); - struct inode *inode = file->f_mapping->host; + struct inode *inode = mapping->host; int ret = 0, ret2; loff_t new_i_size; @@ -1340,7 +1382,7 @@ static int ext4_writeback_write_end(struct file *file, if (new_i_size > EXT4_I(inode)->i_disksize) EXT4_I(inode)->i_disksize = new_i_size; - ret2 = ext4_generic_write_end(file, mapping, pos, len, copied, + ret2 = generic_write_end(file, mapping, pos, len, copied, page, fsdata); copied = ret2; if (ret2 < 0) @@ -1349,8 +1391,6 @@ static int ext4_writeback_write_end(struct file *file, ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - unlock_page(page); - page_cache_release(page); return ret ? ret : copied; } @@ -1389,14 +1429,965 @@ static int ext4_journalled_write_end(struct file *file, ret = ret2; } + unlock_page(page); ret2 = ext4_journal_stop(handle); if (!ret) ret = ret2; - unlock_page(page); page_cache_release(page); return ret ? ret : copied; } +/* + * Calculate the number of metadata blocks need to reserve + * to allocate @blocks for non extent file based file + */ +static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) +{ + int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb); + int ind_blks, dind_blks, tind_blks; + + /* number of new indirect blocks needed */ + ind_blks = (blocks + icap - 1) / icap; + + dind_blks = (ind_blks + icap - 1) / icap; + + tind_blks = 1; + + return ind_blks + dind_blks + tind_blks; +} + +/* + * Calculate the number of metadata blocks need to reserve + * to allocate given number of blocks + */ +static int ext4_calc_metadata_amount(struct inode *inode, int blocks) +{ + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_calc_metadata_amount(inode, blocks); + + return ext4_indirect_calc_metadata_amount(inode, blocks); +} + +static int ext4_da_reserve_space(struct inode *inode, int nrblocks) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + unsigned long md_needed, mdblocks, total = 0; + + /* + * recalculate the amount of metadata blocks to reserve + * in order to allocate nrblocks + * worse case is one extent per block + */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; + mdblocks = ext4_calc_metadata_amount(inode, total); + BUG_ON(mdblocks < EXT4_I(inode)->i_reserved_meta_blocks); + + md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; + total = md_needed + nrblocks; + + if (ext4_has_free_blocks(sbi, total) < total) { + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return -ENOSPC; + } + + /* reduce fs free blocks counter */ + percpu_counter_sub(&sbi->s_freeblocks_counter, total); + + EXT4_I(inode)->i_reserved_data_blocks += nrblocks; + EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; + + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + return 0; /* success */ +} + +void ext4_da_release_space(struct inode *inode, int used, int to_free) +{ + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + int total, mdb, mdb_free, release; + + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + /* recalculate the number of metablocks still need to be reserved */ + total = EXT4_I(inode)->i_reserved_data_blocks - used - to_free; + mdb = ext4_calc_metadata_amount(inode, total); + + /* figure out how many metablocks to release */ + BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); + mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; + + /* Account for allocated meta_blocks */ + mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; + + release = to_free + mdb_free; + + /* update fs free blocks counter for truncate case */ + percpu_counter_add(&sbi->s_freeblocks_counter, release); + + /* update per-inode reservations */ + BUG_ON(used + to_free > EXT4_I(inode)->i_reserved_data_blocks); + EXT4_I(inode)->i_reserved_data_blocks -= (used + to_free); + + BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); + EXT4_I(inode)->i_reserved_meta_blocks = mdb; + EXT4_I(inode)->i_allocated_meta_blocks = 0; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); +} + +static void ext4_da_page_release_reservation(struct page *page, + unsigned long offset) +{ + int to_release = 0; + struct buffer_head *head, *bh; + unsigned int curr_off = 0; + + head = page_buffers(page); + bh = head; + do { + unsigned int next_off = curr_off + bh->b_size; + + if ((offset <= curr_off) && (buffer_delay(bh))) { + to_release++; + clear_buffer_delay(bh); + } + curr_off = next_off; + } while ((bh = bh->b_this_page) != head); + ext4_da_release_space(page->mapping->host, 0, to_release); +} + +/* + * Delayed allocation stuff + */ + +struct mpage_da_data { + struct inode *inode; + struct buffer_head lbh; /* extent of blocks */ + unsigned long first_page, next_page; /* extent of pages */ + get_block_t *get_block; + struct writeback_control *wbc; +}; + +/* + * mpage_da_submit_io - walks through extent of pages and try to write + * them with __mpage_writepage() + * + * @mpd->inode: inode + * @mpd->first_page: first page of the extent + * @mpd->next_page: page after the last page of the extent + * @mpd->get_block: the filesystem's block mapper function + * + * By the time mpage_da_submit_io() is called we expect all blocks + * to be allocated. this may be wrong if allocation failed. + * + * As pages are already locked by write_cache_pages(), we can't use it + */ +static int mpage_da_submit_io(struct mpage_da_data *mpd) +{ + struct address_space *mapping = mpd->inode->i_mapping; + struct mpage_data mpd_pp = { + .bio = NULL, + .last_block_in_bio = 0, + .get_block = mpd->get_block, + .use_writepage = 1, + }; + int ret = 0, err, nr_pages, i; + unsigned long index, end; + struct pagevec pvec; + + BUG_ON(mpd->next_page <= mpd->first_page); + + pagevec_init(&pvec, 0); + index = mpd->first_page; + end = mpd->next_page - 1; + + while (index <= end) { + /* XXX: optimize tail */ + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + index = page->index; + if (index > end) + break; + index++; + + err = __mpage_writepage(page, mpd->wbc, &mpd_pp); + + /* + * In error case, we have to continue because + * remaining pages are still locked + * XXX: unlock and re-dirty them? + */ + if (ret == 0) + ret = err; + } + pagevec_release(&pvec); + } + if (mpd_pp.bio) + mpage_bio_submit(WRITE, mpd_pp.bio); + + return ret; +} + +/* + * mpage_put_bnr_to_bhs - walk blocks and assign them actual numbers + * + * @mpd->inode - inode to walk through + * @exbh->b_blocknr - first block on a disk + * @exbh->b_size - amount of space in bytes + * @logical - first logical block to start assignment with + * + * the function goes through all passed space and put actual disk + * block numbers into buffer heads, dropping BH_Delay + */ +static void mpage_put_bnr_to_bhs(struct mpage_da_data *mpd, sector_t logical, + struct buffer_head *exbh) +{ + struct inode *inode = mpd->inode; + struct address_space *mapping = inode->i_mapping; + int blocks = exbh->b_size >> inode->i_blkbits; + sector_t pblock = exbh->b_blocknr, cur_logical; + struct buffer_head *head, *bh; + unsigned long index, end; + struct pagevec pvec; + int nr_pages, i; + + index = logical >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + end = (logical + blocks - 1) >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + cur_logical = index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + + pagevec_init(&pvec, 0); + + while (index <= end) { + /* XXX: optimize tail */ + nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + index = page->index; + if (index > end) + break; + index++; + + BUG_ON(!PageLocked(page)); + BUG_ON(PageWriteback(page)); + BUG_ON(!page_has_buffers(page)); + + bh = page_buffers(page); + head = bh; + + /* skip blocks out of the range */ + do { + if (cur_logical >= logical) + break; + cur_logical++; + } while ((bh = bh->b_this_page) != head); + + do { + if (cur_logical >= logical + blocks) + break; + if (buffer_delay(bh)) { + bh->b_blocknr = pblock; + clear_buffer_delay(bh); + } else if (buffer_mapped(bh)) + BUG_ON(bh->b_blocknr != pblock); + + cur_logical++; + pblock++; + } while ((bh = bh->b_this_page) != head); + } + pagevec_release(&pvec); + } +} + + +/* + * __unmap_underlying_blocks - just a helper function to unmap + * set of blocks described by @bh + */ +static inline void __unmap_underlying_blocks(struct inode *inode, + struct buffer_head *bh) +{ + struct block_device *bdev = inode->i_sb->s_bdev; + int blocks, i; + + blocks = bh->b_size >> inode->i_blkbits; + for (i = 0; i < blocks; i++) + unmap_underlying_metadata(bdev, bh->b_blocknr + i); +} + +/* + * mpage_da_map_blocks - go through given space + * + * @mpd->lbh - bh describing space + * @mpd->get_block - the filesystem's block mapper function + * + * The function skips space we know is already mapped to disk blocks. + * + * The function ignores errors ->get_block() returns, thus real + * error handling is postponed to __mpage_writepage() + */ +static void mpage_da_map_blocks(struct mpage_da_data *mpd) +{ + struct buffer_head *lbh = &mpd->lbh; + int err = 0, remain = lbh->b_size; + sector_t next = lbh->b_blocknr; + struct buffer_head new; + + /* + * We consider only non-mapped and non-allocated blocks + */ + if (buffer_mapped(lbh) && !buffer_delay(lbh)) + return; + + while (remain) { + new.b_state = lbh->b_state; + new.b_blocknr = 0; + new.b_size = remain; + err = mpd->get_block(mpd->inode, next, &new, 1); + if (err) { + /* + * Rather than implement own error handling + * here, we just leave remaining blocks + * unallocated and try again with ->writepage() + */ + break; + } + BUG_ON(new.b_size == 0); + + if (buffer_new(&new)) + __unmap_underlying_blocks(mpd->inode, &new); + + /* + * If blocks are delayed marked, we need to + * put actual blocknr and drop delayed bit + */ + if (buffer_delay(lbh)) + mpage_put_bnr_to_bhs(mpd, next, &new); + + /* go for the remaining blocks */ + next += new.b_size >> mpd->inode->i_blkbits; + remain -= new.b_size; + } +} + +#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | (1 << BH_Delay)) + +/* + * mpage_add_bh_to_extent - try to add one more block to extent of blocks + * + * @mpd->lbh - extent of blocks + * @logical - logical number of the block in the file + * @bh - bh of the block (used to access block's state) + * + * the function is used to collect contig. blocks in same state + */ +static void mpage_add_bh_to_extent(struct mpage_da_data *mpd, + sector_t logical, struct buffer_head *bh) +{ + struct buffer_head *lbh = &mpd->lbh; + sector_t next; + + next = lbh->b_blocknr + (lbh->b_size >> mpd->inode->i_blkbits); + + /* + * First block in the extent + */ + if (lbh->b_size == 0) { + lbh->b_blocknr = logical; + lbh->b_size = bh->b_size; + lbh->b_state = bh->b_state & BH_FLAGS; + return; + } + + /* + * Can we merge the block to our big extent? + */ + if (logical == next && (bh->b_state & BH_FLAGS) == lbh->b_state) { + lbh->b_size += bh->b_size; + return; + } + + /* + * We couldn't merge the block to our extent, so we + * need to flush current extent and start new one + */ + mpage_da_map_blocks(mpd); + + /* + * Now start a new extent + */ + lbh->b_size = bh->b_size; + lbh->b_state = bh->b_state & BH_FLAGS; + lbh->b_blocknr = logical; +} + +/* + * __mpage_da_writepage - finds extent of pages and blocks + * + * @page: page to consider + * @wbc: not used, we just follow rules + * @data: context + * + * The function finds extents of pages and scan them for all blocks. + */ +static int __mpage_da_writepage(struct page *page, + struct writeback_control *wbc, void *data) +{ + struct mpage_da_data *mpd = data; + struct inode *inode = mpd->inode; + struct buffer_head *bh, *head, fake; + sector_t logical; + + /* + * Can we merge this page to current extent? + */ + if (mpd->next_page != page->index) { + /* + * Nope, we can't. So, we map non-allocated blocks + * and start IO on them using __mpage_writepage() + */ + if (mpd->next_page != mpd->first_page) { + mpage_da_map_blocks(mpd); + mpage_da_submit_io(mpd); + } + + /* + * Start next extent of pages ... + */ + mpd->first_page = page->index; + + /* + * ... and blocks + */ + mpd->lbh.b_size = 0; + mpd->lbh.b_state = 0; + mpd->lbh.b_blocknr = 0; + } + + mpd->next_page = page->index + 1; + logical = (sector_t) page->index << + (PAGE_CACHE_SHIFT - inode->i_blkbits); + + if (!page_has_buffers(page)) { + /* + * There is no attached buffer heads yet (mmap?) + * we treat the page asfull of dirty blocks + */ + bh = &fake; + bh->b_size = PAGE_CACHE_SIZE; + bh->b_state = 0; + set_buffer_dirty(bh); + set_buffer_uptodate(bh); + mpage_add_bh_to_extent(mpd, logical, bh); + } else { + /* + * Page with regular buffer heads, just add all dirty ones + */ + head = page_buffers(page); + bh = head; + do { + BUG_ON(buffer_locked(bh)); + if (buffer_dirty(bh)) + mpage_add_bh_to_extent(mpd, logical, bh); + logical++; + } while ((bh = bh->b_this_page) != head); + } + + return 0; +} + +/* + * mpage_da_writepages - walk the list of dirty pages of the given + * address space, allocates non-allocated blocks, maps newly-allocated + * blocks to existing bhs and issue IO them + * + * @mapping: address space structure to write + * @wbc: subtract the number of written pages from *@wbc->nr_to_write + * @get_block: the filesystem's block mapper function. + * + * This is a library function, which implements the writepages() + * address_space_operation. + * + * In order to avoid duplication of logic that deals with partial pages, + * multiple bio per page, etc, we find non-allocated blocks, allocate + * them with minimal calls to ->get_block() and re-use __mpage_writepage() + * + * It's important that we call __mpage_writepage() only once for each + * involved page, otherwise we'd have to implement more complicated logic + * to deal with pages w/o PG_lock or w/ PG_writeback and so on. + * + * See comments to mpage_writepages() + */ +static int mpage_da_writepages(struct address_space *mapping, + struct writeback_control *wbc, + get_block_t get_block) +{ + struct mpage_da_data mpd; + int ret; + + if (!get_block) + return generic_writepages(mapping, wbc); + + mpd.wbc = wbc; + mpd.inode = mapping->host; + mpd.lbh.b_size = 0; + mpd.lbh.b_state = 0; + mpd.lbh.b_blocknr = 0; + mpd.first_page = 0; + mpd.next_page = 0; + mpd.get_block = get_block; + + ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd); + + /* + * Handle last extent of pages + */ + if (mpd.next_page != mpd.first_page) { + mpage_da_map_blocks(&mpd); + mpage_da_submit_io(&mpd); + } + + return ret; +} + +/* + * this is a special callback for ->write_begin() only + * it's intention is to return mapped block or reserve space + */ +static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + int ret = 0; + + BUG_ON(create == 0); + BUG_ON(bh_result->b_size != inode->i_sb->s_blocksize); + + /* + * first, we need to know whether the block is allocated already + * preallocated blocks are unmapped but should treated + * the same as allocated blocks. + */ + ret = ext4_get_blocks_wrap(NULL, inode, iblock, 1, bh_result, 0, 0, 0); + if ((ret == 0) && !buffer_delay(bh_result)) { + /* the block isn't (pre)allocated yet, let's reserve space */ + /* + * XXX: __block_prepare_write() unmaps passed block, + * is it OK? + */ + ret = ext4_da_reserve_space(inode, 1); + if (ret) + /* not enough space to reserve */ + return ret; + + map_bh(bh_result, inode->i_sb, 0); + set_buffer_new(bh_result); + set_buffer_delay(bh_result); + } else if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + + return ret; +} +#define EXT4_DELALLOC_RSVED 1 +static int ext4_da_get_block_write(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + int ret; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + loff_t disksize = EXT4_I(inode)->i_disksize; + handle_t *handle = NULL; + + handle = ext4_journal_current_handle(); + if (!handle) { + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + bh_result, 0, 0, 0); + BUG_ON(!ret); + } else { + ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks, + bh_result, create, 0, EXT4_DELALLOC_RSVED); + } + + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + + /* + * Update on-disk size along with block allocation + * we don't use 'extend_disksize' as size may change + * within already allocated block -bzzz + */ + disksize = ((loff_t) iblock + ret) << inode->i_blkbits; + if (disksize > i_size_read(inode)) + disksize = i_size_read(inode); + if (disksize > EXT4_I(inode)->i_disksize) { + /* + * XXX: replace with spinlock if seen contended -bzzz + */ + down_write(&EXT4_I(inode)->i_data_sem); + if (disksize > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = disksize; + up_write(&EXT4_I(inode)->i_data_sem); + + if (EXT4_I(inode)->i_disksize == disksize) { + ret = ext4_mark_inode_dirty(handle, inode); + return ret; + } + } + ret = 0; + } + return ret; +} + +static int ext4_bh_unmapped_or_delay(handle_t *handle, struct buffer_head *bh) +{ + /* + * unmapped buffer is possible for holes. + * delay buffer is possible with delayed allocation + */ + return ((!buffer_mapped(bh) || buffer_delay(bh)) && buffer_dirty(bh)); +} + +static int ext4_normal_get_block_write(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + int ret = 0; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + + /* + * we don't want to do block allocation in writepage + * so call get_block_wrap with create = 0 + */ + ret = ext4_get_blocks_wrap(NULL, inode, iblock, max_blocks, + bh_result, 0, 0, 0); + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + return ret; +} + +/* + * get called vi ext4_da_writepages after taking page lock (have journal handle) + * get called via journal_submit_inode_data_buffers (no journal handle) + * get called via shrink_page_list via pdflush (no journal handle) + * or grab_page_cache when doing write_begin (have journal handle) + */ +static int ext4_da_writepage(struct page *page, + struct writeback_control *wbc) +{ + int ret = 0; + loff_t size; + unsigned long len; + struct buffer_head *page_bufs; + struct inode *inode = page->mapping->host; + + size = i_size_read(inode); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + + if (page_has_buffers(page)) { + page_bufs = page_buffers(page); + if (walk_page_buffers(NULL, page_bufs, 0, len, NULL, + ext4_bh_unmapped_or_delay)) { + /* + * We don't want to do block allocation + * So redirty the page and