From 72d0d248ca8232dbd30d35b42d0d86e39b3e322b Mon Sep 17 00:00:00 2001 From: Brian Foster Date: Mon, 16 Jul 2012 15:23:48 -0400 Subject: fuse: add FUSE_AUTO_INVAL_DATA init flag FUSE_AUTO_INVAL_DATA is provided to enable updated/auto cache invalidation logic. Signed-off-by: Brian Foster Signed-off-by: Miklos Szeredi --- include/linux/fuse.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 9303348965fb..e4a9d2af9aaa 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -57,6 +57,9 @@ * * 7.19 * - add FUSE_FALLOCATE + * + * 7.20 + * - add FUSE_AUTO_INVAL_DATA */ #ifndef _LINUX_FUSE_H @@ -88,7 +91,7 @@ #define FUSE_KERNEL_VERSION 7 /** Minor version number of this interface */ -#define FUSE_KERNEL_MINOR_VERSION 19 +#define FUSE_KERNEL_MINOR_VERSION 20 /** The node ID of the root inode */ #define FUSE_ROOT_ID 1 @@ -167,6 +170,7 @@ struct fuse_file_lock { * FUSE_EXPORT_SUPPORT: filesystem handles lookups of "." and ".." * FUSE_DONT_MASK: don't apply umask to file mode on create operations * FUSE_FLOCK_LOCKS: remote locking for BSD style file locks + * FUSE_AUTO_INVAL_DATA: automatically invalidate cached pages */ #define FUSE_ASYNC_READ (1 << 0) #define FUSE_POSIX_LOCKS (1 << 1) @@ -176,6 +180,7 @@ struct fuse_file_lock { #define FUSE_BIG_WRITES (1 << 5) #define FUSE_DONT_MASK (1 << 6) #define FUSE_FLOCK_LOCKS (1 << 10) +#define FUSE_AUTO_INVAL_DATA (1 << 12) /** * CUSE INIT request/reply flags -- cgit v1.2.3 From 69fe05c90ed58aac956dccb9e6d3a325fb3b8767 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 18 Jul 2012 16:09:40 +0200 Subject: fuse: add missing INIT flags Add missing flags that userspace derived from the protocol version number. This makes the protocol more flexible. Signed-off-by: Miklos Szeredi --- fs/fuse/inode.c | 3 ++- include/linux/fuse.h | 8 ++++++++ 2 files changed, 10 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index dd4401650b47..ce0a2838ccd0 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -882,7 +882,8 @@ static void fuse_send_init(struct fuse_conn *fc, struct fuse_req *req) arg->max_readahead = fc->bdi.ra_pages * PAGE_CACHE_SIZE; arg->flags |= FUSE_ASYNC_READ | FUSE_POSIX_LOCKS | FUSE_ATOMIC_O_TRUNC | FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES | FUSE_DONT_MASK | - FUSE_FLOCK_LOCKS | FUSE_AUTO_INVAL_DATA; + FUSE_SPLICE_WRITE | FUSE_SPLICE_MOVE | FUSE_SPLICE_READ | + FUSE_FLOCK_LOCKS | FUSE_IOCTL_DIR | FUSE_AUTO_INVAL_DATA; req->in.h.opcode = FUSE_INIT; req->in.numargs = 1; req->in.args[0].size = sizeof(*arg); diff --git a/include/linux/fuse.h b/include/linux/fuse.h index e4a9d2af9aaa..6455c5b64c2e 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -169,7 +169,11 @@ struct fuse_file_lock { * FUSE_POSIX_LOCKS: remote locking for POSIX file locks * FUSE_EXPORT_SUPPORT: filesystem handles lookups of "." and ".." * FUSE_DONT_MASK: don't apply umask to file mode on create operations + * FUSE_SPLICE_WRITE: kernel supports splice write on the device + * FUSE_SPLICE_MOVE: kernel supports splice move on the device + * FUSE_SPLICE_READ: kernel supports splice read on the device * FUSE_FLOCK_LOCKS: remote locking for BSD style file locks + * FUSE_HAS_IOCTL_DIR: kernel supports ioctl on directories * FUSE_AUTO_INVAL_DATA: automatically invalidate cached pages */ #define FUSE_ASYNC_READ (1 << 0) @@ -179,7 +183,11 @@ struct fuse_file_lock { #define FUSE_EXPORT_SUPPORT (1 << 4) #define FUSE_BIG_WRITES (1 << 5) #define FUSE_DONT_MASK (1 << 6) +#define FUSE_SPLICE_WRITE (1 << 7) +#define FUSE_SPLICE_MOVE (1 << 8) +#define FUSE_SPLICE_READ (1 << 9) #define FUSE_FLOCK_LOCKS (1 << 10) +#define FUSE_HAS_IOCTL_DIR (1 << 11) #define FUSE_AUTO_INVAL_DATA (1 << 12) /** -- cgit v1.2.3 From f3840dc0fb57aef120c5ee8241cdc9aaf3cec8d4 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 18 Jul 2012 16:09:40 +0200 Subject: fuse: add missing INIT flag descriptions Signed-off-by: Miklos Szeredi --- include/linux/fuse.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fuse.h b/include/linux/fuse.h index 6455c5b64c2e..d8c713e148e3 100644 --- a/include/linux/fuse.h +++ b/include/linux/fuse.h @@ -166,8 +166,12 @@ struct fuse_file_lock { /** * INIT request/reply flags * + * FUSE_ASYNC_READ: asynchronous read requests * FUSE_POSIX_LOCKS: remote locking for POSIX file locks + * FUSE_FILE_OPS: kernel sends file handle for fstat, etc... (not yet supported) + * FUSE_ATOMIC_O_TRUNC: handles the O_TRUNC open flag in the filesystem * FUSE_EXPORT_SUPPORT: filesystem handles lookups of "." and ".." + * FUSE_BIG_WRITES: filesystem can handle write size larger than 4kB * FUSE_DONT_MASK: don't apply umask to file mode on create operations * FUSE_SPLICE_WRITE: kernel supports splice write on the device * FUSE_SPLICE_MOVE: kernel supports splice move on the device -- cgit v1.2.3 From dc9b229a58dc0dfed34272ff26c6d5fd17c674e0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 13 Jul 2012 19:29:45 +0200 Subject: genirq: Allow irq chips to mark themself oneshot safe Some interrupt chips like MSI are oneshot safe by implementation. For those interrupts we can avoid the mask/unmask sequence for threaded interrupt handlers. Signed-off-by: Thomas Gleixner Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1207132056540.32033@ionos Cc: Linus Torvalds Cc: Avi Kivity Cc: Marcelo Tosatti Cc: Jan Kiszka --- include/linux/irq.h | 1 + kernel/irq/manage.c | 15 ++++++++++++++- 2 files changed, 15 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/irq.h b/include/linux/irq.h index 553fb66da130..216b0ba109d7 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -349,6 +349,7 @@ enum { IRQCHIP_MASK_ON_SUSPEND = (1 << 2), IRQCHIP_ONOFFLINE_ENABLED = (1 << 3), IRQCHIP_SKIP_SET_WAKE = (1 << 4), + IRQCHIP_ONESHOT_SAFE = (1 << 5), }; /* This include will go away once we isolated irq_desc usage to core code */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 8c548232ba39..2e326d1ebec1 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -959,6 +959,18 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) goto out_thread; } + /* + * Drivers are often written to work w/o knowledge about the + * underlying irq chip implementation, so a request for a + * threaded irq without a primary hard irq context handler + * requires the ONESHOT flag to be set. Some irq chips like + * MSI based interrupts are per se one shot safe. Check the + * chip flags, so we can avoid the unmask dance at the end of + * the threaded handler for those. + */ + if (desc->irq_data.chip->flags & IRQCHIP_ONESHOT_SAFE) + new->flags &= ~IRQF_ONESHOT; + /* * The following block of code has to be executed atomically */ @@ -1033,7 +1045,8 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) */ new->thread_mask = 1 << ffz(thread_mask); - } else if (new->handler == irq_default_primary_handler) { + } else if (new->handler == irq_default_primary_handler && + !(desc->irq_data.chip->flags & IRQCHIP_ONESHOT_SAFE)) { /* * The interrupt was requested with handler = NULL, so * we use the default primary handler for it. But it -- cgit v1.2.3 From 6956dc568f34107f1d02b24f87efe7250803fc87 Mon Sep 17 00:00:00 2001 From: Alex Shi Date: Fri, 20 Jul 2012 14:19:50 +0800 Subject: sched/numa: Add SD_PERFER_SIBLING to CPU domain Commit 8e7fbcbc22c ("sched: Remove stale power aware scheduling remnants and dysfunctional knobs") removed SD_PERFER_SIBLING from the CPU domain. On NUMA machines this causes that load_balance() doesn't perfer LCPU in same physical CPU package. It causes some actual performance regressions on our NUMA machines from Core2 to NHM and SNB. Adding this domain flag again recovers the performance drop. This change doesn't have any bad impact on any of my benchmarks: specjbb, kbuild, fio, hackbench .. etc, on all my machines. Signed-off-by: Alex Shi Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1342765190-21540-1-git-send-email-alex.shi@intel.com Signed-off-by: Ingo Molnar --- include/linux/topology.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/topology.h b/include/linux/topology.h index e91cd43394df..fec12d667211 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -164,6 +164,7 @@ int arch_update_cpu_topology(void); | 0*SD_SHARE_CPUPOWER \ | 0*SD_SHARE_PKG_RESOURCES \ | 0*SD_SERIALIZE \ + | 1*SD_PREFER_SIBLING \ , \ .last_balance = jiffies, \ .balance_interval = 1, \ -- cgit v1.2.3 From a7e4786b937a3ae918a7520cfdba557a80915fa7 Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Sat, 21 Jul 2012 00:54:59 +0530 Subject: sched: Fix comment about PREEMPT_ACTIVE bit location PREEMPT_ACTIVE flag is bit 27, not 28. Fix the comment. Signed-off-by: Srivatsa S. Bhat Cc: paulmck@linux.vnet.ibm.com Cc: josh@joshtriplett.org Cc: rostedt@goodmis.org Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120720192459.6149.14821.stgit@srivatsabhat.in.ibm.com Signed-off-by: Ingo Molnar --- include/linux/hardirq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index bb7f30971858..305f23cd7cff 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -22,7 +22,7 @@ * * - bits 16-25 are the hardirq count (max # of nested hardirqs: 1024) * - bit 26 is the NMI_MASK - * - bit 28 is the PREEMPT_ACTIVE flag + * - bit 27 is the PREEMPT_ACTIVE flag * * PREEMPT_MASK: 0x000000ff * SOFTIRQ_MASK: 0x0000ff00 -- cgit v1.2.3 From 0f26d0e0a715556270d85b7946b99546a2f92888 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 30 Jul 2012 22:44:41 -0500 Subject: kdb: Remove unused KDB_FLAG_ONLY_DO_DUMP This code cleanup was missed in the original kdb merge, and this code is simply not used at all. The code that was previously used to set the KDB_FLAG_ONLY_DO_DUMP was removed prior to the initial kdb merge. Signed-off-by: Jason Wessel --- include/linux/kdb.h | 2 -- kernel/debug/kdb/kdb_main.c | 15 +-------------- 2 files changed, 1 insertion(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kdb.h b/include/linux/kdb.h index 064725854db8..42d9e863a313 100644 --- a/include/linux/kdb.h +++ b/include/linux/kdb.h @@ -75,8 +75,6 @@ extern const char *kdb_diemsg; #define KDB_FLAG_CATASTROPHIC (1 << 