From 4a8fcc2cc34b7546584bf9ebd8f0caf433e4736e Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 19 May 2011 20:19:09 -0700 Subject: [PATCH 001/224] [SCSI] target: Fix multi task->task_sg[] chaining logic bug This patch fixes a bug in transport_do_task_sg_chain() used by HW target mode modules with sg_chain() to provide a single sg_next() walkable memory layout for use with pci_map_sg() and friends. This patch addresses an issue with mapping multiple small block max_sector tasks across multiple struct se_task->task_sg[] mappings for HW target mode operation. This was causing OOPs with (cmd->t_task->t_tasks_no > 1) I/O traffic for HW target drivers using transport_do_task_sg_chain(), and has been tested so far with tcm_fc(openfcoe), tcm_qla2xxx, and ib_srpt fabrics with t_tasks_no > 1 IBLOCK backends using a smaller max_sectors to trigger the original issue. Signed-off-by: Nicholas Bellinger Acked-by: Kiran Patil Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/target/target_core_transport.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index 9583b23c9c84..fefe10a2aa94 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -4776,18 +4776,20 @@ void transport_do_task_sg_chain(struct se_cmd *cmd) sg_end_cur->page_link &= ~0x02; sg_chain(sg_head, task_sg_num, sg_head_cur); - sg_count += (task->task_sg_num + 1); - } else sg_count += task->task_sg_num; + task_sg_num = (task->task_sg_num + 1); + } else { + sg_chain(sg_head, task_sg_num, sg_head_cur); + sg_count += task->task_sg_num; + task_sg_num = task->task_sg_num; + } sg_head = sg_head_cur; sg_link = sg_link_cur; - task_sg_num = task->task_sg_num; continue; } sg_head = sg_first = &task->task_sg[0]; sg_link = &task->task_sg[task->task_sg_num]; - task_sg_num = task->task_sg_num; /* * Check for single task.. */ @@ -4798,9 +4800,12 @@ void transport_do_task_sg_chain(struct se_cmd *cmd) */ sg_end = &task->task_sg[task->task_sg_num - 1]; sg_end->page_link &= ~0x02; - sg_count += (task->task_sg_num + 1); - } else sg_count += task->task_sg_num; + task_sg_num = (task->task_sg_num + 1); + } else { + sg_count += task->task_sg_num; + task_sg_num = task->task_sg_num; + } } /* * Setup the starting pointer and total t_tasks_sg_linked_no including @@ -4809,21 +4814,20 @@ void transport_do_task_sg_chain(struct se_cmd *cmd) T_TASK(cmd)->t_tasks_sg_chained = sg_first; T_TASK(cmd)->t_tasks_sg_chained_no = sg_count; - DEBUG_CMD_M("Setup T_TASK(cmd)->t_tasks_sg_chained: %p and" - " t_tasks_sg_chained_no: %u\n", T_TASK(cmd)->t_tasks_sg_chained, + DEBUG_CMD_M("Setup cmd: %p T_TASK(cmd)->t_tasks_sg_chained: %p and" + " t_tasks_sg_chained_no: %u\n", cmd, T_TASK(cmd)->t_tasks_sg_chained, T_TASK(cmd)->t_tasks_sg_chained_no); for_each_sg(T_TASK(cmd)->t_tasks_sg_chained, sg, T_TASK(cmd)->t_tasks_sg_chained_no, i) { - DEBUG_CMD_M("SG: %p page: %p length: %d offset: %d\n", - sg, sg_page(sg), sg->length, sg->offset); + DEBUG_CMD_M("SG[%d]: %p page: %p length: %d offset: %d, magic: 0x%08x\n", + i, sg, sg_page(sg), sg->length, sg->offset, sg->sg_magic); if (sg_is_chain(sg)) DEBUG_CMD_M("SG: %p sg_is_chain=1\n", sg); if (sg_is_last(sg)) DEBUG_CMD_M("SG: %p sg_is_last=1\n", sg); } - } EXPORT_SYMBOL(transport_do_task_sg_chain); From 1e7de68c57daf75ec4b721f101f88cccf029e38c Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 19 May 2011 20:19:10 -0700 Subject: [PATCH 002/224] [SCSI] target: Fix interrupt context bug with stats_lock and core_tmr_alloc_req This patch fixes two bugs wrt to the interrupt context usage of target core with HW target mode drivers. It first converts the usage of struct se_device->stats_lock in transport_get_lun_for_cmd() and core_tmr_lun_reset() to properly use spin_lock_irq() to address an BUG with CONFIG_LOCKDEP_SUPPORT=y enabled. This patch also adds a 'in_interrupt()' check to allow GFP_ATOMIC usage from core_tmr_alloc_req() to fix a 'sleeping in interrupt context' BUG with HW target fabrics that require this logic to function. Signed-off-by: Nicholas Bellinger Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/target/target_core_device.c | 4 ++-- drivers/target/target_core_tmr.c | 7 ++++--- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index d25e20829012..fc10ed4ac493 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -150,13 +150,13 @@ out: { struct se_device *dev = se_lun->lun_se_dev; - spin_lock(&dev->stats_lock); + spin_lock_irq(&dev->stats_lock); dev->num_cmds++; if (se_cmd->data_direction == DMA_TO_DEVICE) dev->write_bytes += se_cmd->data_length; else if (se_cmd->data_direction == DMA_FROM_DEVICE) dev->read_bytes += se_cmd->data_length; - spin_unlock(&dev->stats_lock); + spin_unlock_irq(&dev->stats_lock); } /* diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c index 4a109835e420..59b8b9c5ad72 100644 --- a/drivers/target/target_core_tmr.c +++ b/drivers/target/target_core_tmr.c @@ -55,7 +55,8 @@ struct se_tmr_req *core_tmr_alloc_req( { struct se_tmr_req *tmr; - tmr = kmem_cache_zalloc(se_tmr_req_cache, GFP_KERNEL); + tmr = kmem_cache_zalloc(se_tmr_req_cache, (in_interrupt()) ? + GFP_ATOMIC : GFP_KERNEL); if (!(tmr)) { printk(KERN_ERR "Unable to allocate struct se_tmr_req\n"); return ERR_PTR(-ENOMEM); @@ -398,9 +399,9 @@ int core_tmr_lun_reset( printk(KERN_INFO "LUN_RESET: SCSI-2 Released reservation\n"); } - spin_lock(&dev->stats_lock); + spin_lock_irq(&dev->stats_lock); dev->num_resets++; - spin_unlock(&dev->stats_lock); + spin_unlock_irq(&dev->stats_lock); DEBUG_LR("LUN_RESET: %s for [%s] Complete\n", (preempt_and_abort_list) ? "Preempt" : "TMR", From 42c6951e2f7a665bcb57b92fe3f806ba48152c0e Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 19 May 2011 20:19:11 -0700 Subject: [PATCH 003/224] [SCSI] target: Fix bug with task_sg chained transport_free_dev_tasks release This patch addresses a bug in the target core release path for HW operation where transport_free_dev_tasks() was incorrectly being called from transport_lun_remove_cmd() while releasing a se_cmd reference and calling struct target_core_fabric_ops->queue_data_in(). This would result in a OOPs with HW target mode when the release of se_task->task_sg[] would happen before pci_unmap_sg() can be called in HW target mode fabric module code. This patch addresses the issue by moving transport_free_dev_tasks() from transport_lun_remove_cmd() into transport_generic_free_cmd(), and adding TRANSPORT_FREE_CMD_INTR and transport_generic_free_cmd_intr() to allow se_cmd descriptor release to happen fromfrom within transport_processing_thread() process context when release of se_cmd is not possible from HW interrupt context. Signed-off-by: Nicholas Bellinger Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/target/target_core_transport.c | 13 ++++++++++++- include/target/target_core_base.h | 1 + include/target/target_core_transport.h | 1 + 3 files changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index fefe10a2aa94..3eeb3e2436f1 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -762,7 +762,6 @@ static void transport_lun_remove_cmd(struct se_cmd *cmd) transport_all_task_dev_remove_state(cmd); spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags); - transport_free_dev_tasks(cmd); check_lun: spin_lock_irqsave(&lun->lun_cmd_lock, flags); @@ -2058,6 +2057,13 @@ int transport_generic_handle_tmr( } EXPORT_SYMBOL(transport_generic_handle_tmr); +void transport_generic_free_cmd_intr( + struct se_cmd *cmd) +{ + transport_add_cmd_to_queue(cmd, TRANSPORT_FREE_CMD_INTR); +} +EXPORT_SYMBOL(transport_generic_free_cmd_intr); + static int transport_stop_tasks_for_cmd(struct se_cmd *cmd) { struct se_task *task, *task_tmp; @@ -5301,6 +5307,8 @@ void transport_generic_free_cmd( if (wait_for_tasks && cmd->transport_wait_for_tasks) cmd->transport_wait_for_tasks(cmd, 0, 0); + transport_free_dev_tasks(cmd); + transport_generic_remove(cmd, release_to_pool, session_reinstatement); } @@ -6136,6 +6144,9 @@ get_cmd: case TRANSPORT_REMOVE: transport_generic_remove(cmd, 1, 0); break; + case TRANSPORT_FREE_CMD_INTR: + transport_generic_free_cmd(cmd, 0, 1, 0); + break; case TRANSPORT_PROCESS_TMR: transport_generic_do_tmr(cmd); break; diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h index 1d3b5b2f0dbc..561ac99def5a 100644 --- a/include/target/target_core_base.h +++ b/include/target/target_core_base.h @@ -98,6 +98,7 @@ enum transport_state_table { TRANSPORT_REMOVE = 14, TRANSPORT_FREE = 15, TRANSPORT_NEW_CMD_MAP = 16, + TRANSPORT_FREE_CMD_INTR = 17, }; /* Used for struct se_cmd->se_cmd_flags */ diff --git a/include/target/target_core_transport.h b/include/target/target_core_transport.h index 59aa464f6ee2..24a1c6cb83c3 100644 --- a/include/target/target_core_transport.h +++ b/include/target/target_core_transport.h @@ -172,6 +172,7 @@ extern int transport_generic_handle_cdb_map(struct se_cmd *); extern int transport_generic_handle_data(struct se_cmd *); extern void transport_new_cmd_failure(struct se_cmd *); extern int transport_generic_handle_tmr(struct se_cmd *); +extern void transport_generic_free_cmd_intr(struct se_cmd *); extern void __transport_stop_task_timer(struct se_task *, unsigned long *); extern unsigned char transport_asciihex_to_binaryhex(unsigned char val[2]); extern int transport_generic_map_mem_to_cmd(struct se_cmd *cmd, struct scatterlist *, u32, From ccf4d680f80941f0073a9bc6a5e0ed41496b46e7 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 19 May 2011 20:19:12 -0700 Subject: [PATCH 004/224] [SCSI] target: Fix task->task_execute_queue=1 clear bug + LUN_RESET OOPs This patch fixes a bug where task->task_execute_queue=1 was not being cleared once se_task had been removed from se_device->execute_task_list, resulting in an OOPs in core_tmr_lun_reset() for the task->task_active=0 case where transport_remove_task_from_execute_queue() was incorrectly being called. This patch fixes two cases in transport_get_task_from_execute_queue() and transport_remove_task_from_execute_queue() to properly clear task->task_execute_queue=0 once list_del(&task->t_execute_list) has been called. It also adds an explict check in transport_remove_task_from_execute_queue() to dump_stack + return if called with task->task_execute_queue=0. Signed-off-by: Nicholas Bellinger Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/target/target_core_transport.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index 3eeb3e2436f1..beaf8fa89124 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -1194,6 +1194,7 @@ transport_get_task_from_execute_queue(struct se_device *dev) break; list_del(&task->t_execute_list); + atomic_set(&task->task_execute_queue, 0); atomic_dec(&dev->execute_tasks); return task; @@ -1209,8 +1210,14 @@ void transport_remove_task_from_execute_queue( { unsigned long flags; + if (atomic_read(&task->task_execute_queue) == 0) { + dump_stack(); + return; + } + spin_lock_irqsave(&dev->execute_task_lock, flags); list_del(&task->t_execute_list); + atomic_set(&task->task_execute_queue, 0); atomic_dec(&dev->execute_tasks); spin_unlock_irqrestore(&dev->execute_task_lock, flags); } From 1078da163fd2c1d3cba76cd4018e897a4f98efc5 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 19 May 2011 20:19:13 -0700 Subject: [PATCH 005/224] [SCSI] target: Convert REPORT_LUNs to use int_to_scsilun This patch converts transport_core_report_lun_response() to use drivers/scsi/scsi_scan.c:int_to_scsilun instead of using the struct target_core_fabric_ops->pack_lun() fabric provided API vector. It also removes the tfo->pack_lun check from target_fabric_tf_ops_check() and removes from struct target_core_fabric_ops->pack_lun() from target_core_fabric_ops.h, and the following mainline scsi-misc fabric modules: *) tcm_loop: Drop tcm_loop_pack_lun() usage *) tcm_fc: Drop ft_pack_lun() usage Reported-by: Mike Christie Signed-off-by: Nicholas A. Bellinger Signed-off-by: James Bottomley --- drivers/target/loopback/tcm_loop.c | 13 ------------- drivers/target/target_core_configfs.c | 4 ---- drivers/target/target_core_device.c | 25 +++++-------------------- drivers/target/tcm_fc/tfc_conf.c | 8 -------- include/target/target_core_fabric_ops.h | 1 - 5 files changed, 5 insertions(+), 46 deletions(-) diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c index aed4e464d31c..09681bafe10d 100644 --- a/drivers/target/loopback/tcm_loop.c +++ b/drivers/target/loopback/tcm_loop.c @@ -939,18 +939,6 @@ static u16 tcm_loop_get_fabric_sense_len(void) return 0; } -static u64 tcm_loop_pack_lun(unsigned int lun) -{ - u64 result; - - /* LSB of lun into byte 1 big-endian */ - result = ((lun & 0xff) << 8); - /* use flat space addressing method */ - result |= 0x40 | ((lun >> 8) & 0x3f); - - return cpu_to_le64(result); -} - static char *tcm_loop_dump_proto_id(struct tcm_loop_hba *tl_hba) { switch (tl_hba->tl_proto_id) { @@ -1481,7 +1469,6 @@ static int tcm_loop_register_configfs(void) fabric->tf_ops.set_fabric_sense_len = &tcm_loop_set_fabric_sense_len; fabric->tf_ops.get_fabric_sense_len = &tcm_loop_get_fabric_sense_len; fabric->tf_ops.is_state_remove = &tcm_loop_is_state_remove; - fabric->tf_ops.pack_lun = &tcm_loop_pack_lun; tf_cg = &fabric->tf_group; /* diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index a5f44a6e6e1d..ee6fad979b50 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -497,10 +497,6 @@ static int target_fabric_tf_ops_check( printk(KERN_ERR "Missing tfo->is_state_remove()\n"); return -EINVAL; } - if (!(tfo->pack_lun)) { - printk(KERN_ERR "Missing tfo->pack_lun()\n"); - return -EINVAL; - } /* * We at least require tfo->fabric_make_wwn(), tfo->fabric_drop_wwn() * tfo->fabric_make_tpg() and tfo->fabric_drop_tpg() in diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index fc10ed4ac493..8407f9ca2b31 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include @@ -658,8 +659,7 @@ int transport_core_report_lun_response(struct se_cmd *se_cmd) struct se_session *se_sess = SE_SESS(se_cmd); struct se_task *se_task; unsigned char *buf = (unsigned char *)T_TASK(se_cmd)->t_task_buf; - u32 cdb_offset = 0, lun_count = 0, offset = 8; - u64 i, lun; + u32 cdb_offset = 0, lun_count = 0, offset = 8, i; list_for_each_entry(se_task, &T_TASK(se_cmd)->t_task_list, t_list) break; @@ -675,15 +675,7 @@ int transport_core_report_lun_response(struct se_cmd *se_cmd) * a $FABRIC_MOD. In that case, report LUN=0 only. */ if (!(se_sess)) { - lun = 0; - buf[offset++] = ((lun >> 56) & 0xff); - buf[offset++] = ((lun >> 48) & 0xff); - buf[offset++] = ((lun >> 40) & 0xff); - buf[offset++] = ((lun >> 32) & 0xff); - buf[offset++] = ((lun >> 24) & 0xff); - buf[offset++] = ((lun >> 16) & 0xff); - buf[offset++] = ((lun >> 8) & 0xff); - buf[offset++] = (lun & 0xff); + int_to_scsilun(0, (struct scsi_lun *)&buf[offset]); lun_count = 1; goto done; } @@ -703,15 +695,8 @@ int transport_core_report_lun_response(struct se_cmd *se_cmd) if ((cdb_offset + 8) >= se_cmd->data_length) continue; - lun = cpu_to_be64(CMD_TFO(se_cmd)->pack_lun(deve->mapped_lun)); - buf[offset++] = ((lun >> 56) & 0xff); - buf[offset++] = ((lun >> 48) & 0xff); - buf[offset++] = ((lun >> 40) & 0xff); - buf[offset++] = ((lun >> 32) & 0xff); - buf[offset++] = ((lun >> 24) & 0xff); - buf[offset++] = ((lun >> 16) & 0xff); - buf[offset++] = ((lun >> 8) & 0xff); - buf[offset++] = (lun & 0xff); + int_to_scsilun(deve->mapped_lun, (struct scsi_lun *)&buf[offset]); + offset += 8; cdb_offset += 8; } spin_unlock_irq(&SE_NODE_ACL(se_sess)->device_list_lock); diff --git a/drivers/target/tcm_fc/tfc_conf.c b/drivers/target/tcm_fc/tfc_conf.c index fcdbbffe88cc..84e868c255dd 100644 --- a/drivers/target/tcm_fc/tfc_conf.c +++ b/drivers/target/tcm_fc/tfc_conf.c @@ -519,13 +519,6 @@ static u32 ft_tpg_get_inst_index(struct se_portal_group *se_tpg) return tpg->index; } -static u64 ft_pack_lun(unsigned int index) -{ - WARN_ON(index >= 256); - /* Caller wants this byte-swapped */ - return cpu_to_le64((index & 0xff) << 8); -} - static struct target_core_fabric_ops ft_fabric_ops = { .get_fabric_name = ft_get_fabric_name, .get_fabric_proto_ident = fc_get_fabric_proto_ident, @@ -564,7 +557,6 @@ static struct target_core_fabric_ops ft_fabric_ops = { .get_fabric_sense_len = ft_get_fabric_sense_len, .set_fabric_sense_len = ft_set_fabric_sense_len, .is_state_remove = ft_is_state_remove, - .pack_lun = ft_pack_lun, /* * Setup function pointers for generic logic in * target_core_fabric_configfs.c diff --git a/include/target/target_core_fabric_ops.h b/include/target/target_core_fabric_ops.h index dc78f77f9450..747e1404dca0 100644 --- a/include/target/target_core_fabric_ops.h +++ b/include/target/target_core_fabric_ops.h @@ -77,7 +77,6 @@ struct target_core_fabric_ops { u16 (*set_fabric_sense_len)(struct se_cmd *, u32); u16 (*get_fabric_sense_len)(void); int (*is_state_remove)(struct se_cmd *); - u64 (*pack_lun)(unsigned int); /* * fabric module calls for target_core_fabric_configfs.c */ From 61db1802bf33bf027cd97ba3f79566b2b2fce5c6 Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 19 May 2011 20:19:14 -0700 Subject: [PATCH 006/224] [SCSI] target: Convert TASK_ATTR to scsi_tcq.h definitions This patch converts target core and follwing scsi-misc upstream fabric modules to use include/scsi/scsi_tcq.h includes for SIMPLE, HEAD_OF_QUEUE and ORDERED SCSI tasks instead of scsi/libsas.h with TASK_ATTR* *) tcm_loop: Convert tcm_loop_allocate_core_cmd() + tcm_loop_device_reset() to scsi_tcq.h *) tcm_fc: Convert ft_send_cmd() from FCP_PTA_* to scsi_tcq.h Reported-by: Christoph Hellwig Signed-off-by: Nicholas A. Bellinger Signed-off-by: James Bottomley --- drivers/target/loopback/tcm_loop.c | 12 ++++++------ drivers/target/target_core_pscsi.c | 4 ++-- drivers/target/target_core_transport.c | 22 +++++++++++----------- drivers/target/tcm_fc/tfc_cmd.c | 20 +++++++++++++++++++- include/scsi/scsi_tcq.h | 1 + 5 files changed, 39 insertions(+), 20 deletions(-) diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c index 09681bafe10d..dee2a2c909f5 100644 --- a/drivers/target/loopback/tcm_loop.c +++ b/drivers/target/loopback/tcm_loop.c @@ -31,7 +31,7 @@ #include #include #include -#include /* For TASK_ATTR_* */ +#include #include #include @@ -95,17 +95,17 @@ static struct se_cmd *tcm_loop_allocate_core_cmd( if (sc->device->tagged_supported) { switch (sc->tag) { case HEAD_OF_QUEUE_TAG: - sam_task_attr = TASK_ATTR_HOQ; + sam_task_attr = MSG_HEAD_TAG; break; case ORDERED_QUEUE_TAG: - sam_task_attr = TASK_ATTR_ORDERED; + sam_task_attr = MSG_ORDERED_TAG; break; default: - sam_task_attr = TASK_ATTR_SIMPLE; + sam_task_attr = MSG_SIMPLE_TAG; break; } } else - sam_task_attr = TASK_ATTR_SIMPLE; + sam_task_attr = MSG_SIMPLE_TAG; /* * Initialize struct se_cmd descriptor from target_core_mod infrastructure @@ -379,7 +379,7 @@ static int tcm_loop_device_reset(struct scsi_cmnd *sc) * Initialize struct se_cmd descriptor from target_core_mod infrastructure */ transport_init_se_cmd(se_cmd, se_tpg->se_tpg_tfo, se_sess, 0, - DMA_NONE, TASK_ATTR_SIMPLE, + DMA_NONE, MSG_SIMPLE_TAG, &tl_cmd->tl_sense_buf[0]); /* * Allocate the LUN_RESET TMR diff --git a/drivers/target/target_core_pscsi.c b/drivers/target/target_core_pscsi.c index 7ff6a35f26ac..331d423fd0e0 100644 --- a/drivers/target/target_core_pscsi.c +++ b/drivers/target/target_core_pscsi.c @@ -41,7 +41,7 @@ #include #include #include -#include /* For TASK_ATTR_* */ +#include #include #include @@ -911,7 +911,7 @@ static int pscsi_do_task(struct se_task *task) * descriptor */ blk_execute_rq_nowait(pdv->pdv_sd->request_queue, NULL, pt->pscsi_req, - (task->task_se_cmd->sam_task_attr == TASK_ATTR_HOQ), + (task->task_se_cmd->sam_task_attr == MSG_HEAD_TAG), pscsi_req_done); return PYX_TRANSPORT_SENT_TO_TRANSPORT; diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index beaf8fa89124..5b1ce81e3aa7 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -42,7 +42,7 @@ #include #include #include -#include /* For TASK_ATTR_* */ +#include #include #include @@ -1074,7 +1074,7 @@ static inline int transport_add_task_check_sam_attr( * head of the struct se_device->execute_task_list, and task_prev * after that for each subsequent task */ - if (task->task_se_cmd->sam_task_attr == TASK_ATTR_HOQ) { + if (task->task_se_cmd->sam_task_attr == MSG_HEAD_TAG) { list_add(&task->t_execute_list, (task_prev != NULL) ? &task_prev->t_execute_list : @@ -1873,7 +1873,7 @@ static int transport_check_alloc_task_attr(struct se_cmd *cmd) if (SE_DEV(cmd)->dev_task_attr_type != SAM_TASK_ATTR_EMULATED) return 0; - if (cmd->sam_task_attr == TASK_ATTR_ACA) { + if (cmd->sam_task_attr == MSG_ACA_TAG) { DEBUG_STA("SAM Task Attribute ACA" " emulation is not supported\n"); return -1; @@ -2517,7 +2517,7 @@ static inline int transport_execute_task_attr(struct se_cmd *cmd) * Check for the existence of HEAD_OF_QUEUE, and if true return 1 * to allow the passed struct se_cmd list of tasks to the front of the list. */ - if (cmd->sam_task_attr == TASK_ATTR_HOQ) { + if (cmd->sam_task_attr == MSG_HEAD_TAG) { atomic_inc(&SE_DEV(cmd)->dev_hoq_count); smp_mb__after_atomic_inc(); DEBUG_STA("Added HEAD_OF_QUEUE for CDB:" @@ -2525,7 +2525,7 @@ static inline int transport_execute_task_attr(struct se_cmd *cmd) T_TASK(cmd)->t_task_cdb[0], cmd->se_ordered_id); return 1; - } else if (cmd->sam_task_attr == TASK_ATTR_ORDERED) { + } else if (cmd->sam_task_attr == MSG_ORDERED_TAG) { spin_lock(&SE_DEV(cmd)->ordered_cmd_lock); list_add_tail(&cmd->se_ordered_list, &SE_DEV(cmd)->ordered_cmd_list); @@ -3424,7 +3424,7 @@ static int transport_generic_cmd_sequencer( * See spc4r17 section 5.3 */ if (SE_DEV(cmd)->dev_task_attr_type == SAM_TASK_ATTR_EMULATED) - cmd->sam_task_attr = TASK_ATTR_HOQ; + cmd->sam_task_attr = MSG_HEAD_TAG; cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB; break; case READ_BUFFER: @@ -3632,7 +3632,7 @@ static int transport_generic_cmd_sequencer( * See spc4r17 section 5.3 */ if (SE_DEV(cmd)->dev_task_attr_type == SAM_TASK_ATTR_EMULATED) - cmd->sam_task_attr = TASK_ATTR_HOQ; + cmd->sam_task_attr = MSG_HEAD_TAG; cmd->se_cmd_flags |= SCF_SCSI_CONTROL_NONSG_IO_CDB; break; default: @@ -3790,21 +3790,21 @@ static void transport_complete_task_attr(struct se_cmd *cmd) struct se_cmd *cmd_p, *cmd_tmp; int new_active_tasks = 0; - if (cmd->sam_task_attr == TASK_ATTR_SIMPLE) { + if (cmd->sam_task_attr == MSG_SIMPLE_TAG) { atomic_dec(&dev->simple_cmds); smp_mb__after_atomic_dec(); dev->dev_cur_ordered_id++; DEBUG_STA("Incremented dev->dev_cur_ordered_id: %u for" " SIMPLE: %u\n", dev->dev_cur_ordered_id, cmd->se_ordered_id); - } else if (cmd->sam_task_attr == TASK_ATTR_HOQ) { + } else if (cmd->sam_task_attr == MSG_HEAD_TAG) { atomic_dec(&dev->dev_hoq_count); smp_mb__after_atomic_dec(); dev->dev_cur_ordered_id++; DEBUG_STA("Incremented dev_cur_ordered_id: %u for" " HEAD_OF_QUEUE: %u\n", dev->dev_cur_ordered_id, cmd->se_ordered_id); - } else if (cmd->sam_task_attr == TASK_ATTR_ORDERED) { + } else if (cmd->sam_task_attr == MSG_ORDERED_TAG) { spin_lock(&dev->ordered_cmd_lock); list_del(&cmd->se_ordered_list); atomic_dec(&dev->dev_ordered_sync); @@ -3837,7 +3837,7 @@ static void transport_complete_task_attr(struct se_cmd *cmd) new_active_tasks++; spin_lock(&dev->delayed_cmd_lock); - if (cmd_p->sam_task_attr == TASK_ATTR_ORDERED) + if (cmd_p->sam_task_attr == MSG_ORDERED_TAG) break; } spin_unlock(&dev->delayed_cmd_lock); diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c index 49e51778f733..c056a1132ae1 100644 --- a/drivers/target/tcm_fc/tfc_cmd.c +++ b/drivers/target/tcm_fc/tfc_cmd.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include @@ -592,8 +593,25 @@ static void ft_send_cmd(struct ft_cmd *cmd) case FCP_CFL_WRDATA | FCP_CFL_RDDATA: goto err; /* TBD not supported by tcm_fc yet */ } + /* + * Locate the SAM Task Attr from fc_pri_ta + */ + switch (fcp->fc_pri_ta & FCP_PTA_MASK) { + case FCP_PTA_HEADQ: + task_attr = MSG_HEAD_TAG; + break; + case FCP_PTA_ORDERED: + task_attr = MSG_ORDERED_TAG; + break; + case FCP_PTA_ACA: + task_attr = MSG_ACA_TAG; + break; + case FCP_PTA_SIMPLE: /* Fallthrough */ + default: + task_attr = MSG_SIMPLE_TAG; + } + - /* FCP_PTA_ maps 1:1 to TASK_ATTR_ */ task_attr = fcp->fc_pri_ta & FCP_PTA_MASK; data_len = ntohl(fcp->fc_dl); cmd->cdb = fcp->fc_cdb; diff --git a/include/scsi/scsi_tcq.h b/include/scsi/scsi_tcq.h index d6e7994aa634..81dd12edc38c 100644 --- a/include/scsi/scsi_tcq.h +++ b/include/scsi/scsi_tcq.h @@ -9,6 +9,7 @@ #define MSG_SIMPLE_TAG 0x20 #define MSG_HEAD_TAG 0x21 #define MSG_ORDERED_TAG 0x22 +#define MSG_ACA_TAG 0x24 /* unsupported */ #define SCSI_NO_TAG (-1) /* identify no tag in use */ From 5aceca9d3cbdacbd017712513387d930f9f944d9 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 23 May 2011 17:12:22 -0700 Subject: [PATCH 007/224] PCI: Fix warning in drivers/pci/probe.c on sparc64 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit IO_SPACE_LIMIT is currently used in two ways: 1) As a way to mask I/O port values read out of PCI base address registers. This value should be 64-bit. 2) As a value which is the upper limit for all I/O "ports" in the system. On sparc64 we store the full 64-bit physical I/O address in the resources. For this reason we define IO_SPACE_LIMIT at a 64-bit "all 1's". This is the right value to use for ioport_resource.end and for the check made in drivers/pcmcia/rsrc_nonstatic.c:adjust_io(). But in driver/pci/probe.c:__pci_read_base() we mask this against a "u32" variable and thus get the following warning: drivers/pci/probe.c: In function ¡__pci_read_base¢: drivers/pci/probe.c:207: warning: large integer implicitly truncated to unsigned type Fix this by using an explicit "u32" cast. I considered changing sparc64 to define a 32-bit "all 1's" like most other systems do, but this wouldn't work because the checks in PCMCIA's rsrc_nonstatic.c would no longer be right since they are testing against fully formed 64-bit resources. As described above, on sparc64 such resources will hold full 64-bit physical I/O addresses, not bus-centric 32-bit ones. Signed-off-by: David S. Miller Signed-off-by: Jesse Barnes --- drivers/pci/probe.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 48849ffdd672..bafb3c3d4a89 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -168,7 +168,7 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type, res->flags |= pci_calc_resource_flags(l) | IORESOURCE_SIZEALIGN; if (type == pci_bar_io) { l &= PCI_BASE_ADDRESS_IO_MASK; - mask = PCI_BASE_ADDRESS_IO_MASK & IO_SPACE_LIMIT; + mask = PCI_BASE_ADDRESS_IO_MASK & (u32) IO_SPACE_LIMIT; } else { l &= PCI_BASE_ADDRESS_MEM_MASK; mask = (u32)PCI_BASE_ADDRESS_MEM_MASK; From 3f37d6229ca309f96b163b943ff982f4697630cd Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 25 May 2011 19:21:25 -0700 Subject: [PATCH 008/224] PCI: fix new kernel-doc warning Fix pci.c kernel-doc warnings: Warning(drivers/pci/pci.c:3292): No description found for parameter 'flags' Warning(drivers/pci/pci.c:3292): Excess function parameter 'change_bridge_flags' description in 'pci_set_vga_state' Signed-off-by: Randy Dunlap Signed-off-by: Jesse Barnes --- drivers/pci/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 56098b3e17c0..e43a7bbeb28f 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -3284,7 +3284,7 @@ static int pci_set_vga_state_arch(struct pci_dev *dev, bool decode, * @dev: the PCI device * @decode: true = enable decoding, false = disable decoding * @command_bits: PCI_COMMAND_IO and/or PCI_COMMAND_MEMORY - * @change_bridge_flags: traverse ancestors and change bridges + * @flags: traverse ancestors and change bridges * CHANGE_BRIDGE_ONLY / CHANGE_BRIDGE */ int pci_set_vga_state(struct pci_dev *dev, bool decode, From 6e33a852a37dee02979ec9d82bea26c07cee5bce Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?M=C3=A1rton=20N=C3=A9meth?= Date: Sat, 14 May 2011 19:27:33 +0200 Subject: [PATCH 009/224] x86/PCI/ACPI: fix type mismatch MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The flags field of struct resource from linux/ioport.h is "unsigned long". Change the "type" parameter of coalesce_windows() function to match that field. This fixes the following warning messages when compiling with "make C=1 W=1 bzImage modules": arch/x86/pci/acpi.c: In function ‘coalesce_windows’: arch/x86/pci/acpi.c:198: warning: conversion to ‘long unsigned int’ from ‘int’ may change the sign of the result arch/x86/pci/acpi.c:203: warning: conversion to ‘long unsigned int’ from ‘int’ may change the sign of the result Signed-off-by: Márton Németh Signed-off-by: Jesse Barnes --- arch/x86/pci/acpi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index 0972315c3860..68c3c1395202 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -188,7 +188,7 @@ static bool resource_contains(struct resource *res, resource_size_t point) return false; } -static void coalesce_windows(struct pci_root_info *info, int type) +static void coalesce_windows(struct pci_root_info *info, unsigned long type) { int i, j; struct resource *res1, *res2; From eb7073db1076777496495483854993165e14790f Mon Sep 17 00:00:00 2001 From: Tomoya MORINAGA Date: Thu, 2 Jun 2011 11:31:29 +0900 Subject: [PATCH 010/224] 8250_pci: add -ENODEV code for Intel EG20T PCH Intel EG20T PCH has UART device which is compatible with 8250. Currently, with general configuration, the PCH UART driver is not loaded but 8250 standard driver is loaded. Therefore, in case of using PCH UART driver, need to disable 8250 pci function. However, this procedure is not best solution. This patch, in 8250_pci, if the device is the PCH or the family IOH, '-ENODEV' is returned. As a result, disabling 8250-pci processing becomes unnecessary. Signed-off-by: Tomoya MORINAGA Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250_pci.c | 59 +++++++++++++++++++++++++++++++++++ 1 file changed, 59 insertions(+) diff --git a/drivers/tty/serial/8250_pci.c b/drivers/tty/serial/8250_pci.c index 4b4968a294b2..d7dc513451a6 100644 --- a/drivers/tty/serial/8250_pci.c +++ b/drivers/tty/serial/8250_pci.c @@ -994,6 +994,15 @@ static int skip_tx_en_setup(struct serial_private *priv, return pci_default_setup(priv, board, port, idx); } +static int pci_eg20t_init(struct pci_dev *dev) +{ +#if defined(CONFIG_SERIAL_PCH_UART) || defined(CONFIG_SERIAL_PCH_UART_MODULE) + return -ENODEV; +#else + return 0; +#endif +} + /* This should be in linux/pci_ids.h */ #define PCI_VENDOR_ID_SBSMODULARIO 0x124B #define PCI_SUBVENDOR_ID_SBSMODULARIO 0x124B @@ -1446,6 +1455,56 @@ static struct pci_serial_quirk pci_serial_quirks[] __refdata = { .init = pci_oxsemi_tornado_init, .setup = pci_default_setup, }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = 0x8811, + .init = pci_eg20t_init, + }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = 0x8812, + .init = pci_eg20t_init, + }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = 0x8813, + .init = pci_eg20t_init, + }, + { + .vendor = PCI_VENDOR_ID_INTEL, + .device = 0x8814, + .init = pci_eg20t_init, + }, + { + .vendor = 0x10DB, + .device = 0x8027, + .init = pci_eg20t_init, + }, + { + .vendor = 0x10DB, + .device = 0x8028, + .init = pci_eg20t_init, + }, + { + .vendor = 0x10DB, + .device = 0x8029, + .init = pci_eg20t_init, + }, + { + .vendor = 0x10DB, + .device = 0x800C, + .init = pci_eg20t_init, + }, + { + .vendor = 0x10DB, + .device = 0x800D, + .init = pci_eg20t_init, + }, + { + .vendor = 0x10DB, + .device = 0x800D, + .init = pci_eg20t_init, + }, /* * Cronyx Omega PCI (PLX-chip based) */ From 5def1360252b974faeb438775c19c14338bc1903 Mon Sep 17 00:00:00 2001 From: Yongqiang Yang Date: Sun, 5 Jun 2011 23:26:40 -0400 Subject: [PATCH 011/224] ext4: correct comments for ext4_free_blocks() metadata is not parameter of ext4_free_blocks() any more. Signed-off-by: Yongqiang Yang Signed-off-by: "Theodore Ts'o" --- fs/ext4/mballoc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 859f2ae8864e..eb71dd59f2e4 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -4448,7 +4448,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, * @inode: inode * @block: start physical block to free * @count: number of blocks to count - * @metadata: Are these metadata blocks + * @flags: flags used by ext4_free_blocks */ void ext4_free_blocks(handle_t *handle, struct inode *inode, struct buffer_head *bh, ext4_fsblk_t block, From ab4bd22d3cce6977dc039664cc2d052e3147d662 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 5 Jun 2011 06:01:13 +0200 Subject: [PATCH 012/224] cfq-iosched: fix locking around ioc->ioc_data assignment Since we are modifying this RCU pointer, we need to hold the lock protecting it around it. This fixes a potential reuse and double free of a cfq io_context structure. The bug has been in CFQ for a long time, it hit very few people but those it did hit seemed to see it a lot. Tracked in RH bugzilla here: https://bugzilla.redhat.com/show_bug.cgi?id=577968 Credit goes to Paul Bolle for figuring out that the issue was around the one-hit ioc->ioc_data cache. Thanks to his hard work the issue is now fixed. Cc: stable@kernel.org Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 3c7b537bf908..545b8d4c829d 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -2772,8 +2772,11 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd, smp_wmb(); cic->key = cfqd_dead_key(cfqd); - if (ioc->ioc_data == cic) + if (rcu_dereference(ioc->ioc_data) == cic) { + spin_lock(&ioc->lock); rcu_assign_pointer(ioc->ioc_data, NULL); + spin_unlock(&ioc->lock); + } if (cic->cfqq[BLK_RW_ASYNC]) { cfq_exit_cfqq(cfqd, cic->cfqq[BLK_RW_ASYNC]); From f17722f917b2f21497deb6edc62fb1683daa08e6 Mon Sep 17 00:00:00 2001 From: Lukas Czerner Date: Mon, 6 Jun 2011 00:05:17 -0400 Subject: [PATCH 013/224] ext4: Fix max file size and logical block counting of extent format file Kazuya Mio reported that he was able to hit BUG_ON(next == lblock) in ext4_ext_put_gap_in_cache() while creating a sparse file in extent format and fill the tail of file up to its end. We will hit the BUG_ON when we write the last block (2^32-1) into the sparse file. The root cause of the problem lies in the fact that we specifically set s_maxbytes so that block at s_maxbytes fit into on-disk extent format, which is 32 bit long. However, we are not storing start and end block number, but rather start block number and length in blocks. It means that in order to cover extent from 0 to EXT_MAX_BLOCK we need EXT_MAX_BLOCK+1 to fit into len (because we counting block 0 as well) - and it does not. The only way to fix it without changing the meaning of the struct ext4_extent members is, as Kazuya Mio suggested, to lower s_maxbytes by one fs block so we can cover the whole extent we can get by the on-disk extent format. Also in many places EXT_MAX_BLOCK is used as length instead of maximum logical block number as the name suggests, it is all a bit messy. So this commit renames it to EXT_MAX_BLOCKS and change its usage in some places to actually be maximum number of blocks in the extent. The bug which this commit fixes can be reproduced as follows: dd if=/dev/zero of=/mnt/mp1/file bs= count=1 seek=$((2**32-2)) sync dd if=/dev/zero of=/mnt/mp1/file bs= count=1 seek=$((2**32-1)) Reported-by: Kazuya Mio Signed-off-by: Lukas Czerner Signed-off-by: "Theodore Ts'o" --- fs/ext4/ext4_extents.h | 7 +++++-- fs/ext4/extents.c | 34 +++++++++++++++++----------------- fs/ext4/move_extent.c | 10 +++++----- fs/ext4/super.c | 15 ++++++++++++--- 4 files changed, 39 insertions(+), 27 deletions(-) diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index 2e29abb30f76..476414644387 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -133,8 +133,11 @@ typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, #define EXT_BREAK 1 #define EXT_REPEAT 2 -/* Maximum logical block in a file; ext4_extent's ee_block is __le32 */ -#define EXT_MAX_BLOCK 0xffffffff +/* + * Maximum number of logical blocks in a file; ext4_extent's ee_block is + * __le32. + */ +#define EXT_MAX_BLOCKS 0xffffffff /* * EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 5199bac7fc62..41575704600a 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1408,7 +1408,7 @@ got_index: /* * ext4_ext_next_allocated_block: - * returns allocated block in subsequent extent or EXT_MAX_BLOCK. + * returns allocated block in subsequent extent or EXT_MAX_BLOCKS. * NOTE: it considers block number from index entry as * allocated block. Thus, index entries have to be consistent * with leaves. @@ -1422,7 +1422,7 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) depth = path->p_depth; if (depth == 0 && path->p_ext == NULL) - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; while (depth >= 0) { if (depth == path->p_depth) { @@ -1439,12 +1439,12 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path) depth--; } - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; } /* * ext4_ext_next_leaf_block: - * returns first allocated block from next leaf or EXT_MAX_BLOCK + * returns first allocated block from next leaf or EXT_MAX_BLOCKS */ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, struct ext4_ext_path *path) @@ -1456,7 +1456,7 @@ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, /* zero-tree has no leaf blocks at all */ if (depth == 0) - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; /* go to index block */ depth--; @@ -1469,7 +1469,7 @@ static ext4_lblk_t ext4_ext_next_leaf_block(struct inode *inode, depth--; } - return EXT_MAX_BLOCK; + return EXT_MAX_BLOCKS; } /* @@ -1677,13 +1677,13 @@ static unsigned int ext4_ext_check_overlap(struct inode *inode, */ if (b2 < b1) { b2 = ext4_ext_next_allocated_block(path); - if (b2 == EXT_MAX_BLOCK) + if (b2 == EXT_MAX_BLOCKS) goto out; } /* check for wrap through zero on extent logical start block*/ if (b1 + len1 < b1) { - len1 = EXT_MAX_BLOCK - b1; + len1 = EXT_MAX_BLOCKS - b1; newext->ee_len = cpu_to_le16(len1); ret = 1; } @@ -1767,7 +1767,7 @@ repeat: fex = EXT_LAST_EXTENT(eh); next = ext4_ext_next_leaf_block(inode, path); if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block) - && next != EXT_MAX_BLOCK) { + && next != EXT_MAX_BLOCKS) { ext_debug("next leaf block - %d\n", next); BUG_ON(npath != NULL); npath = ext4_ext_find_extent(inode, next, NULL); @@ -1887,7 +1887,7 @@ static int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block, BUG_ON(func == NULL); BUG_ON(inode == NULL); - while (block < last && block != EXT_MAX_BLOCK) { + while (block < last && block != EXT_MAX_BLOCKS) { num = last - block; /* find extent for this block */ down_read(&EXT4_I(inode)->i_data_sem); @@ -2020,7 +2020,7 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, if (ex == NULL) { /* there is no extent yet, so gap is [0;-] */ lblock = 0; - len = EXT_MAX_BLOCK; + len = EXT_MAX_BLOCKS; ext_debug("cache gap(whole file):"); } else if (block < le32_to_cpu(ex->ee_block)) { lblock = block; @@ -2350,7 +2350,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, * never happen because at least one of the end points * needs to be on the edge of the extent. */ - if (end == EXT_MAX_BLOCK) { + if (end == EXT_MAX_BLOCKS - 1) { ext_debug(" bad truncate %u:%u\n", start, end); block = 0; @@ -2398,7 +2398,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, * If this is a truncate, this condition * should never happen */ - if (end == EXT_MAX_BLOCK) { + if (end == EXT_MAX_BLOCKS - 1) { ext_debug(" bad truncate %u:%u\n", start, end); err = -EIO; @@ -2478,7 +2478,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, * we need to remove it from the leaf */ if (num == 0) { - if (end != EXT_MAX_BLOCK) { + if (end != EXT_MAX_BLOCKS - 1) { /* * For hole punching, we need to scoot all the * extents up when an extent is removed so that @@ -3699,7 +3699,7 @@ void ext4_ext_truncate(struct inode *inode) last_block = (inode->i_size + sb->s_blocksize - 1) >> EXT4_BLOCK_SIZE_BITS(sb); - err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCK); + err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1); /* In a multi-transaction truncate, we only make the final * transaction synchronous. @@ -4347,8 +4347,8 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, start_blk = start >> inode->i_sb->s_blocksize_bits; last_blk = (start + len - 1) >> inode->i_sb->s_blocksize_bits; - if (last_blk >= EXT_MAX_BLOCK) - last_blk = EXT_MAX_BLOCK-1; + if (last_blk >= EXT_MAX_BLOCKS) + last_blk = EXT_MAX_BLOCKS-1; len_blks = ((ext4_lblk_t) last_blk) - start_blk + 1; /* diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c index 2b8304bf3c50..f57455a1b1b2 100644 --- a/fs/ext4/move_extent.c +++ b/fs/ext4/move_extent.c @@ -1002,12 +1002,12 @@ mext_check_arguments(struct inode *orig_inode, return -EINVAL; } - if ((orig_start > EXT_MAX_BLOCK) || - (donor_start > EXT_MAX_BLOCK) || - (*len > EXT_MAX_BLOCK) || - (orig_start + *len > EXT_MAX_BLOCK)) { + if ((orig_start >= EXT_MAX_BLOCKS) || + (donor_start >= EXT_MAX_BLOCKS) || + (*len > EXT_MAX_BLOCKS) || + (orig_start + *len >= EXT_MAX_BLOCKS)) { ext4_debug("ext4 move extent: Can't handle over [%u] blocks " - "[ino:orig %lu, donor %lu]\n", EXT_MAX_BLOCK, + "[ino:orig %lu, donor %lu]\n", EXT_MAX_BLOCKS, orig_inode->i_ino, donor_inode->i_ino); return -EINVAL; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index cc5c157aa11d..9ea71aa864b3 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -2243,6 +2243,12 @@ static void ext4_orphan_cleanup(struct super_block *sb, * in the vfs. ext4 inode has 48 bits of i_block in fsblock units, * so that won't be a limiting factor. * + * However there is other limiting factor. We do store extents in the form + * of starting block and length, hence the resulting length of the extent + * covering maximum file size must fit into on-disk format containers as + * well. Given that length is always by 1 unit bigger than max unit (because + * we count 0 as well) we have to lower the s_maxbytes by one fs block. + * * Note, this does *not* consider any metadata overhead for vfs i_blocks. */ static loff_t ext4_max_size(int blkbits, int has_huge_files) @@ -2264,10 +2270,13 @@ static loff_t ext4_max_size(int blkbits, int has_huge_files) upper_limit <<= blkbits; } - /* 32-bit extent-start container, ee_block */ - res = 1LL << 32; + /* + * 32-bit extent-start container, ee_block. We lower the maxbytes + * by one fs block, so ee_len can cover the extent of maximum file + * size + */ + res = (1LL << 32) - 1; res <<= blkbits; - res -= 1; /* Sanity check against vm- & vfs- imposed limits */ if (res > upper_limit) From c03f8aa9abdd517477c2021ea1251939b4da49e6 Mon Sep 17 00:00:00 2001 From: Lukas Czerner Date: Mon, 6 Jun 2011 00:06:52 -0400 Subject: [PATCH 014/224] ext4: use FIEMAP_EXTENT_LAST flag for last extent in fiemap Currently we are not marking the extent as the last one (FIEMAP_EXTENT_LAST) if there is a hole at the end of the file. This is because we just do not check for it right now and continue searching for next extent. But at the point we hit the hole at the end of the file, it is too late. This commit adds check for the allocated block in subsequent extent and if there is no more extents (block = EXT_MAX_BLOCKS) just flag the current one as the last one. This behaviour has been spotted unintentionally by 252 xfstest, when the test hangs out, because of wrong loop condition. However on other filesystems (like xfs) it will exit anyway, because we notice the last extent flag and exit. With this patch xfstest 252 does not hang anymore, ext4 fiemap implementation still reports bad extent type in some cases, however this seems to be different issue. Signed-off-by: Lukas Czerner Signed-off-by: "Theodore Ts'o" --- fs/ext4/ext4_extents.h | 2 +- fs/ext4/extents.c | 8 +++----- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index 476414644387..095c36f3b612 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -125,7 +125,7 @@ struct ext4_ext_path { * positive retcode - signal for ext4_ext_walk_space(), see below * callback must return valid extent (passed or newly created) */ -typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *, +typedef int (*ext_prepare_callback)(struct inode *, ext4_lblk_t, struct ext4_ext_cache *, struct ext4_extent *, void *); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 41575704600a..f815cc81e7a2 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -1958,7 +1958,7 @@ static int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block, err = -EIO; break; } - err = func(inode, path, &cbex, ex, cbdata); + err = func(inode, next, &cbex, ex, cbdata); ext4_ext_drop_refs(path); if (err < 0) @@ -3914,14 +3914,13 @@ int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, /* * Callback function called for each extent to gather FIEMAP information. */ -static int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path, +static int ext4_ext_fiemap_cb(struct inode *inode, ext4_lblk_t next, struct ext4_ext_cache *newex, struct ext4_extent *ex, void *data) { __u64 logical; __u64 physical; __u64 length; - loff_t size; __u32 flags = 0; int ret = 0; struct fiemap_extent_info *fieinfo = data; @@ -4103,8 +4102,7 @@ found_delayed_extent: if (ex && ext4_ext_is_uninitialized(ex)) flags |= FIEMAP_EXTENT_UNWRITTEN; - size = i_size_read(inode); - if (logical + length >= size) + if (next == EXT_MAX_BLOCKS) flags |= FIEMAP_EXTENT_LAST; ret = fiemap_fill_next_extent(fieinfo, logical, physical, From a9c667f8f0656631ee5438baaf21bf30d5f67375 Mon Sep 17 00:00:00 2001 From: Lukas Czerner Date: Mon, 6 Jun 2011 09:51:52 -0400 Subject: [PATCH 015/224] ext4: fixed tracepoints cleanup While creating fixed tracepoints for ext3, basically by porting them from ext4, I found a lot of useless retyping, wrong type usage, useless variable passing and other inconsistencies in the ext4 fixed tracepoint code. This patch cleans the fixed tracepoint code for ext4 and also simplify some of them. Signed-off-by: Lukas Czerner Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 2 +- fs/ext4/mballoc.c | 6 +- include/trace/events/ext4.h | 179 +++++++++++++++--------------------- 3 files changed, 80 insertions(+), 107 deletions(-) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index a5763e3505ba..e3126c051006 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2634,7 +2634,7 @@ static int ext4_writepage(struct page *page, struct buffer_head *page_bufs = NULL; struct inode *inode = page->mapping->host; - trace_ext4_writepage(inode, page); + trace_ext4_writepage(page); size = i_size_read(inode); if (page->index == size >> PAGE_CACHE_SHIFT) len = size & ~PAGE_CACHE_MASK; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index eb71dd59f2e4..6ed859d56850 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -3578,8 +3578,8 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh, free += next - bit; trace_ext4_mballoc_discard(sb, NULL, group, bit, next - bit); - trace_ext4_mb_release_inode_pa(sb, pa->pa_inode, pa, - grp_blk_start + bit, next - bit); + trace_ext4_mb_release_inode_pa(pa, grp_blk_start + bit, + next - bit); mb_free_blocks(pa->pa_inode, e4b, bit, next - bit); bit = next + 1; } @@ -3608,7 +3608,7 @@ ext4_mb_release_group_pa(struct ext4_buddy *e4b, ext4_group_t group; ext4_grpblk_t bit; - trace_ext4_mb_release_group_pa(sb, pa); + trace_ext4_mb_release_group_pa(pa); BUG_ON(pa->pa_deleted == 0); ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit); BUG_ON(group != e4b->bd_group && pa->pa_len != 0); diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index e09592d2f916..5ce2b2f5f524 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -26,7 +26,7 @@ TRACE_EVENT(ext4_free_inode, __field( umode_t, mode ) __field( uid_t, uid ) __field( gid_t, gid ) - __field( blkcnt_t, blocks ) + __field( __u64, blocks ) ), TP_fast_assign( @@ -40,9 +40,8 @@ TRACE_EVENT(ext4_free_inode, TP_printk("dev %d,%d ino %lu mode 0%o uid %u gid %u blocks %llu", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - __entry->mode, __entry->uid, __entry->gid, - (unsigned long long) __entry->blocks) + (unsigned long) __entry->ino, __entry->mode, + __entry->uid, __entry->gid, __entry->blocks) ); TRACE_EVENT(ext4_request_inode, @@ -178,7 +177,7 @@ TRACE_EVENT(ext4_begin_ordered_truncate, TP_printk("dev %d,%d ino %lu new_size %lld", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (long long) __entry->new_size) + __entry->new_size) ); DECLARE_EVENT_CLASS(ext4__write_begin, @@ -204,7 +203,7 @@ DECLARE_EVENT_CLASS(ext4__write_begin, __entry->flags = flags; ), - TP_printk("dev %d,%d ino %lu pos %llu len %u flags %u", + TP_printk("dev %d,%d ino %lu pos %lld len %u flags %u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->pos, __entry->len, __entry->flags) @@ -248,7 +247,7 @@ DECLARE_EVENT_CLASS(ext4__write_end, __entry->copied = copied; ), - TP_printk("dev %d,%d ino %lu pos %llu len %u copied %u", + TP_printk("dev %d,%d ino %lu pos %lld len %u copied %u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->pos, __entry->len, __entry->copied) @@ -286,29 +285,6 @@ DEFINE_EVENT(ext4__write_end, ext4_da_write_end, TP_ARGS(inode, pos, len, copied) ); -TRACE_EVENT(ext4_writepage, - TP_PROTO(struct inode *inode, struct page *page), - - TP_ARGS(inode, page), - - TP_STRUCT__entry( - __field( dev_t, dev ) - __field( ino_t, ino ) - __field( pgoff_t, index ) - - ), - - TP_fast_assign( - __entry->dev = inode->i_sb->s_dev; - __entry->ino = inode->i_ino; - __entry->index = page->index; - ), - - TP_printk("dev %d,%d ino %lu page_index %lu", - MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, __entry->index) -); - TRACE_EVENT(ext4_da_writepages, TP_PROTO(struct inode *inode, struct writeback_control *wbc), @@ -341,7 +317,7 @@ TRACE_EVENT(ext4_da_writepages, ), TP_printk("dev %d,%d ino %lu nr_to_write %ld pages_skipped %ld " - "range_start %llu range_end %llu sync_mode %d" + "range_start %lld range_end %lld sync_mode %d" "for_kupdate %d range_cyclic %d writeback_index %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->nr_to_write, @@ -449,7 +425,14 @@ DECLARE_EVENT_CLASS(ext4__page_op, TP_printk("dev %d,%d ino %lu page_index %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->index) + (unsigned long) __entry->index) +); + +DEFINE_EVENT(ext4__page_op, ext4_writepage, + + TP_PROTO(struct page *page), + + TP_ARGS(page) ); DEFINE_EVENT(ext4__page_op, ext4_readpage, @@ -489,7 +472,7 @@ TRACE_EVENT(ext4_invalidatepage, TP_printk("dev %d,%d ino %lu page_index %lu offset %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->index, __entry->offset) + (unsigned long) __entry->index, __entry->offset) ); TRACE_EVENT(ext4_discard_blocks, @@ -562,12 +545,10 @@ DEFINE_EVENT(ext4__mb_new_pa, ext4_mb_new_group_pa, ); TRACE_EVENT(ext4_mb_release_inode_pa, - TP_PROTO(struct super_block *sb, - struct inode *inode, - struct ext4_prealloc_space *pa, + TP_PROTO(struct ext4_prealloc_space *pa, unsigned long long block, unsigned int count), - TP_ARGS(sb, inode, pa, block, count), + TP_ARGS(pa, block, count), TP_STRUCT__entry( __field( dev_t, dev ) @@ -578,8 +559,8 @@ TRACE_EVENT(ext4_mb_release_inode_pa, ), TP_fast_assign( - __entry->dev = sb->s_dev; - __entry->ino = inode->i_ino; + __entry->dev = pa->pa_inode->i_sb->s_dev; + __entry->ino = pa->pa_inode->i_ino; __entry->block = block; __entry->count = count; ), @@ -591,10 +572,9 @@ TRACE_EVENT(ext4_mb_release_inode_pa, ); TRACE_EVENT(ext4_mb_release_group_pa, - TP_PROTO(struct super_block *sb, - struct ext4_prealloc_space *pa), + TP_PROTO(struct ext4_prealloc_space *pa), - TP_ARGS(sb, pa), + TP_ARGS(pa), TP_STRUCT__entry( __field( dev_t, dev ) @@ -604,7 +584,7 @@ TRACE_EVENT(ext4_mb_release_group_pa, ), TP_fast_assign( - __entry->dev = sb->s_dev; + __entry->dev = pa->pa_inode->i_sb->s_dev; __entry->pa_pstart = pa->pa_pstart; __entry->pa_len = pa->pa_len; ), @@ -666,10 +646,10 @@ TRACE_EVENT(ext4_request_blocks, __field( ino_t, ino ) __field( unsigned int, flags ) __field( unsigned int, len ) - __field( __u64, logical ) + __field( __u32, logical ) + __field( __u32, lleft ) + __field( __u32, lright ) __field( __u64, goal ) - __field( __u64, lleft ) - __field( __u64, lright ) __field( __u64, pleft ) __field( __u64, pright ) ), @@ -687,17 +667,13 @@ TRACE_EVENT(ext4_request_blocks, __entry->pright = ar->pright; ), - TP_printk("dev %d,%d ino %lu flags %u len %u lblk %llu goal %llu " - "lleft %llu lright %llu pleft %llu pright %llu ", + TP_printk("dev %d,%d ino %lu flags %u len %u lblk %u goal %llu " + "lleft %u lright %u pleft %llu pright %llu ", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - __entry->flags, __entry->len, - (unsigned long long) __entry->logical, - (unsigned long long) __entry->goal, - (unsigned long long) __entry->lleft, - (unsigned long long) __entry->lright, - (unsigned long long) __entry->pleft, - (unsigned long long) __entry->pright) + (unsigned long) __entry->ino, __entry->flags, + __entry->len, __entry->logical, __entry->goal, + __entry->lleft, __entry->lright, __entry->pleft, + __entry->pright) ); TRACE_EVENT(ext4_allocate_blocks, @@ -711,10 +687,10 @@ TRACE_EVENT(ext4_allocate_blocks, __field( __u64, block ) __field( unsigned int, flags ) __field( unsigned int, len ) - __field( __u64, logical ) + __field( __u32, logical ) + __field( __u32, lleft ) + __field( __u32, lright ) __field( __u64, goal ) - __field( __u64, lleft ) - __field( __u64, lright ) __field( __u64, pleft ) __field( __u64, pright ) ), @@ -733,17 +709,13 @@ TRACE_EVENT(ext4_allocate_blocks, __entry->pright = ar->pright; ), - TP_printk("dev %d,%d ino %lu flags %u len %u block %llu lblk %llu " - "goal %llu lleft %llu lright %llu pleft %llu pright %llu", + TP_printk("dev %d,%d ino %lu flags %u len %u block %llu lblk %u " + "goal %llu lleft %u lright %u pleft %llu pright %llu", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - __entry->flags, __entry->len, __entry->block, - (unsigned long long) __entry->logical, - (unsigned long long) __entry->goal, - (unsigned long long) __entry->lleft, - (unsigned long long) __entry->lright, - (unsigned long long) __entry->pleft, - (unsigned long long) __entry->pright) + (unsigned long) __entry->ino, __entry->flags, + __entry->len, __entry->block, __entry->logical, + __entry->goal, __entry->lleft, __entry->lright, + __entry->pleft, __entry->pright) ); TRACE_EVENT(ext4_free_blocks, @@ -755,10 +727,10 @@ TRACE_EVENT(ext4_free_blocks, TP_STRUCT__entry( __field( dev_t, dev ) __field( ino_t, ino ) - __field( umode_t, mode ) + __field( umode_t, mode ) __field( __u64, block ) __field( unsigned long, count ) - __field( int, flags ) + __field( int, flags ) ), TP_fast_assign( @@ -798,7 +770,7 @@ TRACE_EVENT(ext4_sync_file_enter, __entry->parent = dentry->d_parent->d_inode->i_ino; ), - TP_printk("dev %d,%d ino %ld parent %ld datasync %d ", + TP_printk("dev %d,%d ino %lu parent %lu datasync %d ", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, (unsigned long) __entry->parent, __entry->datasync) @@ -821,7 +793,7 @@ TRACE_EVENT(ext4_sync_file_exit, __entry->dev = inode->i_sb->s_dev; ), - TP_printk("dev %d,%d ino %ld ret %d", + TP_printk("dev %d,%d ino %lu ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->ret) @@ -1005,7 +977,7 @@ DECLARE_EVENT_CLASS(ext4__mballoc, __entry->result_len = len; ), - TP_printk("dev %d,%d inode %lu extent %u/%d/%u ", + TP_printk("dev %d,%d inode %lu extent %u/%d/%d ", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->result_group, __entry->result_start, @@ -1093,7 +1065,7 @@ TRACE_EVENT(ext4_da_update_reserve_space, "allocated_meta_blocks %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->mode, (unsigned long long) __entry->i_blocks, + __entry->mode, __entry->i_blocks, __entry->used_blocks, __entry->reserved_data_blocks, __entry->reserved_meta_blocks, __entry->allocated_meta_blocks) ); @@ -1127,7 +1099,7 @@ TRACE_EVENT(ext4_da_reserve_space, "reserved_data_blocks %d reserved_meta_blocks %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->mode, (unsigned long long) __entry->i_blocks, + __entry->mode, __entry->i_blocks, __entry->md_needed, __entry->reserved_data_blocks, __entry->reserved_meta_blocks) ); @@ -1164,7 +1136,7 @@ TRACE_EVENT(ext4_da_release_space, "allocated_meta_blocks %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - __entry->mode, (unsigned long long) __entry->i_blocks, + __entry->mode, __entry->i_blocks, __entry->freed_blocks, __entry->reserved_data_blocks, __entry->reserved_meta_blocks, __entry->allocated_meta_blocks) ); @@ -1239,14 +1211,15 @@ TRACE_EVENT(ext4_direct_IO_enter, __entry->rw = rw; ), - TP_printk("dev %d,%d ino %lu pos %llu len %lu rw %d", + TP_printk("dev %d,%d ino %lu pos %lld len %lu rw %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, __entry->len, __entry->rw) + __entry->pos, __entry->len, __entry->rw) ); TRACE_EVENT(ext4_direct_IO_exit, - TP_PROTO(struct inode *inode, loff_t offset, unsigned long len, int rw, int ret), + TP_PROTO(struct inode *inode, loff_t offset, unsigned long len, + int rw, int ret), TP_ARGS(inode, offset, len, rw, ret), @@ -1268,10 +1241,10 @@ TRACE_EVENT(ext4_direct_IO_exit, __entry->ret = ret; ), - TP_printk("dev %d,%d ino %lu pos %llu len %lu rw %d ret %d", + TP_printk("dev %d,%d ino %lu pos %lld len %lu rw %d ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, __entry->len, + __entry->pos, __entry->len, __entry->rw, __entry->ret) ); @@ -1296,15 +1269,15 @@ TRACE_EVENT(ext4_fallocate_enter, __entry->mode = mode; ), - TP_printk("dev %d,%d ino %ld pos %llu len %llu mode %d", + TP_printk("dev %d,%d ino %lu pos %lld len %lld mode %d", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, - (unsigned long long) __entry->len, __entry->mode) + (unsigned long) __entry->ino, __entry->pos, + __entry->len, __entry->mode) ); TRACE_EVENT(ext4_fallocate_exit, - TP_PROTO(struct inode *inode, loff_t offset, unsigned int max_blocks, int ret), + TP_PROTO(struct inode *inode, loff_t offset, + unsigned int max_blocks, int ret), TP_ARGS(inode, offset, max_blocks, ret), @@ -1312,7 +1285,7 @@ TRACE_EVENT(ext4_fallocate_exit, __field( ino_t, ino ) __field( dev_t, dev ) __field( loff_t, pos ) - __field( unsigned, blocks ) + __field( unsigned int, blocks ) __field( int, ret ) ), @@ -1324,10 +1297,10 @@ TRACE_EVENT(ext4_fallocate_exit, __entry->ret = ret; ), - TP_printk("dev %d,%d ino %ld pos %llu blocks %d ret %d", + TP_printk("dev %d,%d ino %lu pos %lld blocks %u ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned long long) __entry->pos, __entry->blocks, + __entry->pos, __entry->blocks, __entry->ret) ); @@ -1350,7 +1323,7 @@ TRACE_EVENT(ext4_unlink_enter, __entry->dev = dentry->d_inode->i_sb->s_dev; ), - TP_printk("dev %d,%d ino %ld size %lld parent %ld", + TP_printk("dev %d,%d ino %lu size %lld parent %lu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->size, (unsigned long) __entry->parent) @@ -1373,7 +1346,7 @@ TRACE_EVENT(ext4_unlink_exit, __entry->ret = ret; ), - TP_printk("dev %d,%d ino %ld ret %d", + TP_printk("dev %d,%d ino %lu ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, __entry->ret) @@ -1387,7 +1360,7 @@ DECLARE_EVENT_CLASS(ext4__truncate, TP_STRUCT__entry( __field( ino_t, ino ) __field( dev_t, dev ) - __field( blkcnt_t, blocks ) + __field( __u64, blocks ) ), TP_fast_assign( @@ -1396,9 +1369,9 @@ DECLARE_EVENT_CLASS(ext4__truncate, __entry->blocks = inode->i_blocks; ), - TP_printk("dev %d,%d ino %lu blocks %lu", + TP_printk("dev %d,%d ino %lu blocks %llu", MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long) __entry->ino, (unsigned long) __entry->blocks) + (unsigned long) __entry->ino, __entry->blocks) ); DEFINE_EVENT(ext4__truncate, ext4_truncate_enter, @@ -1417,7 +1390,7 @@ DEFINE_EVENT(ext4__truncate, ext4_truncate_exit, DECLARE_EVENT_CLASS(ext4__map_blocks_enter, TP_PROTO(struct inode *inode, ext4_lblk_t lblk, - unsigned len, unsigned flags), + unsigned int len, unsigned int flags), TP_ARGS(inode, lblk, len, flags), @@ -1425,8 +1398,8 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_enter, __field( ino_t, ino ) __field( dev_t, dev ) __field( ext4_lblk_t, lblk ) - __field( unsigned, len ) - __field( unsigned, flags ) + __field( unsigned int, len ) + __field( unsigned int, flags ) ), TP_fast_assign( @@ -1440,7 +1413,7 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_enter, TP_printk("dev %d,%d ino %lu lblk %u len %u flags %u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned) __entry->lblk, __entry->len, __entry->flags) + __entry->lblk, __entry->len, __entry->flags) ); DEFINE_EVENT(ext4__map_blocks_enter, ext4_ext_map_blocks_enter, @@ -1459,7 +1432,7 @@ DEFINE_EVENT(ext4__map_blocks_enter, ext4_ind_map_blocks_enter, DECLARE_EVENT_CLASS(ext4__map_blocks_exit, TP_PROTO(struct inode *inode, ext4_lblk_t lblk, - ext4_fsblk_t pblk, unsigned len, int ret), + ext4_fsblk_t pblk, unsigned int len, int ret), TP_ARGS(inode, lblk, pblk, len, ret), @@ -1468,7 +1441,7 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_exit, __field( dev_t, dev ) __field( ext4_lblk_t, lblk ) __field( ext4_fsblk_t, pblk ) - __field( unsigned, len ) + __field( unsigned int, len ) __field( int, ret ) ), @@ -1484,7 +1457,7 @@ DECLARE_EVENT_CLASS(ext4__map_blocks_exit, TP_printk("dev %d,%d ino %lu lblk %u pblk %llu len %u ret %d", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned) __entry->lblk, (unsigned long long) __entry->pblk, + __entry->lblk, __entry->pblk, __entry->len, __entry->ret) ); @@ -1524,7 +1497,7 @@ TRACE_EVENT(ext4_ext_load_extent, TP_printk("dev %d,%d ino %lu lblk %u pblk %llu", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long) __entry->ino, - (unsigned) __entry->lblk, (unsigned long long) __entry->pblk) + __entry->lblk, __entry->pblk) ); TRACE_EVENT(ext4_load_inode, From af99d6f0037d970084b03d9690f50e34d6f70dae Mon Sep 17 00:00:00 2001 From: Lennart Sorensen Date: Wed, 1 Jun 2011 14:38:41 -0400 Subject: [PATCH 016/224] serial: ioremap warning fix for jsm driver. I saw a warning about ioremap from the jsm driver on a system which looked like this: resource map sanity check conflict: 0xe0200800 0xe02017ff 0xe0200800 0xe0200fff 0000:01:08.0 Turns out the warning is valid. The jsm driver has been asking to ioremap 0x1000 forever, but in fact only 8 port chips have 0x1000 bytes of memory. 