return + * We may reach here when we do a journal commit + * via journal_submit_inode_data_buffers. + * If we don't have mapping block we just ignore + * them. We can also reach here via shrink_page_list + */ + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + } else { + /* + * The test for page_has_buffers() is subtle: + * We know the page is dirty but it lost buffers. That means + * that at some moment in time after write_begin()/write_end() + * has been called all buffers have been clean and thus they + * must have been written at least once. So they are all + * mapped and we can happily proceed with mapping them + * and writing the page. + * + * Try to initialize the buffer_heads and check whether + * all are mapped and non delay. We don't want to + * do block allocation here. + */ + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, + ext4_normal_get_block_write); + if (!ret) { + page_bufs = page_buffers(page); + /* check whether all are mapped and non delay */ + if (walk_page_buffers(NULL, page_bufs, 0, len, NULL, + ext4_bh_unmapped_or_delay)) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + } else { + /* + * We can't do block allocation here + * so just redity the page and unlock + * and return + */ + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + } + + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_writepage(page, ext4_normal_get_block_write, wbc); + else + ret = block_write_full_page(page, + ext4_normal_get_block_write, + wbc); + + return ret; +} + +/* + * For now just follow the DIO way to estimate the max credits + * needed to write out EXT4_MAX_WRITEBACK_PAGES. + * todo: need to calculate the max credits need for + * extent based files, currently the DIO credits is based on + * indirect-blocks mapping way. + * + * Probably should have a generic way to calculate credits + * for DIO, writepages, and truncate + */ +#define EXT4_MAX_WRITEBACK_PAGES DIO_MAX_BLOCKS +#define EXT4_MAX_WRITEBACK_CREDITS DIO_CREDITS + +static int ext4_da_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct inode *inode = mapping->host; + handle_t *handle = NULL; + int needed_blocks; + int ret = 0; + long to_write; + loff_t range_start = 0; + + /* + * No pages to write? This is mainly a kludge to avoid starting + * a transaction for special inodes like journal inode on last iput() + * because that could violate lock ordering on umount + */ + if (!mapping->nrpages) + return 0; + + /* + * Estimate the worse case needed credits to write out + * EXT4_MAX_BUF_BLOCKS pages + */ + needed_blocks = EXT4_MAX_WRITEBACK_CREDITS; + + to_write = wbc->nr_to_write; + if (!wbc->range_cyclic) { + /* + * If range_cyclic is not set force range_cont + * and save the old writeback_index + */ + wbc->range_cont = 1; + range_start = wbc->range_start; + } + + while (!ret && to_write) { + /* start a new transaction*/ + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_writepages; + } + if (ext4_should_order_data(inode)) { + /* + * With ordered mode we need to add + * the inode to the journal handle + * when we do block allocation. + */ + ret = ext4_jbd2_file_inode(handle, inode); + if (ret) { + ext4_journal_stop(handle); + goto out_writepages; + } + + } + /* + * set the max dirty pages could be write at a time + * to fit into the reserved transaction credits + */ + if (wbc->nr_to_write > EXT4_MAX_WRITEBACK_PAGES) + wbc->nr_to_write = EXT4_MAX_WRITEBACK_PAGES; + + to_write -= wbc->nr_to_write; + ret = mpage_da_writepages(mapping, wbc, + ext4_da_get_block_write); + ext4_journal_stop(handle); + if (wbc->nr_to_write) { + /* + * There is no more writeout needed + * or we requested for a noblocking writeout + * and we found the device congested + */ + to_write += wbc->nr_to_write; + break; + } + wbc->nr_to_write = to_write; + } + +out_writepages: + wbc->nr_to_write = to_write; + if (range_start) + wbc->range_start = range_start; + return ret; +} + +static int ext4_da_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + int ret, retries = 0; + struct page *page; + pgoff_t index; + unsigned from, to; + struct inode *inode = mapping->host; + handle_t *handle; + + index = pos >> PAGE_CACHE_SHIFT; + from = pos & (PAGE_CACHE_SIZE - 1); + to = from + len; + +retry: + /* + * With delayed allocation, we don't log the i_disksize update + * if there is delayed block allocation. But we still need + * to journalling the i_disksize update if writes to the end + * of file which has an already mapped buffer. + */ + handle = ext4_journal_start(inode, 1); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + + page = __grab_cache_page(mapping, index); + if (!page) + return -ENOMEM; + *pagep = page; + + ret = block_write_begin(file, mapping, pos, len, flags, pagep, fsdata, + ext4_da_get_block_prep); + if (ret < 0) { + unlock_page(page); + ext4_journal_stop(handle); + page_cache_release(page); + } + + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; +out: + return ret; +} + +/* + * Check if we should update i_disksize + * when write to the end of file but not require block allocation + */ +static int ext4_da_should_update_i_disksize(struct page *page, + unsigned long offset) +{ + struct buffer_head *bh; + struct inode *inode = page->mapping->host; + unsigned int idx; + int i; + + bh = page_buffers(page); + idx = offset >> inode->i_blkbits; + + for (i=0; i < idx; i++) + bh = bh->b_this_page; + + if (!buffer_mapped(bh) || (buffer_delay(bh))) + return 0; + return 1; +} + +static int ext4_da_write_end(struct file *file, + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = mapping->host; + int ret = 0, ret2; + handle_t *handle = ext4_journal_current_handle(); + loff_t new_i_size; + unsigned long start, end; + + start = pos & (PAGE_CACHE_SIZE - 1); + end = start + copied -1; + + /* + * generic_write_end() will run mark_inode_dirty() if i_size + * changes. So let's piggyback the i_disksize mark_inode_dirty + * into that. + */ + + new_i_size = pos + copied; + if (new_i_size > EXT4_I(inode)->i_disksize) { + if (ext4_da_should_update_i_disksize(page, end)) { + down_write(&EXT4_I(inode)->i_data_sem); + if (new_i_size > EXT4_I(inode)->i_disksize) { + /* + * Updating i_disksize when extending file + * without needing block allocation + */ + if (ext4_should_order_data(inode)) + ret = ext4_jbd2_file_inode(handle, + inode); + + EXT4_I(inode)->i_disksize = new_i_size; + } + up_write(&EXT4_I(inode)->i_data_sem); + } + } + ret2 = generic_write_end(file, mapping, pos, len, copied, + page, fsdata); + copied = ret2; + if (ret2 < 0) + ret = ret2; + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + + return ret ? ret : copied; +} + +static void ext4_da_invalidatepage(struct page *page, unsigned long offset) +{ + /* + * Drop reserved blocks + */ + BUG_ON(!PageLocked(page)); + if (!page_has_buffers(page)) + goto out; + + ext4_da_page_release_reservation(page, offset); + +out: + ext4_invalidatepage(page, offset); + + return; +} + /* * bmap() is special. It gets used by applications such as lilo and by @@ -1418,6 +2409,16 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block) journal_t *journal; int err; + if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY) && + test_opt(inode->i_sb, DELALLOC)) { + /* + * With delalloc we want to sync the file + * so that we can make sure we allocate + * blocks for file + */ + filemap_write_and_wait(mapping); + } + if (EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { /* * This is a REALLY heavyweight approach, but the use of @@ -1462,21 +2463,17 @@ static int bput_one(handle_t *handle, struct buffer_head *bh) return 0; } -static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) -{ - if (buffer_mapped(bh)) - return ext4_journal_dirty_data(handle, bh); - return 0; -} - /* - * Note that we always start a transaction even if we're not journalling - * data. This is to preserve ordering: any hole instantiation within - * __block_write_full_page -> ext4_get_block() should be journalled - * along with the data so we don't crash and then get metadata which - * refers to old data. + * Note that we don't need to start a transaction unless we're journaling data + * because we should have holes filled from ext4_page_mkwrite(). We even don't + * need to file the inode to the transaction's list in ordered mode because if + * we are writing back data added by write(), the inode is already there and if + * we are writing back data modified via mmap(), noone guarantees in which + * transaction the data will hit the disk. In case we are journaling data, we + * cannot start transaction directly because transaction start ranks above page + * lock so we have to do some magic. * - * In all journalling modes block_write_full_page() will start the I/O. + * In all journaling modes block_write_full_page() will start the I/O. * * Problem: * @@ -1518,105 +2515,103 @@ static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) * disastrous. Any write() or metadata operation will sync the fs for * us. * - * AKPM2: if all the page's buffers are mapped to disk and !data=journal, - * we don't need to open a transaction here. */ -static int ext4_ordered_writepage(struct page *page, +static int __ext4_normal_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; - struct buffer_head *page_bufs; - handle_t *handle = NULL; - int ret = 0; - int err; - - J_ASSERT(PageLocked(page)); - - /* - * We give up here if we're reentered, because it might be for a - * different filesystem. - */ - if (ext4_journal_current_handle()) - goto out_fail; - handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); + if (test_opt(inode->i_sb, NOBH)) + return nobh_writepage(page, + ext4_normal_get_block_write, wbc); + else + return block_write_full_page(page, + ext4_normal_get_block_write, + wbc); +} - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out_fail; - } +static int ext4_normal_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + loff_t size = i_size_read(inode); + loff_t len; - if (!page_has_buffers(page)) { - create_empty_buffers(page, inode->i_sb->s_blocksize, - (1 << BH_Dirty)|(1 << BH_Uptodate)); + J_ASSERT(PageLocked(page)); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + + if (page_has_buffers(page)) { + /* if page has buffers it should all be mapped + * and allocated. If there are not buffers attached + * to the page we know the page is dirty but it lost + * buffers. That means that at some moment in time + * after write_begin() / write_end() has been called + * all buffers have been clean and thus they must have been + * written at least once. So they are all mapped and we can + * happily proceed with mapping them and writing the page. + */ + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); } - page_bufs = page_buffers(page); - walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, bget_one); - - ret = block_write_full_page(page, ext4_get_block, wbc); - /* - * The page can become unlocked at any point now, and - * truncate can then come in and change things. So we - * can't touch *page from now on. But *page_bufs is - * safe due to elevated refcount. - */ + if (!ext4_journal_current_handle()) + return __ext4_normal_writepage(page, wbc); - /* - * And attach them to the current transaction. But only if - * block_write_full_page() succeeded. Otherwise they are unmapped, - * and generally junk. - */ - if (ret == 0) { - err = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, - NULL, jbd2_journal_dirty_data_fn); - if (!ret) - ret = err; - } - walk_page_buffers(handle, page_bufs, 0, - PAGE_CACHE_SIZE, NULL, bput_one); - err = ext4_journal_stop(handle); - if (!ret) - ret = err; - return ret; - -out_fail: redirty_page_for_writepage(wbc, page); unlock_page(page); - return ret; + return 0; } -static int ext4_writeback_writepage(struct page *page, +static int __ext4_journalled_writepage(struct page *page, struct writeback_control *wbc) { - struct inode *inode = page->mapping->host; + struct address_space *mapping = page->mapping; + struct inode *inode = mapping->host; + struct buffer_head *page_bufs; handle_t *handle = NULL; int ret = 0; int err; - if (ext4_journal_current_handle()) - goto out_fail; + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, + ext4_normal_get_block_write); + if (ret != 0) + goto out_unlock; + + page_bufs = page_buffers(page); + walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, NULL, + bget_one); + /* As soon as we unlock the page, it can go away, but we have + * references to buffers so we are safe */ + unlock_page(page); handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); - goto out_fail; + goto out; } - if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) - ret = nobh_writepage(page, ext4_get_block, wbc); - else - ret = block_write_full_page(page, ext4_get_block, wbc); + ret = walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); + err = walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, write_end_fn); + if (ret == 0) + ret = err; err = ext4_journal_stop(handle); if (!ret) ret = err; - return ret; -out_fail: - redirty_page_for_writepage(wbc, page); + walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, bput_one); + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + goto out; + +out_unlock: unlock_page(page); +out: return ret; } @@ -1624,59 +2619,53 @@ static int ext4_journalled_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; - handle_t *handle = NULL; - int ret = 0; - int err; + loff_t size = i_size_read(inode); + loff_t len; - if (ext4_journal_current_handle()) - goto no_write; + J_ASSERT(PageLocked(page)); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + + if (page_has_buffers(page)) { + /* if page has buffers it should all be mapped + * and allocated. If there are not buffers attached + * to the page we know the page is dirty but it lost + * buffers. That means that at some moment in time + * after write_begin() / write_end() has been called + * all buffers have been clean and thus they must have been + * written at least once. So they are all mapped and we can + * happily proceed with mapping them and writing the page. + */ + BUG_ON(walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped_or_delay)); + } - handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); + if (ext4_journal_current_handle()) goto no_write; - } - if (!page_has_buffers(page) || PageChecked(page)) { + if (PageChecked(page)) { /* * It's mmapped pagecache. Add buffers and journal it. There * doesn't seem much point in redirtying the page here. */ ClearPageChecked(page); - ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, - ext4_get_block); - if (ret != 0) { - ext4_journal_stop(handle); - goto out_unlock; - } - ret = walk_page_buffers(handle, page_buffers(page), 0, - PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); - - err = walk_page_buffers(handle, page_buffers(page), 0, - PAGE_CACHE_SIZE, NULL, write_end_fn); - if (ret == 0) - ret = err; - EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; - unlock_page(page); + return __ext4_journalled_writepage(page, wbc); } else { /* * It may be a page full of checkpoint-mode buffers. We don't * really know unless we go poke around in the buffer_heads. * But block_write_full_page will do the right thing. */ - ret = block_write_full_page(page, ext4_get_block, wbc); + return block_write_full_page(page, + ext4_normal_get_block_write, + wbc); } - err = ext4_journal_stop(handle); - if (!ret) - ret = err; -out: - return ret; - no_write: redirty_page_for_writepage(wbc, page); -out_unlock: unlock_page(page); - goto out; + return 0; } static int ext4_readpage(struct file *file, struct page *page) @@ -1819,7 +2808,7 @@ static int ext4_journalled_set_page_dirty(struct page *page) static const struct address_space_operations ext4_ordered_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_ordered_writepage, + .writepage = ext4_normal_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_ordered_write_end, @@ -1833,7 +2822,7 @@ static const struct address_space_operations ext4_ordered_aops = { static const struct address_space_operations ext4_writeback_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, - .writepage = ext4_writeback_writepage, + .writepage = ext4_normal_writepage, .sync_page = block_sync_page, .write_begin = ext4_write_begin, .write_end = ext4_writeback_write_end, @@ -1857,10 +2846,31 @@ static const struct address_space_operations ext4_journalled_aops = { .releasepage = ext4_releasepage, }; +static const struct address_space_operations ext4_da_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_da_writepage, + .writepages = ext4_da_writepages, + .sync_page = block_sync_page, + .write_begin = ext4_da_write_begin, + .write_end = ext4_da_write_end, + .bmap = ext4_bmap, + .invalidatepage = ext4_da_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, +}; + void ext4_set_aops(struct inode *inode) { - if (ext4_should_order_data(inode)) + if (ext4_should_order_data(inode) && + test_opt(inode->i_sb, DELALLOC)) + inode->i_mapping->a_ops = &ext4_da_aops; + else if (ext4_should_order_data(inode)) inode->i_mapping->a_ops = &ext4_ordered_aops; + else if (ext4_should_writeback_data(inode) && + test_opt(inode->i_sb, DELALLOC)) + inode->i_mapping->a_ops = &ext4_da_aops; else if (ext4_should_writeback_data(inode)) inode->i_mapping->a_ops = &ext4_writeback_aops; else @@ -1873,7 +2883,7 @@ void ext4_set_aops(struct inode *inode) * This required during truncate. We need to physically zero the tail end * of that block so it doesn't yield old data if the file is later grown. */ -int ext4_block_truncate_page(handle_t *handle, struct page *page, +int ext4_block_truncate_page(handle_t *handle, struct address_space *mapping, loff_t from) { ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; @@ -1882,8 +2892,13 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page, ext4_lblk_t iblock; struct inode *inode = mapping->host; struct buffer_head *bh; + struct page *page; int err = 0; + page = grab_cache_page(mapping, from >> PAGE_CACHE_SHIFT); + if (!page) + return -EINVAL; + blocksize = inode->i_sb->s_blocksize; length = blocksize - (offset & (blocksize - 1)); iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); @@ -1956,7 +2971,7 @@ int