1) /* A catastrophic event has occurred */ #define KDB_FLAG_CMD_INTERRUPT (1 << 2) /* Previous command was interrupted */ #define KDB_FLAG_NOIPI (1 << 3) /* Do not send IPIs */ -#define KDB_FLAG_ONLY_DO_DUMP (1 << 4) /* Only do a dump, used when - * kdb is off */ #define KDB_FLAG_NO_CONSOLE (1 << 5) /* No console is available, * kdb is disabled */ #define KDB_FLAG_NO_VT_CONSOLE (1 << 6) /* No VT console is available, do diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index 1f91413edb87..31df1706b9a9 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -139,11 +139,10 @@ static const int __nkdb_err = sizeof(kdbmsgs) / sizeof(kdbmsg_t); static char *__env[] = { #if defined(CONFIG_SMP) "PROMPT=[%d]kdb> ", - "MOREPROMPT=[%d]more> ", #else "PROMPT=kdb> ", - "MOREPROMPT=more> ", #endif + "MOREPROMPT=more> ", "RADIX=16", "MDCOUNT=8", /* lines of md output */ KDB_PLATFORM_ENV, @@ -1236,18 +1235,6 @@ static int kdb_local(kdb_reason_t reason, int error, struct pt_regs *regs, *cmdbuf = '\0'; *(cmd_hist[cmd_head]) = '\0'; - if (KDB_FLAG(ONLY_DO_DUMP)) { - /* kdb is off but a catastrophic error requires a dump. - * Take the dump and reboot. - * Turn on logging so the kdb output appears in the log - * buffer in the dump. - */ - const char *setargs[] = { "set", "LOGGING", "1" }; - kdb_set(2, setargs); - kdb_reboot(0, NULL); - /*NOTREACHED*/ - } - do_full_getstr: #if defined(CONFIG_SMP) snprintf(kdb_prompt_str, CMD_BUFLEN, kdbgetenv("PROMPT"), -- cgit v1.2.3 From e6dab5ffab59e910ec0e3355f4a6f29f7a7be474 Mon Sep 17 00:00:00 2001 From: Andrew Vagin Date: Wed, 11 Jul 2012 18:14:58 +0400 Subject: perf/trace: Add ability to set a target task for events A few events are interesting not only for a current task. For example, sched_stat_* events are interesting for a task which wakes up. For this reason, it will be good if such events will be delivered to a target task too. Now a target task can be set by using __perf_task(). The original idea and a draft patch belongs to Peter Zijlstra. I need these events for profiling sleep times. sched_switch is used for getting callchains and sched_stat_* is used for getting time periods. These events are combined in user space, then it can be analyzed by perf tools. Inspired-by: Peter Zijlstra Cc: Steven Rostedt Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Cc: Steven Rostedt Cc: Arun Sharma Signed-off-by: Andrew Vagin Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1342016098-213063-1-git-send-email-avagin@openvz.org Signed-off-by: Ingo Molnar --- include/linux/ftrace_event.h | 5 +++-- include/linux/perf_event.h | 3 ++- include/trace/events/sched.h | 4 ++++ include/trace/ftrace.h | 6 +++++- kernel/events/callchain.c | 9 ++++++++- kernel/events/core.c | 30 ++++++++++++++++++++++++++++-- kernel/events/internal.h | 3 ++- kernel/trace/trace_event_perf.c | 2 +- kernel/trace/trace_kprobe.c | 6 ++++-- kernel/trace/trace_syscalls.c | 4 ++-- kernel/trace/trace_uprobe.c | 2 +- 11 files changed, 60 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index af961d6f7ab1..642928cf57b4 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -306,9 +306,10 @@ extern void *perf_trace_buf_prepare(int size, unsigned short type, static inline void perf_trace_buf_submit(void *raw_data, int size, int rctx, u64 addr, - u64 count, struct pt_regs *regs, void *head) + u64 count, struct pt_regs *regs, void *head, + struct task_struct *task) { - perf_tp_event(addr, count, raw_data, size, regs, head, rctx); + perf_tp_event(addr, count, raw_data, size, regs, head, rctx, task); } #endif diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 76c5c8b724a7..7602ccb3f40e 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1272,7 +1272,8 @@ static inline bool perf_paranoid_kernel(void) extern void perf_event_init(void); extern void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, struct pt_regs *regs, - struct hlist_head *head, int rctx); + struct hlist_head *head, int rctx, + struct task_struct *task); extern void perf_bp_event(struct perf_event *event, void *data); #ifndef perf_misc_flags diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index ea7a2035456d..5a8671e8a67f 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -73,6 +73,9 @@ DECLARE_EVENT_CLASS(sched_wakeup_template, __entry->prio = p->prio; __entry->success = success; __entry->target_cpu = task_cpu(p); + ) + TP_perf_assign( + __perf_task(p); ), TP_printk("comm=%s pid=%d prio=%d success=%d target_cpu=%03d", @@ -325,6 +328,7 @@ DECLARE_EVENT_CLASS(sched_stat_template, ) TP_perf_assign( __perf_count(delay); + __perf_task(tsk); ), TP_printk("comm=%s pid=%d delay=%Lu [ns]", diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index c6bc2faaf261..a763888a36f9 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -712,6 +712,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call #undef __perf_count #define __perf_count(c) __count = (c) +#undef __perf_task +#define __perf_task(t) __task = (t) + #undef TP_perf_assign #define TP_perf_assign(args...) args @@ -725,6 +728,7 @@ perf_trace_##call(void *__data, proto) \ struct ftrace_raw_##call *entry; \ struct pt_regs __regs; \ u64 __addr = 0, __count = 1; \ + struct task_struct *__task = NULL; \ struct hlist_head *head; \ int __entry_size; \ int __data_size; \ @@ -752,7 +756,7 @@ perf_trace_##call(void *__data, proto) \ \ head = this_cpu_ptr(event_call->perf_events); \ perf_trace_buf_submit(entry, __entry_size, rctx, __addr, \ - __count, &__regs, head); \ + __count, &__regs, head, __task); \ } /* diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c index 6581a040f399..98d4597f43d6 100644 --- a/kernel/events/callchain.c +++ b/kernel/events/callchain.c @@ -153,7 +153,8 @@ put_callchain_entry(int rctx) put_recursion_context(__get_cpu_var(callchain_recursion), rctx); } -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +struct perf_callchain_entry * +perf_callchain(struct perf_event *event, struct pt_regs *regs) { int rctx; struct perf_callchain_entry *entry; @@ -178,6 +179,12 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) } if (regs) { + /* + * Disallow cross-task user callchains. + */ + if (event->ctx->task && event->ctx->task != current) + goto exit_put; + perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_user(entry, regs); } diff --git a/kernel/events/core.c b/kernel/events/core.c index f1cf0edeb39a..b7935fcec7d9 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4039,7 +4039,7 @@ void perf_prepare_sample(struct perf_event_header *header, if (sample_type & PERF_SAMPLE_CALLCHAIN) { int size = 1; - data->callchain = perf_callchain(regs); + data->callchain = perf_callchain(event, regs); if (data->callchain) size += data->callchain->nr; @@ -5209,7 +5209,8 @@ static int perf_tp_event_match(struct perf_event *event, } void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, - struct pt_regs *regs, struct hlist_head *head, int rctx) + struct pt_regs *regs, struct hlist_head *head, int rctx, + struct task_struct *task) { struct perf_sample_data data; struct perf_event *event; @@ -5228,6 +5229,31 @@ void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, perf_swevent_event(event, count, &data, regs); } + /* + * If we got specified a target task, also iterate its context and + * deliver this event there too. + */ + if (task && task != current) { + struct perf_event_context *ctx; + struct trace_entry *entry = record; + + rcu_read_lock(); + ctx = rcu_dereference(task->perf_event_ctxp[perf_sw_context]); + if (!ctx) + goto unlock; + + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (event->attr.type != PERF_TYPE_TRACEPOINT) + continue; + if (event->attr.config != entry->type) + continue; + if (perf_tp_event_match(event, &data, regs)) + perf_swevent_event(event, count, &data, regs); + } +unlock: + rcu_read_unlock(); + } + perf_swevent_put_recursion_context(rctx); } EXPORT_SYMBOL_GPL(perf_tp_event); diff --git a/kernel/events/internal.h b/kernel/events/internal.h index b0b107f90afc..a096c19f2c2a 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -101,7 +101,8 @@ __output_copy(struct perf_output_handle *handle, } /* Callchain handling */ -extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs); +extern struct perf_callchain_entry * +perf_callchain(struct perf_event *event, struct pt_regs *regs); extern int get_callchain_buffers(void); extern void put_callchain_buffers(void); diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index fee3752ae8f6..8a6d2ee2086c 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -281,7 +281,7 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip) head = this_cpu_ptr(event_function.perf_events); perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0, - 1, ®s, head); + 1, ®s, head, NULL); #undef ENTRY_SIZE } diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index b31d3d5699fe..1a2117043bb1 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1002,7 +1002,8 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp, store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); head = this_cpu_ptr(call->perf_events); - perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head); + perf_trace_buf_submit(entry, size, rctx, + entry->ip, 1, regs, head, NULL); } /* Kretprobe profile handler */ @@ -1033,7 +1034,8 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri, store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); head = this_cpu_ptr(call->perf_events); - perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, regs, head); + perf_trace_buf_submit(entry, size, rctx, + entry->ret_ip, 1, regs, head, NULL); } #endif /* CONFIG_PERF_EVENTS */ diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 96fc73369099..60e4d7875672 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -532,7 +532,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) (unsigned long *)&rec->args); head = this_cpu_ptr(sys_data->enter_event->perf_events); - perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head); + perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head, NULL); } int