4 port chips have 0x800 and 2 port chips have 0x400 according to the data sheet. It makes more sense to map the size of the region rather than a hard coded value. If you happen to have the region legitimately mapped to a base address that is not 4K aligned, ioremap complains otherwise. Signed-off-by: Len Sorensen Acked-by: Alan Cox Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/jsm/jsm_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/jsm/jsm_driver.c b/drivers/tty/serial/jsm/jsm_driver.c index 18f548449c63..96da17868cf3 100644 --- a/drivers/tty/serial/jsm/jsm_driver.c +++ b/drivers/tty/serial/jsm/jsm_driver.c @@ -125,7 +125,7 @@ static int __devinit jsm_probe_one(struct pci_dev *pdev, const struct pci_device brd->bd_uart_offset = 0x200; brd->bd_dividend = 921600; - brd->re_map_membase = ioremap(brd->membase, 0x1000); + brd->re_map_membase = ioremap(brd->membase, pci_resource_len(pdev, 0)); if (!brd->re_map_membase) { dev_err(&pdev->dev, "card has no PCI Memory resources, " From 470f22975448a65a1084a6f0721fa5df15323f02 Mon Sep 17 00:00:00 2001 From: Boojin Kim Date: Fri, 27 May 2011 19:04:03 -0700 Subject: [PATCH 017/224] ARM: SAMSUNG: serial: Fix on handling of one clock source for UART This patch fixes the way of comparison for handling of two or more clock sources for UART. For example, if just only one clock source is defined even though there are two clock sources for UART, the serial driver does not set proper clock up. Of course, it is problem. So this patch changes the condition of comparison to avoid useless setup clock and adds a flag 'NO_NEED_CHECK_CLKSRC' which means selection of source clock is not required. In addition, since the Exynos4210 has only one clock source for UART this patch adds the flag into its common_init_uarts(). Signed-off-by: Boojin Kim Signed-off-by: Kukjin Kim Cc: stable Signed-off-by: Greg Kroah-Hartman --- arch/arm/mach-exynos4/init.c | 1 + arch/arm/plat-samsung/include/plat/regs-serial.h | 2 ++ drivers/tty/serial/s5pv210.c | 4 ++-- 3 files changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-exynos4/init.c b/arch/arm/mach-exynos4/init.c index cf91f50e43ab..a8a83e3881a4 100644 --- a/arch/arm/mach-exynos4/init.c +++ b/arch/arm/mach-exynos4/init.c @@ -35,6 +35,7 @@ void __init exynos4_common_init_uarts(struct s3c2410_uartcfg *cfg, int no) tcfg->clocks = exynos4_serial_clocks; tcfg->clocks_size = ARRAY_SIZE(exynos4_serial_clocks); } + tcfg->flags |= NO_NEED_CHECK_CLKSRC; } s3c24xx_init_uartdevs("s5pv210-uart", s5p_uart_resources, cfg, no); diff --git a/arch/arm/plat-samsung/include/plat/regs-serial.h b/arch/arm/plat-samsung/include/plat/regs-serial.h index c151c5f94a87..116edfe120b9 100644 --- a/arch/arm/plat-samsung/include/plat/regs-serial.h +++ b/arch/arm/plat-samsung/include/plat/regs-serial.h @@ -224,6 +224,8 @@ #define S5PV210_UFSTAT_RXMASK (255<<0) #define S5PV210_UFSTAT_RXSHIFT (0) +#define NO_NEED_CHECK_CLKSRC 1 + #ifndef __ASSEMBLY__ /* struct s3c24xx_uart_clksrc diff --git a/drivers/tty/serial/s5pv210.c b/drivers/tty/serial/s5pv210.c index fb2619f93d84..dd194dc80ee9 100644 --- a/drivers/tty/serial/s5pv210.c +++ b/drivers/tty/serial/s5pv210.c @@ -30,7 +30,7 @@ static int s5pv210_serial_setsource(struct uart_port *port, struct s3c2410_uartcfg *cfg = port->dev->platform_data; unsigned long ucon = rd_regl(port, S3C2410_UCON); - if ((cfg->clocks_size) == 1) + if (cfg->flags & NO_NEED_CHECK_CLKSRC) return 0; if (strcmp(clk->name, "pclk") == 0) @@ -55,7 +55,7 @@ static int s5pv210_serial_getsource(struct uart_port *port, clk->divisor = 1; - if ((cfg->clocks_size) == 1) + if (cfg->flags & NO_NEED_CHECK_CLKSRC) return 0; switch (ucon & S5PV210_UCON_CLKMASK) { From 1798ca13bfae8cc7c0ef82c034c3c4951ecaeb88 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Tue, 24 May 2011 12:35:48 +0100 Subject: [PATCH 018/224] 8250_pci: Fix missing const from merges Signed-off-by: Alan Cox Signed-off-by: Antony Pavlov Signed-off-by: Borislav Petkov Signed-off-by: Vasily Averin Signed-off-by: Guenter Roeck Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250_pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/8250_pci.c b/drivers/tty/serial/8250_pci.c index d7dc513451a6..f41b4259ecdd 100644 --- a/drivers/tty/serial/8250_pci.c +++ b/drivers/tty/serial/8250_pci.c @@ -973,7 +973,7 @@ ce4100_serial_setup(struct serial_private *priv, static int pci_omegapci_setup(struct serial_private *priv, - struct pciserial_board *board, + const struct pciserial_board *board, struct uart_port *port, int idx) { return setup_port(priv, port, 2, idx * 8, 0); From cb01ece3ea5dec16ac7bab30069c7736b59f7dea Mon Sep 17 00:00:00 2001 From: "leitao@linux.vnet.ibm.com" Date: Thu, 26 May 2011 11:18:39 -0300 Subject: [PATCH 019/224] 8250: Fix capabilities when changing the port type When changing the port type, the capabilities flags should be changed also, otherwise the capabilities will not correspond to the port type, which make set_sleep() crash on rmmod. This patch just assign the correct capabilites when the port changes. Signed-off-by: Breno Leitao CC: Michael Reed Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/tty/serial/8250.c b/drivers/tty/serial/8250.c index b40f7b90c81d..b4129f53fb1b 100644 --- a/drivers/tty/serial/8250.c +++ b/drivers/tty/serial/8250.c @@ -3318,6 +3318,7 @@ void serial8250_unregister_port(int line) uart->port.flags &= ~UPF_BOOT_AUTOCONF; uart->port.type = PORT_UNKNOWN; uart->port.dev = &serial8250_isa_devs->dev; + uart->capabilities = uart_config[uart->port.type].flags; uart_add_one_port(&serial8250_reg, &uart->port); } else { uart->port.dev = NULL; From 5daf538a0313509ecdeb5b7a61257f39881f9361 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 22 May 2011 17:38:02 -0700 Subject: [PATCH 020/224] firmware: fix GOOGLE_SMI kconfig dependency warning Is it meaningful/useful to enable EFI_VARS but not EFI? That's what GOOGLE_SMI does. Make it enable EFI also. Fixes this kconfig dependency warning: warning: (GOOGLE_SMI) selects EFI_VARS which has unmet direct dependencies (EFI) Signed-off-by: Randy Dunlap Acked-by: Mike Waychison Signed-off-by: Greg Kroah-Hartman --- drivers/firmware/google/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/firmware/google/Kconfig b/drivers/firmware/google/Kconfig index 87096b6ca5c9..2f21b0bfe653 100644 --- a/drivers/firmware/google/Kconfig +++ b/drivers/firmware/google/Kconfig @@ -13,6 +13,7 @@ menu "Google Firmware Drivers" config GOOGLE_SMI tristate "SMI interface for Google platforms" depends on ACPI && DMI + select EFI select EFI_VARS help Say Y here if you want to enable SMI callbacks for Google From 7316a9f2a94c14e66e9421a777dffc509a2fe0e3 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 23 May 2011 16:45:32 -0400 Subject: [PATCH 021/224] st_kim: Handle case of no device found for ID 0 Running ktest.pl, I hit this bug: [ 19.780654] BUG: unable to handle kernel NULL pointer dereference at 0000000c [ 19.780660] IP: [] dev_get_drvdata+0xc/0x46 [ 19.780669] *pdpt = 0000000031daf001 *pde = 0000000000000000 [ 19.780673] Oops: 0000 [#1] SMP [ 19.780680] Dumping ftrace buffer:^M [ 19.780685] (ftrace buffer empty) [ 19.780687] Modules linked in: ide_pci_generic firewire_ohci firewire_core evbug crc_itu_t e1000 ide_core i2c_i801 iTCO_wdt [ 19.780697] [ 19.780700] Pid: 346, comm: v4l_id Not tainted 2.6.39-test-02740-gcaebc16-dirty #4 /DG965MQ [ 19.780706] EIP: 0060:[] EFLAGS: 00010202 CPU: 0 [ 19.780709] EIP is at dev_get_drvdata+0xc/0x46 [ 19.780712] EAX: 00000008 EBX: f1e37da4 ECX: 00000000 EDX: 00000000 [ 19.780715] ESI: f1c3f200 EDI: c33ec95c EBP: f1e37d80 ESP: f1e37d80 [ 19.780718] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 [ 19.780721] Process v4l_id (pid: 346, ti=f1e36000 task=f2bc2a60 task.ti=f1e36000) [ 19.780723] Stack: [ 19.780725] f1e37d8c c117d395 c33ec93c f1e37db4 c117a0f9 00000002 00000000 c1725e54 [ 19.780732] 00000001 00000007 f2918c90 f1c3f200 c33ec95c f1e37dd4 c1789d3d 22222222 [ 19.780740] 22222222 22222222 f2918c90 f1c3f200 f29194f4 f1e37de8 c178d5c4 c1725e54 [ 19.780747] Call Trace: [ 19.780752] [] st_kim_ref+0x28/0x41 [ 19.780756] [] st_register+0x29/0x562 [ 19.780761] [] ? v4l2_open+0x111/0x1e3 [ 19.780766] [] fmc_prepare+0x97/0x424 [ 19.780770] [] fm_v4l2_fops_open+0x70/0x106 [ 19.780773] [] ? v4l2_open+0x111/0x1e3 [ 19.780777] [] v4l2_open+0x158/0x1e3 [ 19.780782] [] chrdev_open+0x22c/0x276 [ 19.780787] [] __dentry_open+0x35c/0x581 [ 19.780792] [] nameidata_to_filp+0x7c/0x96 [ 19.780795] [] ? cdev_put+0x57/0x57 [ 19.780800] [] do_last+0x743/0x9d4 [ 19.780804] [] ? path_init+0x1ee/0x596 [ 19.780808] [] path_openat+0x10c/0x597 [ 19.780813] [] ? trace_hardirqs_off+0x27/0x37 [ 19.780817] [] ? local_clock+0x78/0xc7 [ 19.780821] [] do_filp_open+0x39/0xc2 [ 19.780827] [] ? _raw_spin_unlock+0x4c/0x5d^M [ 19.780831] [] ? alloc_fd+0x19e/0x1b7 [ 19.780836] [] do_sys_open+0xb7/0x1bd [ 19.780840] [] ? sys_munmap+0x78/0x8d [ 19.780844] [] sys_open+0x36/0x58 [ 19.780849] [] sysenter_do_call+0x12/0x38 [ 19.780852] Code: d8 2f 20 c3 01 83 15 dc 2f 20 c3 00 f0 ff 00 83 05 e0 2f 20 c3 01 83 15 e4 2f 20 c3 00 5d c3 55 89 e5 3e 8d 74 26 00 85 c0 74 28 <8b> 40 04 83 05 e8 2f 20 c3 01 83 15 ec 2f 20 c3 00 85 c0 74 13 ^M [ 19.780889] EIP: [] dev_get_drvdata+0xc/0x46 SS:ESP 0068:f1e37d80 [ 19.780894] CR2: 000000000000000c [ 19.780898] ---[ end trace e7d1d0f6a2d1d390 ]--- The id of 0 passed to st_kim_ref() found no device, keeping pdev null, and causing pdev->dev cause a NULL pointer dereference. After having st_kim_ref() check for NULL, the st_unregister() function needed to be updated to handle the case that st_gdata was not set by the st_kim_ref(). Signed-off-by: Steven Rostedt Signed-off-by: Greg Kroah-Hartman --- drivers/misc/ti-st/st_core.c | 2 +- drivers/misc/ti-st/st_kim.c | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/misc/ti-st/st_core.c b/drivers/misc/ti-st/st_core.c index f91f82eabda7..54c91ffe4a91 100644 --- a/drivers/misc/ti-st/st_core.c +++ b/drivers/misc/ti-st/st_core.c @@ -605,7 +605,7 @@ long st_unregister(struct st_proto_s *proto) pr_debug("%s: %d ", __func__, proto->chnl_id); st_kim_ref(&st_gdata, 0); - if (proto->chnl_id >= ST_MAX_CHANNELS) { + if (!st_gdata || proto->chnl_id >= ST_MAX_CHANNELS) { pr_err(" chnl_id %d not supported", proto->chnl_id); return -EPROTONOSUPPORT; } diff --git a/drivers/misc/ti-st/st_kim.c b/drivers/misc/ti-st/st_kim.c index 5da93ee6f6be..3613c3bc4da3 100644 --- a/drivers/misc/ti-st/st_kim.c +++ b/drivers/misc/ti-st/st_kim.c @@ -604,6 +604,10 @@ void st_kim_ref(struct st_data_s **core_data, int id) struct kim_data_s *kim_gdata; /* get kim_gdata reference from platform device */ pdev = st_get_plat_device(id); + if (!pdev) { + *core_data = NULL; + return; + } kim_gdata = dev_get_drvdata(&pdev->dev); *core_data = kim_gdata->core_data; } From bb2b43fefab723f4a0760146e7bed59d41a50e53 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 23 May 2011 14:44:19 -0700 Subject: [PATCH 022/224] drivers/base/platform.c: don't mark platform_device_register_resndata() as __init_or_module MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts 737a3bb9416ce2a7c7a4 ("Driver core: move platform device creation helpers to .init.text (if MODULE=n)"). That patch assumed that platform_device_register_resndata() is only ever called from __init code but that isn't true in the case ioctl->drm_ioctl->radeon_cp_init(). Addresses https://bugzilla.kernel.org/show_bug.cgi?id=35192 Cc: Uwe Kleine-König Reported-by: Anthony Basile Cc: Greg KH Cc: David Airlie Cc: Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- drivers/base/platform.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 1c291af637b3..6040717b62bb 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -367,7 +367,7 @@ EXPORT_SYMBOL_GPL(platform_device_unregister); * * Returns &struct platform_device pointer on success, or ERR_PTR() on error. */ -struct platform_device *__init_or_module platform_device_register_resndata( +struct platform_device *platform_device_register_resndata( struct device *parent, const char *name, int id, const struct resource *res, unsigned int num, From 9d031d94da453077bbc6108b7822fc751ac85299 Mon Sep 17 00:00:00 2001 From: Shahar Lev Date: Mon, 23 May 2011 11:36:11 +0300 Subject: [PATCH 023/224] drivers:misc: ti-st: fix skipping of change remote baud Before the incrementing of ptr in skip_change_remote_baud, it points to cur_action, but the increment is done by the size of nxt_action instead. This could cause ptr to not point to a bts_action structure, which is harmful for the increment of ptr done in download_firmware. Therefore, the skipping is first done for cur_action. Signed-off-by: Shahar Lev Signed-off-by: Greg Kroah-Hartman --- drivers/misc/ti-st/st_kim.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/misc/ti-st/st_kim.c b/drivers/misc/ti-st/st_kim.c index 3613c3bc4da3..38fd2f04c07e 100644 --- a/drivers/misc/ti-st/st_kim.c +++ b/drivers/misc/ti-st/st_kim.c @@ -245,9 +245,9 @@ void skip_change_remote_baud(unsigned char **ptr, long *len) pr_err("invalid action after change remote baud command"); } else { *ptr = *ptr + sizeof(struct bts_action) + - ((struct bts_action *)nxt_action)->size; + ((struct bts_action *)cur_action)->size; *len = *len - (sizeof(struct bts_action) + - ((struct bts_action *)nxt_action)->size); + ((struct bts_action *)cur_action)->size); /* warn user on not commenting these in firmware */ pr_warn("skipping the wait event of change remote baud"); } From 29021bccea0dc42d7d101004058438a9a4e693b1 Mon Sep 17 00:00:00 2001 From: J Freyensee Date: Wed, 25 May 2011 14:38:18 -0700 Subject: [PATCH 024/224] pti: double-free security PTI fix This patch fixes a double-free error that will not always be seen unless /dev/pti char interface is stressed. Signed-off-by: J Freyensee Signed-off-by: Greg Kroah-Hartman --- drivers/misc/pti.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/misc/pti.c b/drivers/misc/pti.c index bb6f9255c17c..be4857358d91 100644 --- a/drivers/misc/pti.c +++ b/drivers/misc/pti.c @@ -317,7 +317,8 @@ EXPORT_SYMBOL_GPL(pti_request_masterchannel); * a master, channel ID address * used to write to PTI HW. * - * @mc: master, channel apeture ID address to be released. + * @mc: master, channel apeture ID address to be released. This + * will de-allocate the structure via kfree(). */ void pti_release_masterchannel(struct pti_masterchannel *mc) { @@ -581,7 +582,7 @@ static int pti_char_open(struct inode *inode, struct file *filp) static int pti_char_release(struct inode *inode, struct file *filp) { pti_release_masterchannel(filp->private_data); - kfree(filp->private_data); + filp->private_data = NULL; return 0; } From 1dae42bff57f7a61577ee881265985a660d35c07 Mon Sep 17 00:00:00 2001 From: J Freyensee Date: Wed, 25 May 2011 14:45:40 -0700 Subject: [PATCH 025/224] pti: ENXIO error case memory leak PTI fix. This patch fixes a memory leak that can occur in the error case ENXIO is returned in the pti_tty_install() routine. Signed-off-by: J Freyensee Signed-off-by: Greg Kroah-Hartman --- drivers/misc/pti.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/misc/pti.c b/drivers/misc/pti.c index be4857358d91..e74e7d266569 100644 --- a/drivers/misc/pti.c +++ b/drivers/misc/pti.c @@ -476,8 +476,10 @@ static int pti_tty_install(struct tty_driver *driver, struct tty_struct *tty) else pti_tty_data->mc = pti_request_masterchannel(2); - if (pti_tty_data->mc == NULL) + if (pti_tty_data->mc == NULL) { + kfree(pti_tty_data); return -ENXIO; + } tty->driver_data = pti_tty_data; } From 1312ba40c0d00f0a5cfcfe3afc222fcef2a90b56 Mon Sep 17 00:00:00 2001 From: J Freyensee Date: Wed, 25 May 2011 14:56:43 -0700 Subject: [PATCH 026/224] pti: PTI semantics fix in pti_tty_cleanup. This patch fixes a semantics issue in the pti_tty_cleanup() routine. Signed-off-by: J Freyensee Signed-off-by: Greg Kroah-Hartman --- drivers/misc/pti.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/misc/pti.c b/drivers/misc/pti.c index e74e7d266569..374dfcfccd07 100644 --- a/drivers/misc/pti.c +++ b/drivers/misc/pti.c @@ -498,7 +498,7 @@ static void pti_tty_cleanup(struct tty_struct *tty) if (pti_tty_data == NULL) return; pti_release_masterchannel(pti_tty_data->mc); - kfree(tty->driver_data); + kfree(pti_tty_data); tty->driver_data = NULL; } From 92f6fa09bd453ffe3351fa1f1377a1b7cfa911e6 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Sun, 5 Jun 2011 14:16:16 +0200 Subject: [PATCH 027/224] TTY: ldisc, do not close until there are readers We restored tty_ldisc_wait_idle in 100eeae2c5c (TTY: restore tty_ldisc_wait_idle). We used it in the ldisc changing path to fix the case where there are tasks in n_tty_read waiting for data and somebody tries to change ldisc. Similar to the case above, there may be also tasks waiting in n_tty_read while hangup is performed. As 65b770468e98 (tty-ldisc: turn ldisc user count into a proper refcount) removed the wait-until-idle from all paths, hangup path won't wait for them to disappear either now. So add it back even to the hangup path. There is a difference, we need uninterruptible sleep as there is obviously HUP signal pending. So tty_ldisc_wait_idle now sleeps without possibility to be interrupted. This is what original tty_ldisc_wait_idle did. After the wait idle reintroduction (100eeae2c5c), we have had interruptible sleeps for the ldisc changing path. But as there is a 5s timeout anyway, we don't allow it to be interrupted from now on. It's not worth the added complexity of deciding what kind of sleep we want. Before 65b770468e98 tty_ldisc_release was called also from tty_ldisc_release. It is called from tty_release, so I don't think we need to restore that one. This is nicely reproducible after constifying the timing when drivers/tty/n_tty.c is patched as follows ("TTY: ntty, add one more sanity check" patch is needed to actually see it explode): %% -1548,6 +1549,7 @@ static int n_tty_open(struct tty_struct *tty) /* These are ugly. Currently a malloc failure here can panic */ if (!tty->read_buf) { + msleep(100); tty->read_buf = kzalloc(N_TTY_BUF_SIZE, GFP_KERNEL); if (!tty->read_buf) return -ENOMEM; %% -1785,6 +1788,7 @@ do_it_again: break; } timeout = schedule_timeout(timeout); + msleep(20); continue; } __set_current_state(TASK_RUNNING); ===== With a process: ===== while (1) { int fd = open(argv[1], O_RDWR); read(fd, buf, sizeof(buf)); close(fd); } ===== and its child: ===== setsid(); while (1) { int fd = open(tty, O_RDWR|O_NOCTTY); ioctl(fd, TIOCSCTTY, 1); vhangup(); close(fd); usleep(100 * (10 + random() % 1000)); } ===== EOF ===== References: https://bugzilla.novell.com/show_bug.cgi?id=693374 References: https://bugzilla.novell.com/show_bug.cgi?id=694509 Signed-off-by: Jiri Slaby Cc: Alan Cox Cc: Linus Torvalds Cc: stable [32, 33, 34, 39] Signed-off-by: Greg Kroah-Hartman --- drivers/tty/tty_ldisc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/tty/tty_ldisc.c b/drivers/tty/tty_ldisc.c index 5d01d32e2cf0..ef925d581713 100644 --- a/drivers/tty/tty_ldisc.c +++ b/drivers/tty/tty_ldisc.c @@ -555,7 +555,7 @@ static void tty_ldisc_flush_works(struct tty_struct *tty) static int tty_ldisc_wait_idle(struct tty_struct *tty) { int ret; - ret = wait_event_interruptible_timeout(tty_ldisc_idle, + ret = wait_event_timeout(tty_ldisc_idle, atomic_read(&tty->ldisc->users) == 1, 5 * HZ); if (ret < 0) return ret; @@ -763,6 +763,8 @@ static int tty_ldisc_reinit(struct tty_struct *tty, int ldisc) if (IS_ERR(ld)) return -1; + WARN_ON_ONCE(tty_ldisc_wait_idle(tty)); + tty_ldisc_close(tty, tty->ldisc); tty_ldisc_put(tty->ldisc); tty->ldisc = NULL; From 2872628680bad71a6734e7d379168f990a91cc09 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Sun, 5 Jun 2011 14:16:17 +0200 Subject: [PATCH 028/224] TTY: ntty, add one more sanity check With the previous patch, we fixed another bug where read_buf was freed while we still was in n_tty_read. We currently check whether read_buf is NULL at the start of the function. Add one more check after we wake up from waiting for input. Signed-off-by: Jiri Slaby Cc: Alan Cox Signed-off-by: Greg Kroah-Hartman --- drivers/tty/n_tty.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c index 0ad32888091c..c3954fbf6ac4 100644 --- a/drivers/tty/n_tty.c +++ b/drivers/tty/n_tty.c @@ -1815,6 +1815,7 @@ do_it_again: /* FIXME: does n_tty_set_room need locking ? */ n_tty_set_room(tty); timeout = schedule_timeout(timeout); + BUG_ON(!tty->read_buf); continue; } __set_current_state(TASK_RUNNING); From ea2c00095c022846dd8dfd211de05154d3e4e1b8 Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Tue, 17 May 2011 15:25:37 -0700 Subject: [PATCH 029/224] Connector: Set the CN_NETLINK_USERS correctly The CN_NETLINK_USERS must be set to the highest valid index +1. Thanks to Evgeniy for pointing this out. Signed-off-by: K. Y. Srinivasan Acked-by: Evgeniy Polyakov Cc: stable [.39] Signed-off-by: Greg Kroah-Hartman --- include/linux/connector.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/connector.h b/include/linux/connector.h index 7c60d0942adb..f696bccd48cb 100644 --- a/include/linux/connector.h +++ b/include/linux/connector.h @@ -44,7 +44,7 @@ #define CN_VAL_DRBD 0x1 #define CN_KVP_IDX 0x9 /* HyperV KVP */ -#define CN_NETLINK_USERS 9 +#define CN_NETLINK_USERS 10 /* Highest index + 1 */ /* * Maximum connector's message size. From 663dd6dcaf7e95526e469e91f41972a9c0cca30c Mon Sep 17 00:00:00 2001 From: "K. Y. Srinivasan" Date: Tue, 17 May 2011 15:25:38 -0700 Subject: [PATCH 030/224] Connector: Correctly set the error code in case of success when dispatching receive callbacks The recent changes to the connector code introduced this bug where even when a callback was invoked, we would return an error resulting in double freeing of the skb. This patch fixes this bug. Signed-off-by: K. Y. Srinivasan Acked-by: Evgeniy Polyakov Cc: stable [.39] Signed-off-by: Greg Kroah-Hartman --- drivers/connector/connector.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c index 219d88a0eeae..dde6a0fad408 100644 --- a/drivers/connector/connector.c +++ b/drivers/connector/connector.c @@ -139,6 +139,7 @@ static int cn_call_callback(struct sk_buff *skb) spin_unlock_bh(&dev->cbdev->queue_lock); if (cbq != NULL) { + err = 0; cbq->callback(msg, nsp); kfree_skb(skb); cn_queue_release_callback(cbq); From 72887c8644384c0cc43b9298ae0659de383f2e9c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?M=C3=A1rton=20N=C3=A9meth?= Date: Mon, 30 May 2011 20:45:42 +0200 Subject: [PATCH 031/224] usb: musb: host: compare status for negative error values MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Variable d is a struct usb_iso_packet_descriptor. The status filed is usually negative when an error happens. Signed-off-by: Márton Németh Signed-off-by: Felipe Balbi --- drivers/usb/musb/musb_host.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/musb/musb_host.c b/drivers/usb/musb/musb_host.c index 7295e316bdfc..8b2473fa0f47 100644 --- a/drivers/usb/musb/musb_host.c +++ b/drivers/usb/musb/musb_host.c @@ -1575,7 +1575,7 @@ void musb_host_rx(struct musb *musb, u8 epnum) /* even if there was an error, we did the dma * for iso_frame_desc->length */ - if (d->status != EILSEQ && d->status != -EOVERFLOW) + if (d->status != -EILSEQ && d->status != -EOVERFLOW) d->status = 0; if (++qh->iso_idx >= urb->number_of_packets) From 4858f06e7d92ed2ebdb29ccbc079c127e675a89c Mon Sep 17 00:00:00 2001 From: Yauheni Kaliuta Date: Wed, 8 Jun 2011 17:12:02 +0300 Subject: [PATCH 032/224] usb: musb: gadget: clear TXPKTRDY flag when set FLUSHFIFO Fixes mis-use of MUSB's hardware feature where it won't flush FIFOs when TXPKTRDY flag was set before and we are flushing setting both FLUSHFIFO and TXPKTRDY. In other words, we need to ensure that when we try to flush FIFOs, we don't accidentaly set TXPKTRDY bit too due to a read-back of the register. The MUSB Programming Guide says "May be set simultaneously with TxPktRdy to abort the packet that is currently being loaded into the FIFO". This is a situation where TXPKTRDY hasn't been set yet, but some data already loaded into the fifo. It looks, that if TXPKTRDY has been set before, and there is no loading in progress, but we set FLUSHFIFO with the TXPKTRDY, controller tries to use the same logic to abort loading and as the result just does nothing (because there is no packet been loaded currently) Signed-off-by: Yauheni Kaliuta [ balbi@ti.com : fixed one whitespace git complained about improved the commit log slightly ] Signed-off-by: Felipe Balbi --- drivers/usb/musb/musb_gadget.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/usb/musb/musb_gadget.c b/drivers/usb/musb/musb_gadget.c index 0a50a35e1853..6aeb363e63e7 100644 --- a/drivers/usb/musb/musb_gadget.c +++ b/drivers/usb/musb/musb_gadget.c @@ -1524,6 +1524,12 @@ static void musb_gadget_fifo_flush(struct usb_ep *ep) csr = musb_readw(epio, MUSB_TXCSR); if (csr & MUSB_TXCSR_FIFONOTEMPTY) { csr |= MUSB_TXCSR_FLUSHFIFO | MUSB_TXCSR_P_WZC_BITS; + /* + * Setting both TXPKTRDY and FLUSHFIFO makes controller + * to interrupt current FIFO loading, but not flushing + * the already loaded ones. + */ + csr &= ~MUSB_TXCSR_TXPKTRDY; musb_writew(epio, MUSB_TXCSR, csr); /* REVISIT may be inappropriate w/o FIFONOTEMPTY ... */ musb_writew(epio, MUSB_TXCSR, csr); From a9dce2a3b4f0686dd66cb44d4826a59508bce969 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 9 Jun 2011 20:43:54 +0200 Subject: [PATCH 033/224] block: don't use non-syncing event blocking in disk_check_events() This patch is part of fix for triggering of WARN_ON_ONCE() in disk_clear_events() reported in bug#34662. https://bugzilla.kernel.org/show_bug.cgi?id=34662 disk_clear_events() blocks events, schedules and flushes the event work. It expects the work to have started execution on schedule and finished on return from flush. WARN_ON_ONCE() triggers if the event work hasn't executed as expected. This problem happens because __disk_block_events() fails to guarantee that the event work item is not in flight on return from the function in race-free manner. The problem is two-fold and this patch addresses one of them. When __disk_block_events() is called with @sync == %false, it bumps event block count, calls cancel_delayed_work() and return. This makes it impossible to guarantee that event polling is not in flight on return from syncing __disk_block_events() - if the first blocker was non-syncing, polling could still be in progress and later syncing ones would assume that the first blocker already canceled it. Making __disk_block_events() cancel_sync regardless of block count isn't feasible either as it may race with forced event checking in disk_clear_events(). As disk_check_events() is the only user of non-syncing __disk_block_events(), updating it to directly cancel and schedule event work is the easiest way to solve the issue. Note that there's another bug in __disk_block_events() and this patch doesn't fix the issue completely. Later patch will fix the other bug. Signed-off-by: Tejun Heo Tested-by: Sitsofe Wheeler Reported-by: Sitsofe Wheeler Reported-by: Borislav Petkov Reported-by: Meelis Roos Reported-by: Linus Torvalds Cc: Andrew Morton Cc: Jens Axboe Cc: Kay Sievers Signed-off-by: Jens Axboe --- block/genhd.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/block/genhd.c b/block/genhd.c index 95822ae25cfe..3f0933077642 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1508,10 +1508,18 @@ void disk_unblock_events(struct gendisk *disk) */ void disk_check_events(struct gendisk *disk) { - if (disk->ev) { - __disk_block_events(disk, false); - __disk_unblock_events(disk, true); + struct disk_events *ev = disk->ev; + unsigned long flags; + + if (!ev) + return; + + spin_lock_irqsave(&ev->lock, flags); + if (!ev->block) { + cancel_delayed_work(&ev->dwork); + queue_delayed_work(system_nrt_wq, &ev->dwork, 0); } + spin_unlock_irqrestore(&ev->lock, flags); } EXPORT_SYMBOL_GPL(disk_check_events); From c3af54afbac3675337cedf326b7b127ffa7f7327 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 9 Jun 2011 20:43:55 +0200 Subject: [PATCH 034/224] block: remove non-syncing __disk_block_events() and fold it into disk_block_events() After the previous update to disk_check_events(), nobody is using non-syncing __disk_block_events(). Remove @sync and, as this makes __disk_block_events() virtually identical to disk_block_events(), remove the underscore prefixed version. Signed-off-by: Tejun Heo Cc: Jens Axboe Signed-off-by: Jens Axboe --- block/genhd.c | 55 ++++++++++++++++++++++----------------------------- 1 file changed, 24 insertions(+), 31 deletions(-) diff --git a/block/genhd.c b/block/genhd.c index 3f0933077642..ab0731d8976d 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1414,22 +1414,36 @@ static unsigned long disk_events_poll_jiffies(struct gendisk *disk) return msecs_to_jiffies(intv_msecs); } -static void __disk_block_events(struct gendisk *disk, bool sync) +/** + * disk_block_events - block and flush disk event checking + * @disk: disk to block events for + * + * On return from this function, it is guaranteed that event checking + * isn't in progress and won't happen until unblocked by + * disk_unblock_events(). Events blocking is counted and the actual + * unblocking happens after the matching number of unblocks are done. + * + * Note that this intentionally does not block event checking from + * disk_clear_events(). + * + * CONTEXT: + * Might sleep. + */ +void disk_block_events(struct gendisk *disk) { struct disk_events *ev = disk->ev; unsigned long flags; bool cancel; + if (!ev) + return; + spin_lock_irqsave(&ev->lock, flags); cancel = !ev->block++; spin_unlock_irqrestore(&ev->lock, flags); - if (cancel) { - if (sync) - cancel_delayed_work_sync(&disk->ev->dwork); - else - cancel_delayed_work(&disk->ev->dwork); - } + if (cancel) + cancel_delayed_work_sync(&disk->ev->dwork); } static void __disk_unblock_events(struct gendisk *disk, bool check_now) @@ -1460,27 +1474,6 @@ out_unlock: spin_unlock_irqrestore(&ev->lock, flags); } -/** - * disk_block_events - block and flush disk event checking - * @disk: disk to block events for - * - * On return from this function, it is guaranteed that event checking - * isn't in progress and won't happen until unblocked by - * disk_unblock_events(). Events blocking is counted and the actual - * unblocking happens after the matching number of unblocks are done. - * - * Note that this intentionally does not block event checking from - * disk_clear_events(). - * - * CONTEXT: - * Might sleep. - */ -void disk_block_events(struct gendisk *disk) -{ - if (disk->ev) - __disk_block_events(disk, true); -} - /** * disk_unblock_events - unblock disk event checking * @disk: disk to unblock events for @@ -1554,7 +1547,7 @@ unsigned int disk_clear_events(struct gendisk *disk, unsigned int mask) spin_unlock_irq(&ev->lock); /* uncondtionally schedule event check and wait for it to finish */ - __disk_block_events(disk, true); + disk_block_events(disk); queue_delayed_work(system_nrt_wq, &ev->dwork, 0); flush_delayed_work(&ev->dwork); __disk_unblock_events(disk, false); @@ -1672,7 +1665,7 @@ static ssize_t disk_events_poll_msecs_store(struct device *dev, if (intv < 0 && intv != -1) return -EINVAL; - __disk_block_events(disk, true); + disk_block_events(disk); disk->ev->poll_msecs = intv; __disk_unblock_events(disk, true); @@ -1778,7 +1771,7 @@ static void disk_del_events(struct gendisk *disk) if (!disk->ev) return; - __disk_block_events(disk, true); + disk_block_events(disk); mutex_lock(&disk_events_mutex); list_del_init(&disk->ev->node); From fdd514e16bb2531c0c61ae8a1f87740ce217f630 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 9 Jun 2011 20:43:59 +0200 Subject: [PATCH 035/224] block: make disk_block_events() properly wait for work cancellation disk_block_events() should guarantee that the event work is not in flight on return and once blocked it shouldn't issue further cancellations. Because there was no synchronization between the first blocker doing cancel_delayed_work_sync() and the following blockers, the following blockers could finish before cancellation was complete, which broke both guarantees - event work could be in flight and cancellation could happen after return. This bug triggered WARN_ON_ONCE() in disk_clear_events() reported in bug#34662. https://bugzilla.kernel.org/show_bug.cgi?id=34662 Fix it by adding an outer mutex which protects both block count manipulation and work cancellation. -v2: Use outer mutex instead of bit waitqueue per Linus. Signed-off-by: Tejun Heo Tested-by: Sitsofe Wheeler Reported-by: Sitsofe Wheeler Reported-by: Borislav Petkov Reported-by: Meelis Roos Reported-by: Linus Torvalds Cc: Andrew Morton Cc: Jens Axboe Cc: Kay Sievers Signed-off-by: Jens Axboe --- block/genhd.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/block/genhd.c b/block/genhd.c index ab0731d8976d..3608289c8ecd 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1371,6 +1371,7 @@ struct disk_events { struct gendisk *disk; /* the associated disk */ spinlock_t lock; + struct mutex block_mutex; /* protects blocking */ int block; /* event blocking depth */ unsigned int pending; /* events already sent out */ unsigned int clearing; /* events being cleared */ @@ -1438,12 +1439,20 @@ void disk_block_events(struct gendisk *disk) if (!ev) return; + /* + * Outer mutex ensures that the first blocker completes canceling + * the event work before further blockers are allowed to finish. + */ + mutex_lock(&ev->block_mutex); + spin_lock_irqsave(&ev->lock, flags); cancel = !ev->block++; spin_unlock_irqrestore(&ev->lock, flags); if (cancel) cancel_delayed_work_sync(&disk->ev->dwork); + + mutex_unlock(&ev->block_mutex); } static void __disk_unblock_events(struct gendisk *disk, bool check_now) @@ -1751,6 +1760,7 @@ static void disk_add_events(struct gendisk *disk) INIT_LIST_HEAD(&ev->node); ev->disk = disk; spin_lock_init(&ev->lock); + mutex_init(&ev->block_mutex); ev->block = 1; ev->poll_msecs = -1; INIT_DELAYED_WORK(&ev->dwork, disk_events_workfn); From 1fb74cda1b5e9c6207225fda5ef7504e815ce0e0 Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Sun, 12 Jun 2011 22:44:10 -0400 Subject: [PATCH 036/224] jbd2: Remove obsolete parameters in the comments for some jbd2 functions credits isn't a parameter for jbd2_journal_get_write_access and jbd2_journal_get_undo_access. So remove the corresponding comments. Acked-by: Jan Kara Cc: Randy Dunlap Signed-off-by: Tao Ma Signed-off-by: "Theodore Ts'o" --- fs/jbd2/transaction.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 3eec82d32fd4..547b101049e5 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -814,7 +814,6 @@ out: * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. * @handle: transaction to add buffer modifications to * @bh: bh to be used for metadata writes - * @credits: variable that will receive credits for the buffer * * Returns an error code or 0 on success. * @@ -932,7 +931,6 @@ out: * non-rewindable consequences * @handle: transaction * @bh: buffer to undo - * @credits: store the number of taken credits here (if not NULL) * * Sometimes there is a need to distinguish between metadata which has * been committed to disk and that which has not. The ext3fs code uses From 08e8138adebdd511e0955e8d6c051904bb4082af Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 13 Jun 2011 10:42:49 +0200 Subject: [PATCH 037/224] block: Add __attribute__((format(printf...) and fix fallout Use the compiler to verify format strings and arguments. Fix fallout. Signed-off-by: Joe Perches Signed-off-by: Jens Axboe --- block/blk-throttle.c | 4 ++-- block/cfq-iosched.c | 11 ++++++----- include/linux/blktrace_api.h | 3 ++- 3 files changed, 10 insertions(+), 8 deletions(-) diff --git a/block/blk-throttle.c b/block/blk-throttle.c index a62be8d0dc1b..3689f833afdc 100644 --- a/block/blk-throttle.c +++ b/block/blk-throttle.c @@ -927,7 +927,7 @@ static int throtl_dispatch(struct request_queue *q) bio_list_init(&bio_list_on_stack); - throtl_log(td, "dispatch nr_queued=%lu read=%u write=%u", + throtl_log(td, "dispatch nr_queued=%d read=%u write=%u", total_nr_queued(td), td->nr_queued[READ], td->nr_queued[WRITE]); @@ -1204,7 +1204,7 @@ int blk_throtl_bio(struct request_queue *q, struct bio **biop) } queue_bio: - throtl_log_tg(td, tg, "[%c] bio. bdisp=%u sz=%u bps=%llu" + throtl_log_tg(td, tg, "[%c] bio. bdisp=%llu sz=%u bps=%llu" " iodisp=%u iops=%u queued=%d/%d", rw == READ ? 'R' : 'W', tg->bytes_disp[rw], bio->bi_size, tg->bps[rw], diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 545b8d4c829d..f3799432676d 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -988,9 +988,10 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg, cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime, st->min_vdisktime); - cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u disp=%u charge=%u iops=%u" - " sect=%u", used_sl, cfqq->slice_dispatch, charge, - iops_mode(cfqd), cfqq->nr_sectors); + cfq_log_cfqq(cfqq->cfqd, cfqq, + "sl_used=%u disp=%u charge=%u iops=%u sect=%lu", + used_sl, cfqq->slice_dispatch, charge, + iops_mode(cfqd), cfqq->nr_sectors); cfq_blkiocg_update_timeslice_used(&cfqg->blkg, used_sl, unaccounted_sl); cfq_blkiocg_set_start_empty_time(&cfqg->blkg); @@ -2023,8 +2024,8 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) */ if (sample_valid(cic->ttime_samples) && (cfqq->slice_end - jiffies < cic->ttime_mean)) { - cfq_log_cfqq(cfqd, cfqq, "Not idling. think_time:%d", - cic->ttime_mean); + cfq_log_cfqq(cfqd, cfqq, "Not idling. think_time:%lu", + cic->ttime_mean); return; } diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index b22fb0d3db0f..8c7c2de7631a 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -169,7 +169,8 @@ extern void blk_trace_shutdown(struct request_queue *); extern int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, struct block_device *bdev, struct blk_user_trace_setup *buts); -extern void __trace_note_message(struct blk_trace *, const char *fmt, ...); +extern __attribute__((format(printf, 2, 3))) +void __trace_note_message(struct blk_trace *, const char *fmt, ...); /** * blk_add_trace_msg - Add a (simple) message to the blktrace stream From d4c208b86b8be4254eba0e74071496e599f94639 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 13 Jun 2011 12:45:48 +0200 Subject: [PATCH 038/224] block: use the passed in @bdev when claiming if partno is zero 6b4517a791 (block: implement bd_claiming and claiming block) introduced claiming block to support O_EXCL blkdev opens properly. bd_start_claiming() looks up the part 0 bdev and starts claiming block. The function assumed that there is only one part 0 bdev and always used bdget_disk(disk, 0) to look it up; unfortunately, this isn't true for some drivers (floppy) which use multiple block devices to denote different operating parameters for the same physical device. There can be multiple part 0 bdev's for the same device number. This incorrect assumption caused the wrong bdev to be used during claiming leading to unbalanced bd_holders as reported in the following bug. https://bugzilla.kernel.org/show_bug.cgi?id=28522 This patch updates bd_start_claiming() such that it uses the bdev specified as argument if its partno is zero. Note that this means that different bdev's can be used for the same device and O_EXCL check can be effectively bypassed. It has always been broken that way and floppy is fortunately on its way out. Leave that breakage alone. Signed-off-by: Tejun Heo Reported-by: Alex Villacis Lasso Tested-by: Alex Villacis Lasso Cc: stable@kernel.org # >= v2.6.36 Signed-off-by: Jens Axboe --- fs/block_dev.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index 1a2421f908f0..610e8e0b04b8 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -762,7 +762,19 @@ static struct block_device *bd_start_claiming(struct block_device *bdev, if (!disk) return ERR_PTR(-ENXIO); - whole = bdget_disk(disk, 0); + /* + * Normally, @bdev should equal what's returned from bdget_disk() + * if partno is 0; however, some drivers (floppy) use multiple + * bdev's for the same physical device and @bdev may be one of the + * aliases. Keep @bdev if partno is 0. This means claimer + * tracking is broken for those devices but it has always been that + * way. + */ + if (partno) + whole = bdget_disk(disk, 0); + else + whole = bdgrab(bdev); + module_put(disk->fops->owner); put_disk(disk); if (!whole) From e9c039052be59753e6bcc7c8b59763899dc1161c Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Mon, 13 Jun 2011 19:05:58 +0100 Subject: [PATCH 039/224] ASoC: Remove unused and about to be broken SND_SOC_CUSTOM I/O bus This will be removed in -next so let's drop it from mainline as soon as we can in order to minimise surprises. Signed-off-by: Mark Brown --- include/sound/soc.h | 3 +-- sound/soc/soc-cache.c | 3 --- 2 files changed, 1 insertion(+), 5 deletions(-) diff --git a/include/sound/soc.h b/include/sound/soc.h index f1de3e0c75bc..3a4bd3a3c68d 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -248,8 +248,7 @@ typedef int (*hw_write_t)(void *,const char* ,int); extern struct snd_ac97_bus_ops soc_ac97_ops; enum snd_soc_control_type { - SND_SOC_CUSTOM = 1, - SND_SOC_I2C, + SND_SOC_I2C = 1, SND_SOC_SPI, }; diff --git a/sound/soc/soc-cache.c b/sound/soc/soc-cache.c index c005ceb70c9d..039b9532b270 100644 --- a/sound/soc/soc-cache.c +++ b/sound/soc/soc-cache.c @@ -409,9 +409,6 @@ int snd_soc_codec_set_cache_io(struct snd_soc_codec *codec, codec->bulk_write_raw = snd_soc_hw_bulk_write_raw; switch (control) { - case SND_SOC_CUSTOM: - break; - case SND_SOC_I2C: #if defined(CONFIG_I2C) || (defined(CONFIG_I2C_MODULE) && defined(MODULE)) codec->hw_write = (hw_write_t)i2c_master_send; From de1b794130b130e77ffa975bb58cb843744f9ae5 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 13 Jun 2011 15:38:22 -0400 Subject: [PATCH 040/224] jbd2: Fix oops in jbd2_journal_remove_journal_head() jbd2_journal_remove_journal_head() can oops when trying to access journal_head returned by bh2jh(). This is caused for example by the following race: TASK1 TASK2 jbd2_journal_commit_transaction() ... processing t_forget list __jbd2_journal_refile_buffer(jh); if (!jh->b_transaction) { jbd_unlock_bh_state(bh); jbd2_journal_try_to_free_buffers() jbd2_journal_grab_journal_head(bh) jbd_lock_bh_state(bh) __journal_try_to_free_buffer() jbd2_journal_put_journal_head(jh) jbd2_journal_remove_journal_head(bh); jbd2_journal_put_journal_head() in TASK2 sees that b_jcount == 0 and buffer is not part of any transaction and thus frees journal_head before TASK1 gets to doing so. Note that even buffer_head can be released by try_to_free_buffers() after jbd2_journal_put_journal_head() which adds even larger opportunity for oops (but I didn't see this happen in reality). Fix the problem by making transactions hold their own journal_head reference (in b_jcount). That way we don't have to remove journal_head explicitely via jbd2_journal_remove_journal_head() and instead just remove journal_head when b_jcount drops to zero. The result of this is that [__]jbd2_journal_refile_buffer(), [__]jbd2_journal_unfile_buffer(), and __jdb2_journal_remove_checkpoint() can free journal_head which needs modification of a few callers. Also we have to be careful because once journal_head is removed, buffer_head might be freed as well. So we have to get our own buffer_head reference where it matters. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/jbd2/checkpoint.c | 28 +++++++------ fs/jbd2/commit.c | 33 +++++++++------- fs/jbd2/journal.c | 91 ++++++++++++++----------------------------- fs/jbd2/transaction.c | 65 ++++++++++++++++--------------- include/linux/jbd2.h | 2 - 5 files changed, 98 insertions(+), 121 deletions(-) diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 6a79fd0a1a32..2c62c5aae82f 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -97,10 +97,14 @@ static int __try_to_free_cp_buf(struct journal_head *jh) if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh) && !buffer_write_io_error(bh)) { + /* + * Get our reference so that bh cannot be freed before + * we unlock it + */ + get_bh(bh); JBUFFER_TRACE(jh, "remove from checkpoint list"); ret = __jbd2_journal_remove_checkpoint(jh) + 1; jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); BUFFER_TRACE(bh, "release"); __brelse(bh); } else { @@ -223,8 +227,8 @@ restart: spin_lock(&journal->j_list_lock); goto restart; } + get_bh(bh); if (buffer_locked(bh)) { - atomic_inc(&bh->b_count); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); wait_on_buffer(bh); @@ -243,7 +247,6 @@ restart: */ released = __jbd2_journal_remove_checkpoint(jh); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); __brelse(bh); } @@ -284,7 +287,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, int ret = 0; if (buffer_locked(bh)) { - atomic_inc(&bh->b_count); + get_bh(bh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); wait_on_buffer(bh); @@ -316,12 +319,12 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh, ret = 1; if (unlikely(buffer_write_io_error(bh))) ret = -EIO; + get_bh(bh); J_ASSERT_JH(jh, !buffer_jbddirty(bh)); BUFFER_TRACE(bh, "remove from checkpoint"); __jbd2_journal_remove_checkpoint(jh); spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); __brelse(bh); } else { /* @@ -554,7 +557,8 @@ int jbd2_cleanup_journal_tail(journal_t *journal) /* * journal_clean_one_cp_list * - * Find all the written-back checkpoint buffers in the given list and release them. + * Find all the written-back checkpoint buffers in the given list and + * release them. * * Called with the journal locked. * Called with j_list_lock held. @@ -663,8 +667,8 @@ out: * checkpoint lists. * * The function returns 1 if it frees the transaction, 0 otherwise. + * The function can free jh and bh. * - * This function is called with the journal locked. * This function is called with j_list_lock held. * This function is called with jbd_lock_bh_state(jh2bh(jh)) */ @@ -684,13 +688,14 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) } journal = transaction->t_journal; + JBUFFER_TRACE(jh, "removing from transaction"); __buffer_unlink(jh); jh->b_cp_transaction = NULL; + jbd2_journal_put_journal_head(jh); if (transaction->t_checkpoint_list != NULL || transaction->t_checkpoint_io_list != NULL) goto out; - JBUFFER_TRACE(jh, "transaction has no more buffers"); /* * There is one special case to worry about: if we have just pulled the @@ -701,10 +706,8 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) * The locking here around t_state is a bit sleazy. * See the comment at the end of jbd2_journal_commit_transaction(). */ - if (transaction->t_state != T_FINISHED) { - JBUFFER_TRACE(jh, "belongs to running/committing transaction"); + if (transaction->t_state != T_FINISHED) goto out; - } /* OK, that was the last buffer for the transaction: we can now safely remove this transaction from the log */ @@ -723,7 +726,6 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) wake_up(&journal->j_wait_logspace); ret = 1; out: - JBUFFER_TRACE(jh, "exit"); return ret; } @@ -742,6 +744,8 @@ void __jbd2_journal_insert_checkpoint(struct journal_head *jh, J_ASSERT_JH(jh, buffer_dirty(jh2bh(jh)) || buffer_jbddirty(jh2bh(jh))); J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + /* Get reference for checkpointing transaction */ + jbd2_journal_grab_journal_head(jh2bh(jh)); jh->b_cp_transaction = transaction; if (!transaction->t_checkpoint_list) { diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 7f21cf3aaf92..eef6979821a4 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -848,10 +848,16 @@ restart_loop: while (commit_transaction->t_forget) { transaction_t *cp_transaction; struct buffer_head *bh; + int try_to_free = 0; jh = commit_transaction->t_forget; spin_unlock(&journal->j_list_lock); bh = jh2bh(jh); + /* + * Get a reference so that bh cannot be freed before we are + * done with it. + */ + get_bh(bh); jbd_lock_bh_state(bh); J_ASSERT_JH(jh, jh->b_transaction == commit_transaction); @@ -914,28 +920,27 @@ restart_loop: __jbd2_journal_insert_checkpoint(jh, commit_transaction); if (is_journal_aborted(journal)) clear_buffer_jbddirty(bh); - JBUFFER_TRACE(jh, "refile for checkpoint writeback"); - __jbd2_journal_refile_buffer(jh); - jbd_unlock_bh_state(bh); } else { J_ASSERT_BH(bh, !buffer_dirty(bh)); - /* The buffer on BJ_Forget list and not jbddirty means + /* + * The buffer on BJ_Forget list and not jbddirty means * it has been freed by this transaction and hence it * could not have been reallocated until this * transaction has committed. *BUT* it could be * reallocated once we have written all the data to * disk and before we process the buffer on BJ_Forget - * list. */ - JBUFFER_TRACE(jh, "refile or unfile freed buffer"); - __jbd2_journal_refile_buffer(jh); - if (!jh->b_transaction) { - jbd_unlock_bh_state(bh); - /* needs a brelse */ - jbd2_journal_remove_journal_head(bh); - release_buffer_page(bh); - } else - jbd_unlock_bh_state(bh); + * list. + */ + if (!jh->b_next_transaction) + try_to_free = 1; } + JBUFFER_TRACE(jh, "refile or unfile buffer"); + __jbd2_journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + if (try_to_free) + release_buffer_page(bh); /* Drops bh reference */ + else + __brelse(bh); cond_resched_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 9a7826990304..0dfa5b598e68 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -2078,10 +2078,9 @@ static void journal_free_journal_head(struct journal_head *jh) * When a buffer has its BH_JBD bit set it is immune from being released by * core kernel code, mainly via ->b_count. * - * A journal_head may be detached from its buffer_head when the journal_head's - * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL. - * Various places in JBD call jbd2_journal_remove_journal_head() to indicate that the - * journal_head can be dropped if needed. + * A journal_head is detached from its buffer_head when the journal_head's + * b_jcount reaches zero. Running transaction (b_transaction) and checkpoint + * transaction (b_cp_transaction) hold their references to b_jcount. * * Various places in the kernel want to attach a journal_head to a buffer_head * _before_ attaching the journal_head to a transaction. To protect the @@ -2094,17 +2093,16 @@ static void journal_free_journal_head(struct journal_head *jh) * (Attach a journal_head if needed. Increments b_jcount) * struct journal_head *jh = jbd2_journal_add_journal_head(bh); * ... + * (Get another reference for transaction) + * jbd2_journal_grab_journal_head(bh); * jh->b_transaction = xxx; + * (Put original reference) * jbd2_journal_put_journal_head(jh); - * - * Now, the journal_head's b_jcount is zero, but it is safe from being released - * because it has a non-zero b_transaction. */ /* * Give a buffer_head a journal_head. * - * Doesn't need the journal lock. * May sleep. */ struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh) @@ -2168,61 +2166,29 @@ static void __journal_remove_journal_head(struct buffer_head *bh) struct journal_head *jh = bh2jh(bh); J_ASSERT_JH(jh, jh->b_jcount >= 0); - - get_bh(bh); - if (jh->b_jcount == 0) { - if (jh->b_transaction == NULL && - jh->b_next_transaction == NULL && - jh->b_cp_transaction == NULL) { - J_ASSERT_JH(jh, jh->b_jlist == BJ_None); - J_ASSERT_BH(bh, buffer_jbd(bh)); - J_ASSERT_BH(bh, jh2bh(jh) == bh); - BUFFER_TRACE(bh, "remove journal_head"); - if (jh->b_frozen_data) { - printk(KERN_WARNING "%s: freeing " - "b_frozen_data\n", - __func__); - jbd2_free(jh->b_frozen_data, bh->b_size); - } - if (jh->b_committed_data) { - printk(KERN_WARNING "%s: freeing " - "b_committed_data\n", - __func__); - jbd2_free(jh->b_committed_data, bh->b_size); - } - bh->b_private = NULL; - jh->b_bh = NULL; /* debug, really */ - clear_buffer_jbd(bh); - __brelse(bh); - journal_free_journal_head(jh); - } else { - BUFFER_TRACE(bh, "journal_head was locked"); - } + J_ASSERT_JH(jh, jh->b_transaction == NULL); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + J_ASSERT_JH(jh, jh->b_jlist == BJ_None); + J_ASSERT_BH(bh, buffer_jbd(bh)); + J_ASSERT_BH(bh, jh2bh(jh) == bh); + BUFFER_TRACE(bh, "remove journal_head"); + if (jh->b_frozen_data) { + printk(KERN_WARNING "%s: freeing b_frozen_data\n", __func__); + jbd2_free(jh->b_frozen_data, bh->b_size); } + if (jh->b_committed_data) { + printk(KERN_WARNING "%s: freeing b_committed_data\n", __func__); + jbd2_free(jh->b_committed_data, bh->b_size); + } + bh->b_private = NULL; + jh->b_bh = NULL; /* debug, really */ + clear_buffer_jbd(bh); + journal_free_journal_head(jh); } /* - * jbd2_journal_remove_journal_head(): if the buffer isn't attached to a transaction - * and has a zero b_jcount then remove and release its journal_head. If we did - * see that the buffer is not used by any transaction we also "logically" - * decrement ->b_count. - * - * We in fact take an additional increment on ->b_count as a convenience, - * because the caller usually wants to do additional things with the bh - * after calling here. - * The caller of jbd2_journal_remove_journal_head() *must* run __brelse(bh) at some - * time. Once the caller has run __brelse(), the buffer is eligible for - * reaping by try_to_free_buffers(). - */ -void jbd2_journal_remove_journal_head(struct buffer_head *bh) -{ - jbd_lock_bh_journal_head(bh); - __journal_remove_journal_head(bh); - jbd_unlock_bh_journal_head(bh); -} - -/* - * Drop a reference on the passed journal_head. If it fell to zero then try to + * Drop a reference on the passed journal_head. If it fell to zero then * release the journal_head from the buffer_head. */ void jbd2_journal_put_journal_head(struct journal_head *jh) @@ -2232,11 +2198,12 @@ void jbd2_journal_put_journal_head(struct journal_head *jh) jbd_lock_bh_journal_head(bh); J_ASSERT_JH(jh, jh->b_jcount > 0); --jh->b_jcount; - if (!jh->b_jcount && !jh->b_transaction) { + if (!jh->b_jcount) { __journal_remove_journal_head(bh); + jbd_unlock_bh_journal_head(bh); __brelse(bh); - } - jbd_unlock_bh_journal_head(bh); + } else + jbd_unlock_bh_journal_head(bh); } /* diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 547b101049e5..2d7109414cdd 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -30,6 +30,7 @@ #include static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); +static void __jbd2_journal_unfile_buffer(struct journal_head *jh); /* * jbd2_get_transaction: obtain a new transaction_t object. @@ -764,7 +765,6 @@ repeat: if (!jh->b_transaction) { JBUFFER_TRACE(jh, "no transaction"); J_ASSERT_JH(jh, !jh->b_next_transaction); - jh->b_transaction = transaction; JBUFFER_TRACE(jh, "file as BJ_Reserved"); spin_lock(&journal->j_list_lock); __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); @@ -895,8 +895,6 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) * committed and so it's safe to clear the dirty bit. */ clear_buffer_dirty(jh2bh(jh)); - jh->b_transaction = transaction; - /* first access by this transaction */ jh->b_modified = 0; @@ -1230,8 +1228,6 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); } else { __jbd2_journal_unfile_buffer(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); if (!buffer_jbd(bh)) { spin_unlock(&journal->j_list_lock); jbd_unlock_bh_state(bh); @@ -1554,19 +1550,32 @@ void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) mark_buffer_dirty(bh); /* Expose it to the VM */ } -void __jbd2_journal_unfile_buffer(struct journal_head *jh) +/* + * Remove buffer from all transactions. + * + * Called with bh_state lock and j_list_lock + * + * jh and bh may be already freed when this function returns. + */ +static void __jbd2_journal_unfile_buffer(struct journal_head *jh) { __jbd2_journal_temp_unlink_buffer(jh); jh->b_transaction = NULL; + jbd2_journal_put_journal_head(jh); } void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh) { - jbd_lock_bh_state(jh2bh(jh)); + struct buffer_head *bh = jh2bh(jh); + + /* Get reference so that buffer cannot be freed before we unlock it */ + get_bh(bh); + jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); __jbd2_journal_unfile_buffer(jh); spin_unlock(&journal->j_list_lock); - jbd_unlock_bh_state(jh2bh(jh)); + jbd_unlock_bh_state(bh); + __brelse(bh); } /* @@ -1593,8 +1602,6 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) if (jh->b_jlist == BJ_None) { JBUFFER_TRACE(jh, "remove from checkpoint list"); __jbd2_journal_remove_checkpoint(jh); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); } } spin_unlock(&journal->j_list_lock); @@ -1657,7 +1664,6 @@ 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(). */ jh = jbd2_journal_grab_journal_head(bh); @@ -1695,10 +1701,9 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) int may_free = 1; struct buffer_head *bh = jh2bh(jh); - __jbd2_journal_unfile_buffer(jh); - if (jh->b_cp_transaction) { JBUFFER_TRACE(jh, "on running+cp transaction"); + __jbd2_journal_temp_unlink_buffer(jh); /* * We don't want to write the buffer anymore, clear the * bit so that we don't confuse checks in @@ -1709,8 +1714,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) may_free = 0; } else { JBUFFER_TRACE(jh, "on running transaction"); - jbd2_journal_remove_journal_head(bh); - __brelse(bh); + __jbd2_journal_unfile_buffer(jh); } return may_free; } @@ -1988,6 +1992,8 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, if (jh->b_transaction) __jbd2_journal_temp_unlink_buffer(jh); + else + jbd2_journal_grab_journal_head(bh); jh->b_transaction = transaction; switch (jlist) { @@ -2039,9 +2045,10 @@ void jbd2_journal_file_buffer(struct journal_head *jh, * already started to be used by a subsequent transaction, refile the * buffer on that transaction's metadata list. * - * Called under journal->j_list_lock - * + * Called under j_list_lock * Called under jbd_lock_bh_state(jh2bh(jh)) + * + * jh and bh may be already free when this function returns */ void __jbd2_journal_refile_buffer(struct journal_head *jh) { @@ -2065,6 +2072,11 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) was_dirty = test_clear_buffer_jbddirty(bh); __jbd2_journal_temp_unlink_buffer(jh); + /* + * We set b_transaction here because b_next_transaction will inherit + * our jh reference and thus __jbd2_journal_file_buffer() must not + * take a new one. + */ jh->b_transaction = jh->b_next_transaction; jh->b_next_transaction = NULL; if (buffer_freed(bh)) @@ -2081,30 +2093,21 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh) } /* - * For the unlocked version of this call, also make sure that any - * hanging journal_head is cleaned up if necessary. + * __jbd2_journal_refile_buffer() with necessary locking added. We take our + * bh reference so that we can safely unlock bh. * - * __jbd2_journal_refile_buffer is usually called as part of a single locked - * operation on a buffer_head, in which the caller is probably going to - * be hooking the journal_head onto other lists. In that case it is up - * to the caller to remove the journal_head if necessary. For the - * unlocked jbd2_journal_refile_buffer call, the caller isn't going to be - * doing anything else to the buffer so we need to do the cleanup - * ourselves to avoid a jh leak. - * - * *** The journal_head may be freed by this call! *** + * The jh and bh may be freed by this call. */ void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) { struct buffer_head *bh = jh2bh(jh); + /* Get reference so that buffer cannot be freed before we unlock it */ + get_bh(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); - __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); - jbd2_journal_remove_journal_head(bh); - spin_unlock(&journal->j_list_lock); __brelse(bh); } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 4ecb7b16b278..d087c2e7b2aa 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1024,7 +1024,6 @@ struct journal_s /* Filing buffers */ extern void jbd2_journal_unfile_buffer(journal_t *, struct journal_head *); -extern void __jbd2_journal_unfile_buffer(struct journal_head *); extern void __jbd2_journal_refile_buffer(struct journal_head *); extern void jbd2_journal_refile_buffer(journal_t *, struct journal_head *); extern void __jbd2_journal_file_buffer(struct journal_head *, transaction_t *, int); @@ -1165,7 +1164,6 @@ extern void jbd2_journal_release_jbd_inode(journal_t *journal, struct jbd2_in */ struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh); struct journal_head *jbd2_journal_grab_journal_head(struct buffer_head *bh); -void jbd2_journal_remove_journal_head(struct buffer_head *bh); void jbd2_journal_put_journal_head(struct journal_head *jh); /* From 357f45db920393aac983a137bd74095f612d5a01 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Mon, 13 Jun 2011 22:50:41 +0200 Subject: [PATCH 041/224] USB: TI 3410/5052 USB Serial Driver: Fix mem leak when firmware is too big. If the size of the firmware exceeds TI_FIRMWARE_BUF_SIZE we'll leak 'fw_p' by failing to call release_firmware(). This patch fixes the leak. Signed-off-by: Jesper Juhl Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/ti_usb_3410_5052.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/serial/ti_usb_3410_5052.c b/drivers/usb/serial/ti_usb_3410_5052.c index c6d92a530086..ea8445689c85 100644 --- a/drivers/usb/serial/ti_usb_3410_5052.c +++ b/drivers/usb/serial/ti_usb_3410_5052.c @@ -1745,6 +1745,7 @@ static int ti_download_firmware(struct ti_device *tdev) } if (fw_p->size > TI_FIRMWARE_BUF_SIZE) { dev_err(&dev->dev, "%s - firmware too large %zu\n", __func__, fw_p->size); + release_firmware(fw_p); return -ENOENT; } From 9e3bd4e24e94d60d2e0762e919aab6c9a7fc0c5b Mon Sep 17 00:00:00 2001 From: Weston Andros Adamson Date: Tue, 31 May 2011 21:46:50 -0400 Subject: [PATCH 042/224] NFS: fix umount of pnfs filesystems Unmounting a pnfs filesystem hangs using filelayout and possibly others. This fixes the use of the rcu protected node by making use of a new 'tmpnode' for the temporary purge list. Also, the spinlock shouldn't be held when calling synchronize_rcu(). Signed-off-by: Weston Andros Adamson Signed-off-by: Andy Adamson Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.h | 1 + fs/nfs/pnfs_dev.c | 16 +++++++++++----- 2 files changed, 12 insertions(+), 5 deletions(-) diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 48d0a8e4d062..96bf4e6f45be 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -186,6 +186,7 @@ int pnfs_ld_read_done(struct nfs_read_data *); /* pnfs_dev.c */ struct nfs4_deviceid_node { struct hlist_node node; + struct hlist_node tmpnode; const struct pnfs_layoutdriver_type *ld; const struct nfs_client *nfs_client; struct nfs4_deviceid deviceid; diff --git a/fs/nfs/pnfs_dev.c b/fs/nfs/pnfs_dev.c index c65e133ce9c0..5944d4b369a2 100644 --- a/fs/nfs/pnfs_dev.c +++ b/fs/nfs/pnfs_dev.c @@ -174,6 +174,7 @@ nfs4_init_deviceid_node(struct nfs4_deviceid_node *d, const struct nfs4_deviceid *id) { INIT_HLIST_NODE(&d->node); + INIT_HLIST_NODE(&d->tmpnode); d->ld = ld; d->nfs_client = nfs_client; d->deviceid = *id; @@ -238,24 +239,29 @@ static void _deviceid_purge_client(const struct nfs_client *clp, long hash) { struct nfs4_deviceid_node *d; - struct hlist_node *n, *next; + struct hlist_node *n; HLIST_HEAD(tmp); + spin_lock(&nfs4_deviceid_lock); rcu_read_lock(); hlist_for_each_entry_rcu(d, n, &nfs4_deviceid_cache[hash], node) if (d->nfs_client == clp && atomic_read(&d->ref)) { hlist_del_init_rcu(&d->node); - hlist_add_head(&d->node, &tmp); + hlist_add_head(&d->tmpnode, &tmp); } rcu_read_unlock(); + spin_unlock(&nfs4_deviceid_lock); if (hlist_empty(&tmp)) return; synchronize_rcu(); - hlist_for_each_entry_safe(d, n, next, &tmp, node) + while (!hlist_empty(&tmp)) { + d = hlist_entry(tmp.first, struct nfs4_deviceid_node, tmpnode); + hlist_del(&d->tmpnode); if (atomic_dec_and_test(&d->ref)) d->ld->free_deviceid_node(d); + } } void @@ -263,8 +269,8 @@ nfs4_deviceid_purge_client(const struct nfs_client *clp) { long h; - spin_lock(&nfs4_deviceid_lock); + if (!