ext4_block_truncate_page(handle_t *handle, struct page *page, err = ext4_journal_dirty_metadata(handle, bh); } else { if (ext4_should_order_data(inode)) - err = ext4_journal_dirty_data(handle, bh); + err = ext4_jbd2_file_inode(handle, inode); mark_buffer_dirty(bh); } @@ -2179,7 +3194,21 @@ static void ext4_free_data(handle_t *handle, struct inode *inode, if (this_bh) { BUFFER_TRACE(this_bh, "call ext4_journal_dirty_metadata"); - ext4_journal_dirty_metadata(handle, this_bh); + + /* + * The buffer head should have an attached journal head at this + * point. However, if the data is corrupted and an indirect + * block pointed to itself, it would have been detached when + * the block was cleared. Check for this instead of OOPSing. + */ + if (bh2jh(this_bh)) + ext4_journal_dirty_metadata(handle, this_bh); + else + ext4_error(inode->i_sb, __func__, + "circular indirect block detected, " + "inode=%lu, block=%llu", + inode->i_ino, + (unsigned long long) this_bh->b_blocknr); } } @@ -2305,6 +3334,19 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode, } } +int ext4_can_truncate(struct inode *inode) +{ + if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) + return 0; + if (S_ISREG(inode->i_mode)) + return 1; + if (S_ISDIR(inode->i_mode)) + return 1; + if (S_ISLNK(inode->i_mode)) + return !ext4_inode_is_fast_symlink(inode); + return 0; +} + /* * ext4_truncate() * @@ -2347,51 +3389,25 @@ void ext4_truncate(struct inode *inode) int n; ext4_lblk_t last_block; unsigned blocksize = inode->i_sb->s_blocksize; - struct page *page; - if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || - S_ISLNK(inode->i_mode))) - return; - if (ext4_inode_is_fast_symlink(inode)) - return; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) + if (!ext4_can_truncate(inode)) return; - /* - * We have to lock the EOF page here, because lock_page() nests - * outside jbd2_journal_start(). - */ - if ((inode->i_size & (blocksize - 1)) == 0) { - /* Block boundary? Nothing to do */ - page = NULL; - } else { - page = grab_cache_page(mapping, - inode->i_size >> PAGE_CACHE_SHIFT); - if (!page) - return; - } - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) { - ext4_ext_truncate(inode, page); + ext4_ext_truncate(inode); return; } handle = start_transaction(inode); - if (IS_ERR(handle)) { - if (page) { - clear_highpage(page); - flush_dcache_page(page); - unlock_page(page); - page_cache_release(page); - } + if (IS_ERR(handle)) return; /* AKPM: return what? */ - } last_block = (inode->i_size + blocksize-1) >> EXT4_BLOCK_SIZE_BITS(inode->i_sb); - if (page) - ext4_block_truncate_page(handle, page, mapping, inode->i_size); + if (inode->i_size & (blocksize - 1)) + if (ext4_block_truncate_page(handle, mapping, inode->i_size)) + goto out_stop; n = ext4_block_to_path(inode, last_block, offsets, NULL); if (n == 0) @@ -2410,6 +3426,11 @@ void ext4_truncate(struct inode *inode) goto out_stop; /* + * From here we block out all ext4_get_block() callers who want to + * modify the block allocation tree. + */ + down_write(&ei->i_data_sem); + /* * The orphan list entry will now protect us from any crash which * occurs before the truncate completes, so it is now safe to propagate * the new, shorter inode size (held for now in i_size) into the @@ -2418,12 +3439,6 @@ void ext4_truncate(struct inode *inode) */ ei->i_disksize = inode->i_size; - /* - * From here we block out all ext4_get_block() callers who want to - * modify the block allocation tree. - */ - down_write(&ei->i_data_sem); - if (n == 1) { /* direct blocks */ ext4_free_data(handle, inode, NULL, i_data+offsets[0], i_data + EXT4_NDIR_BLOCKS); @@ -3107,7 +4122,14 @@ int ext4_write_inode(struct inode *inode, int wait) * be freed, so we have a strong guarantee that no future commit will * leave these blocks visible to the user.) * - * Called with inode->sem down. + * Another thing we have to assure is that if we are in ordered mode + * and inode is still attached to the committing transaction, we must + * we start writeout of all the dirty pages which are being truncated. + * This way we are sure that all the data written in the previous + * transaction are already on disk (truncate waits for pages under + * writeback). + * + * Called with inode->i_mutex down. */ int ext4_setattr(struct dentry *dentry, struct iattr *attr) { @@ -3173,6 +4195,22 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) if (!error) error = rc; ext4_journal_stop(handle); + + if (ext4_should_order_data(inode)) { + error = ext4_begin_ordered_truncate(inode, + attr->ia_size); + if (error) { + /* Do as much error cleanup as possible */ + handle = ext4_journal_start(inode, 3); + if (IS_ERR(handle)) { + ext4_orphan_del(NULL, inode); + goto err_out; + } + ext4_orphan_del(handle, inode); + ext4_journal_stop(handle); + goto err_out; + } + } } rc = inode_setattr(inode, attr); @@ -3193,6 +4231,32 @@ err_out: return error; } +int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode; + unsigned long delalloc_blocks; + + inode = dentry->d_inode; + generic_fillattr(inode, stat); + + /* + * We can't update i_blocks if the block allocation is delayed + * otherwise in the case of system crash before the real block + * allocation is done, we will have i_blocks inconsistent with + * on-disk file blocks. + * We always keep i_blocks updated together with real + * allocation. But to not confuse with user, stat + * will return the blocks that include the delayed allocation + * blocks for this file. + */ + spin_lock(&EXT4_I(inode)->i_block_reservation_lock); + delalloc_blocks = EXT4_I(inode)->i_reserved_data_blocks; + spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + + stat->blocks += (delalloc_blocks << inode->i_sb->s_blocksize_bits)>>9; + return 0; +} /* * How many blocks doth make a writepage()? @@ -3506,3 +4570,64 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) return err; } + +static int ext4_bh_unmapped(handle_t *handle, struct buffer_head *bh) +{ + return !buffer_mapped(bh); +} + +int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page) +{ + loff_t size; + unsigned long len; + int ret = -EINVAL; + struct file *file = vma->vm_file; + struct inode *inode = file->f_path.dentry->d_inode; + struct address_space *mapping = inode->i_mapping; + + /* + * Get i_alloc_sem to stop truncates messing with the inode. We cannot + * get i_mutex because we are already holding mmap_sem. + */ + down_read(&inode->i_alloc_sem); + size = i_size_read(inode); + if (page->mapping != mapping || size <= page_offset(page) + || !PageUptodate(page)) { + /* page got truncated from under us? */ + goto out_unlock; + } + ret = 0; + if (PageMappedToDisk(page)) + goto out_unlock; + + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + + if (page_has_buffers(page)) { + /* return if we have all the buffers mapped */ + if (!walk_page_buffers(NULL, page_buffers(page), 0, len, NULL, + ext4_bh_unmapped)) + goto out_unlock; + } + /* + * OK, we need to fill the hole... Do write_begin write_end + * to do block allocation/reservation.We are not holding + * inode.i__mutex here. That allow * parallel write_begin, + * write_end call. lock_page prevent this from happening + * on the same page though + */ + ret = mapping->a_ops->write_begin(file, mapping, page_offset(page), + len, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL); + if (ret < 0) + goto out_unlock; + ret = mapping->a_ops->write_end(file, mapping, page_offset(page), + len, len, page, NULL); + if (ret < 0) + goto out_unlock; + ret = 0; +out_unlock: + up_read(&inode->i_alloc_sem); + return ret; +} diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index c9900aade150..8d141a25bbee 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -381,22 +381,28 @@ static inline void mb_clear_bit_atomic(spinlock_t *lock, int bit, void *addr) static inline int mb_find_next_zero_bit(void *addr, int max, int start) { - int fix = 0; + int fix = 0, ret, tmpmax; addr = mb_correct_addr_and_bit(&fix, addr); - max += fix; + tmpmax = max + fix; start += fix; - return ext4_find_next_zero_bit(addr, max, start) - fix; + ret = ext4_find_next_zero_bit(addr, tmpmax, start) - fix; + if (ret > max) + return max; + return ret; } static inline int mb_find_next_bit(void *addr, int max, int start) { - int fix = 0; + int fix = 0, ret, tmpmax; addr = mb_correct_addr_and_bit(&fix, addr); - max += fix; + tmpmax = max + fix; start += fix; - return ext4_find_next_bit(addr, max, start) - fix; + ret = ext4_find_next_bit(addr, tmpmax, start) - fix; + if (ret > max) + return max; + return ret; } static void *mb_find_buddy(struct ext4_buddy *e4b, int order, int *max) @@ -803,6 +809,7 @@ static int ext4_mb_init_cache(struct page *page, char *incore) if (!buffer_uptodate(bh[i])) goto out; + err = 0; first_block = page->index * blocks_per_page; for (i = 0; i < blocks_per_page; i++) { int group; @@ -883,6 +890,7 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, int pnum; int poff; struct page *page; + int ret; mb_debug("load group %lu\n", group); @@ -914,15 +922,21 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, if (page) { BUG_ON(page->mapping != inode->i_mapping); if (!PageUptodate(page)) { - ext4_mb_init_cache(page, NULL); + ret = ext4_mb_init_cache(page, NULL); + if (ret) { + unlock_page(page); + goto err; + } mb_cmp_bitmaps(e4b, page_address(page) + (poff * sb->s_blocksize)); } unlock_page(page); } } - if (page == NULL || !PageUptodate(page)) + if (page == NULL || !PageUptodate(page)) { + ret = -EIO; goto err; + } e4b->bd_bitmap_page = page; e4b->bd_bitmap = page_address(page) + (poff * sb->s_blocksize); mark_page_accessed(page); @@ -938,14 +952,20 @@ ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group, page = find_or_create_page(inode->i_mapping, pnum, GFP_NOFS); if (page) { BUG_ON(page->mapping != inode->i_mapping); - if (!PageUptodate(page)) - ext4_mb_init_cache(page, e4b->bd_bitmap); - + if (!PageUptodate(page)) { + ret = ext4_mb_init_cache(page, e4b->bd_bitmap); + if (ret) { + unlock_page(page); + goto err; + } + } unlock_page(page); } } - if (page == NULL || !PageUptodate(page)) + if (page == NULL || !PageUptodate(page)) { + ret = -EIO; goto err; + } e4b->bd_buddy_page = page; e4b->bd_buddy = page_address(page) + (poff * sb->s_blocksize); mark_page_accessed(page); @@ -962,7 +982,7 @@ err: page_cache_release(e4b->bd_buddy_page); e4b->bd_buddy = NULL; e4b->bd_bitmap = NULL; - return -EIO; + return ret; } static void ext4_mb_release_desc(struct ext4_buddy *e4b) @@ -1031,7 +1051,7 @@ static void mb_set_bits(spinlock_t *lock, void *bm, int cur, int len) } } -static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, +static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, int first, int count) { int block = 0; @@ -1071,11 +1091,12 @@ static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, blocknr += block; blocknr += le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block); - + ext4_unlock_group(sb, e4b->bd_group); ext4_error(sb, __func__, "double-free of inode" " %lu's block %llu(bit %u in group %lu)\n", inode ? inode->i_ino : 0, blocknr, block, e4b->bd_group); + ext4_lock_group(sb, e4b->bd_group); } mb_clear_bit(block, EXT4_MB_BITMAP(e4b)); e4b->bd_info->bb_counters[order]++; @@ -1113,8 +1134,6 @@ static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b, } while (1); } mb_check_buddy(e4b); - - return 0; } static int mb_find_extent(struct ext4_buddy *e4b, int order, int block, @@ -1730,10 +1749,6 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac) ac->ac_g_ex.fe_start = sbi->s_mb_last_start; spin_unlock(&sbi->s_md_lock); } - - /* searching for the right group start from the goal value specified */ - group = ac->ac_g_ex.fe_group; - /* Let's just scan groups to find more-less suitable blocks */ cr = ac->ac_2order ? 0 : 1; /* @@ -1743,6 +1758,12 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac) repeat: for (; cr < 4 && ac->ac_status == AC_STATUS_CONTINUE; cr++) { ac->ac_criteria = cr; + /* + * searching for the right group start + * from the goal value specified + */ + group = ac->ac_g_ex.fe_group; + for (i = 0; i < EXT4_SB(sb)->s_groups_count; group++, i++) { struct ext4_group_info *grp; struct ext4_group_desc *desc; @@ -1963,6 +1984,8 @@ static int ext4_mb_seq_history_open(struct inode *inode, struct file *file) int rc; int size; + if (unlikely(sbi->s_mb_history == NULL)) + return -ENOMEM; s = kmalloc(sizeof(*s), GFP_KERNEL); if (s == NULL) return -ENOMEM; @@ -2165,9 +2188,7 @@ static void ext4_mb_history_init(struct super_block *sb) sbi->s_mb_history_cur = 0; spin_lock_init(&sbi->s_mb_history_lock); i = sbi->s_mb_history_max * sizeof(struct ext4_mb_history); - sbi->s_mb_history = kmalloc(i, GFP_KERNEL); - if (likely(sbi->s_mb_history != NULL)) - memset(sbi->s_mb_history, 0, i); + sbi->s_mb_history = kzalloc(i, GFP_KERNEL); /* if we can't allocate history, then we simple won't use it */ } @@ -2215,21 +2236,192 @@ ext4_mb_store_history(struct ext4_allocation_context *ac) #define ext4_mb_history_init(sb) #endif + +/* Create and initialize ext4_group_info data for the given group. */ +int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group, + struct ext4_group_desc *desc) +{ + int i, len; + int metalen = 0; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_group_info **meta_group_info; + + /* + * First check if this group is the first of a reserved block. + * If it's true, we have to allocate a new table of pointers + * to ext4_group_info structures + */ + if (group % EXT4_DESC_PER_BLOCK(sb) == 0) { + metalen = sizeof(*meta_group_info) << + EXT4_DESC_PER_BLOCK_BITS(sb); + meta_group_info = kmalloc(metalen, GFP_KERNEL); + if (meta_group_info == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate mem for a " + "buddy group\n"); + goto exit_meta_group_info; + } + sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)] = + meta_group_info; + } + + /* + * calculate needed size. if change bb_counters size, + * don't forget about ext4_mb_generate_buddy() + */ + len = offsetof(typeof(**meta_group_info), + bb_counters[sb->s_blocksize_bits + 2]); + + meta_group_info = + sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]; + i = group & (EXT4_DESC_PER_BLOCK(sb) - 1); + + meta_group_info[i] = kzalloc(len, GFP_KERNEL); + if (meta_group_info[i] == NULL) { + printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n"); + goto exit_group_info; + } + set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, + &(meta_group_info[i]->bb_state)); + + /* + * initialize bb_free to be able to skip + * empty groups without initialization + */ + if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { + meta_group_info[i]->bb_free = + ext4_free_blocks_after_init(sb, group, desc); + } else { + meta_group_info[i]->bb_free = + le16_to_cpu(desc->bg_free_blocks_count); + } + + INIT_LIST_HEAD(&meta_group_info[i]->bb_prealloc_list); + +#ifdef DOUBLE_CHECK + { + struct buffer_head *bh; + meta_group_info[i]->bb_bitmap = + kmalloc(sb->s_blocksize, GFP_KERNEL); + BUG_ON(meta_group_info[i]->bb_bitmap == NULL); + bh = ext4_read_block_bitmap(sb, group); + BUG_ON(bh == NULL); + memcpy(meta_group_info[i]->bb_bitmap, bh->b_data, + sb->s_blocksize); + put_bh(bh); + } +#endif + + return 0; + +exit_group_info: + /* If a meta_group_info table has been allocated, release it now */ + if (group % EXT4_DESC_PER_BLOCK(sb) == 0) + kfree(sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]); +exit_meta_group_info: + return -ENOMEM; +} /* ext4_mb_add_groupinfo */ + +/* + * Add a group to the existing groups. + * This function is used for online resize + */ +int ext4_mb_add_more_groupinfo(struct super_block *sb, ext4_group_t group, + struct ext4_group_desc *desc) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct inode *inode = sbi->s_buddy_cache; + int blocks_per_page; + int block; + int pnum; + struct page *page; + int err; + + /* Add group based on group descriptor*/ + err = ext4_mb_add_groupinfo(sb, group, desc); + if (err) + return err; + + /* + * Cache pages containing dynamic mb_alloc datas (buddy and bitmap + * datas) are set not up to date so that they will be re-initilaized + * during the next call to ext4_mb_load_buddy + */ + + /* Set buddy page as not up to date */ + blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; + block = group * 2; + pnum = block / blocks_per_page; + page = find_get_page(inode->i_mapping, pnum); + if (page != NULL) { + ClearPageUptodate(page); + page_cache_release(page); + } + + /* Set bitmap page as not up to date */ + block++; + pnum = block / blocks_per_page; + page = find_get_page(inode->i_mapping, pnum); + if (page != NULL) { + ClearPageUptodate(page); + page_cache_release(page); + } + + return 0; +} + +/* + * Update an existing group. + * This function is used for online resize + */ +void ext4_mb_update_group_info(struct ext4_group_info *grp, ext4_grpblk_t add) +{ + grp->bb_free += add; +} + static int ext4_mb_init_backend(struct super_block *sb) { ext4_group_t i; - int j, len, metalen; + int metalen; struct ext4_sb_info *sbi = EXT4_SB(sb); - int num_meta_group_infos = - (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) >> - EXT4_DESC_PER_BLOCK_BITS(sb); + struct ext4_super_block *es = sbi->s_es; + int num_meta_group_infos; + int num_meta_group_infos_max; + int array_size; struct ext4_group_info **meta_group_info; + struct ext4_group_desc *desc; + + /* This is the number of blocks used by GDT */ + num_meta_group_infos = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - + 1) >> EXT4_DESC_PER_BLOCK_BITS(sb); + + /* + * This is the total number of blocks used by GDT including + * the number of reserved blocks for GDT. + * The s_group_info array is allocated with this value + * to allow a clean online resize without a complex + * manipulation of pointer. + * The drawback is the unused memory when no resize + * occurs but it's very low in terms of pages + * (see comments below) + * Need to handle this properly when META_BG resizing is allowed + */ + num_meta_group_infos_max = num_meta_group_infos + + le16_to_cpu(es->s_reserved_gdt_blocks); + /* + * array_size is the size of s_group_info array. We round it + * to the next power of two because this approximation is done + * internally by kmalloc so