perf_sysenter_enable(struct ftrace_event_call *call) @@ -608,7 +608,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->ret = syscall_get_return_value(current, regs); head = this_cpu_ptr(sys_data->exit_event->perf_events); - perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head); + perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head, NULL); } int perf_sysexit_enable(struct ftrace_event_call *call) diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 2b36ac68549e..03003cd7dd96 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -670,7 +670,7 @@ static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs) call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset); head = this_cpu_ptr(call->perf_events); - perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head); + perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head, NULL); out: preempt_enable(); -- cgit v1.2.3 From a7ea3bbf5d58f4df2265d312f91d5769eabc8144 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Fri, 27 Jul 2012 14:48:09 -0400 Subject: time/jiffies: Allow CLOCK_TICK_RATE to be undefined CLOCK_TICK_RATE is a legacy constant that defines the timer device's granularity. On hardware with particularly coarse granularity, this constant is used to reduce accumulated time error when using jiffies as a clocksource, by calculating the hardware's actual tick length rather then just assuming it is 1sec/HZ. However, for the most part this is unnecessary, as most modern systems don't use jiffies for their clocksource, and their tick device is sufficiently fine grained to avoid major error. Thus, this patch allows an architecture to not define CLOCK_TICK_RATE, in which case ACTHZ defaults to (HZ << 8). Signed-off-by: Catalin Marinas Acked-by: Arnd Bergmann Cc: Richard Cochran Cc: Prarit Bhargava Cc: Andrew Morton [ Commit log & intention tweaks ] Signed-off-by: John Stultz Link: http://lkml.kernel.org/r/1343414893-45779-2-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- include/linux/jiffies.h | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index 265e2c3cbd1c..7d24466fb80b 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -39,9 +39,6 @@ # error Invalid value of HZ. #endif -/* LATCH is used in the interval timer and ftape setup. */ -#define LATCH ((CLOCK_TICK_RATE + HZ/2) / HZ) /* For divider */ - /* Suppose we want to divide two numbers NOM and DEN: NOM/DEN, then we can * improve accuracy by shifting LSH bits, hence calculating: * (NOM << LSH) / DEN @@ -54,8 +51,15 @@ #define SH_DIV(NOM,DEN,LSH) ( (((NOM) / (DEN)) << (LSH)) \ + ((((NOM) % (DEN)) << (LSH)) + (DEN) / 2) / (DEN)) +#ifdef CLOCK_TICK_RATE +/* LATCH is used in the interval timer and ftape setup. */ +# define LATCH ((CLOCK_TICK_RATE + HZ/2) / HZ) /* For divider */ + /* HZ is the requested value. ACTHZ is actual HZ ("<< 8" is for accuracy) */ -#define ACTHZ (SH_DIV (CLOCK_TICK_RATE, LATCH, 8)) +# define ACTHZ (SH_DIV(CLOCK_TICK_RATE, LATCH, 8)) +#else +# define ACTHZ (HZ << 8) +#endif /* TICK_NSEC is the time between ticks in nsec assuming real ACTHZ */ #define TICK_NSEC (SH_DIV (1000000UL * 1000, ACTHZ, 8)) -- cgit v1.2.3 From 02ab20ae38337b99b5c29c81090f594b8fd61283 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 27 Jul 2012 14:48:10 -0400 Subject: time/jiffies: Rename ACTHZ to SHIFTED_HZ Ingo noted that ACTHZ is a confusing name, and requested it be renamed, so this patch renames ACTHZ to SHIFTED_HZ to better describe it. Signed-off-by: John Stultz Cc: Prarit Bhargava Link: http://lkml.kernel.org/r/1343414893-45779-3-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- include/linux/jiffies.h | 21 +++++++++++++-------- include/linux/timex.h | 2 +- kernel/time/jiffies.c | 2 +- kernel/time/ntp.c | 2 +- 4 files changed, 16 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index 7d24466fb80b..82680541576d 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -55,21 +55,26 @@ /* LATCH is used in the interval timer and ftape setup. */ # define LATCH ((CLOCK_TICK_RATE + HZ/2) / HZ) /* For divider */ -/* HZ is the requested value. ACTHZ is actual HZ ("<< 8" is for accuracy) */ -# define ACTHZ (SH_DIV(CLOCK_TICK_RATE, LATCH, 8)) +/* + * HZ is the requested value. However the CLOCK_TICK_RATE may not allow + * for exactly HZ. So SHIFTED_HZ is high res HZ ("<< 8" is for accuracy) + */ +# define SHIFTED_HZ (SH_DIV(CLOCK_TICK_RATE, LATCH, 8)) #else -# define ACTHZ (HZ << 8) +# define SHIFTED_HZ (HZ << 8) #endif -/* TICK_NSEC is the time between ticks in nsec assuming real ACTHZ */ -#define TICK_NSEC (SH_DIV (1000000UL * 1000, ACTHZ, 8)) +/* TICK_NSEC is the time between ticks in nsec assuming SHIFTED_HZ */ +#define TICK_NSEC (SH_DIV(1000000UL * 1000, SHIFTED_HZ, 8)) /* TICK_USEC is the time between ticks in usec assuming fake USER_HZ */ #define TICK_USEC ((1000000UL + USER_HZ/2) / USER_HZ) -/* TICK_USEC_TO_NSEC is the time between ticks in nsec assuming real ACTHZ and */ -/* a value TUSEC for TICK_USEC (can be set bij adjtimex) */ -#define TICK_USEC_TO_NSEC(TUSEC) (SH_DIV (TUSEC * USER_HZ * 1000, ACTHZ, 8)) +/* + * TICK_USEC_TO_NSEC is the time between ticks in nsec assuming SHIFTED_HZ and + * a value TUSEC for TICK_USEC (can be set bij adjtimex) + */ +#define TICK_USEC_TO_NSEC(TUSEC) (SH_DIV(TUSEC * USER_HZ * 1000, SHIFTED_HZ, 8)) /* some arch's have a small-data section that can be accessed register-relative * but that can only take up to, say, 4-byte variables. jiffies being part of diff --git a/include/linux/timex.h b/include/linux/timex.h index 99bc88b1fc02..7c5ceb20e03a 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -232,7 +232,7 @@ struct timex { * estimated error = NTP dispersion. */ extern unsigned long tick_usec; /* USER_HZ period (usec) */ -extern unsigned long tick_nsec; /* ACTHZ period (nsec) */ +extern unsigned long tick_nsec; /* SHIFTED_HZ period (nsec) */ extern void ntp_init(void); extern void ntp_clear(void); diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index a470154e0408..46da0537c10b 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -37,7 +37,7 @@ * requested HZ value. It is also not recommended * for "tick-less" systems. */ -#define NSEC_PER_JIFFY ((u32)((((u64)NSEC_PER_SEC)<<8)/ACTHZ)) +#define NSEC_PER_JIFFY ((u32)((((u64)NSEC_PER_SEC)<<8)/SHIFTED_HZ)) /* Since jiffies uses a simple NSEC_PER_JIFFY multiplier * conversion, the .shift value could be zero. However diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index b7fbadc5c973..24174b4d669b 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -28,7 +28,7 @@ DEFINE_SPINLOCK(ntp_lock); /* USER_HZ period (usecs): */ unsigned long tick_usec = TICK_USEC; -/* ACTHZ period (nsecs): */ +/* SHIFTED_HZ period (nsecs): */ unsigned long tick_nsec; static u64 tick_length; -- cgit v1.2.3 From 392a325c4351339cfbf182bb5a1444df1cf65dbb Mon Sep 17 00:00:00 2001 From: Andres Salomon Date: Tue, 10 Jul 2012 19:31:51 -0700 Subject: Platform: OLPC: add a stub to drivers/platform/ for the OLPC EC driver The OLPC EC driver has outgrown arch/x86/platform/. It's time to both share common code amongst different architectures, as well as move it out of arch/x86/. The XO-1.75 is ARM-based, and the EC driver shares a lot of code with the x86 code. Signed-off-by: Andres Salomon Acked-by: Paul Fox Reviewed-by: Thomas Gleixner --- arch/x86/include/asm/olpc.h | 19 +++---------------- arch/x86/platform/olpc/olpc.c | 4 ++-- drivers/platform/Makefile | 1 + drivers/platform/olpc/Makefile | 4 ++++ drivers/platform/olpc/olpc-ec.c | 16 ++++++++++++++++ include/linux/olpc-ec.h | 29 +++++++++++++++++++++++++++++ 6 files changed, 55 insertions(+), 18 deletions(-) create mode 100644 drivers/platform/olpc/Makefile create mode 100644 drivers/platform/olpc/olpc-ec.c create mode 100644 include/linux/olpc-ec.h (limited to 'include/linux') diff --git a/arch/x86/include/asm/olpc.h b/arch/x86/include/asm/olpc.h index 87bdbca72f94..513e9992771d 100644 --- a/arch/x86/include/asm/olpc.h +++ b/arch/x86/include/asm/olpc.h @@ -4,6 +4,7 @@ #define _ASM_X86_OLPC_H #include +#include struct olpc_platform_t { int flags; @@ -102,22 +103,8 @@ extern int pci_olpc_init(void); /* EC related functions */ -extern int olpc_ec_cmd(unsigned char cmd, unsigned char *inbuf, size_t inlen, - unsigned char *outbuf, size_t outlen); - -/* EC commands */ - -#define EC_FIRMWARE_REV 0x08 -#define EC_WRITE_SCI_MASK 0x1b -#define EC_WAKE_UP_WLAN 0x24 -#define EC_WLAN_LEAVE_RESET 0x25 -#define EC_READ_EB_MODE 0x2a -#define EC_SET_SCI_INHIBIT 0x32 -#define EC_SET_SCI_INHIBIT_RELEASE 0x34 -#define EC_WLAN_ENTER_RESET 0x35 -#define EC_WRITE_EXT_SCI_MASK 0x38 -#define EC_SCI_QUERY 0x84 -#define EC_EXT_SCI_QUERY 0x85 +extern int olpc_ec_cmd_x86(unsigned char cmd, unsigned char *inbuf, + size_t inlen, unsigned char *outbuf, size_t outlen); /* SCI source values */ diff --git a/arch/x86/platform/olpc/olpc.c b/arch/x86/platform/olpc/olpc.c index a4bee53c2e54..796e199ac77a 100644 --- a/arch/x86/platform/olpc/olpc.c +++ b/arch/x86/platform/olpc/olpc.c @@ -125,7 +125,7 @@ static int __wait_on_obf(unsigned int line, unsigned int port, int desired) * . Unfortunately, while * OpenFirmware's source is available, the EC's is not. */ -int olpc_ec_cmd(unsigned char cmd, unsigned char *inbuf, size_t inlen, +int olpc_ec_cmd_x86(unsigned char cmd, unsigned char *inbuf, size_t inlen, unsigned char *outbuf, size_t outlen) { unsigned long flags; @@ -201,7 +201,7 @@ err: spin_unlock_irqrestore(&ec_lock, flags); return ret; } -EXPORT_SYMBOL_GPL(olpc_ec_cmd); +EXPORT_SYMBOL_GPL(olpc_ec_cmd_x86); void olpc_ec_wakeup_set(u16 value) { diff --git a/drivers/platform/Makefile b/drivers/platform/Makefile index 782953ae4c03..b17c16ce54ad 100644 --- a/drivers/platform/Makefile +++ b/drivers/platform/Makefile @@ -3,3 +3,4 @@ # obj-$(CONFIG_X86) += x86/ +obj-$(CONFIG_OLPC) += olpc/ diff --git a/drivers/platform/olpc/Makefile b/drivers/platform/olpc/Makefile new file mode 100644 index 000000000000..dc8b26bc7209 --- /dev/null +++ b/drivers/platform/olpc/Makefile @@ -0,0 +1,4 @@ +# +# OLPC XO platform-specific drivers +# +obj-$(CONFIG_OLPC) += olpc-ec.o diff --git a/drivers/platform/olpc/olpc-ec.c b/drivers/platform/olpc/olpc-ec.c new file mode 100644 index 000000000000..42026036cd3e --- /dev/null +++ b/drivers/platform/olpc/olpc-ec.c @@ -0,0 +1,16 @@ +/* + * Generic driver for the OLPC Embedded Controller. + * + * Copyright (C) 2011-2012 One Laptop per Child Foundation. + * + * Licensed under the GPL v2 or later. + */ +#include +#include + +int olpc_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf, size_t outlen) +{ + /* Currently a stub; this will be expanded upon later. */ + return olpc_ec_cmd_x86(cmd, inbuf, inlen, outbuf, outlen); +} +EXPORT_SYMBOL_GPL(olpc_ec_cmd); diff --git a/include/linux/olpc-ec.h b/include/linux/olpc-ec.h new file mode 100644 index 000000000000..6d4e426d9fdc --- /dev/null +++ b/include/linux/olpc-ec.h @@ -0,0 +1,29 @@ +#ifndef _LINUX_OLPC_EC_H +#define _LINUX_OLPC_EC_H + +/* XO-1 EC commands */ +#define EC_FIRMWARE_REV 0x08 +#define EC_WRITE_SCI_MASK 0x1b +#define EC_WAKE_UP_WLAN 0x24 +#define EC_WLAN_LEAVE_RESET 0x25 +#define EC_READ_EB_MODE 0x2a +#define EC_SET_SCI_INHIBIT 0x32 +#define EC_SET_SCI_INHIBIT_RELEASE 0x34 +#define EC_WLAN_ENTER_RESET 0x35 +#define EC_WRITE_EXT_SCI_MASK 0x38 +#define EC_SCI_QUERY 0x84 +#define EC_EXT_SCI_QUERY 0x85 + +#ifdef CONFIG_OLPC + +extern int olpc_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf, + size_t outlen); + +#else + +static inline int olpc_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf, + size_t outlen) { return -ENODEV; } + +#endif /* CONFIG_OLPC */ + +#endif /* _LINUX_OLPC_EC_H */ -- cgit v1.2.3 From 3d26c20bae9e97c98f7240184427d3a38515d406 Mon Sep 17 00:00:00 2001 From: Andres Salomon Date: Wed, 11 Jul 2012 17:40:25 -0700 Subject: Platform: OLPC: allow EC cmd to be overridden, and create a workqueue to call it This provides a new API allows different OLPC architectures to override the EC driver. x86 and ARM OLPC machines use completely different EC backends. The olpc_ec_cmd is synchronous, and waits for the workqueue to send the command to the EC. Multiple callers can run olpc_ec_cmd() at once, and they will by serialized and sleep while only one executes on the EC at a time. We don't provide an unregister function, as that doesn't make sense within the context of OLPC machines - there's only ever 1 EC, it's critical to functionality, and it certainly not hotpluggable. Signed-off-by: Andres Salomon Acked-by: Paul Fox Reviewed-by: Thomas Gleixner --- drivers/platform/olpc/olpc-ec.c | 112 +++++++++++++++++++++++++++++++++++++++- include/linux/olpc-ec.h | 6 +++ 2 files changed, 116 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/platform/olpc/olpc-ec.c b/drivers/platform/olpc/olpc-ec.c index 42026036cd3e..44e6a4fae79b 100644 --- a/drivers/platform/olpc/olpc-ec.c +++ b/drivers/platform/olpc/olpc-ec.c @@ -5,12 +5,120 @@ * * Licensed under the GPL v2 or later. */ +#include +#include +#include +#include #include +#include +#include #include +struct ec_cmd_desc { + u8 cmd; + u8 *inbuf, *outbuf; + size_t inlen, outlen; + + int err; + struct completion finished; + struct list_head node; + + void *priv; +}; + +static void olpc_ec_worker(struct work_struct *w); + +static DECLARE_WORK(ec_worker, olpc_ec_worker); +static LIST_HEAD(ec_cmd_q); +static DEFINE_SPINLOCK(ec_cmd_q_lock); + +static struct olpc_ec_driver *ec_driver; +static void *ec_cb_arg; +static DEFINE_MUTEX(ec_cb_lock); + +void olpc_ec_driver_register(struct olpc_ec_driver *drv, void *arg) +{ + ec_driver = drv; + ec_cb_arg = arg; +} +EXPORT_SYMBOL_GPL(olpc_ec_driver_register); + +static void olpc_ec_worker(struct work_struct *w) +{ + struct ec_cmd_desc *desc = NULL; + unsigned long flags; + + /* Grab the first pending command from the queue */ + spin_lock_irqsave(&ec_cmd_q_lock, flags); + if (!list_empty(&ec_cmd_q)) { + desc = list_first_entry(&ec_cmd_q, struct ec_cmd_desc, node); + list_del(&desc->node); + } + spin_unlock_irqrestore(&ec_cmd_q_lock, flags); + + /* Do we actually have anything to do? */ + if (!desc) + return; + + /* Protect the EC hw with a mutex; only run one cmd at a time */ + mutex_lock(&ec_cb_lock); + desc->err = ec_driver->ec_cmd(desc->cmd, desc->inbuf, desc->inlen, + desc->outbuf, desc->outlen, ec_cb_arg); + mutex_unlock(&ec_cb_lock); + + /* Finished, wake up olpc_ec_cmd() */ + complete(&desc->finished); + + /* Run the worker thread again in case there are more cmds pending */ + schedule_work(&ec_worker); +} + +/* + * Throw a cmd descripter onto the list. We now have SMP OLPC machines, so + * locking is pretty critical. + */ +static void queue_ec_descriptor(struct ec_cmd_desc *desc) +{ + unsigned long flags; + + INIT_LIST_HEAD(&desc->node); + + spin_lock_irqsave(&ec_cmd_q_lock, flags); + list_add_tail(&desc->node, &ec_cmd_q); + spin_unlock_irqrestore(&ec_cmd_q_lock, flags); + + schedule_work(&ec_worker); +} + int olpc_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf, size_t outlen) { - /* Currently a stub; this will be expanded upon later. */ - return olpc_ec_cmd_x86(cmd, inbuf, inlen, outbuf, outlen); + struct ec_cmd_desc desc; + + /* XXX: this will be removed in later patches */ + /* Are we using old-style callers? */ + if (!ec_driver || !ec_driver->ec_cmd) + return olpc_ec_cmd_x86(cmd, inbuf, inlen, outbuf, outlen); + + /* Ensure a driver and ec hook have been registered */ + if (WARN_ON(!ec_driver || !ec_driver->ec_cmd)) + return -ENODEV; + + might_sleep(); + + desc.cmd = cmd; + desc.inbuf = inbuf; + desc.outbuf = outbuf; + desc.inlen = inlen; + desc.outlen = outlen; + desc.err = 0; + init_completion(&desc.finished); + + queue_ec_descriptor(&desc); + + /* Timeouts must be handled in the platform-specific EC hook */ + wait_for_completion(&desc.finished); + + /* The worker thread dequeues the cmd; no need to do anything here */ + return desc.err; } EXPORT_SYMBOL_GPL(olpc_ec_cmd); diff --git a/include/linux/olpc-ec.h b/include/linux/olpc-ec.h index 6d4e426d9fdc..231e96f5dfe2 100644 --- a/include/linux/olpc-ec.h +++ b/include/linux/olpc-ec.h @@ -14,8 +14,14 @@ #define EC_SCI_QUERY 0x84 #define EC_EXT_SCI_QUERY 0x85 +struct olpc_ec_driver { + int (*ec_cmd)(u8, u8 *, size_t, u8 *, size_t, void *); +}; + #ifdef CONFIG_OLPC +extern void olpc_ec_driver_register(struct olpc_ec_driver *drv, void *arg); + extern int olpc_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf, size_t outlen); -- cgit v1.2.3 From ac2504151f5af27bbf0c0362b7da5951e05dfc43 Mon Sep 17 00:00:00 2001 From: Andres Salomon Date: Fri, 13 Jul 2012 05:57:17 -0700 Subject: Platform: OLPC: turn EC driver into a platform_driver The 1.75-based OLPC EC driver already does this; let's do it for all EC drivers. This gives us nice suspend/resume hooks, amongst other things. We want to run the EC's suspend hooks later than other drivers (which may be setting wakeup masks or be running EC commands). We also want to run the EC's resume hooks earlier than other drivers (which may want to run EC commands). Signed-off-by: Andres Salomon Acked-by: Paul Fox Reviewed-by: Thomas Gleixner --- drivers/platform/olpc/olpc-ec.c | 48 +++++++++++++++++++++++++++++++++++++++++ include/linux/olpc-ec.h | 6 ++++++ 2 files changed, 54 insertions(+) (limited to 'include/linux') diff --git a/drivers/platform/olpc/olpc-ec.c b/drivers/platform/olpc/olpc-ec.c index 44e6a4fae79b..d00523c65191 100644 --- a/drivers/platform/olpc/olpc-ec.c +++ b/drivers/platform/olpc/olpc-ec.c @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -122,3 +123,50 @@ int olpc_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf, size_t outlen) return desc.err; } EXPORT_SYMBOL_GPL(olpc_ec_cmd); + +static int olpc_ec_probe(struct platform_device *pdev) +{ + int err; + + if (!ec_driver) + return -ENODEV; + + err = ec_driver->probe ? ec_driver->probe(pdev) : 0; + + return err; +} + +static int olpc_ec_suspend(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + return ec_driver->suspend ? ec_driver->suspend(pdev) : 0; +} + +static int olpc_ec_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + return ec_driver->resume ? ec_driver->resume(pdev) : 0; +} + +static const struct dev_pm_ops olpc_ec_pm_ops = { + .suspend_late = olpc_ec_suspend, + .resume_early = olpc_ec_resume, +}; + +static struct platform_driver olpc_ec_plat_driver = { + .probe = olpc_ec_probe, + .driver = { + .name = "olpc-ec", + .pm = &olpc_ec_pm_ops, + }, +}; + +static int __init olpc_ec_init_module(void) +{ + return platform_driver_register(&olpc_ec_plat_driver); +} + +module_init(olpc_ec_init_module); + +MODULE_AUTHOR("Andres Salomon "); +MODULE_LICENSE("GPL"); diff --git a/include/linux/olpc-ec.h b/include/linux/olpc-ec.h index 231e96f5dfe2..5bb6e760aa61 100644 --- a/include/linux/olpc-ec.h +++ b/include/linux/olpc-ec.h @@ -14,7 +14,13 @@ #define EC_SCI_QUERY 0x84 #define EC_EXT_SCI_QUERY 0x85 +struct platform_device; + struct olpc_ec_driver { + int (*probe)(struct platform_device *); + int (*suspend)(struct platform_device *); + int (*resume)(struct platform_device *); + int (*ec_cmd)(u8, u8 *, size_t, u8 *, size_t, void *); }; -- cgit v1.2.3 From 4f46f8ac80416b0e8fd3aba6a0d842205fb29140 Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Mon, 30 Jul 2012 21:28:27 +0200 Subject: dmaengine: shdma: restore partial transfer calculation The recent shdma driver split has mistakenly removed support for partial DMA transfer size calculation on forced termination. This patch restores it. Signed-off-by: Guennadi Liakhovetski Acked-by: Vinod Koul Signed-off-by: Paul Mundt --- drivers/dma/sh/shdma-base.c | 9 +++++++++ drivers/dma/sh/shdma.c | 12 ++++++++++++ include/linux/shdma-base.h | 2 ++ 3 files changed, 23 insertions(+) (limited to 'include/linux') diff --git a/drivers/dma/sh/shdma-base.c b/drivers/dma/sh/shdma-base.c index 27f5c781fd73..f4cd946d259d 100644 --- a/drivers/dma/sh/shdma-base.c +++ b/drivers/dma/sh/shdma-base.c @@ -483,6 +483,7 @@ static struct shdma_desc *shdma_add_desc(struct shdma_chan *schan, new->mark = DESC_PREPARED; new->async_tx.flags = flags; new->direction = direction; + new->partial = 0; *len -= copy_size; if (direction == DMA_MEM_TO_MEM || direction == DMA_MEM_TO_DEV) @@ -644,6 +645,14 @@ static int shdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd, case DMA_TERMINATE_ALL: spin_lock_irqsave(&schan->chan_lock, flags); ops->halt_channel(schan); + + if (ops->get_partial && !list_empty(&schan->ld_queue)) { + /* Record partial transfer */ + struct shdma_desc *desc = list_first_entry(&schan->ld_queue, + struct shdma_desc, node); + desc->partial = ops->get_partial(schan, desc); + } + spin_unlock_irqrestore(&schan->chan_lock, flags); shdma_chan_ld_cleanup(schan, true); diff --git a/drivers/dma/sh/shdma.c b/drivers/dma/sh/shdma.c index 027c9be97654..f41bcc5267fd 100644 --- a/drivers/dma/sh/shdma.c +++ b/drivers/dma/sh/shdma.c @@ -381,6 +381,17 @@ static bool sh_dmae_chan_irq(struct shdma_chan *schan, int irq) return true; } +static size_t sh_dmae_get_partial(struct shdma_chan *schan, + struct shdma_desc *sdesc) +{ + struct sh_dmae_chan *sh_chan = container_of(schan, struct sh_dmae_chan, + shdma_chan); + struct sh_dmae_desc *sh_desc = container_of(sdesc, + struct sh_dmae_desc, shdma_desc); + return (sh_desc->hw.tcr - sh_dmae_readl(sh_chan, TCR)) << + sh_chan->xmit_shift; +} + /* Called from error IRQ or NMI */ static bool sh_dmae_reset(struct sh_dmae_device *shdev) { @@ -632,6 +643,7 @@ static const struct shdma_ops sh_dmae_shdma_ops = { .start_xfer = sh_dmae_start_xfer, .embedded_desc = sh_dmae_embedded_desc, .chan_irq = sh_dmae_chan_irq, + .get_partial = sh_dmae_get_partial, }; static int __devinit sh_dmae_probe(struct platform_device *pdev) diff --git a/include/linux/shdma-base.h b/include/linux/shdma-base.h index 93f9821554b6..a3728bf66f0e 100644 --- a/include/linux/shdma-base.h +++ b/include/linux/shdma-base.h @@ -50,6 +50,7 @@ struct shdma_desc { struct list_head node; struct dma_async_tx_descriptor async_tx; enum dma_transfer_direction direction; + size_t partial; dma_cookie_t cookie; int chunks; int mark; @@ -98,6 +99,7 @@ struct shdma_ops { void (*start_xfer)(struct shdma_chan *, struct shdma_desc *); struct shdma_desc *(*embedded_desc)(void *, int); bool (*chan_irq)(struct shdma_chan *, int); + size_t (*get_partial)(struct shdma_chan *, struct shdma_desc *); }; struct shdma_dev { -- cgit v1.2.3 From 095adbb6441172985f5ddc3b9e88cb3191bdeac4 Mon Sep 17 00:00:00 2001 From: Thomas Renninger Date: Tue, 31 Jul 2012 17:41:09 +0200 Subject: ACPI: Only count valid srat memory structures Otherwise you could run into: WARN_ON in numa_register_memblks(), because node_possible_map is zero References: https://bugzilla.novell.com/show_bug.cgi?id=757888 On this machine (ProLiant ML570 G3) the SRAT table contains: - No processor affinities - One memory affinity structure (which is set disabled) CC: Per Jessen CC: Andi Kleen Signed-off-by: Thomas Renninger Signed-off-by: Len Brown --- arch/ia64/kernel/acpi.c | 5 +++-- arch/x86/mm/srat.c | 15 ++++++++------- drivers/acpi/numa.c | 8 +++++--- include/linux/acpi.h | 2 +- 4 files changed, 17 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 6f38b6120d96..440578850ae5 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -497,7 +497,7 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa) srat_num_cpus++; } -void __init +int __init acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) { unsigned long paddr, size; @@ -512,7 +512,7 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) /* Ignore disabled entries */ if (!(ma->flags & ACPI_SRAT_MEM_ENABLED)) - return; + return -1; /* record this node in proximity bitmap */ pxm_bit_set(pxm); @@ -531,6 +531,7 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) p->size = size; p->nid = pxm; num_node_memblks++; + return 0; } void __init acpi_numa_arch_fixup(void) diff --git a/arch/x86/mm/srat.c b/arch/x86/mm/srat.c index 4599c3e8bcb6..4ddf497ca65b 100644 --- a/arch/x86/mm/srat.c +++ b/arch/x86/mm/srat.c @@ -142,23 +142,23 @@ static inline int save_add_info(void) {return 0;} #endif /* Callback for parsing of the Proximity Domain <-> Memory Area mappings */ -void __init +int __init acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) { u64 start, end; int node, pxm; if (srat_disabled()) - return; + return -1; if (ma->header.length != sizeof(struct acpi_srat_mem_affinity)) { bad_srat(); - return; + return -1; } if ((ma->flags & ACPI_SRAT_MEM_ENABLED) == 0) - return; + return -1; if ((ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) && !save_add_info()) - return; + return -1; start = ma->base_address; end = start + ma->length; pxm = ma->proximity_domain; @@ -168,12 +168,12 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) if (node < 0) { printk(KERN_ERR "SRAT: Too many proximity domains.\n"); bad_srat(); - return; + return -1; } if (numa_add_memblk(node, start, end) < 0) { bad_srat(); - return; + return -1; } node_set(node, numa_nodes_parsed); @@ -181,6 +181,7 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) printk(KERN_INFO "SRAT: Node %u PXM %u [mem %#010Lx-%#010Lx]\n", node, pxm, (unsigned long long) start, (unsigned long long) end - 1); + return 0; } void __init acpi_numa_arch_fixup(void) {} diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c index 2a6399345c85..cb31298ca684 100644 --- a/drivers/acpi/numa.c +++ b/drivers/acpi/numa.c @@ -237,6 +237,8 @@ acpi_parse_processor_affinity(struct acpi_subtable_header *header, return 0; } +static int __initdata parsed_numa_memblks; + static int __init acpi_parse_memory_affinity(struct acpi_subtable_header * header, const unsigned long end) @@ -250,8 +252,8 @@ acpi_parse_memory_affinity(struct acpi_subtable_header * header, acpi_table_print_srat_entry(header); /* let architecture-dependent part to do it */ - acpi_numa_memory_affinity_init(memory_affinity); - + if (!acpi_numa_memory_affinity_init(memory_affinity)) + parsed_numa_memblks++; return 0; } @@ -306,7 +308,7 @@ int __init acpi_numa_init(void) if (cnt < 0) return cnt; - else if (cnt == 0) + else if (!parsed_numa_memblks) return -ENOENT; return 0; } diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 3ad510b25283..4f2a76224509 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -96,7 +96,7 @@ void acpi_table_print_madt_entry (struct acpi_subtable_header *madt); void acpi_numa_slit_init (struct acpi_table_slit *slit); void acpi_numa_processor_affinity_init (struct acpi_srat_cpu_affinity *pa); void acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa); -void acpi_numa_memory_affinity_init (struct acpi_srat_mem_affinity *ma); +int acpi_numa_memory_affinity_init (struct acpi_srat_mem_affinity *ma); void acpi_numa_arch_fixup(void); #ifdef CONFIG_ACPI_HOTPLUG_CPU -- cgit v1.2.3 From 76582d0a68ab6c46987dd678165c26ad81c44cdb Mon Sep 17 00:00:00 2001 From: Thierry Reding Date: Wed, 25 Jul 2012 16:24:49 +0200 Subject: iommu: Include linux/types.h The linux/iommu.h header uses types defined in linux/types.h but doesn't include it. Signed-off-by: Thierry Reding Signed-off-by: Joerg Roedel --- include/linux/iommu.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/iommu.h b/include/linux/iommu.h index 54d6d690073c..0a85199191b9 100644 --- a/include/linux/iommu.h +++ b/include/linux/iommu.h @@ -20,6 +20,7 @@ #define __LINUX_IOMMU_H #include +#include #define IOMMU_READ (1) #define IOMMU_WRITE (2) -- cgit v1.2.3 From ba1eabfade3272596000bf3c62b68ca375e48b08 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Fri, 3 Aug 2012 15:55:41 +0200 Subject: iommu: Add missing forward declaration in include file The 'struct notifier_block' is not used in linux/iommu.h but not declared anywhere. Add a forward declaration for it. Reported-by: Thierry Reding Signed-off-by: Joerg Roedel --- include/linux/iommu.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/iommu.h b/include/linux/iommu.h index 0a85199191b9..7e83370e6fd2 100644 --- a/include/linux/iommu.h +++ b/include/linux/iommu.h @@ -31,6 +31,7 @@ struct iommu_group; struct bus_type; struct device; struct iommu_domain; +struct notifier_block; /* iommu fault flags */ #define IOMMU_FAULT_READ 0x0 -- cgit v1.2.3 From f0cd2dbb6cf387c11f87265462e370bb5469299e Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 25 Jul 2012 18:11:59 +0300 Subject: vfs: kill write_super and sync_supers Finally we can kill the 'sync_supers' kernel thread along with the '->write_super()' superblock operation because all the users are gone. Now every file-system is supposed to self-manage own superblock and its dirty state. The nice thing about killing this thread is that it improves power management. Indeed, 'sync_supers' is a source of monotonic system wake-ups - it woke up every 5 seconds no matter what - even if there were no dirty superblocks and even if there were no file-systems using this service (e.g., btrfs and journalled ext4 do not need it). So it was wasting power most of the time. And because the thread was in the core of the kernel, all systems had to have it. So I am quite happy to make it go away. Interestingly, this thread is a left-over from the pdflush kernel thread which was a self-forking kernel thread responsible for all the write-back in old Linux kernels. It was turned into per-block device BDI threads, and 'sync_supers' was a left-over. Thus, R.I.P, pdflush as well. Signed-off-by: Artem Bityutskiy Signed-off-by: Al Viro --- fs/super.c | 40 ---------------------------------- include/linux/backing-dev.h | 1 - include/linux/fs.h | 3 --- mm/backing-dev.c | 52 --------------------------------------------- mm/page-writeback.c | 1 - 5 files changed, 97 deletions(-) (limited to 'include/linux') diff --git a/fs/super.c b/fs/super.c index b05cf47463d0..0902cfa6a12e 100644 --- a/fs/super.c +++ b/fs/super.c @@ -536,46 +536,6 @@ void drop_super(struct super_block *sb) EXPORT_SYMBOL(drop_super); -/** - * sync_supers - helper for periodic superblock writeback - * - * Call the write_super method if present on all dirty superblocks in - * the system. This is for the periodic writeback