(clp->cl_exchange_flags & EXCHGID4_FLAG_USE_PNFS_MDS)) + return; for (h = 0; h < NFS4_DEVICE_ID_HASH_SIZE; h++) _deviceid_purge_client(clp, h); - spin_unlock(&nfs4_deviceid_lock); } From 0b760113a3a155269a3fba93a409c640031dd68f Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 31 May 2011 15:15:34 -0400 Subject: [PATCH 043/224] NLM: Don't hang forever on NLM unlock requests If the NLM daemon is killed on the NFS server, we can currently end up hanging forever on an 'unlock' request, instead of aborting. Basically, if the rpcbind request fails, or the server keeps returning garbage, we really want to quit instead of retrying. Tested-by: Vasily Averin Signed-off-by: Trond Myklebust Cc: stable@kernel.org --- fs/lockd/clntproc.c | 8 +++++++- include/linux/sunrpc/sched.h | 3 ++- net/sunrpc/clnt.c | 3 +++ net/sunrpc/sched.c | 1 + 4 files changed, 13 insertions(+), 2 deletions(-) diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index adb45ec9038c..e374050a911c 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -708,7 +708,13 @@ static void nlmclnt_unlock_callback(struct rpc_task *task, void *data) if (task->tk_status < 0) { dprintk("lockd: unlock failed (err = %d)\n", -task->tk_status); - goto retry_rebind; + switch (task->tk_status) { + case -EACCES: + case -EIO: + goto die; + default: + goto retry_rebind; + } } if (status == NLM_LCK_DENIED_GRACE_PERIOD) { rpc_delay(task, NLMCLNT_GRACE_WAIT); diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index f73c482ec9c6..fe2d8e6b923b 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -84,7 +84,8 @@ struct rpc_task { #endif unsigned char tk_priority : 2,/* Task priority */ tk_garb_retry : 2, - tk_cred_retry : 2; + tk_cred_retry : 2, + tk_rebind_retry : 2; }; #define tk_xprt tk_client->cl_xprt diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index b84d7395535e..566bcfd067f6 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1175,6 +1175,9 @@ call_bind_status(struct rpc_task *task) status = -EOPNOTSUPP; break; } + if (task->tk_rebind_retry == 0) + break; + task->tk_rebind_retry--; rpc_delay(task, 3*HZ); goto retry_timeout; case -ETIMEDOUT: diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 6b43ee7221d5..a27406b1654f 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -792,6 +792,7 @@ static void rpc_init_task(struct rpc_task *task, const struct rpc_task_setup *ta /* Initialize retry counters */ task->tk_garb_retry = 2; task->tk_cred_retry = 2; + task->tk_rebind_retry = 2; task->tk_priority = task_setup_data->priority - RPC_PRIORITY_LOW; task->tk_owner = current->tgid; From 0f66b5984df2fe1617c05900a39a7ef493ca9de9 Mon Sep 17 00:00:00 2001 From: Peng Tao Date: Sat, 16 Oct 2010 22:07:46 -0700 Subject: [PATCH 044/224] NFS41: do not update isize if inode needs layoutcommit nfs_update_inode will update isize if there is no queued pages. For pNFS, layoutcommit is supposed to change file size on server, the same effect as queued pages. nfs_update_inode may be called when dirty pages are written back (nfsi->npages==0) but layoutcommit is not sent, and it will change client file size according to server file size. Then client ends up losing what it just writes back in pNFS path. So we should skip updating client file size if file needs layoutcommit. Signed-off-by: Peng Tao Cc: stable@kernel.org [2.6.39] Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 144f2a3c7185..3f1eb8180373 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1294,7 +1294,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (new_isize != cur_isize) { /* Do we perhaps have any outstanding writes, or has * the file grown beyond our last write? */ - if (nfsi->npages == 0 || new_isize > cur_isize) { + if ((nfsi->npages == 0 && !test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) || + new_isize > cur_isize) { i_size_write(inode, new_isize); invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; } From c9c30dd5f73dccaa326a54dfcf490316946aea87 Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Sat, 11 Jun 2011 17:08:39 -0400 Subject: [PATCH 045/224] NFSv4.1: deprecate headerpadsz in CREATE_SESSION We don't support header padding yet so better off ditching it Reported-by: Sid Moore Signed-off-by: Benny Halevy Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 4 ---- fs/nfs/nfs4xdr.c | 10 ++++++---- include/linux/nfs_xdr.h | 1 - 3 files changed, 6 insertions(+), 9 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index d2c4b59c896d..3ca449789545 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -5098,7 +5098,6 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) if (mxresp_sz == 0) mxresp_sz = NFS_MAX_FILE_IO_SIZE; /* Fore channel attributes */ - args->fc_attrs.headerpadsz = 0; args->fc_attrs.max_rqst_sz = mxrqst_sz; args->fc_attrs.max_resp_sz = mxresp_sz; args->fc_attrs.max_ops = NFS4_MAX_OPS; @@ -5111,7 +5110,6 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args) args->fc_attrs.max_ops, args->fc_attrs.max_reqs); /* Back channel attributes */ - args->bc_attrs.headerpadsz = 0; args->bc_attrs.max_rqst_sz = PAGE_SIZE; args->bc_attrs.max_resp_sz = PAGE_SIZE; args->bc_attrs.max_resp_sz_cached = 0; @@ -5131,8 +5129,6 @@ static int nfs4_verify_fore_channel_attrs(struct nfs41_create_session_args *args struct nfs4_channel_attrs *sent = &args->fc_attrs; struct nfs4_channel_attrs *rcvd = &session->fc_attrs; - if (rcvd->headerpadsz > sent->headerpadsz) - return -EINVAL; if (rcvd->max_resp_sz > sent->max_resp_sz) return -EINVAL; /* diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index d869a5e5464b..c4b7d6c04948 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -1725,7 +1725,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(args->flags); /*flags */ /* Fore Channel */ - *p++ = cpu_to_be32(args->fc_attrs.headerpadsz); /* header padding size */ + *p++ = cpu_to_be32(0); /* header padding size */ *p++ = cpu_to_be32(args->fc_attrs.max_rqst_sz); /* max req size */ *p++ = cpu_to_be32(args->fc_attrs.max_resp_sz); /* max resp size */ *p++ = cpu_to_be32(max_resp_sz_cached); /* Max resp sz cached */ @@ -1734,7 +1734,7 @@ static void encode_create_session(struct xdr_stream *xdr, *p++ = cpu_to_be32(0); /* rdmachannel_attrs */ /* Back Channel */ - *p++ = cpu_to_be32(args->fc_attrs.headerpadsz); /* header padding size */ + *p++ = cpu_to_be32(0); /* header padding size */ *p++ = cpu_to_be32(args->bc_attrs.max_rqst_sz); /* max req size */ *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz); /* max resp size */ *p++ = cpu_to_be32(args->bc_attrs.max_resp_sz_cached); /* Max resp sz cached */ @@ -4997,12 +4997,14 @@ static int decode_chan_attrs(struct xdr_stream *xdr, struct nfs4_channel_attrs *attrs) { __be32 *p; - u32 nr_attrs; + u32 nr_attrs, val; p = xdr_inline_decode(xdr, 28); if (unlikely(!p)) goto out_overflow; - attrs->headerpadsz = be32_to_cpup(p++); + val = be32_to_cpup(p++); /* headerpadsz */ + if (val) + return -EINVAL; /* no support for header padding yet */ attrs->max_rqst_sz = be32_to_cpup(p++); attrs->max_resp_sz = be32_to_cpup(p++); attrs->max_resp_sz_cached = be32_to_cpup(p++); diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 5e8444a11adf..00848d86ffb2 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -158,7 +158,6 @@ struct nfs_seqid; /* nfs41 sessions channel attributes */ struct nfs4_channel_attrs { - u32 headerpadsz; u32 max_rqst_sz; u32 max_resp_sz; u32 max_resp_sz_cached; From 1d92a08da23848a38eece4df7eaa4e8ec0e6c699 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 14 Jun 2011 12:07:38 -0400 Subject: [PATCH 046/224] NFSv4.1: Fix a refcounting issue in the pNFS device id cache When we add something to the global device id cache, we need to bump the reference count, so that the cache itself holds a reference. Signed-off-by: Trond Myklebust --- fs/nfs/pnfs_dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/nfs/pnfs_dev.c b/fs/nfs/pnfs_dev.c index 5944d4b369a2..f0f8e1e22f6c 100644 --- a/fs/nfs/pnfs_dev.c +++ b/fs/nfs/pnfs_dev.c @@ -209,6 +209,7 @@ nfs4_insert_deviceid_node(struct nfs4_deviceid_node *new) hlist_add_head_rcu(&new->node, &nfs4_deviceid_cache[hash]); spin_unlock(&nfs4_deviceid_lock); + atomic_inc(&new->ref); return new; } From 533eb4611c9eea53072eb6a61d5a6393b6a77ed7 Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Mon, 13 Jun 2011 18:25:56 -0400 Subject: [PATCH 047/224] NFSv4.1: allow nfs_fhget to succeed with mounted on fileid Commit 28331a46d88459788c8fca72dbb0415cd7f514c9 "Ensure we request the ordinary fileid when doing readdirplus" changed the meaning of NFS_ATTR_FATTR_FILEID which used to be set when FATTR4_WORD1_MOUNTED_ON_FILED was requested. Allow nfs_fhget to succeed with only a mounted on fileid when crossing a mountpoint or a referral. Ask for the fileid of the absent file system if mounted_on_fileid is not supported. Signed-off-by: Andy Adamson cc:stable@kernel.org [2.6.39] Signed-off-by: Trond Myklebust --- fs/nfs/inode.c | 3 ++- fs/nfs/internal.h | 11 +++++++++++ fs/nfs/nfs4proc.c | 33 +++++++++++++++++++++++---------- 3 files changed, 36 insertions(+), 11 deletions(-) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 3f1eb8180373..6f4850deb272 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -256,7 +256,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) nfs_attr_check_mountpoint(sb, fattr); - if ((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0 && (fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0) + if (((fattr->valid & NFS_ATTR_FATTR_FILEID) == 0) && + !nfs_attr_use_mounted_on_fileid(fattr)) goto out_no_inode; if ((fattr->valid & NFS_ATTR_FATTR_TYPE) == 0) goto out_no_inode; diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index b9056cbe68d6..2a55347a2daa 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -45,6 +45,17 @@ static inline void nfs_attr_check_mountpoint(struct super_block *parent, struct fattr->valid |= NFS_ATTR_FATTR_MOUNTPOINT; } +static inline int nfs_attr_use_mounted_on_fileid(struct nfs_fattr *fattr) +{ + if (((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) == 0) || + (((fattr->valid & NFS_ATTR_FATTR_MOUNTPOINT) == 0) && + ((fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL) == 0))) + return 0; + + fattr->fileid = fattr->mounted_on_fileid; + return 1; +} + struct nfs_clone_mount { const struct super_block *sb; const struct dentry *dentry; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 3ca449789545..289c539bd6b3 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2265,12 +2265,14 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, return nfs4_map_errors(status); } +static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr); /* * Get locations and (maybe) other attributes of a referral. * Note that we'll actually follow the referral later when * we detect fsid mismatch in inode revalidation */ -static int nfs4_get_referral(struct inode *dir, const struct qstr *name, struct nfs_fattr *fattr, struct nfs_fh *fhandle) +static int nfs4_get_referral(struct inode *dir, const struct qstr *name, + struct nfs_fattr *fattr, struct nfs_fh *fhandle) { int status = -ENOMEM; struct page *page = NULL; @@ -2288,15 +2290,16 @@ static int nfs4_get_referral(struct inode *dir, const struct qstr *name, struct goto out; /* Make sure server returned a different fsid for the referral */ if (nfs_fsid_equal(&NFS_SERVER(dir)->fsid, &locations->fattr.fsid)) { - dprintk("%s: server did not return a different fsid for a referral at %s\n", __func__, name->name); + dprintk("%s: server did not return a different fsid for" + " a referral at %s\n", __func__, name->name); status = -EIO; goto out; } + /* Fixup attributes for the nfs_lookup() call to nfs_fhget() */ + nfs_fixup_referral_attributes(&locations->fattr); + /* replace the lookup nfs_fattr with the locations nfs_fattr */ memcpy(fattr, &locations->fattr, sizeof(struct nfs_fattr)); - fattr->valid |= NFS_ATTR_FATTR_V4_REFERRAL; - if (!fattr->mode) - fattr->mode = S_IFDIR; memset(fhandle, 0, sizeof(struct nfs_fh)); out: if (page) @@ -4667,11 +4670,15 @@ static size_t nfs4_xattr_list_nfs4_acl(struct dentry *dentry, char *list, return len; } +/* + * nfs_fhget will use either the mounted_on_fileid or the fileid + */ static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr) { - if (!((fattr->valid & NFS_ATTR_FATTR_FILEID) && - (fattr->valid & NFS_ATTR_FATTR_FSID) && - (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL))) + if (!(((fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID) || + (fattr->valid & NFS_ATTR_FATTR_FILEID)) && + (fattr->valid & NFS_ATTR_FATTR_FSID) && + (fattr->valid & NFS_ATTR_FATTR_V4_REFERRAL))) return; fattr->valid |= NFS_ATTR_FATTR_TYPE | NFS_ATTR_FATTR_MODE | @@ -4686,7 +4693,6 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, struct nfs_server *server = NFS_SERVER(dir); u32 bitmask[2] = { [0] = FATTR4_WORD0_FSID | FATTR4_WORD0_FS_LOCATIONS, - [1] = FATTR4_WORD1_MOUNTED_ON_FILEID, }; struct nfs4_fs_locations_arg args = { .dir_fh = NFS_FH(dir), @@ -4705,11 +4711,18 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name, int status; dprintk("%s: start\n", __func__); + + /* Ask for the fileid of the absent filesystem if mounted_on_fileid + * is not supported */ + if (NFS_SERVER(dir)->attr_bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID) + bitmask[1] |= FATTR4_WORD1_MOUNTED_ON_FILEID; + else + bitmask[0] |= FATTR4_WORD0_FILEID; + nfs_fattr_init(&fs_locations->fattr); fs_locations->server = server; fs_locations->nlocations = 0; status = nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); - nfs_fixup_referral_attributes(&fs_locations->fattr); dprintk("%s: returned status = %d\n", __func__, status); return status; } From cec765cf5891c7fc3d905832b481bfb6fd55825d Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Mon, 13 Jun 2011 18:36:17 -0400 Subject: [PATCH 048/224] NFSv4.1: allow zero fh array in filelayout decode layout Signed-off-by: Andy Adamson cc:stable@kernel.org [2.6.39] Signed-off-by: Trond Myklebust --- fs/nfs/nfs4filelayout.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 426908809c97..5d6f369b15d0 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -552,13 +552,18 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo, __func__, nfl_util, fl->num_fh, fl->first_stripe_index, fl->pattern_offset); - if (!fl->num_fh) + /* Note that a zero value for num_fh is legal for STRIPE_SPARSE. + * Futher checking is done in filelayout_check_layout */ + if (fl->num_fh < 0 || fl->num_fh > + max(NFS4_PNFS_MAX_STRIPE_CNT, NFS4_PNFS_MAX_MULTI_CNT)) goto out_err; - fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), - gfp_flags); - if (!fl->fh_array) - goto out_err; + if (fl->num_fh > 0) { + fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *), + gfp_flags); + if (!fl->fh_array) + goto out_err; + } for (i = 0; i < fl->num_fh; i++) { /* Do we want to use a mempool here? */ From a2e1d4f2e5ed83850de92a491ef225824cb457bd Mon Sep 17 00:00:00 2001 From: Fred Isaman Date: Mon, 13 Jun 2011 18:54:53 -0400 Subject: [PATCH 049/224] nfs4.1: fix several problems with _pnfs_return_layout _pnfs_return_layout had the following problems: - it did not call pnfs_free_lseg_list on all paths - it unintentionally did a forgetful return when there was no outstanding io - it raced with concurrent LAYOUTGETS Signed-off-by: Fred Isaman Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 8 ++++---- fs/nfs/pnfs.c | 8 +++++--- 2 files changed, 9 insertions(+), 7 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 289c539bd6b3..5879b23e0c99 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -5706,6 +5706,7 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) { struct nfs4_layoutreturn *lrp = calldata; struct nfs_server *server; + struct pnfs_layout_hdr *lo = NFS_I(lrp->args.inode)->layout; dprintk("--> %s\n", __func__); @@ -5717,16 +5718,15 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) nfs_restart_rpc(task, lrp->clp); return; } + spin_lock(&lo->plh_inode->i_lock); if (task->tk_status == 0) { - struct pnfs_layout_hdr *lo = NFS_I(lrp->args.inode)->layout; - if (lrp->res.lrs_present) { - spin_lock(&lo->plh_inode->i_lock); pnfs_set_layout_stateid(lo, &lrp->res.stateid, true); - spin_unlock(&lo->plh_inode->i_lock); } else BUG_ON(!list_empty(&lo->plh_segs)); } + lo->plh_block_lgets--; + spin_unlock(&lo->plh_inode->i_lock); dprintk("<-- %s\n", __func__); } diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 8c1309d852a6..25de6b27bdf4 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -634,12 +634,14 @@ _pnfs_return_layout(struct inode *ino) spin_lock(&ino->i_lock); lo = nfsi->layout; - if (!lo || !mark_matching_lsegs_invalid(lo, &tmp_list, NULL)) { + if (!lo) { spin_unlock(&ino->i_lock); - dprintk("%s: no layout segments to return\n", __func__); - goto out; + dprintk("%s: no layout to return\n", __func__); + return status; } stateid = nfsi->layout->plh_stateid; + mark_matching_lsegs_invalid(lo, &tmp_list, NULL); + lo->plh_block_lgets++; /* Reference matched in nfs4_layoutreturn_release */ get_layout_hdr(lo); spin_unlock(&ino->i_lock); From d771e3a43e23a37398b7e05a9d1b1036d698263c Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Tue, 14 Jun 2011 16:30:16 -0400 Subject: [PATCH 050/224] NFSv4.1: fix break condition in pnfs_find_lseg The break condition to skip out of the loop got broken when cmp_layout was change. Essentially, we want to stop looking once we know no layout on the remainder of the list can match the first byte of the looked-up range. Reported-by: Peng Tao Signed-off-by: Benny Halevy Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 25de6b27bdf4..d066aad608ad 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -889,7 +889,7 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, ret = get_lseg(lseg); break; } - if (cmp_layout(range, &lseg->pls_range) > 0) + if (lseg->pls_range.offset > range->offset) break; } From c7fd06228b994190d8369a2a0acf5224e4e13d1a Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 15 Jun 2011 00:55:44 +0100 Subject: [PATCH 051/224] NFS: (d)printks should use %zd for ssize_t arguments (d)printks should use %zd for ssize_t arguments not %ld, otherwise they might get a warning. I see the following with MN10300. fs/nfs/objlayout/objlayout.c: In function 'objlayout_read_done': fs/nfs/objlayout/objlayout.c:294: warning: format '%ld' expects type 'long int', but argument 3 has type 'ssize_t' Signed-off-by: David Howells cc: Trond Myklebust cc: linux-nfs@vger.kernel.org Signed-off-by: Trond Myklebust --- fs/nfs/objlayout/objlayout.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c index dc3956c0de80..1d06f8e2adea 100644 --- a/fs/nfs/objlayout/objlayout.c +++ b/fs/nfs/objlayout/objlayout.c @@ -291,7 +291,7 @@ objlayout_read_done(struct objlayout_io_state *state, ssize_t status, bool sync) struct nfs_read_data *rdata; state->status = status; - dprintk("%s: Begin status=%ld eof=%d\n", __func__, status, eof); + dprintk("%s: Begin status=%zd eof=%d\n", __func__, status, eof); rdata = state->rpcdata; rdata->task.tk_status = status; if (status >= 0) { From 1ed3a8539af7b36aa5c977f304e80f7fc8d27bfc Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Wed, 15 Jun 2011 11:39:57 -0400 Subject: [PATCH 052/224] NFSv4.1: need to put_layout_hdr on _pnfs_return_layout error path We always get a reference on the layout header and we rely on nfs4_layoutreturn_release to put it. If we hit an allocation error before starting the rpc proc we bail out early without dereferncing the layout header properly. Signed-off-by: Benny Halevy Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index d066aad608ad..8f9582281252 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -652,6 +652,7 @@ _pnfs_return_layout(struct inode *ino) lrp = kzalloc(sizeof(*lrp), GFP_KERNEL); if (unlikely(lrp == NULL)) { status = -ENOMEM; + put_layout_hdr(lo); goto out; } From ea0ded748bdea78f9e2fefb571f7d6ce9edb4f89 Mon Sep 17 00:00:00 2001 From: Fred Isaman Date: Wed, 15 Jun 2011 12:31:02 -0400 Subject: [PATCH 053/224] nfs4.1: prevent race that allowed use of freed layout in _pnfs_return_layout mark_matching_lsegs_invalid could put the last ref to the layout, so the get_layout_hdr needs to be called first. Signed-off-by: Fred Isaman Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 8f9582281252..730d4dbbaf68 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -640,10 +640,10 @@ _pnfs_return_layout(struct inode *ino) return status; } stateid = nfsi->layout->plh_stateid; - mark_matching_lsegs_invalid(lo, &tmp_list, NULL); - lo->plh_block_lgets++; /* Reference matched in nfs4_layoutreturn_release */ get_layout_hdr(lo); + mark_matching_lsegs_invalid(lo, &tmp_list, NULL); + lo->plh_block_lgets++; spin_unlock(&ino->i_lock); pnfs_free_lseg_list(&tmp_list); From 9e2dfdb3081edfae66a49013517e80dd8a0469fa Mon Sep 17 00:00:00 2001 From: Fred Isaman Date: Wed, 15 Jun 2011 14:32:02 -0400 Subject: [PATCH 054/224] nfs4.1: mark layout as bad on error path in _pnfs_return_layout Signed-off-by: Fred Isaman Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 730d4dbbaf68..539b94cb6c0f 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -652,6 +652,8 @@ _pnfs_return_layout(struct inode *ino) lrp = kzalloc(sizeof(*lrp), GFP_KERNEL); if (unlikely(lrp == NULL)) { status = -ENOMEM; + set_bit(NFS_LAYOUT_RW_FAILED, &lo->plh_flags); + set_bit(NFS_LAYOUT_RO_FAILED, &lo->plh_flags); put_layout_hdr(lo); goto out; } From fa75ac379e63c2864e9049b5e8615e40f65c1e70 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Sun, 5 Jun 2011 23:10:04 -0700 Subject: [PATCH 055/224] xhci: Reject double add of active endpoints. While trying to switch a UAS device from the BOT configuration to the UAS configuration via the bConfigurationValue file, Tanya ran into an issue in the USB core. usb_disable_device() sets entries in udev->ep_out and udev->ep_out to NULL, but doesn't call into the xHCI bandwidth management functions to remove the BOT configuration endpoints from the xHCI host's internal structures. The USB core would then attempt to add endpoints for the UAS configuration, and some of the endpoints had the same address as endpoints in the BOT configuration. The xHCI driver blindly added the endpoints again, but the xHCI host controller rejected the Configure Endpoint command because active endpoints were added without being dropped. Make the xHCI driver reject calls to xhci_add_endpoint() that attempt to add active endpoints without first calling xhci_drop_endpoint(). This should be backported to kernels as old as 2.6.31. Signed-off-by: Sarah Sharp Reported-by: Tanya Brokhman Cc: stable@kernel.org --- drivers/usb/host/xhci.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 06e7023258d0..e5a01713f937 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -1401,6 +1401,7 @@ int xhci_add_endpoint(struct usb_hcd *hcd, struct usb_device *udev, u32 added_ctxs; unsigned int last_ctx; u32 new_add_flags, new_drop_flags, new_slot_info; + struct xhci_virt_device *virt_dev; int ret = 0; ret = xhci_check_args(hcd, udev, ep, 1, true, __func__); @@ -1425,11 +1426,25 @@ int xhci_add_endpoint(struct usb_hcd *hcd, struct usb_device *udev, return 0; } - in_ctx = xhci->devs[udev->slot_id]->in_ctx; - out_ctx = xhci->devs[udev->slot_id]->out_ctx; + virt_dev = xhci->devs[udev->slot_id]; + in_ctx = virt_dev->in_ctx; + out_ctx = virt_dev->out_ctx; ctrl_ctx = xhci_get_input_control_ctx(xhci, in_ctx); ep_index = xhci_get_endpoint_index(&ep->desc); ep_ctx = xhci_get_ep_ctx(xhci, out_ctx, ep_index); + + /* If this endpoint is already in use, and the upper layers are trying + * to add it again without dropping it, reject the addition. + */ + if (virt_dev->eps[ep_index].ring && + !(le32_to_cpu(ctrl_ctx->drop_flags) & + xhci_get_endpoint_flag(&ep->desc))) { + xhci_warn(xhci, "Trying to add endpoint 0x%x " + "without dropping it.\n", + (unsigned int) ep->desc.bEndpointAddress); + return -EINVAL; + } + /* If the HCD has already noted the endpoint is enabled, * ignore this request. */ @@ -1445,8 +1460,7 @@ int xhci_add_endpoint(struct usb_hcd *hcd, struct usb_device *udev, * process context, not interrupt context (or so documenation * for usb_set_interface() and usb_set_configuration() claim). */ - if (xhci_endpoint_init(xhci, xhci->devs[udev->slot_id], - udev, ep, GFP_NOIO) < 0) { + if (xhci_endpoint_init(xhci, virt_dev, udev, ep, GFP_NOIO) < 0) { dev_dbg(&udev->dev, "%s - could not initialize ep %#x\n", __func__, ep->desc.bEndpointAddress); return -ENOMEM; From fccf4e86200b8f5edd9a65da26f150e32ba79808 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Sun, 5 Jun 2011 23:22:22 -0700 Subject: [PATCH 056/224] USB: Free bandwidth when usb_disable_device is called. Tanya ran into an issue when trying to switch a UAS device from the BOT configuration to the UAS configuration via the bConfigurationValue sysfs file. Before installing the UAS configuration, set_bConfigurationValue() calls usb_disable_device(). That function is supposed to remove all host controller resources associated with that device, but it leaves some state in the xHCI host controller. Commit 0791971ba8fbc44e4f476079f856335ed45e6324 usb: allow drivers to use allocated bandwidth until unbound added a call to usb_disable_device() in usb_set_configuration(), before the xHCI bandwidth functions were invoked. That commit fixed a bug, but also introduced a bug that is triggered when a configured device is switched to a new configuration. usb_disable_device() goes through all the motions of unbinding the drivers attached to active interfaces and removing the USB core structures associated with those interfaces, but it doesn't actually remove the endpoints from the internal xHCI host controller bandwidth structures. When usb_disable_device() calls usb_disable_endpoint() with reset_hardware set to true, the entries in udev->ep_out and udev->ep_in will be set to NULL. Usually, when the USB core installs a new configuration, usb_hcd_alloc_bandwidth() will drop all non-NULL endpoints in udev->ep_out and udev->ep_in before adding any new endpoints. However, when the new UAS configuration was added, all those entries were null, so none of the old endpoints in the BOT configuration were dropped. The xHCI driver blindly added the UAS configuration endpoints, and some of the endpoint addresses overlapped with the old BOT configuration endpoints. This caused the xHCI host to reject the Configure Endpoint command. Now that the xHCI driver code is cleaned up to reject a double-add of active endpoints, we need to fix the USB core to properly drop old endpoints in usb_disable_device(). If the host controller driver needs bandwidth checking support, make usb_disable_device() call usb_disable_endpoint() with reset_hardware set to false, drop the endpoints from the xHCI host controller, and then call usb_disable_endpoint() again with reset_hardware set to true. The first call to usb_disable_endpoint() will cancel any pending URBs and wait on them to be freed in usb_hcd_disable_endpoint(), but will keep the pointers in udev->ep_out and udev->ep in intact. Then usb_hcd_alloc_bandwidth() will use those pointers to know which endpoints to drop. The final call to usb_disable_endpoint() will do two things: 1. It will call usb_hcd_disable_endpoint() again, which should be harmless since the ep->urb_list should be empty after the first call to usb_disable_endpoint() returns. 2. It will set the entries in udev->ep_out and udev->ep in to NULL, and call usb_hcd_disable_endpoint(). That call will have no effect, since the xHCI driver doesn't set the endpoint_disable function pointer. Note that usb_disable_device() will now need to be called with hcd->bandwidth_mutex held. This should be backported to kernels as old as 2.6.32. Signed-off-by: Sarah Sharp Reported-by: Tanya Brokhman Cc: ablay@codeaurora.org Cc: Alan Stern Cc: stable@kernel.org --- drivers/usb/core/hub.c | 3 +++ drivers/usb/core/message.c | 15 ++++++++++++++- 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 90ae1753dda1..ca339bc799e4 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -1634,6 +1634,7 @@ void usb_disconnect(struct usb_device **pdev) { struct usb_device *udev = *pdev; int i; + struct usb_hcd *hcd = bus_to_hcd(udev->bus); if (!udev) { pr_debug ("%s nodev\n", __func__); @@ -1661,7 +1662,9 @@ void usb_disconnect(struct usb_device **pdev) * so that the hardware is now fully quiesced. */ dev_dbg (&udev->dev, "unregistering device\n"); + mutex_lock(hcd->bandwidth_mutex); usb_disable_device(udev, 0); + mutex_unlock(hcd->bandwidth_mutex); usb_hcd_synchronize_unlinks(udev); usb_remove_ep_devs(&udev->ep0); diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index 5701e857392b..64c7ab4702df 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -1135,10 +1135,13 @@ void usb_disable_interface(struct usb_device *dev, struct usb_interface *intf, * Deallocates hcd/hardware state for the endpoints (nuking all or most * pending urbs) and usbcore state for the interfaces, so that usbcore * must usb_set_configuration() before any interfaces could be used. + * + * Must be called with hcd->bandwidth_mutex held. */ void usb_disable_device(struct usb_device *dev, int skip_ep0) { int i; + struct usb_hcd *hcd = bus_to_hcd(dev->bus); /* getting rid of interfaces will disconnect * any drivers bound to them (a key side effect) @@ -1172,6 +1175,16 @@ void usb_disable_device(struct usb_device *dev, int skip_ep0) dev_dbg(&dev->dev, "%s nuking %s URBs\n", __func__, skip_ep0 ? "non-ep0" : "all"); + if (hcd->driver->check_bandwidth) { + /* First pass: Cancel URBs, leave endpoint pointers intact. */ + for (i = skip_ep0; i < 16; ++i) { + usb_disable_endpoint(dev, i, false); + usb_disable_endpoint(dev, i + USB_DIR_IN, false); + } + /* Remove endpoints from the host controller internal state */ + usb_hcd_alloc_bandwidth(dev, NULL, NULL, NULL); + /* Second pass: remove endpoint pointers */ + } for (i = skip_ep0; i < 16; ++i) { usb_disable_endpoint(dev, i, true); usb_disable_endpoint(dev, i + USB_DIR_IN, true); @@ -1727,6 +1740,7 @@ free_interfaces: /* if it's already configured, clear out old state first. * getting rid of old interfaces means unbinding their drivers. */ + mutex_lock(hcd->bandwidth_mutex); if (dev->state != USB_STATE_ADDRESS) usb_disable_device(dev, 1); /* Skip ep0 */ @@ -1739,7 +1753,6 @@ free_interfaces: * host controller will not allow submissions to dropped endpoints. If * this call fails, the device state is unchanged. */ - mutex_lock(hcd->bandwidth_mutex); ret = usb_hcd_alloc_bandwidth(dev, cp, NULL, NULL); if (ret < 0) { mutex_unlock(hcd->bandwidth_mutex); From d23336329fa4c157ed6256d4279a73b87486a1b6 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Mon, 6 Jun 2011 00:53:47 -0700 Subject: [PATCH 057/224] xhci: Don't warn about zeroed bMaxBurst descriptor field. The USB 3.0 specification says that the bMaxBurst field in the SuperSpeed Endpoint Companion descriptor is supposed to indicate how many packets a SS device can handle before it needs to wait for an explicit handshake from the host controller. A zero value means the device can only handle one packet before it needs a handshake. Remove a warning in the xHCI driver that implies this is an invalid value. Signed-off-by: Sarah Sharp --- drivers/usb/host/xhci-mem.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 0f8e1d29a858..fcb7f7efc86d 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -1215,8 +1215,6 @@ int xhci_endpoint_init(struct xhci_hcd *xhci, ep_ctx->ep_info2 |= cpu_to_le32(MAX_PACKET(max_packet)); /* dig out max burst from ep companion desc */ max_packet = ep->ss_ep_comp.bMaxBurst; - if (!max_packet) - xhci_warn(xhci, "WARN no SS endpoint bMaxBurst\n"); ep_ctx->ep_info2 |= cpu_to_le32(MAX_BURST(max_packet)); break; case USB_SPEED_HIGH: From e1cf486d881d853d710e2d86a7adfc5fd260990f Mon Sep 17 00:00:00 2001 From: Alex He Date: Fri, 3 Jun 2011 15:58:25 +0800 Subject: [PATCH 058/224] xHCI 1.0: Force Stopped Event(FSE) FSE shall occur on the TD natural boundary. The software ep_ring dequeue pointer exceed the hardware ep_ring dequeue pointer in these cases of Table-3. As a result, the event_trb(pointed by hardware dequeue pointer) of the FSE can't be found in the current TD(pointed by software dequeue pointer). What should we do is to figured out the FSE case and skip over it. Signed-off-by: Alex He Signed-off-by: Sarah Sharp --- drivers/usb/host/xhci-ring.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 800f417c7309..0c008497edfd 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -2063,6 +2063,20 @@ static int handle_tx_event(struct xhci_hcd *xhci, /* Is this a TRB in the currently executing TD? */ event_seg = trb_in_td(ep_ring->deq_seg, ep_ring->dequeue, td->last_trb, event_dma); + + /* + * Skip the Force Stopped Event. The event_trb(event_dma) of FSE + * is not in the current TD pointed by ep_ring->dequeue because + * that the hardware dequeue pointer still at the previous TRB + * of the current TD. The previous TRB maybe a Link TD or the + * last TRB of the previous TD. The command completion handle + * will take care the rest. + */ + if (!event_seg && trb_comp_code == COMP_STOP_INVAL) { + ret = 0; + goto cleanup; + } + if (!event_seg) { if (!ep->skip || !usb_endpoint_xfer_isoc(&td->urb->ep->desc)) { From 578333ab95f70db13951d30a9ad6b565b61639a9 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 15 Jun 2011 16:32:46 -0400 Subject: [PATCH 059/224] USB: change maintainership of ohci-hcd and ehci-hcd Following the loss of David Brownell, I volunteer to maintain the ohci-hcd and ehci-hcd drivers. This patch (as1472) makes it official. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- MAINTAINERS | 6 ++++-- drivers/usb/host/ehci-hcd.c | 4 ++++ drivers/usb/host/ohci-hcd.c | 4 +++- 3 files changed, 11 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index e50fc6eb99e0..4307673415f7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6430,8 +6430,9 @@ S: Maintained F: drivers/usb/misc/rio500* USB EHCI DRIVER +M: Alan Stern L: linux-usb@vger.kernel.org -S: Orphan +S: Maintained F: Documentation/usb/ehci.txt F: drivers/usb/host/ehci* @@ -6490,8 +6491,9 @@ S: Maintained F: sound/usb/midi.* USB OHCI DRIVER +M: Alan Stern L: linux-usb@vger.kernel.org -S: Orphan +S: Maintained F: Documentation/usb/ohci.txt F: drivers/usb/host/ohci* diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c index b435ed67dd5c..f8030ee928e8 100644 --- a/drivers/usb/host/ehci-hcd.c +++ b/drivers/usb/host/ehci-hcd.c @@ -1,4 +1,8 @@ /* + * Enhanced Host Controller Interface (EHCI) driver for USB. + * + * Maintainer: Alan Stern + * * Copyright (c) 2000-2004 by David Brownell * * This program is free software; you can redistribute it and/or modify it diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c index 9aa10bdf3918..f9cf3f04b742 100644 --- a/drivers/usb/host/ohci-hcd.c +++ b/drivers/usb/host/ohci-hcd.c @@ -1,5 +1,7 @@ /* - * OHCI HCD (Host Controller Driver) for USB. + * Open Host Controller Interface (OHCI) driver for USB. + * + * Maintainer: Alan Stern * * (C) Copyright 1999 Roman Weissgaerber * (C) Copyright 2000-2004 David Brownell From cbb330045e5df8f665ac60227ff898421fc8fb92 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 15 Jun 2011 16:29:16 -0400 Subject: [PATCH 060/224] USB: don't let the hub driver prevent system sleep This patch (as1465) continues implementation of the policy that errors during suspend or hibernation should not prevent the system from going to sleep. In this case, failure to turn on the Suspend feature for a hub port shouldn't be reported as an error. There are situations where this does actually occur (such as when the device plugged into that port was disconnected in the recent past), and it turns out to be harmless. There's no reason for it to prevent a system sleep. Also, don't allow the hub driver to fail a system suspend if the downstream ports aren't all suspended. This is also harmless (and should never happen, given the change mentioned above); printing a warning message in the kernel log is all we really need to do. Signed-off-by: Alan Stern CC: Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hub.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 90ae1753dda1..c2ac08755f27 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -2362,6 +2362,10 @@ int usb_port_suspend(struct usb_device *udev, pm_message_t msg) USB_DEVICE_REMOTE_WAKEUP, 0, NULL, 0, USB_CTRL_SET_TIMEOUT); + + /* System sleep transitions should never fail */ + if (!(msg.event & PM_EVENT_AUTO)) + status = 0; } else { /* device has up to 10 msec to fully suspend */ dev_dbg(&udev->dev, "usb %ssuspend\n", @@ -2611,16 +2615,15 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg) struct usb_device *hdev = hub->hdev; unsigned port1; - /* fail if children aren't already suspended */ + /* Warn if children aren't already suspended */ for (port1 = 1; port1 <= hdev->maxchild; port1++) { struct usb_device *udev; udev = hdev->children [port1-1]; if (udev && udev->can_submit) { - if (!(msg.event & PM_EVENT_AUTO)) - dev_dbg(&intf->dev, "port %d nyet suspended\n", - port1); - return -EBUSY; + dev_warn(&intf->dev, "port %d nyet suspended\n", port1); + if (msg.event & PM_EVENT_AUTO) + return -EBUSY; } } From 0af212ba8f123c2eba151af7726c34a50b127962 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 15 Jun 2011 16:27:43 -0400 Subject: [PATCH 061/224] USB: don't let errors prevent system sleep This patch (as1464) implements the recommended policy that most errors during suspend or hibernation should not prevent the system from going to sleep. In particular, failure to suspend a USB driver or a USB device should not prevent the sleep from succeeding: Failure to suspend a device won't matter, because the device will automatically go into suspend mode when the USB bus stops carrying packets. (This might be less true for USB-3.0 devices, but let's not worry about them now.) Failure of a driver to suspend might lead to trouble later on when the system wakes up, but it isn't sufficient reason to prevent the system from going to sleep. Signed-off-by: Alan Stern CC: Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/driver.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index e35a17687c05..81add81ab040 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -1187,13 +1187,22 @@ static int usb_suspend_both(struct usb_device *udev, pm_message_t msg) for (i = n - 1; i >= 0; --i) { intf = udev->actconfig->interface[i]; status = usb_suspend_interface(udev, intf, msg); + + /* Ignore errors during system sleep transitions */ + if (!(msg.event & PM_EVENT_AUTO)) + status = 0; if (status != 0) break; } } - if (status == 0) + if (status == 0) { status = usb_suspend_device(udev, msg); + /* Again, ignore errors during system sleep transitions */ + if (!(msg.event & PM_EVENT_AUTO)) + status = 0; + } + /* If the suspend failed, resume interfaces that did get suspended */ if (status != 0) { msg.event ^= (PM_EVENT_SUSPEND | PM_EVENT_RESUME); From ee7b75fc4f3ae49e1f25bf56219bb5de3c29afaf Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 16 Jun 2011 13:15:41 -0400 Subject: [PATCH 062/224] NFSv4: Fix a readdir regression Commit 7ebb9315 (NFS: use secinfo when crossing mountpoints) introduces a regression when decoding an NFSv4 readdir entry that sets the rdattr_error field. By treating the resulting value as if it is a decoding error, the current code may cause us to skip valid readdir entries. Reported-by: Andy Adamson Cc: stable@kernel.org [2.6.39] Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index c4b7d6c04948..5db44a8dddf9 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -3098,7 +3098,7 @@ out_overflow: return -EIO; } -static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap) +static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap, int32_t *res) { __be32 *p; @@ -3109,7 +3109,7 @@ static int decode_attr_error(struct xdr_stream *xdr, uint32_t *bitmap) if (unlikely(!p)) goto out_overflow; bitmap[0] &= ~FATTR4_WORD0_RDATTR_ERROR; - return -be32_to_cpup(p); + *res = -be32_to_cpup(p); } return 0; out_overflow: @@ -4070,6 +4070,7 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, int status; umode_t fmode = 0; uint32_t type; + int32_t err; status = decode_attr_type(xdr, bitmap, &type); if (status < 0) @@ -4095,13 +4096,12 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap, goto xdr_error; fattr->valid |= status; - status = decode_attr_error(xdr, bitmap); - if (status == -NFS4ERR_WRONGSEC) { - nfs_fixup_secinfo_attributes(fattr, fh); - status = 0; - } + err = 0; + status = decode_attr_error(xdr, bitmap, &err); if (status < 0) goto xdr_error; + if (err == -NFS4ERR_WRONGSEC) + nfs_fixup_secinfo_attributes(fattr, fh); status = decode_attr_filehandle(xdr, bitmap, fh); if (status < 0) From 7263287af93db4d5cf324a30546f2143419b7900 Mon Sep 17 00:00:00 2001 From: Russ Gorby Date: Tue, 14 Jun 2011 13:23:28 -0700 Subject: [PATCH 063/224] tty: n_gsm: Fixed logic to decode break signal from modem status The modem status can be one or 2 octets and contains the V.24 signals and in the 2 octet case also the break signal. We were improperly decoding the break signal from the modem in the 2 octet case. Signed-off-by: Russ Gorby Signed-off-by: Greg Kroah-Hartman --- drivers/tty/n_gsm.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c index 09e8c7d53af3..7290394e3131 100644 --- a/drivers/tty/n_gsm.c +++ b/drivers/tty/n_gsm.c @@ -984,10 +984,22 @@ static void gsm_control_reply(struct gsm_mux *gsm, int cmd, u8 *data, */ static void gsm_process_modem(struct tty_struct *tty, struct gsm_dlci *dlci, - u32 modem) + u32 modem, int clen) { int mlines = 0; - u8 brk = modem >> 6; + u8 brk = 0; + + /* The modem status command can either contain one octet (v.24 signals) + or two octets (v.24 signals + break signals). The length field will + either be 2 or 3 respectively. This is specified in section + 5.4.6.3.7 of the 27.010 mux spec. */ + + if (clen == 2) + modem = modem & 0x7f; + else { + brk = modem & 0x7f; + modem = (modem >> 7) & 0x7f; + }; /* Flow control/ready to communicate */ if (modem & MDM_FC) { @@ -1061,7 +1073,7 @@ static void gsm_control_modem(struct gsm_mux *gsm, u8 *data, int clen) return; } tty = tty_port_tty_get(&dlci->port); - gsm_process_modem(tty, dlci, modem); + gsm_process_modem(tty, dlci, modem, clen); if (tty) { tty_wakeup(tty); tty_kref_put(tty); @@ -1482,12 +1494,13 @@ static void gsm_dlci_begin_close(struct gsm_dlci *dlci) * open we shovel the bits down it, if not we drop them. */ -static void gsm_dlci_data(struct gsm_dlci *dlci, u8 *data, int len) +static void gsm_dlci_data(struct gsm_dlci *dlci, u8 *data, int clen) { /* krefs .. */ struct tty_port *port = &dlci->port; struct tty_struct *tty = tty_port_tty_get(port); unsigned int modem = 0; + int len = clen; if (debug & 16) pr_debug("%d bytes for tty %p\n", len, tty); @@ -1507,7 +1520,7 @@ static void gsm_dlci_data(struct gsm_dlci *dlci, u8 *data, int len) if (len == 0) return; } - gsm_process_modem(tty, dlci, modem); + gsm_process_modem(tty, dlci, modem, clen); /* Line state will go via DLCI 0 controls only */ case 1: default: From 57f2104f39995bac332ddc492fbf60aa28e0c35e Mon Sep 17 00:00:00 2001 From: Russ Gorby Date: Tue, 14 Jun 2011 13:23:29 -0700 Subject: [PATCH 064/224] tty: n_gsm: improper skb_pull() use was leaking framed data gsm_dlci_data_output_framed() was doing: memcpy(dp, skb_pull(dlci->skb, len), len); The problem is skb_pull() returns the post-increment data ptr so the first chunk of dlci->skb->data is leaked. Signed-off-by: Russ Gorby Signed-off-by: Greg Kroah-Hartman --- drivers/tty/n_gsm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/tty/n_gsm.c b/drivers/tty/n_gsm.c index 7290394e3131..19b4ae052af8 100644 --- a/drivers/tty/n_gsm.c +++ b/drivers/tty/n_gsm.c @@ -875,7 +875,8 @@ static int gsm_dlci_data_output_framed(struct gsm_mux *gsm, *dp++ = last << 7 | first << 6 | 1; /* EA */ len--; } - memcpy(dp, skb_pull(dlci->skb, len), len); + memcpy(dp, dlci->skb->data, len); + skb_pull(dlci->skb, len); __gsm_data_queue(dlci, msg); if (last) dlci->skb = NULL; From c16d51a32bbb61ac8fd96f78b5ce2fccfe0fb4c3 Mon Sep 17 00:00:00 2001 From: Shreshtha Kumar Sahu Date: Mon, 13 Jun 2011 10:11:33 +0200 Subject: [PATCH 065/224] amba pl011: workaround for uart registers lockup This workaround aims to break the deadlock situation which raises during continuous transfer of data for long duration over uart with hardware flow control. It is observed that CTS interrupt cannot be cleared in uart interrupt register (ICR). Hence further transfer over uart gets blocked. It is seen that during such deadlock condition ICR don't get cleared even on multiple write. This leads pass_counter to decrease and finally reach zero. This can be taken as trigger point to run this UART_BT_WA. Workaround backups the register configuration, does soft reset of UART using BIT-0 of PRCC_K_SOFTRST_SET/CLEAR registers and restores the registers. This patch also provides support for uart init and exit function calls if present. Signed-off-by: Shreshtha Kumar Sahu Signed-off-by: Linus Walleij Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/amba-pl011.c | 123 +++++++++++++++++++++++++++++++- include/linux/amba/serial.h | 3 + 2 files changed, 125 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/amba-pl011.c b/drivers/tty/serial/amba-pl011.c index 8dc0541feecc..f5f6831b0a64 100644 --- a/drivers/tty/serial/amba-pl011.c +++ b/drivers/tty/serial/amba-pl011.c @@ -50,6 +50,7 @@ #include #include #include +#include #include #include @@ -65,6 +66,30 @@ #define UART_DR_ERROR (UART011_DR_OE|UART011_DR_BE|UART011_DR_PE|UART011_DR_FE) #define UART_DUMMY_DR_RX (1 << 16) + +#define UART_WA_SAVE_NR 14 + +static void pl011_lockup_wa(unsigned long data); +static const u32 uart_wa_reg[UART_WA_SAVE_NR] = { + ST_UART011_DMAWM, + ST_UART011_TIMEOUT, + ST_UART011_LCRH_RX, + UART011_IBRD, + UART011_FBRD, + ST_UART011_LCRH_TX, + UART011_IFLS, + ST_UART011_XFCR, + ST_UART011_XON1, + ST_UART011_XON2, + ST_UART011_XOFF1, + ST_UART011_XOFF2, + UART011_CR, + UART011_IMSC +}; + +static u32 uart_wa_regdata[UART_WA_SAVE_NR]; +static DECLARE_TASKLET(pl011_lockup_tlet, pl011_lockup_wa, 0); + /* There is by now at least one vendor with differing details, so handle it */ struct vendor_data { unsigned int ifls; @@ -72,6 +97,7 @@ struct vendor_data { unsigned int lcrh_tx; unsigned int lcrh_rx; bool oversampling; + bool interrupt_may_hang; /* vendor-specific */ bool dma_threshold; }; @@ -90,9 +116,12 @@ static struct vendor_data vendor_st = { .lcrh_tx = ST_UART011_LCRH_TX, .lcrh_rx = ST_UART011_LCRH_RX, .oversampling = true, + .interrupt_may_hang = true, .dma_threshold = true, }; +static struct uart_amba_port *amba_ports[UART_NR]; + /* Deals with DMA transactions */ struct pl011_sgbuf { @@ -132,6 +161,7 @@ struct uart_amba_port { unsigned int lcrh_rx; /* vendor-specific */ bool autorts; char type[12]; + bool interrupt_may_hang; /* vendor-specific */ #ifdef CONFIG_DMA_ENGINE /* DMA stuff */ bool using_tx_dma; @@ -1008,6 +1038,68 @@ static inline bool pl011_dma_rx_running(struct uart_amba_port *uap) #endif +/* + * pl011_lockup_wa + * This workaround aims to break the deadlock situation + * when after long transfer over uart in hardware flow + * control, uart interrupt registers cannot be cleared. + * Hence uart transfer gets blocked. + * + * It is seen that during such deadlock condition ICR + * don't get cleared even on multiple write. This leads + * pass_counter to decrease and finally reach zero. This + * can be taken as trigger point to run this UART_BT_WA. + * + */ +static void pl011_lockup_wa(unsigned long data) +{ + struct uart_amba_port *uap = amba_ports[0]; + void __iomem *base = uap->port.membase; + struct circ_buf *xmit = &uap->port.state->xmit; + struct tty_struct *tty = uap->port.state->port.tty; + int buf_empty_retries = 200; + int loop; + + /* Stop HCI layer from submitting data for tx */ + tty->hw_stopped = 1; + while (!uart_circ_empty(xmit)) { + if (buf_empty_retries-- == 0) + break; + udelay(100); + } + + /* Backup registers */ + for (loop = 0; loop < UART_WA_SAVE_NR; loop++) + uart_wa_regdata[loop] = readl(base + uart_wa_reg[loop]); + + /* Disable UART so that FIFO data is flushed out */ + writew(0x00, uap->port.membase + UART011_CR); + + /* Soft reset UART module */ + if (uap->port.dev->platform_data) { + struct amba_pl011_data *plat; + + plat = uap->port.dev->platform_data; + if (plat->reset) + plat->reset(); + } + + /* Restore registers */ + for (loop = 0; loop < UART_WA_SAVE_NR; loop++) + writew(uart_wa_regdata[loop] , + uap->port.membase + uart_wa_reg[loop]); + + /* Initialise the old status of the modem signals */ + uap->old_status = readw(uap->port.membase + UART01x_FR) & + UART01x_FR_MODEM_ANY; + + if (readl(base + UART011_MIS) & 0x2) + printk(KERN_EMERG "UART_BT_WA: ***FAILED***\n"); + + /* Start Tx/Rx */ + tty->hw_stopped = 0; +} + static void pl011_stop_tx(struct uart_port *port) { struct uart_amba_port *uap = (struct uart_amba_port *)port; @@ -1158,8 +1250,11 @@ static irqreturn_t pl011_int(int irq, void *dev_id) if (status & UART011_TXIS) pl011_tx_chars(uap); - if (pass_counter-- == 0) + if (pass_counter-- == 0) { + if (uap->interrupt_may_hang) + tasklet_schedule(&pl011_lockup_tlet); break; + } status = readw(uap->port.membase + UART011_MIS); } while (status != 0); @@ -1339,6 +1434,14 @@ static int pl011_startup(struct uart_port *port) writew(uap->im, uap->port.membase + UART011_IMSC); spin_unlock_irq(&uap->port.lock); + if (uap->port.dev->platform_data) { + struct amba_pl011_data *plat; + + plat = uap->port.dev->platform_data; + if (plat->init) + plat->init(); + } + return 0; clk_dis: @@ -1394,6 +1497,15 @@ static void pl011_shutdown(struct uart_port *port) * Shut down the clock producer */ clk_disable(uap->clk); + + if (uap->port.dev->platform_data) { + struct amba_pl011_data *plat; + + plat = uap->port.dev->platform_data; + if (plat->exit) + plat->exit(); + } + } static void @@ -1700,6 +1812,14 @@ static int __init pl011_console_setup(struct console *co, char *options) if (!uap) return -ENODEV; + if (uap->port.dev->platform_data) { + struct amba_pl011_data *plat; + + plat = uap->port.dev->platform_data; + if (plat->init) + plat->init(); + } + uap->port.uartclk = clk_get_rate(uap->clk); if (options) @@ -1774,6 +1894,7 @@ static int pl011_probe(struct amba_device *dev, const struct amba_id *id) uap->lcrh_rx = vendor->lcrh_rx; uap->lcrh_tx = vendor->lcrh_tx; uap->fifosize = vendor->fifosize; + uap->interrupt_may_hang = vendor->interrupt_may_hang; uap->port.dev = &dev->dev; uap->port.mapbase = dev->res.start; uap->port.membase = base; diff --git a/include/linux/amba/serial.h b/include/linux/amba/serial.h index 5479fdc849e9..514ed45c462e 100644 --- a/include/linux/amba/serial.h +++ b/include/linux/amba/serial.h @@ -201,6 +201,9 @@ struct amba_pl011_data { bool (*dma_filter)(struct dma_chan *chan, void *filter_param); void *dma_rx_param; void *dma_tx_param; + void (*init) (void); + void (*exit) (void); + void (*reset) (void); }; #endif From 1a7d4369b3fe1f8e5efe7f11a1c482055693852f Mon Sep 17 00:00:00 2001 From: Shreshtha Kumar Sahu Date: Mon, 13 Jun 2011 10:11:44 +0200 Subject: [PATCH 066/224] amba pl011: platform data for reg lockup and glitch v2 This patch provides platform data for following - uart reset function to assist uart register lockup workaround - init/exit function to fix glitch in the tx pin in tty_open when tty port0 is opened a glitch is seen in the tx line of uart0. This happens in pl011_startup() when tx fifo interrupt is provoked into asserting. Now uart0 pins are enabled (alt function) only when init is complete and turned back to gpio when closed. Signed-off-by: Shreshtha Kumar Sahu Signed-off-by: Linus Walleij Signed-off-by: Greg Kroah-Hartman --- arch/arm/mach-ux500/board-mop500-pins.c | 16 ++++++-- arch/arm/mach-ux500/board-mop500.c | 54 +++++++++++++++++++++++++ 2 files changed, 66 insertions(+), 4 deletions(-) diff --git a/arch/arm/mach-ux500/board-mop500-pins.c b/arch/arm/mach-ux500/board-mop500-pins.c index fd4cf1ca5efd..70cdbd60596a 100644 --- a/arch/arm/mach-ux500/board-mop500-pins.c +++ b/arch/arm/mach-ux500/board-mop500-pins.c @@ -110,10 +110,18 @@ static pin_cfg_t mop500_pins_common[] = { GPIO168_KP_O0, /* UART */ - GPIO0_U0_CTSn | PIN_INPUT_PULLUP, - GPIO1_U0_RTSn | PIN_OUTPUT_HIGH, - GPIO2_U0_RXD | PIN_INPUT_PULLUP, - GPIO3_U0_TXD | PIN_OUTPUT_HIGH, + /* uart-0 pins gpio configuration should be + * kept intact to prevent glitch in tx line + * when tty dev is opened. Later these pins + * are configured to uart mop500_pins_uart0 + * + * It will be replaced with uart configuration + * once the issue is solved. + */ + GPIO0_GPIO | PIN_INPUT_PULLUP, + GPIO1_GPIO | PIN_OUTPUT_HIGH, + GPIO2_GPIO | PIN_INPUT_PULLUP, + GPIO3_GPIO | PIN_OUTPUT_HIGH, GPIO29_U2_RXD | PIN_INPUT_PULLUP, GPIO30_U2_TXD | PIN_OUTPUT_HIGH, diff --git a/arch/arm/mach-ux500/board-mop500.c b/arch/arm/mach-ux500/board-mop500.c index bb26f40493e6..2a08c07dec6d 100644 --- a/arch/arm/mach-ux500/board-mop500.c +++ b/arch/arm/mach-ux500/board-mop500.c @@ -27,18 +27,21 @@ #include #include #include +#include #include #include #include #include +#include #include #include #include #include +#include "pins-db8500.h" #include "ste-dma40-db8500.h" #include "devices-db8500.h" #include "board-mop500.h" @@ -393,12 +396,63 @@ static struct stedma40_chan_cfg uart2_dma_cfg_tx = { }; #endif + +static pin_cfg_t mop500_pins_uart0[] = { + GPIO0_U0_CTSn | PIN_INPUT_PULLUP, + GPIO1_U0_RTSn | PIN_OUTPUT_HIGH, + GPIO2_U0_RXD | PIN_INPUT_PULLUP, + GPIO3_U0_TXD | PIN_OUTPUT_HIGH, +}; + +#define PRCC_K_SOFTRST_SET 0x18 +#define PRCC_K_SOFTRST_CLEAR 0x1C +static void ux500_uart0_reset(void) +{ + void __iomem *prcc_rst_set, *prcc_rst_clr; + + prcc_rst_set = (void __iomem *)IO_ADDRESS(U8500_CLKRST1_BASE + + PRCC_K_SOFTRST_SET); + prcc_rst_clr = (void __iomem *)IO_ADDRESS(U8500_CLKRST1_BASE + + PRCC_K_SOFTRST_CLEAR); + + /* Activate soft reset PRCC_K_SOFTRST_CLEAR */ + writel((readl(prcc_rst_clr) | 0x1), prcc_rst_clr); + udelay(1); + + /* Release soft reset PRCC_K_SOFTRST_SET */ + writel((readl(prcc_rst_set) | 0x1), prcc_rst_set); + udelay(1); +} + +static void ux500_uart0_init(void) +{ + int ret; + + ret = nmk_config_pins(mop500_pins_uart0, + ARRAY_SIZE(mop500_pins_uart0)); + if (ret < 0) + pr_err("pl011: uart pins_enable failed\n"); +} + +static void ux500_uart0_exit(void) +{ + int ret; + + ret = nmk_config_pins_sleep(mop500_pins_uart0, + ARRAY_SIZE(mop500_pins_uart0)); + if (ret < 0) + pr_err("pl011: uart pins_disable failed\n"); +} + static struct amba_pl011_data uart0_plat = { #ifdef CONFIG_STE_DMA40 .dma_filter = stedma40_filter, .dma_rx_param = &uart0_dma_cfg_rx, .dma_tx_param = &uart0_dma_cfg_tx, #endif + .init = ux500_uart0_init, + .exit = ux500_uart0_exit, + .reset = ux500_uart0_reset, }; static struct amba_pl011_data uart1_plat = { From 3bc46b312b1486b1fe2db4246a34a30160d26d8d Mon Sep 17 00:00:00 2001 From: Maxime Bizon Date: Fri, 10 Jun 2011 23:17:58 +0200 Subject: [PATCH 067/224] serial: bcm63xx_uart: fix irq storm after rx fifo overrun. RX fifo reset is required to clear irq. Signed-off-by: Maxime Bizon Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/bcm63xx_uart.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/drivers/tty/serial/bcm63xx_uart.c b/drivers/tty/serial/bcm63xx_uart.c index a1a0e55d0807..c0b68b9cad91 100644 --- a/drivers/tty/serial/bcm63xx_uart.c +++ b/drivers/tty/serial/bcm63xx_uart.c @@ -250,6 +250,20 @@ static void bcm_uart_do_rx(struct uart_port *port) /* get overrun/fifo empty information from ier * register */ iestat = bcm_uart_readl(port, UART_IR_REG); + + if (unlikely(iestat & UART_IR_STAT(UART_IR_RXOVER))) { + unsigned int val; + + /* fifo reset is required to clear + * interrupt */ + val = bcm_uart_readl(port, UART_CTL_REG); + val |= UART_CTL_RSTRXFIFO_MASK; + bcm_uart_writel(port, val, UART_CTL_REG); + + port->icount.overrun++; + tty_insert_flip_char(tty, 0, TTY_OVERRUN); + } + if (!(iestat & UART_IR_STAT(UART_IR_RXNOTEMPTY))) break; @@ -284,10 +298,6 @@ static void bcm_uart_do_rx(struct uart_port *port) if (uart_handle_sysrq_char(port, c)) continue; - if (unlikely(iestat & UART_IR_STAT(UART_IR_RXOVER))) { - port->icount.overrun++; - tty_insert_flip_char(tty, 0, TTY_OVERRUN); - } if ((cstat & port->ignore_status_mask) == 0) tty_insert_flip_char(tty, c, flag); From 2bc58a6fd76f89052c7f151d78fb2d8b804aacfe Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Mon, 13 Jun 2011 06:46:44 +0100 Subject: [PATCH 068/224] ARM: 6959/1: SMP build fix for entry-macro-multi.S The assembly code in entry-macro-multi.S does not build without the include asm/assembler.h in the case of CONFIG_SMP=y. Fixes the rather theoretical SMP build of mach-shmobile/entry-intc.c: arch/arm/include/asm/entry-macro-multi.S: Assembler messages: arch/arm/include/asm/entry-macro-multi.S:20: Error: bad instruction `alt_smp(test_for_ipi r0,r6,r5,lr)' arch/arm/include/asm/entry-macro-multi.S:20: Error: bad instruction `alt_up_b(9997f)' make[1]: *** [arch/arm/mach-shmobile/entry-intc.o] Error 1 make: *** [arch/arm/mach-shmobile] Error 2 make: *** Waiting for unfinished jobs.... Signed-off-by: Magnus Damm Signed-off-by: Russell King --- arch/arm/include/asm/assembler.h | 4 ++++ arch/arm/include/asm/entry-macro-multi.S | 2 ++ 2 files changed, 6 insertions(+) diff --git a/arch/arm/include/asm/assembler.h b/arch/arm/include/asm/assembler.h index bc2d2d75f706..65c3f2474f5e 100644 --- a/arch/arm/include/asm/assembler.h +++ b/arch/arm/include/asm/assembler.h @@ -13,6 +13,9 @@ * Do not include any C declarations in this file - it is included by * assembler source. */ +#ifndef __ASM_ASSEMBLER_H__ +#define __ASM_ASSEMBLER_H__ + #ifndef __ASSEMBLY__ #error "Only include this from assembly code" #endif @@ -290,3 +293,4 @@ .macro ldrusr, reg, ptr, inc, cond=al, rept=1, abort=9001f usracc ldr, \reg, \ptr, \inc, \cond, \rept, \abort .endm +#endif /* __ASM_ASSEMBLER_H__ */ diff --git a/arch/arm/include/asm/entry-macro-multi.S b/arch/arm/include/asm/entry-macro-multi.S index ec0bbf79c71f..2da8547de6d6 100644 --- a/arch/arm/include/asm/entry-macro-multi.S +++ b/arch/arm/include/asm/entry-macro-multi.S @@ -1,3 +1,5 @@ +#include + /* * Interrupt handling. Preserves r7, r8, r9 */ From 343fda59823ca20f48578f816ec12e741214f509 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 16 Jun 2011 08:26:06 +0100 Subject: [PATCH 069/224] ARM: 6962/1: mach-h720x: fix build error The h7201/h7202 machines did not build since they define ARM_DMA_ZONE_OFFSET but do not select ZONE_DMA. Fix it up by selecting ZONE_DMA in their Kconfig. Cc: Sascha Hauer Signed-off-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/mach-h720x/Kconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/mach-h720x/Kconfig b/arch/arm/mach-h720x/Kconfig index 9b6982efbd22..abf356c02343 100644 --- a/arch/arm/mach-h720x/Kconfig +++ b/arch/arm/mach-h720x/Kconfig @@ -6,12 +6,14 @@ config ARCH_H7201 bool "gms30c7201" depends on ARCH_H720X select CPU_H7201 + select ZONE_DMA help Say Y here if you are using the Hynix GMS30C7201 Reference Board config ARCH_H7202 bool "hms30c7202" select CPU_H7202 + select ZONE_DMA depends on ARCH_H720X help Say Y here if you are using the Hynix HMS30C7202 Reference Board From 9a00318eadbb43db4e9c163c262a22a3c8b5a672 Mon Sep 17 00:00:00 2001 From: Dave Martin Date: Thu, 16 Jun 2011 12:09:37 +0100 Subject: [PATCH 070/224] ARM: 6963/1: Thumb-2: Relax relocation requirements for non-function symbols The "Thumb bit" of a symbol is only really meaningful for function symbols (STT_FUNC). However, sometimes a branch is relocated against a non-function symbol; for example, PC-relative branches to anonymous assembler local symbols are typically fixed up against the start-of-section symbol, which is not a function symbol. Some inline assembler generates references of this type, such as fixup code generated by macros in . The existing relocation code for R_ARM_THM_CALL/R_ARM_THM_JUMP24 interprets this case as an error, because the target symbol appears to be an ARM symbol; but this is really not the case, since the target symbol is just a base in these cases. The addend defines the precise offset to the target location, but since the addend is encoded in a non-interworking Thumb branch instruction, there is no explicit Thumb bit in the addend. Because these instructions never interwork, the implied Thumb bit in the addend is 1, and the destination is Thumb by definition. This patch removes the extraneous Thumb bit check for non-function symbols, enabling modules containing the affected relocation types to be loaded. No modification to the actual relocation code is required, since this code does not take bit[0] of the location->destination offset into account in any case. Function symbols are always checked for interworking conflicts, as before. Signed-off-by: Dave Martin Acked-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/kernel/module.