we can have some more memory + * for free here (e.g. may be used for META_BG resize). + */ + array_size = 1; + while (array_size < sizeof(*sbi->s_group_info) * + num_meta_group_infos_max) + array_size = array_size << 1; /* An 8TB filesystem with 64-bit pointers requires a 4096 byte * kmalloc. A 128kb malloc should suffice for a 256TB filesystem. * So a two level scheme suffices for now. */ - sbi->s_group_info = kmalloc(sizeof(*sbi->s_group_info) * - num_meta_group_infos, GFP_KERNEL); + sbi->s_group_info = kmalloc(array_size, GFP_KERNEL); if (sbi->s_group_info == NULL) { printk(KERN_ERR "EXT4-fs: can't allocate buddy meta group\n"); return -ENOMEM; @@ -2256,63 +2448,15 @@ static int ext4_mb_init_backend(struct super_block *sb) sbi->s_group_info[i] = meta_group_info; } - /* - * calculate needed size. if change bb_counters size, - * don't forget about ext4_mb_generate_buddy() - */ - len = sizeof(struct ext4_group_info); - len += sizeof(unsigned short) * (sb->s_blocksize_bits + 2); for (i = 0; i < sbi->s_groups_count; i++) { - struct ext4_group_desc *desc; - - meta_group_info = - sbi->s_group_info[i >> EXT4_DESC_PER_BLOCK_BITS(sb)]; - j = i & (EXT4_DESC_PER_BLOCK(sb) - 1); - - meta_group_info[j] = kzalloc(len, GFP_KERNEL); - if (meta_group_info[j] == NULL) { - printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n"); - goto err_freebuddy; - } desc = ext4_get_group_desc(sb, i, NULL); if (desc == NULL) { printk(KERN_ERR "EXT4-fs: can't read descriptor %lu\n", i); - i++; goto err_freebuddy; } - memset(meta_group_info[j], 0, len); - set_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, - &(meta_group_info[j]->bb_state)); - - /* - * initialize bb_free to be able to skip - * empty groups without initialization - */ - if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) { - meta_group_info[j]->bb_free = - ext4_free_blocks_after_init(sb, i, desc); - } else { - meta_group_info[j]->bb_free = - le16_to_cpu(desc->bg_free_blocks_count); - } - - INIT_LIST_HEAD(&meta_group_info[j]->bb_prealloc_list); - -#ifdef DOUBLE_CHECK - { - struct buffer_head *bh; - meta_group_info[j]->bb_bitmap = - kmalloc(sb->s_blocksize, GFP_KERNEL); - BUG_ON(meta_group_info[j]->bb_bitmap == NULL); - bh = read_block_bitmap(sb, i); - BUG_ON(bh == NULL); - memcpy(meta_group_info[j]->bb_bitmap, bh->b_data, - sb->s_blocksize); - put_bh(bh); - } -#endif - + if (ext4_mb_add_groupinfo(sb, i, desc) != 0) + goto err_freebuddy; } return 0; @@ -2336,6 +2480,7 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) unsigned i; unsigned offset; unsigned max; + int ret; if (!test_opt(sb, MBALLOC)) return 0; @@ -2370,12 +2515,12 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) } while (i <= sb->s_blocksize_bits + 1); /* init file for buddy data */ - i = ext4_mb_init_backend(sb); - if (i) { + ret = ext4_mb_init_backend(sb); + if (ret != 0) { clear_opt(sbi->s_mount_opt, MBALLOC); kfree(sbi->s_mb_offsets); kfree(sbi->s_mb_maxs); - return i; + return ret; } spin_lock_init(&sbi->s_md_lock); @@ -2548,8 +2693,7 @@ ext4_mb_free_committed_blocks(struct super_block *sb) ext4_lock_group(sb, md->group); for (i = 0; i < md->num; i++) { mb_debug(" %u", md->blocks[i]); - err = mb_free_blocks(NULL, &e4b, md->blocks[i], 1); - BUG_ON(err != 0); + mb_free_blocks(NULL, &e4b, md->blocks[i], 1); } mb_debug("\n"); ext4_unlock_group(sb, md->group); @@ -2575,25 +2719,24 @@ ext4_mb_free_committed_blocks(struct super_block *sb) -#define MB_PROC_VALUE_READ(name) \ -static int ext4_mb_read_##name(char *page, char **start, \ - off_t off, int count, int *eof, void *data) \ +#define MB_PROC_FOPS(name) \ +static int ext4_mb_##name##_proc_show(struct seq_file *m, void *v) \ { \ - struct ext4_sb_info *sbi = data; \ - int len; \ - *eof = 1; \ - if (off != 0) \ - return 0; \ - len = sprintf(page, "%ld\n", sbi->s_mb_##name); \ - *start = page; \ - return len; \ -} - -#define MB_PROC_VALUE_WRITE(name) \ -static int ext4_mb_write_##name(struct file *file, \ - const char __user *buf, unsigned long cnt, void *data) \ + struct ext4_sb_info *sbi = m->private; \ + \ + seq_printf(m, "%ld\n", sbi->s_mb_##name); \ + return 0; \ +} \ + \ +static int ext4_mb_##name##_proc_open(struct inode *inode, struct file *file)\ +{ \ + return single_open(file, ext4_mb_##name##_proc_show, PDE(inode)->data);\ +} \ + \ +static ssize_t ext4_mb_##name##_proc_write(struct file *file, \ + const char __user *buf, size_t cnt, loff_t *ppos) \ { \ - struct ext4_sb_info *sbi = data; \ + struct ext4_sb_info *sbi = PDE(file->f_path.dentry->d_inode)->data;\ char str[32]; \ long value; \ if (cnt >= sizeof(str)) \ @@ -2605,31 +2748,32 @@ static int ext4_mb_write_##name(struct file *file, \ return -ERANGE; \ sbi->s_mb_##name = value; \ return cnt; \ -} +} \ + \ +static const struct file_operations ext4_mb_##name##_proc_fops = { \ + .owner = THIS_MODULE, \ + .open = ext4_mb_##name##_proc_open, \ + .read = seq_read, \ + .llseek = seq_lseek, \ + .release = single_release, \ + .write = ext4_mb_##name##_proc_write, \ +}; -MB_PROC_VALUE_READ(stats); -MB_PROC_VALUE_WRITE(stats); -MB_PROC_VALUE_READ(max_to_scan); -MB_PROC_VALUE_WRITE(max_to_scan); -MB_PROC_VALUE_READ(min_to_scan); -MB_PROC_VALUE_WRITE(min_to_scan); -MB_PROC_VALUE_READ(order2_reqs); -MB_PROC_VALUE_WRITE(order2_reqs); -MB_PROC_VALUE_READ(stream_request); -MB_PROC_VALUE_WRITE(stream_request); -MB_PROC_VALUE_READ(group_prealloc); -MB_PROC_VALUE_WRITE(group_prealloc); +MB_PROC_FOPS(stats); +MB_PROC_FOPS(max_to_scan); +MB_PROC_FOPS(min_to_scan); +MB_PROC_FOPS(order2_reqs); +MB_PROC_FOPS(stream_request); +MB_PROC_FOPS(group_prealloc); #define MB_PROC_HANDLER(name, var) \ do { \ - proc = create_proc_entry(name, mode, sbi->s_mb_proc); \ + proc = proc_create_data(name, mode, sbi->s_mb_proc, \ + &ext4_mb_##var##_proc_fops, sbi); \ if (proc == NULL) { \ printk(KERN_ERR "EXT4-fs: can't to create %s\n", name); \ goto err_out; \ } \ - proc->data = sbi; \ - proc->read_proc = ext4_mb_read_##var ; \ - proc->write_proc = ext4_mb_write_##var; \ } while (0) static int ext4_mb_init_per_dev_proc(struct super_block *sb) @@ -2639,6 +2783,10 @@ static int ext4_mb_init_per_dev_proc(struct super_block *sb) struct proc_dir_entry *proc; char devname[64]; + if (proc_root_ext4 == NULL) { + sbi->s_mb_proc = NULL; + return -EINVAL; + } bdevname(sb->s_bdev, devname); sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4); @@ -2747,7 +2895,7 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, err = -EIO; - bitmap_bh = read_block_bitmap(sb, ac->ac_b_ex.fe_group); + bitmap_bh = ext4_read_block_bitmap(sb, ac->ac_b_ex.fe_group); if (!bitmap_bh) goto out_err; @@ -2816,7 +2964,23 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len); gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp); spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group)); - percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len); + + /* + * free blocks account has already be reduced/reserved + * at write_begin() time for delayed allocation + * do not double accounting + */ + if (!(ac->ac_flags & EXT4_MB_DELALLOC_RESERVED)) + percpu_counter_sub(&sbi->s_freeblocks_counter, + ac->ac_b_ex.fe_len); + + if (sbi->s_log_groups_per_flex) { + ext4_group_t flex_group = ext4_flex_group(sbi, + ac->ac_b_ex.fe_group); + spin_lock(sb_bgl_lock(sbi, flex_group)); + sbi->s_flex_groups[flex_group].free_blocks -= ac->ac_b_ex.fe_len; + spin_unlock(sb_bgl_lock(sbi, flex_group)); + } err = ext4_journal_dirty_metadata(handle, bitmap_bh); if (err) @@ -3473,8 +3637,6 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, if (bit >= end) break; next = mb_find_next_bit(bitmap_bh->b_data, end, bit); - if (next > end) - next = end; start = group * EXT4_BLOCKS_PER_GROUP(sb) + bit + le32_to_cpu(sbi->s_es->s_first_data_block); mb_debug(" free preallocated %u/%u in group %u\n", @@ -3569,7 +3731,7 @@ ext4_mb_discard_group_preallocations(struct super_block *sb, if (list_empty(&grp->bb_prealloc_list)) return 0; - bitmap_bh = read_block_bitmap(sb, group); + bitmap_bh = ext4_read_block_bitmap(sb, group); if (bitmap_bh == NULL) { /* error handling here */ ext4_mb_release_desc(&e4b); @@ -3743,7 +3905,7 @@ repeat: err = ext4_mb_load_buddy(sb, group, &e4b); BUG_ON(err != 0); /* error handling here */ - bitmap_bh = read_block_bitmap(sb, group); + bitmap_bh = ext4_read_block_bitmap(sb, group); if (bitmap_bh == NULL) { /* error handling here */ ext4_mb_release_desc(&e4b); @@ -4011,10 +4173,21 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, sbi = EXT4_SB(sb); if (!test_opt(sb, MBALLOC)) { - block = ext4_new_blocks_old(handle, ar->inode, ar->goal, + block = ext4_old_new_blocks(handle, ar->inode, ar->goal, &(ar->len), errp); return block; } + if (!EXT4_I(ar->inode)->i_delalloc_reserved_flag) { + /* + * With delalloc we already reserved the blocks + */ + ar->len = ext4_has_free_blocks(sbi, ar->len); + } + + if (ar->len == 0) { + *errp = -ENOSPC; + return 0; + } while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) { ar->flags |= EXT4_MB_HINT_NOPREALLOC; @@ -4026,10 +4199,14 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, } inquota = ar->len; + if (EXT4_I(ar->inode)->i_delalloc_reserved_flag) + ar->flags |= EXT4_MB_DELALLOC_RESERVED; + ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS); if (!ac) { + ar->len = 0; *errp = -ENOMEM; - return 0; + goto out1; } ext4_mb_poll_new_transaction(sb, handle); @@ -4037,12 +4214,11 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, *errp = ext4_mb_initialize_context(ac, ar); if (*errp) { ar->len = 0; - goto out; + goto out2; } ac->ac_op = EXT4_MB_HISTORY_PREALLOC; if (!ext4_mb_use_preallocated(ac)) { - ac->ac_op = EXT4_MB_HISTORY_ALLOC; ext4_mb_normalize_request(ac, ar); repeat: @@ -4085,11 +4261,12 @@ repeat: ext4_mb_release_context(ac); -out: +out2: + kmem_cache_free(ext4_ac_cachep, ac); +out1: if (ar->len < inquota) DQUOT_FREE_BLOCK(ar->inode, inquota - ar->len); - kmem_cache_free(ext4_ac_cachep, ac); return block; } static void ext4_mb_poll_new_transaction(struct super_block *sb, @@ -4242,7 +4419,7 @@ do_more: overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb); count -= overflow; } - bitmap_bh = read_block_bitmap(sb, block_group); + bitmap_bh = ext4_read_block_bitmap(sb, block_group); if (!bitmap_bh) goto error_return; gdp = ext4_get_group_desc(sb, block_group, &gd_bh); @@ -4309,10 +4486,9 @@ do_more: ext4_mb_free_metadata(handle, &e4b, block_group, bit, count); } else { ext4_lock_group(sb, block_group); - err = mb_free_blocks(inode, &e4b, bit, count); + mb_free_blocks(inode, &e4b, bit, count); ext4_mb_return_to_preallocation(inode, &e4b, block, count); ext4_unlock_group(sb, block_group); - BUG_ON(err != 0); } spin_lock(sb_bgl_lock(sbi, block_group)); @@ -4321,6 +4497,13 @@ do_more: spin_unlock(sb_bgl_lock(sbi, block_group)); percpu_counter_add(&sbi->s_freeblocks_counter, count); + if (sbi->s_log_groups_per_flex) { + ext4_group_t flex_group = ext4_flex_group(sbi, block_group); + spin_lock(sb_bgl_lock(sbi, flex_group)); + sbi->s_flex_groups[flex_group].free_blocks += count; + spin_unlock(sb_bgl_lock(sbi, flex_group)); + } + ext4_mb_release_desc(&e4b); *freed += count; diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index ab16beaa830d..387ad98350c3 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -183,6 +183,16 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, struct inode *inode); /* + * p is at least 6 bytes before the end of page + */ +static inline struct ext4_dir_entry_2 * +ext4_next_entry(struct ext4_dir_entry_2 *p) +{ + return (struct ext4_dir_entry_2 *)((char *)p + + ext4_rec_len_from_disk(p->rec_len)); +} + +/* * Future: use high four bits of block for coalesce-on-delete flags * Mask them off for now. */ @@ -231,13 +241,13 @@ static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize) { unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1) - EXT4_DIR_REC_LEN(2) - infosize; - return 0? 20: entry_space / sizeof(struct dx_entry); + return entry_space / sizeof(struct dx_entry); } static inline unsigned dx_node_limit (struct inode *dir) { unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0); - return 0? 22: entry_space / sizeof(struct dx_entry); + return entry_space / sizeof(struct dx_entry); } /* @@ -554,15 +564,6 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash, /* - * p is at least 6 bytes before the end of page - */ -static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 *p) -{ - return (struct ext4_dir_entry_2 *)((char *)p + - ext4_rec_len_from_disk(p->rec_len)); -} - -/* * This function fills a red-black tree with information from a * directory block. It returns the number directory entries loaded * into the tree. If there is an error it is returned in err. @@ -993,19 +994,21 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, de = (struct ext4_dir_entry_2 *) bh->b_data; top = (struct ext4_dir_entry_2 *) ((char *) de + sb->s_blocksize - EXT4_DIR_REC_LEN(0)); - for (; de < top; de = ext4_next_entry(de)) - if (ext4_match (namelen, name, de)) { - if (!ext4_check_dir_entry("ext4_find_entry", - dir, de, bh, - (block<<EXT4_BLOCK_SIZE_BITS(sb)) - +((char *)de - bh->b_data))) { - brelse (bh); + for (; de < top; de = ext4_next_entry(de)) { + int off = (block << EXT4_BLOCK_SIZE_BITS(sb)) + + ((char *) de - bh->b_data); + + if (!ext4_check_dir_entry(__func__, dir, de, bh, off)) { + brelse(bh); *err = ERR_BAD_DX_DIR; goto errout; } - *res_dir = de; - dx_release (frames); - return bh; + + if (ext4_match(namelen, name, de)) { + *res_dir = de; + dx_release(frames); + return bh; + } } brelse (bh); /* Check to see if we should continue to search */ diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 9ff7b1c04239..f000fbe2cd93 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -866,6 +866,15 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) gdp->bg_checksum = ext4_group_desc_csum(sbi, input->group, gdp); /* + * We can allocate memory for mb_alloc based on the new group + * descriptor + */ + if (test_opt(sb, MBALLOC)) { + err = ext4_mb_add_more_groupinfo(sb, input->group, gdp); + if (err) + goto exit_journal; + } + /* * Make the new blocks and inodes valid next. We do this before * increasing the group count so that once the group is enabled, * all of its blocks and inodes are already valid. @@ -957,6 +966,8 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, handle_t *handle; int err; unsigned long freed_blocks; + ext4_group_t group; + struct ext4_group_info *grp; /* We don't need to worry about locking wrt other resizers just * yet: we're going to revalidate es->s_blocks_count after @@ -988,7 +999,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, } /* Handle the remaining blocks in the last group only. */ - ext4_get_group_no_and_offset(sb, o_blocks_count, NULL, &last); + ext4_get_group_no_and_offset(sb, o_blocks_count, &group, &last); if (last == 0) { ext4_warning(sb, __func__, @@ -1060,6 +1071,45 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, o_blocks_count + add); if ((err = ext4_journal_stop(handle))) goto exit_put; + + /* + * Mark mballoc pages as not up to date so that they will be updated + * next time they are loaded by ext4_mb_load_buddy. + */ + if (test_opt(sb, MBALLOC)) { + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct inode *inode = sbi->s_buddy_cache; + int blocks_per_page; + int block; + int pnum; + struct page *page; + + /* Set buddy page as not up to date */ + blocks_per_page = PAGE_CACHE_SIZE / sb->s_blocksize; + block = group * 2; + pnum = block / blocks_per_page; + page = find_get_page(inode->i_mapping, pnum); + if (page != NULL) { + ClearPageUptodate(page); + page_cache_release(page); + } + + /* Set bitmap page as not up to date */ + block++; + pnum = block / blocks_per_page; + page = find_get_page(inode->i_mapping, pnum); + if (page != NULL) { + ClearPageUptodate(page); + page_cache_release(page); + } + + /* Get the info on the last group */ + grp = ext4_get_group_info(sb, group); + + /* Update free blocks in group info */ + ext4_mb_update_group_info(grp, add); + } + if (test_opt(sb, DEBUG)) printk(KERN_DEBUG "EXT4-fs: extended group to %llu blocks\n", ext4_blocks_count(es)); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 02bf24343979..1cb371dcd609 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -506,6 +506,7 @@ static void ext4_put_super (struct super_block * sb) ext4_ext_release(sb); ext4_xattr_put_super(sb); jbd2_journal_destroy(sbi->s_journal); + sbi->s_journal = NULL; if (!