used by most older - * filesystems. For data integrity superblock writeback use - * sync_filesystems() instead. - * - * Note: check the dirty flag before waiting, so we don't - * hold up the sync while mounting a device. (The newly - * mounted device won't need syncing.) - */ -void sync_supers(void) -{ - struct super_block *sb, *p = NULL; - - spin_lock(&sb_lock); - list_for_each_entry(sb, &super_blocks, s_list) { - if (hlist_unhashed(&sb->s_instances)) - continue; - if (sb->s_op->write_super && sb->s_dirt) { - sb->s_count++; - spin_unlock(&sb_lock); - - down_read(&sb->s_umount); - if (sb->s_root && sb->s_dirt && (sb->s_flags & MS_BORN)) - sb->s_op->write_super(sb); - up_read(&sb->s_umount); - - spin_lock(&sb_lock); - if (p) - __put_super(p); - p = sb; - } - } - if (p) - __put_super(p); - spin_unlock(&sb_lock); -} - /** * iterate_supers - call function for all active superblocks * @f: function to call diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index c97c6b9cd38e..2a9a9abc9126 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -124,7 +124,6 @@ void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages, void bdi_start_background_writeback(struct backing_dev_info *bdi); int bdi_writeback_thread(void *data); int bdi_has_dirty_io(struct backing_dev_info *bdi); -void bdi_arm_supers_timer(void); void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi); void bdi_lock_two(struct bdi_writeback *wb1, struct bdi_writeback *wb2); diff --git a/include/linux/fs.h b/include/linux/fs.h index 38dba16c4176..aa110476a95b 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1491,7 +1491,6 @@ struct sb_writers { struct super_block { struct list_head s_list; /* Keep this first */ dev_t s_dev; /* search index; _not_ kdev_t */ - unsigned char s_dirt; unsigned char s_blocksize_bits; unsigned long s_blocksize; loff_t s_maxbytes; /* Max file size */ @@ -1861,7 +1860,6 @@ struct super_operations { int (*drop_inode) (struct inode *); void (*evict_inode) (struct inode *); void (*put_super) (struct super_block *); - void (*write_super) (struct super_block *); int (*sync_fs)(struct super_block *sb, int wait); int (*freeze_fs) (struct super_block *); int (*unfreeze_fs) (struct super_block *); @@ -2397,7 +2395,6 @@ extern int vfs_fsync_range(struct file *file, loff_t start, loff_t end, int datasync); extern int vfs_fsync(struct file *file, int datasync); extern int generic_write_sync(struct file *file, loff_t pos, loff_t count); -extern void sync_supers(void); extern void emergency_sync(void); extern void emergency_remount(void); #ifdef CONFIG_BLOCK diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 6b4718e2ee34..b41823cc05e6 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -39,12 +39,6 @@ DEFINE_SPINLOCK(bdi_lock); LIST_HEAD(bdi_list); LIST_HEAD(bdi_pending_list); -static struct task_struct *sync_supers_tsk; -static struct timer_list sync_supers_timer; - -static int bdi_sync_supers(void *); -static void sync_supers_timer_fn(unsigned long); - void bdi_lock_two(struct bdi_writeback *wb1, struct bdi_writeback *wb2) { if (wb1 < wb2) { @@ -250,12 +244,6 @@ static int __init default_bdi_init(void) { int err; - sync_supers_tsk = kthread_run(bdi_sync_supers, NULL, "sync_supers"); - BUG_ON(IS_ERR(sync_supers_tsk)); - - setup_timer(&sync_supers_timer, sync_supers_timer_fn, 0); - bdi_arm_supers_timer(); - err = bdi_init(&default_backing_dev_info); if (!err) bdi_register(&default_backing_dev_info, NULL, "default"); @@ -270,46 +258,6 @@ int bdi_has_dirty_io(struct backing_dev_info *bdi) return wb_has_dirty_io(&bdi->wb); } -/* - * kupdated() used to do this. We cannot do it from the bdi_forker_thread() - * or we risk deadlocking on ->s_umount. The longer term solution would be - * to implement sync_supers_bdi() or similar and simply do it from the - * bdi writeback thread individually. - */ -static int bdi_sync_supers(void *unused) -{ - set_user_nice(current, 0); - - while (!kthread_should_stop()) { - set_current_state(TASK_INTERRUPTIBLE); - schedule(); - - /* - * Do this periodically, like kupdated() did before. - */ - sync_supers(); - } - - return 0; -} - -void bdi_arm_supers_timer(void) -{ - unsigned long next; - - if (!dirty_writeback_interval) - return; - - next = msecs_to_jiffies(dirty_writeback_interval * 10) + jiffies; - mod_timer(&sync_supers_timer, round_jiffies_up(next)); -} - -static void sync_supers_timer_fn(unsigned long unused) -{ - wake_up_process(sync_supers_tsk); - bdi_arm_supers_timer(); -} - static void wakeup_timer_fn(unsigned long data) { struct backing_dev_info *bdi = (struct backing_dev_info *)data; diff --git a/mm/page-writeback.c b/mm/page-writeback.c index e5363f34e025..5ad5ce23c1e0 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -1532,7 +1532,6 @@ int dirty_writeback_centisecs_handler(ctl_table *table, int write, void __user *buffer, size_t *length, loff_t *ppos) { proc_dointvec(table, write, buffer, length, ppos); - bdi_arm_supers_timer(); return 0; } -- cgit v1.2.3 From 0d5c3eba2e1e5aa74e097f49bc90b58f607e101c Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 25 Jul 2012 18:12:08 +0300 Subject: vfs: nuke pdflush from comments The pdflush thread is long gone, so this patch removes references to pdflush from vfs comments. Signed-off-by: Artem Bityutskiy Signed-off-by: Al Viro --- fs/bio.c | 2 +- include/linux/writeback.h | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/fs/bio.c b/fs/bio.c index 73922abba832..5eaa70c9d96e 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -1312,7 +1312,7 @@ EXPORT_SYMBOL(bio_copy_kern); * Note that this code is very hard to test under normal circumstances because * direct-io pins the pages with get_user_pages(). This makes * is_page_cache_freeable return false, and the VM will not clean the pages. - * But other code (eg, pdflush) could clean the pages if they are mapped + * But other code (eg, flusher threads) could clean the pages if they are mapped * pagecache. * * Simply disabling the call to bio_set_pages_dirty() is a good way to test the diff --git a/include/linux/writeback.h b/include/linux/writeback.h index c66fe3332d83..50c3e8fa06a8 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -104,7 +104,6 @@ static inline void wait_on_inode(struct inode *inode) wait_on_bit(&inode->i_state, __I_NEW, inode_wait, TASK_UNINTERRUPTIBLE); } - /* * mm/page-writeback.c */ -- cgit v1.2.3 From d796c52ef0b71a988364f6109aeb63d79c5b116b Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sun, 5 Aug 2012 19:04:57 -0400 Subject: ext4: make sure the journal sb is written in ext4_clear_journal_err() After we transfer set the EXT4_ERROR_FS bit in the file system superblock, it's not enough to call jbd2_journal_clear_err() to clear the error indication from journal superblock --- we need to call jbd2_journal_update_sb_errno() as well. Otherwise, when the root file system is mounted read-only, the journal is replayed, and the error indicator is transferred to the superblock --- but the s_errno field in the jbd2 superblock is left set (since although we cleared it in memory, we never flushed it out to disk). This can end up confusing e2fsck. We should make e2fsck more robust in this case, but the kernel shouldn't be leaving things in this confused state, either. Signed-off-by: "Theodore Ts'o" Cc: stable@kernel.org --- fs/ext4/super.c | 1 + fs/jbd2/journal.c | 3 ++- include/linux/jbd2.h | 1 + 3 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index d76ec8277d3f..ccc4bcad5616 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4430,6 +4430,7 @@ static void ext4_clear_journal_err(struct super_block *sb, ext4_commit_super(sb, 1); jbd2_journal_clear_err(journal); + jbd2_journal_update_sb_errno(journal); } } diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index e9a3c4c85594..bd23f2ebaa67 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1377,7 +1377,7 @@ static void jbd2_mark_journal_empty(journal_t *journal) * Update a journal's errno. Write updated superblock to disk waiting for IO * to complete. */ -static void jbd2_journal_update_sb_errno(journal_t *journal) +void jbd2_journal_update_sb_errno(journal_t *journal) { journal_superblock_t *sb = journal->j_superblock; @@ -1390,6 +1390,7 @@ static void jbd2_journal_update_sb_errno(journal_t *journal) jbd2_write_superblock(journal, WRITE_SYNC); } +EXPORT_SYMBOL(jbd2_journal_update_sb_errno); /* * Read the superblock for a given journal, performing initial diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index f334c7fab967..3efc43f3f162 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1125,6 +1125,7 @@ extern int jbd2_journal_destroy (journal_t *); extern int jbd2_journal_recover (journal_t *journal); extern int jbd2_journal_wipe (journal_t *, int); extern int jbd2_journal_skip_recovery (journal_t *); +extern void jbd2_journal_update_sb_errno(journal_t *); extern void jbd2_journal_update_sb_log_tail (journal_t *, tid_t, unsigned long, int); extern void __jbd2_journal_abort_hard (journal_t *); -- cgit v1.2.3 From 5d299f3d3c8a2fbc732b1bf03af36333ccec3130 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 Aug 2012 05:09:33 +0000 Subject: net: ipv6: fix TCP early demux IPv6 needs a cookie in dst_check() call. We need to add rx_dst_cookie and provide a family independent sk_rx_dst_set(sk, skb) method to properly support IPv6 TCP early demux. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/ipv6.h | 1 + include/net/inet_connection_sock.h | 1 + include/net/inet_sock.h | 9 --------- net/ipv4/tcp_input.c | 4 +++- net/ipv4/tcp_ipv4.c | 13 ++++++++++--- net/ipv4/tcp_minisocks.c | 2 +- net/ipv6/tcp_ipv6.c | 27 +++++++++++++++++++++++++-- 7 files changed, 41 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 379e433e15e0..879db26ec401 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -369,6 +369,7 @@ struct ipv6_pinfo { __u8 rcv_tclass; __u32 dst_cookie; + __u32 rx_dst_cookie; struct ipv6_mc_socklist __rcu *ipv6_mc_list; struct ipv6_ac_socklist *ipv6_ac_list; diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 5ee66f517b4f..ba1d3615acbb 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -39,6 +39,7 @@ struct inet_connection_sock_af_ops { int (*queue_xmit)(struct sk_buff *skb, struct flowi *fl); void (*send_check)(struct sock *sk, struct sk_buff *skb); int (*rebuild_header)(struct sock *sk); + void (*sk_rx_dst_set)(struct sock *sk, const struct sk_buff *skb); int (*conn_request)(struct sock *sk, struct sk_buff *skb); struct sock *(*syn_recv_sock)(struct sock *sk, struct sk_buff *skb, struct request_sock *req, diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 83b567fe1941..613cfa401672 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -249,13 +249,4 @@ static inline __u8 inet_sk_flowi_flags(const struct sock *sk) return flags; } -static inline void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) -{ - struct dst_entry *dst = skb_dst(skb); - - dst_hold(dst); - sk->sk_rx_dst = dst; - inet_sk(sk)->rx_dst_ifindex = skb->skb_iif; -} - #endif /* _INET_SOCK_H */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2fd2bc9e3c64..85308b90df80 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5392,6 +5392,8 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, { struct tcp_sock *tp = tcp_sk(sk); + if (unlikely(sk->sk_rx_dst == NULL)) + inet_csk(sk)->icsk_af_ops->sk_rx_dst_set(sk, skb); /* * Header prediction. * The code loosely follows the one in the famous @@ -5605,7 +5607,7 @@ void tcp_finish_connect(struct sock *sk, struct sk_buff *skb) tcp_set_state(sk, TCP_ESTABLISHED); if (skb != NULL) { - inet_sk_rx_dst_set(sk, skb); + icsk->icsk_af_ops->sk_rx_dst_set(sk, skb); security_inet_conn_established(sk, skb); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 42b2a6a73092..272241f16fcb 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1627,9 +1627,6 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) sk->sk_rx_dst = NULL; } } - if (unlikely(sk->sk_rx_dst == NULL)) - inet_sk_rx_dst_set(sk, skb); - if (tcp_rcv_established(sk, skb, tcp_hdr(skb), skb->len)) { rsk = sk; goto reset; @@ -1872,10 +1869,20 @@ static struct timewait_sock_ops tcp_timewait_sock_ops = { .twsk_destructor= tcp_twsk_destructor, }; +static void inet_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) +{ + struct dst_entry *dst = skb_dst(skb); + + dst_hold(dst); + sk->sk_rx_dst = dst; + inet_sk(sk)->rx_dst_ifindex = skb->skb_iif; +} + const struct inet_connection_sock_af_ops ipv4_specific = { .queue_xmit = ip_queue_xmit, .send_check = tcp_v4_send_check, .rebuild_header = inet_sk_rebuild_header, + .sk_rx_dst_set = inet_sk_rx_dst_set, .conn_request = tcp_v4_conn_request, .syn_recv_sock = tcp_v4_syn_recv_sock, .net_header_len = sizeof(struct iphdr), diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 232a90c3ec86..d9c9dcef2de3 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -387,7 +387,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, struct tcp_sock *oldtp = tcp_sk(sk); struct tcp_cookie_values *oldcvp = oldtp->cookie_values; - inet_sk_rx_dst_set(newsk, skb); + newicsk->icsk_af_ops->sk_rx_dst_set(newsk, skb); /* TCP Cookie Transactions require space for the cookie pair, * as it differs for each connection. There is no need to diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index c66b90f71c9b..5a439e9a4c01 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1447,7 +1447,17 @@ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) opt_skb = skb_clone(skb, sk_gfp_atomic(sk, GFP_ATOMIC)); if (sk->sk_state == TCP_ESTABLISHED) { /* Fast path */ + struct dst_entry *dst = sk->sk_rx_dst; + sock_rps_save_rxhash(sk, skb); + if (dst) { + if (inet_sk(sk)->rx_dst_ifindex != skb->skb_iif || + dst->ops->check(dst, np->rx_dst_cookie) == NULL) { + dst_release(dst); + sk->sk_rx_dst = NULL; + } + } + if (tcp_rcv_established(sk, skb, tcp_hdr(skb), skb->len)) goto reset; if (opt_skb) @@ -1705,9 +1715,9 @@ static void tcp_v6_early_demux(struct sk_buff *skb) struct dst_entry *dst = sk->sk_rx_dst; struct inet_sock *icsk = inet_sk(sk); if (dst) - dst = dst_check(dst, 0); + dst = dst_check(dst, inet6_sk(sk)->rx_dst_cookie); if (dst && - icsk->rx_dst_ifindex == inet6_iif(skb)) + icsk->rx_dst_ifindex == skb->skb_iif) skb_dst_set_noref(skb, dst); } } @@ -1719,10 +1729,23 @@ static struct timewait_sock_ops tcp6_timewait_sock_ops = { .twsk_destructor= tcp_twsk_destructor, }; +static void inet6_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) +{ + struct dst_entry *dst = skb_dst(skb); + const struct rt6_info *rt = (const struct rt6_info *)dst; + + dst_hold(dst); + sk->sk_rx_dst = dst; + inet_sk(sk)->rx_dst_ifindex = skb->skb_iif; + if (rt->rt6i_node) + inet6_sk(sk)->rx_dst_cookie = rt->rt6i_node->fn_sernum; +} + static const struct inet_connection_sock_af_ops ipv6_specific = { .queue_xmit = inet6_csk_xmit, .send_check = tcp_v6_send_check, .rebuild_header = inet6_sk_rebuild_header, + .sk_rx_dst_set = inet6_sk_rx_dst_set, .conn_request = tcp_v6_conn_request, .syn_recv_sock = tcp_v6_syn_recv_sock, .net_header_len = sizeof(struct ipv6hdr), -- cgit v1.2.3 From 035534ed3377d9def2c17717899fd64a111a785b Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Mon, 6 Aug 2012 18:02:29 +0200 Subject: canfd: remove redundant CAN FD flag The first idea of the CAN FD implementation started with a new struct canfd_frame to be used for both CAN FD frames and legacy CAN frames. The now mainlined implementation supports both CAN frame types simultaneously and distinguishes them only by their required sizes: CAN_MTU and CANFD_MTU. Only the struct canfd_frame contains a flags element which is needed for the additional CAN FD information. As CAN FD implicitly means that the 'Extened Data Length' mode is enabled the formerly defined CANFD_EDL bit became redundant and also confusing as an unset bit would be an error and would always need to be tested. This patch removes the obsolete CANFD_EDL bit and clarifies the documentation for the use of struct canfd_frame and the CAN FD relevant flags. Signed-off-by: Oliver Hartkopp Signed-off-by: Marc Kleine-Budde --- include/linux/can.h | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/can.h b/include/linux/can.h index 018055efc034..e52958d7c2d1 100644 --- a/include/linux/can.h +++ b/include/linux/can.h @@ -74,20 +74,21 @@ struct can_frame { /* * defined bits for canfd_frame.flags * - * As the default for CAN FD should be to support the high data rate in the - * payload section of the frame (HDR) and to support up to 64 byte in the - * data section (EDL) the bits are only set in the non-default case. - * Btw. as long as there's no real implementation for CAN FD network driver - * these bits are only preliminary. + * The use of struct canfd_frame implies the Extended Data Length (EDL) bit to + * be set in the CAN frame bitstream on the wire. The EDL bit switch turns + * the CAN controllers bitstream processor into the CAN FD mode which creates + * two new options within the CAN FD frame specification: * - * RX: NOHDR/NOEDL - info about received CAN FD frame - * ESI - bit from originating CAN controller - * TX: NOHDR/NOEDL - control per-frame settings if supported by CAN controller - * ESI - bit is set by local CAN controller + * Bit Rate Switch - to indicate a second bitrate is/was used for the payload + * Error State Indicator - represents the error state of the transmitting node + * + * As the CANFD_ESI bit is internally generated by the transmitting CAN + * controller only the CANFD_BRS bit is relevant for real CAN controllers when + * building a CAN FD frame for transmission. Setting the CANFD_ESI bit can make + * sense for virtual CAN interfaces to test applications with echoed frames. */ -#define CANFD_NOHDR 0x01 /* frame without high data rate */ -#define CANFD_NOEDL 0x02 /* frame without extended data length */ -#define CANFD_ESI 0x04 /* error state indicator */ +#define CANFD_BRS 0x01 /* bit rate switch (second bitrate for payload data) */ +#define CANFD_ESI 0x02 /* error state indicator of the transmitting node */ /** * struct canfd_frame - CAN flexible data rate frame structure -- cgit v1.2.3 From 817fea2df3c24b22f6123dc0106eb063b7132883 Mon Sep 17 00:00:00 2001 From: Alex Williamson Date: Tue, 7 Aug 2012 11:48:33 -0600 Subject: vfio: Include vfio.h in installed headers Signed-off-by: Alex Williamson --- include/linux/Kbuild | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index d9a754474878..fa217607c582 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -391,6 +391,7 @@ header-y += v4l2-dv-timings.h header-y += v4l2-mediabus.h header-y += v4l2-subdev.h header-y += veth.h +header-y += vfio.h header-y += vhost.h header-y += videodev2.h header-y += virtio_9p.h -- cgit v1.2.3 From 300d3739e873d50d4c6e3656f89007a217fb1d29 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 7 Aug 2012 13:50:22 +0200 Subject: Revert "NMI watchdog: fix for lockup detector breakage on resume" Revert commit 45226e9 (NMI watchdog: fix for lockup detector breakage on resume) which breaks resume from system suspend on my SH7372 Mackerel board (by causing a NULL pointer dereference to happen) and is generally wrong, because it abuses the CPU hotplug functionality in a shamelessly blatant way. The original issue should be addressed through appropriate syscore resume callback instead. Signed-off-by: Rafael J. Wysocki --- include/linux/sched.h | 8 -------- kernel/power/suspend.c | 3 --- kernel/watchdog.c | 21 ++------------------- 3 files changed, 2 insertions(+), 30 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index c147e7024f11..b8c86648a2f9 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -334,14 +334,6 @@ static inline void lockup_detector_init(void) } #endif -#if defined(CONFIG_LOCKUP_DETECTOR) && defined(CONFIG_SUSPEND) -void lockup_detector_bootcpu_resume(void); -#else -static inline void lockup_detector_bootcpu_resume(void) -{ -} -#endif - #ifdef CONFIG_DETECT_HUNG_TASK extern unsigned int sysctl_hung_task_panic; extern unsigned long sysctl_hung_task_check_count; diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 