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c index fee7c36349eb..016d6a0830a3 100644 --- a/arch/arm/kernel/module.c +++ b/arch/arm/kernel/module.c @@ -193,8 +193,17 @@ apply_relocate(Elf32_Shdr *sechdrs, const char *strtab, unsigned int symindex, offset -= 0x02000000; offset += sym->st_value - loc; - /* only Thumb addresses allowed (no interworking) */ - if (!(offset & 1) || + /* + * For function symbols, only Thumb addresses are + * allowed (no interworking). + * + * For non-function symbols, the destination + * has no specific ARM/Thumb disposition, so + * the branch is resolved under the assumption + * that interworking is not required. + */ + if ((ELF32_ST_TYPE(sym->st_info) == STT_FUNC && + !(offset & 1)) || offset <= (s32)0xff000000 || offset >= (s32)0x01000000) { pr_err("%s: section %u reloc %u sym '%s': relocation %u out of range (%#lx -> %#x)\n", From 5afa9133cfe67f1bfead6049a9640c9262a7101c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 17 Jun 2011 10:14:59 -0400 Subject: [PATCH 071/224] SUNRPC: Ensure the RPC client only quits on fatal signals Fix a couple of instances where we were exiting the RPC client on arbitrary signals. We should only do so on fatal signals. Cc: stable@kernel.org Signed-off-by: Trond Myklebust --- net/sunrpc/auth_gss/auth_gss.c | 4 ++-- net/sunrpc/clnt.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index 339ba64cce1e..5daf6cc4faea 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -577,13 +577,13 @@ retry: } inode = &gss_msg->inode->vfs_inode; for (;;) { - prepare_to_wait(&gss_msg->waitqueue, &wait, TASK_INTERRUPTIBLE); + prepare_to_wait(&gss_msg->waitqueue, &wait, TASK_KILLABLE); spin_lock(&inode->i_lock); if (gss_msg->ctx != NULL || gss_msg->msg.errno < 0) { break; } spin_unlock(&inode->i_lock); - if (signalled()) { + if (fatal_signal_pending(current)) { err = -ERESTARTSYS; goto out_intr; } diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 566bcfd067f6..8c9141583d6f 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -1061,7 +1061,7 @@ call_allocate(struct rpc_task *task) dprintk("RPC: %5u rpc_buffer allocation failed\n", task->tk_pid); - if (RPC_IS_ASYNC(task) || !signalled()) { + if (RPC_IS_ASYNC(task) || !fatal_signal_pending(current)) { task->tk_action = call_allocate; rpc_delay(task, HZ>>4); return; From f6ba6fe2d913da6707a71a413d6ec8ae98d6ce18 Mon Sep 17 00:00:00 2001 From: Alex He Date: Wed, 8 Jun 2011 18:34:06 +0800 Subject: [PATCH 072/224] xHCI 1.0: Incompatible Device Error It is one new TRB Completion Code for the xHCI spec v1.0. Asserted if the xHC detects a problem with a device that does not allow it to be successfully accessed, e.g. due to a device compliance or compatibility problem. This error may be returned by any command or transfer, and is fatal as far as the Slot is concerned. Return -EPROTO by urb->status or frame->status of ISOC for transfer case. And return -ENODEV for configure endpoint command, evaluate context command and address device command if there is an incompatible Device Error. The error codes will be sent back to the USB core to decide how to do. It's unnecessary for other commands because after the three commands run successfully means that the device has been accepted. Signed-off-by: Alex He Signed-off-by: Sarah Sharp --- drivers/usb/host/xhci-ring.c | 5 +++++ drivers/usb/host/xhci.c | 15 +++++++++++++++ drivers/usb/host/xhci.h | 2 ++ 3 files changed, 22 insertions(+) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 0c008497edfd..436332aa341b 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -1733,6 +1733,7 @@ static int process_isoc_td(struct xhci_hcd *xhci, struct xhci_td *td, frame->status = -EOVERFLOW; skip_td = true; break; + case COMP_DEV_ERR: case COMP_STALL: frame->status = -EPROTO; skip_td = true; @@ -2016,6 +2017,10 @@ static int handle_tx_event(struct xhci_hcd *xhci, TRB_TO_SLOT_ID(le32_to_cpu(event->flags)), ep_index); goto cleanup; + case COMP_DEV_ERR: + xhci_warn(xhci, "WARN: detect an incompatible device"); + status = -EPROTO; + break; case COMP_MISSED_INT: /* * When encounter missed service error, one or more isoc tds diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index e5a01713f937..15eb4c3d793c 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -1551,6 +1551,11 @@ static int xhci_configure_endpoint_result(struct xhci_hcd *xhci, "and endpoint is not disabled.\n"); ret = -EINVAL; break; + case COMP_DEV_ERR: + dev_warn(&udev->dev, "ERROR: Incompatible device for endpoint " + "configure command.\n"); + ret = -ENODEV; + break; case COMP_SUCCESS: dev_dbg(&udev->dev, "Successful Endpoint Configure command\n"); ret = 0; @@ -1585,6 +1590,11 @@ static int xhci_evaluate_context_result(struct xhci_hcd *xhci, xhci_dbg_ctx(xhci, virt_dev->out_ctx, 1); ret = -EINVAL; break; + case COMP_DEV_ERR: + dev_warn(&udev->dev, "ERROR: Incompatible device for evaluate " + "context command.\n"); + ret = -ENODEV; + break; case COMP_MEL_ERR: /* Max Exit Latency too large error */ dev_warn(&udev->dev, "WARN: Max Exit Latency too large\n"); @@ -2867,6 +2877,11 @@ int xhci_address_device(struct usb_hcd *hcd, struct usb_device *udev) dev_warn(&udev->dev, "Device not responding to set address.\n"); ret = -EPROTO; break; + case COMP_DEV_ERR: + dev_warn(&udev->dev, "ERROR: Incompatible device for address " + "device command.\n"); + ret = -ENODEV; + break; case COMP_SUCCESS: xhci_dbg(xhci, "Successful Address Device command\n"); break; diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index 7d1ea3bf5e1f..ba90af1c34b4 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -874,6 +874,8 @@ struct xhci_transfer_event { #define COMP_PING_ERR 20 /* Event Ring is full */ #define COMP_ER_FULL 21 +/* Incompatible Device Error */ +#define COMP_DEV_ERR 22 /* Missed Service Error - HC couldn't service an isoc ep within interval */ #define COMP_MISSED_INT 23 /* Successfully stopped command ring */ From c877b3b2ad5cb9d4fe523c5496185cc328ff3ae9 Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Wed, 15 Jun 2011 23:47:21 +0200 Subject: [PATCH 073/224] xhci: Add reset on resume quirk for asrock p67 host The asrock p67 xhci controller completely dies on resume, add a quirk for this, to bring the host back online after a suspend. This should be backported to stable kernels as old as 2.6.37. Signed-off-by: Maarten Lankhorst Signed-off-by: Sarah Sharp Cc: stable@kernel.org --- drivers/usb/host/xhci-pci.c | 8 ++++++++ drivers/usb/host/xhci.c | 2 ++ drivers/usb/host/xhci.h | 1 + 3 files changed, 11 insertions(+) diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c index 17541d09eabb..cb16de213f64 100644 --- a/drivers/usb/host/xhci-pci.c +++ b/drivers/usb/host/xhci-pci.c @@ -29,6 +29,9 @@ #define PCI_VENDOR_ID_FRESCO_LOGIC 0x1b73 #define PCI_DEVICE_ID_FRESCO_LOGIC_PDK 0x1000 +#define PCI_VENDOR_ID_ETRON 0x1b6f +#define PCI_DEVICE_ID_ASROCK_P67 0x7023 + static const char hcd_name[] = "xhci_hcd"; /* called after powerup, by probe or system-pm "wakeup" */ @@ -134,6 +137,11 @@ static int xhci_pci_setup(struct usb_hcd *hcd) xhci->quirks |= XHCI_EP_LIMIT_QUIRK; xhci->limit_active_eps = 64; } + if (pdev->vendor == PCI_VENDOR_ID_ETRON && + pdev->device == PCI_DEVICE_ID_ASROCK_P67) { + xhci->quirks |= XHCI_RESET_ON_RESUME; + xhci_dbg(xhci, "QUIRK: Resetting on resume\n"); + } /* Make sure the HC is halted. */ retval = xhci_halt(xhci); diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 15eb4c3d793c..f5fe1ac301ab 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -759,6 +759,8 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) msleep(100); spin_lock_irq(&xhci->lock); + if (xhci->quirks & XHCI_RESET_ON_RESUME) + hibernated = true; if (!hibernated) { /* step 1: restore register */ diff --git a/drivers/usb/host/xhci.h b/drivers/usb/host/xhci.h index ba90af1c34b4..d8bbf5ccb10d 100644 --- a/drivers/usb/host/xhci.h +++ b/drivers/usb/host/xhci.h @@ -1310,6 +1310,7 @@ struct xhci_hcd { */ #define XHCI_EP_LIMIT_QUIRK (1 << 5) #define XHCI_BROKEN_MSI (1 << 6) +#define XHCI_RESET_ON_RESUME (1 << 7) unsigned int num_active_eps; unsigned int limit_active_eps; /* There are two roothubs to keep track of bus suspend info for */ From b3df3f9c7df9a8d85e03e158d35487618a160901 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Wed, 15 Jun 2011 19:57:46 -0700 Subject: [PATCH 074/224] xhci: Always set urb->status to zero for isoc endpoints. When the xHCI driver encounters a Missed Service Interval event for an isochronous endpoint ring, it means the host controller skipped over one or more isochronous TDs. For TD that is skipped, skip_isoc_td() is called. This sets the frame descriptor status to -EXDEV, and also sets the value stored in the int pointed to by status to -EXDEV. If the isochronous TD happens to be the last TD in an URB, handle_tx_event() will use the status variable to give back the URB to the USB core. That means drivers will see urb->status as -EXDEV. It turns out that EHCI, UHCI, and OHCI always set urb->status to zero for an isochronous urb, regardless of what the frame status is. See itd_complete() in ehci-sched.c: } else { /* URB was too late */ desc->status = -EXDEV; } } /* handle completion now? */ if (likely ((urb_index + 1) != urb->number_of_packets)) goto done; /* ASSERT: it's really the last itd for this urb list_for_each_entry (itd, &stream->td_list, itd_list) BUG_ON (itd->urb == urb); */ /* give urb back to the driver; completion often (re)submits */ dev = urb->dev; ehci_urb_done(ehci, urb, 0); ehci_urb_done() completes the URB with the status of the third argument, which is always zero in this case. It turns out that many USB webcam drivers, such as uvcvideo, cannot handle urb->status set to a non-zero value. They will not resubmit their isochronous URBs in that case, and userspace will see a frozen video. Change the xHCI driver to be consistent with the EHCI and UHCI driver, and always set urb->status to 0 for isochronous URBs. This patch should be backported to kernels as old as 2.6.36 Signed-off-by: Sarah Sharp Cc: "Xu, Andiry" Cc: stable@kernel.org --- drivers/usb/host/xhci-ring.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 436332aa341b..70cacbbe7fb9 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -1768,9 +1768,6 @@ static int process_isoc_td(struct xhci_hcd *xhci, struct xhci_td *td, } } - if ((idx == urb_priv->length - 1) && *status == -EINPROGRESS) - *status = 0; - return finish_td(xhci, td, event_trb, event, ep, status, false); } @@ -1788,8 +1785,7 @@ static int skip_isoc_td(struct xhci_hcd *xhci, struct xhci_td *td, idx = urb_priv->td_cnt; frame = &td->urb->iso_frame_desc[idx]; - /* The transfer is partly done */ - *status = -EXDEV; + /* The transfer is partly done. */ frame->status = -EXDEV; /* calc actual length */ @@ -2177,6 +2173,11 @@ cleanup: urb->transfer_buffer_length, status); spin_unlock(&xhci->lock); + /* EHCI, UHCI, and OHCI always unconditionally set the + * urb->status of an isochronous endpoint to 0. + */ + if (usb_pipetype(urb->pipe) == PIPE_ISOCHRONOUS) + status = 0; usb_hcd_giveback_urb(bus_to_hcd(urb->dev->bus), urb, status); spin_lock(&xhci->lock); } From a9e758634f464ffb09344821a9f0b5a5c6df2b3e Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Thu, 16 Jun 2011 13:06:04 -0700 Subject: [PATCH 075/224] USB: Fix up URB error codes to reflect implementation. Documentation/usb/error-codes.txt mentions that urb->status can be set to -EXDEV, if the isochronous transfer was not fully completed. However, in practice, EHCI, UHCI, and OHCI all only set -EXDEV in the individual frame status, never in the URB status. Those host controller actually always pass in a zero status to usb_hcd_giveback_urb, and rely on the core to set the appropriate status value. The xHCI driver ran into issues with the uvcvideo driver when it tried to set -EXDEV in urb->status, because the driver refused to submit URBs, and the userspace camera application's video froze. Clean up the documentation to reflect the actual implementation. Signed-off-by: Sarah Sharp Acked-by: Alan Stern --- Documentation/usb/error-codes.txt | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/Documentation/usb/error-codes.txt b/Documentation/usb/error-codes.txt index d83703ea74b2..b3f606b81a03 100644 --- a/Documentation/usb/error-codes.txt +++ b/Documentation/usb/error-codes.txt @@ -76,6 +76,13 @@ A transfer's actual_length may be positive even when an error has been reported. That's because transfers often involve several packets, so that one or more packets could finish before an error stops further endpoint I/O. +For isochronous URBs, the urb status value is non-zero only if the URB is +unlinked, the device is removed, the host controller is disabled, or the total +transferred length is less than the requested length and the URB_SHORT_NOT_OK +flag is set. Completion handlers for isochronous URBs should only see +urb->status set to zero, -ENOENT, -ECONNRESET, -ESHUTDOWN, or -EREMOTEIO. +Individual frame descriptor status fields may report more status codes. + 0 Transfer completed successfully @@ -132,7 +139,7 @@ one or more packets could finish before an error stops further endpoint I/O. device removal events immediately. -EXDEV ISO transfer only partially completed - look at individual frame status for details + (only set in iso_frame_desc[n].status, not urb->status) -EINVAL ISO madness, if this happens: Log off and go home From 2ff7d09a1b0f20f2d9c1bde0e003d4e384de2313 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Wed, 1 Jun 2011 17:49:14 +0000 Subject: [PATCH 076/224] RDMA/cxgb4: Don't exceed hw IQ depth limit for user CQs Memory allocated for user CQs gets rounded up to the next page boundary. And after rounding, we recalculate the resulting IQ depth and we need to make sure we don't exceed the HW limits. This bug can result a much smaller CQ allocated than was expected if the HW size field is exceeded, resulting in CQ overflow failures. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier --- drivers/infiniband/hw/cxgb4/cq.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 8d8f8add6fcd..1720dc790d13 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -801,6 +801,10 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, int entries, if (ucontext) { memsize = roundup(memsize, PAGE_SIZE); hwentries = memsize / sizeof *chp->cq.queue; + while (hwentries > T4_MAX_IQ_SIZE) { + memsize -= PAGE_SIZE; + hwentries = memsize / sizeof *chp->cq.queue; + } } chp->cq.size = hwentries; chp->cq.memsize = memsize; From 301c2c3f039a1f9478f6cbef60f2ccd4da9bd4a1 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Tue, 14 Jun 2011 20:59:21 +0000 Subject: [PATCH 077/224] RDMA/cxgb4: Don't truncate MR lengths Remove left-over code from T3 that limited MR sizes to 32b. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier --- drivers/infiniband/hw/cxgb4/mem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index 273ffe49525a..0347eed4a167 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -625,7 +625,7 @@ pbl_done: mhp->attr.perms = c4iw_ib_to_tpt_access(acc); mhp->attr.va_fbo = virt; mhp->attr.page_size = shift - 12; - mhp->attr.len = (u32) length; + mhp->attr.len = length; err = register_mem(rhp, php, mhp, shift); if (err) From 8da7e7a55231543b84ac84e93ad5ca9d340773d7 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Tue, 14 Jun 2011 20:59:27 +0000 Subject: [PATCH 078/224] RDMA/cxgb4: Couple of abort fixes - fix a race where the driver could end up sending a close_con_req after an abort_rpl. In c4iw_ep_disconnect(), send abort or close request with the ep mutex held. - fix a hang where driver fails to wake up when a connection is reset during a normal close. Wake up any waiters in the interrupt path, and correctly cleanup after rdma_fini() failures. Signed-off-by: Steve Wise Signed-off-by: Roland Dreier --- drivers/infiniband/hw/cxgb4/cm.c | 46 +++++++++++++++++++++++++------- drivers/infiniband/hw/cxgb4/qp.c | 5 +--- 2 files changed, 38 insertions(+), 13 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index f660cd04ec2f..31fb44085c9b 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -1463,9 +1463,9 @@ static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb) struct c4iw_qp_attributes attrs; int disconnect = 1; int release = 0; - int abort = 0; struct tid_info *t = dev->rdev.lldi.tids; unsigned int tid = GET_TID(hdr); + int ret; ep = lookup_tid(t, tid); PDBG("%s ep %p tid %u\n", __func__, ep, ep->hwtid); @@ -1501,10 +1501,12 @@ static int peer_close(struct c4iw_dev *dev, struct sk_buff *skb) start_ep_timer(ep); __state_set(&ep->com, CLOSING); attrs.next_state = C4IW_QP_STATE_CLOSING; - abort = c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp, + ret = c4iw_modify_qp(ep->com.qp->rhp, ep->com.qp, C4IW_QP_ATTR_NEXT_STATE, &attrs, 1); - peer_close_upcall(ep); - disconnect = 1; + if (ret != -ECONNRESET) { + peer_close_upcall(ep); + disconnect = 1; + } break; case ABORTING: disconnect = 0; @@ -2109,15 +2111,16 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp) break; } - mutex_unlock(&ep->com.mutex); if (close) { - if (abrupt) - ret = abort_connection(ep, NULL, gfp); - else + if (abrupt) { + close_complete_upcall(ep); + ret = send_abort(ep, NULL, gfp); + } else ret = send_halfclose(ep, gfp); if (ret) fatal = 1; } + mutex_unlock(&ep->com.mutex); if (fatal) release_ep_resources(ep); return ret; @@ -2301,6 +2304,31 @@ static int fw6_msg(struct c4iw_dev *dev, struct sk_buff *skb) return 0; } +static int peer_abort_intr(struct c4iw_dev *dev, struct sk_buff *skb) +{ + struct cpl_abort_req_rss *req = cplhdr(skb); + struct c4iw_ep *ep; + struct tid_info *t = dev->rdev.lldi.tids; + unsigned int tid = GET_TID(req); + + ep = lookup_tid(t, tid); + if (is_neg_adv_abort(req->status)) { + PDBG("%s neg_adv_abort ep %p tid %u\n", __func__, ep, + ep->hwtid); + kfree_skb(skb); + return 0; + } + PDBG("%s ep %p tid %u state %u\n", __func__, ep, ep->hwtid, + ep->com.state); + + /* + * Wake up any threads in rdma_init() or rdma_fini(). + */ + c4iw_wake_up(&ep->com.wr_wait, -ECONNRESET); + sched(dev, skb); + return 0; +} + /* * Most upcalls from the T4 Core go to sched() to * schedule the processing on a work queue. @@ -2317,7 +2345,7 @@ c4iw_handler_func c4iw_handlers[NUM_CPL_CMDS] = { [CPL_PASS_ESTABLISH] = sched, [CPL_PEER_CLOSE] = sched, [CPL_CLOSE_CON_RPL] = sched, - [CPL_ABORT_REQ_RSS] = sched, + [CPL_ABORT_REQ_RSS] = peer_abort_intr, [CPL_RDMA_TERMINATE] = sched, [CPL_FW4_ACK] = sched, [CPL_SET_TCB_RPL] = set_tcb_rpl, diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index 3b773b05a898..a41578e48c7b 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -1207,11 +1207,8 @@ int c4iw_modify_qp(struct c4iw_dev *rhp, struct c4iw_qp *qhp, c4iw_get_ep(&qhp->ep->com); } ret = rdma_fini(rhp, qhp, ep); - if (ret) { - if (internal) - c4iw_get_ep(&qhp->ep->com); + if (ret) goto err; - } break; case C4IW_QP_STATE_TERMINATE: set_state(qhp, C4IW_QP_STATE_TERMINATE); From 3126448451105fae59de0058c68692aa09aa4c37 Mon Sep 17 00:00:00 2001 From: Mitko Haralanov Date: Thu, 9 Jun 2011 20:27:26 +0000 Subject: [PATCH 079/224] IB/qib: Ensure that LOS and DFE are being turned off Due to timing, it is possible for the LOS and DFE to remain on. This is due to the link progressing to LinkUP prior to the driver getting the first Status Changed interrupt. By expanding the conditions under which LOS is turned off and DFE timeout is being set, timing is no longer an issue. Signed-off-by: Mitko Haralanov Signed-off-by: Mike Marciniszyn Signed-off-by: Roland Dreier --- drivers/infiniband/hw/qib/qib_iba7322.c | 25 ++++++++++++++++++------- drivers/infiniband/hw/qib/qib_intr.c | 6 +++++- 2 files changed, 23 insertions(+), 8 deletions(-) diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index 9f53e68a096a..8ec5237031a0 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -469,6 +469,8 @@ static u8 ib_rate_to_delay[IB_RATE_120_GBPS + 1] = { #define IB_7322_LT_STATE_RECOVERIDLE 0x0f #define IB_7322_LT_STATE_CFGENH 0x10 #define IB_7322_LT_STATE_CFGTEST 0x11 +#define IB_7322_LT_STATE_CFGWAITRMTTEST 0x12 +#define IB_7322_LT_STATE_CFGWAITENH 0x13 /* link state machine states from IBC */ #define IB_7322_L_STATE_DOWN 0x0 @@ -498,8 +500,10 @@ static const u8 qib_7322_physportstate[0x20] = { IB_PHYSPORTSTATE_LINK_ERR_RECOVER, [IB_7322_LT_STATE_CFGENH] = IB_PHYSPORTSTATE_CFG_ENH, [IB_7322_LT_STATE_CFGTEST] = IB_PHYSPORTSTATE_CFG_TRAIN, - [0x12] = IB_PHYSPORTSTATE_CFG_TRAIN, - [0x13] = IB_PHYSPORTSTATE_CFG_WAIT_ENH, + [IB_7322_LT_STATE_CFGWAITRMTTEST] = + IB_PHYSPORTSTATE_CFG_TRAIN, + [IB_7322_LT_STATE_CFGWAITENH] = + IB_PHYSPORTSTATE_CFG_WAIT_ENH, [0x14] = IB_PHYSPORTSTATE_CFG_TRAIN, [0x15] = IB_PHYSPORTSTATE_CFG_TRAIN, [0x16] = IB_PHYSPORTSTATE_CFG_TRAIN, @@ -1692,7 +1696,9 @@ static void handle_serdes_issues(struct qib_pportdata *ppd, u64 ibcst) break; } - if (ibclt == IB_7322_LT_STATE_CFGTEST && + if (((ibclt >= IB_7322_LT_STATE_CFGTEST && + ibclt <= IB_7322_LT_STATE_CFGWAITENH) || + ibclt == IB_7322_LT_STATE_LINKUP) && (ibcst & SYM_MASK(IBCStatusA_0, LinkSpeedQDR))) { force_h1(ppd); ppd->cpspec->qdr_reforce = 1; @@ -7301,12 +7307,17 @@ static void ibsd_wr_allchans(struct qib_pportdata *ppd, int addr, unsigned data, static void serdes_7322_los_enable(struct qib_pportdata *ppd, int enable) { u64 data = qib_read_kreg_port(ppd, krp_serdesctrl); - printk(KERN_INFO QIB_DRV_NAME " IB%u:%u Turning LOS %s\n", - ppd->dd->unit, ppd->port, (enable ? "on" : "off")); - if (enable) + u8 state = SYM_FIELD(data, IBSerdesCtrl_0, RXLOSEN); + + if (enable && !state) { + printk(KERN_INFO QIB_DRV_NAME " IB%u:%u Turning LOS on\n", + ppd->dd->unit, ppd->port); data |= SYM_MASK(IBSerdesCtrl_0, RXLOSEN); - else + } else if (!enable && state) { + printk(KERN_INFO QIB_DRV_NAME " IB%u:%u Turning LOS off\n", + ppd->dd->unit, ppd->port); data &= ~SYM_MASK(IBSerdesCtrl_0, RXLOSEN); + } qib_write_kreg_port(ppd, krp_serdesctrl, data); } diff --git a/drivers/infiniband/hw/qib/qib_intr.c b/drivers/infiniband/hw/qib/qib_intr.c index a693c56ec8a6..6ae57d23004a 100644 --- a/drivers/infiniband/hw/qib/qib_intr.c +++ b/drivers/infiniband/hw/qib/qib_intr.c @@ -96,8 +96,12 @@ void qib_handle_e_ibstatuschanged(struct qib_pportdata *ppd, u64 ibcs) * states, or if it transitions from any of the up (INIT or better) * states into any of the down states (except link recovery), then * call the chip-specific code to take appropriate actions. + * + * ppd->lflags could be 0 if this is the first time the interrupt + * handlers has been called but the link is already up. */ - if (lstate >= IB_PORT_INIT && (ppd->lflags & QIBL_LINKDOWN) && + if (lstate >= IB_PORT_INIT && + (!ppd->lflags || (ppd->lflags & QIBL_LINKDOWN)) && ltstate == IB_PHYSPORTSTATE_LINKUP) { /* transitioned to UP */ if (dd->f_ib_updown(ppd, 1, ibcs)) From c44048dea28b3febd38417f36f7b2924d93f9bb2 Mon Sep 17 00:00:00 2001 From: Chris Ball Date: Thu, 26 May 2011 23:15:49 -0400 Subject: [PATCH 080/224] mmc: vub300: fix null dereferences in error handling Reported-by: Dan Carpenter Signed-off-by: Chris Ball --- drivers/mmc/host/vub300.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/drivers/mmc/host/vub300.c b/drivers/mmc/host/vub300.c index cbb03305b77b..d4455ffbefd8 100644 --- a/drivers/mmc/host/vub300.c +++ b/drivers/mmc/host/vub300.c @@ -2096,7 +2096,7 @@ static struct mmc_host_ops vub300_mmc_ops = { static int vub300_probe(struct usb_interface *interface, const struct usb_device_id *id) { /* NOT irq */ - struct vub300_mmc_host *vub300 = NULL; + struct vub300_mmc_host *vub300; struct usb_host_interface *iface_desc; struct usb_device *udev = usb_get_dev(interface_to_usbdev(interface)); int i; @@ -2118,23 +2118,20 @@ static int vub300_probe(struct usb_interface *interface, command_out_urb = usb_alloc_urb(0, GFP_KERNEL); if (!command_out_urb) { retval = -ENOMEM; - dev_err(&vub300->udev->dev, - "not enough memory for the command_out_urb\n"); + dev_err(&udev->dev, "not enough memory for command_out_urb\n"); goto error0; } command_res_urb = usb_alloc_urb(0, GFP_KERNEL); if (!command_res_urb) { retval = -ENOMEM; - dev_err(&vub300->udev->dev, - "not enough memory for the command_res_urb\n"); + dev_err(&udev->dev, "not enough memory for command_res_urb\n"); goto error1; } /* this also allocates memory for our VUB300 mmc host device */ mmc = mmc_alloc_host(sizeof(struct vub300_mmc_host), &udev->dev); if (!mmc) { retval = -ENOMEM; - dev_err(&vub300->udev->dev, - "not enough memory for the mmc_host\n"); + dev_err(&udev->dev, "not enough memory for the mmc_host\n"); goto error4; } /* MMC core transfer sizes tunable parameters */ From b9c350a0a9d3564d20f5301f472b8841f8934c7d Mon Sep 17 00:00:00 2001 From: Wanlong Gao Date: Tue, 24 May 2011 21:03:22 +0800 Subject: [PATCH 081/224] mmc: of_mmc_spi: add NO_IRQ define to of_mmc_spi.c Provide a dummy value of NO_IRQ for architectures that don't support it (such as MIPS). Fixes the build error for MIPS. Signed-off-by: Wanlong Gao Acked-by: Grant Likely Signed-off-by: Chris Ball --- drivers/mmc/host/of_mmc_spi.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/mmc/host/of_mmc_spi.c b/drivers/mmc/host/of_mmc_spi.c index e2aecb7f1d5c..ab66f2454dc4 100644 --- a/drivers/mmc/host/of_mmc_spi.c +++ b/drivers/mmc/host/of_mmc_spi.c @@ -25,6 +25,11 @@ #include #include +/* For archs that don't support NO_IRQ (such as mips), provide a dummy value */ +#ifndef NO_IRQ +#define NO_IRQ 0 +#endif + MODULE_LICENSE("GPL"); enum { From 742a0c7cae9bd0f913c71722289b8ec862d70a2b Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Fri, 27 May 2011 16:54:05 +0200 Subject: [PATCH 082/224] mmc: sdhi: fix module unloading MMC host drivers must be able to process interrupts during mmc_remove_host(). Signed-off-by: Guennadi Liakhovetski Signed-off-by: Chris Ball --- drivers/mmc/host/sh_mobile_sdhi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/host/sh_mobile_sdhi.c b/drivers/mmc/host/sh_mobile_sdhi.c index b3654293017b..2353bb797d6c 100644 --- a/drivers/mmc/host/sh_mobile_sdhi.c +++ b/drivers/mmc/host/sh_mobile_sdhi.c @@ -165,13 +165,14 @@ static int sh_mobile_sdhi_remove(struct platform_device *pdev) p->pdata = NULL; + tmio_mmc_host_remove(host); + for (i = 0; i < 3; i++) { irq = platform_get_irq(pdev, i); if (irq >= 0) free_irq(irq, host); } - tmio_mmc_host_remove(host); clk_disable(priv->clk); clk_put(priv->clk); kfree(priv); From be98ca652faa6468916a9b7608befff215a8ca70 Mon Sep 17 00:00:00 2001 From: Manoj Iyer Date: Thu, 26 May 2011 11:19:05 -0500 Subject: [PATCH 083/224] mmc: Add PCI fixup quirks for Ricoh 1180:e823 reader Signed-off-by: Manoj Iyer Cc: Signed-off-by: Chris Ball --- drivers/pci/quirks.c | 2 ++ include/linux/pci_ids.h | 1 + 2 files changed, 3 insertions(+) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index e8a140669f90..02145e9697a9 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -2761,6 +2761,8 @@ static void ricoh_mmc_fixup_r5c832(struct pci_dev *dev) } DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5C832, ricoh_mmc_fixup_r5c832); DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5C832, ricoh_mmc_fixup_r5c832); +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5CE823, ricoh_mmc_fixup_r5c832); +DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5CE823, ricoh_mmc_fixup_r5c832); #endif /*CONFIG_MMC_RICOH_MMC*/ #if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP) diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index a311008af5e1..f8910e155566 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1537,6 +1537,7 @@ #define PCI_DEVICE_ID_RICOH_RL5C476 0x0476 #define PCI_DEVICE_ID_RICOH_RL5C478 0x0478 #define PCI_DEVICE_ID_RICOH_R5C822 0x0822 +#define PCI_DEVICE_ID_RICOH_R5CE823 0xe823 #define PCI_DEVICE_ID_RICOH_R5C832 0x0832 #define PCI_DEVICE_ID_RICOH_R5C843 0x0843 From df18d127f4fed7a0284bcfa8d2843800cdb63b72 Mon Sep 17 00:00:00 2001 From: Boaz Harrosh Date: Fri, 17 Jun 2011 16:25:51 -0400 Subject: [PATCH 084/224] pnfs-obj: No longer needed to take an extra ref at add_device Andy's last device_cache patches, already take an extra reference on the newly inserted device_id. So we can remove it from obj-io. Without this patch the device_ids are leaked. Andy's patches are not in Linus tree yet. So I'm not sure if they are scheduled for this Kernel or the next. This patch should be added as part of these. CC: Andy Adamson Signed-off-by: Boaz Harrosh Signed-off-by: Trond Myklebust --- fs/nfs/objlayout/objio_osd.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c index 9cf208df1f25..eb4aafa9f521 100644 --- a/fs/nfs/objlayout/objio_osd.c +++ b/fs/nfs/objlayout/objio_osd.c @@ -108,7 +108,6 @@ _dev_list_add(const struct nfs_server *nfss, de = n; } - atomic_inc(&de->id_node.ref); return de; } From 155d109b5f52ffd749219b27702462dcd9cf4f8d Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Mon, 20 Jun 2011 13:23:14 +0200 Subject: [PATCH 085/224] block: add REQ_SECURE to REQ_COMMON_MASK Add REQ_SECURE flag to REQ_COMMON_MASK so that init_request_from_bio() can pass it to @req->cmd_flags. Signed-off-by: Namhyung Kim Acked-by: Adrian Hunter Cc: stable@kernel.org # 2.6.36 and newer Signed-off-by: Jens Axboe --- include/linux/blk_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 2a7cea53ca0d..6395692b2e7a 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -167,7 +167,7 @@ enum rq_flag_bits { (REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | REQ_FAILFAST_DRIVER) #define REQ_COMMON_MASK \ (REQ_WRITE | REQ_FAILFAST_MASK | REQ_SYNC | REQ_META | REQ_DISCARD | \ - REQ_NOIDLE | REQ_FLUSH | REQ_FUA) + REQ_NOIDLE | REQ_FLUSH | REQ_FUA | REQ_SECURE) #define REQ_CLONE_MASK REQ_COMMON_MASK #define REQ_RAHEAD (1 << __REQ_RAHEAD) From 6f2e810ad5d162c2bfa063c1811087277b299e4e Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Mon, 20 Jun 2011 10:27:07 +0200 Subject: [PATCH 086/224] ALSA: HDA: Remove quirk for an HP device The reporter, who is running kernel 2.6.38, reports that he needs to set model=auto for the headphone output to work correctly. BugLink: http://bugs.launchpad.net/bugs/761022 Cc: stable@kernel.org (v2.6.38+) Reported-by: Jo Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 61a774b3d3cb..c923b2cc9e53 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -4883,7 +4883,6 @@ static const struct snd_pci_quirk alc880_cfg_tbl[] = { SND_PCI_QUIRK(0x1025, 0xe309, "ULI", ALC880_3ST_DIG), SND_PCI_QUIRK(0x1025, 0xe310, "ULI", ALC880_3ST), SND_PCI_QUIRK(0x1039, 0x1234, NULL, ALC880_6ST_DIG), - SND_PCI_QUIRK(0x103c, 0x2a09, "HP", ALC880_5ST), SND_PCI_QUIRK(0x1043, 0x10b3, "ASUS W1V", ALC880_ASUS_W1V), SND_PCI_QUIRK(0x1043, 0x10c2, "ASUS W6A", ALC880_ASUS_DIG), SND_PCI_QUIRK(0x1043, 0x10c3, "ASUS Wxx", ALC880_ASUS_DIG), From 206b6310fd0268a6ca50cf36f03b0f4eee5602ec Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Mon, 20 Jun 2011 10:30:04 -0500 Subject: [PATCH 087/224] jfs: old_agsize should be 64 bits in jfs_extendfs Signed-off-by: Dave Kleikamp --- fs/jfs/resize.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/jfs/resize.c b/fs/jfs/resize.c index 8ea5efb5a34e..8d0c1c7c0820 100644 --- a/fs/jfs/resize.c +++ b/fs/jfs/resize.c @@ -80,7 +80,7 @@ int jfs_extendfs(struct super_block *sb, s64 newLVSize, int newLogSize) int log_formatted = 0; struct inode *iplist[1]; struct jfs_superblock *j_sb, *j_sb2; - uint old_agsize; + s64 old_agsize; int agsizechanged = 0; struct buffer_head *bh, *bh2; From 28e0fa894cd5996d3007ce82f07226f79beb7286 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Mon, 20 Jun 2011 10:32:46 -0500 Subject: [PATCH 088/224] jfs: Update agstart when resizing volume A comment indicates that the IAG's agstart does not need to be updated since it will always point to a block in the same aggregate group, but jfs_fsck isn't so forgiving and reports it as an error. I'm fixing this in jfsutils as well, so either a new kernel or new utilities will be sufficient to fix the problem. Signed-off-by: Dave Kleikamp --- fs/jfs/jfs_imap.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index ed53a4740168..0533e8f3d19e 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -2921,10 +2921,9 @@ int diExtendFS(struct inode *ipimap, struct inode *ipbmap) continue; } - /* agstart that computes to the same ag is treated as same; */ agstart = le64_to_cpu(iagp->agstart); - /* iagp->agstart = agstart & ~(mp->db_agsize - 1); */ n = agstart >> mp->db_agl2size; + iagp->agstart = cpu_to_le64((s64)n << mp->db_agl2size); /* compute backed inodes */ numinos = (EXTSPERIAG - le32_to_cpu(iagp->nfreeexts)) From d31b53e3cd069e02290ed8a648aa8c7618d6fe77 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Mon, 20 Jun 2011 10:53:46 -0500 Subject: [PATCH 089/224] JFS: Don't save agno in the inode Resizing the file system can result in an in-memory inode being remapped to a different aggregate group (AG). A cached AG number can cause problems when trying to free or allocate inodes. Instead, save the IAG's agstart address and calculate the agno when we need it. Signed-off-by: Dave Kleikamp --- fs/jfs/file.c | 6 +++--- fs/jfs/jfs_imap.c | 9 ++++----- fs/jfs/jfs_incore.h | 3 ++- 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/fs/jfs/file.c b/fs/jfs/file.c index c5ce6c1d1ff4..2f3f531f3606 100644 --- a/fs/jfs/file.c +++ b/fs/jfs/file.c @@ -66,9 +66,9 @@ static int jfs_open(struct inode *inode, struct file *file) struct jfs_inode_info *ji = JFS_IP(inode); spin_lock_irq(&ji->ag_lock); if (ji->active_ag == -1) { - ji->active_ag = ji->agno; - atomic_inc( - &JFS_SBI(inode->i_sb)->bmap->db_active[ji->agno]); + struct jfs_sb_info *jfs_sb = JFS_SBI(inode->i_sb); + ji->active_ag = BLKTOAG(addressPXD(&ji->ixpxd), jfs_sb); + atomic_inc( &jfs_sb->bmap->db_active[ji->active_ag]); } spin_unlock_irq(&ji->ag_lock); } diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 0533e8f3d19e..b78b2f978f04 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -397,7 +397,7 @@ int diRead(struct inode *ip) release_metapage(mp); /* set the ag for the inode */ - JFS_IP(ip)->agno = BLKTOAG(agstart, sbi); + JFS_IP(ip)->agstart = agstart; JFS_IP(ip)->active_ag = -1; return (rc); @@ -901,7 +901,7 @@ int diFree(struct inode *ip) /* get the allocation group for this ino. */ - agno = JFS_IP(ip)->agno; + agno = BLKTOAG(JFS_IP(ip)->agstart, JFS_SBI(ip->i_sb)); /* Lock the AG specific inode map information */ @@ -1315,12 +1315,11 @@ int diFree(struct inode *ip) static inline void diInitInode(struct inode *ip, int iagno, int ino, int extno, struct iag * iagp) { - struct jfs_sb_info *sbi = JFS_SBI(ip->i_sb); struct jfs_inode_info *jfs_ip = JFS_IP(ip); ip->i_ino = (iagno << L2INOSPERIAG) + ino; jfs_ip->ixpxd = iagp->inoext[extno]; - jfs_ip->agno = BLKTOAG(le64_to_cpu(iagp->agstart), sbi); + jfs_ip->agstart = le64_to_cpu(iagp->agstart); jfs_ip->active_ag = -1; } @@ -1379,7 +1378,7 @@ int diAlloc(struct inode *pip, bool dir, struct inode *ip) */ /* get the ag number of this iag */ - agno = JFS_IP(pip)->agno; + agno = BLKTOAG(JFS_IP(pip)->agstart, JFS_SBI(pip->i_sb)); if (atomic_read(&JFS_SBI(pip->i_sb)->bmap->db_active[agno])) { /* diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h index 1439f119ec83..5097839b7709 100644 --- a/fs/jfs/jfs_incore.h +++ b/fs/jfs/jfs_incore.h @@ -50,8 +50,9 @@ struct jfs_inode_info { short btindex; /* btpage entry index*/ struct inode *ipimap; /* inode map */ unsigned long cflag; /* commit flags */ + long agstart; /* agstart of the containing IAG */ u16 bxflag; /* xflag of pseudo buffer? */ - unchar agno; /* ag number */ + unchar pad; signed char active_ag; /* ag currently allocating from */ lid_t blid; /* lid of pseudo buffer? */ lid_t atlhead; /* anonymous tlock list head */ From 19345cb299e8234006c5125151ab723e851a1d24 Mon Sep 17 00:00:00 2001 From: Benny Halevy Date: Sun, 19 Jun 2011 18:33:46 -0400 Subject: [PATCH 090/224] NFSv4.1: file layout must consider pg_bsize for coalescing Otherwise we end up overflowing the rpc buffer size on the receive end. Signed-off-by: Benny Halevy Signed-off-by: Trond Myklebust --- fs/nfs/nfs4filelayout.c | 6 ++++-- fs/nfs/pagelist.c | 3 ++- include/linux/nfs_page.h | 3 +++ 3 files changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 5d6f369b15d0..0bafcc91c27f 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -30,6 +30,7 @@ */ #include +#include #include "internal.h" #include "nfs4filelayout.h" @@ -666,8 +667,9 @@ filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, u64 p_stripe, r_stripe; u32 stripe_unit; - if (!pnfs_generic_pg_test(pgio, prev, req)) - return 0; + if (!pnfs_generic_pg_test(pgio, prev, req) || + !nfs_generic_pg_test(pgio, prev, req)) + return false; if (!pgio->pg_lseg) return 1; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 7913961aff22..009855716286 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -204,7 +204,7 @@ nfs_wait_on_request(struct nfs_page *req) TASK_UNINTERRUPTIBLE); } -static bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_page *prev, struct nfs_page *req) +bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_page *prev, struct nfs_page *req) { /* * FIXME: ideally we should be able to coalesce all requests @@ -218,6 +218,7 @@ static bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_p return desc->pg_count + req->wb_bytes <= desc->pg_bsize; } +EXPORT_SYMBOL_GPL(nfs_generic_pg_test); /** * nfs_pageio_init - initialise a page io descriptor diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 3a34e80ae92f..25311b3bedf8 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -92,6 +92,9 @@ extern int nfs_pageio_add_request(struct nfs_pageio_descriptor *, struct nfs_page *); extern void nfs_pageio_complete(struct nfs_pageio_descriptor *desc); extern void nfs_pageio_cond_complete(struct nfs_pageio_descriptor *, pgoff_t); +extern bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, + struct nfs_page *prev, + struct nfs_page *req); extern int nfs_wait_on_request(struct nfs_page *); extern void nfs_unlock_request(struct nfs_page *req); extern int nfs_set_page_tag_locked(struct nfs_page *req); From ecc90462b428db2ad2ee5081c45496ed10f3a633 Mon Sep 17 00:00:00 2001 From: Dave Kleikamp Date: Mon, 20 Jun 2011 17:53:24 -0500 Subject: [PATCH 091/224] jfs: agstart field must be 64 bits The previous patch added the agstart field to jfs_ip, but declared it a long. We need to make sure its 64 bits on every platform. Signed-off-by: Dave Kleikamp --- fs/jfs/jfs_incore.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h index 5097839b7709..584a4a1a6e81 100644 --- a/fs/jfs/jfs_incore.h +++ b/fs/jfs/jfs_incore.h @@ -50,7 +50,7 @@ struct jfs_inode_info { short btindex; /* btpage entry index*/ struct inode *ipimap; /* inode map */ unsigned long cflag; /* commit flags */ - long agstart; /* agstart of the containing IAG */ + u64 agstart; /* agstart of the containing IAG */ u16 bxflag; /* xflag of pseudo buffer? */ unchar pad; signed char active_ag; /* ag currently allocating from */ From c933790614529c06b221f73ff36e2456aecee30d Mon Sep 17 00:00:00 2001 From: Tony Vroon Date: Mon, 20 Jun 2011 22:11:11 +0100 Subject: [PATCH 092/224] ALSA: hda - Remove ALC268 model override for CPR2000 The "diverse" Quanta ID 0x0763 is overridden to ALC268_ACER. This keeps headphone automute and microphone input from operating on at least one laptop from Opti Systems. Without the override, the BIOS parser does a fine job setting the card up and everything works. Tested-By: Peter Schneider Signed-off-by: Tony Vroon Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 1 - 1 file changed, 1 deletion(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index c923b2cc9e53..475ed1e8ffc6 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -13871,7 +13871,6 @@ static const struct snd_pci_quirk alc268_cfg_tbl[] = { SND_PCI_QUIRK(0x1043, 0x1205, "ASUS W7J", ALC268_3ST), SND_PCI_QUIRK(0x1170, 0x0040, "ZEPTO", ALC268_ZEPTO), SND_PCI_QUIRK(0x14c0, 0x0025, "COMPAL IFL90/JFL-92", ALC268_TOSHIBA), - SND_PCI_QUIRK(0x152d, 0x0763, "Diverse (CPR2000)", ALC268_ACER), SND_PCI_QUIRK(0x152d, 0x0771, "Quanta IL1", ALC267_QUANTA_IL1), {} }; From 42467b32ce4f1ba933673b396f807110e3618ff5 Mon Sep 17 00:00:00 2001 From: Lydia Wang Date: Mon, 20 Jun 2011 14:14:37 +0800 Subject: [PATCH 093/224] ALSA: VIA HDA: Modify initial verbs list for VT1718S. Remove some invalid initial verbs and correct some wrong initial verbs for VT1718S codec. Signed-off-by: Lydia Wang Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_via.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c index c952582fb218..abee9ac15902 100644 --- a/sound/pci/hda/patch_via.c +++ b/sound/pci/hda/patch_via.c @@ -4283,9 +4283,6 @@ static const struct hda_verb vt1718S_volume_init_verbs[] = { {0x21, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, {0x21, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, {0x21, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(5)}, - - /* Setup default input of Front HP to MW9 */ - {0x28, AC_VERB_SET_CONNECT_SEL, 0x1}, /* PW9 PW10 Output enable */ {0x2d, AC_VERB_SET_PIN_WIDGET_CONTROL, AC_PINCTL_OUT_EN}, {0x2e, AC_VERB_SET_PIN_WIDGET_CONTROL, AC_PINCTL_OUT_EN}, @@ -4294,10 +4291,10 @@ static const struct hda_verb vt1718S_volume_init_verbs[] = { /* Enable Boost Volume backdoor */ {0x1, 0xf88, 0x8}, /* MW0/1/2/3/4: un-mute index 0 (AOWx), mute index 1 (MW9) */ - {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, - {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, {0x1c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, @@ -4307,8 +4304,6 @@ static const struct hda_verb vt1718S_volume_init_verbs[] = { /* set MUX1 = 2 (AOW4), MUX2 = 1 (AOW3) */ {0x34, AC_VERB_SET_CONNECT_SEL, 0x2}, {0x35, AC_VERB_SET_CONNECT_SEL, 0x1}, - /* Unmute MW4's index 0 */ - {0x1c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, { } }; From ba31a60d0fd8a3976d44d32f2b82491c62646b2a Mon Sep 17 00:00:00 2001 From: Lydia Wang Date: Mon, 20 Jun 2011 14:16:33 +0800 Subject: [PATCH 094/224] ALSA: VIA HDA: Mute/unmute mixer conncted to Headphone for VT1718S. When switch HP independent mode, mute/unmute connctions of mixer which is connected to headphone for VT1718S. Signed-off-by: Lydia Wang Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_via.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c index abee9ac15902..f1a80cd6afe4 100644 --- a/sound/pci/hda/patch_via.c +++ b/sound/pci/hda/patch_via.c @@ -745,12 +745,23 @@ static int via_independent_hp_put(struct snd_kcontrol *kcontrol, struct via_spec *spec = codec->spec; hda_nid_t nid = kcontrol->private_value; unsigned int pinsel = ucontrol->value.enumerated.item[0]; + unsigned int parm0, parm1; /* Get Independent Mode index of headphone pin widget */ spec->hp_independent_mode = spec->hp_independent_mode_index == pinsel ? 1 : 0; - if (spec->codec_type == VT1718S) + if (spec->codec_type == VT1718S) { snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_CONNECT_SEL, pinsel ? 2 : 0); + /* Set correct mute switch for MW3 */ + parm0 = spec->hp_independent_mode ? + AMP_IN_UNMUTE(0) : AMP_IN_MUTE(0); + parm1 = spec->hp_independent_mode ? + AMP_IN_MUTE(1) : AMP_IN_UNMUTE(1); + snd_hda_codec_write(codec, 0x1b, 0, + AC_VERB_SET_AMP_GAIN_MUTE, parm0); + snd_hda_codec_write(codec, 0x1b, 0, + AC_VERB_SET_AMP_GAIN_MUTE, parm1); + } else snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_CONNECT_SEL, pinsel); From e905a83acd7bf8989c3d5ba3099b72675f5d7d29 Mon Sep 17 00:00:00 2001 From: Lydia Wang Date: Mon, 20 Jun 2011 14:17:56 +0800 Subject: [PATCH 095/224] ALSA: VIA HDA: Create a master amplifier control for VT1718S. Create a master volume and mute control of playback for VT1718S. Signed-off-by: Lydia Wang Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_via.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/sound/pci/hda/patch_via.c b/sound/pci/hda/patch_via.c index f1a80cd6afe4..f43bb0eaed8b 100644 --- a/sound/pci/hda/patch_via.c +++ b/sound/pci/hda/patch_via.c @@ -4462,6 +4462,19 @@ static int vt1718S_auto_create_multi_out_ctls(struct via_spec *spec, if (err < 0) return err; } else if (i == AUTO_SEQ_FRONT) { + /* add control to mixer index 0 */ + err = via_add_control(spec, VIA_CTL_WIDGET_VOL, + "Master Front Playback Volume", + HDA_COMPOSE_AMP_VAL(0x21, 3, 5, + HDA_INPUT)); + if (err < 0) + return err; + err = via_add_control(spec, VIA_CTL_WIDGET_MUTE, + "Master Front Playback Switch", + HDA_COMPOSE_AMP_VAL(0x21, 3, 5, + HDA_INPUT)); + if (err < 0) + return err; /* Front */ sprintf(name, "%s Playback Volume", chname[i]); err = via_add_control( From 573619d165b85152eeddd3b3871002c48cd94e42 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 20 Jun 2011 16:46:01 +0100 Subject: [PATCH 096/224] ARM: SMP: wait for CPU to be marked active When we bring a CPU online, we should wait for it to become active before entering the idle thread, so we know that the scheduler and thread migration is going to work. Signed-off-by: Russell King --- arch/arm/kernel/smp.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 344e52b16c8c..e7f92a4321f3 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -318,9 +318,13 @@ asmlinkage void __cpuinit secondary_start_kernel(void) smp_store_cpu_info(cpu); /* - * OK, now it's safe to let the boot CPU continue + * OK, now it's safe to let the boot CPU continue. Wait for + * the CPU migration code to notice that the CPU is online + * before we continue. */ set_cpu_online(cpu, true); + while (!cpu_active(cpu)) + cpu_relax(); /* * OK, it's off to the idle thread for us From 946a105e16651c35e9cc670bff23812761f1ad35 Mon Sep 17 00:00:00 2001 From: Dave Martin Date: Tue, 14 Jun 2011 14:20:44 +0100 Subject: [PATCH 097/224] ARM: 6961/1: zImage: Add build-time check for correctly-sized proc_type entries It is easy to mis-maintain the proc_types table such that the entries become wrongly-sized and misaligned when the kernel is built in Thumb-2. This patch adds an assembly-time check which will turn most common size/alignment mistakes in this table into build failures, to avoid having to debug the boot-time kernel hang which would happen if the resulting kernel were actually booted. Signed-off-by: Dave Martin Acked-by: Nicolas Pitre Signed-off-by: Russell King --- arch/arm/boot/compressed/head.S | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S index 942fad97e447..940b20178107 100644 --- a/arch/arm/boot/compressed/head.S +++ b/arch/arm/boot/compressed/head.S @@ -597,6 +597,8 @@ __common_mmu_cache_on: sub pc, lr, r0, lsr #32 @ properly flush pipeline #endif +#define PROC_ENTRY_SIZE (4*5) + /* * Here follow the relocatable cache support functions for the * various processors. This is a generic hook for locating an @@ -624,7 +626,7 @@ call_cache_fn: adr r12, proc_types ARM( addeq pc, r12, r3 ) @ call cache function THUMB( addeq r12, r3 ) THUMB( moveq pc, r12 ) @ call cache function - add r12, r12, #4*5 + add r12, r12, #PROC_ENTRY_SIZE b 1b /* @@ -794,6 +796,16 @@ proc_types: .size proc_types, . - proc_types + /* + * If you get a "non-constant expression in ".if" statement" + * error from the assembler on this line, check that you have + * not accidentally written a "b" instruction where you should + * have written W(b). + */ + .if (. - proc_types) % PROC_ENTRY_SIZE != 0 + .error "The size of one or more proc_types entries is wrong." + .endif + /* * Turn off the Cache and MMU. ARMv3 does not support * reading the control register, but ARMv4 does. From 082763a80ad11adbe8418fff1cfe466343035b7a Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 20 Jun 2011 08:13:00 +0100 Subject: [PATCH 098/224] ARM: 6969/1: plat-iop: fix build error The iop13xx_defconfig didn't build since the platform code uses defines from . Simply add the include so it compiles. Signed-off-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/plat-iop/cp6.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/plat-iop/cp6.c b/arch/arm/plat-iop/cp6.c index 9612a87e2a88..bab73e2c79db 100644 --- a/arch/arm/plat-iop/cp6.c +++ b/arch/arm/plat-iop/cp6.c @@ -18,6 +18,7 @@ */ #include #include +#include static int cp6_trap(struct pt_regs *regs, unsigned int instr) { From 8f7d5efbef8718a774ac5e347b4ec069f17fd9b4 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 10 Jun 2011 13:30:22 -0400 Subject: [PATCH 099/224] NFSv4.1: Fix some issues with pnfs_generic_pg_test 1. If the intention is to coalesce requests 'prev' and 'req' then we have to ensure at least that we have a layout starting at req_offset(prev). 2. If we're only requesting a minimal layout of length desc->pg_count, we need to test the length actually returned by the server before we allow the coalescing to occur. 3. We need to deal correctly with (pgio->lseg == NULL) 4. Fixup the test guarding the pnfs_update_layout. Signed-off-by: Trond Myklebust --- fs/nfs/objlayout/objio_osd.c | 3 +++ fs/nfs/pnfs.c | 12 +++++++----- 2 files changed, 10 insertions(+), 5 deletions(-) diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c index eb4aafa9f521..8ff2ea3f10ef 100644 --- a/fs/nfs/objlayout/objio_osd.c +++ b/fs/nfs/objlayout/objio_osd.c @@ -1000,6 +1000,9 @@ static bool objio_pg_test(struct nfs_pageio_descriptor *pgio, if (!pnfs_generic_pg_test(pgio, prev, req)) return false; + if (pgio->pg_lseg == NULL) + return true; + return pgio->pg_count + req->wb_bytes <= OBJIO_LSEG(pgio->pg_lseg)->max_io_size; } diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 539b94cb6c0f..0f42e02436ee 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1064,19 +1064,21 @@ pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, gfp_flags = GFP_NOFS; } - if (pgio->pg_count == prev->wb_bytes) { + if (pgio->pg_lseg == NULL) { + if (pgio->pg_count != prev->wb_bytes) + return true; /* This is first coelesce call for a series of nfs_pages */ pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, prev->wb_context, - req_offset(req), + req_offset(prev), pgio->pg_count, access_type, gfp_flags); - return true; + if (pgio->pg_lseg == NULL) + return true; } - if (pgio->pg_lseg && - req_offset(req) > end_offset(pgio->pg_lseg->pls_range.offset, + if (req_offset(req) > end_offset(pgio->pg_lseg->pls_range.offset, pgio->pg_lseg->pls_range.length)) return false; From 19982ba8562e33083cb5bbb59a74855d8a9624ea Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 10 Jun 2011 13:30:23 -0400 Subject: [PATCH 100/224] NFSv4.1: Fix an off-by-one error in pnfs_generic_pg_test And document what is going on there... Signed-off-by: Trond Myklebust --- fs/nfs/pnfs.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 0f42e02436ee..29c0ca7fc347 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1078,11 +1078,22 @@ pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, return true; } - if (req_offset(req) > end_offset(pgio->pg_lseg->pls_range.offset, - pgio->pg_lseg->pls_range.length)) - return false; - - return true; + /* + * Test if a nfs_page is fully contained in the pnfs_layout_range. + * Note that this test makes several assumptions: + * - that the previous nfs_page in the struct nfs_pageio_descriptor + * is known to lie within the range. + * - that the nfs_page being tested is known to be contiguous with the + * previous nfs_page. + * - Layout ranges are page aligned, so we only have to test the + * start offset of the request. + * + * Please also note that 'end_offset' is actually the offset of the + * first byte that lies outside the pnfs_layout_range. FIXME? + * + */ + return req_offset(req) < end_offset(pgio->pg_lseg->pls_range.offset, + pgio->pg_lseg->pls_range.length); } EXPORT_SYMBOL_GPL(pnfs_generic_pg_test); From 1650add23578b5ca35c1f1e863987180a8c03779 Mon Sep 17 00:00:00 2001 From: Bryan Schumaker Date: Thu, 2 Jun 2011 15:07:35 -0400 Subject: [PATCH 101/224] NFS: Fix decode_secinfo_maxsz I initially did the calculation in bytes, and not words Signed-off-by: Bryan Schumaker Signed-off-by: Trond Myklebust --- fs/nfs/nfs4xdr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 5db44a8dddf9..6870bc61ceec 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -255,7 +255,7 @@ static int nfs4_stat_to_errno(int); #define decode_fs_locations_maxsz \ (0) #define encode_secinfo_maxsz (op_encode_hdr_maxsz + nfs4_name_maxsz) -#define decode_secinfo_maxsz (op_decode_hdr_maxsz + 4 + (NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN))) +#define decode_secinfo_maxsz (op_decode_hdr_maxsz + 1 + ((NFS_MAX_SECFLAVORS * (16 + GSS_OID_MAX_LEN)) / 4)) #if defined(CONFIG_NFS_V4_1) #define NFS4_MAX_MACHINE_NAME_LEN (64) From 129b656a0de9a229a72fe4bb6bacd134a1477b44 Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Fri, 10 Jun 2011 16:05:51 -0700 Subject: [PATCH 102/224] PM / Runtime: Update doc: usage count no longer incremented across system PM Commit e8665002477f0278f84f898145b1f141ba26ee26 (PM: Allow pm_runtime_suspend() to succeed during system suspend) removed usage count increment across system PM. Update doc to reflect this. Signed-off-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- Documentation/power/runtime_pm.txt | 5 ----- 1 file changed, 5 deletions(-) diff --git a/Documentation/power/runtime_pm.txt b/Documentation/power/runtime_pm.txt index 654097b130b4..22accb3eb40e 100644 --- a/Documentation/power/runtime_pm.txt +++ b/Documentation/power/runtime_pm.txt @@ -566,11 +566,6 @@ to do this is: pm_runtime_set_active(dev); pm_runtime_enable(dev); -The PM core always increments the run-time usage counter before calling the -->prepare() callback and decrements it after calling the ->complete() callback. -Hence disabling run-time PM temporarily like this will not cause any run-time -suspend callbacks to be lost. - 7. Generic subsystem callbacks Subsystems may wish to conserve code space by using the set of generic power From 78420884e680da8fbc3240de2d3106437042381e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 18 Jun 2011 19:53:57 +0200 Subject: [PATCH 103/224] PM: Update documentation regarding sysdevs The part of Documentation/power/devices.txt regarding sysdevs is not valid any more after commit 2e711c04dbbf7a7732a3f7073b1fc285d12b369d (PM: Remove sysdev suspend, resume and shutdown operations), so remove it. Signed-off-by: Rafael J. Wysocki --- Documentation/power/devices.txt | 26 -------------------------- 1 file changed, 26 deletions(-) diff --git a/Documentation/power/devices.txt b/Documentation/power/devices.txt index 88880839ece4..ff923fe67d98 100644 --- a/Documentation/power/devices.txt +++ b/Documentation/power/devices.txt @@ -549,32 +549,6 @@ callbacks. The other platforms need not implement it or take it into account in any way. -System Devices --------------- -System devices (sysdevs) follow a slightly different API, which can be found in - - include/linux/sysdev.h - drivers/base/sys.c - -System devices will be suspended with interrupts disabled, and after all other -devices have been suspended. On resume, they will be resumed before any other -devices, and also with interrupts disabled. These things occur in special -"sysdev_driver" phases, which affect only system devices. - -Thus, after the suspend_noirq (or freeze_noirq or poweroff_noirq) phase, when -the non-boot CPUs are all offline and IRQs are disabled on the remaining online -CPU, then a sysdev_driver.suspend phase is carried out, and the system enters a -sleep state (or a system image is created). During resume (or after the image -has been created or loaded) a sysdev_driver.resume phase is carried out, IRQs -are enabled on the only online CPU, the non-boot CPUs are enabled, and the -resume_noirq (or thaw_noirq or restore_noirq) phase begins. - -Code to actually enter and exit the system-wide low power state sometimes -involves hardware details that are only known to the boot firmware, and -may leave a CPU running software (from SRAM or flash memory) that monitors -the system and manages its wakeup sequence. - - Device Low Power (suspend) States --------------------------------- Device low-power states aren't standard. One device might only handle From f76b168b6f117a49d36307053e1acbe30580ea5b Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sat, 18 Jun 2011 20:22:23 +0200 Subject: [PATCH 104/224] PM: Rename dev_pm_info.in_suspend to is_prepared This patch (as1473) renames the "in_suspend" field in struct dev_pm_info to "is_prepared", in preparation for an upcoming change. The new name is more descriptive of what the field really means. Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org --- drivers/base/power/main.c | 14 +++++++++----- drivers/usb/core/driver.c | 6 +++--- include/linux/device.h | 4 ++-- include/linux/pm.h | 2 +- 4 files changed, 15 insertions(+), 11 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index aa6320207745..bf5a59ac1957 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -57,7 +57,7 @@ static int async_error; */ void device_pm_init(struct device *dev) { - dev->power.in_suspend = false; + dev->power.is_prepared = false; init_completion(&dev->power.completion); complete_all(&dev->power.completion); dev->power.wakeup = NULL; @@ -91,7 +91,7 @@ void device_pm_add(struct device *dev) pr_debug("PM: Adding info for %s:%s\n", dev->bus ? dev->bus->name : "No Bus", dev_name(dev)); mutex_lock(&dpm_list_mtx); - if (dev->parent && dev->parent->power.in_suspend) + if (dev->parent && dev->parent->power.is_prepared) dev_warn(dev, "parent %s should not be sleeping\n", dev_name(dev->parent)); list_add_tail(&dev->power.entry, &dpm_list); @@ -511,7 +511,11 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) dpm_wait(dev->parent, async); device_lock(dev); - dev->power.in_suspend = false; + /* + * This is a fib. But we'll allow new children to be added below + * a resumed device, even if the device hasn't been completed yet. + */ + dev->power.is_prepared = false; if (dev->pwr_domain) { pm_dev_dbg(dev, state, "power domain "); @@ -670,7 +674,7 @@ void dpm_complete(pm_message_t state) struct device *dev = to_device(dpm_prepared_list.prev); get_device(dev); - dev->power.in_suspend = false; + dev->power.is_prepared = false; list_move(&dev->power.entry, &list); mutex_unlock(&dpm_list_mtx); @@ -1042,7 +1046,7 @@ int dpm_prepare(pm_message_t state) put_device(dev); break; } - dev->power.in_suspend = true; + dev->power.is_prepared = true; if (!list_empty(&dev->power.entry)) list_move_tail(&dev->power.entry, &dpm_prepared_list); put_device(dev); diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c index e35a17687c05..aa3cc465a601 100644 --- a/drivers/usb/core/driver.c +++ b/drivers/usb/core/driver.c @@ -375,7 +375,7 @@ static int usb_unbind_interface(struct device *dev) * Just re-enable it without affecting the endpoint toggles. */ usb_enable_interface(udev, intf, false); - } else if (!error && !intf->dev.power.in_suspend) { + } else if (!error && !intf->dev.power.is_prepared) { r = usb_set_interface(udev, intf->altsetting[0]. desc.bInterfaceNumber, 0); if (r < 0) @@ -960,7 +960,7 @@ void usb_rebind_intf(struct usb_interface *intf) } /* Try to rebind the interface */ - if (!intf->dev.power.in_suspend) { + if (!intf->dev.power.is_prepared) { intf->needs_binding = 0; rc = device_attach(&intf->dev); if (rc < 0) @@ -1107,7 +1107,7 @@ static int usb_resume_interface(struct usb_device *udev, if (intf->condition == USB_INTERFACE_UNBOUND) { /* Carry out a deferred switch to altsetting 0 */ - if (intf->needs_altsetting0 && !intf->dev.power.in_suspend) { + if (intf->needs_altsetting0 && !intf->dev.power.is_prepared) { usb_set_interface(udev, intf->altsetting[0]. desc.bInterfaceNumber, 0); intf->needs_altsetting0 = 0; diff --git a/include/linux/device.h b/include/linux/device.h index c66111affca9..553fd37b173b 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -654,13 +654,13 @@ static inline int device_is_registered(struct device *dev) static inline void device_enable_async_suspend(struct device *dev) { - if (!dev->power.in_suspend) + if (!dev->power.is_prepared) dev->power.async_suspend = true; } static inline void device_disable_async_suspend(struct device *dev) { - if (!dev->power.in_suspend) + if (!dev->power.is_prepared) dev->power.async_suspend = false; } diff --git a/include/linux/pm.h b/include/linux/pm.h index 3160648ccdda..cc536bd80984 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -425,7 +425,7 @@ struct dev_pm_info { pm_message_t power_state; unsigned int can_wakeup:1; unsigned int async_suspend:1; - unsigned int in_suspend:1; /* Owned by the PM core */ + bool is_prepared:1; /* Owned by the PM core */ spinlock_t lock; #ifdef CONFIG_PM_SLEEP struct list_head entry; From 8440f4b19494467883f8541b7aa28c7bbf6ac92b Mon Sep 17 00:00:00 2001 From: Michal Kubecek Date: Sat, 18 Jun 2011 20:34:01 +0200 Subject: [PATCH 105/224] PM: Free memory bitmaps if opening /dev/snapshot fails When opening /dev/snapshot device, snapshot_open() creates memory bitmaps which are freed in snapshot_release(). But if any of the callbacks called by pm_notifier_call_chain() returns NOTIFY_BAD, open() fails, snapshot_release() is never called and bitmaps are not freed. Next attempt to open /dev/snapshot then triggers BUG_ON() check in create_basic_memory_bitmaps(). This happens e.g. when vmwatchdog module is active on s390x. Signed-off-by: Michal Kubecek Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org --- kernel/power/user.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/power/user.c b/kernel/power/user.c index 7d02d33be699..42ddbc6f0de6 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -113,8 +113,10 @@ static int snapshot_open(struct inode *inode, struct file *filp) if (error) pm_notifier_call_chain(PM_POST_RESTORE); } - if (error) + if (error) { + free_basic_memory_bitmaps(); atomic_inc(&snapshot_device_available); + } data->frozen = 0; data->ready = 0; data->platform_support = 0; From 6d0e0e84f66d32c33511984dd3badd32364b863c Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Sat, 18 Jun 2011 22:42:09 +0200 Subject: [PATCH 106/224] PM: Fix async resume following suspend failure The PM core doesn't handle suspend failures correctly when it comes to asynchronously suspended devices. These devices are moved onto the dpm_suspended_list as soon as the corresponding async thread is started up, and they remain on the list even if they fail to suspend or the sleep transition is cancelled before they get suspended. As a result, when the PM core unwinds the transition, it tries to resume the devices even though they were never suspended. This patch (as1474) fixes the problem by adding a new "is_suspended" flag to dev_pm_info. Devices are resumed only if the flag is set. [rjw: * Moved the dev->power.is_suspended check into device_resume(), because we need to complete dev->power.completion and clear dev->power.is_prepared too for devices whose dev->power.is_suspended flags are unset. * Fixed __device_suspend() to avoid setting dev->power.is_suspended if async_error is different from zero.] Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org --- drivers/base/power/main.c | 14 ++++++++++++-- include/linux/pm.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index bf5a59ac1957..06f09bf89cb2 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -58,6 +58,7 @@ static int async_error; void device_pm_init(struct device *dev) { dev->power.is_prepared = false; + dev->power.is_suspended = false; init_completion(&dev->power.completion); complete_all(&dev->power.completion); dev->power.wakeup = NULL; @@ -517,6 +518,9 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) */ dev->power.is_prepared = false; + if (!dev->power.is_suspended) + goto Unlock; + if (dev->pwr_domain) { pm_dev_dbg(dev, state, "power domain "); error = pm_op(dev, &dev->pwr_domain->ops, state); @@ -552,6 +556,9 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) } End: + dev->power.is_suspended = false; + + Unlock: device_unlock(dev); complete_all(&dev->power.completion); @@ -839,11 +846,11 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) device_lock(dev); if (async_error) - goto End; + goto Unlock; if (pm_wakeup_pending()) { async_error = -EBUSY; - goto End; + goto Unlock; } if (dev->pwr_domain) { @@ -881,6 +888,9 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) } End: + dev->power.is_suspended = !error; + + Unlock: device_unlock(dev); complete_all(&dev->power.completion); diff --git a/include/linux/pm.h b/include/linux/pm.h index cc536bd80984..411e4f4be52b 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -426,6 +426,7 @@ struct dev_pm_info { unsigned int can_wakeup:1; unsigned int async_suspend:1; bool is_prepared:1; /* Owned by the PM core */ + bool is_suspended:1; /* Ditto */ spinlock_t lock; #ifdef CONFIG_PM_SLEEP struct list_head entry; From 4d1518f5668ef1b3dff6c3b30fa761fe5573cdaa Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 21 Jun 2011 23:24:33 +0200 Subject: [PATCH 107/224] PM / Runtime: Handle clocks correctly if CONFIG_PM_RUNTIME is unset Commit 85eb8c8d0b0900c073b0e6f89979ac9c439ade1a (PM / Runtime: Generic clock manipulation rountines for runtime PM (v6)) converted the shmobile platform to using generic code for runtime PM clock management, but it changed the behavior for CONFIG_PM_RUNTIME unset incorrectly. Specifically, for CONFIG_PM_RUNTIME unset pm_runtime_clk_notify() should enable clocks for action equal to BUS_NOTIFY_BIND_DRIVER and it should disable them for action equal to BUS_NOTIFY_UNBOUND_DRIVER (instead of BUS_NOTIFY_ADD_DEVICE and BUS_NOTIFY_DEL_DEVICE, respectively). Make this function behave as appropriate. Signed-off-by: Rafael J. Wysocki Acked-by: Magnus Damm --- drivers/base/power/clock_ops.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/base/power/clock_ops.c b/drivers/base/power/clock_ops.c index eaa8a854af03..ad367c4139b1 100644 --- a/drivers/base/power/clock_ops.c +++ b/drivers/base/power/clock_ops.c @@ -387,7 +387,7 @@ static int pm_runtime_clk_notify(struct notifier_block *nb, clknb = container_of(nb, struct pm_clk_notifier_block, nb); switch (action) { - case BUS_NOTIFY_ADD_DEVICE: + case BUS_NOTIFY_BIND_DRIVER: if (clknb->con_ids[0]) { for (con_id = clknb->con_ids; *con_id; con_id++) enable_clock(dev, *con_id); @@ -395,7 +395,7 @@ static int pm_runtime_clk_notify(struct notifier_block *nb, enable_clock(dev, NULL); } break; - case BUS_NOTIFY_DEL_DEVICE: + case BUS_NOTIFY_UNBOUND_DRIVER: if (clknb->con_ids[0]) { for (con_id = clknb->con_ids; *con_id; con_id++) disable_clock(dev, *con_id); From ca9c6890b598997165a7c85c001f382c910f12b0 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 21 Jun 2011 23:25:32 +0200 Subject: [PATCH 108/224] PM / Domains: Update documentation Commit 4d27e9dcff00a6425d779b065ec8892e4f391661 (PM: Make power domain callbacks take precedence over subsystem ones) forgot to update the device power management documentation to take changes made by it into account. Correct that mistake. Signed-off-by: Rafael J. Wysocki --- Documentation/power/devices.txt | 39 +++++++++++---------------------- 1 file changed, 13 insertions(+), 26 deletions(-) diff --git a/Documentation/power/devices.txt b/Documentation/power/devices.txt index ff923fe67d98..64565aac6e40 100644 --- a/Documentation/power/devices.txt +++ b/Documentation/power/devices.txt @@ -520,33 +520,20 @@ Support for power domains is provided through the pwr_domain field of struct device. This field is a pointer to an object of type struct dev_power_domain, defined in include/linux/pm.h, providing a set of power management callbacks analogous to the subsystem-level and device driver callbacks that are executed -for the given device during all power transitions, in addition to the respective -subsystem-level callbacks. Specifically, the power domain "suspend" callbacks -(i.e. ->runtime_suspend(), ->suspend(), ->freeze(), ->poweroff(), etc.) are -executed after the analogous subsystem-level callbacks, while the power domain -"resume" callbacks (i.e. ->runtime_resume(), ->resume(), ->thaw(), ->restore, -etc.) are executed before the analogous subsystem-level callbacks. Error codes -returned by the "suspend" and "resume" power domain callbacks are ignored. +for the given device during all power transitions, instead of the respective +subsystem-level callbacks. Specifically, if a device's pm_domain pointer is +not NULL, the ->suspend() callback from the object pointed to by it will be +executed instead of its subsystem's (e.g. bus type's) ->suspend() callback and +anlogously for all of the remaining callbacks. In other words, power management +domain callbacks, if defined for the given device, always take precedence over +the callbacks provided by the device's subsystem (e.g. bus type). -Power domain ->runtime_idle() callback is executed before the subsystem-level -->runtime_idle() callback and the result returned by it is not ignored. Namely, -if it returns error code, the subsystem-level ->runtime_idle() callback will not -be called and the helper function rpm_idle() executing it will return error -code. This mechanism is intended to help platforms where saving device state -is a time consuming operation and should only be carried out if all devices -in the power domain are idle, before turning off the shared power resource(s). -Namely, the power domain ->runtime_idle() callback may return error code until -the pm_runtime_idle() helper (or its asychronous version) has been called for -all devices in the power domain (it is recommended that the returned error code -be -EBUSY in those cases), preventing the subsystem-level ->runtime_idle() -callback from being run prematurely. - -The support for device power domains is only relevant to platforms needing to -use the same subsystem-level (e.g. platform bus type) and device driver power -management callbacks in many different power domain configurations and wanting -to avoid incorporating the support for power domains into the subsystem-level -callbacks. The other platforms need not implement it or take it into account -in any way. +The support for device power management domains is only relevant to platforms +needing to use the same device driver power management callbacks in many +different power domain configurations and wanting to avoid incorporating the +support for power domains into subsystem-level callbacks, for example by +modifying the platform bus type. Other platforms need not implement it or take +it into account in any way. Device Low Power (suspend) States From a5f76d5eba157bf637beb2dd18026db2917c512e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 21 Jun 2011 23:47:15 +0200 Subject: [PATCH 109/224] PCI / PM: Block races between runtime PM and system sleep After commit e8665002477f0278f84f898145b1f141ba26ee26 (PM: Allow pm_runtime_suspend() to succeed during system suspend) it is possible that a device resumed by the pm_runtime_resume(dev) in pci_pm_prepare() will be suspended immediately from a work item, timer function or otherwise, defeating the very purpose of calling pm_runtime_resume(dev) from there. To prevent that from happening it is necessary to increment the runtime PM usage counter of the device by replacing pm_runtime_resume() with pm_runtime_get_sync(). Moreover, the incremented runtime PM usage counter has to be decremented by the corresponding pci_pm_complete(), via pm_runtime_put_sync(). Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org Acked-by: Jesse Barnes --- drivers/pci/pci-driver.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 135df164a4c1..46767c53917a 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -624,7 +624,7 @@ static int pci_pm_prepare(struct device *dev) * system from the sleep state, we'll have to prevent it from signaling * wake-up. */ - pm_runtime_resume(dev); + pm_runtime_get_sync(dev); if (drv && drv->pm && drv->pm->prepare) error = drv->pm->prepare(dev); @@ -638,6 +638,8 @@ static void pci_pm_complete(struct device *dev) if (drv && drv->pm && drv->pm->complete) drv->pm->complete(dev); + + pm_runtime_put_sync(dev); } #else /* !CONFIG_PM_SLEEP */ From c008ba58af24dc5d0d8e9fe6e59d876910254761 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 16 Jun 2011 18:27:09 -0700 Subject: [PATCH 110/224] alarmtimers: Handle late rtc module loading The alarmtimers code currently picks a rtc device to use at late init time. However, if your rtc driver is loaded as a module, it may be registered after the alarmtimers late init code, leaving the alarmtimers nonfunctional. This patch moves the the rtcdevice selection to when we actually try to use it, allowing us to make use of rtc modules that may have been loaded at any point since bootup. CC: Thomas Gleixner CC: Meelis Roos Reported-by: Meelis Roos Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 137 +++++++++++++++++++-------------------- 1 file changed, 67 insertions(+), 70 deletions(-) diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 2d966244ea60..98ecf4e36f2f 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -42,15 +42,72 @@ static struct alarm_base { clockid_t base_clockid; } alarm_bases[ALARM_NUMTYPE]; +/* freezer delta & lock used to handle clock_nanosleep triggered wakeups */ +static ktime_t freezer_delta; +static DEFINE_SPINLOCK(freezer_delta_lock); + #ifdef CONFIG_RTC_CLASS /* rtc timer and device for setting alarm wakeups at suspend */ static struct rtc_timer rtctimer; static struct rtc_device *rtcdev; -#endif +static DEFINE_SPINLOCK(rtcdev_lock); -/* freezer delta & lock used to handle clock_nanosleep triggered wakeups */ -static ktime_t freezer_delta; -static DEFINE_SPINLOCK(freezer_delta_lock); +/** + * has_wakealarm - check rtc device has wakealarm ability + * @dev: current device + * @name_ptr: name to be returned + * + * This helper function checks to see if the rtc device can wake + * from suspend. + */ +static int has_wakealarm(struct device *dev, void *name_ptr) +{ + struct rtc_device *candidate = to_rtc_device(dev); + + if (!candidate->ops->set_alarm) + return 0; + if (!device_may_wakeup(candidate->dev.parent)) + return 0; + + *(const char **)name_ptr = dev_name(dev); + return 1; +} + +/** + * alarmtimer_get_rtcdev - Return selected rtcdevice + * + * This function returns the rtc device to use for wakealarms. + * If one has not already been chosen, it checks to see if a + * functional rtc device is available. + */ +static struct rtc_device *alarmtimer_get_rtcdev(void) +{ + struct device *dev; + char *str; + unsigned long flags; + struct rtc_device *ret; + + spin_lock_irqsave(&rtcdev_lock, flags); + if (!rtcdev) { + /* Find an rtc device and init the rtc_timer */ + dev = class_find_device(rtc_class, NULL, &str, has_wakealarm); + /* If we have a device then str is valid. See has_wakealarm() */ + if (dev) { + rtcdev = rtc_class_open(str); + /* + * Drop the reference we got in class_find_device, + * rtc_open takes its own. + */ + put_device(dev); + rtc_timer_init(&rtctimer, NULL, NULL); + } + } + ret = rtcdev; + spin_unlock_irqrestore(&rtcdev_lock, flags); + + return ret; +} +#endif /** @@ -166,6 +223,7 @@ static int alarmtimer_suspend(struct device *dev) struct rtc_time tm; ktime_t min, now; unsigned long flags; + struct rtc_device *rtc; int i; spin_lock_irqsave(&freezer_delta_lock, flags); @@ -173,8 +231,9 @@ static int alarmtimer_suspend(struct device *dev) freezer_delta = ktime_set(0, 0); spin_unlock_irqrestore(&freezer_delta_lock, flags); + rtc = alarmtimer_get_rtcdev(); /* If we have no rtcdev, just return */ - if (!rtcdev) + if (!rtc) return 0; /* Find the soonest timer to expire*/ @@ -199,12 +258,12 @@ static int alarmtimer_suspend(struct device *dev) WARN_ON(min.tv64 < NSEC_PER_SEC); /* Setup an rtc timer to fire that far in the future */ - rtc_timer_cancel(rtcdev, &rtctimer); - rtc_read_time(rtcdev, &tm); + rtc_timer_cancel(rtc, &rtctimer); + rtc_read_time(rtc, &tm); now = rtc_tm_to_ktime(tm); now = ktime_add(now, min); - rtc_timer_start(rtcdev, &rtctimer, now, ktime_set(0, 0)); + rtc_timer_start(rtc, &rtctimer, now, ktime_set(0, 0)); return 0; } @@ -638,65 +697,3 @@ static int __init alarmtimer_init(void) } device_initcall(alarmtimer_init); -#ifdef CONFIG_RTC_CLASS -/** - * has_wakealarm - check rtc device has wakealarm ability - * @dev: current device - * @name_ptr: name to be returned - * - * This helper function checks to see if the rtc device can wake - * from suspend. - */ -static int __init has_wakealarm(struct device *dev, void *name_ptr) -{ - struct rtc_device *candidate = to_rtc_device(dev); - - if (!candidate->ops->set_alarm) - return 0; - if (!device_may_wakeup(candidate->dev.parent)) - return 0; - - *(const char **)name_ptr = dev_name(dev); - return 1; -} - -/** - * alarmtimer_init_late - Late initializing of alarmtimer code - * - * This function locates a rtc device to use for wakealarms. - * Run as late_initcall to make sure rtc devices have been - * registered. - */ -static int __init alarmtimer_init_late(void) -{ - struct device *dev; - char *str; - - /* Find an rtc device and init the rtc_timer */ - dev = class_find_device(rtc_class, NULL, &str, has_wakealarm); - /* If we have a device then str is valid. See has_wakealarm() */ - if (dev) { - rtcdev = rtc_class_open(str); - /* - * Drop the reference we got in class_find_device, - * rtc_open takes its own. - */ - put_device(dev); - } - if (!rtcdev) { - printk(KERN_WARNING "No RTC device found, ALARM timers will" - " not wake from suspend"); - } - rtc_timer_init(&rtctimer, NULL, NULL); - - return 0; -} -#else -static int __init alarmtimer_init_late(void) -{ - printk(KERN_WARNING "Kernel not built with RTC support, ALARM timers" - " will not wake from suspend"); - return 0; -} -#endif -late_initcall(alarmtimer_init_late); From 1c6b39ad3f01514fd8dd84b5b412bafb75c19388 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 16 Jun 2011 18:47:37 -0700 Subject: [PATCH 111/224] alarmtimers: Return -ENOTSUPP if no RTC device is present MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Toralf Förster and Richard Weinberger noted that if there is no RTC device, the alarm timers core prints out an annoying "ALARM timers will not wake from suspend" message. This warning has been removed in a previous patch, however the issue still remains: The original idea was to support alarm timers even if there was no rtc device, as long as the system didn't go into suspend. However, after further consideration, communicating to the application that alarmtimers are not fully functional seems like the better solution. So this patch makes it so we return -ENOTSUPP to any posix _ALARM clockid calls if there is no backing RTC device on the system. Further this changes the behavior where when there is no rtc device we will check for one on clock_getres, clock_gettime, timer_create, and timer_nsleep instead of on suspend. CC: Toralf Förster CC: Richard Weinberger CC: Thomas Gleixner Reported-by: Toralf Förster Reported by: Richard Weinberger Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-) diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 98ecf4e36f2f..59f369f98a04 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -107,6 +107,9 @@ static struct rtc_device *alarmtimer_get_rtcdev(void) return ret; } +#else +#define alarmtimer_get_rtcdev() (0) +#define rtcdev (0) #endif @@ -231,7 +234,7 @@ static int alarmtimer_suspend(struct device *dev) freezer_delta = ktime_set(0, 0); spin_unlock_irqrestore(&freezer_delta_lock, flags); - rtc = alarmtimer_get_rtcdev(); + rtc = rtcdev; /* If we have no rtcdev, just return */ if (!rtc) return 0; @@ -381,6 +384,9 @@ static int alarm_clock_getres(const clockid_t which_clock, struct timespec *tp) { clockid_t baseid = alarm_bases[clock2alarm(which_clock)].base_clockid; + if (!alarmtimer_get_rtcdev()) + return -ENOTSUPP; + return hrtimer_get_res(baseid, tp); } @@ -395,6 +401,9 @@ static int alarm_clock_get(clockid_t which_clock, struct timespec *tp) { struct alarm_base *base = &alarm_bases[clock2alarm(which_clock)]; + if (!alarmtimer_get_rtcdev()) + return -ENOTSUPP; + *tp = ktime_to_timespec(base->gettime()); return 0; } @@ -410,6 +419,9 @@ static int alarm_timer_create(struct k_itimer *new_timer) enum alarmtimer_type type; struct alarm_base *base; + if (!alarmtimer_get_rtcdev()) + return -ENOTSUPP; + if (!capable(CAP_WAKE_ALARM)) return -EPERM; @@ -444,6 +456,9 @@ static void alarm_timer_get(struct k_itimer *timr, */ static int alarm_timer_del(struct k_itimer *timr) { + if (!rtcdev) + return -ENOTSUPP; + alarm_cancel(&timr->it.alarmtimer); return 0; } @@ -461,6 +476,9 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, struct itimerspec *new_setting, struct itimerspec *old_setting) { + if (!rtcdev) + return -ENOTSUPP; + /* Save old values */ old_setting->it_interval = ktime_to_timespec(timr->it.alarmtimer.period); @@ -600,6 +618,9 @@ static int alarm_timer_nsleep(const clockid_t which_clock, int flags, int ret = 0; struct restart_block *restart; + if (!alarmtimer_get_rtcdev()) + return -ENOTSUPP; + if (!capable(CAP_WAKE_ALARM)) return -EPERM; From 35052cffe0081904f3362c05818db900dd9dc7de Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 Jun 2011 10:29:51 +0100 Subject: [PATCH 112/224] MN10300: asm/uaccess.h needs to #include linux/kernel.h for might_sleep() MN10300's asm/uaccess.h needs to #include linux/kernel.h to get might_sleep() otherwise it fails to build on MN10300 allyesconfig. This fails in a few places with messages like the following: In file included from security/keys/trusted.c:14: include/linux/uaccess.h: In function '__copy_from_user_nocache': include/linux/uaccess.h:52: error: implicit declaration of function 'might_sleep' Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- arch/mn10300/include/asm/uaccess.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mn10300/include/asm/uaccess.h b/arch/mn10300/include/asm/uaccess.h index 3d6e60dad9d9..780560b330d9 100644 --- a/arch/mn10300/include/asm/uaccess.h +++ b/arch/mn10300/include/asm/uaccess.h @@ -15,6 +15,7 @@ * User space memory access functions */ #include +#include #include #include From b1d7dd80aadb9042e83f9778b484a2f92e0b04d4 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 21 Jun 2011 14:32:05 +0100 Subject: [PATCH 113/224] KEYS: Fix error handling in construct_key_and_link() Fix error handling in construct_key_and_link(). If construct_alloc_key() returns an error, it shouldn't pass out through the normal path as the key_serial() called by the kleave() statement will oops when it gets an error code in the pointer: BUG: unable to handle kernel paging request at ffffffffffffff84 IP: [] request_key_and_link+0x4d7/0x52f .. Call Trace: [] request_key+0x41/0x75 [] cifs_get_spnego_key+0x206/0x226 [cifs] [] CIFS_SessSetup+0x511/0x1234 [cifs] [] cifs_setup_session+0x90/0x1ae [cifs] [] cifs_get_smb_ses+0x34b/0x40f [cifs] [] cifs_mount+0x13f/0x504 [cifs] [] cifs_do_mount+0xc4/0x672 [cifs] [] mount_fs+0x69/0x155 [] vfs_kern_mount+0x63/0xa0 [] do_kern_mount+0x4d/0xdf [] do_mount+0x63c/0x69f [] sys_mount+0x88/0xc2 [] system_call_fastpath+0x16/0x1b Signed-off-by: David Howells Acked-by: Jeff Layton Signed-off-by: Linus Torvalds --- security/keys/request_key.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/security/keys/request_key.c b/security/keys/request_key.c index 8e319a416eec..82465328c39b 100644 --- a/security/keys/request_key.c +++ b/security/keys/request_key.c @@ -469,7 +469,7 @@ static struct key *construct_key_and_link(struct key_type *type, } else if (ret == -EINPROGRESS) { ret = 0; } else { - key = ERR_PTR(ret); + goto couldnt_alloc_key; } key_put(dest_keyring); @@ -479,6 +479,7 @@ static struct key *construct_key_and_link(struct key_type *type, construction_failed: key_negate_and_link(key, key_negative_timeout, NULL, NULL); key_put(key); +couldnt_alloc_key: key_put(dest_keyring); kleave(" = %d", ret); return ERR_PTR(ret); From 80629b0b0fd5ca868dc8eced28e6101e39ac2ef6 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Wed, 22 Jun 2011 16:24:07 +0200 Subject: [PATCH 114/224] [S390] kvm-s390: fix kconfig dependencies A user can create the Kconfig combination !VIRTUALIZATION, S390_GUEST which results in the following warnings: warning: (S390_GUEST) selects VIRTIO which has unmet direct dependencies (VIRTUALIZATION) warning: (S390_GUEST && VIRTIO_PCI && VIRTIO_BALLOON) selects VIRTIO_RING which has unmet direct dependencies (VIRTUALIZATION && VIRTIO) warning: (S390_GUEST) selects VIRTIO which has unmet direct dependencies (VIRTUALIZATION) warning: (S390_GUEST && VIRTIO_PCI && VIRTIO_BALLOON) selects VIRTIO_RING which has unmet direct dependencies (VIRTUALIZATION && VIRTIO) S390_GUEST has to select VIRTUALIZATION before selecting VIRTIO and friends. Reported-by: Jan Glauber Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 90d77bd078f5..c03fef7a9c22 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -579,6 +579,7 @@ config S390_GUEST def_bool y prompt "s390 guest support for KVM (EXPERIMENTAL)" depends on 64BIT && EXPERIMENTAL + select VIRTUALIZATION select VIRTIO select VIRTIO_RING select VIRTIO_CONSOLE From b530ce7a1af5a9355be518557d86b33c6d2cf088 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Wed, 22 Jun 2011 16:24:08 +0200 Subject: [PATCH 115/224] [S390] s390: enforce HW limits for the initial sampling rate On specific configurations with hwsampler opcontrol --start returns an error on "echo 1 >/dev/oprofile/enable". Turns out that the hw sampling interval is not checked against the hardware limits. Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky --- arch/s390/oprofile/init.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/s390/oprofile/init.c b/arch/s390/oprofile/init.c index 5995e9bc72d9..5e2ab03eaffd 100644 --- a/arch/s390/oprofile/init.c +++ b/arch/s390/oprofile/init.c @@ -151,6 +151,12 @@ static int oprofile_hwsampler_init(struct oprofile_operations *ops) if (oprofile_max_interval == 0) return -ENODEV; + /* The initial value should be sane */ + if (oprofile_hw_interval < oprofile_min_interval) + oprofile_hw_interval = oprofile_min_interval; + if (oprofile_hw_interval > oprofile_max_interval) + oprofile_hw_interval = oprofile_max_interval; + if (oprofile_timer_init(ops)) return -ENODEV; From 6815823431296082fa20c2f14007e194424660b8 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Wed, 22 Jun 2011 16:24:09 +0200 Subject: [PATCH 116/224] [S390] hwsampler: Set a sane default sampling rate The sampling interval for the hardware sampler is specified in cycles. (see SA23-2260-01 The Load-Program-Parameter and the CPU-Measurement Facilities) The current default value will therefore result in millions of samples. This patch changes the default sampling interval to 4M, which will result in ~1500 samples per second on a z196 reducing the overhead of sampling. Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky --- arch/s390/oprofile/init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/s390/oprofile/init.c b/arch/s390/oprofile/init.c index 5e2ab03eaffd..0e358c2cffeb 100644 --- a/arch/s390/oprofile/init.c +++ b/arch/s390/oprofile/init.c @@ -25,7 +25,7 @@ extern void s390_backtrace(struct pt_regs * const regs, unsigned int depth); #include "hwsampler.h" -#define DEFAULT_INTERVAL 4096 +#define DEFAULT_INTERVAL 4127518 #define DEFAULT_SDBT_BLOCKS 1 #define DEFAULT_SDB_BLOCKS 511 From 859c965149ab5004b58b1fffd98190b6664cb717 Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Wed, 22 Jun 2011 16:24:10 +0200 Subject: [PATCH 117/224] [S390] allow setting of upper 32 bit in smp_ctl_set_bit The bit shift operation in smp_ctl_set_bit does not specify the type of the shifted bit so integer is used as default. Therefore it is not possible to set bits in the upper 32 bit of the control register if the kernel runs in 64 bit mode. Fix this by specifying the type as unsigned long. Signed-off-by: Jan Glauber Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/smp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 52420d2785b3..1d55c95f617c 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -262,7 +262,7 @@ void smp_ctl_set_bit(int cr, int bit) memset(&parms.orvals, 0, sizeof(parms.orvals)); memset(&parms.andvals, 0xff, sizeof(parms.andvals)); - parms.orvals[cr] = 1 << bit; + parms.orvals[cr] = 1UL << bit; on_each_cpu(smp_ctl_bit_callback, &parms, 1); } EXPORT_SYMBOL(smp_ctl_set_bit); @@ -276,7 +276,7 @@ void smp_ctl_clear_bit(int cr, int bit) memset(&parms.orvals, 0, sizeof(parms.orvals)); memset(&parms.andvals, 0xff, sizeof(parms.andvals)); - parms.andvals[cr] = ~(1L << bit); + parms.andvals[cr] = ~(1UL << bit); on_each_cpu(smp_ctl_bit_callback, &parms, 1); } EXPORT_SYMBOL(smp_ctl_clear_bit); From 446b23a75804d7ffa4cca2d4d8f0afb822108c7e Mon Sep 17 00:00:00 2001 From: Pavel Shilovsky Date: Mon, 20 Jun 2011 12:33:16 +0400 Subject: [PATCH 118/224] CIFS: Fix problem with 3.0-rc1 null user mount failure Figured it out: it was broken by b946845a9dc523c759cae2b6a0f6827486c3221a commit - "cifs: cifs_parse_mount_options: do not tokenize mount options in-place". So, as a quick fix I suggest to apply this patch. [PATCH] CIFS: Fix kfree() with constant string in a null user case Signed-off-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/connect.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 12cf72dd0c42..19fdbda7aa95 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2937,7 +2937,11 @@ int cifs_setup_volume_info(struct smb_vol **pvolume_info, char *mount_data, if (volume_info->nullauth) { cFYI(1, "null user"); - volume_info->username = ""; + volume_info->username = kzalloc(1, GFP_KERNEL); + if (volume_info->username == NULL) { + rc = -ENOMEM; + goto out; + } } else if (volume_info->username) { /* BB fixme parse for domain name here */ cFYI(1, "Username: %s", volume_info->username); From 7553e8f2d5161a2b7a9b7a9f37be1b77e735552f Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Wed, 22 Jun 2011 18:13:01 -0700 Subject: [PATCH 119/224] mm, hotplug: fix error handling in mem_online_node() The error handling in mem_online_node() is incorrect: hotadd_new_pgdat() returns NULL if the new pgdat could not have been allocated and a pointer to it otherwise. mem_online_node() should fail if hotadd_new_pgdat() fails, not the inverse. This fixes an issue when memoryless nodes are not onlined and their sysfs interface is not registered when their first cpu is brought up. The bug was introduced by commit cf23422b9d76 ("cpu/mem hotplug: enable CPUs online before local memory online") iow v2.6.35. Signed-off-by: David Rientjes Reviewed-by: KOSAKI Motohiro Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 02159c755136..78dba9f04b5b 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -521,7 +521,7 @@ int mem_online_node(int nid) lock_memory_hotplug(); pgdat = hotadd_new_pgdat(nid, 0); - if (pgdat) { + if (!pgdat) { ret = -ENOMEM; goto out; } From f957db4fcdd8f03e186aa8f041f4049e76ab741c Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Wed, 22 Jun 2011 18:13:04 -0700 Subject: [PATCH 120/224] mm, hotplug: protect zonelist building with zonelists_mutex Commit 959ecc48fc75 ("mm/memory_hotplug.c: fix building of node hotplug zonelist") does not protect the build_all_zonelists() call with zonelists_mutex as needed. This can lead to races in constructing zonelist ordering if a concurrent build is underway. Protecting this with lock_memory_hotplug() is insufficient since zonelists can be rebuild though sysfs as well. Signed-off-by: David Rientjes Reviewed-by: KOSAKI Motohiro Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 78dba9f04b5b..c46887b5a11e 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -498,7 +498,9 @@ static pg_data_t __ref *hotadd_new_pgdat(int nid, u64 start) * The node we allocated has no zone fallback lists. For avoiding * to access not-initialized zonelist, build here. */ + mutex_lock(&zonelists_mutex); build_all_zonelists(NULL); + mutex_unlock(&zonelists_mutex); return pgdat; } From d2a19da79d3ea5b7859248b0f132c479ed4505e2 Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Wed, 22 Jun 2011 09:58:37 +0200 Subject: [PATCH 121/224] ALSA: HDA: Pinfix quirk for HP Z200 Workstation BIOS lists the internal speaker as an internal line-out. Change to internal speaker + model=auto for better auto-mute capabilities. BugLink: http://bugs.launchpad.net/bugs/754964 Reported-by: Marc Legris Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 475ed1e8ffc6..d21191dcfe88 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -12599,6 +12599,7 @@ static const struct hda_verb alc262_toshiba_rx1_unsol_verbs[] = { */ enum { PINFIX_FSC_H270, + PINFIX_HP_Z200, }; static const struct alc_fixup alc262_fixups[] = { @@ -12611,9 +12612,17 @@ static const struct alc_fixup alc262_fixups[] = { { } } }, + [PINFIX_HP_Z200] = { + .type = ALC_FIXUP_PINS, + .v.pins = (const struct alc_pincfg[]) { + { 0x16, 0x99130120 }, /* internal speaker */ + { } + } + }, }; static const struct snd_pci_quirk alc262_fixup_tbl[] = { + SND_PCI_QUIRK(0x103c, 0x170b, "HP Z200", PINFIX_HP_Z200), SND_PCI_QUIRK(0x1734, 0x1147, "FSC Celsius H270", PINFIX_FSC_H270), {} }; @@ -12730,6 +12739,8 @@ static const struct snd_pci_quirk alc262_cfg_tbl[] = { ALC262_HP_BPC), SND_PCI_QUIRK_MASK(0x103c, 0xff00, 0x1500, "HP z series", ALC262_HP_BPC), + SND_PCI_QUIRK(0x103c, 0x170b, "HP Z200", + ALC262_AUTO), SND_PCI_QUIRK_MASK(0x103c, 0xff00, 0x1700, "HP xw series", ALC262_HP_BPC), SND_PCI_QUIRK(0x103c, 0x2800, "HP D7000", ALC262_HP_BPC_D7000_WL), From 0bb04bf3dfdfe1c981087cdfb0d9d772c3a0ba55 Mon Sep 17 00:00:00 2001 From: William Douglas Date: Thu, 23 Jun 2011 13:38:36 +0100 Subject: [PATCH 122/224] mrst_max3110: Change max missing message priority. Change print message to notice instead of error to clean up non critical messages showing on startup. The MAX3111 not being present is a normal path for end user systems. Signed-off-by: William Douglas [rebased on 3.0, switched to dev_dbg()] Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds --- drivers/tty/serial/mrst_max3110.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/mrst_max3110.c b/drivers/tty/serial/mrst_max3110.c index 1bd28450ca40..3be1b0dd6f22 100644 --- a/drivers/tty/serial/mrst_max3110.c +++ b/drivers/tty/serial/mrst_max3110.c @@ -823,7 +823,7 @@ static int __devinit serial_m3110_probe(struct spi_device *spi) res = RC_TAG; ret = max3110_write_then_read(max, (u8 *)&res, (u8 *)&res, 2, 0); if (ret < 0 || res == 0 || res == 0xffff) { - printk(KERN_ERR "MAX3111 deemed not present (conf reg %04x)", + dev_dbg(&spi->dev, "MAX3111 deemed not present (conf reg %04x)", res); ret = -ENODEV; goto err_get_page; From 33b1e6939f5c37ab8e64280fd3d54046607b5c80 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 23 Jun 2011 13:39:00 +0100 Subject: [PATCH 123/224] serial: mrst_max3110: initialize waitqueue earlier The driver went to initialize its waitqueue at the start of the main processing thread. However, it is possible that this thread is not scheduled on a CPU before the write function is called which leads to a following error: BUG: spinlock bad magic on CPU#1, swapper/1 lock: f5f3ebdc, .magic: 00000000, .owner: /-1, .owner_cpu: 0 Pid: 1, comm: swapper Not tainted 3.0.0-rc2+ #67 Call Trace: [] spin_bug+0xa3/0xf0 [] do_raw_spin_lock+0x7d/0x150 [] _raw_spin_lock_irqsave+0x4e/0x60 [] __wake_up+0x1b/0x50 [] serial_m3110_con_write+0x55/0x60 [] __call_console_drivers+0x75/0x90 [] _call_console_drivers+0x49/0x80 [] console_unlock+0xca/0x1f0 [] vprintk+0x18f/0x4f0 [] printk+0x18/0x1a [] register_console+0x2e0/0x350 [] uart_add_one_port+0x33e/0x3d0 [] serial_m3110_probe+0x1c2/0x1df [] spi_drv_probe+0x17/0x20 ... Fix this by initializing the waitqueue before the main thread is created. Signed-off-by: Mika Westerberg Signed-off-by: Alan Cox Signed-off-by: Linus Torvalds --- drivers/tty/serial/mrst_max3110.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/mrst_max3110.c b/drivers/tty/serial/mrst_max3110.c index 3be1b0dd6f22..a764bf99743b 100644 --- a/drivers/tty/serial/mrst_max3110.c +++ b/drivers/tty/serial/mrst_max3110.c @@ -421,7 +421,6 @@ static int max3110_main_thread(void *_max) int ret = 0; struct circ_buf *xmit = &max->con_xmit; - init_waitqueue_head(wq); pr_info(PR_FMT "start main thread\n"); do { @@ -838,6 +837,8 @@ static int __devinit serial_m3110_probe(struct spi_device *spi) max->con_xmit.head = 0; max->con_xmit.tail = 0; + init_waitqueue_head(&max->wq); + max->main_thread = kthread_run(max3110_main_thread, max, "max3110_main"); if (IS_ERR(max->main_thread)) { From 1b19ca9f0bdab7d5035821e1ec8f39df9a6e3ee0 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 22 Jun 2011 11:55:50 +0100 Subject: [PATCH 124/224] Fix CPU spinlock lockups on secondary CPU bringup Secondary CPU bringup typically calls calibrate_delay() during its initialization. However, calibrate_delay() modifies a global variable (loops_per_jiffy) used for udelay() and __delay(). A side effect of 71c696b1 ("calibrate: extract fall-back calculation into own helper") introduced in the 2.6.39 merge window means that we end up with a substantial period where loops_per_jiffy is zero. This causes the spinlock debugging code to malfunction: u64 loops = loops_per_jiffy * HZ; for (;;) { for (i = 0; i < loops; i++) { if (arch_spin_trylock(&lock->raw_lock)) return; __delay(1); } ... } by never calling arch_spin_trylock() - resulting in the CPU locking up in an infinite loop inside __spin_lock_debug(). Work around this by only writing to loops_per_jiffy only once we have completed all the calibration decisions. Tested-by: Santosh Shilimkar Signed-off-by: Russell King Cc: (2.6.39-stable) -- Better solutions (such as omitting the calibration for secondary CPUs, or arranging for calibrate_delay() to return the LPJ value and leave it to the caller to decide where to store it) are a possibility, but would be much more invasive into each architecture. I think this is the best solution for -rc and stable, but it should be revisited for the next merge window. init/calibrate.c | 14 ++++++++------ 1 files changed, 8 insertions(+), 6 deletions(-) Signed-off-by: Linus Torvalds --- init/calibrate.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/init/calibrate.c b/init/calibrate.c index 2568d22a304e..aae2f40fea4c 100644 --- a/init/calibrate.c +++ b/init/calibrate.c @@ -245,30 +245,32 @@ recalibrate: void __cpuinit calibrate_delay(void) { + unsigned long lpj; static bool printed; if (preset_lpj) { - loops_per_jiffy = preset_lpj; + lpj = preset_lpj; if (!printed) pr_info("Calibrating delay loop (skipped) " "preset value.. "); } else if ((!printed) && lpj_fine) { - loops_per_jiffy = lpj_fine; + lpj = lpj_fine; pr_info("Calibrating delay loop (skipped), " "value calculated using timer frequency.. "); - } else if ((loops_per_jiffy = calibrate_delay_direct()) != 0) { + } else if ((lpj = calibrate_delay_direct()) != 0) { if (!printed) pr_info("Calibrating delay using timer " "specific routine.. "); } else { if (!printed) pr_info("Calibrating delay loop... "); - loops_per_jiffy = calibrate_delay_converge(); + lpj = calibrate_delay_converge(); } if (!printed) pr_cont("%lu.%02lu BogoMIPS (lpj=%lu)\n", - loops_per_jiffy/(500000/HZ), - (loops_per_jiffy/(5000/HZ)) % 100, loops_per_jiffy); + lpj/(500000/HZ), + (lpj/(5000/HZ)) % 100, lpj); + loops_per_jiffy = lpj; printed = true; } From 1190f6a067bf27b2ee7e06ec0776a17fe0f6c4d8 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Wed, 22 Jun 2011 17:33:57 -0400 Subject: [PATCH 125/224] cifs: fix wsize negotiation to respect max buffer size and active signing (try #4) Hopefully last version. Base signing check on CAP_UNIX instead of tcon->unix_ext, also clean up the comments a bit more. According to Hongwei Sun's blog posting here: http://blogs.msdn.com/b/openspecification/archive/2009/04/10/smb-maximum-transmit-buffer-size-and-performance-tuning.aspx CAP_LARGE_WRITEX is ignored when signing is active. Also, the maximum size for a write without CAP_LARGE_WRITEX should be the maxBuf that the server sent in the NEGOTIATE request. Fix the wsize negotiation to take this into account. While we're at it, alter the other wsize definitions to use sizeof(WRITE_REQ) to allow for slightly larger amounts of data to potentially be written per request. Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/connect.c | 33 ++++++++++++++++++++------------- 1 file changed, 20 insertions(+), 13 deletions(-) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 19fdbda7aa95..c761935eab8c 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2750,21 +2750,21 @@ void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, /* * When the server supports very large writes via POSIX extensions, we can - * allow up to 2^24 - PAGE_CACHE_SIZE. + * allow up to 2^24-1, minus the size of a WRITE_AND_X header, not including + * the RFC1001 length. * * Note that this might make for "interesting" allocation problems during - * writeback however (as we have to allocate an array of pointers for the - * pages). A 16M write means ~32kb page array with PAGE_CACHE_SIZE == 4096. + * writeback however as we have to allocate an array of pointers for the + * pages. A 16M write means ~32kb page array with PAGE_CACHE_SIZE == 4096. */ -#define CIFS_MAX_WSIZE ((1<<24) - PAGE_CACHE_SIZE) +#define CIFS_MAX_WSIZE ((1<<24) - 1 - sizeof(WRITE_REQ) + 4) /* - * When the server doesn't allow large posix writes, default to a wsize of - * 128k - PAGE_CACHE_SIZE -- one page less than the largest frame size - * described in RFC1001. This allows space for the header without going over - * that by default. + * When the server doesn't allow large posix writes, only allow a wsize of + * 128k minus the size of the WRITE_AND_X header. That allows for a write up + * to the maximum size described by RFC1002. */ -#define CIFS_MAX_RFC1001_WSIZE (128 * 1024 - PAGE_CACHE_SIZE) +#define CIFS_MAX_RFC1002_WSIZE (128 * 1024 - sizeof(WRITE_REQ) + 4) /* * The default wsize is 1M. find_get_pages seems to return a maximum of 256 @@ -2783,11 +2783,18 @@ cifs_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *pvolume_info) /* can server support 24-bit write sizes? (via UNIX extensions) */ if (!tcon->unix_ext || !(unix_cap & CIFS_UNIX_LARGE_WRITE_CAP)) - wsize = min_t(unsigned int, wsize, CIFS_MAX_RFC1001_WSIZE); + wsize = min_t(unsigned int, wsize, CIFS_MAX_RFC1002_WSIZE); - /* no CAP_LARGE_WRITE_X? Limit it to 16 bits */ - if (!(server->capabilities & CAP_LARGE_WRITE_X)) - wsize = min_t(unsigned int, wsize, USHRT_MAX); + /* + * no CAP_LARGE_WRITE_X or is signing enabled without CAP_UNIX set? + * Limit it to max buffer offered by the server, minus the size of the + * WRITEX header, not including the 4 byte RFC1001 length. + */ + if (!(server->capabilities & CAP_LARGE_WRITE_X) || + (!(server->capabilities & CAP_UNIX) && + (server->sec_mode & (SECMODE_SIGN_ENABLED|SECMODE_SIGN_REQUIRED)))) + wsize = min_t(unsigned int, wsize, + server->maxBuf - sizeof(WRITE_REQ) + 4); /* hard limit of CIFS_MAX_WSIZE */ wsize = min_t(unsigned int, wsize, CIFS_MAX_WSIZE); From f6d96e0da1ee3cfe67b719570fba3bb2ea057131 Mon Sep 17 00:00:00 2001 From: "Arnaud Patard (Rtp)" Date: Wed, 22 Jun 2011 22:21:48 +0200 Subject: [PATCH 126/224] ASoC: imx: Remove unused Kconfig SND_MXC_SOC_SSI entry SND_MXC_SOC_SSI looks to be unused, so kill it. Signed-off-by: Arnaud Patard Acked-by: Sascha Hauer Acked-by: Liam Girdwood Signed-off-by: Mark Brown --- sound/soc/imx/Kconfig | 7 ------- 1 file changed, 7 deletions(-) diff --git a/sound/soc/imx/Kconfig b/sound/soc/imx/Kconfig index d8f130d39dd9..bb699bb55a50 100644 --- a/sound/soc/imx/Kconfig +++ b/sound/soc/imx/Kconfig @@ -11,9 +11,6 @@ menuconfig SND_IMX_SOC if SND_IMX_SOC -config SND_MXC_SOC_SSI - tristate - config SND_MXC_SOC_FIQ tristate @@ -24,7 +21,6 @@ config SND_MXC_SOC_WM1133_EV1 tristate "Audio on the the i.MX31ADS with WM1133-EV1 fitted" depends on MACH_MX31ADS_WM1133_EV1 && EXPERIMENTAL select SND_SOC_WM8350 - select SND_MXC_SOC_SSI select SND_MXC_SOC_FIQ help Enable support for audio on the i.MX31ADS with the WM1133-EV1 @@ -34,7 +30,6 @@ config SND_SOC_MX27VIS_AIC32X4 tristate "SoC audio support for Visstrim M10 boards" depends on MACH_IMX27_VISSTRIM_M10 select SND_SOC_TVL320AIC32X4 - select SND_MXC_SOC_SSI select SND_MXC_SOC_MX2 help Say Y if you want to add support for SoC audio on Visstrim SM10 @@ -44,7 +39,6 @@ config SND_SOC_PHYCORE_AC97 tristate "SoC Audio support for Phytec phyCORE (and phyCARD) boards" depends on MACH_PCM043 || MACH_PCA100 select SND_SOC_WM9712 - select SND_MXC_SOC_SSI select SND_MXC_SOC_FIQ help Say Y if you want to add support for SoC audio on Phytec phyCORE @@ -57,7 +51,6 @@ config SND_SOC_EUKREA_TLV320 || MACH_EUKREA_MBIMXSD35_BASEBOARD \ || MACH_EUKREA_MBIMXSD51_BASEBOARD select SND_SOC_TLV320AIC23 - select SND_MXC_SOC_SSI select SND_MXC_SOC_FIQ help Enable I2S based access to the TLV320AIC23B codec attached From 96dcabb99b9f63f2b65f2b0bfe5d4eb48f11b177 Mon Sep 17 00:00:00 2001 From: "Arnaud Patard (Rtp)" Date: Wed, 22 Jun 2011 22:21:49 +0200 Subject: [PATCH 127/224] ASoC: imx: add missing module informations - add some modules aliases - add module license to avoid tainted kernel when loading the imx-pcm-audio driver Signed-off-by: Arnaud Patard Acked-by: Sascha Hauer Acked-by: Liam Girdwood Signed-off-by: Mark Brown --- sound/soc/imx/imx-pcm-dma-mx2.c | 2 ++ sound/soc/imx/imx-ssi.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/sound/soc/imx/imx-pcm-dma-mx2.c b/sound/soc/imx/imx-pcm-dma-mx2.c index aab7765f401a..4173b3d87f97 100644 --- a/sound/soc/imx/imx-pcm-dma-mx2.c +++ b/sound/soc/imx/imx-pcm-dma-mx2.c @@ -337,3 +337,5 @@ static void __exit snd_imx_pcm_exit(void) platform_driver_unregister(&imx_pcm_driver); } module_exit(snd_imx_pcm_exit); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("platform:imx-pcm-audio"); diff --git a/sound/soc/imx/imx-ssi.c b/sound/soc/imx/imx-ssi.c index 5b13feca7537..61fceb09cdb5 100644 --- a/sound/soc/imx/imx-ssi.c +++ b/sound/soc/imx/imx-ssi.c @@ -774,4 +774,4 @@ module_exit(imx_ssi_exit); MODULE_AUTHOR("Sascha Hauer, "); MODULE_DESCRIPTION("i.MX I2S/ac97 SoC Interface"); MODULE_LICENSE("GPL"); - +MODULE_ALIAS("platform:imx-ssi"); From 53dea36c70c1857149a8c447224e3936eb8b5339 Mon Sep 17 00:00:00 2001 From: Daniel Mack Date: Wed, 22 Jun 2011 20:48:25 +0200 Subject: [PATCH 128/224] ASoC: pxa-ssp: Correct check for stream presence Don't rely on the codec's channels_min information to decide wheter or not allocate a substream's DMA buffer. Rather check if the substream itself was allocated previously. Signed-off-by: Daniel Mack Acked-by: Liam Girdwood Signed-off-by: Mark Brown Cc: stable@kernel.org --- sound/soc/pxa/pxa2xx-pcm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/pxa/pxa2xx-pcm.c b/sound/soc/pxa/pxa2xx-pcm.c index 2ce0b2d891d5..fab20a54e863 100644 --- a/sound/soc/pxa/pxa2xx-pcm.c +++ b/sound/soc/pxa/pxa2xx-pcm.c @@ -95,14 +95,14 @@ static int pxa2xx_soc_pcm_new(struct snd_card *card, struct snd_soc_dai *dai, if (!card->dev->coherent_dma_mask) card->dev->coherent_dma_mask = DMA_BIT_MASK(32); - if (dai->driver->playback.channels_min) { + if (pcm->streams[SNDRV_PCM_STREAM_PLAYBACK].substream) { ret = pxa2xx_pcm_preallocate_dma_buffer(pcm, SNDRV_PCM_STREAM_PLAYBACK); if (ret) goto out; } - if (dai->driver->capture.channels_min) { + if (pcm->streams[SNDRV_PCM_STREAM_CAPTURE].substream) { ret = pxa2xx_pcm_preallocate_dma_buffer(pcm, SNDRV_PCM_STREAM_CAPTURE); if (ret) From 8618ccd352dcd01628f39eb1fca4f9a7bc077ea1 Mon Sep 17 00:00:00 2001 From: "Justin P. Mattock" Date: Wed, 22 Jun 2011 23:29:20 -0700 Subject: [PATCH 129/224] drivers/ata/sata_dwc_460ex: Fix typo 'corrresponding' The patch below fixes a typo. Signed-off-by: Justin P. Mattock Signed-off-by: Jeff Garzik --- drivers/ata/sata_dwc_460ex.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/sata_dwc_460ex.c b/drivers/ata/sata_dwc_460ex.c index 1c4b3aa4c7c4..dc88a39e7db8 100644 --- a/drivers/ata/sata_dwc_460ex.c +++ b/drivers/ata/sata_dwc_460ex.c @@ -389,7 +389,7 @@ static void sata_dwc_tf_dump(struct ata_taskfile *tf) /* * Function: get_burst_length_encode * arguments: datalength: length in bytes of data - * returns value to be programmed in register corrresponding to data length + * returns value to be programmed in register corresponding to data length * This value is effectively the log(base 2) of the length */ static int get_burst_length_encode(int datalength) From cd691876d73e24b4c0a2e96993251abbe3a320df Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 22 Jun 2011 12:13:10 +0200 Subject: [PATCH 130/224] libata: apply NOSETXFER horkage to the affected Pioneer drives regardless of firmware revision It's unlikely NOSETXFER works for a revision of drive but doesn't for another and pioneer doesn't seem to be fixing firmwares for the affected drives. Apply NOSETXFER to the affected pioneer drives regardless of firmware revision. http://article.gmane.org/gmane.linux.ide/49734 Signed-off-by: Tejun Heo Reported-by: fl-00@gmx.de Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 736bee5dafeb..000d03ae6653 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4143,9 +4143,9 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = { * Devices which choke on SETXFER. Applies only if both the * device and controller are SATA. */ - { "PIONEER DVD-RW DVRTD08", "1.00", ATA_HORKAGE_NOSETXFER }, - { "PIONEER DVD-RW DVR-212D", "1.28", ATA_HORKAGE_NOSETXFER }, - { "PIONEER DVD-RW DVR-216D", "1.08", ATA_HORKAGE_NOSETXFER }, + { "PIONEER DVD-RW DVRTD08", NULL, ATA_HORKAGE_NOSETXFER }, + { "PIONEER DVD-RW DVR-212D", NULL, ATA_HORKAGE_NOSETXFER }, + { "PIONEER DVD-RW DVR-216D", NULL, ATA_HORKAGE_NOSETXFER }, /* End Marker */ { } From 3f1e046ad3370d22d39529103667354eb50abc08 Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Thu, 16 Jun 2011 08:28:36 -0700 Subject: [PATCH 131/224] libata/sas: only set FROZEN flag if new EH is supported On 16.06.2011 [08:28:39 -0500], Brian King wrote: > On 06/16/2011 02:51 AM, Tejun Heo wrote: > > On Wed, Jun 15, 2011 at 04:34:17PM -0700, Nishanth Aravamudan wrote: > >>> That looks like the right thing to do. For ipr's usage of > >>> libata, we don't have the concept of a port frozen state, so this flag > >>> should really never get set. The alternate way to fix this would be to > >>> only set ATA_PFLAG_FROZEN in ata_port_alloc if ap->ops->error_handler > >>> is not NULL. > >> > >> It seemed like ipr is as you say, but I wasn't sure if it was > >> appropriate to make the change above in the common libata-scis code or > >> not. I don't want to break some other device on accident. > >> > >> Also, I tried your suggestion, but I don't think that can happen in > >> ata_port_alloc? ata_port_alloc is allocated ap itself, and it seems like > >> ap->ops typically gets set only after ata_port_alloc returns? > > > > Maybe we can test error_handler in ata_sas_port_start()? > > Good point. Since libsas is converted to the new eh now, we would need to have > this test. Commit 7b3a24c57d2eeda8dba9c205342b12689c4679f9 ("ahci: don't enable port irq before handler is registered") caused a regression for CD-ROMs attached to the IPR SATA bus on Power machines: ata_port_alloc: ENTER ata_port_probe: ata1: bus probe begin ata1.00: ata_dev_read_id: ENTER ata1.00: failed to IDENTIFY (I/O error, err_mask=0x40) ata1.00: ata_dev_read_id: ENTER ata1.00: failed to IDENTIFY (I/O error, err_mask=0x40) ata1.00: limiting speed to UDMA7:PIO5 ata1.00: ata_dev_read_id: ENTER ata1.00: failed to IDENTIFY (I/O error, err_mask=0x40) ata1.00: disabled ata_port_probe: ata1: bus probe end scsi_alloc_sdev: Allocation failure during SCSI scanning, some SCSI devices might not be configured The FROZEN flag added in that commit is only cleared by the new EH code, which is not used by ipr. Clear this flag in the SAS code if we don't support new EH. Reported-by: Benjamin Herrenschmidt Signed-off-by: Nishanth Aravamudan Signed-off-by: Jeff Garzik --- drivers/ata/libata-scsi.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index d51f9795c064..927f968e99d9 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -3797,6 +3797,12 @@ EXPORT_SYMBOL_GPL(ata_sas_port_alloc); */ int ata_sas_port_start(struct ata_port *ap) { + /* + * the port is marked as frozen at allocation time, but if we don't + * have new eh, we won't thaw it + */ + if (!ap->ops->error_handler) + ap->pflags &= ~ATA_PFLAG_FROZEN; return 0; } EXPORT_SYMBOL_GPL(ata_sas_port_start); From 03ad2d501e7189245bf6bf9dfc2a30511dd50602 Mon Sep 17 00:00:00 2001 From: Alexey Charkov Date: Wed, 22 Jun 2011 23:20:13 +0200 Subject: [PATCH 132/224] rtc: vt8500: Fix build error & cleanup rtc_class_ops->update_irq_enable() Now that the generic code handles UIE mode irqs via periodic alarm interrupts, no one calls the rtc_class_ops->update_irq_enable() method anymore. Further the rtc_class_ops doesn't have a update_irq_enable element anymore, so this causes a build error. This patch removes the driver hooks and implementations of update_irq_enable and the associated setup. [wsa: updated commit-message and removed update_irq_enable-function, too] [jstultz: improve commit message, clarifying build issue] Signed-off-by: Alexey Charkov Signed-off-by: Wolfram Sang Signed-off-by: John Stultz --- drivers/rtc/rtc-vt8500.c | 45 +++------------------------------------- 1 file changed, 3 insertions(+), 42 deletions(-) diff --git a/drivers/rtc/rtc-vt8500.c b/drivers/rtc/rtc-vt8500.c index b8bc862903ae..efd6066b5cd2 100644 --- a/drivers/rtc/rtc-vt8500.c +++ b/drivers/rtc/rtc-vt8500.c @@ -78,7 +78,6 @@ struct vt8500_rtc { void __iomem *regbase; struct resource *res; int irq_alarm; - int irq_hz; struct rtc_device *rtc; spinlock_t lock; /* Protects this structure */ }; @@ -100,10 +99,6 @@ static irqreturn_t vt8500_rtc_irq(int irq, void *dev_id) if (isr & 1) events |= RTC_AF | RTC_IRQF; - /* Only second/minute interrupts are supported */ - if (isr & 2) - events |= RTC_UF | RTC_IRQF; - rtc_update_irq(vt8500_rtc->rtc, 1, events); return IRQ_HANDLED; @@ -199,27 +194,12 @@ static int vt8500_alarm_irq_enable(struct device *dev, unsigned int enabled) return 0; } -static int vt8500_update_irq_enable(struct device *dev, unsigned int enabled) -{ - struct vt8500_rtc *vt8500_rtc = dev_get_drvdata(dev); - unsigned long tmp = readl(vt8500_rtc->regbase + VT8500_RTC_CR); - - if (enabled) - tmp |= VT8500_RTC_CR_SM_SEC | VT8500_RTC_CR_SM_ENABLE; - else - tmp &= ~VT8500_RTC_CR_SM_ENABLE; - - writel(tmp, vt8500_rtc->regbase + VT8500_RTC_CR); - return 0; -} - static const struct rtc_class_ops vt8500_rtc_ops = { .read_time = vt8500_rtc_read_time, .set_time = vt8500_rtc_set_time, .read_alarm = vt8500_rtc_read_alarm, .set_alarm = vt8500_rtc_set_alarm, .alarm_irq_enable = vt8500_alarm_irq_enable, - .update_irq_enable = vt8500_update_irq_enable, }; static int __devinit vt8500_rtc_probe(struct platform_device *pdev) @@ -248,13 +228,6 @@ static int __devinit vt8500_rtc_probe(struct platform_device *pdev) goto err_free; } - vt8500_rtc->irq_hz = platform_get_irq(pdev, 1); - if (vt8500_rtc->irq_hz < 0) { - dev_err(&pdev->dev, "No 1Hz IRQ resource defined\n"); - ret = -ENXIO; - goto err_free; - } - vt8500_rtc->res = request_mem_region(vt8500_rtc->res->start, resource_size(vt8500_rtc->res), "vt8500-rtc"); @@ -272,9 +245,8 @@ static int __devinit vt8500_rtc_probe(struct platform_device *pdev) goto err_release; } - /* Enable the second/minute interrupt generation and enable RTC */ - writel(VT8500_RTC_CR_ENABLE | VT8500_RTC_CR_24H - | VT8500_RTC_CR_SM_ENABLE | VT8500_RTC_CR_SM_SEC, + /* Enable RTC and set it to 24-hour mode */ + writel(VT8500_RTC_CR_ENABLE | VT8500_RTC_CR_24H, vt8500_rtc->regbase + VT8500_RTC_CR); vt8500_rtc->rtc = rtc_device_register("vt8500-rtc", &pdev->dev, @@ -286,26 +258,16 @@ static int __devinit vt8500_rtc_probe(struct platform_device *pdev) goto err_unmap; } - ret = request_irq(vt8500_rtc->irq_hz, vt8500_rtc_irq, 0, - "rtc 1Hz", vt8500_rtc); - if (ret < 0) { - dev_err(&pdev->dev, "can't get irq %i, err %d\n", - vt8500_rtc->irq_hz, ret); - goto err_unreg; - } - ret = request_irq(vt8500_rtc->irq_alarm, vt8500_rtc_irq, 0, "rtc alarm", vt8500_rtc); if (ret < 0) { dev_err(&pdev->dev, "can't get irq %i, err %d\n", vt8500_rtc->irq_alarm, ret); - goto err_free_hz; + goto err_unreg; } return 0; -err_free_hz: - free_irq(vt8500_rtc->irq_hz, vt8500_rtc); err_unreg: rtc_device_unregister(vt8500_rtc->rtc); err_unmap: @@ -323,7 +285,6 @@ static int __devexit vt8500_rtc_remove(struct platform_device *pdev) struct vt8500_rtc *vt8500_rtc = platform_get_drvdata(pdev); free_irq(vt8500_rtc->irq_alarm, vt8500_rtc); - free_irq(vt8500_rtc->irq_hz, vt8500_rtc); rtc_device_unregister(vt8500_rtc->rtc); From 7fd29aa920273b70be50c14c4b7e2213fb6623ce Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 23 Jun 2011 23:48:32 +0000 Subject: [PATCH 133/224] target: Fix transport_get_lun_for_tmr failure cases This patch fixes two possible NULL pointer dereferences in target v4.0 code where se_tmr release path in core_tmr_release_req() can OOPs upon transport_get_lun_for_tmr() failure by attempting to access se_device or se_tmr->tmr_list without a valid member of se_device->tmr_list during transport_free_se_cmd() release. This patch moves the se_tmr->tmr_dev pointer assignment in transport_get_lun_for_tmr() until after possible -ENODEV failures during unpacked_lun lookup. This addresses an OOPs originally reported with LIO v4.1 upstream on .39 code here: TARGET_CORE[qla2xxx]: Detected NON_EXISTENT_LUN Access for 0x00000000 BUG: unable to handle kernel NULL pointer dereference at 0000000000000550 IP: [] __ticket_spin_trylock+0x4/0x20 PGD 0 Oops: 0000 [#1] SMP last sysfs file: /sys/devices/system/cpu/cpu23/cache/index2/shared_cpu_map CPU 1 Modules linked in: netconsole target_core_pscsi target_core_file tcm_qla2xxx target_core_iblock tcm_loop target_core_mod configfs ipmi_devintf ipmi_si ipmi_msghandler serio_raw i7core_edac ioatdma dca edac_core ps_bdrv ses enclosure usbhid usb_storage ahci qla2xxx hid uas e1000e mpt2sas libahci mlx4_core scsi_transport_fc scsi_transport_sas raid_class scsi_tgt [last unloaded: netconsole] Pid: 0, comm: kworker/0:0 Tainted: G W 2.6.39+ #1 Xyratex Storage Server RIP: 0010:[] []__ticket_spin_trylock+0x4/0x20 RSP: 0018:ffff88063e803c08 EFLAGS: 00010286 RAX: ffff880619ab45e0 RBX: 0000000000000550 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000550 RBP: ffff88063e803c08 R08: 0000000000000002 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000568 R13: 0000000000000001 R14: 0000000000000000 R15: ffff88060cd96a20 FS: 0000000000000000(0000) GS:ffff88063e800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000550 CR3: 0000000001a03000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process kworker/0:0 (pid: 0, threadinfo ffff880619ab8000, task ffff880619ab45e0) Stack: ffff88063e803c28 ffffffff812cf039 0000000000000550 0000000000000568 ffff88063e803c58 ffffffff8157071e ffffffffa028a1dc ffff88060f7e4600 0000000000000550 ffff880616961480 ffff88063e803c78 ffffffffa028a1dc Call Trace: [] do_raw_spin_trylock+0x19/0x50 [] _raw_spin_lock+0x3e/0x70 [] ? core_tmr_release_req+0x2c/0x60 [target_core_mod] [] core_tmr_release_req+0x2c/0x60 [target_core_mod] [] transport_free_se_cmd+0x22/0x50 [target_core_mod] [] transport_release_cmd_to_pool+0x20/0x40 [target_core_mod] [] transport_generic_free_cmd+0xa5/0xb0 [target_core_mod] [] tcm_qla2xxx_handle_tmr+0xc4/0xd0 [tcm_qla2xxx] [] __qla24xx_handle_abts+0xd3/0x150 [qla2xxx] [] qla_tgt_response_pkt+0x171/0x520 [qla2xxx] [] qla_tgt_response_pkt_all_vps+0x2d/0x220 [qla2xxx] [] qla24xx_process_response_queue+0x1a3/0x670 [qla2xxx] [] ? qla24xx_atio_pkt+0x81/0x120 [qla2xxx] [] ? qla24xx_msix_default+0x45/0x2a0 [qla2xxx] [] qla24xx_msix_default+0x1b8/0x2a0 [qla2xxx] [] handle_irq_event_percpu+0x54/0x210 [] handle_irq_event+0x48/0x70 [] ? handle_edge_irq+0x1e/0x110 [] handle_edge_irq+0x77/0x110 [] handle_irq+0x22/0x40 [] do_IRQ+0x5d/0xe0 [] common_interrupt+0x13/0x13 [] ? intel_idle+0xd7/0x130 [] ? intel_idle+0xd0/0x130 [] cpuidle_idle_call+0xab/0x1c0 [] cpu_idle+0xab/0xf0 [] start_secondary+0x1cb/0x1d2 Reported-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_device.c | 3 ++- drivers/target/target_core_tmr.c | 8 +++++++- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index 8407f9ca2b31..f8d8af7a0d41 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -192,7 +192,7 @@ int transport_get_lun_for_tmr( &SE_NODE_ACL(se_sess)->device_list[unpacked_lun]; if (deve->lun_flags & TRANSPORT_LUNFLAGS_INITIATOR_ACCESS) { se_lun = se_cmd->se_lun = se_tmr->tmr_lun = deve->se_lun; - dev = se_tmr->tmr_dev = se_lun->lun_se_dev; + dev = se_lun->lun_se_dev; se_cmd->pr_res_key = deve->pr_res_key; se_cmd->orig_fe_lun = unpacked_lun; se_cmd->se_orig_obj_ptr = SE_LUN(se_cmd)->lun_se_dev; @@ -216,6 +216,7 @@ int transport_get_lun_for_tmr( se_cmd->se_cmd_flags |= SCF_SCSI_CDB_EXCEPTION; return -1; } + se_tmr->tmr_dev = dev; spin_lock(&dev->se_tmr_lock); list_add_tail(&se_tmr->tmr_list, &dev->dev_tmr_list); diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c index 59b8b9c5ad72..179063d81cdd 100644 --- a/drivers/target/target_core_tmr.c +++ b/drivers/target/target_core_tmr.c @@ -75,10 +75,16 @@ void core_tmr_release_req( { struct se_device *dev = tmr->tmr_dev; + if (!dev) { + kmem_cache_free(se_tmr_req_cache, tmr); + return; + } + spin_lock(&dev->se_tmr_lock); list_del(&tmr->tmr_list); - kmem_cache_free(se_tmr_req_cache, tmr); spin_unlock(&dev->se_tmr_lock); + + kmem_cache_free(se_tmr_req_cache, tmr); } static void core_tmr_handle_tas_abort( From 233888644d80cc44330062e5e978c9e3a14c9cb9 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 22 Jun 2011 01:02:21 -0700 Subject: [PATCH 134/224] target: Convert transport_deregister_session_configfs nacl_sess_lock to save irq state This patch converts transport_deregister_session_configfs() to save/restore spinlock IRQ state for struct se_node_acl->nacl_sess_lock access as tcm_qla2xxx logic expects to call transport_deregister_session_configfs() code with irq save already held for struct qla_hw_data. Reported-by: Roland Dreier Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_transport.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c index 4dafeb8b5638..4b9b7169bdd9 100644 --- a/drivers/target/target_core_transport.c +++ b/drivers/target/target_core_transport.c @@ -536,13 +536,13 @@ EXPORT_SYMBOL(transport_register_session); void transport_deregister_session_configfs(struct se_session *se_sess) { struct se_node_acl *se_nacl; - + unsigned long flags; /* * Used by struct se_node_acl's under ConfigFS to locate active struct se_session */ se_nacl = se_sess->se_node_acl; if ((se_nacl)) { - spin_lock_irq(&se_nacl->nacl_sess_lock); + spin_lock_irqsave(&se_nacl->nacl_sess_lock, flags); list_del(&se_sess->sess_acl_list); /* * If the session list is empty, then clear the pointer. @@ -556,7 +556,7 @@ void transport_deregister_session_configfs(struct se_session *se_sess) se_nacl->acl_sess_list.prev, struct se_session, sess_acl_list); } - spin_unlock_irq(&se_nacl->nacl_sess_lock); + spin_unlock_irqrestore(&se_nacl->nacl_sess_lock, flags); } } EXPORT_SYMBOL(transport_deregister_session_configfs); From 552523dcbf0f33d44d816da310be8227a2c1502a Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 15 Jun 2011 09:41:33 -0700 Subject: [PATCH 135/224] target: Fix ERR_PTR dereferencing bugs transport_init_session() and core_tmr_alloc_req() never return NULL, they only return ERR_PTRs on error. v2: Fix patch to return PTR_ERR(tl_nexus->se_sess) from Ankit Jain's feedback. Signed-off-by: Dan Carpenter Signed-off-by: Ankit Jain Signed-off-by: Nicholas Bellinger --- drivers/target/loopback/tcm_loop.c | 9 ++++++--- drivers/target/tcm_fc/tfc_sess.c | 2 +- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c index dee2a2c909f5..d4fee2a9d66a 100644 --- a/drivers/target/loopback/tcm_loop.c +++ b/drivers/target/loopback/tcm_loop.c @@ -386,7 +386,7 @@ static int tcm_loop_device_reset(struct scsi_cmnd *sc) */ se_cmd->se_tmr_req = core_tmr_alloc_req(se_cmd, (void *)tl_tmr, TMR_LUN_RESET); - if (!se_cmd->se_tmr_req) + if (IS_ERR(se_cmd->se_tmr_req)) goto release; /* * Locate the underlying TCM struct se_lun from sc->device->lun @@ -1017,6 +1017,7 @@ static int tcm_loop_make_nexus( struct se_portal_group *se_tpg; struct tcm_loop_hba *tl_hba = tl_tpg->tl_hba; struct tcm_loop_nexus *tl_nexus; + int ret = -ENOMEM; if (tl_tpg->tl_hba->tl_nexus) { printk(KERN_INFO "tl_tpg->tl_hba->tl_nexus already exists\n"); @@ -1033,8 +1034,10 @@ static int tcm_loop_make_nexus( * Initialize the struct se_session pointer */ tl_nexus->se_sess = transport_init_session(); - if (!tl_nexus->se_sess) + if (IS_ERR(tl_nexus->se_sess)) { + ret = PTR_ERR(tl_nexus->se_sess); goto out; + } /* * Since we are running in 'demo mode' this call with generate a * struct se_node_acl for the tcm_loop struct se_portal_group with the SCSI @@ -1060,7 +1063,7 @@ static int tcm_loop_make_nexus( out: kfree(tl_nexus); - return -ENOMEM; + return ret; } static int tcm_loop_drop_nexus( diff --git a/drivers/target/tcm_fc/tfc_sess.c b/drivers/target/tcm_fc/tfc_sess.c index a3bd57f2ea32..03744c23095c 100644 --- a/drivers/target/tcm_fc/tfc_sess.c +++ b/drivers/target/tcm_fc/tfc_sess.c @@ -229,7 +229,7 @@ static struct ft_sess *ft_sess_create(struct ft_tport *tport, u32 port_id, return NULL; sess->se_sess = transport_init_session(); - if (!sess->se_sess) { + if (IS_ERR(sess->se_sess)) { kfree(sess); return NULL; } From 5eff5be0b1993f4291f2b8c6d035b408010f96c5 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 13 Jun 2011 23:10:49 +0300 Subject: [PATCH 136/224] target: Drop bogus ERR_PTR usage in target_fabric_configfs_init In the original code, there were several places inside the target_fabric_configfs_init() function that returned NULL on error and one place the returned an ERR_PTR. There are two places that call this function and they only check for NULL returns; they don't check for ERR_PTRs. So I've changed the ERR_PTR so now the function only returns NULL on error. Signed-off-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/target_core_configfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index ee6fad979b50..b17abd13c8e3 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -312,7 +312,7 @@ struct target_fabric_configfs *target_fabric_configfs_init( tf = kzalloc(sizeof(struct target_fabric_configfs), GFP_KERNEL); if (!