(sb->s_flags & MS_RDONLY)) { EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); es->s_state = cpu_to_le16(sbi->s_mount_state); @@ -517,6 +518,7 @@ static void ext4_put_super (struct super_block * sb) for (i = 0; i < sbi->s_gdb_count; i++) brelse(sbi->s_group_desc[i]); kfree(sbi->s_group_desc); + kfree(sbi->s_flex_groups); percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); @@ -571,6 +573,12 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); INIT_LIST_HEAD(&ei->i_prealloc_list); spin_lock_init(&ei->i_prealloc_lock); + jbd2_journal_init_jbd_inode(&ei->jinode, &ei->vfs_inode); + ei->i_reserved_data_blocks = 0; + ei->i_reserved_meta_blocks = 0; + ei->i_allocated_meta_blocks = 0; + ei->i_delalloc_reserved_flag = 0; + spin_lock_init(&(ei->i_block_reservation_lock)); return &ei->vfs_inode; } @@ -635,6 +643,8 @@ static void ext4_clear_inode(struct inode *inode) EXT4_I(inode)->i_block_alloc_info = NULL; if (unlikely(rsv)) kfree(rsv); + jbd2_journal_release_jbd_inode(EXT4_SB(inode->i_sb)->s_journal, + &EXT4_I(inode)->jinode); } static inline void ext4_show_quota_options(struct seq_file *seq, struct super_block *sb) @@ -671,7 +681,6 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) unsigned long def_mount_opts; struct super_block *sb = vfs->mnt_sb; struct ext4_sb_info *sbi = EXT4_SB(sb); - journal_t *journal = sbi->s_journal; struct ext4_super_block *es = sbi->s_es; def_mount_opts = le32_to_cpu(es->s_default_mount_opts); @@ -747,6 +756,9 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_puts(seq, ",nomballoc"); if (test_opt(sb, I_VERSION)) seq_puts(seq, ",i_version"); + if (!test_opt(sb, DELALLOC)) + seq_puts(seq, ",nodelalloc"); + if (sbi->s_stripe) seq_printf(seq, ",stripe=%lu", sbi->s_stripe); @@ -894,7 +906,7 @@ enum { Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version, - Opt_mballoc, Opt_nomballoc, Opt_stripe, + Opt_mballoc, Opt_nomballoc, Opt_stripe, Opt_delalloc, Opt_nodelalloc, }; static match_table_t tokens = { @@ -953,6 +965,8 @@ static match_table_t tokens = { {Opt_nomballoc, "nomballoc"}, {Opt_stripe, "stripe=%u"}, {Opt_resize, "resize"}, + {Opt_delalloc, "delalloc"}, + {Opt_nodelalloc, "nodelalloc"}, {Opt_err, NULL}, }; @@ -990,6 +1004,7 @@ static int parse_options (char *options, struct super_block *sb, int qtype, qfmt; char *qname; #endif + ext4_fsblk_t last_block; if (!options) return 1; @@ -1309,15 +1324,39 @@ set_qf_format: clear_opt(sbi->s_mount_opt, NOBH); break; case Opt_extents: + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_EXTENTS)) { + ext4_warning(sb, __func__, + "extents feature not enabled " + "on this filesystem, use tune2fs\n"); + return 0; + } set_opt (sbi->s_mount_opt, EXTENTS); break; case Opt_noextents: + /* + * When e2fsprogs support resizing an already existing + * ext3 file system to greater than 2**32 we need to + * add support to block allocator to handle growing + * already existing block mapped inode so that blocks + * allocated for them fall within 2**32 + */ + last_block = ext4_blocks_count(sbi->s_es) - 1; + if (last_block > 0xffffffffULL) { + printk(KERN_ERR "EXT4-fs: Filesystem too " + "large to mount with " + "-o noextents options\n"); + return 0; + } clear_opt (sbi->s_mount_opt, EXTENTS); break; case Opt_i_version: set_opt(sbi->s_mount_opt, I_VERSION); sb->s_flags |= MS_I_VERSION; break; + case Opt_nodelalloc: + clear_opt(sbi->s_mount_opt, DELALLOC); + break; case Opt_mballoc: set_opt(sbi->s_mount_opt, MBALLOC); break; @@ -1331,6 +1370,9 @@ set_qf_format: return 0; sbi->s_stripe = option; break; + case Opt_delalloc: + set_opt(sbi->s_mount_opt, DELALLOC); + break; default: printk (KERN_ERR "EXT4-fs: Unrecognized mount option \"%s\" " @@ -1443,6 +1485,54 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, return res; } +static int ext4_fill_flex_info(struct super_block *sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_group_desc *gdp = NULL; + struct buffer_head *bh; + ext4_group_t flex_group_count; + ext4_group_t flex_group; + int groups_per_flex = 0; + __u64 block_bitmap = 0; + int i; + + if (!sbi->s_es->s_log_groups_per_flex) { + sbi->s_log_groups_per_flex = 0; + return 1; + } + + sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex; + groups_per_flex = 1 << sbi->s_log_groups_per_flex; + + flex_group_count = (sbi->s_groups_count + groups_per_flex - 1) / + groups_per_flex; + sbi->s_flex_groups = kmalloc(flex_group_count * + sizeof(struct flex_groups), GFP_KERNEL); + if (sbi->s_flex_groups == NULL) { + printk(KERN_ERR "EXT4-fs: not enough memory\n"); + goto failed; + } + memset(sbi->s_flex_groups, 0, flex_group_count * + sizeof(struct flex_groups)); + + gdp = ext4_get_group_desc(sb, 1, &bh); + block_bitmap = ext4_block_bitmap(sb, gdp) - 1; + + for (i = 0; i < sbi->s_groups_count; i++) { + gdp = ext4_get_group_desc(sb, i, &bh); + + flex_group = ext4_flex_group(sbi, i); + sbi->s_flex_groups[flex_group].free_inodes += + le16_to_cpu(gdp->bg_free_inodes_count); + sbi->s_flex_groups[flex_group].free_blocks += + le16_to_cpu(gdp->bg_free_blocks_count); + } + + return 1; +failed: + return 0; +} + __le16 ext4_group_desc_csum(struct ext4_sb_info *sbi, __u32 block_group, struct ext4_group_desc *gdp) { @@ -1810,8 +1900,8 @@ static unsigned long ext4_get_stripe_size(struct ext4_sb_info *sbi) } static int ext4_fill_super (struct super_block *sb, void *data, int silent) - __releases(kernel_sem) - __acquires(kernel_sem) + __releases(kernel_lock) + __acquires(kernel_lock) { struct buffer_head * bh; @@ -1851,11 +1941,6 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) goto out_fail; } - if (!sb_set_blocksize(sb, blocksize)) { - printk(KERN_ERR "EXT4-fs: bad blocksize %d.\n", blocksize); - goto out_fail; - } - /* * The ext4 superblock will not be buffer aligned for other than 1kB * block sizes. We need to calculate the offset from buffer start. @@ -1919,15 +2004,28 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) /* * turn on extents feature by default in ext4 filesystem - * User -o noextents to turn it off + * only if feature flag already set by mkfs or tune2fs. + * Use -o noextents to turn it off */ - set_opt(sbi->s_mount_opt, EXTENTS); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) + set_opt(sbi->s_mount_opt, EXTENTS); + else + ext4_warning(sb, __func__, + "extents feature not enabled on this filesystem, " + "use tune2fs.\n"); /* - * turn on mballoc feature by default in ext4 filesystem - * User -o nomballoc to turn it off + * turn on mballoc code by default in ext4 filesystem + * Use -o nomballoc to turn it off */ set_opt(sbi->s_mount_opt, MBALLOC); + /* + * enable delayed allocation by default + * Use -o nodelalloc to turn it off + */ + set_opt(sbi->s_mount_opt, DELALLOC); + + if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum, NULL, 0)) goto failed_mount; @@ -2138,6 +2236,14 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n"); goto failed_mount2; } + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) + if (!ext4_fill_flex_info(sb)) { + printk(KERN_ERR + "EXT4-fs: unable to initialize " + "flex_bg meta info!\n"); + goto failed_mount2; + } + sbi->s_gdb_count = db_count; get_random_bytes(&sbi->s_next_generation, sizeof(u32)); spin_lock_init(&sbi->s_next_gen_lock); @@ -2358,6 +2464,13 @@ static int ext4_fill_super (struct super_block *sb, void *data, int silent) test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered": "writeback"); + if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) { + printk(KERN_WARNING "EXT4-fs: Ignoring delalloc option - " + "requested data journaling mode\n"); + clear_opt(sbi->s_mount_opt, DELALLOC); + } else if (test_opt(sb, DELALLOC)) + printk(KERN_INFO "EXT4-fs: delayed allocation enabled\n"); + ext4_ext_init(sb); ext4_mb_init(sb, needs_recovery); @@ -2372,6 +2485,7 @@ cantfind_ext4: failed_mount4: jbd2_journal_destroy(sbi->s_journal); + sbi->s_journal = NULL; failed_mount3: percpu_counter_destroy(&sbi->s_freeblocks_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); @@ -3325,7 +3439,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type, err = ext4_journal_dirty_metadata(handle, bh); else { /* Always do at least ordered writes for quotas */ - err = ext4_journal_dirty_data(handle, bh); + err = ext4_jbd2_file_inode(handle, inode); mark_buffer_dirty(bh); } brelse(bh); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index ff08633f398e..93c5fdcdad2e 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -810,7 +810,7 @@ inserted: /* We need to allocate a new block */ ext4_fsblk_t goal = ext4_group_first_block_no(sb, EXT4_I(inode)->i_block_group); - ext4_fsblk_t block = ext4_new_block(handle, inode, + ext4_fsblk_t block = ext4_new_meta_block(handle, inode, goal, &error); if (error) goto cleanup; diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c index fff33382cadc..ac1a52cf2a37 100644 --- a/fs/ext4/xattr_trusted.c +++ b/fs/ext4/xattr_trusted.c @@ -13,13 +13,11 @@ #include "ext4.h" #include "xattr.h" -#define XATTR_TRUSTED_PREFIX "trusted." - static size_t ext4_xattr_trusted_list(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { - const size_t prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1; + const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN; const size_t total_len = prefix_len + name_len + 1; if (!capable(CAP_SYS_ADMIN)) diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c index 67be723fcc4e..d91aa61b42aa 100644 --- a/fs/ext4/xattr_user.c +++ b/fs/ext4/xattr_user.c @@ -12,13 +12,11 @@ #include "ext4.h" #include "xattr.h" -#define XATTR_USER_PREFIX "user." - static size_t ext4_xattr_user_list(struct inode *inode, char *list, size_t list_size, const char *name, size_t name_len) { - const size_t prefix_len = sizeof(XATTR_USER_PREFIX)-1; + const size_t prefix_len = XATTR_USER_PREFIX_LEN; const size_t total_len = prefix_len + name_len + 1; if (!test_opt(inode->i_sb, XATTR_USER)) diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 6914598022ce..91389c8aee8a 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -688,7 +688,6 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact J_ASSERT(transaction->t_state == T_FINISHED); J_ASSERT(transaction->t_buffers == NULL); - J_ASSERT(transaction->t_sync_datalist == NULL); J_ASSERT(transaction->t_forget == NULL); J_ASSERT(transaction->t_iobuf_list == NULL); J_ASSERT(transaction->t_shadow_list == NULL); diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index a2ed72f7ceee..f8b3be873226 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -22,6 +22,8 @@ #include <linux/pagemap.h> #include <linux/jiffies.h> #include <linux/crc32.h> +#include <linux/writeback.h> +#include <linux/backing-dev.h> /* * Default IO end handler for temporary BJ_IO buffer_heads. @@ -37,8 +39,8 @@ static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate) } /* - * When an ext3-ordered file is truncated, it is possible that many pages are - * not sucessfully freed, because they are attached to a committing transaction. + * When an ext4 file is truncated, it is possible that some pages are not + * successfully freed, because they are attached to a committing transaction. * After the transaction commits, these pages are left on the LRU, with no * ->mapping, and with attached buffers. These pages are trivially reclaimable * by the VM, but their apparent absence upsets the VM accounting, and it makes @@ -80,21 +82,6 @@ nope: } /* - * Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is - * held. For ranking reasons we must trylock. If we lose, schedule away and - * return 0. j_list_lock is dropped in this case. - */ -static int inverted_lock(journal_t *journal, struct buffer_head *bh) -{ - if (!jbd_trylock_bh_state(bh)) { - spin_unlock(&journal->j_list_lock); - schedule(); - return 0; - } - return 1; -} - -/* * Done it all: now submit the commit record. We should have * cleaned up our previous buffers by now, so if we are in abort * mode we can now just skip the rest of the journal write @@ -112,6 +99,7 @@ static int journal_submit_commit_record(journal_t *journal, struct buffer_head *bh; int ret; int barrier_done = 0; + struct timespec now = current_kernel_time(); if (is_journal_aborted(journal)) return 0; @@ -126,6 +114,8 @@ static int journal_submit_commit_record(journal_t *journal, tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + tmp->h_commit_sec = cpu_to_be64(now.tv_sec); + tmp->h_commit_nsec = cpu_to_be32(now.tv_nsec); if (JBD2_HAS_COMPAT_FEATURE(journal, JBD2_FEATURE_COMPAT_CHECKSUM)) { @@ -197,159 +187,104 @@ static int journal_wait_on_commit_record(struct buffer_head *bh) } /* - * Wait for all submitted IO to complete. + * write the filemap data using writepage() address_space_operations. + * We don't do block allocation here even for delalloc. We don't + * use writepages() because with dealyed allocation we may be doing + * block allocation in writepages(). */ -static int journal_wait_on_locked_list(journal_t *journal, - transaction_t *commit_transaction) +static int journal_submit_inode_data_buffers(struct address_space *mapping) { - int ret = 0; - struct journal_head *jh; - - while (commit_transaction->t_locked_list) { - struct buffer_head *bh; - - jh = commit_transaction->t_locked_list->b_tprev; - bh = jh2bh(jh); - get_bh(bh); - if (buffer_locked(bh)) { - spin_unlock(&journal->j_list_lock); - wait_on_buffer(bh); - if (unlikely(!buffer_uptodate(bh))) - ret = -EIO; - spin_lock(&journal->j_list_lock); - } - if (!inverted_lock(journal, bh)) { - put_bh(bh); - spin_lock(&journal->j_list_lock); - continue; - } - if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { - __jbd2_journal_unfile_buffer(jh); - jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); - put_bh(bh); - } else { - jbd_unlock_bh_state(bh); - } - put_bh(bh); - cond_resched_lock(&journal->j_list_lock); - } + int ret; + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = mapping->nrpages * 2, + .range_start = 0, + .range_end = i_size_read(mapping->host), + .for_writepages = 1, + }; + + ret = generic_writepages(mapping, &wbc); return ret; - } +} -static void journal_do_submit_data(struct buffer_head **wbuf, int bufs) +/* + * Submit all the data buffers of inode associated with the transaction to + * disk. + * + * We are in a committing transaction. Therefore no new inode can be added to + * our inode list. We use JI_COMMIT_RUNNING flag to protect inode we currently + * operate on from being released while we write out pages. + */ +static int journal_submit_data_buffers(journal_t *journal, + transaction_t *commit_transaction) { - int i; + struct jbd2_inode *jinode; + int err, ret = 0; + struct address_space *mapping; - for (i = 0; i < bufs; i++) { - wbuf[i]->b_end_io = end_buffer_write_sync; - /* We use-up our safety reference in submit_bh() */ - submit_bh(WRITE, wbuf[i]); + spin_lock(&journal->j_list_lock); + list_for_each_entry(jinode, &commit_transaction->t_inode_list, i_list) { + mapping = jinode->i_vfs_inode->i_mapping; + jinode->i_flags |= JI_COMMIT_RUNNING; + spin_unlock(&journal->j_list_lock); + /* + * submit the inode data buffers. We use writepage + * instead of writepages. Because writepages can do + * block allocation with delalloc. We need to write + * only allocated blocks here. + */ + err = journal_submit_inode_data_buffers(mapping); + if (!ret) + ret = err; + spin_lock(&journal->j_list_lock); + J_ASSERT(jinode->i_transaction == commit_transaction); + jinode->i_flags &= ~JI_COMMIT_RUNNING; + wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); } + spin_unlock(&journal->j_list_lock); + return ret; } /* - * Submit all the data buffers to disk + * Wait for data submitted for writeout, refile inodes to proper + * transaction if needed. + * */ -static void journal_submit_data_buffers(journal_t *journal, - transaction_t *commit_transaction) +static int journal_finish_inode_data_buffers(journal_t *journal, + transaction_t *commit_transaction) { - struct journal_head *jh; - struct buffer_head *bh; - int locked; - int bufs = 0; - struct buffer_head **wbuf = journal->j_wbuf; + struct jbd2_inode *jinode, *next_i; + int err, ret = 0; - /* - * Whenever we unlock the journal and sleep, things can get added - * onto ->t_sync_datalist, so we have to keep looping back to - * write_out_data until we *know* that the list is empty. - * - * Cleanup any flushed data buffers from the data list. Even in - * abort mode, we want to flush this out as soon as possible. - */ -write_out_data: - cond_resched(); + /* For locking, see the comment in journal_submit_data_buffers() */ spin_lock(&journal->j_list_lock); + list_for_each_entry(jinode, &commit_transaction->t_inode_list, i_list) { + jinode->i_flags |= JI_COMMIT_RUNNING; + spin_unlock(&journal->j_list_lock); + err = filemap_fdatawait(jinode->i_vfs_inode->i_mapping); + if (!ret) + ret = err; + spin_lock(&journal->j_list_lock); + jinode->i_flags &= ~JI_COMMIT_RUNNING; + wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); + } - while (commit_transaction->t_sync_datalist) { - jh = commit_transaction->t_sync_datalist; - bh = jh2bh(jh); - locked = 0; - - /* Get reference just to make sure buffer does not disappear - * when we are forced to drop various locks */ - get_bh(bh); - /* If the buffer is dirty, we need to submit IO and hence - * we need the buffer lock. We try to lock the buffer without - * blocking. If we fail, we need to drop j_list_lock and do - * blocking lock_buffer(). - */ - if (buffer_dirty(bh)) { - if (test_set_buffer_locked(bh)) { - BUFFER_TRACE(bh, "needs blocking lock"); - spin_unlock(&journal->j_list_lock); - /* Write out all data to prevent deadlocks */ - journal_do_submit_data(wbuf, bufs); - bufs = 0; - lock_buffer(bh); - spin_lock(&journal->j_list_lock); - } - locked = 1; - } - /* We have to get bh_state lock. Again out of order, sigh. */ - if (!inverted_lock(journal, bh)) { - jbd_lock_bh_state(bh); - spin_lock(&journal->j_list_lock); - } - /* Someone already cleaned up the buffer? */ - if (!buffer_jbd(bh) - || jh->b_transaction != commit_transaction - || jh->b_jlist != BJ_SyncData) { - jbd_unlock_bh_state(bh); - if (locked) - unlock_buffer(bh); - BUFFER_TRACE(bh, "already cleaned up"); - put_bh(bh); - continue; - } - if (locked && test_clear_buffer_dirty(bh)) { - BUFFER_TRACE(bh, "needs writeout, adding to array"); - wbuf[bufs++] = bh; - __jbd2_journal_file_buffer(jh, commit_transaction, - BJ_Locked); - jbd_unlock_bh_state(bh); - if (bufs == journal->j_wbufsize) { - spin_unlock(&journal->j_list_lock); - journal_do_submit_data(wbuf, bufs); - bufs = 0; - goto write_out_data; - } - } else if (!locked && buffer_locked(bh)) { - __jbd2_journal_file_buffer(jh, commit_transaction, - BJ_Locked); - jbd_unlock_bh_state(bh); - put_bh(bh); + /* Now refile inode to proper lists */ + list_for_each_entry_safe(jinode, next_i, + &commit_transaction->t_inode_list, i_list) { + list_del(&jinode->i_list); + if (jinode->i_next_transaction) { + jinode->i_transaction = jinode->i_next_transaction; + jinode->i_next_transaction = NULL; + list_add(&jinode->i_list, + &jinode->i_transaction->t_inode_list); } else { - BUFFER_TRACE(bh, "writeout complete: unfile"); - __jbd2_journal_unfile_buffer(jh); - jbd_unlock_bh_state(bh); - if (locked) - unlock_buffer(bh); - jbd2_journal_remove_journal_head(bh); - /* Once for our safety reference, once for - * jbd2_journal_remove_journal_head() */ - put_bh(bh); - put_bh(bh); - } - - if (need_resched() || spin_needbreak(&journal->j_list_lock)) { - spin_unlock(&journal->j_list_lock); - goto write_out_data; + jinode->i_transaction = NULL; } } spin_unlock(&journal->j_list_lock); - journal_do_submit_data(wbuf, bufs); + + return ret; } static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh) @@ -524,21 +459,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) * Now start flushing things to disk, in the order they appear * on the transaction lists. Data blocks go first. */ - err = 0; - journal_submit_data_buffers(journal, commit_transaction); - - /* - * Wait for all previously submitted IO to complete if commit - * record is to be written synchronously. - */ - spin_lock(&journal->j_list_lock); - if (!JBD2_HAS_INCOMPAT_FEATURE(journal, - JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) - err = journal_wait_on_locked_list(journal, - commit_transaction); - - spin_unlock(&journal->j_list_lock); - + err = journal_submit_data_buffers(journal, commit_transaction); if (err) jbd2_journal_abort(journal, err); @@ -547,16 +468,6 @@ void jbd2_journal_commit_transaction(journal_t *journal) jbd_debug(3, "JBD: commit phase 2\n"); /* - * If we found any dirty or locked buffers, then we should have - * looped back up to the write_out_data label. If there weren't - * any then journal_clean_data_list should have wiped the list - * clean by now, so check that it is in fact empty. - */ - J_ASSERT (commit_transaction->t_sync_datalist == NULL); - - jbd_debug (3, "JBD: commit phase 3\n"); - - /* * Way to go: we have now written out all of the data for a * transaction! Now comes the tricky part: we need to write out * metadata. Loop over the transaction's entire buffer list: @@ -574,6 +485,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) J_ASSERT(commit_transaction->t_nr_buffers <= commit_transaction->t_outstanding_credits); + err = 0; descriptor = NULL; bufs = 0; while (commit_transaction->t_buffers) { @@ -748,15 +660,19 @@ start_journal_io: &cbh, crc32_sum); if (err) __jbd2_journal_abort_hard(journal); - - spin_lock(&journal->j_list_lock); - err = journal_wait_on_locked_list(journal, - commit_transaction); - spin_unlock(&journal->j_list_lock); - if (err) - __jbd2_journal_abort_hard(journal); } + /* + * This is the right place to wait for data buffers both for ASYNC + * and !ASYNC commit. If commit is ASYNC, we need to wait only after + * the commit block went to disk (which happens above). If commit is + * SYNC, we need to wait for data buffers before we start writing + * commit block, which happens below in such setting. + */ + err = journal_finish_inode_data_buffers(journal, commit_transaction); + if (err) + jbd2_journal_abort(journal, err); + /* Lo and behold: we have just managed to send a transaction to the log. Before we can commit it, wait for the IO so far to complete. Control buffers being written are on the @@ -768,7 +684,7 @@ start_journal_io: so we incur less scheduling load. */ - jbd_debug(3, "JBD: commit phase 4\n"); + jbd_debug(3, "JBD: commit phase 3\n"); /* * akpm: these are BJ_IO, and j_list_lock is not needed. @@ -827,7 +743,7 @@ wait_for_iobuf: J_ASSERT (commit_transaction->t_shadow_list == NULL); - jbd_debug(3, "JBD: commit phase 5\n"); + jbd_debug(3, "JBD: commit phase 4\n"); /* Here we wait for the revoke record and descriptor record buffers */ wait_for_ctlbuf: @@ -854,7 +770,7 @@ wait_for_iobuf: /* AKPM: bforget here */ } - jbd_debug(3, "JBD: commit phase 6\n"); + jbd_debug(3, "JBD: commit phase 5\n"); if (!JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { @@ -874,9 +790,9 @@ wait_for_iobuf: transaction can be removed from any checkpoint list it was on before. */ - jbd_debug(3, "JBD: commit phase 7\n"); + jbd_debug(3, "JBD: commit phase 6\n"); - J_ASSERT(commit_transaction->t_sync_datalist == NULL); + J_ASSERT(list_empty(&commit_transaction->t_inode_list)); J_ASSERT(commit_transaction->t_buffers == NULL); J_ASSERT(commit_transaction->t_checkpoint_list == NULL); J_ASSERT(commit_transaction->t_iobuf_list == NULL); @@ -997,7 +913,7 @@ restart_loop: /* Done with this transaction! */ - jbd_debug(3, "JBD: commit phase 8\n"); + jbd_debug(3, "JBD: commit phase 7\n"); J_ASSERT(commit_transaction->t_state == T_COMMIT); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 2e24567c4a79..b26c6d9fe6ae 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -50,7 +50,6 @@ EXPORT_SYMBOL(jbd2_journal_unlock_updates); EXPORT_SYMBOL(jbd2_journal_get_write_access); EXPORT_SYMBOL(jbd2_journal_get_create_access); EXPORT_SYMBOL(jbd2_journal_get_undo_access); -EXPORT_SYMBOL(jbd2_journal_dirty_data); EXPORT_SYMBOL(jbd2_journal_dirty_metadata); EXPORT_SYMBOL(jbd2_journal_release_buffer); EXPORT_SYMBOL(jbd2_journal_forget); @@ -82,6 +81,10 @@ EXPORT_SYMBOL(jbd2_journal_blocks_per_page); EXPORT_SYMBOL(jbd2_journal_invalidatepage); EXPORT_SYMBOL(jbd2_journal_try_to_free_buffers); EXPORT_SYMBOL(jbd2_journal_force_commit); +EXPORT_SYMBOL(jbd2_journal_file_inode); +EXPORT_SYMBOL(jbd2_journal_init_jbd_inode); +EXPORT_SYMBOL(jbd2_journal_release_jbd_inode); +EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate); static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *); static void __journal_abort_soft (journal_t *journal, int errno); @@ -2195,6 +2198,54 @@ void jbd2_journal_put_journal_head(struct journal_head *jh) } /* + * Initialize jbd inode head + */ +void jbd2_journal_init_jbd_inode(struct jbd2_inode *jinode, struct inode *inode) +{ + jinode->i_transaction = NULL; + jinode->i_next_transaction = NULL; + jinode->i_vfs_inode = inode; + jinode->i_flags = 0; + INIT_LIST_HEAD(&jinode->i_list); +} + +/* + * Function to be called before we start removing inode from memory (i.e., + * clear_inode() is a fine place to be called from). It removes inode from + * transaction's lists. + */ +void jbd2_journal_release_jbd_inode(journal_t *journal, + struct jbd2_inode *jinode) +{ + int writeout = 0; + + if (!journal) + return; +restart: + spin_lock(&journal->j_list_lock); + /* Is commit writing out inode - we have to wait */ + if (jinode->i_flags & JI_COMMIT_RUNNING) { + wait_queue_head_t *wq; + DEFINE_WAIT_BIT(wait, &jinode->i_flags, __JI_COMMIT_RUNNING); + wq = bit_waitqueue(&jinode->i_flags, __JI_COMMIT_RUNNING); + prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE); + spin_unlock(&journal->j_list_lock); + schedule(); + finish_wait(wq, &wait.wait); + goto restart; + } + + /* Do we need to wait for data writeback? */ + if (journal->j_committing_transaction == jinode->i_transaction) + writeout = 1; + if (jinode->i_transaction) { + list_del(&jinode->i_list); + jinode->i_transaction = NULL; + } + spin_unlock(&journal->j_list_lock); +} + +/* * debugfs tunables */ #ifdef CONFIG_JBD2_DEBUG diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index d6e006e67804..4f7cadbb19fa 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -41,7 +41,6 @@ static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); * new transaction and we can't block without protecting against other * processes trying to touch the journal while it is in transition. * - * Called under j_state_lock */ static transaction_t * @@ -52,6 +51,7 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) transaction->t_tid = journal->j_transaction_sequence++; transaction->t_expires = jiffies + journal->j_commit_interval; spin_lock_init(&transaction->t_handle_lock); + INIT_LIST_HEAD(&transaction->t_inode_list); /* Set up the commit timer for the new transaction. */ journal->j_commit_timer.expires = round_jiffies(transaction->t_expires); @@ -943,183 +943,6 @@ out: } /** - * int jbd2_journal_dirty_data() - mark a buffer as containing dirty data which - * needs to be flushed before we can commit the - * current transaction. - * @handle: transaction - * @bh: bufferhead to mark - * - * The buffer is placed on the transaction's data list and is marked as - * belonging to the transaction. - * - * Returns error number or 0 on success. - * - * jbd2_journal_dirty_data() can be called via page_launder->ext3_writepage - * by kswapd. - */ -int jbd2_journal_dirty_data(handle_t *handle, struct buffer_head *bh) -{ - journal_t *journal = handle->h_transaction->t_journal; - int need_brelse = 0; - struct journal_head *jh; - - if (is_handle_aborted(handle)) - return 0; - - jh = jbd2_journal_add_journal_head(bh); - JBUFFER_TRACE(jh, "entry"); - - /* - * The buffer could *already* be dirty. Writeout can start - * at any time. - */ - jbd_debug(4, "jh: %p, tid:%d\n", jh, handle->h_transaction->t_tid); - - /* - * What if the buffer is already part of a running transaction? - * - * There are two cases: - * 1) It is part of the current running transaction. Refile it, - * just in case we have allocated it as metadata, deallocated - * it, then reallocated it as data. - * 2) It is part of the previous, still-committing transaction. - * If all we want to do is to guarantee that the buffer will be - * written to disk before this new transaction commits, then - * being sure that the *previous* transaction has this same - * property is sufficient for us! Just leave it on its old - * transaction. - * - * In case (2), the buffer must not already exist as metadata - * --- that would violate write ordering (a transaction is free - * to write its data at any point, even before the previous - * committing transaction has committed). The caller must - * never, ever allow this to happen: there's nothing we can do - * about it in this layer. - */ - jbd_lock_bh_state(bh); - spin_lock(&journal->j_list_lock); - - /* Now that we have bh_state locked, are we really still mapped? */ - if (!buffer_mapped(bh)) { - JBUFFER_TRACE(jh, "unmapped buffer, bailing out"); - goto no_journal; - } - - if (jh->b_transaction) { - JBUFFER_TRACE(jh, "has transaction"); - if (jh->b_transaction != handle->h_transaction) { - JBUFFER_TRACE(jh, "belongs to older transaction"); - J_ASSERT_JH(jh, jh->b_transaction == - journal->j_committing_transaction); - - /* @@@ IS THIS TRUE ? */ - /* - * Not any more. Scenario: someone does a write() - * in data=journal mode. The buffer's transaction has - * moved into commit. Then someone does another - * write() to the file. We do the frozen data copyout - * and set b_next_transaction to point to j_running_t. - * And while we're in that state, someone does a - * writepage() in an attempt to pageout the same area - * of the file via a shared mapping. At present that - * calls jbd2_journal_dirty_data(), and we get right here. - * It may be too late to journal the data. Simply - * falling through to the next test will suffice: the - * data will be dirty and wil be checkpointed. The - * ordering comments in the next comment block still - * apply. - */ - //J_ASSERT_JH(jh, jh->b_next_transaction == NULL); - - /* - * If we're journalling data, and this buffer was - * subject to a write(), it could be metadata, forget - * or shadow against the committing transaction. Now, - * someone has dirtied the same darn page via a mapping - * and it is being writepage()'d. - * We *could* just steal the page from commit, with some - * fancy locking there. Instead, we just skip it - - * don't tie the page's buffers to the new transaction - * at all. - * Implication: if we crash before the writepage() data - * is written into the filesystem, recovery will replay - * the write() data. - */ - if (jh->b_jlist != BJ_None && - jh->b_jlist != BJ_SyncData && - jh->b_jlist != BJ_Locked) { - JBUFFER_TRACE(jh, "Not stealing"); - goto no_journal; - } - - /* - * This buffer may be undergoing writeout in commit. We - * can't return from here and let the caller dirty it - * again because that can cause the write-out loop in - * commit to never terminate. - */ - if (buffer_dirty(bh)) { - get_bh(bh); - spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(bh); - need_brelse = 1; - sync_dirty_buffer(bh); - jbd_lock_bh_state(bh); - spin_lock(&journal->j_list_lock); - /* Since we dropped the lock... */ - if (!buffer_mapped(bh)) { - JBUFFER_TRACE(jh, "buffer got unmapped"); - goto no_journal; - } - /* The buffer may become locked again at any - time if it is redirtied */ - } - - /* journal_clean_data_list() may have got there first */ - if (jh->b_transaction != NULL) { - JBUFFER_TRACE(jh, "unfile from commit"); - __jbd2_journal_temp_unlink_buffer(jh); - /* It still points to the committing - * transaction; move it to this one so - * that the refile assert checks are - * happy. */ - jh->b_transaction = handle->h_transaction; - } - /* The buffer will be refiled below */ - - } - /* - * Special case --- the buffer might actually have been - * allocated and then immediately deallocated in the previous, - * committing transaction, so might still be left on that - * transaction's metadata lists. - */ - if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) { - JBUFFER_TRACE(jh, "not on correct data list: unfile"); - J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow); - __jbd2_journal_temp_unlink_buffer(jh); - jh->b_transaction = handle->h_transaction; - JBUFFER_TRACE(jh, "file as data"); - __jbd2_journal_file_buffer(jh, handle->h_transaction, - BJ_SyncData); - } - } else { - JBUFFER_TRACE(jh, "not on a transaction"); - __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); - } -no_journal: - spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(bh); - if (need_brelse) { - BUFFER_TRACE(bh, "brelse"); - __brelse(bh); - } - JBUFFER_TRACE(jh, "exit"); - jbd2_journal_put_journal_head(jh); - return 0; -} - -/** * int jbd2_journal_dirty_metadata() - mark a buffer as containing dirty metadata * @handle: transaction to add buffer to. * @bh: buffer to mark @@ -1541,10 +1364,10 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh) * Remove a buffer from the appropriate transaction list. * * Note that this function can *change* the value of - * bh->b_transaction->t_sync_datalist, t_buffers, t_forget, - * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list. If the caller - * is holding onto a copy of one of thee pointers, it could go bad. - * Generally the caller needs to re-read the pointer from the transaction_t. + * bh->b_transaction->t_buffers, t_forget, t_iobuf_list, t_shadow_list, + * t_log_list or t_reserved_list. If the caller is holding onto a copy of one + * of these pointers, it could go bad. Generally the caller needs to re-read + * the pointer from the transaction_t. * * Called under j_list_lock. The journal may not be locked. */ @@ -1566,9 +1389,6 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) switch (jh->b_jlist) { case BJ_None: return; - case BJ_SyncData: - list = &transaction->t_sync_datalist; - break; case BJ_Metadata: transaction->t_nr_buffers--; J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0); @@ -1589,9 +1409,6 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) case BJ_Reserved: list = &transaction->t_reserved_list; break; - case BJ_Locked: - list = &transaction->t_locked_list; - break; } __blist_del_buffer(list, jh); @@ -1634,15 +1451,7 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) goto out; spin_lock(&journal->j_list_lock); - if (jh->b_transaction != NULL && jh->b_cp_transaction == NULL) { - if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) { - /* A written-back ordered data buffer */ - JBUFFER_TRACE(jh, "release data"); - __jbd2_journal_unfile_buffer(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); - } - } else if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) { + if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) { /* written-back checkpointed metadata buffer */ if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); @@ -1656,12 +1465,43 @@ out: return; } +/* + * jbd2_journal_try_to_free_buffers() could race with + * jbd2_journal_commit_transaction(). The later might still hold the + * reference count to the buffers when inspecting them on + * t_syncdata_list or t_locked_list. + * + * jbd2_journal_try_to_free_buffers() will call this function to + * wait for the current transaction to finish syncing data buffers, before + * try to free that buffer. + * + * Called with journal->j_state_lock hold. + */ +static void jbd2_journal_wait_for_transaction_sync_data(journal_t *journal) +{ + transaction_t *transaction; + tid_t tid; + + spin_lock(&journal->j_state_lock); + transaction = journal->j_committing_transaction; + + if (!transaction) { + spin_unlock(&journal->j_state_lock); + return; + } + + tid = transaction->t_tid; + spin_unlock(&journal->j_state_lock); + jbd2_log_wait_commit(journal, tid); +} /** * int jbd2_journal_try_to_free_buffers() - try to free page buffers. * @journal: journal for operation * @page: to try and free - * @unused_gfp_mask: unused + * @gfp_mask: we use the mask to detect how hard should we try to release + * buffers. If __GFP_WAIT and __GFP_FS is set, we wait for commit code to + * release the buffers. * * * For all the buffers on this page, @@ -1690,9 +1530,11 @@ out: * journal_try_to_free_buffer() is changing its state. But that * cannot happen because we never reallocate freed data as metadata * while the data is part of a transaction. Yes? + * + * Return 0 on failure, 1 on success */ int jbd2_journal_try_to_free_buffers(journal_t *journal, - struct page *page, gfp_t unused_gfp_mask) + struct page *page, gfp_t gfp_mask) { struct buffer_head *head; struct buffer_head *bh; @@ -1708,7 +1550,8 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal, /* * We take our own ref against the journal_head here to avoid * having to add tons of locking around each instance of - * jbd2_journal_remove_journal_head() and jbd2_journal_put_journal_head(). + * jbd2_journal_remove_journal_head() and + * jbd2_journal_put_journal_head(). */ jh = jbd2_journal_grab_journal_head(bh); if (!jh) @@ -1721,7 +1564,28 @@ int jbd2_journal_try_to_free_buffers(journal_t *journal, if (buffer_jbd(bh)) goto busy; } while ((bh = bh->b_this_page) != head); + ret = try_to_free_buffers(page); + + /* + * There are a number of places where jbd2_journal_try_to_free_buffers() + * could race with jbd2_journal_commit_transaction(), the later still + * holds the reference to the buffers to free while processing them. + * try_to_free_buffers() failed to free those buffers. Some of the + * caller of releasepage() request page buffers to be dropped, otherwise + * treat the fail-to-free as errors (such as generic_file_direct_IO()) + * + * So, if the caller of try_to_release_page() wants the synchronous + * behaviour(i.e make sure buffers are dropped upon return), + * let's wait for the current transaction to finish flush of + * dirty data buffers, then try to free those buffers again, + * with the journal locked. + */ + if (ret == 0 && (gfp_mask & __GFP_WAIT) && (gfp_mask & __GFP_FS)) { + jbd2_journal_wait_for_transaction_sync_data(journal); + ret = try_to_free_buffers(page); + } + busy: return