(tf)) - return ERR_PTR(-ENOMEM); + return NULL; INIT_LIST_HEAD(&tf->tf_list); atomic_set(&tf->tf_access_cnt, 0); From 60d645a4e9e7e7ddc20e534fea82aa4e6947f911 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 15 Jun 2011 10:03:05 -0700 Subject: [PATCH 137/224] target: Fix incorrect strlen() NULL terminator checks This patch fixes a number of cases in target core using an incorrectly if (strlen(foo) > SOME_MAX_SIZE) As strlen() returns the number of characters in the string not counting the NULL character at the end. So if you do something like: char buf[10]; if (strlen("0123456789") > 10) return -ETOOLONG; snprintf(buf, 10, "0123456789"); printf("%s\n", buf); then the last "9" gets chopped off and only "012345678" is printed. Plus I threw in one small related cleanup. Signed-off-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/loopback/tcm_loop.c | 4 ++-- drivers/target/target_core_configfs.c | 22 +++++++++++----------- drivers/target/target_core_device.c | 2 +- drivers/target/target_core_pr.c | 6 +++--- 4 files changed, 17 insertions(+), 17 deletions(-) diff --git a/drivers/target/loopback/tcm_loop.c b/drivers/target/loopback/tcm_loop.c index d4fee2a9d66a..70c2e7fa6664 100644 --- a/drivers/target/loopback/tcm_loop.c +++ b/drivers/target/loopback/tcm_loop.c @@ -1143,7 +1143,7 @@ static ssize_t tcm_loop_tpg_store_nexus( * the fabric protocol_id set in tcm_loop_make_scsi_hba(), and call * tcm_loop_make_nexus() */ - if (strlen(page) > TL_WWN_ADDR_LEN) { + if (strlen(page) >= TL_WWN_ADDR_LEN) { printk(KERN_ERR "Emulated NAA Sas Address: %s, exceeds" " max: %d\n", page, TL_WWN_ADDR_LEN); return -EINVAL; @@ -1324,7 +1324,7 @@ struct se_wwn *tcm_loop_make_scsi_hba( return ERR_PTR(-EINVAL); check_len: - if (strlen(name) > TL_WWN_ADDR_LEN) { + if (strlen(name) >= TL_WWN_ADDR_LEN) { printk(KERN_ERR "Emulated NAA %s Address: %s, exceeds" " max: %d\n", name, tcm_loop_dump_proto_id(tl_hba), TL_WWN_ADDR_LEN); diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c index b17abd13c8e3..25c1f49a7d8b 100644 --- a/drivers/target/target_core_configfs.c +++ b/drivers/target/target_core_configfs.c @@ -304,7 +304,7 @@ struct target_fabric_configfs *target_fabric_configfs_init( printk(KERN_ERR "Unable to locate passed fabric name\n"); return NULL; } - if (strlen(name) > TARGET_FABRIC_NAME_SIZE) { + if (strlen(name) >= TARGET_FABRIC_NAME_SIZE) { printk(KERN_ERR "Passed name: %s exceeds TARGET_FABRIC" "_NAME_SIZE\n", name); return NULL; @@ -851,7 +851,7 @@ static ssize_t target_core_dev_wwn_store_attr_vpd_unit_serial( return -EOPNOTSUPP; } - if ((strlen(page) + 1) > INQUIRY_VPD_SERIAL_LEN) { + if (strlen(page) >= INQUIRY_VPD_SERIAL_LEN) { printk(KERN_ERR "Emulated VPD Unit Serial exceeds" " INQUIRY_VPD_SERIAL_LEN: %d\n", INQUIRY_VPD_SERIAL_LEN); return -EOVERFLOW; @@ -917,7 +917,7 @@ static ssize_t target_core_dev_wwn_show_attr_vpd_protocol_identifier( transport_dump_vpd_proto_id(vpd, buf, VPD_TMP_BUF_SIZE); - if ((len + strlen(buf) > PAGE_SIZE)) + if ((len + strlen(buf) >= PAGE_SIZE)) break; len += sprintf(page+len, "%s", buf); @@ -962,19 +962,19 @@ static ssize_t target_core_dev_wwn_show_attr_##_name( \ \ memset(buf, 0, VPD_TMP_BUF_SIZE); \ transport_dump_vpd_assoc(vpd, buf, VPD_TMP_BUF_SIZE); \ - if ((len + strlen(buf) > PAGE_SIZE)) \ + if ((len + strlen(buf) >= PAGE_SIZE)) \ break; \ len += sprintf(page+len, "%s", buf); \ \ memset(buf, 0, VPD_TMP_BUF_SIZE); \ transport_dump_vpd_ident_type(vpd, buf, VPD_TMP_BUF_SIZE); \ - if ((len + strlen(buf) > PAGE_SIZE)) \ + if ((len + strlen(buf) >= PAGE_SIZE)) \ break; \ len += sprintf(page+len, "%s", buf); \ \ memset(buf, 0, VPD_TMP_BUF_SIZE); \ transport_dump_vpd_ident(vpd, buf, VPD_TMP_BUF_SIZE); \ - if ((len + strlen(buf) > PAGE_SIZE)) \ + if ((len + strlen(buf) >= PAGE_SIZE)) \ break; \ len += sprintf(page+len, "%s", buf); \ } \ @@ -1299,7 +1299,7 @@ static ssize_t target_core_dev_pr_show_attr_res_pr_registered_i_pts( &i_buf[0] : "", pr_reg->pr_res_key, pr_reg->pr_res_generation); - if ((len + strlen(buf) > PAGE_SIZE)) + if ((len + strlen(buf) >= PAGE_SIZE)) break; len += sprintf(page+len, "%s", buf); @@ -1496,7 +1496,7 @@ static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata( ret = -ENOMEM; goto out; } - if (strlen(i_port) > PR_APTPL_MAX_IPORT_LEN) { + if (strlen(i_port) >= PR_APTPL_MAX_IPORT_LEN) { printk(KERN_ERR "APTPL metadata initiator_node=" " exceeds PR_APTPL_MAX_IPORT_LEN: %d\n", PR_APTPL_MAX_IPORT_LEN); @@ -1510,7 +1510,7 @@ static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata( ret = -ENOMEM; goto out; } - if (strlen(isid) > PR_REG_ISID_LEN) { + if (strlen(isid) >= PR_REG_ISID_LEN) { printk(KERN_ERR "APTPL metadata initiator_isid" "= exceeds PR_REG_ISID_LEN: %d\n", PR_REG_ISID_LEN); @@ -1571,7 +1571,7 @@ static ssize_t target_core_dev_pr_store_attr_res_aptpl_metadata( ret = -ENOMEM; goto out; } - if (strlen(t_port) > PR_APTPL_MAX_TPORT_LEN) { + if (strlen(t_port) >= PR_APTPL_MAX_TPORT_LEN) { printk(KERN_ERR "APTPL metadata target_node=" " exceeds PR_APTPL_MAX_TPORT_LEN: %d\n", PR_APTPL_MAX_TPORT_LEN); @@ -3052,7 +3052,7 @@ static struct config_group *target_core_call_addhbatotarget( int ret; memset(buf, 0, TARGET_CORE_NAME_MAX_LEN); - if (strlen(name) > TARGET_CORE_NAME_MAX_LEN) { + if (strlen(name) >= TARGET_CORE_NAME_MAX_LEN) { printk(KERN_ERR "Passed *name strlen(): %d exceeds" " TARGET_CORE_NAME_MAX_LEN: %d\n", (int)strlen(name), TARGET_CORE_NAME_MAX_LEN); diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c index f8d8af7a0d41..ba698ea62bb2 100644 --- a/drivers/target/target_core_device.c +++ b/drivers/target/target_core_device.c @@ -1431,7 +1431,7 @@ struct se_lun_acl *core_dev_init_initiator_node_lun_acl( struct se_lun_acl *lacl; struct se_node_acl *nacl; - if (strlen(initiatorname) > TRANSPORT_IQN_LEN) { + if (strlen(initiatorname) >= TRANSPORT_IQN_LEN) { printk(KERN_ERR "%s InitiatorName exceeds maximum size.\n", TPG_TFO(tpg)->get_fabric_name()); *ret = -EOVERFLOW; diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c index a79f518ca6e2..b662db3a320b 100644 --- a/drivers/target/target_core_pr.c +++ b/drivers/target/target_core_pr.c @@ -1916,7 +1916,7 @@ static int __core_scsi3_update_aptpl_buf( pr_reg->pr_res_mapped_lun); } - if ((len + strlen(tmp) > pr_aptpl_buf_len)) { + if ((len + strlen(tmp) >= pr_aptpl_buf_len)) { printk(KERN_ERR "Unable to update renaming" " APTPL metadata\n"); spin_unlock(&T10_RES(su_dev)->registration_lock); @@ -1934,7 +1934,7 @@ static int __core_scsi3_update_aptpl_buf( TPG_TFO(tpg)->tpg_get_tag(tpg), lun->lun_sep->sep_rtpi, lun->unpacked_lun, reg_count); - if ((len + strlen(tmp) > pr_aptpl_buf_len)) { + if ((len + strlen(tmp) >= pr_aptpl_buf_len)) { printk(KERN_ERR "Unable to update renaming" " APTPL metadata\n"); spin_unlock(&T10_RES(su_dev)->registration_lock); @@ -1986,7 +1986,7 @@ static int __core_scsi3_write_aptpl_to_file( memset(iov, 0, sizeof(struct iovec)); memset(path, 0, 512); - if (strlen(&wwn->unit_serial[0]) > 512) { + if (strlen(&wwn->unit_serial[0]) >= 512) { printk(KERN_ERR "WWN value for struct se_device does not fit" " into path buffer\n"); return -1; From 61db952713a8bc1b18515db3f2eac354ec8990bd Mon Sep 17 00:00:00 2001 From: Kiran Patil Date: Wed, 22 Jun 2011 16:30:22 -0700 Subject: [PATCH 138/224] tcm_fc: Fix ft_send_tm LUN lookup OOPs This patch fixes a bug in ft_send_tm() that was incorrectly calling ft_get_lun_for_cmd() -> transport_get_lun_for_cmd(), instead of using transport_get_lun_for_tmr() for the proper struct se_lun lookup that was triggering an OOPs in the se_cmd->tmr_req failure path. This patch fixes the issue by re-arranging the codepath where transport_get_lun_for_tmr() is called after tmr request is allocated and made it available as part of se_cmd. It also drops the now unnecessary ft_get_lun_for_cmd() unpacking code, and uses scsilun_to_int() directly ahead of transport_get_lun_for_cmd() and transport_get_lun_for_tmr() usage. Signed-off-by: Patil, Kiran Signed-off-by: Robert Love Signed-off-by: Nicholas A. Bellinger --- drivers/target/tcm_fc/tcm_fc.h | 2 +- drivers/target/tcm_fc/tfc_cmd.c | 64 +++++++++++++++++---------------- 2 files changed, 34 insertions(+), 32 deletions(-) diff --git a/drivers/target/tcm_fc/tcm_fc.h b/drivers/target/tcm_fc/tcm_fc.h index defff32b7880..7b82f1b7fef8 100644 --- a/drivers/target/tcm_fc/tcm_fc.h +++ b/drivers/target/tcm_fc/tcm_fc.h @@ -144,7 +144,7 @@ enum ft_cmd_state { */ struct ft_cmd { enum ft_cmd_state state; - u16 lun; /* LUN from request */ + u32 lun; /* LUN from request */ struct ft_sess *sess; /* session held for cmd */ struct fc_seq *seq; /* sequence in exchange mgr */ struct se_cmd se_cmd; /* Local TCM I/O descriptor */ diff --git a/drivers/target/tcm_fc/tfc_cmd.c b/drivers/target/tcm_fc/tfc_cmd.c index c056a1132ae1..b2a106729d49 100644 --- a/drivers/target/tcm_fc/tfc_cmd.c +++ b/drivers/target/tcm_fc/tfc_cmd.c @@ -94,29 +94,6 @@ void ft_dump_cmd(struct ft_cmd *cmd, const char *caller) 16, 4, cmd->cdb, MAX_COMMAND_SIZE, 0); } -/* - * Get LUN from CDB. - */ -static int ft_get_lun_for_cmd(struct ft_cmd *cmd, u8 *lunp) -{ - u64 lun; - - lun = lunp[1]; - switch (lunp[0] >> 6) { - case 0: - break; - case 1: - lun |= (lunp[0] & 0x3f) << 8; - break; - default: - return -1; - } - if (lun >= TRANSPORT_MAX_LUNS_PER_TPG) - return -1; - cmd->lun = lun; - return transport_get_lun_for_cmd(&cmd->se_cmd, NULL, lun); -} - static void ft_queue_cmd(struct ft_sess *sess, struct ft_cmd *cmd) { struct se_queue_obj *qobj; @@ -418,6 +395,7 @@ static void ft_send_tm(struct ft_cmd *cmd) { struct se_tmr_req *tmr; struct fcp_cmnd *fcp; + struct ft_sess *sess; u8 tm_func; fcp = fc_frame_payload_get(cmd->req_frame, sizeof(*fcp)); @@ -425,13 +403,6 @@ static void ft_send_tm(struct ft_cmd *cmd) switch (fcp->fc_tm_flags) { case FCP_TMF_LUN_RESET: tm_func = TMR_LUN_RESET; - if (ft_get_lun_for_cmd(cmd, fcp->fc_lun) < 0) { - ft_dump_cmd(cmd, __func__); - transport_send_check_condition_and_sense(&cmd->se_cmd, - cmd->se_cmd.scsi_sense_reason, 0); - ft_sess_put(cmd->sess); - return; - } break; case FCP_TMF_TGT_RESET: tm_func = TMR_TARGET_WARM_RESET; @@ -463,6 +434,36 @@ static void ft_send_tm(struct ft_cmd *cmd) return; } cmd->se_cmd.se_tmr_req = tmr; + + switch (fcp->fc_tm_flags) { + case FCP_TMF_LUN_RESET: + cmd->lun = scsilun_to_int((struct scsi_lun *)fcp->fc_lun); + if (transport_get_lun_for_tmr(&cmd->se_cmd, cmd->lun) < 0) { + /* + * Make sure to clean up newly allocated TMR request + * since "unable to handle TMR request because failed + * to get to LUN" + */ + FT_TM_DBG("Failed to get LUN for TMR func %d, " + "se_cmd %p, unpacked_lun %d\n", + tm_func, &cmd->se_cmd, cmd->lun); + ft_dump_cmd(cmd, __func__); + sess = cmd->sess; + transport_send_check_condition_and_sense(&cmd->se_cmd, + cmd->se_cmd.scsi_sense_reason, 0); + transport_generic_free_cmd(&cmd->se_cmd, 0, 1, 0); + ft_sess_put(sess); + return; + } + break; + case FCP_TMF_TGT_RESET: + case FCP_TMF_CLR_TASK_SET: + case FCP_TMF_ABT_TASK_SET: + case FCP_TMF_CLR_ACA: + break; + default: + return; + } transport_generic_handle_tmr(&cmd->se_cmd); } @@ -635,7 +636,8 @@ static void ft_send_cmd(struct ft_cmd *cmd) fc_seq_exch(cmd->seq)->lp->tt.seq_set_resp(cmd->seq, ft_recv_seq, cmd); - ret = ft_get_lun_for_cmd(cmd, fcp->fc_lun); + cmd->lun = scsilun_to_int((struct scsi_lun *)fcp->fc_lun); + ret = transport_get_lun_for_cmd(&cmd->se_cmd, NULL, cmd->lun); if (ret < 0) { ft_dump_cmd(cmd, __func__); transport_send_check_condition_and_sense(&cmd->se_cmd, From 7c7cf3b9c31ed09822e5c186297991093ee13c49 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 13 Jun 2011 23:08:46 +0300 Subject: [PATCH 139/224] tcm_fc: Fix possible lock to unlock type deadlock There is a typo here, it should be an unlock instead of a lock. The original code will deadlock. Signed-off-by: Dan Carpenter Signed-off-by: Nicholas Bellinger --- drivers/target/tcm_fc/tfc_sess.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/tcm_fc/tfc_sess.c b/drivers/target/tcm_fc/tfc_sess.c index 03744c23095c..7491e21cc6ae 100644 --- a/drivers/target/tcm_fc/tfc_sess.c +++ b/drivers/target/tcm_fc/tfc_sess.c @@ -332,7 +332,7 @@ void ft_sess_close(struct se_session *se_sess) lport = sess->tport->lport; port_id = sess->port_id; if (port_id == -1) { - mutex_lock(&ft_lport_lock); + mutex_unlock(&ft_lport_lock); return; } FT_SESS_DBG("port_id %x\n", port_id); From 95efa2863996b643083957079b9304fb3c01130f Mon Sep 17 00:00:00 2001 From: Nicholas Bellinger Date: Thu, 23 Jun 2011 23:28:46 +0000 Subject: [PATCH 140/224] tcm_fc: Fix conversion spec warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch fixes the following conversion specification warning for size_t drivers/target/tcm_fc/tfc_io.c: In function ‘ft_queue_data_in’: drivers/target/tcm_fc/tfc_io.c:209: warning: format ‘%x’ expects type ‘unsigned int’, but argument 5 has type ‘size_t’ Reported-by: Randy Dunlap Reported-by: Stephen Rothwell Signed-off-by: Nicholas Bellinger --- drivers/target/tcm_fc/tfc_io.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/target/tcm_fc/tfc_io.c b/drivers/target/tcm_fc/tfc_io.c index 4c3c0efbe13f..8c4a24077d9d 100644 --- a/drivers/target/tcm_fc/tfc_io.c +++ b/drivers/target/tcm_fc/tfc_io.c @@ -203,7 +203,7 @@ int ft_queue_data_in(struct se_cmd *se_cmd) /* XXX For now, initiator will retry */ if (printk_ratelimit()) printk(KERN_ERR "%s: Failed to send frame %p, " - "xid <0x%x>, remaining <0x%x>, " + "xid <0x%x>, remaining %zu, " "lso_max <0x%x>\n", __func__, fp, ep->xid, remaining, lport->lso_max); From d698a34da7c5626c3c16a6311c6290522e902e7f Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 23 Jun 2011 00:49:29 -0400 Subject: [PATCH 141/224] drm/radeon/kms: fix num_banks tiling config for fusion The field is encoded: 0 = 4 banks 1 = 8 banks 2 = 16 banks Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/evergreen.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 445af7981637..12d2fdc52414 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2013,9 +2013,9 @@ static void evergreen_gpu_init(struct radeon_device *rdev) rdev->config.evergreen.tile_config |= (3 << 0); break; } - /* num banks is 8 on all fusion asics */ + /* num banks is 8 on all fusion asics. 0 = 4, 1 = 8, 2 = 16 */ if (rdev->flags & RADEON_IS_IGP) - rdev->config.evergreen.tile_config |= 8 << 4; + rdev->config.evergreen.tile_config |= 1 << 4; else rdev->config.evergreen.tile_config |= ((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT) << 4; From ee4017f4ac8163737793cc64df535cd246792887 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 23 Jun 2011 12:19:32 -0400 Subject: [PATCH 142/224] drm/radeon/kms: handle special cases for vddc A voltage value of 0xff01 requires that the driver look up the max voltage for the board based using the atom SetVoltage command table. Setting the proper voltage should fix stability on some newer asics. Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 1 + drivers/gpu/drm/radeon/radeon_atombios.c | 36 ++++++++++++++++++++++++ 2 files changed, 37 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 27f45579e64b..ef0e0e016914 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -179,6 +179,7 @@ void radeon_pm_resume(struct radeon_device *rdev); void radeon_combios_get_power_modes(struct radeon_device *rdev); void radeon_atombios_get_power_modes(struct radeon_device *rdev); void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 voltage_type); +int radeon_atom_get_max_vddc(struct radeon_device *rdev, u16 *voltage); void rs690_pm_info(struct radeon_device *rdev); extern int rv6xx_get_temp(struct radeon_device *rdev); extern int rv770_get_temp(struct radeon_device *rdev); diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index 1e725d9f767f..bf2b61584cdb 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -2320,6 +2320,14 @@ static bool radeon_atombios_parse_pplib_clock_info(struct radeon_device *rdev, le16_to_cpu(clock_info->r600.usVDDC); } + /* patch up vddc if necessary */ + if (rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage == 0xff01) { + u16 vddc; + + if (radeon_atom_get_max_vddc(rdev, &vddc) == 0) + rdev->pm.power_state[state_index].clock_info[mode_index].voltage.voltage = vddc; + } + if (rdev->flags & RADEON_IS_IGP) { /* skip invalid modes */ if (rdev->pm.power_state[state_index].clock_info[mode_index].sclk == 0) @@ -2630,7 +2638,35 @@ void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 v atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); } +int radeon_atom_get_max_vddc(struct radeon_device *rdev, + u16 *voltage) +{ + union set_voltage args; + int index = GetIndexIntoMasterTable(COMMAND, SetVoltage); + u8 frev, crev; + if (!atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev, &crev)) + return -EINVAL; + + switch (crev) { + case 1: + return -EINVAL; + case 2: + args.v2.ucVoltageType = SET_VOLTAGE_GET_MAX_VOLTAGE; + args.v2.ucVoltageMode = 0; + args.v2.usVoltageLevel = 0; + + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + + *voltage = le16_to_cpu(args.v2.usVoltageLevel); + break; + default: + DRM_ERROR("Unknown table version %d, %d\n", frev, crev); + return -EINVAL; + } + + return 0; +} void radeon_atom_initialize_bios_scratch_regs(struct drm_device *dev) { From 778e24bb6dd8682318bb496d4bfdc32b501a6420 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Thu, 23 Jun 2011 01:34:59 +0000 Subject: [PATCH 143/224] xfs: reset inode per-lifetime state when recycling it XFS inodes has several per-lifetime state fields that determine the behaviour of the inode. These state fields are not all reset when an inode is reused from the reclaimable state. This can lead to unexpected behaviour of the new inode such as speculative preallocation not being truncated away in the expected manner for local files until the inode is subsequently truncated, freed or cycles out of the cache. It can also lead to an inode being considered to be a filestream inode or having been truncated when that is not the case. Rework the reinitialisation of the inode when it is recycled to ensure that it is pristine before it is reused. While there, also fix the resetting of state flags in the recycling error paths so the inode does not become unreclaimable. Signed-off-by: Dave Chinner Signed-off-by: Alex Elder --- fs/xfs/xfs_iget.c | 13 +++++++++---- fs/xfs/xfs_inode.h | 10 ++++++++++ 2 files changed, 19 insertions(+), 4 deletions(-) diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index cb9b6d1469f7..3631783b2b53 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -253,16 +253,21 @@ xfs_iget_cache_hit( rcu_read_lock(); spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~XFS_INEW; - ip->i_flags |= XFS_IRECLAIMABLE; - __xfs_inode_set_reclaim_tag(pag, ip); + ip->i_flags &= ~(XFS_INEW | XFS_IRECLAIM); + ASSERT(ip->i_flags & XFS_IRECLAIMABLE); trace_xfs_iget_reclaim_fail(ip); goto out_error; } spin_lock(&pag->pag_ici_lock); spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~(XFS_IRECLAIMABLE | XFS_IRECLAIM); + + /* + * Clear the per-lifetime state in the inode as we are now + * effectively a new inode and need to return to the initial + * state before reuse occurs. + */ + ip->i_flags &= ~XFS_IRECLAIM_RESET_FLAGS; ip->i_flags |= XFS_INEW; __xfs_inode_clear_reclaim_tag(mp, pag, ip); inode->i_state = I_NEW; diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 3ae6d58e5473..964cfea77686 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -383,6 +383,16 @@ static inline void xfs_ifunlock(xfs_inode_t *ip) #define XFS_ITRUNCATED 0x0020 /* truncated down so flush-on-close */ #define XFS_IDIRTY_RELEASE 0x0040 /* dirty release already seen */ +/* + * Per-lifetime flags need to be reset when re-using a reclaimable inode during + * inode lookup. Thi prevents unintended behaviour on the new inode from + * ocurring. + */ +#define XFS_IRECLAIM_RESET_FLAGS \ + (XFS_IRECLAIMABLE | XFS_IRECLAIM | \ + XFS_IDIRTY_RELEASE | XFS_ITRUNCATED | \ + XFS_IFILESTREAM); + /* * Flags for inode locking. * Bit ranges: 1<<1 - 1<<16-1 -- iolock/ilock modes (bitfield) From df4368a146d2b350b8398babfe11e2088f741d67 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Thu, 23 Jun 2011 01:35:00 +0000 Subject: [PATCH 144/224] xfs: clear XFS_IDIRTY_RELEASE on truncate down When an inode is truncated down, speculative preallocation is removed from the inode. This should also reset the state bits for controlling whether preallocation is subsequently removed when the file is next closed. The flag is not being cleared, so repeated operations on a file that first involve a truncate (e.g. multiple repeated dd invocations on a file) give different file layouts for the second and subsequent invocations. Fix this by clearing the XFS_IDIRTY_RELEASE state bit when the XFS_ITRUNCATED bit is detected in xfs_release() and hence ensure that speculative delalloc is removed on files that have been truncated down. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/xfs_vnodeops.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index b7a5fe7c52c8..619720705bc6 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -960,8 +960,11 @@ xfs_release( * be exposed to that problem. */ truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED); - if (truncated && VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) - xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE); + if (truncated) { + xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE); + if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) + xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE); + } } if (ip->i_d.di_nlink == 0) From 4a33821236f2ef3af0081e8a5eec1301cbed3125 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Thu, 23 Jun 2011 01:35:01 +0000 Subject: [PATCH 145/224] xfs: prevent bogus assert when trying to remove non-existent attribute If the attribute fork on an inode is in btree format and has multiple levels (i.e node format rather than leaf format), then a lookup failure will trigger an assert failure in xfs_da_path_shift if the flag XFS_DA_OP_OKNOENT is not set. This flag is used to indicate to the directory btree code that not finding an entry is not a fatal error. In the case of doing a lookup for a directory name removal, this is valid as a user cannot insert an arbitrary name to remove from the directory btree. However, in the case of the attribute tree, a user has direct control over the attribute name and can ask for any random name to be removed without any validation. In this case, fsstress is asking for a non-existent user.selinux attribute to be removed, and that is causing xfs_da_path_shift() to fall off the bottom of the tree where it asserts that a lookup failure is allowed. Because the flag is not set, we die a horrible death on a debug enable kernel. Prevent this assert from firing on attribute removes by adding the op_flag XFS_DA_OP_OKNOENT to atribute removal operations. Discovered when testing on a SELinux enabled system by fsstress in test 070 by trying to remove a non-existent user.selinux attribute. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/xfs_attr.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index c86375378810..01d2072fb6d4 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -489,6 +489,13 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags) args.total = 0; args.whichfork = XFS_ATTR_FORK; + /* + * we have no control over the attribute names that userspace passes us + * to remove, so we have to allow the name lookup prior to attribute + * removal to fail. + */ + args.op_flags = XFS_DA_OP_OKNOENT; + /* * Attach the dquots to the inode. */ From f920fe1cb74191a780d88937f36994231a8faba1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Pawe=C5=82=20Drewniak?= Date: Fri, 24 Jun 2011 02:07:35 -0400 Subject: [PATCH 146/224] pata_marvell: Add support for 88SE91A0, 88SE91A4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch enables support for Marvell IDE PATA controllers found on Asus P8P67LE motherboard. The formatting has been corrected and I also received a report from two users of this motherboard that the patch works. Signed-off-by: Paweł Drewniak Signed-off-by: Jeff Garzik --- drivers/ata/pata_marvell.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/ata/pata_marvell.c b/drivers/ata/pata_marvell.c index 75a6a0c0094f..5d7f58a7e34d 100644 --- a/drivers/ata/pata_marvell.c +++ b/drivers/ata/pata_marvell.c @@ -161,6 +161,9 @@ static const struct pci_device_id marvell_pci_tbl[] = { { PCI_DEVICE(0x11AB, 0x6121), }, { PCI_DEVICE(0x11AB, 0x6123), }, { PCI_DEVICE(0x11AB, 0x6145), }, + { PCI_DEVICE(0x1B4B, 0x91A0), }, + { PCI_DEVICE(0x1B4B, 0x91A4), }, + { } /* terminate list */ }; From 7a0ee92b4a510bc2dd026333f90031e883e0cde0 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 23 Jun 2011 22:00:20 +0100 Subject: [PATCH 147/224] ARM: pm: proc-v7: fix missing struct processor pointers for suspend code Add the missing suspend/resume pointers for the suspend code. This is needed when building for multiple CPUs. Tested-by: Kevin Hilman Acked-by: Jean Pihet Signed-off-by: Russell King --- arch/arm/mm/proc-v7.S | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index 3c3867850a30..e27c011a759c 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -418,9 +418,9 @@ ENTRY(v7_processor_functions) .word cpu_v7_dcache_clean_area .word cpu_v7_switch_mm .word cpu_v7_set_pte_ext - .word 0 - .word 0 - .word 0 + .word cpu_v7_suspend_size + .word cpu_v7_do_suspend + .word cpu_v7_do_resume .size v7_processor_functions, . - v7_processor_functions .section ".rodata" From 111b20d01346b9635b3223c7af4e40e43bee8dc6 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 22 Jun 2011 15:41:58 +0100 Subject: [PATCH 148/224] ARM: pm: ensure ARMv7 CPUs save and restore the TLS register Ensure that the TLS register is saved and restored over a suspend cycle, so that userspace programs don't see a corrupted TLS value. Tested-by: Kevin Hilman Acked-by: Jean Pihet Signed-off-by: Russell King --- arch/arm/mm/proc-v7.S | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index e27c011a759c..089c0b5e454f 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -210,19 +210,21 @@ cpu_v7_name: /* Suspend/resume support: derived from arch/arm/mach-s5pv210/sleep.S */ .globl cpu_v7_suspend_size -.equ cpu_v7_suspend_size, 4 * 8 +.equ cpu_v7_suspend_size, 4 * 9 #ifdef CONFIG_PM_SLEEP ENTRY(cpu_v7_do_suspend) stmfd sp!, {r4 - r11, lr} mrc p15, 0, r4, c13, c0, 0 @ FCSE/PID mrc p15, 0, r5, c13, c0, 1 @ Context ID + mrc p15, 0, r6, c13, c0, 3 @ User r/o thread ID + stmia r0!, {r4 - r6} mrc p15, 0, r6, c3, c0, 0 @ Domain ID mrc p15, 0, r7, c2, c0, 0 @ TTB 0 mrc p15, 0, r8, c2, c0, 1 @ TTB 1 mrc p15, 0, r9, c1, c0, 0 @ Control register mrc p15, 0, r10, c1, c0, 1 @ Auxiliary control register mrc p15, 0, r11, c1, c0, 2 @ Co-processor access control - stmia r0, {r4 - r11} + stmia r0, {r6 - r11} ldmfd sp!, {r4 - r11, pc} ENDPROC(cpu_v7_do_suspend) @@ -230,9 +232,11 @@ ENTRY(cpu_v7_do_resume) mov ip, #0 mcr p15, 0, ip, c8, c7, 0 @ invalidate TLBs mcr p15, 0, ip, c7, c5, 0 @ invalidate I cache - ldmia r0, {r4 - r11} + ldmia r0!, {r4 - r6} mcr p15, 0, r4, c13, c0, 0 @ FCSE/PID mcr p15, 0, r5, c13, c0, 1 @ Context ID + mcr p15, 0, r6, c13, c0, 3 @ User r/o thread ID + ldmia r0, {r6 - r11} mcr p15, 0, r6, c3, c0, 0 @ Domain ID mcr p15, 0, r7, c2, c0, 0 @ TTB 0 mcr p15, 0, r8, c2, c0, 1 @ TTB 1 From 16866741bda5d16f3d30d1656ce941faf5dad34c Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 23 Jun 2011 23:54:40 +0200 Subject: [PATCH 149/224] ALSA: Remove unneeded version.h includes from sound/ In the sound/ directory there are two files (flagged by 'make versioncheck'); sound/pci/asihpi/asihpi.c and sound/soc/codecs/wm8991.c that include linux/version.h although they don't need it. This patch removes the unneeded includes. Signed-off-by: Jesper Juhl Signed-off-by: Takashi Iwai --- sound/pci/asihpi/asihpi.c | 1 - sound/soc/codecs/wm8991.c | 1 - 2 files changed, 2 deletions(-) diff --git a/sound/pci/asihpi/asihpi.c b/sound/pci/asihpi/asihpi.c index 2ca6f4f85b41..e3569bdd3b64 100644 --- a/sound/pci/asihpi/asihpi.c +++ b/sound/pci/asihpi/asihpi.c @@ -27,7 +27,6 @@ #include "hpioctl.h" #include -#include #include #include #include diff --git a/sound/soc/codecs/wm8991.c b/sound/soc/codecs/wm8991.c index 3c2ee1bb73cd..6af23d06870f 100644 --- a/sound/soc/codecs/wm8991.c +++ b/sound/soc/codecs/wm8991.c @@ -13,7 +13,6 @@ #include #include -#include #include #include #include From 50bc03ab5c7529fdfe4e01621efca7d26439ea00 Mon Sep 17 00:00:00 2001 From: Benjamin Tissoires Date: Tue, 21 Jun 2011 15:01:53 +0200 Subject: [PATCH 150/224] HID: hid-multitouch: ensure slots are initialized In case a device does not provide the feature "Maximum Contact Count", or set it at 0, the maxcontacts field may be at 0 while calling input_mt_init_slots. This patch ensures that hid-multitouch will allways report ABS_MT_SLOT and ABS_MT_TRACKING_ID to the user space. This corrects a bug found with some Ilitek devices that has been integrated in 3.0-rc0. Signed-off-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- drivers/hid/hid-multitouch.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c index 0b2dcd0ee591..386765654cf6 100644 --- a/drivers/hid/hid-multitouch.c +++ b/drivers/hid/hid-multitouch.c @@ -271,6 +271,8 @@ static int mt_input_mapping(struct hid_device *hdev, struct hid_input *hi, } return 1; case HID_DG_CONTACTID: + if (!td->maxcontacts) + td->maxcontacts = MT_DEFAULT_MAXCONTACT; input_mt_init_slots(hi->input, td->maxcontacts); td->last_slot_field = usage->hid; td->last_field_index = field->index; @@ -547,9 +549,6 @@ static int mt_probe(struct hid_device *hdev, const struct hid_device_id *id) if (ret) goto fail; - if (!td->maxcontacts) - td->maxcontacts = MT_DEFAULT_MAXCONTACT; - td->slots = kzalloc(td->maxcontacts * sizeof(struct mt_slot), GFP_KERNEL); if (!td->slots) { From 85a600825b425d52e466c6093dcdfeba85eb0044 Mon Sep 17 00:00:00 2001 From: Benjamin Tissoires Date: Tue, 21 Jun 2011 15:01:54 +0200 Subject: [PATCH 151/224] HID: hid-multitouch: correct VID for Stantum panels while merging hid-stantum into hid-multitouch, I did not correctly copy/paste the VIDs for those devices. This patch fixes it. Signed-off-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- drivers/hid/hid-multitouch.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c index 386765654cf6..467b518a43b3 100644 --- a/drivers/hid/hid-multitouch.c +++ b/drivers/hid/hid-multitouch.c @@ -706,10 +706,10 @@ static const struct hid_device_id mt_devices[] = { HID_USB_DEVICE(USB_VENDOR_ID_STANTUM, USB_DEVICE_ID_MTP)}, { .driver_data = MT_CLS_CONFIDENCE, - HID_USB_DEVICE(USB_VENDOR_ID_STANTUM, + HID_USB_DEVICE(USB_VENDOR_ID_STANTUM_STM, USB_DEVICE_ID_MTP_STM)}, { .driver_data = MT_CLS_CONFIDENCE, - HID_USB_DEVICE(USB_VENDOR_ID_STANTUM, + HID_USB_DEVICE(USB_VENDOR_ID_STANTUM_SITRONIX, USB_DEVICE_ID_MTP_SITRONIX)}, /* Touch International panels */ From c3ead6de4f6bd1c08a81f84e629e3dbf4a9078f0 Mon Sep 17 00:00:00 2001 From: Benjamin Tissoires Date: Tue, 21 Jun 2011 15:01:55 +0200 Subject: [PATCH 152/224] HID: hid-multitouch: add support for a new Lumio dual-touch panel Signed-off-by: Benjamin Tissoires Signed-off-by: Jiri Kosina --- drivers/hid/hid-core.c | 1 + drivers/hid/hid-ids.h | 1 + drivers/hid/hid-multitouch.c | 3 +++ 3 files changed, 5 insertions(+) diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index f7440e8ce3e7..6f3289a57888 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -1423,6 +1423,7 @@ static const struct hid_device_id hid_have_special_driver[] = { { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_SPACETRAVELLER) }, { HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_SPACENAVIGATOR) }, { HID_USB_DEVICE(USB_VENDOR_ID_LUMIO, USB_DEVICE_ID_CRYSTALTOUCH) }, + { HID_USB_DEVICE(USB_VENDOR_ID_LUMIO, USB_DEVICE_ID_CRYSTALTOUCH_DUAL) }, { HID_USB_DEVICE(USB_VENDOR_ID_MICROCHIP, USB_DEVICE_ID_PICOLCD) }, { HID_USB_DEVICE(USB_VENDOR_ID_MICROCHIP, USB_DEVICE_ID_PICOLCD_BOOTLOADER) }, { HID_USB_DEVICE(USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_SIDEWINDER_GV) }, diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index aecb5a4b8d6d..a756ee6c7df5 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -449,6 +449,7 @@ #define USB_VENDOR_ID_LUMIO 0x202e #define USB_DEVICE_ID_CRYSTALTOUCH 0x0006 +#define USB_DEVICE_ID_CRYSTALTOUCH_DUAL 0x0007 #define USB_VENDOR_ID_MCC 0x09db #define USB_DEVICE_ID_MCC_PMD1024LS 0x0076 diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c index 467b518a43b3..62cac4dc3b62 100644 --- a/drivers/hid/hid-multitouch.c +++ b/drivers/hid/hid-multitouch.c @@ -676,6 +676,9 @@ static const struct hid_device_id mt_devices[] = { { .driver_data = MT_CLS_CONFIDENCE_MINUS_ONE, HID_USB_DEVICE(USB_VENDOR_ID_LUMIO, USB_DEVICE_ID_CRYSTALTOUCH) }, + { .driver_data = MT_CLS_CONFIDENCE_MINUS_ONE, + HID_USB_DEVICE(USB_VENDOR_ID_LUMIO, + USB_DEVICE_ID_CRYSTALTOUCH_DUAL) }, /* MosArt panels */ { .driver_data = MT_CLS_CONFIDENCE_MINUS_ONE, From 46e4edbf7ea9cf26665eb9f90c0fc7688d1a51ed Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 23 Jun 2011 23:59:32 +0200 Subject: [PATCH 153/224] Remove unneeded version.h includes from fs/ It was pointed out by 'make versioncheck' that some includes of linux/version.h were not needed in fs/ (fs/btrfs/ctree.h and fs/omfs/file.c). This patch removes them. Signed-off-by: Jesper Juhl Acked-by: Bob Copeland Signed-off-by: Linus Torvalds --- fs/btrfs/ctree.h | 1 - fs/omfs/file.c | 1 - 2 files changed, 2 deletions(-) diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 300628795fdb..f30ac05dbda7 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -19,7 +19,6 @@ #ifndef __BTRFS_CTREE__ #define __BTRFS_CTREE__ -#include #include #include #include diff --git a/fs/omfs/file.c b/fs/omfs/file.c index d738a7e493dd..2c6d95257a4d 100644 --- a/fs/omfs/file.c +++ b/fs/omfs/file.c @@ -4,7 +4,6 @@ * Released under GPL v2. */ -#include #include #include #include From e4fb0edb7c03e5ec19b6f732f1dfbe911212dbde Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 20 Jun 2011 14:33:16 -0400 Subject: [PATCH 154/224] cifs: free blkcipher in smbhash This is currently leaked in the rc == 0 case. Reported-by: J. Bruce Fields Signed-off-by: Jeff Layton Reviewed-by: Shirish Pargaonkar Signed-off-by: Steve French --- fs/cifs/smbencrypt.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index 1525d5e662b6..1c5b770c3141 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -90,12 +90,10 @@ smbhash(unsigned char *out, const unsigned char *in, unsigned char *key) sg_init_one(&sgout, out, 8); rc = crypto_blkcipher_encrypt(&desc, &sgout, &sgin, 8); - if (rc) { + if (rc) cERROR(1, "could not encrypt crypt key rc: %d\n", rc); - crypto_free_blkcipher(tfm_des); - goto smbhash_err; - } + crypto_free_blkcipher(tfm_des); smbhash_err: return rc; } From 1973f0faeb4a5f35597793c65d3c94d8fd386e10 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Fri, 24 Jun 2011 13:13:29 -0400 Subject: [PATCH 155/224] Btrfs: make sure to record the transid in new inodes When we create a new inode, we aren't filling in the field that records the transaction that last changed this inode. If we then go to fsync that inode, it will be skipped because the field isn't filled in. Signed-off-by: Chris Mason --- fs/btrfs/inode.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5813dec5101c..87f1e0cf26f8 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4520,6 +4520,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, inode_tree_add(inode); trace_btrfs_inode_new(inode); + btrfs_set_inode_last_trans(trans, inode); return inode; fail: From 9b8e072a31180eb5cd6991d08524d9c4fa235ade Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 21 Jun 2011 07:18:26 -0400 Subject: [PATCH 156/224] cifs: mark CONFIG_CIFS_NFSD_EXPORT as BROKEN This does not work properly with CIFS as current servers do not enable support for the FILE_OPEN_BY_FILE_ID on SMB NTCreateX and not all NFS clients handle ESTALE. For now, it just plain doesn't work. Mark it BROKEN to discourage distros from enabling it. Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 53ed1ad2c112..f66cc1625150 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -156,6 +156,6 @@ config CIFS_ACL config CIFS_NFSD_EXPORT bool "Allow nfsd to export CIFS file system (EXPERIMENTAL)" - depends on CIFS && EXPERIMENTAL + depends on CIFS && EXPERIMENTAL && BROKEN help Allows NFS server to export a CIFS mounted share (nfsd over cifs) From dd8544661947ad6d8d87b3c9d4333bfa1583d1bc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 08:24:42 -0400 Subject: [PATCH 157/224] take bdi setup/destruction into cifs_mount/cifs_umount Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 12 +----------- fs/cifs/connect.c | 10 ++++++++++ 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 2f0c58646c10..5d3c4fa4b546 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -116,18 +116,12 @@ cifs_read_super(struct super_block *sb, struct smb_vol *volume_info, spin_lock_init(&cifs_sb->tlink_tree_lock); cifs_sb->tlink_tree = RB_ROOT; - rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY); - if (rc) - return rc; - - cifs_sb->bdi.ra_pages = default_backing_dev_info.ra_pages; - rc = cifs_mount(sb, cifs_sb, volume_info, devname); if (rc) { if (!silent) cERROR(1, "cifs_mount failed w/return code = %d", rc); - goto out_mount_failed; + return rc; } sb->s_magic = CIFS_MAGIC_NUMBER; @@ -171,9 +165,6 @@ out_no_root: iput(inode); cifs_umount(sb, cifs_sb); - -out_mount_failed: - bdi_destroy(&cifs_sb->bdi); return rc; } @@ -199,7 +190,6 @@ cifs_put_super(struct super_block *sb) } unload_nls(cifs_sb->local_nls); - bdi_destroy(&cifs_sb->bdi); kfree(cifs_sb); } diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 12cf72dd0c42..78fd7557e35d 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2983,6 +2983,13 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, struct tcon_link *tlink; #ifdef CONFIG_CIFS_DFS_UPCALL int referral_walks_count = 0; + + rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY); + if (rc) + return rc; + + cifs_sb->bdi.ra_pages = default_backing_dev_info.ra_pages; + try_mount_again: /* cleanup activities if we're chasing a referral */ if (referral_walks_count) { @@ -3007,6 +3014,7 @@ try_mount_again: srvTcp = cifs_get_tcp_session(volume_info); if (IS_ERR(srvTcp)) { rc = PTR_ERR(srvTcp); + bdi_destroy(&cifs_sb->bdi); goto out; } @@ -3161,6 +3169,7 @@ mount_fail_check: cifs_put_smb_ses(pSesInfo); else cifs_put_tcp_session(srvTcp); + bdi_destroy(&cifs_sb->bdi); goto out; } @@ -3357,6 +3366,7 @@ cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) } spin_unlock(&cifs_sb->tlink_tree_lock); + bdi_destroy(&cifs_sb->bdi); return 0; } From 6d6861757dfadb7d6aec6bb34acd471210a755f9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 08:34:57 -0400 Subject: [PATCH 158/224] cifs: double free on mount failure if we get to out_super with ->s_root already set (e.g. with cifs_get_root() failure), we'll end up with cifs_put_super() called and ->mountdata freed twice. We'll also get cifs_sb freed twice and cifs_sb->local_nls dropped twice. The problem is, we can get to out_super both with and without ->s_root, which makes ->put_super() a bad place for such work. Switch to ->kill_sb(), have all that work done there after kill_anon_super(). Unlike ->put_super(), ->kill_sb() is called by deactivate_locked_super() whether we have ->s_root or not. Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 5d3c4fa4b546..dc76c7bccb15 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -184,11 +184,13 @@ cifs_put_super(struct super_block *sb) rc = cifs_umount(sb, cifs_sb); if (rc) cERROR(1, "cifs_umount failed with return code %d", rc); - if (cifs_sb->mountdata) { - kfree(cifs_sb->mountdata); - cifs_sb->mountdata = NULL; - } +} +static void cifs_kill_sb(struct super_block *sb) +{ + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); + kill_anon_super(sb); + kfree(cifs_sb->mountdata); unload_nls(cifs_sb->local_nls); kfree(cifs_sb); } @@ -729,8 +731,8 @@ out_shared: goto out; out_super: - kfree(cifs_sb->mountdata); deactivate_locked_super(sb); + goto out; out_cifs_sb: unload_nls(cifs_sb->local_nls); @@ -827,7 +829,7 @@ struct file_system_type cifs_fs_type = { .owner = THIS_MODULE, .name = "cifs", .mount = cifs_do_mount, - .kill_sb = kill_anon_super, + .kill_sb = cifs_kill_sb, /* .fs_flags */ }; const struct inode_operations cifs_dir_inode_ops = { From ca171baaad1420a29cca98be5bdf5596cd70b294 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 08:49:12 -0400 Subject: [PATCH 159/224] cifs: don't leak nls on mount failure if cifs_sb allocation fails, we still need to drop nls we'd stashed into volume_info - the one we would've copied to cifs_sb if we could allocate the latter. Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index dc76c7bccb15..bfab2bc83726 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -672,6 +672,7 @@ cifs_do_mount(struct file_system_type *fs_type, cifs_sb = kzalloc(sizeof(struct cifs_sb_info), GFP_KERNEL); if (cifs_sb == NULL) { root = ERR_PTR(-ENOMEM); + unload_nls(volume_info->local_nls); goto out; } From 2c6292ae4be00454882246d07f38cdf15a823c2a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:05:48 -0400 Subject: [PATCH 160/224] cifs: don't pass superblock to cifs_mount() To close sget() races we'll need to be able to set cifs_sb up before we get the superblock, so we'll want to be able to do cifs_mount() earlier. Fortunately, it's easy to do - setting ->s_maxbytes can be done in cifs_read_super(), ditto for ->s_time_gran and as for putting MS_POSIXACL into ->s_flags, we can mirror it in ->mnt_cifs_flags until cifs_read_super() is called. Kill unused 'devname' argument, while we are at it... Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifs_fs_sb.h | 1 + fs/cifs/cifsfs.c | 13 ++++++++++++- fs/cifs/cifsproto.h | 6 +++--- fs/cifs/connect.c | 28 ++++++++++------------------ 4 files changed, 26 insertions(+), 22 deletions(-) diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index ffb1459dc6ec..7260e11e21f8 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -42,6 +42,7 @@ #define CIFS_MOUNT_MULTIUSER 0x20000 /* multiuser mount */ #define CIFS_MOUNT_STRICT_IO 0x40000 /* strict cache mode */ #define CIFS_MOUNT_RWPIDFORWARD 0x80000 /* use pid forwarding for rw */ +#define CIFS_MOUNT_POSIXACL 0x100000 /* mirror of MS_POSIXACL in mnt_cifs_flags */ struct cifs_sb_info { struct rb_root tlink_tree; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index bfab2bc83726..8f7451f3c8e6 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -116,7 +116,7 @@ cifs_read_super(struct super_block *sb, struct smb_vol *volume_info, spin_lock_init(&cifs_sb->tlink_tree_lock); cifs_sb->tlink_tree = RB_ROOT; - rc = cifs_mount(sb, cifs_sb, volume_info, devname); + rc = cifs_mount(cifs_sb, volume_info); if (rc) { if (!silent) @@ -124,6 +124,17 @@ cifs_read_super(struct super_block *sb, struct smb_vol *volume_info, return rc; } + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIXACL) + sb->s_flags |= MS_POSIXACL; + + if (cifs_sb_master_tcon(cifs_sb)->ses->capabilities & CAP_LARGE_FILES) + sb->s_maxbytes = MAX_LFS_FILESIZE; + else + sb->s_maxbytes = MAX_NON_LFS; + + /* BB FIXME fix time_gran to be larger for LANMAN sessions */ + sb->s_time_gran = 100; + sb->s_magic = CIFS_MAGIC_NUMBER; sb->s_op = &cifs_super_ops; sb->s_bdi = &cifs_sb->bdi; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 953f84413c77..5814fe543f95 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -157,8 +157,7 @@ extern int cifs_match_super(struct super_block *, void *); extern void cifs_cleanup_volume_info(struct smb_vol **pvolume_info); extern int cifs_setup_volume_info(struct smb_vol **pvolume_info, char *mount_data, const char *devname); -extern int cifs_mount(struct super_block *, struct cifs_sb_info *, - struct smb_vol *, const char *); +extern int cifs_mount(struct cifs_sb_info *, struct smb_vol *); extern int cifs_umount(struct super_block *, struct cifs_sb_info *); extern void cifs_dfs_release_automount_timer(void); void cifs_proc_init(void); @@ -218,7 +217,8 @@ extern int get_dfs_path(int xid, struct cifs_ses *pSesInfo, struct dfs_info3_param **preferrals, int remap); extern void reset_cifs_unix_caps(int xid, struct cifs_tcon *tcon, - struct super_block *sb, struct smb_vol *vol); + struct cifs_sb_info *cifs_sb, + struct smb_vol *vol); extern int CIFSSMBQFSInfo(const int xid, struct cifs_tcon *tcon, struct kstatfs *FSData); extern int SMBOldQFSInfo(const int xid, struct cifs_tcon *tcon, diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 78fd7557e35d..3011ac8c9249 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2546,7 +2546,7 @@ ip_connect(struct TCP_Server_Info *server) } void reset_cifs_unix_caps(int xid, struct cifs_tcon *tcon, - struct super_block *sb, struct smb_vol *vol_info) + struct cifs_sb_info *cifs_sb, struct smb_vol *vol_info) { /* if we are reconnecting then should we check to see if * any requested capabilities changed locally e.g. via @@ -2600,22 +2600,23 @@ void reset_cifs_unix_caps(int xid, struct cifs_tcon *tcon, cap &= ~CIFS_UNIX_POSIX_ACL_CAP; else if (CIFS_UNIX_POSIX_ACL_CAP & cap) { cFYI(1, "negotiated posix acl support"); - if (sb) - sb->s_flags |= MS_POSIXACL; + if (cifs_sb) + cifs_sb->mnt_cifs_flags |= + CIFS_MOUNT_POSIXACL; } if (vol_info && vol_info->posix_paths == 0) cap &= ~CIFS_UNIX_POSIX_PATHNAMES_CAP; else if (cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) { cFYI(1, "negotiate posix pathnames"); - if (sb) - CIFS_SB(sb)->mnt_cifs_flags |= + if (cifs_sb) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_POSIX_PATHS; } - if (sb && (CIFS_SB(sb)->rsize > 127 * 1024)) { + if (cifs_sb && (cifs_sb->rsize > 127 * 1024)) { if ((cap & CIFS_UNIX_LARGE_READ_CAP) == 0) { - CIFS_SB(sb)->rsize = 127 * 1024; + cifs_sb->rsize = 127 * 1024; cFYI(DBG2, "larger reads not supported by srv"); } } @@ -2971,8 +2972,7 @@ out: } int -cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, - struct smb_vol *volume_info, const char *devname) +cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *volume_info) { int rc = 0; int xid; @@ -3026,14 +3026,6 @@ try_mount_again: goto mount_fail_check; } - if (pSesInfo->capabilities & CAP_LARGE_FILES) - sb->s_maxbytes = MAX_LFS_FILESIZE; - else - sb->s_maxbytes = MAX_NON_LFS; - - /* BB FIXME fix time_gran to be larger for LANMAN sessions */ - sb->s_time_gran = 100; - /* search for existing tcon to this server share */ tcon = cifs_get_tcon(pSesInfo, volume_info); if (IS_ERR(tcon)) { @@ -3046,7 +3038,7 @@ try_mount_again: if (tcon->ses->capabilities & CAP_UNIX) { /* reset of caps checks mount to see if unix extensions disabled for just this mount */ - reset_cifs_unix_caps(xid, tcon, sb, volume_info); + reset_cifs_unix_caps(xid, tcon, cifs_sb, volume_info); if ((tcon->ses->server->tcpStatus == CifsNeedReconnect) && (le64_to_cpu(tcon->fsUnixInfo.Capability) & CIFS_UNIX_TRANSPORT_ENCRYPTION_MANDATORY_CAP)) { From d687ca380f1a8f3043f42efd2403cbe58c846e70 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:14:27 -0400 Subject: [PATCH 161/224] cifs: leak on mount if we share superblock cifs_sb and nls end up leaked... Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 8f7451f3c8e6..4162ee45d043 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -701,6 +701,8 @@ cifs_do_mount(struct file_system_type *fs_type, if (sb->s_fs_info) { cFYI(1, "Use existing superblock"); + unload_nls(cifs_sb->local_nls); + kfree(cifs_sb); goto out_shared; } From 5d3bc605cafe3f367b1c43b673bf643245c81626 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:17:28 -0400 Subject: [PATCH 162/224] cifs: allocate mountdata earlier pull mountdata allocation up, so that it won't stand in the way when we lift cifs_mount() to location before sget(). Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 4162ee45d043..ec19161dd278 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -687,6 +687,14 @@ cifs_do_mount(struct file_system_type *fs_type, goto out; } + cifs_sb->mountdata = kstrndup(data, PAGE_SIZE, GFP_KERNEL); + if (cifs_sb->mountdata == NULL) { + root = ERR_PTR(-ENOMEM); + unload_nls(volume_info->local_nls); + kfree(cifs_sb); + goto out; + } + cifs_setup_cifs_sb(volume_info, cifs_sb); mnt_data.vol = volume_info; @@ -701,22 +709,12 @@ cifs_do_mount(struct file_system_type *fs_type, if (sb->s_fs_info) { cFYI(1, "Use existing superblock"); + kfree(cifs_sb->mountdata); unload_nls(cifs_sb->local_nls); kfree(cifs_sb); goto out_shared; } - /* - * Copy mount params for use in submounts. Better to do - * the copy here and deal with the error before cleanup gets - * complicated post-mount. - */ - cifs_sb->mountdata = kstrndup(data, PAGE_SIZE, GFP_KERNEL); - if (cifs_sb->mountdata == NULL) { - root = ERR_PTR(-ENOMEM); - goto out_super; - } - sb->s_flags = flags; /* BB should we make this contingent on mount parm? */ sb->s_flags |= MS_NODIRATIME | MS_NOATIME; @@ -749,6 +747,7 @@ out_super: goto out; out_cifs_sb: + kfree(cifs_sb->mountdata); unload_nls(cifs_sb->local_nls); kfree(cifs_sb); From 2ced6f693581357b2a5bf8b031a702c624b12d0d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:20:04 -0400 Subject: [PATCH 163/224] cifs: initialize ->tlink_tree in cifs_setup_cifs_sb() no need to wait until cifs_read_super() and we need it done by the time cifs_mount() will be called. Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 3 --- fs/cifs/connect.c | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index ec19161dd278..61c7aa870f1c 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -113,9 +113,6 @@ cifs_read_super(struct super_block *sb, struct smb_vol *volume_info, cifs_sb = CIFS_SB(sb); - spin_lock_init(&cifs_sb->tlink_tree_lock); - cifs_sb->tlink_tree = RB_ROOT; - rc = cifs_mount(cifs_sb, volume_info); if (rc) { diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 3011ac8c9249..9f09adf51edc 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2663,6 +2663,9 @@ void cifs_setup_cifs_sb(struct smb_vol *pvolume_info, { INIT_DELAYED_WORK(&cifs_sb->prune_tlinks, cifs_prune_tlinks); + spin_lock_init(&cifs_sb->tlink_tree_lock); + cifs_sb->tlink_tree = RB_ROOT; + if (pvolume_info->rsize > CIFSMaxBufSize) { cERROR(1, "rsize %d too large, using MaxBufSize", pvolume_info->rsize); From 2a9b99516c662d1713d58648e4a4c9aef72051bc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:27:16 -0400 Subject: [PATCH 164/224] sanitize cifs_umount() prototype a) superblock argument is unused b) it always returns 0 Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 6 ++---- fs/cifs/cifsproto.h | 2 +- fs/cifs/connect.c | 5 ++--- 3 files changed, 5 insertions(+), 8 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 61c7aa870f1c..2af14d4577a7 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -172,7 +172,7 @@ out_no_root: if (inode) iput(inode); - cifs_umount(sb, cifs_sb); + cifs_umount(cifs_sb); return rc; } @@ -189,9 +189,7 @@ cifs_put_super(struct super_block *sb) return; } - rc = cifs_umount(sb, cifs_sb); - if (rc) - cERROR(1, "cifs_umount failed with return code %d", rc); + cifs_umount(cifs_sb); } static void cifs_kill_sb(struct super_block *sb) diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 5814fe543f95..257f312ede42 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -158,7 +158,7 @@ extern void cifs_cleanup_volume_info(struct smb_vol **pvolume_info); extern int cifs_setup_volume_info(struct smb_vol **pvolume_info, char *mount_data, const char *devname); extern int cifs_mount(struct cifs_sb_info *, struct smb_vol *); -extern int cifs_umount(struct super_block *, struct cifs_sb_info *); +extern void cifs_umount(struct cifs_sb_info *); extern void cifs_dfs_release_automount_timer(void); void cifs_proc_init(void); void cifs_proc_clean(void); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 9f09adf51edc..b2702226634e 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3339,8 +3339,8 @@ CIFSTCon(unsigned int xid, struct cifs_ses *ses, return rc; } -int -cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) +void +cifs_umount(struct cifs_sb_info *cifs_sb) { struct rb_root *root = &cifs_sb->tlink_tree; struct rb_node *node; @@ -3362,7 +3362,6 @@ cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) spin_unlock(&cifs_sb->tlink_tree_lock); bdi_destroy(&cifs_sb->bdi); - return 0; } int cifs_negotiate_protocol(unsigned int xid, struct cifs_ses *ses) From 97d1152acec0647b72f8c6ecc57da0d6fed574de Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:29:57 -0400 Subject: [PATCH 165/224] cifs: pull cifs_mount() call up ... to the point prior to sget(). Now we have cifs_sb set up early enough. Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 2af14d4577a7..4004bc647a72 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -104,8 +104,7 @@ cifs_sb_deactive(struct super_block *sb) } static int -cifs_read_super(struct super_block *sb, struct smb_vol *volume_info, - const char *devname, int silent) +cifs_read_super(struct super_block *sb) { struct inode *inode; struct cifs_sb_info *cifs_sb; @@ -113,14 +112,6 @@ cifs_read_super(struct super_block *sb, struct smb_vol *volume_info, cifs_sb = CIFS_SB(sb); - rc = cifs_mount(cifs_sb, volume_info); - - if (rc) { - if (!silent) - cERROR(1, "cifs_mount failed w/return code = %d", rc); - return rc; - } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIXACL) sb->s_flags |= MS_POSIXACL; @@ -692,6 +683,17 @@ cifs_do_mount(struct file_system_type *fs_type, cifs_setup_cifs_sb(volume_info, cifs_sb); + rc = cifs_mount(cifs_sb, volume_info); + if (rc) { + if (!(flags & MS_SILENT)) + cERROR(1, "cifs_mount failed w/return code = %d", rc); + root = ERR_PTR(rc); + unload_nls(volume_info->local_nls); + kfree(cifs_sb->mountdata); + kfree(cifs_sb); + goto out; + } + mnt_data.vol = volume_info; mnt_data.cifs_sb = cifs_sb; mnt_data.flags = flags; @@ -699,11 +701,13 @@ cifs_do_mount(struct file_system_type *fs_type, sb = sget(fs_type, cifs_match_super, set_anon_super, &mnt_data); if (IS_ERR(sb)) { root = ERR_CAST(sb); + cifs_umount(cifs_sb); goto out_cifs_sb; } if (sb->s_fs_info) { cFYI(1, "Use existing superblock"); + cifs_umount(cifs_sb); kfree(cifs_sb->mountdata); unload_nls(cifs_sb->local_nls); kfree(cifs_sb); @@ -715,8 +719,7 @@ cifs_do_mount(struct file_system_type *fs_type, sb->s_flags |= MS_NODIRATIME | MS_NOATIME; sb->s_fs_info = cifs_sb; - rc = cifs_read_super(sb, volume_info, dev_name, - flags & MS_SILENT ? 1 : 0); + rc = cifs_read_super(sb); if (rc) { root = ERR_PTR(rc); goto out_super; From 98ab494dd1d25388981114057cf9446250cc7dc7 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:32:10 -0400 Subject: [PATCH 166/224] cifs: move cifs_umount() call into ->kill_sb() instead of calling it manually in case if cifs_read_super() fails to set ->s_root, just call it from ->kill_sb(). cifs_put_super() is gone now *and* we have cifs_sb shutdown and destruction done after the superblock is gone from ->s_instances. Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 19 +------------------ 1 file changed, 1 insertion(+), 18 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 4004bc647a72..15de4561dbc5 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -163,30 +163,14 @@ out_no_root: if (inode) iput(inode); - cifs_umount(cifs_sb); return rc; } -static void -cifs_put_super(struct super_block *sb) -{ - int rc = 0; - struct cifs_sb_info *cifs_sb; - - cFYI(1, "In cifs_put_super"); - cifs_sb = CIFS_SB(sb); - if (cifs_sb == NULL) { - cFYI(1, "Empty cifs superblock info passed to unmount"); - return; - } - - cifs_umount(cifs_sb); -} - static void cifs_kill_sb(struct super_block *sb) { struct cifs_sb_info *cifs_sb = CIFS_SB(sb); kill_anon_super(sb); + cifs_umount(cifs_sb); kfree(cifs_sb->mountdata); unload_nls(cifs_sb->local_nls); kfree(cifs_sb); @@ -537,7 +521,6 @@ static int cifs_drop_inode(struct inode *inode) } static const struct super_operations cifs_super_ops = { - .put_super = cifs_put_super, .statfs = cifs_statfs, .alloc_inode = cifs_alloc_inode, .destroy_inode = cifs_destroy_inode, From d757d71bfc30669a500b72792067e8d1c5d401a5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:42:43 -0400 Subject: [PATCH 167/224] cifs: pull freeing mountdata/dropping nls/freeing cifs_sb into cifs_umount() all callers of cifs_umount() proceed to do the same thing; pull it into cifs_umount() itself. Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 15 +-------------- fs/cifs/connect.c | 3 +++ 2 files changed, 4 insertions(+), 14 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 15de4561dbc5..46960b7ee433 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -171,9 +171,6 @@ static void cifs_kill_sb(struct super_block *sb) struct cifs_sb_info *cifs_sb = CIFS_SB(sb); kill_anon_super(sb); cifs_umount(cifs_sb); - kfree(cifs_sb->mountdata); - unload_nls(cifs_sb->local_nls); - kfree(cifs_sb); } static int @@ -685,15 +682,12 @@ cifs_do_mount(struct file_system_type *fs_type, if (IS_ERR(sb)) { root = ERR_CAST(sb); cifs_umount(cifs_sb); - goto out_cifs_sb; + goto out; } if (sb->s_fs_info) { cFYI(1, "Use existing superblock"); cifs_umount(cifs_sb); - kfree(cifs_sb->mountdata); - unload_nls(cifs_sb->local_nls); - kfree(cifs_sb); goto out_shared; } @@ -725,13 +719,6 @@ out_shared: out_super: deactivate_locked_super(sb); - goto out; - -out_cifs_sb: - kfree(cifs_sb->mountdata); - unload_nls(cifs_sb->local_nls); - kfree(cifs_sb); - out: cifs_cleanup_volume_info(&volume_info); return root; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index b2702226634e..ca7fbe3d51a5 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3362,6 +3362,9 @@ cifs_umount(struct cifs_sb_info *cifs_sb) spin_unlock(&cifs_sb->tlink_tree_lock); bdi_destroy(&cifs_sb->bdi); + kfree(cifs_sb->mountdata); + unload_nls(cifs_sb->local_nls); + kfree(cifs_sb); } int cifs_negotiate_protocol(unsigned int xid, struct cifs_ses *ses) From ee01a14d9ddcf3f832f9ceb837888501cb496e27 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:47:23 -0400 Subject: [PATCH 168/224] cifs: close sget() races have ->s_fs_info set by the set() callback passed to sget() Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 46960b7ee433..ba2b2da360d3 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -629,6 +629,13 @@ out: return dparent; } +static int cifs_set_super(struct super_block *sb, void *data) +{ + struct cifs_mnt_data *mnt_data = data; + sb->s_fs_info = mnt_data->cifs_sb; + return set_anon_super(sb, NULL); +} + static struct dentry * cifs_do_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) @@ -678,14 +685,14 @@ cifs_do_mount(struct file_system_type *fs_type, mnt_data.cifs_sb = cifs_sb; mnt_data.flags = flags; - sb = sget(fs_type, cifs_match_super, set_anon_super, &mnt_data); + sb = sget(fs_type, cifs_match_super, cifs_set_super, &mnt_data); if (IS_ERR(sb)) { root = ERR_CAST(sb); cifs_umount(cifs_sb); goto out; } - if (sb->s_fs_info) { + if (sb->s_root) { cFYI(1, "Use existing superblock"); cifs_umount(cifs_sb); goto out_shared; @@ -694,7 +701,6 @@ cifs_do_mount(struct file_system_type *fs_type, sb->s_flags = flags; /* BB should we make this contingent on mount parm? */ sb->s_flags |= MS_NODIRATIME | MS_NOATIME; - sb->s_fs_info = cifs_sb; rc = cifs_read_super(sb); if (rc) { From fa18f1bdce898f0efd0c8639c901d826d01be04f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:50:44 -0400 Subject: [PATCH 169/224] cifs: more breakage on mount failures if cifs_get_root() fails, we end up with ->mount() returning NULL, which is not what callers expect. Moreover, in case of superblock reuse we end up leaking a superblock reference... Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index ba2b2da360d3..234e9d08db76 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -710,17 +710,14 @@ cifs_do_mount(struct file_system_type *fs_type, sb->s_flags |= MS_ACTIVE; - root = cifs_get_root(volume_info, sb); - if (root == NULL) - goto out_super; - - cFYI(1, "dentry root is: %p", root); - goto out; - out_shared: root = cifs_get_root(volume_info, sb); - if (root) - cFYI(1, "dentry root is: %p", root); + if (root == NULL) { + root = ERR_PTR(-EINVAL); /* XXX */ + goto out_super; + } + + cFYI(1, "dentry root is: %p", root); goto out; out_super: From 5c4f1ad7c6aa3b729bd3a93b80f9417d7e978c32 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 09:56:55 -0400 Subject: [PATCH 170/224] cifs: tidy cifs_do_mount() up a bit Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 46 +++++++++++++++++++++++----------------------- 1 file changed, 23 insertions(+), 23 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 234e9d08db76..9a6696a5eb78 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -656,16 +656,13 @@ cifs_do_mount(struct file_system_type *fs_type, cifs_sb = kzalloc(sizeof(struct cifs_sb_info), GFP_KERNEL); if (cifs_sb == NULL) { root = ERR_PTR(-ENOMEM); - unload_nls(volume_info->local_nls); - goto out; + goto out_nls; } cifs_sb->mountdata = kstrndup(data, PAGE_SIZE, GFP_KERNEL); if (cifs_sb->mountdata == NULL) { root = ERR_PTR(-ENOMEM); - unload_nls(volume_info->local_nls); - kfree(cifs_sb); - goto out; + goto out_cifs_sb; } cifs_setup_cifs_sb(volume_info, cifs_sb); @@ -675,10 +672,7 @@ cifs_do_mount(struct file_system_type *fs_type, if (!