ret; } @@ -1823,6 +1687,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) if (!buffer_jbd(bh)) goto zap_buffer_unlocked; + /* OK, we have data buffer in journaled mode */ spin_lock(&journal->j_state_lock); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); @@ -1886,15 +1751,6 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) } } else if (transaction == journal->j_committing_transaction) { JBUFFER_TRACE(jh, "on committing transaction"); - if (jh->b_jlist == BJ_Locked) { - /* - * The buffer is on the committing transaction's locked - * list. We have the buffer locked, so I/O has - * completed. So we can nail the buffer now. - */ - may_free = __dispose_buffer(jh, transaction); - goto zap_buffer; - } /* * If it is committing, we simply cannot touch it. We * can remove it's next_transaction pointer from the @@ -2027,9 +1883,6 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, J_ASSERT_JH(jh, !jh->b_committed_data); J_ASSERT_JH(jh, !jh->b_frozen_data); return; - case BJ_SyncData: - list = &transaction->t_sync_datalist; - break; case BJ_Metadata: transaction->t_nr_buffers++; list = &transaction->t_buffers; @@ -2049,9 +1902,6 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, case BJ_Reserved: list = &transaction->t_reserved_list; break; - case BJ_Locked: - list = &transaction->t_locked_list; - break; } __blist_add_buffer(list, jh); @@ -2141,3 +1991,88 @@ void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) spin_unlock(&journal->j_list_lock); __brelse(bh); } + +/* + * File inode in the inode list of the handle's transaction + */ +int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + + if (is_handle_aborted(handle)) + return -EIO; + + jbd_debug(4, "Adding inode %lu, tid:%d\n", jinode->i_vfs_inode->i_ino, + transaction->t_tid); + + /* + * First check whether inode isn't already on the transaction's + * lists without taking the lock. Note that this check is safe + * without the lock as we cannot race with somebody removing inode + * from the transaction. The reason is that we remove inode from the + * transaction only in journal_release_jbd_inode() and when we commit + * the transaction. We are guarded from the first case by holding + * a reference to the inode. We are safe against the second case + * because if jinode->i_transaction == transaction, commit code + * cannot touch the transaction because we hold reference to it, + * and if jinode->i_next_transaction == transaction, commit code + * will only file the inode where we want it. + */ + if (jinode->i_transaction == transaction || + jinode->i_next_transaction == transaction) + return 0; + + spin_lock(&journal->j_list_lock); + + if (jinode->i_transaction == transaction || + jinode->i_next_transaction == transaction) + goto done; + + /* On some different transaction's list - should be + * the committing one */ + if (jinode->i_transaction) { + J_ASSERT(jinode->i_next_transaction == NULL); + J_ASSERT(jinode->i_transaction == + journal->j_committing_transaction); + jinode->i_next_transaction = transaction; + goto done; + } + /* Not on any transaction list... */ + J_ASSERT(!jinode->i_next_transaction); + jinode->i_transaction = transaction; + list_add(&jinode->i_list, &transaction->t_inode_list); +done: + spin_unlock(&journal->j_list_lock); + + return 0; +} + +/* + * This function must be called when inode is journaled in ordered mode + * before truncation happens. It starts writeout of truncated part in + * case it is in the committing transaction so that we stand to ordered + * mode consistency guarantees. + */ +int jbd2_journal_begin_ordered_truncate(struct jbd2_inode *inode, + loff_t new_size) +{ + journal_t *journal; + transaction_t *commit_trans; + int ret = 0; + + if (!inode->i_transaction && !inode->i_next_transaction) + goto out; + journal = inode->i_transaction->t_journal; + spin_lock(&journal->j_state_lock); + commit_trans = journal->j_committing_transaction; + spin_unlock(&journal->j_state_lock); + if (inode->i_transaction == commit_trans) { + ret = filemap_fdatawrite_range(inode->i_vfs_inode->i_mapping, + new_size, LLONG_MAX); + if (ret) + jbd2_journal_abort(journal, ret); + } +out: + return ret; +} diff --git a/fs/mpage.c b/fs/mpage.c index 235e4d3873a8..dbcc7af76a15 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -82,7 +82,7 @@ static void mpage_end_io_write(struct bio *bio, int err) bio_put(bio); } -static struct bio *mpage_bio_submit(int rw, struct bio *bio) +struct bio *mpage_bio_submit(int rw, struct bio *bio) { bio->bi_end_io = mpage_end_io_read; if (rw == WRITE) @@ -90,6 +90,7 @@ static struct bio *mpage_bio_submit(int rw, struct bio *bio) submit_bio(rw, bio); return NULL; } +EXPORT_SYMBOL(mpage_bio_submit); static struct bio * mpage_alloc(struct block_device *bdev, @@ -435,15 +436,9 @@ EXPORT_SYMBOL(mpage_readpage); * written, so it can intelligently allocate a suitably-sized BIO. For now, * just allocate full-size (16-page) BIOs. */ -struct mpage_data { - struct bio *bio; - sector_t last_block_in_bio; - get_block_t *get_block; - unsigned use_writepage; -}; -static int __mpage_writepage(struct page *page, struct writeback_control *wbc, - void *data) +int __mpage_writepage(struct page *page, struct writeback_control *wbc, + void *data) { struct mpage_data *mpd = data; struct bio *bio = mpd->bio; @@ -651,6 +646,7 @@ out: mpd->bio = bio; return ret; } +EXPORT_SYMBOL(__mpage_writepage); /** * mpage_writepages - walk the list of dirty pages of the given address space & writepage() all of them diff --git a/include/drm/drmP.h b/include/drm/drmP.h index 0764b662b339..1c1b13e29223 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -1089,6 +1089,7 @@ extern int drm_mm_remove_space_from_tail(struct drm_mm *mm, unsigned long size); extern int drm_mm_add_space_to_tail(struct drm_mm *mm, unsigned long size); extern void drm_core_ioremap(struct drm_map *map, struct drm_device *dev); +extern void drm_core_ioremap_wc(struct drm_map *map, struct drm_device *dev); extern void drm_core_ioremapfree(struct drm_map *map, struct drm_device *dev); static __inline__ struct drm_map *drm_core_findmap(struct drm_device *dev, diff --git a/include/linux/fs.h b/include/linux/fs.h index faac13e2cc5c..52e510a0aec2 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1740,6 +1740,8 @@ extern int wait_on_page_writeback_range(struct address_space *mapping, pgoff_t start, pgoff_t end); extern int __filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end, int sync_mode); +extern int filemap_fdatawrite_range(struct address_space *mapping, + loff_t start, loff_t end); extern long do_fsync(struct file *file, int datasync); extern void sync_supers(void); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index d147f0f90360..3dd209007098 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -168,6 +168,8 @@ struct commit_header { unsigned char h_chksum_size; unsigned char h_padding[2]; __be32 h_chksum[JBD2_CHECKSUM_BYTES]; + __be64 h_commit_sec; + __be32 h_commit_nsec; }; /* @@ -379,6 +381,38 @@ static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh) bit_spin_unlock(BH_JournalHead, &bh->b_state); } +/* Flags in jbd_inode->i_flags */ +#define __JI_COMMIT_RUNNING 0 +/* Commit of the inode data in progress. We use this flag to protect us from + * concurrent deletion of inode. We cannot use reference to inode for this + * since we cannot afford doing last iput() on behalf of kjournald + */ +#define JI_COMMIT_RUNNING (1 << __JI_COMMIT_RUNNING) + +/** + * struct jbd_inode is the structure linking inodes in ordered mode + * present in a transaction so that we can sync them during commit. + */ +struct jbd2_inode { + /* Which transaction does this inode belong to? Either the running + * transaction or the committing one. [j_list_lock] */ + transaction_t *i_transaction; + + /* Pointer to the running transaction modifying inode's data in case + * there is already a committing transaction touching it. [j_list_lock] */ + transaction_t *i_next_transaction; + + /* List of inodes in the i_transaction [j_list_lock] */ + struct list_head i_list; + + /* VFS inode this inode belongs to [constant during the lifetime + * of the structure] */ + struct inode *i_vfs_inode; + + /* Flags of inode [j_list_lock] */ + unsigned int i_flags; +}; + struct jbd2_revoke_table_s; /** @@ -509,24 +543,12 @@ struct transaction_s struct journal_head *t_reserved_list; /* - * Doubly-linked circular list of all buffers under writeout during - * commit [j_list_lock] - */ - struct journal_head *t_locked_list; - - /* * Doubly-linked circular list of all metadata buffers owned by this * transaction [j_list_lock] */ struct journal_head *t_buffers; /* - * Doubly-linked circular list of all data buffers still to be - * flushed before this transaction can be committed [j_list_lock] - */ - struct journal_head *t_sync_datalist; - - /* * Doubly-linked circular list of all forget buffers (superseded * buffers which we can un-checkpoint once this transaction commits) * [j_list_lock] @@ -565,6 +587,12 @@ struct transaction_s struct journal_head *t_log_list; /* + * List of inodes whose data we've modified in data=ordered mode. + * [j_list_lock] + */ + struct list_head t_inode_list; + + /* * Protects info related to handles */ spinlock_t t_handle_lock; @@ -1004,7 +1032,6 @@ extern int jbd2_journal_extend (handle_t *, int nblocks); extern int jbd2_journal_get_write_access(handle_t *, struct buffer_head *); extern int jbd2_journal_get_create_access (handle_t *, struct buffer_head *); extern int jbd2_journal_get_undo_access(handle_t *, struct buffer_head *); -extern int jbd2_journal_dirty_data (handle_t *, struct buffer_head *); extern int jbd2_journal_dirty_metadata (handle_t *, struct buffer_head *); extern void jbd2_journal_release_buffer (handle_t *, struct buffer_head *); extern int jbd2_journal_forget (handle_t *, struct buffer_head *); @@ -1044,6 +1071,10 @@ extern void jbd2_journal_ack_err (journal_t *); extern int jbd2_journal_clear_err (journal_t *); extern int jbd2_journal_bmap(journal_t *, unsigned long, unsigned long long *); extern int jbd2_journal_force_commit(journal_t *); +extern int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *inode); +extern int jbd2_journal_begin_ordered_truncate(struct jbd2_inode *inode, loff_t new_size); +extern void jbd2_journal_init_jbd_inode(struct jbd2_inode *jinode, struct inode *inode); +extern void jbd2_journal_release_jbd_inode(journal_t *journal, struct jbd2_inode *jinode); /* * journal_head management @@ -1179,15 +1210,13 @@ static inline int jbd_space_needed(journal_t *journal) /* journaling buffer types */ #define BJ_None 0 /* Not journaled */ -#define BJ_SyncData 1 /* Normal data: flush before commit */ -#define BJ_Metadata 2 /* Normal journaled metadata */ -#define BJ_Forget 3 /* Buffer superseded by this transaction */ -#define BJ_IO 4 /* Buffer is for temporary IO use */ -#define BJ_Shadow 5 /* Buffer contents being shadowed to the log */ -#define BJ_LogCtl 6 /* Buffer contains log descriptors */ -#define BJ_Reserved 7 /* Buffer is reserved for access by journal */ -#define BJ_Locked 8 /* Locked for I/O during commit */ -#define BJ_Types 9 +#define BJ_Metadata 1 /* Normal journaled metadata */ +#define BJ_Forget 2 /* Buffer superseded by this transaction */ +#define BJ_IO 3 /* Buffer is for temporary IO use */ +#define BJ_Shadow 4 /* Buffer contents being shadowed to the log */ +#define BJ_LogCtl 5 /* Buffer contains log descriptors */ +#define BJ_Reserved 6 /* Buffer is reserved for access by journal */ +#define BJ_Types 7 extern int jbd_blocks_per_page(struct inode *inode); diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index a744383d16e9..81b3dd5206e0 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -398,7 +398,8 @@ int mlx4_srq_query(struct mlx4_dev *dev, struct mlx4_srq *srq, int *limit_waterm int mlx4_INIT_PORT(struct mlx4_dev *dev, int port); int mlx4_CLOSE_PORT(struct mlx4_dev *dev, int port); -int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]); +int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16], + int block_mcast_loopback); int mlx4_multicast_detach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16]); int mlx4_map_phys_fmr(struct mlx4_dev *dev, struct mlx4_fmr *fmr, u64 *page_list, diff --git a/include/linux/mpage.h b/include/linux/mpage.h index 068a0c9946af..5c42821da2d1 100644 --- a/include/linux/mpage.h +++ b/include/linux/mpage.h @@ -11,11 +11,21 @@ */ #ifdef CONFIG_BLOCK +struct mpage_data { + struct bio *bio; + sector_t last_block_in_bio; + get_block_t *get_block; + unsigned use_writepage; +}; + struct writeback_control; +struct bio *mpage_bio_submit(int rw, struct bio *bio); int mpage_readpages(struct address_space *mapping, struct list_head *pages, unsigned nr_pages, get_block_t get_block); int mpage_readpage(struct page *page, get_block_t get_block); +int __mpage_writepage(struct page *page, struct writeback_control *wbc, + void *data); int mpage_writepages(struct address_space *mapping, struct writeback_control *wbc, get_block_t get_block); int mpage_writepage(struct page *page, get_block_t *get_block, diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h index 9007ccdfc112..208388835357 100644 --- a/include/linux/percpu_counter.h +++ b/include/linux/percpu_counter.h @@ -35,7 +35,7 @@ int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount); void percpu_counter_destroy(struct percpu_counter *fbc); void percpu_counter_set(struct percpu_counter *fbc, s64 amount); void __percpu_counter_add(struct percpu_counter *fbc, s64 amount, s32 batch); -s64 __percpu_counter_sum(struct percpu_counter *fbc); +s64 __percpu_counter_sum(struct percpu_counter *fbc, int set); static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount) { @@ -44,13 +44,19 @@ static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount) static inline s64 percpu_counter_sum_positive(struct percpu_counter *fbc) { - s64 ret = __percpu_counter_sum(fbc); + s64 ret = __percpu_counter_sum(fbc, 0); return ret < 0 ? 0 : ret; } +static inline s64 percpu_counter_sum_and_set(struct percpu_counter *fbc) +{ + return __percpu_counter_sum(fbc, 1); +} + + static inline s64 percpu_counter_sum(struct percpu_counter *fbc) { - return __percpu_counter_sum(fbc); + return __percpu_counter_sum(fbc, 0); } static inline s64 percpu_counter_read(struct percpu_counter *fbc) diff --git a/include/linux/writeback.h b/include/linux/writeback.h index bd91987c065f..12b15c561a1f 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -63,6 +63,7 @@ struct writeback_control { unsigned for_writepages:1; /* This is a writepages() call */ unsigned range_cyclic:1; /* range_start is cyclic */ unsigned more_io:1; /* more io to be dispatched */ + unsigned range_cont:1; }; /* diff --git a/include/rdma/ib_addr.h b/include/rdma/ib_addr.h index c36750ff6ae8..483057b2f4b4 100644 --- a/include/rdma/ib_addr.h +++ b/include/rdma/ib_addr.h @@ -2,29 +2,33 @@ * Copyright (c) 2005 Voltaire Inc. All rights reserved. * Copyright (c) 2005 Intel Corporation. All rights reserved. * - * This Software is licensed under one of the following licenses: + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: * - * 1) under the terms of the "Common Public License 1.0" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/cpl.php. + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: * - * 2) under the terms of the "The BSD License" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/bsd-license.php. + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * 3) under the terms of the "GNU General Public License (GPL) Version 2" a - * copy of which is available from the Open Source Initiative, see - * http://www.opensource.org/licenses/gpl-license.php. - * - * Licensee has the right to choose one of the above licenses. - * - * Redistributions of source code must retain the above copyright - * notice and one of the license notices. - * - * Redistributions in binary form must reproduce both the above copyright - * notice, one of the license notices in the documentation - * and/or other materials provided with the distribution. + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. */ #if !defined(IB_ADDR_H) @@ -57,6 +61,7 @@ struct rdma_dev_addr { unsigned char dst_dev_addr[MAX_ADDR_LEN]; unsigned char broadcast[MAX_ADDR_LEN]; enum rdma_node_type dev_type; + struct net_device *src_dev; }; /** diff --git a/include/rdma/ib_cache.h b/include/rdma/ib_cache.h index f179d233ffc3..00a2b8ec327f 100644 --- a/include/rdma/ib_cache.h +++ b/include/rdma/ib_cache.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_cache.