(flags & MS_SILENT)) cERROR(1, "cifs_mount failed w/return code = %d", rc); root = ERR_PTR(rc); - unload_nls(volume_info->local_nls); - kfree(cifs_sb->mountdata); - kfree(cifs_sb); - goto out; + goto out_mountdata; } mnt_data.vol = volume_info; @@ -695,22 +689,20 @@ cifs_do_mount(struct file_system_type *fs_type, if (sb->s_root) { cFYI(1, "Use existing superblock"); cifs_umount(cifs_sb); - goto out_shared; + } else { + sb->s_flags = flags; + /* BB should we make this contingent on mount parm? */ + sb->s_flags |= MS_NODIRATIME | MS_NOATIME; + + rc = cifs_read_super(sb); + if (rc) { + root = ERR_PTR(rc); + goto out_super; + } + + sb->s_flags |= MS_ACTIVE; } - sb->s_flags = flags; - /* BB should we make this contingent on mount parm? */ - sb->s_flags |= MS_NODIRATIME | MS_NOATIME; - - rc = cifs_read_super(sb); - if (rc) { - root = ERR_PTR(rc); - goto out_super; - } - - sb->s_flags |= MS_ACTIVE; - -out_shared: root = cifs_get_root(volume_info, sb); if (root == NULL) { root = ERR_PTR(-EINVAL); /* XXX */ @@ -725,6 +717,14 @@ out_super: out: cifs_cleanup_volume_info(&volume_info); return root; + +out_mountdata: + kfree(cifs_sb->mountdata); +out_cifs_sb: + kfree(cifs_sb); +out_nls: + unload_nls(volume_info->local_nls); + goto out; } static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, From 9403c9c598e91d473c0582066e47ed2289292e45 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 17 Jun 2011 10:02:59 -0400 Subject: [PATCH 171/224] cifs: propagate errors from cifs_get_root() to mount(2) ... instead of just failing with -EINVAL Acked-by: Pavel Shilovsky Reviewed-by: Jeff Layton Signed-off-by: Al Viro --- fs/cifs/cifsfs.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 9a6696a5eb78..35f9154615fa 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -554,7 +554,7 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) full_path = cifs_build_path_to_root(vol, cifs_sb, cifs_sb_master_tcon(cifs_sb)); if (full_path == NULL) - return NULL; + return ERR_PTR(-ENOMEM); cFYI(1, "Get root dentry for %s", full_path); @@ -583,7 +583,7 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) dchild = d_alloc(dparent, &name); if (dchild == NULL) { dput(dparent); - dparent = NULL; + dparent = ERR_PTR(-ENOMEM); goto out; } } @@ -601,7 +601,7 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) if (rc) { dput(dchild); dput(dparent); - dparent = NULL; + dparent = ERR_PTR(rc); goto out; } alias = d_materialise_unique(dchild, inode); @@ -609,7 +609,7 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) dput(dchild); if (IS_ERR(alias)) { dput(dparent); - dparent = NULL; + dparent = ERR_PTR(-EINVAL); /* XXX */ goto out; } dchild = alias; @@ -704,10 +704,8 @@ cifs_do_mount(struct file_system_type *fs_type, } root = cifs_get_root(volume_info, sb); - if (root == NULL) { - root = ERR_PTR(-EINVAL); /* XXX */ + if (IS_ERR(root)) goto out_super; - } cFYI(1, "dentry root is: %p", root); goto out; From e0f5406727f1dfdc47b8ba4a0ff6eae4b0b5ed4c Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Sat, 18 Jun 2011 20:26:38 +0000 Subject: [PATCH 172/224] Btrfs: fix type mismatch in find_free_extent() data parameter should be u64 because a full-sized chunk flags field is passed instead of 0/1 for distinguishing data from metadata. All underlying functions expect u64. Signed-off-by: Ilya Dryomov Signed-off-by: Chris Mason --- fs/btrfs/extent-tree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 1f61bf5b4960..71cd456fdb60 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -4842,7 +4842,7 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, u64 num_bytes, u64 empty_size, u64 search_start, u64 search_end, u64 hint_byte, struct btrfs_key *ins, - int data) + u64 data) { int ret = 0; struct btrfs_root *root = orig_root->fs_info->extent_root; @@ -4869,7 +4869,7 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, space_info = __find_space_info(root->fs_info, data); if (!space_info) { - printk(KERN_ERR "No space info for %d\n", data); + printk(KERN_ERR "No space info for %llu\n", data); return -ENOSPC; } From 9b90f5135320bc74dc6c9a8c74d69fd4821d9282 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 24 Jun 2011 16:02:51 +0000 Subject: [PATCH 173/224] Btrfs: make sure to update total_bitmaps when freeing cache V3 A user reported this bug again where we have more bitmaps than we are supposed to. This is because we failed to load the free space cache, but don't update the ctl->total_bitmaps counter when we remove entries from the tree. This patch fixes this problem and we should be good to go again. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/free-space-cache.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 9f985a429877..bf0d61567f3d 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -1893,9 +1893,12 @@ void __btrfs_remove_free_space_cache_locked(struct btrfs_free_space_ctl *ctl) while ((node = rb_last(&ctl->free_space_offset)) != NULL) { info = rb_entry(node, struct btrfs_free_space, offset_index); - unlink_free_space(ctl, info); - kfree(info->bitmap); - kmem_cache_free(btrfs_free_space_cachep, info); + if (!info->bitmap) { + unlink_free_space(ctl, info); + kmem_cache_free(btrfs_free_space_cachep, info); + } else { + free_bitmap(ctl, info); + } if (need_resched()) { spin_unlock(&ctl->tree_lock); cond_resched(); From 297c7f2f158f7dfa9ab5813260ff954f9c2f83d2 Mon Sep 17 00:00:00 2001 From: Ohad Ben-Cohen Date: Thu, 9 Jun 2011 23:40:27 +0000 Subject: [PATCH 174/224] mmc: sdio: fix runtime PM path during driver removal After commit e1866b3 "PM / Runtime: Rework runtime PM handling during driver removal" was introduced, the driver core stopped incrementing the runtime PM usage counter of the device during the invocation of the ->remove() callback. This indirectly broke SDIO's runtime PM path during driver removal, because no one calls _put_sync() anymore after ->remove() completes. This means that the power of runtime-PM-managed SDIO cards is kept high after their driver is removed (even if it was powered down beforehand). Fix that by directly calling _put_sync() when the last usage counter is downref'ed by the SDIO bus. Reported-and-tested-by: Daniel Drake Signed-off-by: Ohad Ben-Cohen Signed-off-by: Chris Ball --- drivers/mmc/core/sdio_bus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c index d29b9c36919a..d2565df8a7fb 100644 --- a/drivers/mmc/core/sdio_bus.c +++ b/drivers/mmc/core/sdio_bus.c @@ -189,7 +189,7 @@ static int sdio_bus_remove(struct device *dev) /* Then undo the runtime PM settings in sdio_bus_probe() */ if (func->card->host->caps & MMC_CAP_POWER_OFF_CARD) - pm_runtime_put_noidle(dev); + pm_runtime_put_sync(dev); out: return ret; From 4a3dc6ef344ccfe80bf986aec0b05f51af000720 Mon Sep 17 00:00:00 2001 From: Anand Gadiyar Date: Tue, 14 Jun 2011 15:59:59 +0530 Subject: [PATCH 175/224] mmc: omap_hsmmc: fix ocr mask usage The OMAP HSMMC driver uses an ocr_mask to determine the list of voltages supported by the card. It populates this mask based on the list of voltages supported by the regulator that supplies the voltage. Commit 64be97822b (omap4 hsmmc: Update ocr mask for MMC2 for regulator to use) passed a fixed ocr_mask from the OMAP4 SDP board file to limit the voltage to 2.9-3.0 Volts, and updated the driver to use this mask if provided, instead of using the regulator's supported voltages. However the commit is buggy - the ocr_mask is overridden by the regulator's capabilities anyway. Fix this. (The bug shows up when a system-wide suspend is attempted on the OMAP4 SDP/Blaze platforms. The eMMC card comes up at 3V, but drops to 1.65V after the system resumes). Signed-off-by: Anand Gadiyar Acked-by: Balaji T K Acked-by: Venkatraman S Tested-by: Kishore Kadiyala Signed-off-by: Sourav Poddar Signed-off-by: Chris Ball --- drivers/mmc/host/omap_hsmmc.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 5b2e2155b413..8707bcdebc43 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -429,7 +429,6 @@ static int omap_hsmmc_reg_get(struct omap_hsmmc_host *host) return -EINVAL; } } - mmc_slot(host).ocr_mask = mmc_regulator_get_ocrmask(reg); /* Allow an aux regulator */ reg = regulator_get(host->dev, "vmmc_aux"); From a9120c33ffbb0b3448d833dae392edc90d6cfac2 Mon Sep 17 00:00:00 2001 From: Per Forlin Date: Fri, 17 Jun 2011 20:14:21 +0200 Subject: [PATCH 176/224] mmc: omap_hsmmc: use original sg_len for dma_unmap_sg Don't use the returned sg_len from dma_map_sg() as inparameter to dma_unmap_sg(). Use the original sg_len for both dma_map_sg and dma_unmap_sg according to the documentation in DMA-API.txt. Signed-off-by: Per Forlin Reviewed-by: Venkatraman S Signed-off-by: Chris Ball --- drivers/mmc/host/omap_hsmmc.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 8707bcdebc43..dedf3dab8a3b 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -961,7 +961,8 @@ static void omap_hsmmc_dma_cleanup(struct omap_hsmmc_host *host, int errno) spin_unlock(&host->irq_lock); if (host->use_dma && dma_ch != -1) { - dma_unmap_sg(mmc_dev(host->mmc), host->data->sg, host->dma_len, + dma_unmap_sg(mmc_dev(host->mmc), host->data->sg, + host->data->sg_len, omap_hsmmc_get_dma_dir(host, host->data)); omap_free_dma(dma_ch); } @@ -1345,7 +1346,7 @@ static void omap_hsmmc_dma_cb(int lch, u16 ch_status, void *cb_data) return; } - dma_unmap_sg(mmc_dev(host->mmc), data->sg, host->dma_len, + dma_unmap_sg(mmc_dev(host->mmc), data->sg, data->sg_len, omap_hsmmc_get_dma_dir(host, data)); req_in_progress = host->req_in_progress; From 7d8b4c2a4b73da8e3632603691838ca5b2a8c26d Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Mon, 20 Jun 2011 16:51:10 +0200 Subject: [PATCH 177/224] mmc: tmio: fix regression in TMIO_MMC_WRPROTECT_DISABLE handling Commit b6147490e6aac82 ("mmc: tmio: split core functionality, DMA and MFD glue") broke handling of the TMIO_MMC_WRPROTECT_DISABLE flag by the tmio-mmc driver. This patch restores the original behaviour. Signed-off-by: Guennadi Liakhovetski Cc: Signed-off-by: Chris Ball --- drivers/mmc/host/tmio_mmc_pio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/host/tmio_mmc_pio.c b/drivers/mmc/host/tmio_mmc_pio.c index ad6347bb02dd..0b09e8239aa0 100644 --- a/drivers/mmc/host/tmio_mmc_pio.c +++ b/drivers/mmc/host/tmio_mmc_pio.c @@ -824,8 +824,8 @@ static int tmio_mmc_get_ro(struct mmc_host *mmc) struct tmio_mmc_host *host = mmc_priv(mmc); struct tmio_mmc_data *pdata = host->pdata; - return ((pdata->flags & TMIO_MMC_WRPROTECT_DISABLE) || - !(sd_ctrl_read32(host, CTL_STATUS) & TMIO_STAT_WRPROTECT)); + return !((pdata->flags & TMIO_MMC_WRPROTECT_DISABLE) || + (sd_ctrl_read32(host, CTL_STATUS) & TMIO_STAT_WRPROTECT)); } static int tmio_mmc_get_cd(struct mmc_host *mmc) From 3e713373ce07b9f59c3901e7d39bc1edccda28da Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Tue, 24 May 2011 12:24:07 +0200 Subject: [PATCH 178/224] mmc: sdhi: DMA slave ID 0 is invalid Don't try to allocate DMA resources if the platform didn't specify positive DMA slave IDs. Signed-off-by: Guennadi Liakhovetski Signed-off-by: Chris Ball --- drivers/mmc/host/sh_mobile_sdhi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/host/sh_mobile_sdhi.c b/drivers/mmc/host/sh_mobile_sdhi.c index 2353bb797d6c..ce500f03df85 100644 --- a/drivers/mmc/host/sh_mobile_sdhi.c +++ b/drivers/mmc/host/sh_mobile_sdhi.c @@ -92,7 +92,7 @@ static int __devinit sh_mobile_sdhi_probe(struct platform_device *pdev) mmc_data->ocr_mask = p->tmio_ocr_mask; mmc_data->capabilities |= p->tmio_caps; - if (p->dma_slave_tx >= 0 && p->dma_slave_rx >= 0) { + if (p->dma_slave_tx > 0 && p->dma_slave_rx > 0) { priv->param_tx.slave_id = p->dma_slave_tx; priv->param_rx.slave_id = p->dma_slave_rx; priv->dma_priv.chan_priv_tx = &priv->param_tx; From e9e8bcb8178e197d889ec31e79fa1ddc1732c8f9 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Tue, 21 Jun 2011 10:55:34 +0100 Subject: [PATCH 179/224] mmc: cb710: fix #ifdef HAVE_EFFICIENT_UNALIGNED_ACCESS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit HAVE_EFFICIENT_UNALIGNED_ACCESS is a config option, therefore it needs the CONFIG_ before it when used by the preprocessor. Signed-off-by: James Hogan Acked-by: Michał Mirosław Signed-off-by: Chris Ball --- drivers/misc/cb710/sgbuf2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/misc/cb710/sgbuf2.c b/drivers/misc/cb710/sgbuf2.c index d019746551f3..2a40d0efdff5 100644 --- a/drivers/misc/cb710/sgbuf2.c +++ b/drivers/misc/cb710/sgbuf2.c @@ -47,7 +47,7 @@ static uint32_t sg_dwiter_read_buffer(struct sg_mapping_iter *miter) static inline bool needs_unaligned_copy(const void *ptr) { -#ifdef HAVE_EFFICIENT_UNALIGNED_ACCESS +#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS return false; #else return ((ptr - NULL) & 3) != 0; From c6e633ad916e2af244dbfd11abd2bc077870bdfd Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sat, 25 Jun 2011 19:20:11 +0100 Subject: [PATCH 180/224] mmc: sdio: reset card during power_restore mmc_sdio_power_restore() skips some steps that are performed in other power-related codepaths which are necessary to fully reset the card. Without this, runtime PM fails for SD8686 SDIO wifi on OLPC XO-1.5. Signed-off-by: Daniel Drake Signed-off-by: Chris Ball --- drivers/mmc/core/sdio.c | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index 4d0c15bfa514..262fff019177 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -691,15 +691,54 @@ static int mmc_sdio_resume(struct mmc_host *host) static int mmc_sdio_power_restore(struct mmc_host *host) { int ret; + u32 ocr; BUG_ON(!host); BUG_ON(!host->card); mmc_claim_host(host); + + /* + * Reset the card by performing the same steps that are taken by + * mmc_rescan_try_freq() and mmc_attach_sdio() during a "normal" probe. + * + * sdio_reset() is technically not needed. Having just powered up the + * hardware, it should already be in reset state. However, some + * platforms (such as SD8686 on OLPC) do not instantly cut power, + * meaning that a reset is required when restoring power soon after + * powering off. It is harmless in other cases. + * + * The CMD5 reset (mmc_send_io_op_cond()), according to the SDIO spec, + * is not necessary for non-removable cards. However, it is required + * for OLPC SD8686 (which expects a [CMD5,5,3,7] init sequence), and + * harmless in other situations. + * + * With these steps taken, mmc_select_voltage() is also required to + * restore the correct voltage setting of the card. + */ + sdio_reset(host); + mmc_go_idle(host); + mmc_send_if_cond(host, host->ocr_avail); + + ret = mmc_send_io_op_cond(host, 0, &ocr); + if (ret) + goto out; + + if (host->ocr_avail_sdio) + host->ocr_avail = host->ocr_avail_sdio; + + host->ocr = mmc_select_voltage(host, ocr & ~0x7F); + if (!host->ocr) { + ret = -EINVAL; + goto out; + } + ret = mmc_sdio_init_card(host, host->ocr, host->card, mmc_card_keep_power(host)); if (!ret && host->sdio_irqs) mmc_signal_sdio_irq(host); + +out: mmc_release_host(host); return ret; From ddd6fa7e794e62af3ec3eb4ffdc78489885701f2 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 23 Jun 2011 13:40:26 +0300 Subject: [PATCH 181/224] mmc: block: switch card to User Data Area when removing the block driver The MMC block driver and other drivers (e.g. mmc-test) will expect the card to be switched to the User Data Area eMMC partition when they start. Hence the MMC block driver should ensure it is that way when it is removed. Signed-off-by: Adrian Hunter Cc: Andrei Warkentin Signed-off-by: Chris Ball --- drivers/mmc/card/block.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index 71da5641e258..77cd083a51e4 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -1297,6 +1297,9 @@ static void mmc_blk_remove(struct mmc_card *card) struct mmc_blk_data *md = mmc_get_drvdata(card); mmc_blk_remove_parts(card, md); + mmc_claim_host(card->host); + mmc_blk_part_switch(card, md); + mmc_release_host(card->host); mmc_blk_remove_req(md); mmc_set_drvdata(card, NULL); } From 4cf8c6dd2e261da94b87c4deadcc136ab022b6ac Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 23 Jun 2011 13:40:27 +0300 Subject: [PATCH 182/224] mmc: core: make erase timeout calculation allow for gated clock The erase timeout calculation may depend on clock rate which is zero if the clock is gated, so use mmc_host_clk_rate() which allows for that case. Signed-off-by: Adrian Hunter Signed-off-by: Chris Ball --- drivers/mmc/core/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 68091dda3f31..7843efe22359 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -1245,7 +1245,7 @@ static unsigned int mmc_mmc_erase_timeout(struct mmc_card *card, */ timeout_clks <<= 1; timeout_us += (timeout_clks * 1000) / - (card->host->ios.clock / 1000); + (mmc_host_clk_rate(card->host) / 1000); erase_timeout = timeout_us / 1000; From d09408ade08a08a710a247fb52aa50101e73ebf7 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 23 Jun 2011 13:40:28 +0300 Subject: [PATCH 183/224] mmc: queue: append partition subname to queue thread name For example, an eMMC with 2 boot partitions will have 3 threads. The names change from: 40 ? 00:00:00 mmcqd/0 41 ? 00:00:00 mmcqd/0 42 ? 00:00:00 mmcqd/0 to: 40 ? 00:00:00 mmcqd/0 41 ? 00:00:00 mmcqd/0boot0 42 ? 00:00:00 mmcqd/0boot1 Signed-off-by: Adrian Hunter Cc: Andrei Warkentin Signed-off-by: Chris Ball --- drivers/mmc/card/block.c | 2 +- drivers/mmc/card/queue.c | 8 +++++--- drivers/mmc/card/queue.h | 3 ++- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index 77cd083a51e4..f85e42224559 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -1024,7 +1024,7 @@ static struct mmc_blk_data *mmc_blk_alloc_req(struct mmc_card *card, INIT_LIST_HEAD(&md->part); md->usage = 1; - ret = mmc_init_queue(&md->queue, card, &md->lock); + ret = mmc_init_queue(&md->queue, card, &md->lock, subname); if (ret) goto err_putdisk; diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c index c07322c2658c..3e2db1cf1cbb 100644 --- a/drivers/mmc/card/queue.c +++ b/drivers/mmc/card/queue.c @@ -106,10 +106,12 @@ static void mmc_request(struct request_queue *q) * @mq: mmc queue * @card: mmc card to attach this queue * @lock: queue lock + * @subname: partition subname * * Initialise a MMC card request queue. */ -int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock) +int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, + spinlock_t *lock, const char *subname) { struct mmc_host *host = card->host; u64 limit = BLK_BOUNCE_HIGH; @@ -209,8 +211,8 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock sema_init(&mq->thread_sem, 1); - mq->thread = kthread_run(mmc_queue_thread, mq, "mmcqd/%d", - host->index); + mq->thread = kthread_run(mmc_queue_thread, mq, "mmcqd/%d%s", + host->index, subname ? subname : ""); if (IS_ERR(mq->thread)) { ret = PTR_ERR(mq->thread); diff --git a/drivers/mmc/card/queue.h b/drivers/mmc/card/queue.h index 64e66e0d4994..6223ef8dc9cd 100644 --- a/drivers/mmc/card/queue.h +++ b/drivers/mmc/card/queue.h @@ -19,7 +19,8 @@ struct mmc_queue { unsigned int bounce_sg_len; }; -extern int mmc_init_queue(struct mmc_queue *, struct mmc_card *, spinlock_t *); +extern int mmc_init_queue(struct mmc_queue *, struct mmc_card *, spinlock_t *, + const char *); extern void mmc_cleanup_queue(struct mmc_queue *); extern void mmc_queue_suspend(struct mmc_queue *); extern void mmc_queue_resume(struct mmc_queue *); From c31b55cd4eaf050bb5a15bd8251da1b3c7edeb1c Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 23 Jun 2011 13:40:29 +0300 Subject: [PATCH 184/224] mmc: queue: bring discard_granularity/alignment into line with SCSI SCSI defines discard alignment as the offset to the first optimal discard. In the case of SD/MMC, that is always zero which is the default. SCSI defines discard granularity as a hint of a optimal discard size. That is much better expressed by the MMC "preferred erase size" (pref_erase) field. Signed-off-by: Adrian Hunter Signed-off-by: Chris Ball --- drivers/mmc/card/queue.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c index 3e2db1cf1cbb..6413afa318d2 100644 --- a/drivers/mmc/card/queue.c +++ b/drivers/mmc/card/queue.c @@ -135,12 +135,7 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, mq->queue->limits.max_discard_sectors = UINT_MAX; if (card->erased_byte == 0) mq->queue->limits.discard_zeroes_data = 1; - if (!mmc_can_trim(card) && is_power_of_2(card->erase_size)) { - mq->queue->limits.discard_granularity = - card->erase_size << 9; - mq->queue->limits.discard_alignment = - card->erase_size << 9; - } + mq->queue->limits.discard_granularity = card->pref_erase << 9; if (mmc_can_secure_erase_trim(card)) queue_flag_set_unlocked(QUEUE_FLAG_SECDISCARD, mq->queue); From 2f7e33d432d097a2a7f467b031bf18be91cb3d49 Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Thu, 23 Jun 2011 07:27:13 +0000 Subject: [PATCH 185/224] btrfs: fix inconsonant inode information When iputting the inode, We may leave the delayed nodes if they have some delayed items that have not been dealt with. So when the inode is read again, we must look up the relative delayed node, and use the information in it to initialize the inode. Or we will get inconsonant inode information, it may cause that the same directory index number is allocated again, and hit the following oops: [ 5447.554187] err add delayed dir index item(name: pglog_0.965_0) into the insertion tree of the delayed node(root id: 262, inode id: 258, errno: -17) [ 5447.569766] ------------[ cut here ]------------ [ 5447.575361] kernel BUG at fs/btrfs/delayed-inode.c:1301! [SNIP] [ 5447.790721] Call Trace: [ 5447.793191] [] btrfs_insert_dir_item+0x189/0x1bb [btrfs] [ 5447.800156] [] btrfs_add_link+0x12b/0x191 [btrfs] [ 5447.806517] [] btrfs_add_nondir+0x31/0x58 [btrfs] [ 5447.812876] [] btrfs_create+0xf9/0x197 [btrfs] [ 5447.818961] [] vfs_create+0x72/0x92 [ 5447.824090] [] do_last+0x22c/0x40b [ 5447.829133] [] path_openat+0xc0/0x2ef [ 5447.834438] [] ? __perf_event_task_sched_out+0x24/0x44 [ 5447.841216] [] ? perf_event_task_sched_out+0x59/0x67 [ 5447.847846] [] do_filp_open+0x3d/0x87 [ 5447.853156] [] ? strncpy_from_user+0x43/0x4d [ 5447.859072] [] ? getname_flags+0x2e/0x80 [ 5447.864636] [] ? do_getname+0x14b/0x173 [ 5447.870112] [] ? audit_getname+0x16/0x26 [ 5447.875682] [] ? spin_lock+0xe/0x10 [ 5447.880882] [] do_sys_open+0x69/0xae [ 5447.886153] [] sys_open+0x20/0x22 [ 5447.891114] [] system_call_fastpath+0x16/0x1b Fix it by reusing the old delayed node. Reported-by: Jim Schutt Signed-off-by: Miao Xie Tested-by: Jim Schutt Signed-off-by: Chris Mason --- fs/btrfs/delayed-inode.c | 128 ++++++++++++++++++++++++++++----------- fs/btrfs/delayed-inode.h | 1 + fs/btrfs/inode.c | 12 +++- 3 files changed, 103 insertions(+), 38 deletions(-) diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index f1cbd028f7b3..98c68e658a9b 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -82,6 +82,39 @@ static inline struct btrfs_delayed_root *btrfs_get_delayed_root( return root->fs_info->delayed_root; } +static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode) +{ + struct btrfs_inode *btrfs_inode = BTRFS_I(inode); + struct btrfs_root *root = btrfs_inode->root; + u64 ino = btrfs_ino(inode); + struct btrfs_delayed_node *node; + + node = ACCESS_ONCE(btrfs_inode->delayed_node); + if (node) { + atomic_inc(&node->refs); + return node; + } + + spin_lock(&root->inode_lock); + node = radix_tree_lookup(&root->delayed_nodes_tree, ino); + if (node) { + if (btrfs_inode->delayed_node) { + atomic_inc(&node->refs); /* can be accessed */ + BUG_ON(btrfs_inode->delayed_node != node); + spin_unlock(&root->inode_lock); + return node; + } + btrfs_inode->delayed_node = node; + atomic_inc(&node->refs); /* can be accessed */ + atomic_inc(&node->refs); /* cached in the inode */ + spin_unlock(&root->inode_lock); + return node; + } + spin_unlock(&root->inode_lock); + + return NULL; +} + static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node( struct inode *inode) { @@ -92,26 +125,9 @@ static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node( int ret; again: - node = ACCESS_ONCE(btrfs_inode->delayed_node); - if (node) { - atomic_inc(&node->refs); /* can be accessed */ + node = btrfs_get_delayed_node(inode); + if (node) return node; - } - - spin_lock(&root->inode_lock); - node = radix_tree_lookup(&root->delayed_nodes_tree, ino); - if (node) { - if (btrfs_inode->delayed_node) { - spin_unlock(&root->inode_lock); - goto again; - } - btrfs_inode->delayed_node = node; - atomic_inc(&node->refs); /* can be accessed */ - atomic_inc(&node->refs); /* cached in the inode */ - spin_unlock(&root->inode_lock); - return node; - } - spin_unlock(&root->inode_lock); node = kmem_cache_alloc(delayed_node_cache, GFP_NOFS); if (!node) @@ -548,19 +564,6 @@ struct btrfs_delayed_item *__btrfs_next_delayed_item( return next; } -static inline struct btrfs_delayed_node *btrfs_get_delayed_node( - struct inode *inode) -{ - struct btrfs_inode *btrfs_inode = BTRFS_I(inode); - struct btrfs_delayed_node *delayed_node; - - delayed_node = btrfs_inode->delayed_node; - if (delayed_node) - atomic_inc(&delayed_node->refs); - - return delayed_node; -} - static inline struct btrfs_root *btrfs_get_fs_root(struct btrfs_root *root, u64 root_id) { @@ -1404,8 +1407,7 @@ end: int btrfs_inode_delayed_dir_index_count(struct inode *inode) { - struct btrfs_delayed_node *delayed_node = BTRFS_I(inode)->delayed_node; - int ret = 0; + struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode); if (!delayed_node) return -ENOENT; @@ -1415,11 +1417,14 @@ int btrfs_inode_delayed_dir_index_count(struct inode *inode) * a new directory index is added into the delayed node and index_cnt * is updated now. So we needn't lock the delayed node. */ - if (!delayed_node->index_cnt) + if (!delayed_node->index_cnt) { + btrfs_release_delayed_node(delayed_node); return -EINVAL; + } BTRFS_I(inode)->index_cnt = delayed_node->index_cnt; - return ret; + btrfs_release_delayed_node(delayed_node); + return 0; } void btrfs_get_delayed_items(struct inode *inode, struct list_head *ins_list, @@ -1613,6 +1618,57 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans, inode->i_ctime.tv_nsec); } +int btrfs_fill_inode(struct inode *inode, u32 *rdev) +{ + struct btrfs_delayed_node *delayed_node; + struct btrfs_inode_item *inode_item; + struct btrfs_timespec *tspec; + + delayed_node = btrfs_get_delayed_node(inode); + if (!delayed_node) + return -ENOENT; + + mutex_lock(&delayed_node->mutex); + if (!delayed_node->inode_dirty) { + mutex_unlock(&delayed_node->mutex); + btrfs_release_delayed_node(delayed_node); + return -ENOENT; + } + + inode_item = &delayed_node->inode_item; + + inode->i_uid = btrfs_stack_inode_uid(inode_item); + inode->i_gid = btrfs_stack_inode_gid(inode_item); + btrfs_i_size_write(inode, btrfs_stack_inode_size(inode_item)); + inode->i_mode = btrfs_stack_inode_mode(inode_item); + inode->i_nlink = btrfs_stack_inode_nlink(inode_item); + inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item)); + BTRFS_I(inode)->generation = btrfs_stack_inode_generation(inode_item); + BTRFS_I(inode)->sequence = btrfs_stack_inode_sequence(inode_item); + inode->i_rdev = 0; + *rdev = btrfs_stack_inode_rdev(inode_item); + BTRFS_I(inode)->flags = btrfs_stack_inode_flags(inode_item); + + tspec = btrfs_inode_atime(inode_item); + inode->i_atime.tv_sec = btrfs_stack_timespec_sec(tspec); + inode->i_atime.tv_nsec = btrfs_stack_timespec_nsec(tspec); + + tspec = btrfs_inode_mtime(inode_item); + inode->i_mtime.tv_sec = btrfs_stack_timespec_sec(tspec); + inode->i_mtime.tv_nsec = btrfs_stack_timespec_nsec(tspec); + + tspec = btrfs_inode_ctime(inode_item); + inode->i_ctime.tv_sec = btrfs_stack_timespec_sec(tspec); + inode->i_ctime.tv_nsec = btrfs_stack_timespec_nsec(tspec); + + inode->i_generation = BTRFS_I(inode)->generation; + BTRFS_I(inode)->index_cnt = (u64)-1; + + mutex_unlock(&delayed_node->mutex); + btrfs_release_delayed_node(delayed_node); + return 0; +} + int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode) { diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h index d1a6a2915c66..8d27af4bd8b9 100644 --- a/fs/btrfs/delayed-inode.h +++ b/fs/btrfs/delayed-inode.h @@ -119,6 +119,7 @@ void btrfs_kill_delayed_inode_items(struct inode *inode); int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode); +int btrfs_fill_inode(struct inode *inode, u32 *rdev); /* Used for drop dead root */ void btrfs_kill_all_delayed_nodes(struct btrfs_root *root); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 87f1e0cf26f8..447612d3a16a 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2509,6 +2509,11 @@ static void btrfs_read_locked_inode(struct inode *inode) int maybe_acls; u32 rdev; int ret; + bool filled = false; + + ret = btrfs_fill_inode(inode, &rdev); + if (!ret) + filled = true; path = btrfs_alloc_path(); BUG_ON(!path); @@ -2520,6 +2525,10 @@ static void btrfs_read_locked_inode(struct inode *inode) goto make_bad; leaf = path->nodes[0]; + + if (filled) + goto cache_acl; + inode_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_inode_item); if (!leaf->map_token) @@ -2556,7 +2565,7 @@ static void btrfs_read_locked_inode(struct inode *inode) BTRFS_I(inode)->index_cnt = (u64)-1; BTRFS_I(inode)->flags = btrfs_inode_flags(leaf, inode_item); - +cache_acl: /* * try to precache a NULL acl entry for files that don't have * any xattrs or acls @@ -2572,7 +2581,6 @@ static void btrfs_read_locked_inode(struct inode *inode) } btrfs_free_path(path); - inode_item = NULL; switch (inode->i_mode & S_IFMT) { case S_IFREG: From 8b1ab60c767f33d79894ef66037ef851a7a88fc8 Mon Sep 17 00:00:00 2001 From: Arvid Brodin Date: Fri, 17 Jun 2011 18:45:37 +0200 Subject: [PATCH 186/224] usb/isp1760: Fix bug preventing the unlinking of control urbs Both control and bulk transfers use isp1760 slots of type ATL, but the driver unlink code for ATL slots only acts on urbs describing a bulk transfer, letting the code for INT slots take care of the unlink instead, which often ended up removing the interrupt transfer for root hub events instead. That's not good, and gets fixed by this patch. Signed-off-by: Arvid Brodin Cc: Sebastian Andrzej Siewior Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/isp1760-hcd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/host/isp1760-hcd.c b/drivers/usb/host/isp1760-hcd.c index c9e6e454c625..55d3d5859ac5 100644 --- a/drivers/usb/host/isp1760-hcd.c +++ b/drivers/usb/host/isp1760-hcd.c @@ -1555,7 +1555,7 @@ static void kill_transfer(struct usb_hcd *hcd, struct urb *urb, /* We need to forcefully reclaim the slot since some transfers never return, e.g. interrupt transfers and NAKed bulk transfers. */ - if (usb_pipebulk(urb->pipe)) { + if (usb_pipecontrol(urb->pipe) || usb_pipebulk(urb->pipe)) { skip_map = reg_read32(hcd->regs, HC_ATL_PTD_SKIPMAP_REG); skip_map |= (1 << qh->slot); reg_write32(hcd->regs, HC_ATL_PTD_SKIPMAP_REG, skip_map); From 309427b6351b763917caac3e4b2ab5651df99823 Mon Sep 17 00:00:00 2001 From: Uwe Bonnes Date: Fri, 17 Jun 2011 17:30:23 +0200 Subject: [PATCH 187/224] USB: Add new FT232H chip to drivers/usb/serial/ftdi_sio.c appended patch adds support for the new FTDI FT232H chip. This chip is a single channel version of the dual FT2232H/quad FT4232H, coming with it's own default PID 0x6014 (FT2232H uses the same PID 0x6010 like FT2232C, FT4232H has also it's own PID). The patch was checked on an UM232H module and a terminal program with TX/RX shorted to that typing in the terminal reproduced the characters. Signed-off-by: Uwe Bonnes Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/ftdi_sio.c | 19 ++++++++++++++----- drivers/usb/serial/ftdi_sio.h | 3 ++- drivers/usb/serial/ftdi_sio_ids.h | 1 + 3 files changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 162728977553..2e06b90aa1f8 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -179,6 +179,7 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(FTDI_VID, FTDI_232RL_PID) }, { USB_DEVICE(FTDI_VID, FTDI_8U2232C_PID) }, { USB_DEVICE(FTDI_VID, FTDI_4232H_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_232H_PID) }, { USB_DEVICE(FTDI_VID, FTDI_MICRO_CHAMELEON_PID) }, { USB_DEVICE(FTDI_VID, FTDI_RELAIS_PID) }, { USB_DEVICE(FTDI_VID, FTDI_OPENDCC_PID) }, @@ -848,7 +849,8 @@ static const char *ftdi_chip_name[] = { [FT2232C] = "FT2232C", [FT232RL] = "FT232RL", [FT2232H] = "FT2232H", - [FT4232H] = "FT4232H" + [FT4232H] = "FT4232H", + [FT232H] = "FT232H" }; @@ -1168,6 +1170,7 @@ static __u32 get_ftdi_divisor(struct tty_struct *tty, break; case FT2232H: /* FT2232H chip */ case FT4232H: /* FT4232H chip */ + case FT232H: /* FT232H chip */ if ((baud <= 12000000) & (baud >= 1200)) { div_value = ftdi_2232h_baud_to_divisor(baud); } else if (baud < 1200) { @@ -1429,9 +1432,12 @@ static void ftdi_determine_type(struct usb_serial_port *port) } else if (version < 0x600) { /* Assume it's an FT232BM (or FT245BM) */ priv->chip_type = FT232BM; - } else { - /* Assume it's an FT232R */ + } else if (version < 0x900) { + /* Assume it's an FT232RL */ priv->chip_type = FT232RL; + } else { + /* Assume it's an FT232H */ + priv->chip_type = FT232H; } dev_info(&udev->dev, "Detected %s\n", ftdi_chip_name[priv->chip_type]); } @@ -1559,7 +1565,8 @@ static int create_sysfs_attrs(struct usb_serial_port *port) priv->chip_type == FT2232C || priv->chip_type == FT232RL || priv->chip_type == FT2232H || - priv->chip_type == FT4232H)) { + priv->chip_type == FT4232H || + priv->chip_type == FT232H)) { retval = device_create_file(&port->dev, &dev_attr_latency_timer); } @@ -1580,7 +1587,8 @@ static void remove_sysfs_attrs(struct usb_serial_port *port) priv->chip_type == FT2232C || priv->chip_type == FT232RL || priv->chip_type == FT2232H || - priv->chip_type == FT4232H) { + priv->chip_type == FT4232H || + priv->chip_type == FT232H) { device_remove_file(&port->dev, &dev_attr_latency_timer); } } @@ -2212,6 +2220,7 @@ static int ftdi_tiocmget(struct tty_struct *tty) case FT232RL: case FT2232H: case FT4232H: + case FT232H: len = 2; break; default: diff --git a/drivers/usb/serial/ftdi_sio.h b/drivers/usb/serial/ftdi_sio.h index 213fe3d61282..19584faa86f9 100644 --- a/drivers/usb/serial/ftdi_sio.h +++ b/drivers/usb/serial/ftdi_sio.h @@ -156,7 +156,8 @@ enum ftdi_chip_type { FT2232C = 4, FT232RL = 5, FT2232H = 6, - FT4232H = 7 + FT4232H = 7, + FT232H = 8 }; enum ftdi_sio_baudrate { diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h index ab1fcdf3c378..19156d1049fe 100644 --- a/drivers/usb/serial/ftdi_sio_ids.h +++ b/drivers/usb/serial/ftdi_sio_ids.h @@ -22,6 +22,7 @@ #define FTDI_8U232AM_ALT_PID 0x6006 /* FTDI's alternate PID for above */ #define FTDI_8U2232C_PID 0x6010 /* Dual channel device */ #define FTDI_4232H_PID 0x6011 /* Quad channel hi-speed device */ +#define FTDI_232H_PID 0x6014 /* Single channel hi-speed device */ #define FTDI_SIO_PID 0x8372 /* Product Id SIO application of 8U100AX */ #define FTDI_232RL_PID 0xFBFA /* Product ID for FT232RL */ From c27d5368ef1f7ea40bcd1fb2626f4c99d77e0080 Mon Sep 17 00:00:00 2001 From: Gabor Juhos Date: Mon, 20 Jun 2011 17:22:36 +0200 Subject: [PATCH 188/224] USB: ehci-ath79: fix a NULL pointer dereference Loading the ehci-hcd module on the ath79 platform causes a NULL pointer dereference: CPU 0 Unable to handle kernel paging request at virtual address 00000000, epc == c0252928, ra == c00de968 Oops[#1]: Cpu 0 $ 0 : 00000000 00000070 00000001 00000000 $ 4 : 802cf870 0000117e ffffffff 8019c7bc $ 8 : 0000000a 00000002 00000001 fffffffb $12 : 8026ef20 0000000f ffffff80 802dad3c $16 : 8077a2d4 8077a200 c00f3484 8019ed84 $20 : c00f0000 00000003 000000a0 80262c2c $24 : 00000002 80079da0 $28 : 80788000 80789c80 80262b14 c00de968 Hi : 00000000 Lo : b61f0000 epc : c0252928 __mod_vermagic5+0xc260/0xc7e8 [ehci_hcd] Not tainted ra : c00de968 usb_add_hcd+0x2a4/0x858 [usbcore] Status: 1000c003 KERNEL EXL IE Cause : 00800008 BadVA : 00000000 PrId : 00019374 (MIPS 24Kc) Modules linked in: ehci_hcd(+) pppoe pppox ipt_REJECT xt_TCPMSS ipt_LOG xt_comment xt_multiport xt_mac xt_limit iptable_mangle iptable_filte r ip_tables xt_tcpudp x_tables ppp_async ppp_generic slhc ath mac80211 usbcore nls_base input_polldev crc_ccitt cfg80211 compat input_core a rc4 aes_generic crypto_algapi Process insmod (pid: 379, threadinfo=80788000, task=80ca2180, tls=77fe52d0) Stack : c0253184 80c57d80 80789cac 8077a200 00000001 8019edc0 807fa800 8077a200 8077a290 c00f3484 8019ed84 c00f0000 00000003 000000a0 80262c2c c00de968 802d0000 800878cc c0253228 c02528e4 c0253184 80c57d80 80bf6800 80ca2180 8007b75c 00000000 8077a200 802cf830 802d0000 00000003 fffffff4 00000015 00000348 00000124 800b189c c024bb4c c0255000 801a27e8 c0253228 c02528e4 ... Call Trace: [] __mod_vermagic5+0xc260/0xc7e8 [ehci_hcd] It is caused by: commit c430131a02d677aa708f56342c1565edfdacb3c0 Author: Jan Andersson Date: Tue May 3 20:11:57 2011 +0200 USB: EHCI: Support controllers with big endian capability regs The two first HC capability registers (CAPLENGTH and HCIVERSION) are defined as one 8-bit and one 16-bit register. Most HC implementations have selected to treat these registers as part of a 32-bit register, giving the same layout for both big and small endian systems. This patch adds a new quirk, big_endian_capbase, to support controllers with big endian register interfaces that treat HCIVERSION and CAPLENGTH as individual registers. Signed-off-by: Jan Andersson Acked-by: Alan Stern Signed-off-by: Greg Kroah-Hartman The reading of the HC capability register has been moved by that commit to a place where the ehci->caps field is not initialized yet. This patch moves the reading of the register back to the original place. Acked-by: Jan Andersson Cc: Alan Stern Signed-off-by: Gabor Juhos Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/ehci-ath79.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/usb/host/ehci-ath79.c b/drivers/usb/host/ehci-ath79.c index 98cc8a13169c..aa248c2f2c60 100644 --- a/drivers/usb/host/ehci-ath79.c +++ b/drivers/usb/host/ehci-ath79.c @@ -44,7 +44,6 @@ static int ehci_ath79_init(struct usb_hcd *hcd) struct ehci_hcd *ehci = hcd_to_ehci(hcd); struct platform_device *pdev = to_platform_device(hcd->self.controller); const struct platform_device_id *id; - int hclength; int ret; id = platform_get_device_id(pdev); @@ -53,20 +52,23 @@ static int ehci_ath79_init(struct usb_hcd *hcd) return -EINVAL; } - hclength = HC_LENGTH(ehci, ehci_readl(ehci, &ehci->caps->hc_capbase)); switch (id->driver_data) { case EHCI_ATH79_IP_V1: ehci->has_synopsys_hc_bug = 1; ehci->caps = hcd->regs; - ehci->regs = hcd->regs + hclength; + ehci->regs = hcd->regs + + HC_LENGTH(ehci, + ehci_readl(ehci, &ehci->caps->hc_capbase)); break; case EHCI_ATH79_IP_V2: hcd->has_tt = 1; ehci->caps = hcd->regs + 0x100; - ehci->regs = hcd->regs + 0x100 + hclength; + ehci->regs = hcd->regs + 0x100 + + HC_LENGTH(ehci, + ehci_readl(ehci, &ehci->caps->hc_capbase)); break; default: From 87abd0a92c72eea71542d6c972c67e39e584e989 Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Tue, 21 Jun 2011 13:24:42 +0900 Subject: [PATCH 189/224] usb: r8a66597-hcd: fix cannot detect low/full speed device This controller can control "Transaction Translators", but the hcd->has_tt is not set. Since the commit d199c96d41d80a567493e12b8e96ea056a1350c1 ("USB: prevent buggy from crashing the USB stack") has checked it, the driver could not work the low/full speed device. Signed-off-by: Yoshihiro Shimoda Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/r8a66597-hcd.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/host/r8a66597-hcd.c b/drivers/usb/host/r8a66597-hcd.c index db6f8b9c19b6..4586369dda00 100644 --- a/drivers/usb/host/r8a66597-hcd.c +++ b/drivers/usb/host/r8a66597-hcd.c @@ -2517,6 +2517,7 @@ static int __devinit r8a66597_probe(struct platform_device *pdev) INIT_LIST_HEAD(&r8a66597->child_device); hcd->rsrc_start = res->start; + hcd->has_tt = 1; ret = usb_add_hcd(hcd, irq, IRQF_DISABLED | irq_trigger); if (ret != 0) { From 857aab34f04ff86666aa80e751ee696eff0113dd Mon Sep 17 00:00:00 2001 From: matt mooney Date: Fri, 17 Jun 2011 15:50:46 -0700 Subject: [PATCH 190/224] MAINTAINERS: add myself as maintainer of USB/IP Signed-off-by: matt mooney Signed-off-by: Greg Kroah-Hartman --- MAINTAINERS | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 4307673415f7..1f93f5084d26 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6462,6 +6462,12 @@ S: Maintained F: Documentation/usb/hiddev.txt F: drivers/hid/usbhid/ +USB/IP DRIVERS +M: Matt Mooney +L: linux-usb@vger.kernel.org +S: Maintained +F: drivers/staging/usbip/ + USB ISP116X DRIVER M: Olav Kongas L: linux-usb@vger.kernel.org From c6830c22603aaecf65405af23f6da2d55892f9cb Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 16 Jun 2011 17:28:07 +0900 Subject: [PATCH 191/224] Fix node_start/end_pfn() definition for mm/page_cgroup.c commit 21a3c96 uses node_start/end_pfn(nid) for detection start/end of nodes. But, it's not defined in linux/mmzone.h but defined in /arch/???/include/mmzone.h which is included only under CONFIG_NEED_MULTIPLE_NODES=y. Then, we see mm/page_cgroup.c: In function 'page_cgroup_init': mm/page_cgroup.c:308: error: implicit declaration of function 'node_start_pfn' mm/page_cgroup.c:309: error: implicit declaration of function 'node_end_pfn' So, fixiing page_cgroup.c is an idea... But node_start_pfn()/node_end_pfn() is a very generic macro and should be implemented in the same manner for all archs. (m32r has different implementation...) This patch removes definitions of node_start/end_pfn() in each archs and defines a unified one in linux/mmzone.h. It's not under CONFIG_NEED_MULTIPLE_NODES, now. A result of macro expansion is here (mm/page_cgroup.c) for !NUMA start_pfn = ((&contig_page_data)->node_start_pfn); end_pfn = ({ pg_data_t *__pgdat = (&contig_page_data); __pgdat->node_start_pfn + __pgdat->node_spanned_pages;}); for NUMA (x86-64) start_pfn = ((node_data[nid])->node_start_pfn); end_pfn = ({ pg_data_t *__pgdat = (node_data[nid]); __pgdat->node_start_pfn + __pgdat->node_spanned_pages;}); Changelog: - fixed to avoid using "nid" twice in node_end_pfn() macro. Reported-and-acked-by: Randy Dunlap Reported-and-tested-by: Ingo Molnar Acked-by: Mel Gorman Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Linus Torvalds --- arch/alpha/include/asm/mmzone.h | 1 - arch/m32r/include/asm/mmzone.h | 8 +------- arch/parisc/include/asm/mmzone.h | 7 ------- arch/powerpc/include/asm/mmzone.h | 7 ------- arch/sh/include/asm/mmzone.h | 4 ---- arch/sparc/include/asm/mmzone.h | 2 -- arch/tile/include/asm/mmzone.h | 11 ----------- arch/x86/include/asm/mmzone_32.h | 11 ----------- arch/x86/include/asm/mmzone_64.h | 3 --- include/linux/mmzone.h | 7 +++++++ 10 files changed, 8 insertions(+), 53 deletions(-) diff --git a/arch/alpha/include/asm/mmzone.h b/arch/alpha/include/asm/mmzone.h index 8af56ce346ad..445dc42e0334 100644 --- a/arch/alpha/include/asm/mmzone.h +++ b/arch/alpha/include/asm/mmzone.h @@ -56,7 +56,6 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n) * Given a kernel address, find the home node of the underlying memory. */ #define kvaddr_to_nid(kaddr) pa_to_nid(__pa(kaddr)) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) /* * Given a kaddr, LOCAL_BASE_ADDR finds the owning node of the memory diff --git a/arch/m32r/include/asm/mmzone.h b/arch/m32r/include/asm/mmzone.h index 9f3b5accda88..115ced33febd 100644 --- a/arch/m32r/include/asm/mmzone.h +++ b/arch/m32r/include/asm/mmzone.h @@ -14,12 +14,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) #define node_localnr(pfn, nid) ((pfn) - NODE_DATA(nid)->node_start_pfn) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages - 1; \ -}) #define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT)) /* @@ -44,7 +38,7 @@ static __inline__ int pfn_to_nid(unsigned long pfn) int node; for (node = 0 ; node < MAX_NUMNODES ; node++) - if (pfn >= node_start_pfn(node) && pfn <= node_end_pfn(node)) + if (pfn >= node_start_pfn(node) && pfn < node_end_pfn(node)) break; return node; diff --git a/arch/parisc/include/asm/mmzone.h b/arch/parisc/include/asm/mmzone.h index 9608d2cf214a..e67eb9c3d1bf 100644 --- a/arch/parisc/include/asm/mmzone.h +++ b/arch/parisc/include/asm/mmzone.h @@ -14,13 +14,6 @@ extern struct node_map_data node_data[]; #define NODE_DATA(nid) (&node_data[nid].pg_data) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages; \ -}) - /* We have these possible memory map layouts: * Astro: 0-3.75, 67.75-68, 4-64 * zx1: 0-1, 257-260, 4-256 diff --git a/arch/powerpc/include/asm/mmzone.h b/arch/powerpc/include/asm/mmzone.h index fd3fd58bad84..7b589178be46 100644 --- a/arch/powerpc/include/asm/mmzone.h +++ b/arch/powerpc/include/asm/mmzone.h @@ -38,13 +38,6 @@ u64 memory_hotplug_max(void); #define memory_hotplug_max() memblock_end_of_DRAM() #endif -/* - * Following are macros that each numa implmentation must define. - */ - -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_end_pfn) - #else #define memory_hotplug_max() memblock_end_of_DRAM() #endif /* CONFIG_NEED_MULTIPLE_NODES */ diff --git a/arch/sh/include/asm/mmzone.h b/arch/sh/include/asm/mmzone.h index 8887baff5eff..15a8496960e6 100644 --- a/arch/sh/include/asm/mmzone.h +++ b/arch/sh/include/asm/mmzone.h @@ -9,10 +9,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ - NODE_DATA(nid)->node_spanned_pages) - static inline int pfn_to_nid(unsigned long pfn) { int nid; diff --git a/arch/sparc/include/asm/mmzone.h b/arch/sparc/include/asm/mmzone.h index e8c648741ed4..99d9b9f577bf 100644 --- a/arch/sparc/include/asm/mmzone.h +++ b/arch/sparc/include/asm/mmzone.h @@ -8,8 +8,6 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_end_pfn) extern int numa_cpu_lookup_table[]; extern cpumask_t numa_cpumask_lookup_table[]; diff --git a/arch/tile/include/asm/mmzone.h b/arch/tile/include/asm/mmzone.h index c6344c4f32ac..9d3dbce8f953 100644 --- a/arch/tile/include/asm/mmzone.h +++ b/arch/tile/include/asm/mmzone.h @@ -40,17 +40,6 @@ static inline int pfn_to_nid(unsigned long pfn) return highbits_to_node[__pfn_to_highbits(pfn)]; } -/* - * Following are macros that each numa implmentation must define. - */ - -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages; \ -}) - #define kern_addr_valid(kaddr) virt_addr_valid((void *)kaddr) static inline int pfn_valid(int pfn) diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 5e83a416eca8..224e8c5eb307 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -48,17 +48,6 @@ static inline int pfn_to_nid(unsigned long pfn) #endif } -/* - * Following are macros that each numa implmentation must define. - */ - -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) \ -({ \ - pg_data_t *__pgdat = NODE_DATA(nid); \ - __pgdat->node_start_pfn + __pgdat->node_spanned_pages; \ -}) - static inline int pfn_valid(int pfn) { int nid = pfn_to_nid(pfn); diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index b3f88d7867c7..129d9aa3ceb3 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -13,8 +13,5 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) -#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) -#define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ - NODE_DATA(nid)->node_spanned_pages) #endif #endif /* _ASM_X86_MMZONE_64_H */ diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index c928dac6cad0..9f7c3ebcbbad 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -647,6 +647,13 @@ typedef struct pglist_data { #endif #define nid_page_nr(nid, pagenr) pgdat_page_nr(NODE_DATA(nid),(pagenr)) +#define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) + +#define node_end_pfn(nid) ({\ + pg_data_t *__pgdat = NODE_DATA(nid);\ + __pgdat->node_start_pfn + __pgdat->node_spanned_pages;\ +}) + #include extern struct mutex zonelists_mutex; From 89089158d191ba0195b33ebf35f91ed217d44848 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 24 Jun 2011 09:28:32 -0700 Subject: [PATCH 192/224] Staging: fix more iio builds when IIO_RING_BUFFER is not enabled Fix lots more build errors in staging/iio when CONFIG_IIO_RING_BUFFER is not enabled by moving enums and defines outside of the CONFIG_IIO_RING_BUFFER ifdef block. Examples (one from each driver; there were 116 total errors): drivers/staging/iio/accel/adis16204_core.c:437: error: 'ADIS16204_SCAN_SUPPLY' undeclared here (not in a function) drivers/staging/iio/accel/adis16209_core.c:410: error: 'ADIS16209_SCAN_SUPPLY' undeclared here (not in a function) drivers/staging/iio/gyro/adis16260_core.c:420: error: 'ADIS16260_SCAN_GYRO' undeclared here (not in a function) drivers/staging/iio/imu/adis16400_core.c:565: error: 'ADIS16400_SCAN_SUPPLY' undeclared here (not in a function) Signed-off-by: Randy Dunlap Acked-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman --- drivers/staging/iio/accel/adis16204.h | 2 +- drivers/staging/iio/accel/adis16209.h | 4 ++-- drivers/staging/iio/gyro/adis16260.h | 2 +- drivers/staging/iio/imu/adis16400.h | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/staging/iio/accel/adis16204.h b/drivers/staging/iio/accel/adis16204.h index 5310a4297688..1690c0d15690 100644 --- a/drivers/staging/iio/accel/adis16204.h +++ b/drivers/staging/iio/accel/adis16204.h @@ -84,7 +84,6 @@ struct adis16204_state { int adis16204_set_irq(struct iio_dev *indio_dev, bool enable); -#ifdef CONFIG_IIO_RING_BUFFER enum adis16204_scan { ADIS16204_SCAN_SUPPLY, ADIS16204_SCAN_ACC_X, @@ -93,6 +92,7 @@ enum adis16204_scan { ADIS16204_SCAN_TEMP, }; +#ifdef CONFIG_IIO_RING_BUFFER void adis16204_remove_trigger(struct iio_dev *indio_dev); int adis16204_probe_trigger(struct iio_dev *indio_dev); diff --git a/drivers/staging/iio/accel/adis16209.h b/drivers/staging/iio/accel/adis16209.h index 58d08db6f9b5..3153cbee0957 100644 --- a/drivers/staging/iio/accel/adis16209.h +++ b/drivers/staging/iio/accel/adis16209.h @@ -121,8 +121,6 @@ struct adis16209_state { int adis16209_set_irq(struct iio_dev *indio_dev, bool enable); -#ifdef CONFIG_IIO_RING_BUFFER - #define ADIS16209_SCAN_SUPPLY 0 #define ADIS16209_SCAN_ACC_X 1 #define ADIS16209_SCAN_ACC_Y 2 @@ -132,6 +130,8 @@ int adis16209_set_irq(struct iio_dev *indio_dev, bool enable); #define ADIS16209_SCAN_INCLI_Y 6 #define ADIS16209_SCAN_ROT 7 +#ifdef CONFIG_IIO_RING_BUFFER + void adis16209_remove_trigger(struct iio_dev *indio_dev); int adis16209_probe_trigger(struct iio_dev *indio_dev); diff --git a/drivers/staging/iio/gyro/adis16260.h b/drivers/staging/iio/gyro/adis16260.h index 702dc982f62f..24bf70e4b29b 100644 --- a/drivers/staging/iio/gyro/adis16260.h +++ b/drivers/staging/iio/gyro/adis16260.h @@ -104,7 +104,6 @@ struct adis16260_state { int adis16260_set_irq(struct iio_dev *indio_dev, bool enable); -#ifdef CONFIG_IIO_RING_BUFFER /* At the moment triggers are only used for ring buffer * filling. This may change! */ @@ -115,6 +114,7 @@ int adis16260_set_irq(struct iio_dev *indio_dev, bool enable); #define ADIS16260_SCAN_TEMP 3 #define ADIS16260_SCAN_ANGL 4 +#ifdef CONFIG_IIO_RING_BUFFER void adis16260_remove_trigger(struct iio_dev *indio_dev); int adis16260_probe_trigger(struct iio_dev *indio_dev); diff --git a/drivers/staging/iio/imu/adis16400.h b/drivers/staging/iio/imu/adis16400.h index db184d11dfc0..e87715b9acc6 100644 --- a/drivers/staging/iio/imu/adis16400.h +++ b/drivers/staging/iio/imu/adis16400.h @@ -158,7 +158,6 @@ struct adis16400_state { int adis16400_set_irq(struct iio_dev *indio_dev, bool enable); -#ifdef CONFIG_IIO_RING_BUFFER /* At the moment triggers are only used for ring buffer * filling. This may change! */ @@ -182,6 +181,7 @@ int adis16400_set_irq(struct iio_dev *indio_dev, bool enable); #define ADIS16300_SCAN_INCLI_X 12 #define ADIS16300_SCAN_INCLI_Y 13 +#ifdef CONFIG_IIO_RING_BUFFER void adis16400_remove_trigger(struct iio_dev *indio_dev); int adis16400_probe_trigger(struct iio_dev *indio_dev); From 6fab3e1fbc83e6d61e8c28ceab513274183a96f6 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 23 Jun 2011 12:45:37 +0100 Subject: [PATCH 193/224] Staging: Comedi: Build only on arches providing PAGE_KERNEL_NOCACHE MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On architectures that don't define PAGE_KERNEL_NOCACHE, the Comedi driver turns into tragedy: CC [M] drivers/staging/comedi/drivers.o drivers/staging/comedi/drivers.c: In function ‘comedi_buf_alloc’: drivers/staging/comedi/drivers.c:505:41: error: ‘PAGE_KERNEL_NOCACHE’ undeclared (first use in this function) drivers/staging/comedi/drivers.c:505:41: note: each undeclared identifier is rep orted only once for each function it appears in make[3]: *** [drivers/staging/comedi/drivers.o] Error 1 Restrict the driver to only those architectures that define PAGE_KERNEL_NOCACHE. PAGE_KERNEL_NOCACHE is a kludge - some system architectures such as SGI IP27 are even uable to offer uncached operation - at least in the way an unwitting driver might assume. I haven't looked in details how the driver is using the area vmaped with PAGE_KERNEL_NOCACHE but maybe doing it XFS-style using cached memory and the flush_kernel_vmap_range / invalidate_kernel_vmap_range APIs in conjunction with the DMA API is a practical alternative. Signed-off-by: Ralf Baechle Reported-by: David Miller Reported-by: Stephen Rothwell Acked-by: Geert Uytterhoeven Cc: Martyn Welch Signed-off-by: Greg Kroah-Hartman --- drivers/staging/comedi/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/staging/comedi/Kconfig b/drivers/staging/comedi/Kconfig index 1502d80f6f78..bccdc12e843a 100644 --- a/drivers/staging/comedi/Kconfig +++ b/drivers/staging/comedi/Kconfig @@ -2,6 +2,7 @@ config COMEDI tristate "Data acquisition support (comedi)" default N depends on m + depends on BROKEN || FRV || M32R || MN10300 || SUPERH || TILE || X86 ---help--- Enable support a wide range of data acquisition devices for Linux. From fdf2df0e8ce634c9ee5f740a9303e8518ca78932 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 14 Jun 2011 15:00:18 -0700 Subject: [PATCH 194/224] Staging: fix iio builds when IIO_RING_BUFFER is not enabled Fix build by moving enum list outside of #ifdef CONFIG_IIO_RING_BUFFER. drivers/staging/iio/accel/adis16201_core.c:413: error: 'ADIS16201_SCAN_SUPPLY' undeclared here (not in a function) drivers/staging/iio/accel/adis16201_core.c:417: error: 'ADIS16201_SCAN_TEMP' undeclared here (not in a function) drivers/staging/iio/accel/adis16201_core.c:422: error: 'ADIS16201_SCAN_ACC_X' undeclared here (not in a function) drivers/staging/iio/accel/adis16201_core.c:427: error: 'ADIS16201_SCAN_ACC_Y' undeclared here (not in a function) drivers/staging/iio/accel/adis16201_core.c:432: error: 'ADIS16201_SCAN_AUX_ADC' undeclared here (not in a function) drivers/staging/iio/accel/adis16201_core.c:436: error: 'ADIS16201_SCAN_INCLI_X' undeclared here (not in a function) drivers/staging/iio/accel/adis16201_core.c:441: error: 'ADIS16201_SCAN_INCLI_Y' undeclared here (not in a function) vers/staging/iio/accel/adis16203_core.c:374: error: 'ADIS16203_SCAN_SUPPLY' undeclared here (not in a function) drivers/staging/iio/accel/adis16203_core.c:378: error: 'ADIS16203_SCAN_AUX_ADC' undeclared here (not in a function) drivers/staging/iio/accel/adis16203_core.c:382: error: 'ADIS16203_SCAN_INCLI_X' undeclared here (not in a function) drivers/staging/iio/accel/adis16203_core.c:388: error: 'ADIS16203_SCAN_INCLI_Y' undeclared here (not in a function) drivers/staging/iio/accel/adis16203_core.c:392: error: 'ADIS16203_SCAN_TEMP' undeclared here (not in a function) Signed-off-by: Randy Dunlap Acked-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman --- drivers/staging/iio/accel/adis16201.h | 2 +- drivers/staging/iio/accel/adis16203.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/staging/iio/accel/adis16201.h b/drivers/staging/iio/accel/adis16201.h index 0b9b85424dfa..4cc1a5bfab40 100644 --- a/drivers/staging/iio/accel/adis16201.h +++ b/drivers/staging/iio/accel/adis16201.h @@ -81,7 +81,6 @@ struct adis16201_state { int adis16201_set_irq(struct iio_dev *indio_dev, bool enable); -#ifdef CONFIG_IIO_RING_BUFFER enum adis16201_scan { ADIS16201_SCAN_SUPPLY, ADIS16201_SCAN_ACC_X, @@ -92,6 +91,7 @@ enum adis16201_scan { ADIS16201_SCAN_INCLI_Y, }; +#ifdef CONFIG_IIO_RING_BUFFER void adis16201_remove_trigger(struct iio_dev *indio_dev); int adis16201_probe_trigger(struct iio_dev *indio_dev); diff --git a/drivers/staging/iio/accel/adis16203.h b/drivers/staging/iio/accel/adis16203.h index 8bb8ce50c248..175e21bb9b40 100644 --- a/drivers/staging/iio/accel/adis16203.h +++ b/drivers/staging/iio/accel/adis16203.h @@ -76,7 +76,6 @@ struct adis16203_state { int adis16203_set_irq(struct iio_dev *indio_dev, bool enable); -#ifdef CONFIG_IIO_RING_BUFFER enum adis16203_scan { ADIS16203_SCAN_SUPPLY, ADIS16203_SCAN_AUX_ADC, @@ -85,6 +84,7 @@ enum adis16203_scan { ADIS16203_SCAN_INCLI_Y, }; +#ifdef CONFIG_IIO_RING_BUFFER void adis16203_remove_trigger(struct iio_dev *indio_dev); int adis16203_probe_trigger(struct iio_dev *indio_dev); From a534bb6eea72c0d082dd2faab85450e5554ba1c8 Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Mon, 13 Jun 2011 16:39:31 +0300 Subject: [PATCH 195/224] Staging: mei: fix suspend failure wait_event_interruptible_timeout return value was wrongly used. The remaining timeout was used as the error code. This fix translated wait_event_interruptible_timeout return value into error code that can be propagated. [10291.674121] pci_pm_suspend(): mei_pci_suspend+0x0/0x8b [mei] returns 2500 It's thinkpad t400 with 00:03.0 Communication controller [0780]: Intel Corporation Mobile 4 Series Chipset MEI Controller [8086:2a44] (rev 07) Reported-by: Arkadiusz Miskiewicz Signed-off-by: Tomas Winkler Signed-off-by: Greg Kroah-Hartman --- drivers/staging/mei/init.c | 2 +- drivers/staging/mei/wd.c | 13 +++++++++---- 2 files changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/staging/mei/init.c b/drivers/staging/mei/init.c index d1ffa32cd141..685fcf639644 100644 --- a/drivers/staging/mei/init.c +++ b/drivers/staging/mei/init.c @@ -189,7 +189,7 @@ int mei_hw_init(struct mei_device *dev) mutex_lock(&dev->device_lock); } - if (!err && !dev->recvd_msg) { + if (err <= 0 && !dev->recvd_msg) { dev->mei_state = MEI_DISABLED; dev_dbg(&dev->pdev->dev, "wait_event_interruptible_timeout failed" diff --git a/drivers/staging/mei/wd.c b/drivers/staging/mei/wd.c index 2564b038636a..fff53d0b5c6e 100644 --- a/drivers/staging/mei/wd.c +++ b/drivers/staging/mei/wd.c @@ -169,10 +169,15 @@ int mei_wd_stop(struct mei_device *dev, bool preserve) ret = wait_event_interruptible_timeout(dev->wait_stop_wd, dev->wd_stopped, 10 * HZ); mutex_lock(&dev->device_lock); - if (!dev->wd_stopped) - dev_dbg(&dev->pdev->dev, "stop wd failed to complete.\n"); - else - dev_dbg(&dev->pdev->dev, "stop wd complete.\n"); + if (dev->wd_stopped) { + dev_dbg(&dev->pdev->dev, "stop wd complete ret=%d.\n", ret); + ret = 0; + } else { + if (!ret) + ret = -ETIMEDOUT; + dev_warn(&dev->pdev->dev, + "stop wd failed to complete ret=%d.\n", ret); + } if (preserve) dev->wd_timeout = wd_timeout; From 928f9111662574c1ac1dfa221c4e502ec713aed9 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 31 May 2011 11:37:40 +0200 Subject: [PATCH 196/224] Staging: iio: Make IIO depend on GENERIC_HARDIRQS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On m68k (which doesn't support generic hardirqs yet): drivers/staging/iio/industrialio-trigger.c: In function ‘iio_trigger_poll’: drivers/staging/iio/industrialio-trigger.c:180: error: implicit declaration of function ‘generic_handle_irq’ drivers/staging/iio/industrialio-trigger.c: In function ‘iio_trigger_poll_chained’: drivers/staging/iio/industrialio-trigger.c:200: error: implicit declaration of function ‘handle_nested_irq’ drivers/staging/iio/industrialio-trigger.c: In function ‘iio_trig_release’: drivers/staging/iio/industrialio-trigger.c:379: error: implicit declaration of function ‘irq_modify_status’ drivers/staging/iio/industrialio-trigger.c:382: error: implicit declaration of function ‘irq_set_chip’ drivers/staging/iio/industrialio-trigger.c:384: error: implicit declaration of function ‘irq_set_handler’ drivers/staging/iio/industrialio-trigger.c:388: error: implicit declaration of function ‘irq_free_descs’ drivers/staging/iio/industrialio-trigger.c: In function ‘iio_trig_subirqmask’: drivers/staging/iio/industrialio-trigger.c:402: error: implicit declaration of function ‘irq_data_get_irq_chip’ drivers/staging/iio/industrialio-trigger.c:402: warning: initialization makes pointer from integer without a cast drivers/staging/iio/industrialio-trigger.c: In function ‘iio_trig_subirqunmask’: drivers/staging/iio/industrialio-trigger.c:411: warning: initialization makes pointer from integer without a cast drivers/staging/iio/industrialio-trigger.c: In function ‘iio_allocate_trigger’: drivers/staging/iio/industrialio-trigger.c:432: error: implicit declaration of function ‘irq_alloc_descs’ drivers/staging/iio/industrialio-trigger.c:455: error: ‘handle_simple_irq’ undeclared (first use in this function) drivers/staging/iio/industrialio-trigger.c:455: error: (Each undeclared identifier is reported only once drivers/staging/iio/industrialio-trigger.c:455: error: for each function it appears in.) Hence IIO_TRIGGER should depend on GENERIC_HARDIRQS. But as IIO_TRIGGER and IIO_RING_BUFFER form a maze of dependencies and selects, just make the whole IIO subsystem depend on GENERIC_HARDIRQS. This dependency also covers !S390, so that one can be removed again. Signed-off-by: Geert Uytterhoeven Acked-by: Jonathan Cameron Signed-off-by: Greg Kroah-Hartman --- drivers/staging/iio/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/iio/Kconfig b/drivers/staging/iio/Kconfig index f96d5b5d5141..d329635fb5c4 100644 --- a/drivers/staging/iio/Kconfig +++ b/drivers/staging/iio/Kconfig @@ -4,7 +4,7 @@ menuconfig IIO tristate "Industrial I/O support" - depends on !S390 + depends on GENERIC_HARDIRQS help The industrial I/O subsystem provides a unified framework for drivers for many different types of embedded sensors using a From 15e5201dfd4eebfce5e50b068ed40cb5f574fee6 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 9 Jun 2011 13:16:13 -0700 Subject: [PATCH 197/224] Staging: brcm80211: disable drivers for PPC platforms Right now, bad things happen if you try to build these drivers for the PPC platform as it seems that the code only has been tested and built on the MIPS big endian platform. So disable it on the PPC32 and PPC64 platforms for now, hopefully this will be resolved in the future as I'm sure someone will want to use these chips with that platform someday. Reported-by: Stephen Rothwell Cc: Henry Ptasinski Cc: Brett Rudley Cc: Roland Vossen Signed-off-by: Greg Kroah-Hartman --- drivers/staging/brcm80211/Kconfig | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/staging/brcm80211/Kconfig b/drivers/staging/brcm80211/Kconfig index f4cf9b23481e..2d1a29bdc0ee 100644 --- a/drivers/staging/brcm80211/Kconfig +++ b/drivers/staging/brcm80211/Kconfig @@ -7,6 +7,7 @@ config BRCMSMAC default n depends on PCI depends on WLAN && MAC80211 + depends on !PPC64 && !PPC32 select BRCMUTIL select FW_LOADER select CRC_CCITT @@ -20,6 +21,7 @@ config BRCMFMAC default n depends on MMC depends on WLAN && CFG80211 + depends on !PPC64 && !PPC32 select BRCMUTIL select FW_LOADER select WIRELESS_EXT From d4620396d7e9759dd3f37a8155827477cf2b5309 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 27 Jun 2011 14:34:42 -0700 Subject: [PATCH 198/224] Staging: brcm80211: disable drivers except for X86 or MIPS platforms As David points out, the driver is also broken on SPARC, so might as well just only enable it on platforms where people have reported it working, instead of trying to list all of the ones where it doesn't work, as the working platform list is much smaller... Reported-by: David Miller Reported-by: Stephen Rothwell Cc: Henry Ptasinski Cc: Brett Rudley Cc: Roland Vossen Signed-off-by: Greg Kroah-Hartman --- drivers/staging/brcm80211/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/staging/brcm80211/Kconfig b/drivers/staging/brcm80211/Kconfig index 2d1a29bdc0ee..379cf16e89f7 100644 --- a/drivers/staging/brcm80211/Kconfig +++ b/drivers/staging/brcm80211/Kconfig @@ -7,7 +7,7 @@ config BRCMSMAC default n depends on PCI depends on WLAN && MAC80211 - depends on !PPC64 && !PPC32 + depends on X86 || MIPS select BRCMUTIL select FW_LOADER select CRC_CCITT @@ -21,7 +21,7 @@ config BRCMFMAC default n depends on MMC depends on WLAN && CFG80211 - depends on !PPC64 && !PPC32 + depends on X86 || MIPS select BRCMUTIL select FW_LOADER select WIRELESS_EXT From a73738e9f85fa6ba3f1f9a6ea95b5e5fa99579a2 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 27 Jun 2011 14:43:34 -0700 Subject: [PATCH 199/224] Staging: comedi: fix build breakages on some platforms Some platforms do not have virt_to_bus(), so properly depend on CONFIG_VIRT_TO_BUS for the Comedi drivers that need this function. Reported-by: Stephen Rothwell Signed-off-by: Greg Kroah-Hartman --- drivers/staging/comedi/Kconfig | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/drivers/staging/comedi/Kconfig b/drivers/staging/comedi/Kconfig index bccdc12e843a..20008a4376e8 100644 --- a/drivers/staging/comedi/Kconfig +++ b/drivers/staging/comedi/Kconfig @@ -161,6 +161,7 @@ config COMEDI_PCL730 config COMEDI_PCL812 tristate "Advantech PCL-812/813 and ADlink ACL-8112/8113/8113/8216" + depends on VIRT_TO_BUS default N ---help--- Enable support for Advantech PCL-812/PG, PCL-813/B, ADLink @@ -172,6 +173,7 @@ config COMEDI_PCL812 config COMEDI_PCL816 tristate "Advantech PCL-814 and PCL-816 ISA card support" + depends on VIRT_TO_BUS default N ---help--- Enable support for Advantech PCL-814 and PCL-816 ISA cards @@ -181,6 +183,7 @@ config COMEDI_PCL816 config COMEDI_PCL818 tristate "Advantech PCL-718 and PCL-818 ISA card support" + depends on VIRT_TO_BUS default N ---help--- Enable support for Advantech PCL-818 ISA cards @@ -270,6 +273,7 @@ config COMEDI_DAS800 config COMEDI_DAS1800 tristate "DAS1800 and compatible ISA card support" + depends on VIRT_TO_BUS select COMEDI_FC default N ---help--- @@ -341,6 +345,7 @@ config COMEDI_DT2817 config COMEDI_DT282X tristate "Data Translation DT2821 series and DT-EZ ISA card support" select COMEDI_FC + depends on VIRT_TO_BUS default N ---help--- Enable support for Data Translation DT2821 series including DT-EZ @@ -420,6 +425,7 @@ config COMEDI_ADQ12B config COMEDI_NI_AT_A2150 tristate "NI AT-A2150 ISA card support" depends on COMEDI_NI_COMMON + depends on VIRT_TO_BUS default N ---help--- Enable support for National Instruments AT-A2150 cards @@ -537,6 +543,7 @@ if COMEDI_PCI_DRIVERS && PCI config COMEDI_ADDI_APCI_035 tristate "ADDI-DATA APCI_035 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_035 cards @@ -546,6 +553,7 @@ config COMEDI_ADDI_APCI_035 config COMEDI_ADDI_APCI_1032 tristate "ADDI-DATA APCI_1032 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_1032 cards @@ -555,6 +563,7 @@ config COMEDI_ADDI_APCI_1032 config COMEDI_ADDI_APCI_1500 tristate "ADDI-DATA APCI_1500 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_1500 cards @@ -564,6 +573,7 @@ config COMEDI_ADDI_APCI_1500 config COMEDI_ADDI_APCI_1516 tristate "ADDI-DATA APCI_1516 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_1516 cards @@ -573,6 +583,7 @@ config COMEDI_ADDI_APCI_1516 config COMEDI_ADDI_APCI_1564 tristate "ADDI-DATA APCI_1564 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_1564 cards @@ -582,6 +593,7 @@ config COMEDI_ADDI_APCI_1564 config COMEDI_ADDI_APCI_16XX tristate "ADDI-DATA APCI_16xx support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_16xx cards @@ -591,6 +603,7 @@ config COMEDI_ADDI_APCI_16XX config COMEDI_ADDI_APCI_2016 tristate "ADDI-DATA APCI_2016 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_2016 cards @@ -600,6 +613,7 @@ config COMEDI_ADDI_APCI_2016 config COMEDI_ADDI_APCI_2032 tristate "ADDI-DATA APCI_2032 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_2032 cards @@ -609,6 +623,7 @@ config COMEDI_ADDI_APCI_2032 config COMEDI_ADDI_APCI_2200 tristate "ADDI-DATA APCI_2200 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_2200 cards @@ -618,6 +633,7 @@ config COMEDI_ADDI_APCI_2200 config COMEDI_ADDI_APCI_3001 tristate "ADDI-DATA APCI_3001 support" + depends on VIRT_TO_BUS select COMEDI_FC default N ---help--- @@ -628,6 +644,7 @@ config COMEDI_ADDI_APCI_3001 config COMEDI_ADDI_APCI_3120 tristate "ADDI-DATA APCI_3520 support" + depends on VIRT_TO_BUS select COMEDI_FC default N ---help--- @@ -638,6 +655,7 @@ config COMEDI_ADDI_APCI_3120 config COMEDI_ADDI_APCI_3501 tristate "ADDI-DATA APCI_3501 support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_3501 cards @@ -647,6 +665,7 @@ config COMEDI_ADDI_APCI_3501 config COMEDI_ADDI_APCI_3XXX tristate "ADDI-DATA APCI_3xxx support" + depends on VIRT_TO_BUS default N ---help--- Enable support for ADDI-DATA APCI_3xxx cards @@ -713,6 +732,7 @@ config COMEDI_ADL_PCI9111 config COMEDI_ADL_PCI9118 tristate "ADLink PCI-9118DG, PCI-9118HG, PCI-9118HR support" select COMEDI_FC + depends on VIRT_TO_BUS default N ---help--- Enable support for ADlink PCI-9118DG, PCI-9118HG, PCI-9118HR cards @@ -1288,6 +1308,7 @@ config COMEDI_NI_LABPC depends on COMEDI_MITE select COMEDI_8255 select COMEDI_FC + depends on VIRT_TO_BUS default N ---help--- Enable support for National Instruments Lab-PC and compatibles From 4d258b25d947521c8b913154db61ec55198243f8 Mon Sep 17 00:00:00 2001 From: Vitaliy Ivanov Date: Mon, 27 Jun 2011 19:07:08 +0300 Subject: [PATCH 200/224] Fix some kernel-doc warnings Fix 'make htmldocs' warnings: Warning(/include/linux/hrtimer.h:153): No description found for parameter 'clockid' Warning(/include/linux/device.h:604): Excess struct/union/enum/typedef member 'of_match' description in 'device' Warning(/include/net/sock.h:349): Excess struct/union/enum/typedef member 'sk_rmem_alloc' description in 'sock' Signed-off-by: Vitaliy Ivanov Acked-by: Grant Likely Acked-by: David S. Miller Signed-off-by: Linus Torvalds --- include/linux/device.h | 1 - include/linux/hrtimer.h | 1 + include/net/sock.h | 1 - 3 files changed, 1 insertion(+), 2 deletions(-) diff --git a/include/linux/device.h b/include/linux/device.h index 553fd37b173b..e4f62d8896b7 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -530,7 +530,6 @@ struct device_dma_parameters { * @dma_mem: Internal for coherent mem override. * @archdata: For arch-specific additions. * @of_node: Associated device tree node. - * @of_match: Matching of_device_id from driver. * @devt: For creating the sysfs "dev". * @devres_lock: Spinlock to protect the resource of the device. * @devres_head: The resources list of the device. diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h index 51932e5acf7c..fd0dc30c9f15 100644 --- a/include/linux/hrtimer.h +++ b/include/linux/hrtimer.h @@ -135,6 +135,7 @@ struct hrtimer_sleeper { * @cpu_base: per cpu clock base * @index: clock type index for per_cpu support when moving a * timer to a base on another cpu. + * @clockid: clock id for per_cpu support * @active: red black tree root node for the active timers * @resolution: the resolution of the clock, in nanoseconds * @get_time: function to retrieve the current time of the clock diff --git a/include/net/sock.h b/include/net/sock.h index f2046e404a61..c0b938cb4b1a 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -178,7 +178,6 @@ struct sock_common { * @sk_dst_cache: destination cache * @sk_dst_lock: destination cache lock * @sk_policy: flow policy - * @sk_rmem_alloc: receive queue bytes committed * @sk_receive_queue: incoming packets * @sk_wmem_alloc: transmit queue bytes committed * @sk_write_queue: Packet sending queue From 5b8ba10198a109f8a02380648c5d29000caa9c55 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:01 -0700 Subject: [PATCH 201/224] mm: move vmtruncate_range to truncate.c You would expect to find vmtruncate_range() next to vmtruncate() in mm/truncate.c: move it there. Signed-off-by: Hugh Dickins Acked-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory.c | 24 ------------------------ mm/truncate.c | 24 ++++++++++++++++++++++++ 2 files changed, 24 insertions(+), 24 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 87d935333f0d..40b7531ee8ba 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2798,30 +2798,6 @@ void unmap_mapping_range(struct address_space *mapping, } EXPORT_SYMBOL(unmap_mapping_range); -int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end) -{ - struct address_space *mapping = inode->i_mapping; - - /* - * If the underlying filesystem is not going to provide - * a way to truncate a range of blocks (punch a hole) - - * we should return failure right now. - */ - if (!inode->i_op->truncate_range) - return -ENOSYS; - - mutex_lock(&inode->i_mutex); - down_write(&inode->i_alloc_sem); - unmap_mapping_range(mapping, offset, (end - offset), 1); - truncate_inode_pages_range(mapping, offset, end); - unmap_mapping_range(mapping, offset, (end - offset), 1); - inode->i_op->truncate_range(inode, offset, end); - up_write(&inode->i_alloc_sem); - mutex_unlock(&inode->i_mutex); - - return 0; -} - /* * We enter with non-exclusive mmap_sem (to exclude vma changes, * but allow concurrent faults), and pte mapped but not yet locked. diff --git a/mm/truncate.c b/mm/truncate.c index 3a29a6180212..5b4c3a4847e9 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -603,3 +603,27 @@ int vmtruncate(struct inode *inode, loff_t offset) return 0; } EXPORT_SYMBOL(vmtruncate); + +int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end) +{ + struct address_space *mapping = inode->i_mapping; + + /* + * If the underlying filesystem is not going to provide + * a way to truncate a range of blocks (punch a hole) - + * we should return failure right now. + */ + if (!inode->i_op->truncate_range) + return -ENOSYS; + + mutex_lock(&inode->i_mutex); + down_write(&inode->i_alloc_sem); + unmap_mapping_range(mapping, offset, (end - offset), 1); + truncate_inode_pages_range(mapping, offset, end); + unmap_mapping_range(mapping, offset, (end - offset), 1); + inode->i_op->truncate_range(inode, offset, end); + up_write(&inode->i_alloc_sem); + mutex_unlock(&inode->i_mutex); + + return 0; +} From 072441e21ddcd1140606b7d4ef6eab579a86b0b3 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:02 -0700 Subject: [PATCH 202/224] mm: move shmem prototypes to shmem_fs.h Before adding any more global entry points into shmem.c, gather such prototypes into shmem_fs.h. Remove mm's own declarations from swap.h, but for now leave the ones in mm.h: because shmem_file_setup() and shmem_zero_setup() are called from various places, and we should not force other subsystems to update immediately. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 17 +++++++++++++++++ include/linux/swap.h | 10 ---------- mm/memcontrol.c | 1 + mm/swapfile.c | 2 +- 4 files changed, 19 insertions(+), 11 deletions(-) diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 2b7fec840517..cae65dc42bcc 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -5,6 +5,13 @@ #include #include +struct page; +struct file; +struct inode; +struct super_block; +struct user_struct; +struct vm_area_struct; + /* inode in-kernel data */ #define SHMEM_NR_DIRECT 16 @@ -45,7 +52,17 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode) return container_of(inode, struct shmem_inode_info, vfs_inode); } +/* + * Functions in mm/shmem.c called directly from elsewhere: + */ extern int init_tmpfs(void); extern int shmem_fill_super(struct super_block *sb, void *data, int silent); +extern struct file *shmem_file_setup(const char *name, + loff_t size, unsigned long flags); +extern int shmem_zero_setup(struct vm_area_struct *); +extern int shmem_lock(struct file *file, int lock, struct user_struct *user); +extern int shmem_unuse(swp_entry_t entry, struct page *page); +extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, + struct page **pagep, swp_entry_t *ent); #endif diff --git a/include/linux/swap.h b/include/linux/swap.h index e70564647039..a273468f8285 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -300,16 +300,6 @@ static inline void scan_unevictable_unregister_node(struct node *node) extern int kswapd_run(int nid); extern void kswapd_stop(int nid); -#ifdef CONFIG_MMU -/* linux/mm/shmem.c */ -extern int shmem_unuse(swp_entry_t entry, struct page *page); -#endif /* CONFIG_MMU */ - -#ifdef CONFIG_CGROUP_MEM_RES_CTLR -extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, - struct page **pagep, swp_entry_t *ent); -#endif - #ifdef CONFIG_SWAP /* linux/mm/page_io.c */ extern int swap_readpage(struct page *); diff --git a/mm/memcontrol.c b/mm/memcontrol.c index cf7d027a8844..ddffc74cdebe 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include diff --git a/mm/swapfile.c b/mm/swapfile.c index d537d29e9b7b..ff8dc1a18cb4 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include #include #include From 94c1e62df4494b79782cb9c7279f827212d1de70 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:03 -0700 Subject: [PATCH 203/224] tmpfs: take control of its truncate_range 2.6.35's new truncate convention gave tmpfs the opportunity to control its file truncation, no longer enforced from outside by vmtruncate(). We shall want to build upon that, to handle pagecache and swap together. Slightly redefine the ->truncate_range interface: let it now be called between the unmap_mapping_range()s, with the filesystem responsible for doing the truncate_inode_pages_range() from it - just as the filesystem is nowadays responsible for doing that from its ->setattr. Let's rename shmem_notify_change() to shmem_setattr(). Instead of calling the generic truncate_setsize(), bring that code in so we can call shmem_truncate_range() - which will later be updated to perform its own variant of truncate_inode_pages_range(). Remove the punch_hole unmap_mapping_range() from shmem_truncate_range(): now that the COW's unmap_mapping_range() comes after ->truncate_range, there is no need to call it a third time. Export shmem_truncate_range() and add it to the list in shmem_fs.h, so that i915_gem_object_truncate() can call it explicitly in future; get this patch in first, then update drm/i915 once this is available (until then, i915 will just be doing the truncate_inode_pages() twice). Though introduced five years ago, no other filesystem is implementing ->truncate_range, and its only other user is madvise(,,MADV_REMOVE): we expect to convert it to fallocate(,FALLOC_FL_PUNCH_HOLE,,) shortly, whereupon ->truncate_range can be removed from inode_operations - shmem_truncate_range() will help i915 across that transition too. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 1 + mm/shmem.c | 51 +++++++++++++++++++++++----------------- mm/truncate.c | 4 ++-- 3 files changed, 32 insertions(+), 24 deletions(-) diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index cae65dc42bcc..22a20af4d785 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -61,6 +61,7 @@ extern struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); extern int shmem_zero_setup(struct vm_area_struct *); extern int shmem_lock(struct file *file, int lock, struct user_struct *user); +extern void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end); extern int shmem_unuse(swp_entry_t entry, struct page *page); extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent); diff --git a/mm/shmem.c b/mm/shmem.c index d221a1cfd7b1..f1714758ea96 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -539,7 +539,7 @@ static void shmem_free_pages(struct list_head *next) } while (next); } -static void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) +void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) { struct shmem_inode_info *info = SHMEM_I(inode); unsigned long idx; @@ -562,6 +562,8 @@ static void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) spinlock_t *punch_lock; unsigned long upper_limit; + truncate_inode_pages_range(inode->i_mapping, start, end); + inode->i_ctime = inode->i_mtime = CURRENT_TIME; idx = (start + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; if (idx >= info->next_index) @@ -738,16 +740,8 @@ done2: * lowered next_index. Also, though shmem_getpage checks * i_size before adding to cache, no recheck after: so fix the * narrow window there too. - * - * Recalling truncate_inode_pages_range and unmap_mapping_range - * every time for punch_hole (which never got a chance to clear - * SHMEM_PAGEIN at the start of vmtruncate_range) is expensive, - * yet hardly ever necessary: try to optimize them out later. */ truncate_inode_pages_range(inode->i_mapping, start, end); - if (punch_hole) - unmap_mapping_range(inode->i_mapping, start, - end - start, 1); } spin_lock(&info->lock); @@ -766,22 +760,23 @@ done2: shmem_free_pages(pages_to_free.next); } } +EXPORT_SYMBOL_GPL(shmem_truncate_range); -static int shmem_notify_change(struct dentry *dentry, struct iattr *attr) +static int shmem_setattr(struct dentry *dentry, struct iattr *attr) { struct inode *inode = dentry->d_inode; - loff_t newsize = attr->ia_size; int error; error = inode_change_ok(inode, attr); if (error) return error; - if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE) - && newsize != inode->i_size) { + if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) { + loff_t oldsize = inode->i_size; + loff_t newsize = attr->ia_size; struct page *page = NULL; - if (newsize < inode->i_size) { + if (newsize < oldsize) { /* * If truncating down to a partial page, then * if that page is already allocated, hold it @@ -810,12 +805,19 @@ static int shmem_notify_change(struct dentry *dentry, struct iattr *attr) spin_unlock(&info->lock); } } - - /* XXX(truncate): truncate_setsize should be called last */ - truncate_setsize(inode, newsize); + if (newsize != oldsize) { + i_size_write(inode, newsize); + inode->i_ctime = inode->i_mtime = CURRENT_TIME; + } + if (newsize < oldsize) { + loff_t holebegin = round_up(newsize, PAGE_SIZE); + unmap_mapping_range(inode->i_mapping, holebegin, 0, 1); + shmem_truncate_range(inode, newsize, (loff_t)-1); + /* unmap again to remove racily COWed private pages */ + unmap_mapping_range(inode->i_mapping, holebegin, 0, 1); + } if (page) page_cache_release(page); - shmem_truncate_range(inode, newsize, (loff_t)-1); } setattr_copy(inode, attr); @@ -832,7 +834,6 @@ static void shmem_evict_inode(struct inode *inode) struct shmem_xattr *xattr, *nxattr; if (inode->i_mapping->a_ops == &shmem_aops) { - truncate_inode_pages(inode->i_mapping, 0); shmem_unacct_size(info->flags, inode->i_size); inode->i_size = 0; shmem_truncate_range(inode, 0, (loff_t)-1); @@ -2706,7 +2707,7 @@ static const struct file_operations shmem_file_operations = { }; static const struct inode_operations shmem_inode_operations = { - .setattr = shmem_notify_change, + .setattr = shmem_setattr, .truncate_range = shmem_truncate_range, #ifdef CONFIG_TMPFS_XATTR .setxattr = shmem_setxattr, @@ -2739,7 +2740,7 @@ static const struct inode_operations shmem_dir_inode_operations = { .removexattr = shmem_removexattr, #endif #ifdef CONFIG_TMPFS_POSIX_ACL - .setattr = shmem_notify_change, + .setattr = shmem_setattr, .check_acl = generic_check_acl, #endif }; @@ -2752,7 +2753,7 @@ static const struct inode_operations shmem_special_inode_operations = { .removexattr = shmem_removexattr, #endif #ifdef CONFIG_TMPFS_POSIX_ACL - .setattr = shmem_notify_change, + .setattr = shmem_setattr, .check_acl = generic_check_acl, #endif }; @@ -2908,6 +2909,12 @@ int shmem_lock(struct file *file, int lock, struct user_struct *user) return 0; } +void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end) +{ + truncate_inode_pages_range(inode->i_mapping, start, end); +} +EXPORT_SYMBOL_GPL(shmem_truncate_range); + #ifdef CONFIG_CGROUP_MEM_RES_CTLR /** * mem_cgroup_get_shmem_target - find a page or entry assigned to the shmem file diff --git a/mm/truncate.c b/mm/truncate.c index 5b4c3a4847e9..29a9b8a5a31a 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -619,9 +619,9 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end) mutex_lock(&inode->i_mutex); down_write(&inode->i_alloc_sem); unmap_mapping_range(mapping, offset, (end - offset), 1); - truncate_inode_pages_range(mapping, offset, end); - unmap_mapping_range(mapping, offset, (end - offset), 1); inode->i_op->truncate_range(inode, offset, end); + /* unmap again to remove racily COWed private pages */ + unmap_mapping_range(mapping, offset, (end - offset), 1); up_write(&inode->i_alloc_sem); mutex_unlock(&inode->i_mutex); From d9d90e5eb70e09903dadff42099b6c948f814050 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:04 -0700 Subject: [PATCH 204/224] tmpfs: add shmem_read_mapping_page_gfp Although it is used (by i915) on nothing but tmpfs, read_cache_page_gfp() is unsuited to tmpfs, because it inserts a page into pagecache before calling the filesystem's ->readpage: tmpfs may have pages in swapcache which only it knows how to locate and switch to filecache. At present tmpfs provides a ->readpage method, and copes with this by copying pages; but soon we can simplify it by removing its ->readpage. Provide shmem_read_mapping_page_gfp() now, ready for that transition, Export shmem_read_mapping_page_gfp() and add it to list in shmem_fs.h, with shmem_read_mapping_page() inline for the common mapping_gfp case. (shmem_read_mapping_page_gfp or shmem_read_cache_page_gfp? Generally the read_mapping_page functions use the mapping's ->readpage, and the read_cache_page functions use the supplied filler, so I think read_cache_page_gfp was slightly misnamed.) Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/shmem_fs.h | 17 ++++++++++------- mm/shmem.c | 23 +++++++++++++++++++++++ 2 files changed, 33 insertions(+), 7 deletions(-) diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index 22a20af4d785..aa08fa8fd79b 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -3,15 +3,9 @@ #include #include +#include #include -struct page; -struct file; -struct inode; -struct super_block; -struct user_struct; -struct vm_area_struct; - /* inode in-kernel data */ #define SHMEM_NR_DIRECT 16 @@ -61,9 +55,18 @@ extern struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags); extern int shmem_zero_setup(struct vm_area_struct *); extern int shmem_lock(struct file *file, int lock, struct user_struct *user); +extern struct page *shmem_read_mapping_page_gfp(struct address_space *mapping, + pgoff_t index, gfp_t gfp_mask); extern void shmem_truncate_range(struct inode *inode, loff_t start, loff_t end); extern int shmem_unuse(swp_entry_t entry, struct page *page); extern void mem_cgroup_get_shmem_target(struct inode *inode, pgoff_t pgoff, struct page **pagep, swp_entry_t *ent); +static inline struct page *shmem_read_mapping_page( + struct address_space *mapping, pgoff_t index) +{ + return shmem_read_mapping_page_gfp(mapping, index, + mapping_gfp_mask(mapping)); +} + #endif diff --git a/mm/shmem.c b/mm/shmem.c index f1714758ea96..fcedf5464eb7 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3035,3 +3035,26 @@ int shmem_zero_setup(struct vm_area_struct *vma) vma->vm_flags |= VM_CAN_NONLINEAR; return 0; } + +/** + * shmem_read_mapping_page_gfp - read into page cache, using specified page allocation flags. + * @mapping: the page's address_space + * @index: the page index + * @gfp: the page allocator flags to use if allocating + * + * This behaves as a tmpfs "read_cache_page_gfp(mapping, index, gfp)", + * with any new page allocations done using the specified allocation flags. + * But read_cache_page_gfp() uses the ->readpage() method: which does not + * suit tmpfs, since it may have pages in swapcache, and needs to find those + * for itself; although drivers/gpu/drm i915 and ttm rely upon this support. + * + * Provide a stub for those callers to start using now, then later + * flesh it out to call shmem_getpage() with additional gfp mask, when + * shmem_file_splice_read() is added and shmem_readpage() is removed. + */ +struct page *shmem_read_mapping_page_gfp(struct address_space *mapping, + pgoff_t index, gfp_t gfp) +{ + return read_cache_page_gfp(mapping, index, gfp); +} +EXPORT_SYMBOL_GPL(shmem_read_mapping_page_gfp); From 31c1771cdbf0edea96e109e24dae387d331b6d96 Mon Sep 17 00:00:00 2001 From: Priyanka Jain Date: Mon, 27 Jun 2011 16:18:04 -0700 Subject: [PATCH 205/224] drivers/rtc/rtc-ds1307.c: add support for RTC device pt7c4338 PT7C4338 chip is being manufactured by Pericom Technology Inc. It is a serial real-time clock which provides: 1) Low-power clock/calendar. 2) Programmable square-wave output. It has 56 bytes of nonvolatile RAM. Its register set is same as that of rtc device: DS1307. Signed-off-by: Priyanka Jain Acked-by: Timur Tabi Reviewed-by: Wolfram Sang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-ds1307.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/rtc/rtc-ds1307.c b/drivers/rtc/rtc-ds1307.c index 4724ba3acf1a..b2005b44e4f7 100644 --- a/drivers/rtc/rtc-ds1307.c +++ b/drivers/rtc/rtc-ds1307.c @@ -149,6 +149,7 @@ static const struct i2c_device_id ds1307_id[] = { { "ds1340", ds_1340 }, { "ds3231", ds_3231 }, { "m41t00", m41t00 }, + { "pt7c4338", ds_1307 }, { "rx8025", rx_8025 }, { } }; From 8c95aa60d2ae3c4e63705ee0f18e155898b09662 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Mon, 27 Jun 2011 16:18:05 -0700 Subject: [PATCH 206/224] um: add asm/percpu.h To make SLUB work on UML we need this_cpu_cmpxchg from asm-generic/percpu.h. Signed-off-by: Richard Weinberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/asm/percpu.h | 6 ++++++ 1 file changed, 6 insertions(+) create mode 100644 arch/um/include/asm/percpu.h diff --git a/arch/um/include/asm/percpu.h b/arch/um/include/asm/percpu.h new file mode 100644 index 000000000000..efe7508d8abd --- /dev/null +++ b/arch/um/include/asm/percpu.h @@ -0,0 +1,6 @@ +#ifndef __UM_PERCPU_H +#define __UM_PERCPU_H + +#include + +#endif /* __UM_PERCPU_H */ From 2b4b2482e70eba10dd98653a3a5ac68126565e24 Mon Sep 17 00:00:00 2001 From: Bob Liu Date: Mon, 27 Jun 2011 16:18:06 -0700 Subject: [PATCH 207/224] romfs: fix romfs_get_unmapped_area() argument check romfs_get_unmapped_area() checks argument `len' without considering PAGE_ALIGN which will cause do_mmap_pgoff() return -EINVAL error after commit f67d9b1576c ("nommu: add page_align to mmap"). Fix the check by changing it in same way ramfs_nommu_get_unmapped_area() was changed in ramfs/file-nommu.c. Signed-off-by: Bob Liu Cc: David Howells Cc: Paul Mundt Acked-by: Greg Ungerer Cc: Geert Uytterhoeven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/romfs/mmap-nommu.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/romfs/mmap-nommu.c b/fs/romfs/mmap-nommu.c index f0511e816967..eed99428f104 100644 --- a/fs/romfs/mmap-nommu.c +++ b/fs/romfs/mmap-nommu.c @@ -27,14 +27,18 @@ static unsigned long romfs_get_unmapped_area(struct file *file, { struct inode *inode = file->f_mapping->host; struct mtd_info *mtd = inode->i_sb->s_mtd; - unsigned long isize, offset; + unsigned long isize, offset, maxpages, lpages; if (!mtd) goto cant_map_directly; + /* the mapping mustn't extend beyond the EOF */ + lpages = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; isize = i_size_read(inode); offset = pgoff << PAGE_SHIFT; - if (offset > isize || len > isize || offset > isize - len) + + maxpages = (isize + PAGE_SIZE - 1) >> PAGE_SHIFT; + if ((pgoff >= maxpages) || (maxpages - pgoff < lpages)) return (unsigned long) -EINVAL; /* we need to call down to the MTD layer to do the actual mapping */ From 507c5f1224014f9956e604ee8703b3bbea7da4a4 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Mon, 27 Jun 2011 16:18:07 -0700 Subject: [PATCH 208/224] include/linux/compat.h: declare compat_sys_sendmmsg() This is required for tilegx to be able to use the compat unistd.h header where compat_sys_sendmmsg() is now mentioned. Signed-off-by: Chris Metcalf Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/compat.h b/include/linux/compat.h index ddcb7db38e67..846bb1792572 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -467,6 +467,8 @@ asmlinkage long compat_sys_setsockopt(int fd, int level, int optname, char __user *optval, unsigned int optlen); asmlinkage long compat_sys_sendmsg(int fd, struct compat_msghdr __user *msg, unsigned flags); +asmlinkage long compat_sys_sendmmsg(int fd, struct compat_mmsghdr __user *mmsg, + unsigned vlen, unsigned int flags); asmlinkage long compat_sys_recvmsg(int fd, struct compat_msghdr __user *msg, unsigned int flags); asmlinkage long compat_sys_recv(int fd, void __user *buf, size_t len, From aa2c96d6f329e66cc59352b0f12e8f04e6a9593b Mon Sep 17 00:00:00 2001 From: Josh Hunt Date: Mon, 27 Jun 2011 16:18:08 -0700 Subject: [PATCH 209/224] drivers/misc/lkdtm.c: fix race when crashpoint is hit multiple times before checking count We observed the crash point count going negative in cases where the crash point is hit multiple times before the check of "count == 0" is done. Because of this we never call lkdtm_do_action(). This patch just adds a spinlock to protect count. Reported-by: Tapan Dhimant Signed-off-by: Josh Hunt Acked-by: Ankita Garg Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/lkdtm.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/misc/lkdtm.c b/drivers/misc/lkdtm.c index 81d7fa4ec0db..150cd7061b80 100644 --- a/drivers/misc/lkdtm.c +++ b/drivers/misc/lkdtm.c @@ -120,6 +120,7 @@ static int recur_count = REC_NUM_DEFAULT; static enum cname cpoint = CN_INVALID; static enum ctype cptype = CT_NONE; static int count = DEFAULT_COUNT; +static DEFINE_SPINLOCK(count_lock); module_param(recur_count, int, 0644); MODULE_PARM_DESC(recur_count, " Recursion level for the stack overflow test, "\ @@ -230,11 +231,14 @@ static const char *cp_name_to_str(enum cname name) static int lkdtm_parse_commandline(void) { int i; + unsigned long flags; if (cpoint_count < 1 || recur_count < 1) return -EINVAL; + spin_lock_irqsave(&count_lock, flags); count = cpoint_count; + spin_unlock_irqrestore(&count_lock, flags); /* No special parameters */ if (!cpoint_type && !cpoint_name) @@ -349,6 +353,9 @@ static void lkdtm_do_action(enum ctype which) static void lkdtm_handler(void) { + unsigned long flags; + + spin_lock_irqsave(&count_lock, flags); count--; printk(KERN_INFO "lkdtm: Crash point %s of type %s hit, trigger in %d rounds\n", cp_name_to_str(cpoint), cp_type_to_str(cptype), count); @@ -357,6 +364,7 @@ static void lkdtm_handler(void) lkdtm_do_action(cptype); count = cpoint_count; } + spin_unlock_irqrestore(&count_lock, flags); } static int lkdtm_register_cpoint(enum cname which) From 9b679320a5fbf46454011e5c62e0b8991b0956d1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 27 Jun 2011 16:18:09 -0700 Subject: [PATCH 210/224] mm/memory-failure.c: fix spinlock vs mutex order We cannot take a mutex while holding a spinlock, so flip the order and fix the locking documentation. Signed-off-by: Peter Zijlstra Acked-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory-failure.c | 21 ++++++--------------- mm/rmap.c | 5 ++--- 2 files changed, 8 insertions(+), 18 deletions(-) diff --git a/mm/memory-failure.c b/mm/memory-failure.c index eac0ba561491..740c4f52059c 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -391,10 +391,11 @@ static void collect_procs_anon(struct page *page, struct list_head *to_kill, struct task_struct *tsk; struct anon_vma *av; - read_lock(&tasklist_lock); av = page_lock_anon_vma(page); if (av == NULL) /* Not actually mapped anymore */ - goto out; + return; + + read_lock(&tasklist_lock); for_each_process (tsk) { struct anon_vma_chain *vmac; @@ -408,9 +409,8 @@ static void collect_procs_anon(struct page *page, struct list_head *to_kill, add_to_kill(tsk, page, vma, to_kill, tkc); } } - page_unlock_anon_vma(av); -out: read_unlock(&tasklist_lock); + page_unlock_anon_vma(av); } /* @@ -424,17 +424,8 @@ static void collect_procs_file(struct page *page, struct list_head *to_kill, struct prio_tree_iter iter; struct address_space *mapping = page->mapping; - /* - * A note on the locking order between the two locks. - * We don't rely on this particular order. - * If you have some other code that needs a different order - * feel free to switch them around. Or add a reverse link - * from mm_struct to task_struct, then this could be all - * done without taking tasklist_lock and looping over all tasks. - */ - - read_lock(&tasklist_lock); mutex_lock(&mapping->i_mmap_mutex); + read_lock(&tasklist_lock); for_each_process(tsk) { pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT); @@ -454,8 +445,8 @@ static void collect_procs_file(struct page *page, struct list_head *to_kill, add_to_kill(tsk, page, vma, to_kill, tkc); } } - mutex_unlock(&mapping->i_mmap_mutex); read_unlock(&tasklist_lock); + mutex_unlock(&mapping->i_mmap_mutex); } /* diff --git a/mm/rmap.c b/mm/rmap.c index 27dfd3b82b0f..23295f65ae43 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -38,9 +38,8 @@ * in arch-dependent flush_dcache_mmap_lock, * within inode_wb_list_lock in __sync_single_inode) * - * (code doesn't rely on that order so it could be switched around) - * ->tasklist_lock - * anon_vma->mutex (memory_failure, collect_procs_anon) + * anon_vma->mutex,mapping->i_mutex (memory_failure, collect_procs_anon) + * ->tasklist_lock * pte map lock */ From 08142579b6ca35883c1ed066a2681de6f6917062 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Mon, 27 Jun 2011 16:18:10 -0700 Subject: [PATCH 211/224] mm: fix assertion mapping->nrpages == 0 in end_writeback() Under heavy memory and filesystem load, users observe the assertion mapping->nrpages == 0 in end_writeback() trigger. This can be caused by page reclaim reclaiming the last page from a mapping in the following race: CPU0 CPU1 ... shrink_page_list() __remove_mapping() __delete_from_page_cache() radix_tree_delete() evict_inode() truncate_inode_pages() truncate_inode_pages_range() pagevec_lookup() - finds nothing end_writeback() mapping->nrpages != 0 -> BUG page->mapping = NULL mapping->nrpages-- Fix the problem by doing a reliable check of mapping->nrpages under mapping->tree_lock in end_writeback(). Analyzed by Jay , lost in LKML, and dug out by Miklos Szeredi . Cc: Jay Cc: Miklos Szeredi Signed-off-by: Jan Kara Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inode.c | 7 +++++++ include/linux/fs.h | 1 + mm/truncate.c | 5 +++++ 3 files changed, 13 insertions(+) diff --git a/fs/inode.c b/fs/inode.c index 0f7e88a7803f..43566d17d1b8 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -423,7 +423,14 @@ EXPORT_SYMBOL(remove_inode_hash); void end_writeback(struct inode *inode) { might_sleep(); + /* + * We have to cycle tree_lock here because reclaim can be still in the + * process of removing the last page (in __delete_from_page_cache()) + * and we must not free mapping under it. + */ + spin_lock_irq(&inode->i_data.tree_lock); BUG_ON(inode->i_data.nrpages); + spin_unlock_irq(&inode->i_data.tree_lock); BUG_ON(!list_empty(&inode->i_data.private_list)); BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); diff --git a/include/linux/fs.h b/include/linux/fs.h index 6e73e2e9ae33..b5b979247863 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -639,6 +639,7 @@ struct address_space { struct prio_tree_root i_mmap; /* tree of private and shared mappings */ struct list_head i_mmap_nonlinear;/*list VM_NONLINEAR mappings */ struct mutex i_mmap_mutex; /* protect tree, count, list */ + /* Protected by tree_lock together with the radix tree */ unsigned long nrpages; /* number of total pages */ pgoff_t writeback_index;/* writeback starts here */ const struct address_space_operations *a_ops; /* methods */ diff --git a/mm/truncate.c b/mm/truncate.c index 29a9b8a5a31a..e13f22efaad7 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -304,6 +304,11 @@ EXPORT_SYMBOL(truncate_inode_pages_range); * @lstart: offset from which to truncate * * Called under (and serialised by) inode->i_mutex. + * + * Note: When this function returns, there can be a page in the process of + * deletion (inside __delete_from_page_cache()) in the specified range. Thus + * mapping->nrpages can be non-zero when this function returns even after + * truncation of the whole mapping. */ void truncate_inode_pages(struct address_space *mapping, loff_t lstart) { From 26c4caea9d697043cc5a458b96411b86d7f6babd Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Mon, 27 Jun 2011 16:18:11 -0700 Subject: [PATCH 212/224] taskstats: don't allow duplicate entries in listener mode Currently a single process may register exit handlers unlimited times. It may lead to a bloated listeners chain and very slow process terminations. Eg after 10KK sent TASKSTATS_CMD_ATTR_REGISTER_CPUMASKs ~300 Mb of kernel memory is stolen for the handlers chain and "time id" shows 2-7 seconds instead of normal 0.003. It makes it possible to exhaust all kernel memory and to eat much of CPU time by triggerring numerous exits on a single CPU. The patch limits the number of times a single process may register itself on a single CPU to one. One little issue is kept unfixed - as taskstats_exit() is called before exit_files() in do_exit(), the orphaned listener entry (if it was not explicitly deregistered) is kept until the next someone's exit() and implicit deregistration in send_cpu_listeners(). So, if a process registered itself as a listener exits and the next spawned process gets the same pid, it would inherit taskstats attributes. Signed-off-by: Vasiliy Kulikov Cc: Balbir Singh Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 9ffea360a778..fc0f22005417 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -285,16 +285,18 @@ ret: static int add_del_listener(pid_t pid, const struct cpumask *mask, int isadd) { struct listener_list *listeners; - struct listener *s, *tmp; + struct listener *s, *tmp, *s2; unsigned int cpu; if (!cpumask_subset(mask, cpu_possible_mask)) return -EINVAL; + s = NULL; if (isadd == REGISTER) { for_each_cpu(cpu, mask) { - s = kmalloc_node(sizeof(struct listener), GFP_KERNEL, - cpu_to_node(cpu)); + if (!s) + s = kmalloc_node(sizeof(struct listener), + GFP_KERNEL, cpu_to_node(cpu)); if (!s) goto cleanup; s->pid = pid; @@ -303,9 +305,16 @@ static int add_del_listener(pid_t pid, const struct cpumask *mask, int isadd) listeners = &per_cpu(listener_array, cpu); down_write(&listeners->sem); + list_for_each_entry_safe(s2, tmp, &listeners->list, list) { + if (s2->pid == pid) + goto next_cpu; + } list_add(&s->list, &listeners->list); + s = NULL; +next_cpu: up_write(&listeners->sem); } + kfree(s); return 0; } From ac34a1a3c39da0a1b9188d12a9ce85506364ed2a Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Mon, 27 Jun 2011 16:18:12 -0700 Subject: [PATCH 213/224] memcg: fix direct softlimit reclaim to be called in limit path Commit d149e3b25d7c ("memcg: add the soft_limit reclaim in global direct reclaim") adds a softlimit hook to shrink_zones(). By this, soft limit is called as try_to_free_pages() do_try_to_free_pages() shrink_zones() mem_cgroup_soft_limit_reclaim() Then, direct reclaim is memcg softlimit hint aware, now. But, the memory cgroup's "limit" path can call softlimit shrinker. try_to_free_mem_cgroup_pages() do_try_to_free_pages() shrink_zones() mem_cgroup_soft_limit_reclaim() This will cause a global reclaim when a memcg hits limit. This is bug. soft_limit_reclaim() should be called when scanning_global_lru(sc) == true. And the commit adds a variable "total_scanned" for counting softlimit scanned pages....it's not "total". This patch removes the variable and update sc->nr_scanned instead of it. This will affect shrink_slab()'s scan condition but, global LRU is scanned by softlimit and I think this change makes sense. TODO: avoid too much scanning of a zone when softlimit did enough work. Signed-off-by: KAMEZAWA Hiroyuki Cc: Daisuke Nishimura Cc: Ying Han Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index 8ff834e19c24..4f49535d4cd3 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1995,14 +1995,13 @@ restart: * If a zone is deemed to be full of pinned pages then just give it a light * scan then give up on it. */ -static unsigned long shrink_zones(int priority, struct zonelist *zonelist, +static void shrink_zones(int priority, struct zonelist *zonelist, struct scan_control *sc) { struct zoneref *z; struct zone *zone; unsigned long nr_soft_reclaimed; unsigned long nr_soft_scanned; - unsigned long total_scanned = 0; for_each_zone_zonelist_nodemask(zone, z, zonelist, gfp_zone(sc->gfp_mask), sc->nodemask) { @@ -2017,19 +2016,23 @@ static unsigned long shrink_zones(int priority, struct zonelist *zonelist, continue; if (zone->all_unreclaimable && priority != DEF_PRIORITY) continue; /* Let kswapd poll it */ + /* + * This steals pages from memory cgroups over softlimit + * and returns the number of reclaimed pages and + * scanned pages. This works for global memory pressure + * and balancing, not for a memcg's limit. + */ + nr_soft_scanned = 0; + nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone, + sc->order, sc->gfp_mask, + &nr_soft_scanned); + sc->nr_reclaimed += nr_soft_reclaimed; + sc->nr_scanned += nr_soft_scanned; + /* need some check for avoid more shrink_zone() */ } - nr_soft_scanned = 0; - nr_soft_reclaimed = mem_cgroup_soft_limit_reclaim(zone, - sc->order, sc->gfp_mask, - &nr_soft_scanned); - sc->nr_reclaimed += nr_soft_reclaimed; - total_scanned += nr_soft_scanned; - shrink_zone(priority, zone, sc); } - - return total_scanned; } static bool zone_reclaimable(struct zone *zone) @@ -2094,7 +2097,7 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist, sc->nr_scanned = 0; if (!priority) disable_swap_token(sc->mem_cgroup); - total_scanned += shrink_zones(priority, zonelist, sc); + shrink_zones(priority, zonelist, sc); /* * Don't shrink slabs when reclaiming memory from * over limit cgroups From 5286bd953645408634daa880d04c73dd18d0224a Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 27 Jun 2011 16:18:13 -0700 Subject: [PATCH 214/224] drivers/leds/leds-lp5521.c: fix section mismatches Fix this section mismatch: WARNING: drivers/leds/leds-lp5521.o(.text+0xf2c): Section mismatch in reference from the function lp5521_probe() to the function .init.text:lp5521_init_led() The function lp5521_probe() references the function __init lp5521_init_led(). This is often because lp5521_probe lacks a __init annotation or the annotation of lp5521_init_led is wrong. Fixing this mismatch triggers one more mismatch, fix that one as well. Signed-off-by: Ralf Baechle Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/leds/leds-lp5521.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/leds/leds-lp5521.c b/drivers/leds/leds-lp5521.c index c0cff64a1ae6..cc1dc4817fac 100644 --- a/drivers/leds/leds-lp5521.c +++ b/drivers/leds/leds-lp5521.c @@ -593,7 +593,7 @@ static void lp5521_unregister_sysfs(struct i2c_client *client) &lp5521_led_attribute_group); } -static int __init lp5521_init_led(struct lp5521_led *led, +static int __devinit lp5521_init_led(struct lp5521_led *led, struct i2c_client *client, int chan, struct lp5521_platform_data *pdata) { @@ -637,7 +637,7 @@ static int __init lp5521_init_led(struct lp5521_led *led, return 0; } -static int lp5521_probe(struct i2c_client *client, +static int __devinit lp5521_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct lp5521_chip *chip; From 33721bd3d00e7a235f70ba4ec19eb64bcd060c0b Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 27 Jun 2011 16:18:14 -0700 Subject: [PATCH 215/224] drivers/leds/leds-lp5523.c: fix section mismatches Fix this section mismatch: WARNING: drivers/leds/leds-lp5523.o(.text+0x12f4): Section mismatch in reference from the function lp5523_probe() to the function .init.text:lp5523_init_led() The function lp5523_probe() references the function __init lp5523_init_led(). This is often because lp5523_probe lacks a __init annotation or the annotation of lp5523_init_led is wrong. Fixing this one triggers one more mismatch, fix that one as well. Signed-off-by: Ralf Baechle Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/leds/leds-lp5523.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/leds/leds-lp5523.c b/drivers/leds/leds-lp5523.c index e19fed25f137..5971e309b234 100644 --- a/drivers/leds/leds-lp5523.c +++ b/drivers/leds/leds-lp5523.c @@ -826,7 +826,7 @@ static int __init lp5523_init_engine(struct lp5523_engine *engine, int id) return 0; } -static int __init lp5523_init_led(struct lp5523_led *led, struct device *dev, +static int __devinit lp5523_init_led(struct lp5523_led *led, struct device *dev, int chan, struct lp5523_platform_data *pdata) { char name[32]; @@ -872,7 +872,7 @@ static int __init lp5523_init_led(struct lp5523_led *led, struct device *dev, static struct i2c_driver lp5523_driver; -static int lp5523_probe(struct i2c_client *client, +static int __devinit lp5523_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct lp5523_chip *chip; From 1fc6e987d8f606371337211f52ff74c6753298a6 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 27 Jun 2011 16:18:15 -0700 Subject: [PATCH 216/224] drivers/misc/ioc4.c: fix section mismatch / race condition Fix this section mismatch: WARNING: drivers/misc/ioc4.o(.data+0x144): Section mismatch in reference from the variable ioc4_load_modules_work to the function .devinit.text:ioc4_load_modules() The variable ioc4_load_modules_work references the function __devinit ioc4_load_modules() If the reference is valid then annotate the variable with __init* or __refdata (see linux/init.h) or name the variable: *driver, *_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console This one is potentially fatal; by the time ioc4_load_modules is invoked it may already have been freed. For that reason ioc4_load_modules_work can't be turned to __devinitdata but also because it's referenced in ioc4_exit. Signed-off-by: Ralf Baechle Acked-by: Brent Casavant Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/ioc4.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/misc/ioc4.c b/drivers/misc/ioc4.c index 668d41e594a9..df03dd3bd0e2 100644 --- a/drivers/misc/ioc4.c +++ b/drivers/misc/ioc4.c @@ -270,7 +270,7 @@ ioc4_variant(struct ioc4_driver_data *idd) return IOC4_VARIANT_PCI_RT; } -static void __devinit +static void ioc4_load_modules(struct work_struct *work) { request_module("sgiioc4"); From a39bce7bf60e728cb33b6b0415c3f44e7f1a102b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 27 Jun 2011 16:18:16 -0700 Subject: [PATCH 217/224] drivers/tty/serial/8250_pci.c: fix warning Fis the warning drivers/tty/serial/8250_pci.c:1457: warning: initialization from incompatible pointer type Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/tty/serial/8250_pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/8250_pci.c b/drivers/tty/serial/8250_pci.c index 4b4968a294b2..78e98a5cef96 100644 --- a/drivers/tty/serial/8250_pci.c +++ b/drivers/tty/serial/8250_pci.c @@ -973,7 +973,7 @@ ce4100_serial_setup(struct serial_private *priv, static int pci_omegapci_setup(struct serial_private *priv, - struct pciserial_board *board, + const struct pciserial_board *board, struct uart_port *port, int idx) { return setup_port(priv, port, 2, idx * 8, 0); From 3142b651ad2232cf0e375c291ee4b893c8559df5 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:17 -0700 Subject: [PATCH 218/224] drm/ttm: use shmem_read_mapping_page Soon tmpfs will stop supporting ->readpage and read_mapping_page(): once "tmpfs: add shmem_read_mapping_page_gfp" has been applied, this patch can be applied to ease the transition. ttm_tt_swapin() and ttm_tt_swapout() use shmem_read_mapping_page() in place of read_mapping_page(), since their swap_space has been created with shmem_file_setup(). Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Cc: Thomas Hellstrom Cc: Dave Airlie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpu/drm/ttm/ttm_tt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index 90e23e0bfadb..58c271ebc0f7 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -484,7 +485,7 @@ static int ttm_tt_swapin(struct ttm_tt *ttm) swap_space = swap_storage->f_path.dentry->d_inode->i_mapping; for (i = 0; i < ttm->num_pages; ++i) { - from_page = read_mapping_page(swap_space, i, NULL); + from_page = shmem_read_mapping_page(swap_space, i); if (IS_ERR(from_page)) { ret = PTR_ERR(from_page); goto out_err; @@ -557,7 +558,7 @@ int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage) from_page = ttm->pages[i]; if (unlikely(from_page == NULL)) continue; - to_page = read_mapping_page(swap_space, i, NULL); + to_page = shmem_read_mapping_page(swap_space, i); if (unlikely(IS_ERR(to_page))) { ret = PTR_ERR(to_page); goto out_err; From 5949eac4d9b5bf936c12cb7ec3a09084c1326834 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:18 -0700 Subject: [PATCH 219/224] drm/i915: use shmem_read_mapping_page Soon tmpfs will stop supporting ->readpage and read_cache_page_gfp(): once "tmpfs: add shmem_read_mapping_page_gfp" has been applied, this patch can be applied to ease the transition. Make i915_gem_object_get_pages_gtt() use shmem_read_mapping_page_gfp() in the one place it's needed; elsewhere use shmem_read_mapping_page(), with the mapping's gfp_mask properly initialized. Forget about __GFP_COLD: since tmpfs initializes its pages with memset, asking for a cold page is counter-productive. Include linux/shmem_fs.h also in drm_gem.c: with shmem_file_setup() now declared there too, we shall remove the prototype from linux/mm.h later. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Cc: Chris Wilson Cc: Keith Packard Cc: Dave Airlie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpu/drm/drm_gem.c | 1 + drivers/gpu/drm/i915/i915_gem.c | 31 ++++++++++++++----------------- 2 files changed, 15 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 74e4ff578017..4012fe423460 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -34,6 +34,7 @@ #include #include #include +#include #include "drmP.h" /** @file drm_gem.c diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index c6389de53161..fa560ceba667 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -31,6 +31,7 @@ #include "i915_drv.h" #include "i915_trace.h" #include "intel_drv.h" +#include #include #include #include @@ -359,8 +360,7 @@ i915_gem_shmem_pread_fast(struct drm_device *dev, if ((page_offset + remain) > PAGE_SIZE) page_length = PAGE_SIZE - page_offset; - page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT, - GFP_HIGHUSER | __GFP_RECLAIMABLE); + page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT); if (IS_ERR(page)) return PTR_ERR(page); @@ -463,8 +463,7 @@ i915_gem_shmem_pread_slow(struct drm_device *dev, if ((data_page_offset + page_length) > PAGE_SIZE) page_length = PAGE_SIZE - data_page_offset; - page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT, - GFP_HIGHUSER | __GFP_RECLAIMABLE); + page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT); if (IS_ERR(page)) { ret = PTR_ERR(page); goto out; @@ -797,8 +796,7 @@ i915_gem_shmem_pwrite_fast(struct drm_device *dev, if ((page_offset + remain) > PAGE_SIZE) page_length = PAGE_SIZE - page_offset; - page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT, - GFP_HIGHUSER | __GFP_RECLAIMABLE); + page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT); if (IS_ERR(page)) return PTR_ERR(page); @@ -907,8 +905,7 @@ i915_gem_shmem_pwrite_slow(struct drm_device *dev, if ((data_page_offset + page_length) > PAGE_SIZE) page_length = PAGE_SIZE - data_page_offset; - page = read_cache_page_gfp(mapping, offset >> PAGE_SHIFT, - GFP_HIGHUSER | __GFP_RECLAIMABLE); + page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT); if (IS_ERR(page)) { ret = PTR_ERR(page); goto out; @@ -1558,12 +1555,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj, inode = obj->base.filp->f_path.dentry->d_inode; mapping = inode->i_mapping; + gfpmask |= mapping_gfp_mask(mapping); + for (i = 0; i < page_count; i++) { - page = read_cache_page_gfp(mapping, i, - GFP_HIGHUSER | - __GFP_COLD | - __GFP_RECLAIMABLE | - gfpmask); + page = shmem_read_mapping_page_gfp(mapping, i, gfpmask); if (IS_ERR(page)) goto err_pages; @@ -3565,6 +3560,7 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, { struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_gem_object *obj; + struct address_space *mapping; obj = kzalloc(sizeof(*obj), GFP_KERNEL); if (obj == NULL) @@ -3575,6 +3571,9 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, return NULL; } + mapping = obj->base.filp->f_path.dentry->d_inode->i_mapping; + mapping_set_gfp_mask(mapping, GFP_HIGHUSER | __GFP_RECLAIMABLE); + i915_gem_info_add_obj(dev_priv, size); obj->base.write_domain = I915_GEM_DOMAIN_CPU; @@ -3950,8 +3949,7 @@ void i915_gem_detach_phys_object(struct drm_device *dev, page_count = obj->base.size / PAGE_SIZE; for (i = 0; i < page_count; i++) { - struct page *page = read_cache_page_gfp(mapping, i, - GFP_HIGHUSER | __GFP_RECLAIMABLE); + struct page *page = shmem_read_mapping_page(mapping, i); if (!IS_ERR(page)) { char *dst = kmap_atomic(page); memcpy(dst, vaddr + i*PAGE_SIZE, PAGE_SIZE); @@ -4012,8 +4010,7 @@ i915_gem_attach_phys_object(struct drm_device *dev, struct page *page; char *dst, *src; - page = read_cache_page_gfp(mapping, i, - GFP_HIGHUSER | __GFP_RECLAIMABLE); + page = shmem_read_mapping_page(mapping, i); if (IS_ERR(page)) return PTR_ERR(page); From e2377fe0b65e3c7577ff6df1701c56ef477d336f Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:19 -0700 Subject: [PATCH 220/224] drm/i915: use shmem_truncate_range The interface to ->truncate_range is changing very slightly: once "tmpfs: take control of its truncate_range" has been applied, this can be applied. For now there is only a slight inefficiency while this remains unapplied, but it will soon become essential for managing shmem's use of swap. Change i915_gem_object_truncate() to use shmem_truncate_range() directly: which should also spare i915 later change if we switch from inode_operations->truncate_range to file_operations->fallocate. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Cc: Chris Wilson Cc: Keith Packard Cc: Dave Airlie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpu/drm/i915/i915_gem.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index fa560ceba667..85f713746a1f 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1696,13 +1696,10 @@ i915_gem_object_truncate(struct drm_i915_gem_object *obj) /* Our goal here is to return as much of the memory as * is possible back to the system as we are called from OOM. * To do this we must instruct the shmfs to drop all of its - * backing pages, *now*. Here we mirror the actions taken - * when by shmem_delete_inode() to release the backing store. + * backing pages, *now*. */ inode = obj->base.filp->f_path.dentry->d_inode; - truncate_inode_pages(inode->i_mapping, 0); - if (inode->i_op->truncate_range) - inode->i_op->truncate_range(inode, 0, (loff_t)-1); + shmem_truncate_range(inode, 0, (loff_t)-1); obj->madv = __I915_MADV_PURGED; } From ecbec53b1d00ba582f71b210ed96cafc05ebd189 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 27 Jun 2011 16:18:20 -0700 Subject: [PATCH 221/224] drm/i915: more struct_mutex locking When auditing the locking in i915_gem.c (for a prospective change which I then abandoned), I noticed two places where struct_mutex is not held across GEM object manipulations that would usually require it. Since one is in initial setup and the other in driver unload, I'm guessing the mutex is not required for either; but post a patch in case it is. Signed-off-by: Hugh Dickins Cc: Chris Wilson Cc: Keith Packard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpu/drm/i915/i915_dma.c | 3 +-- drivers/gpu/drm/i915/intel_overlay.c | 5 +++++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 0239e9974bf2..2b79588541e7 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -2182,9 +2182,8 @@ int i915_driver_unload(struct drm_device *dev) /* Flush any outstanding unpin_work. */ flush_workqueue(dev_priv->wq); - i915_gem_free_all_phys_object(dev); - mutex_lock(&dev->struct_mutex); + i915_gem_free_all_phys_object(dev); i915_gem_cleanup_ringbuffer(dev); mutex_unlock(&dev->struct_mutex); if (I915_HAS_FBC(dev) && i915_powersave) diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c index a670c006982e..56a8e2aea19c 100644 --- a/drivers/gpu/drm/i915/intel_overlay.c +++ b/drivers/gpu/drm/i915/intel_overlay.c @@ -1416,6 +1416,8 @@ void intel_setup_overlay(struct drm_device *dev) goto out_free; overlay->reg_bo = reg_bo; + mutex_lock(&dev->struct_mutex); + if (OVERLAY_NEEDS_PHYSICAL(dev)) { ret = i915_gem_attach_phys_object(dev, reg_bo, I915_GEM_PHYS_OVERLAY_REGS, @@ -1440,6 +1442,8 @@ void intel_setup_overlay(struct drm_device *dev) } } + mutex_unlock(&dev->struct_mutex); + /* init all values */ overlay->color_key = 0x0101fe; overlay->brightness = -19; @@ -1464,6 +1468,7 @@ out_unpin_bo: i915_gem_object_unpin(reg_bo); out_free_bo: drm_gem_object_unreference(®_bo->base); + mutex_unlock(&dev->struct_mutex); out_free: kfree(overlay); return; From b0af8dfdd67699e25083478c63eedef2e72ebd85 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 27 Jun 2011 19:12:22 -0700 Subject: [PATCH 222/224] Linux 3.0-rc5 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 41330a06e4ec..dc670467fb8c 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 0 SUBLEVEL = 0 -EXTRAVERSION = -rc4 +EXTRAVERSION = -rc5 NAME = Sneaky Weasel # *DOCUMENTATION* From 4274215d24633df7302069e51426659d4759c5ed Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 28 Jun 2011 16:59:42 +1000 Subject: [PATCH 223/224] md: avoid endless recovery loop when waiting for fail device to complete. If a device fails in a way that causes pending request to take a while to complete, md will not be able to immediately remove it from the array in remove_and_add_spares. It will then incorrectly look like a spare device and md will try to recover it even though it is failed. This leads to a recovery process starting and instantly aborting over and over again. We should check if the device is faulty before considering it to be a spare. This will avoid trying to start a recovery that cannot proceed. This bug was introduced in 2.6.26 so that patch is suitable for any kernel since then. Cc: stable@kernel.org Reported-by: Jim Paradis Signed-off-by: NeilBrown --- drivers/md/md.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/md/md.c b/drivers/md/md.c index 4332fc2f25d4..91e31e260b4a 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -7088,6 +7088,7 @@ static int remove_and_add_spares(mddev_t *mddev) list_for_each_entry(rdev, &mddev->disks, same_set) { if (rdev->raid_disk >= 0 && !test_bit(In_sync, &rdev->flags) && + !test_bit(Faulty, &rdev->flags) && !test_bit(Blocked, &rdev->flags)) spares++; if (rdev->raid_disk < 0 From 1d1221f375c94ef961ba8574ac4f85c8870ddd51 Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Fri, 24 Jun 2011 16:08:38 +0400 Subject: [PATCH 224/224] proc: restrict access to /proc/PID/io /proc/PID/io may be used for gathering private information. E.g. for openssh and vsftpd daemons wchars/rchars may be used to learn the precise password length. Restrict it to processes being able to ptrace the target process. ptrace_may_access() is needed to prevent keeping open file descriptor of "io" file, executing setuid binary and gathering io information of the setuid'ed process. Signed-off-by: Vasiliy Kulikov Signed-off-by: Linus Torvalds --- fs/proc/base.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 8a84210ca080..fc5bc2767692 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2708,6 +2708,9 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole) struct task_io_accounting acct = task->ioac; unsigned long flags; + if (!ptrace_may_access(task, PTRACE_MODE_READ)) + return -EACCES; + if (whole && lock_task_sighand(task, &flags)) { struct task_struct *t = task; @@ -2839,7 +2842,7 @@ static const struct pid_entry tgid_base_stuff[] = { REG("coredump_filter", S_IRUGO|S_IWUSR, proc_coredump_filter_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING - INF("io", S_IRUGO, proc_tgid_io_accounting), + INF("io", S_IRUSR, proc_tgid_io_accounting), #endif #ifdef CONFIG_HARDWALL INF("hardwall", S_IRUGO, proc_pid_hardwall), @@ -3181,7 +3184,7 @@ static const struct pid_entry tid_base_stuff[] = { REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING - INF("io", S_IRUGO, proc_tid_io_accounting), + INF("io", S_IRUSR, proc_tid_io_accounting), #endif #ifdef CONFIG_HARDWALL INF("hardwall", S_IRUGO, proc_pid_hardwall),