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef _IB_CACHE_H diff --git a/include/rdma/ib_cm.h b/include/rdma/ib_cm.h index a627c8682d2f..ec7c6d99ed3f 100644 --- a/include/rdma/ib_cm.h +++ b/include/rdma/ib_cm.h @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_cm.h 4311 2005-12-05 18:42:01Z sean.hefty $ */ #if !defined(IB_CM_H) #define IB_CM_H diff --git a/include/rdma/ib_fmr_pool.h b/include/rdma/ib_fmr_pool.h index 00dadbf94e1d..f62b842e6596 100644 --- a/include/rdma/ib_fmr_pool.h +++ b/include/rdma/ib_fmr_pool.h @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_fmr_pool.h 2730 2005-06-28 16:43:03Z sean.hefty $ */ #if !defined(IB_FMR_POOL_H) @@ -61,7 +59,7 @@ struct ib_fmr_pool_param { int pool_size; int dirty_watermark; void (*flush_function)(struct ib_fmr_pool *pool, - void * arg); + void *arg); void *flush_arg; unsigned cache:1; }; diff --git a/include/rdma/ib_mad.h b/include/rdma/ib_mad.h index 7228c056b9e9..5f6c40fffcf4 100644 --- a/include/rdma/ib_mad.h +++ b/include/rdma/ib_mad.h @@ -32,11 +32,9 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_mad.h 5596 2006-03-03 01:00:07Z sean.hefty $ */ -#if !defined( IB_MAD_H ) +#if !defined(IB_MAD_H) #define IB_MAD_H #include <linux/list.h> @@ -194,8 +192,7 @@ struct ib_vendor_mad { u8 data[IB_MGMT_VENDOR_DATA]; }; -struct ib_class_port_info -{ +struct ib_class_port_info { u8 base_version; u8 class_version; __be16 capability_mask; @@ -614,11 +611,11 @@ int ib_process_mad_wc(struct ib_mad_agent *mad_agent, * any class specific header, and MAD data area. * If @rmpp_active is set, the RMPP header will be initialized for sending. */ -struct ib_mad_send_buf * ib_create_send_mad(struct ib_mad_agent *mad_agent, - u32 remote_qpn, u16 pkey_index, - int rmpp_active, - int hdr_len, int data_len, - gfp_t gfp_mask); +struct ib_mad_send_buf *ib_create_send_mad(struct ib_mad_agent *mad_agent, + u32 remote_qpn, u16 pkey_index, + int rmpp_active, + int hdr_len, int data_len, + gfp_t gfp_mask); /** * ib_is_mad_class_rmpp - returns whether given management class diff --git a/include/rdma/ib_pack.h b/include/rdma/ib_pack.h index f926020d6331..d7fc45c4eba9 100644 --- a/include/rdma/ib_pack.h +++ b/include/rdma/ib_pack.h @@ -28,8 +28,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_pack.h 1349 2004-12-16 21:09:43Z roland $ */ #ifndef IB_PACK_H diff --git a/include/rdma/ib_sa.h b/include/rdma/ib_sa.h index 942692b0b92e..3841c1aff692 100644 --- a/include/rdma/ib_sa.h +++ b/include/rdma/ib_sa.h @@ -30,8 +30,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_sa.h 2811 2005-07-06 18:11:43Z halr $ */ #ifndef IB_SA_H diff --git a/include/rdma/ib_smi.h b/include/rdma/ib_smi.h index f29af135ba83..aaca0878668f 100644 --- a/include/rdma/ib_smi.h +++ b/include/rdma/ib_smi.h @@ -32,11 +32,9 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_smi.h 1389 2004-12-27 22:56:47Z roland $ */ -#if !defined( IB_SMI_H ) +#if !defined(IB_SMI_H) #define IB_SMI_H #include <rdma/ib_mad.h> diff --git a/include/rdma/ib_user_cm.h b/include/rdma/ib_user_cm.h index 37650afb982c..bd3d380781e0 100644 --- a/include/rdma/ib_user_cm.h +++ b/include/rdma/ib_user_cm.h @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_user_cm.h 4019 2005-11-11 00:33:09Z sean.hefty $ */ #ifndef IB_USER_CM_H diff --git a/include/rdma/ib_user_mad.h b/include/rdma/ib_user_mad.h index 29d2c7205a90..d6fce1cbdb90 100644 --- a/include/rdma/ib_user_mad.h +++ b/include/rdma/ib_user_mad.h @@ -29,8 +29,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_user_mad.h 2814 2005-07-06 19:14:09Z halr $ */ #ifndef IB_USER_MAD_H diff --git a/include/rdma/ib_user_verbs.h b/include/rdma/ib_user_verbs.h index 8d65bf0a625b..a17f77106149 100644 --- a/include/rdma/ib_user_verbs.h +++ b/include/rdma/ib_user_verbs.h @@ -31,8 +31,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_user_verbs.h 4019 2005-11-11 00:33:09Z sean.hefty $ */ #ifndef IB_USER_VERBS_H @@ -291,7 +289,10 @@ struct ib_uverbs_wc { __u32 opcode; __u32 vendor_err; __u32 byte_len; - __u32 imm_data; + union { + __u32 imm_data; + __u32 invalidate_rkey; + } ex; __u32 qp_num; __u32 src_qp; __u32 wc_flags; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 31d30b1852e8..90b529f7a154 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -34,8 +34,6 @@ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. - * - * $Id: ib_verbs.h 1349 2004-12-16 21:09:43Z roland $ */ #if !defined(IB_VERBS_H) @@ -93,7 +91,7 @@ enum ib_device_cap_flags { IB_DEVICE_RC_RNR_NAK_GEN = (1<<12), IB_DEVICE_SRQ_RESIZE = (1<<13), IB_DEVICE_N_NOTIFY_CQ = (1<<14), - IB_DEVICE_ZERO_STAG = (1<<15), + IB_DEVICE_LOCAL_DMA_LKEY = (1<<15), IB_DEVICE_RESERVED = (1<<16), /* old SEND_W_INV */ IB_DEVICE_MEM_WINDOW = (1<<17), /* @@ -105,6 +103,8 @@ enum ib_device_cap_flags { */ IB_DEVICE_UD_IP_CSUM = (1<<18), IB_DEVICE_UD_TSO = (1<<19), + IB_DEVICE_MEM_MGT_EXTENSIONS = (1<<21), + IB_DEVICE_BLOCK_MULTICAST_LOOPBACK = (1<<22), }; enum ib_atomic_cap { @@ -150,6 +150,7 @@ struct ib_device_attr { int max_srq; int max_srq_wr; int max_srq_sge; + unsigned int max_fast_reg_page_list_len; u16 max_pkeys; u8 local_ca_ack_delay; }; @@ -226,6 +227,57 @@ static inline int ib_width_enum_to_int(enum ib_port_width width) } } +struct ib_protocol_stats { + /* TBD... */ +}; + +struct iw_protocol_stats { + u64 ipInReceives; + u64 ipInHdrErrors; + u64 ipInTooBigErrors; + u64 ipInNoRoutes; + u64 ipInAddrErrors; + u64 ipInUnknownProtos; + u64 ipInTruncatedPkts; + u64 ipInDiscards; + u64 ipInDelivers; + u64 ipOutForwDatagrams; + u64 ipOutRequests; + u64 ipOutDiscards; + u64 ipOutNoRoutes; + u64 ipReasmTimeout; + u64 ipReasmReqds; + u64 ipReasmOKs; + u64 ipReasmFails; + u64 ipFragOKs; + u64 ipFragFails; + u64 ipFragCreates; + u64 ipInMcastPkts; + u64 ipOutMcastPkts; + u64 ipInBcastPkts; + u64 ipOutBcastPkts; + + u64 tcpRtoAlgorithm; + u64 tcpRtoMin; + u64 tcpRtoMax; + u64 tcpMaxConn; + u64 tcpActiveOpens; + u64 tcpPassiveOpens; + u64 tcpAttemptFails; + u64 tcpEstabResets; + u64 tcpCurrEstab; + u64 tcpInSegs; + u64 tcpOutSegs; + u64 tcpRetransSegs; + u64 tcpInErrs; + u64 tcpOutRsts; +}; + +union rdma_protocol_stats { + struct ib_protocol_stats ib; + struct iw_protocol_stats iw; +}; + struct ib_port_attr { enum ib_port_state state; enum ib_mtu max_mtu; @@ -413,6 +465,8 @@ enum ib_wc_opcode { IB_WC_FETCH_ADD, IB_WC_BIND_MW, IB_WC_LSO, + IB_WC_LOCAL_INV, + IB_WC_FAST_REG_MR, /* * Set value of IB_WC_RECV so consumers can test if a completion is a * receive by testing (opcode & IB_WC_RECV). @@ -423,7 +477,8 @@ enum ib_wc_opcode { enum ib_wc_flags { IB_WC_GRH = 1, - IB_WC_WITH_IMM = (1<<1) + IB_WC_WITH_IMM = (1<<1), + IB_WC_WITH_INVALIDATE = (1<<2), }; struct ib_wc { @@ -433,7 +488,10 @@ struct ib_wc { u32 vendor_err; u32 byte_len; struct ib_qp *qp; - __be32 imm_data; + union { + __be32 imm_data; + u32 invalidate_rkey; + } ex; u32 src_qp; int wc_flags; u16 pkey_index; @@ -498,7 +556,8 @@ enum ib_qp_type { }; enum ib_qp_create_flags { - IB_QP_CREATE_IPOIB_UD_LSO = 1 << 0, + IB_QP_CREATE_IPOIB_UD_LSO = 1 << 0, + IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK = 1 << 1, }; struct ib_qp_init_attr { @@ -627,6 +686,9 @@ enum ib_wr_opcode { IB_WR_ATOMIC_FETCH_AND_ADD, IB_WR_LSO, IB_WR_SEND_WITH_INV, + IB_WR_RDMA_READ_WITH_INV, + IB_WR_LOCAL_INV, + IB_WR_FAST_REG_MR, }; enum ib_send_flags { @@ -643,6 +705,12 @@ struct ib_sge { u32 lkey; }; +struct ib_fast_reg_page_list { + struct ib_device *device; + u64 *page_list; + unsigned int max_page_list_len; +}; + struct ib_send_wr { struct ib_send_wr *next; u64 wr_id; @@ -675,6 +743,15 @@ struct ib_send_wr { u16 pkey_index; /* valid for GSI only */ u8 port_num; /* valid for DR SMPs on switch only */ } ud; + struct { + u64 iova_start; + struct ib_fast_reg_page_list *page_list; + unsigned int page_shift; + unsigned int page_list_len; + u32 length; + int access_flags; + u32 rkey; + } fast_reg; } wr; }; @@ -777,7 +854,7 @@ struct ib_cq { struct ib_uobject *uobject; ib_comp_handler comp_handler; void (*event_handler)(struct ib_event *, void *); - void * cq_context; + void *cq_context; int cqe; atomic_t usecnt; /* count number of work queues */ }; @@ -883,7 +960,7 @@ struct ib_dma_mapping_ops { void (*sync_single_for_cpu)(struct ib_device *dev, u64 dma_handle, size_t size, - enum dma_data_direction dir); + enum dma_data_direction dir); void (*sync_single_for_device)(struct ib_device *dev, u64 dma_handle, size_t size, @@ -919,6 +996,8 @@ struct ib_device { struct iw_cm_verbs *iwcm; + int (*get_protocol_stats)(struct ib_device *device, + union rdma_protocol_stats *stats); int (*query_device)(struct ib_device *device, struct ib_device_attr *device_attr); int (*query_port)(struct ib_device *device, @@ -1013,6 +1092,11 @@ struct ib_device { int (*query_mr)(struct ib_mr *mr, struct ib_mr_attr *mr_attr); int (*dereg_mr)(struct ib_mr *mr); + struct ib_mr * (*alloc_fast_reg_mr)(struct ib_pd *pd, + int max_page_list_len); + struct ib_fast_reg_page_list * (*alloc_fast_reg_page_list)(struct ib_device *device, + int page_list_len); + void (*free_fast_reg_page_list)(struct ib_fast_reg_page_list *page_list); int (*rereg_phys_mr)(struct ib_mr *mr, int mr_rereg_mask, struct ib_pd *pd, @@ -1065,6 +1149,7 @@ struct ib_device { char node_desc[64]; __be64 node_guid; + u32 local_dma_lkey; u8 node_type; u8 phys_port_cnt; }; @@ -1807,6 +1892,54 @@ int ib_query_mr(struct ib_mr *mr, struct ib_mr_attr *mr_attr); int ib_dereg_mr(struct ib_mr *mr); /** + * ib_alloc_fast_reg_mr - Allocates memory region usable with the + * IB_WR_FAST_REG_MR send work request. + * @pd: The protection domain associated with the region. + * @max_page_list_len: requested max physical buffer list length to be + * used with fast register work requests for this MR. + */ +struct ib_mr *ib_alloc_fast_reg_mr(struct ib_pd *pd, int max_page_list_len); + +/** + * ib_alloc_fast_reg_page_list - Allocates a page list array + * @device - ib device pointer. + * @page_list_len - size of the page list array to be allocated. + * + * This allocates and returns a struct ib_fast_reg_page_list * and a + * page_list array that is at least page_list_len in size. The actual + * size is returned in max_page_list_len. The caller is responsible + * for initializing the contents of the page_list array before posting + * a send work request with the IB_WC_FAST_REG_MR opcode. + * + * The page_list array entries must be translated using one of the + * ib_dma_*() functions just like the addresses passed to + * ib_map_phys_fmr(). Once the ib_post_send() is issued, the struct + * ib_fast_reg_page_list must not be modified by the caller until the + * IB_WC_FAST_REG_MR work request completes. + */ +struct ib_fast_reg_page_list *ib_alloc_fast_reg_page_list( + struct ib_device *device, int page_list_len); + +/** + * ib_free_fast_reg_page_list - Deallocates a previously allocated + * page list array. + * @page_list - struct ib_fast_reg_page_list pointer to be deallocated. + */ +void ib_free_fast_reg_page_list(struct ib_fast_reg_page_list *page_list); + +/** + * ib_update_fast_reg_key - updates the key portion of the fast_reg MR + * R_Key and L_Key. + * @mr - struct ib_mr pointer to be updated. + * @newkey - new key to be used. + */ +static inline void ib_update_fast_reg_key(struct ib_mr *mr, u8 newkey) +{ + mr->lkey = (mr->lkey & 0xffffff00) | newkey; + mr->rkey = (mr->rkey & 0xffffff00) | newkey; +} + +/** * ib_alloc_mw - Allocates a memory window. * @pd: The protection domain associated with the memory window. */ diff --git a/include/rdma/iw_cm.h b/include/rdma/iw_cm.h index aeefa9b740dc..cbb822e8d791 100644 --- a/include/rdma/iw_cm.h +++ b/include/rdma/iw_cm.h @@ -62,7 +62,7 @@ struct iw_cm_event { struct sockaddr_in remote_addr; void *private_data; u8 private_data_len; - void* provider_data; + void *provider_data; }; /** diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index 010f876f41d8..22bb2e7bab1a 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -2,29 +2,33 @@ * Copyright (c) 2005 Voltaire Inc. All rights reserved. * Copyright (c) 2005 Intel Corporation. All rights reserved. * - * This Software is licensed under one of the following licenses: + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: * - * 1) under the terms of the "Common Public License 1.0" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/cpl.php. + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: * - * 2) under the terms of the "The BSD License" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/bsd-license.php. + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. * - * 3) under the terms of the "GNU General Public License (GPL) Version 2" a - * copy of which is available from the Open Source Initiative, see - * http://www.opensource.org/licenses/gpl-license.php. - * - * Licensee has the right to choose one of the above licenses. - * - * Redistributions of source code must retain the above copyright - * notice and one of the license notices. - * - * Redistributions in binary form must reproduce both the above copyright - * notice, one of the license notices in the documentation - * and/or other materials provided with the distribution. + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. */ #if !defined(RDMA_CM_H) @@ -57,11 +61,11 @@ enum rdma_cm_event_type { }; enum rdma_port_space { - RDMA_PS_SDP = 0x0001, - RDMA_PS_IPOIB= 0x0002, - RDMA_PS_TCP = 0x0106, - RDMA_PS_UDP = 0x0111, - RDMA_PS_SCTP = 0x0183 + RDMA_PS_SDP = 0x0001, + RDMA_PS_IPOIB = 0x0002, + RDMA_PS_TCP = 0x0106, + RDMA_PS_UDP = 0x0111, + RDMA_PS_SCTP = 0x0183 }; struct rdma_addr { diff --git a/include/rdma/rdma_cm_ib.h b/include/rdma/rdma_cm_ib.h index 950424b38f16..2389c3b45404 100644 --- a/include/rdma/rdma_cm_ib.h +++ b/include/rdma/rdma_cm_ib.h @@ -1,29 +1,33 @@ /* * Copyright (c) 2006 Intel Corporation. All rights reserved. * - * This Software is licensed under one of the following licenses: - * - * 1) under the terms of the "Common Public License 1.0" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/cpl.php. - * - * 2) under the terms of the "The BSD License" a copy of which is - * available from the Open Source Initiative, see - * http://www.opensource.org/licenses/bsd-license.php. - * - * 3) under the terms of the "GNU General Public License (GPL) Version 2" a - * copy of which is available from the Open Source Initiative, see - * http://www.opensource.org/licenses/gpl-license.php. - * - * Licensee has the right to choose one of the above licenses. - * - * Redistributions of source code must retain the above copyright - * notice and one of the license notices. - * - * Redistributions in binary form must reproduce both the above copyright - * notice, one of the license notices in the documentation - * and/or other materials provided with the distribution. - * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. */ #if !defined(RDMA_CM_IB_H) diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c index 119174494cb5..4a8ba4bf5f6f 100644 --- a/lib/percpu_counter.c +++ b/lib/percpu_counter.c @@ -52,7 +52,7 @@ EXPORT_SYMBOL(__percpu_counter_add); * Add up all the per-cpu counts, return the result. This is a more accurate * but much slower version of percpu_counter_read_positive() */ -s64 __percpu_counter_sum(struct percpu_counter *fbc) +s64 __percpu_counter_sum(struct percpu_counter *fbc, int set) { s64 ret; int cpu; @@ -62,7 +62,12 @@ s64 __percpu_counter_sum(struct percpu_counter *fbc) for_each_online_cpu(cpu) { s32 *pcount = per_cpu_ptr(fbc->counters, cpu); ret += *pcount; + if (set) + *pcount = 0; } + if (set) + fbc->count = ret; + spin_unlock(&fbc->lock); return ret; } diff --git a/mm/filemap.c b/mm/filemap.c index 1e6a7d34874f..65d9d9e2b755 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -236,11 +236,12 @@ int filemap_fdatawrite(struct address_space *mapping) } EXPORT_SYMBOL(filemap_fdatawrite); -static int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, +int filemap_fdatawrite_range(struct address_space *mapping, loff_t start, loff_t end) { return __filemap_fdatawrite_range(mapping, start, end, WB_SYNC_ALL); } +EXPORT_SYMBOL(filemap_fdatawrite_range); /** * filemap_flush - mostly a non-blocking flush diff --git a/mm/page-writeback.c b/mm/page-writeback.c index b38f700825fc..94c6d8988ab3 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -960,6 +960,9 @@ retry: } if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0)) mapping->writeback_index = index; + + if (wbc->range_cont) + wbc->range_start = index << PAGE_CACHE_SHIFT; return ret; } EXPORT_SYMBOL(write_cache_pages); diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 91200feb3f9c..63f131fc42e4 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -555,15 +555,13 @@ static int selinux_set_mnt_opts(struct super_block *sb, struct task_security_struct *tsec = current->security; struct superblock_security_struct *sbsec = sb->s_security; const char *name = sb->s_type->name; - struct dentry *root = sb->s_root; - struct inode *root_inode = root->d_inode; - struct inode_security_struct *root_isec = root_inode->i_security; + struct inode *inode = sbsec->sb->s_root->d_inode; + struct inode_security_struct *root_isec = inode->i_security; u32 fscontext_sid = 0, context_sid = 0, rootcontext_sid = 0; u32 defcontext_sid = 0; char **mount_options = opts->mnt_opts; int *flags = opts->mnt_opts_flags; int num_opts = opts->num_mnt_opts; - bool can_xattr = false; mutex_lock(&sbsec->lock); @@ -667,24 +665,14 @@ static int selinux_set_mnt_opts(struct super_block *sb, goto out; } - if (strcmp(name, "proc") == 0) + if (strcmp(sb->s_type->name, "proc") == 0) sbsec->proc = 1; - /* - * test if the fs supports xattrs, fs_use might make use of this if the - * fs has no definition in policy. - */ - if (root_inode->i_op->getxattr) { - rc = root_inode->i_op->getxattr(root, XATTR_NAME_SELINUX, NULL, 0); - if (rc >= 0 || rc == -ENODATA) - can_xattr = true; - } - /* Determine the labeling behavior to use for this filesystem type. */ - rc = security_fs_use(name, &sbsec->behavior, &sbsec->sid, can_xattr); + rc = security_fs_use(sb->s_type->name, &sbsec->behavior, &sbsec->sid); if (rc) { printk(KERN_WARNING "%s: security_fs_use(%s) returned %d\n", - __func__, name, rc); + __func__, sb->s_type->name, rc); goto out; } diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h index 44cba2e21dcf..7c543003d653 100644 --- a/security/selinux/include/security.h +++ b/security/selinux/include/security.h @@ -136,7 +136,7 @@ int security_get_allow_unknown(void); #define SECURITY_FS_USE_MNTPOINT 6 /* use mountpoint labeling */ int security_fs_use(const char *fstype, unsigned int *behavior, - u32 *sid, bool can_xattr); + u32 *sid); int security_genfs_sid(const char *fstype, char *name, u16 sclass, u32 *sid); diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 8e42da120101..b52f923ce680 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -1934,8 +1934,7 @@ out: int security_fs_use( const char *fstype, unsigned int *behavior, - u32 *sid, - bool can_xattr) + u32 *sid) { int rc = 0; struct ocontext *c; @@ -1949,7 +1948,6 @@ int security_fs_use( c = c->next; } - /* look for labeling behavior defined in policy */ if (c) { *behavior = c->v.behavior; if (!c->sid[0]) { @@ -1960,23 +1958,14 @@ int security_fs_use( goto out; } *sid = c->sid[0]; - goto out; - } - - /* labeling behavior not in policy, use xattrs if possible */ - if (can_xattr) { - *behavior = SECURITY_FS_USE_XATTR; - *sid = SECINITSID_FS; - goto out; - } - - /* no behavior in policy and can't use xattrs, try GENFS */ - rc = security_genfs_sid(fstype, "/", SECCLASS_DIR, sid); - if (rc) { - *behavior = SECURITY_FS_USE_NONE; - rc = 0; } else { - *behavior = SECURITY_FS_USE_GENFS; + rc = security_genfs_sid(fstype, "/", SECCLASS_DIR, sid); + if (rc) { + *behavior = SECURITY_FS_USE_NONE; + rc = 0; + } else { + *behavior = SECURITY_FS_USE_GENFS; + } } out: |