1da39ea248fd..c8b7446b27df 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -178,9 +178,6 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) arch_suspend_enable_irqs(); BUG_ON(irqs_disabled()); - /* Kick the lockup detector */ - lockup_detector_bootcpu_resume(); - Enable_cpus: enable_nonboot_cpus(); diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 69add8a9da68..4b1dfba70f7c 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -575,7 +575,7 @@ out: /* * Create/destroy watchdog threads as CPUs come and go: */ -static int +static int __cpuinit cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { int hotcpu = (unsigned long)hcpu; @@ -610,27 +610,10 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_OK; } -static struct notifier_block cpu_nfb = { +static struct notifier_block __cpuinitdata cpu_nfb = { .notifier_call = cpu_callback }; -#ifdef CONFIG_SUSPEND -/* - * On exit from suspend we force an offline->online transition on the boot CPU - * so that the PMU state that was lost while in suspended state gets set up - * properly for the boot CPU. This information is required for restarting the - * NMI watchdog. - */ -void lockup_detector_bootcpu_resume(void) -{ - void *cpu = (void *)(long)smp_processor_id(); - - cpu_callback(&cpu_nfb, CPU_DEAD_FROZEN, cpu); - cpu_callback(&cpu_nfb, CPU_UP_PREPARE_FROZEN, cpu); - cpu_callback(&cpu_nfb, CPU_ONLINE_FROZEN, cpu); -} -#endif - void __init lockup_detector_init(void) { void *cpu = (void *)(long)smp_processor_id(); -- cgit v1.2.3 From 59ee93a528b94ef4e81a08db252b0326feff171f Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sun, 5 Aug 2012 14:58:37 +0000 Subject: ARM: pxa: remove irq_to_gpio from ezx-pcap driver The irq_to_gpio function was removed from the pxa platform in linux-3.2, and this driver has been broken since. There is actually no in-tree user of this driver that adds this platform device, but the driver can and does get enabled on some platforms. Without this patch, building ezx_defconfig results in: drivers/mfd/ezx-pcap.c: In function 'pcap_isr_work': drivers/mfd/ezx-pcap.c:205:2: error: implicit declaration of function 'irq_to_gpio' [-Werror=implicit-function-declaration] Signed-off-by: Arnd Bergmann Acked-by: Haojian Zhuang Cc: stable@vger.kernel.org (v3.2+) Cc: Samuel Ortiz Cc: Daniel Ribeiro --- drivers/mfd/ezx-pcap.c | 2 +- include/linux/mfd/ezx-pcap.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/mfd/ezx-pcap.c b/drivers/mfd/ezx-pcap.c index 43a76c41cfcc..db662e2dcfa5 100644 --- a/drivers/mfd/ezx-pcap.c +++ b/drivers/mfd/ezx-pcap.c @@ -202,7 +202,7 @@ static void pcap_isr_work(struct work_struct *work) } local_irq_enable(); ezx_pcap_write(pcap, PCAP_REG_MSR, pcap->msr); - } while (gpio_get_value(irq_to_gpio(pcap->spi->irq))); + } while (gpio_get_value(pdata->gpio)); } static void pcap_irq_handler(unsigned int irq, struct irq_desc *desc) diff --git a/include/linux/mfd/ezx-pcap.h b/include/linux/mfd/ezx-pcap.h index 40c372165f3e..32a1b5cfeba1 100644 --- a/include/linux/mfd/ezx-pcap.h +++ b/include/linux/mfd/ezx-pcap.h @@ -16,6 +16,7 @@ struct pcap_subdev { struct pcap_platform_data { unsigned int irq_base; unsigned int config; + int gpio; void (*init) (void *); /* board specific init */ int num_subdevs; struct pcap_subdev *subdevs; -- cgit v1.2.3 From 4eef6cbfcc03b294d9d334368a851b35b496ce53 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 30 Apr 2012 16:21:37 +0000 Subject: Input: eeti_ts: pass gpio value instead of IRQ The EETI touchscreen asserts its IRQ line as soon as it has data in its internal buffers. The line is automatically deasserted once all data has been read via I2C. Hence, the driver has to monitor the GPIO line and cannot simply rely on the interrupt handler reception. In the current implementation of the driver, irq_to_gpio() is used to determine the GPIO number from the i2c_client's IRQ value. As irq_to_gpio() is not available on all platforms, this patch changes this and makes the driver ignore the passed in IRQ. Instead, a GPIO is added to the platform_data struct and gpio_to_irq is used to derive the IRQ from that GPIO. If this fails, bail out. The driver is only able to work in environments where the touchscreen GPIO can be mapped to an IRQ. Without this patch, building raumfeld_defconfig results in: drivers/input/touchscreen/eeti_ts.c: In function 'eeti_ts_irq_active': drivers/input/touchscreen/eeti_ts.c:65:2: error: implicit declaration of function 'irq_to_gpio' [-Werror=implicit-function-declaration] Signed-off-by: Daniel Mack Signed-off-by: Arnd Bergmann Cc: stable@vger.kernel.org (v3.2+) Cc: Dmitry Torokhov Cc: Sven Neumann Cc: linux-input@vger.kernel.org Cc: Haojian Zhuang --- arch/arm/mach-pxa/raumfeld.c | 2 +- drivers/input/touchscreen/eeti_ts.c | 21 +++++++++++++-------- include/linux/input/eeti_ts.h | 1 + 3 files changed, 15 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/mach-pxa/raumfeld.c b/arch/arm/mach-pxa/raumfeld.c index 5905ed130e94..d89d87ae144c 100644 --- a/arch/arm/mach-pxa/raumfeld.c +++ b/arch/arm/mach-pxa/raumfeld.c @@ -953,12 +953,12 @@ static struct i2c_board_info raumfeld_connector_i2c_board_info __initdata = { static struct eeti_ts_platform_data eeti_ts_pdata = { .irq_active_high = 1, + .irq_gpio = GPIO_TOUCH_IRQ, }; static struct i2c_board_info raumfeld_controller_i2c_board_info __initdata = { .type = "eeti_ts", .addr = 0x0a, - .irq = PXA_GPIO_TO_IRQ(GPIO_TOUCH_IRQ), .platform_data = &eeti_ts_pdata, }; diff --git a/drivers/input/touchscreen/eeti_ts.c b/drivers/input/touchscreen/eeti_ts.c index 503c7096ed36..908407efc672 100644 --- a/drivers/input/touchscreen/eeti_ts.c +++ b/drivers/input/touchscreen/eeti_ts.c @@ -48,7 +48,7 @@ struct eeti_ts_priv { struct input_dev *input; struct work_struct work; struct mutex mutex; - int irq, irq_active_high; + int irq_gpio, irq, irq_active_high; }; #define EETI_TS_BITDEPTH (11) @@ -62,7 +62,7 @@ struct eeti_ts_priv { static inline int eeti_ts_irq_active(struct eeti_ts_priv *priv) { - return gpio_get_value(irq_to_gpio(priv->irq)) == priv->irq_active_high; + return gpio_get_value(priv->irq_gpio) == priv->irq_active_high; } static void eeti_ts_read(struct work_struct *work) @@ -157,7 +157,7 @@ static void eeti_ts_close(struct input_dev *dev) static int __devinit eeti_ts_probe(struct i2c_client *client, const struct i2c_device_id *idp) { - struct eeti_ts_platform_data *pdata; + struct eeti_ts_platform_data *pdata = client->dev.platform_data; struct eeti_ts_priv *priv; struct input_dev *input; unsigned int irq_flags; @@ -199,9 +199,12 @@ static int __devinit eeti_ts_probe(struct i2c_client *client, priv->client = client; priv->input = input; - priv->irq = client->irq; + priv->irq_gpio = pdata->irq_gpio; + priv->irq = gpio_to_irq(pdata->irq_gpio); - pdata = client->dev.platform_data; + err = gpio_request_one(pdata->irq_gpio, GPIOF_IN, client->name); + if (err < 0) + goto err1; if (pdata) priv->irq_active_high = pdata->irq_active_high; @@ -215,13 +218,13 @@ static int __devinit eeti_ts_probe(struct i2c_client *client, err = input_register_device(input); if (err) - goto err1; + goto err2; err = request_irq(priv->irq, eeti_ts_isr, irq_flags, client->name, priv); if (err) { dev_err(&client->dev, "Unable to request touchscreen IRQ.\n"); - goto err2; + goto err3; } /* @@ -233,9 +236,11 @@ static int __devinit eeti_ts_probe(struct i2c_client *client, device_init_wakeup(&client->dev, 0); return 0; -err2: +err3: input_unregister_device(input); input = NULL; /* so we dont try to free it below */ +err2: + gpio_free(pdata->irq_gpio); err1: input_free_device(input); kfree(priv); diff --git a/include/linux/input/eeti_ts.h b/include/linux/input/eeti_ts.h index f875b316249d..16625d799b6f 100644 --- a/include/linux/input/eeti_ts.h +++ b/include/linux/input/eeti_ts.h @@ -2,6 +2,7 @@ #define LINUX_INPUT_EETI_TS_H struct eeti_ts_platform_data { + int irq_gpio; unsigned int irq_active_high; }; -- cgit v1.2.3 From 02b69cbdc2fb2e1bfbfd9ac0c246d7be1b08d3cd Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 9 Aug 2012 10:08:46 +0000 Subject: netfilter: nf_ct_sip: fix IPv6 address parsing Within SIP messages IPv6 addresses are enclosed in square brackets in most cases, with the exception of the "received=" header parameter. Currently the helper fails to parse enclosed addresses. This patch: - changes the SIP address parsing function to enforce square brackets when required, and accept them when not required but present, as recommended by RFC 5118. - adds a new SDP address parsing function that never accepts square brackets since SDP doesn't use them. With these changes, the SIP helper correctly parses all test messages from RFC 5118 (Session Initiation Protocol (SIP) Torture Test Messages for Internet Protocol Version 6 (IPv6)). Signed-off-by: Patrick McHardy Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/nf_conntrack_sip.h | 2 +- net/ipv4/netfilter/nf_nat_sip.c | 4 +- net/netfilter/nf_conntrack_sip.c | 87 ++++++++++++++++++++++++------ 3 files changed, 73 insertions(+), 20 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 0dfc8b7210a3..89f2a627f3f0 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -164,7 +164,7 @@ extern int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr unsigned int dataoff, unsigned int datalen, const char *name, unsigned int *matchoff, unsigned int *matchlen, - union nf_inet_addr *addr); + union nf_inet_addr *addr, bool delim); extern int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr, unsigned int off, unsigned int datalen, const char *name, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index ea4a23813d26..eef8f29e8bf8 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -173,7 +173,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, unsigned int dataoff, * the reply. */ if (ct_sip_parse_address_param(ct, *dptr, matchend, *datalen, "maddr=", &poff, &plen, - &addr) > 0 && + &addr, true) > 0 && addr.ip == ct->tuplehash[dir].tuple.src.u3.ip && addr.ip != ct->tuplehash[!dir].tuple.dst.u3.ip) { buflen = sprintf(buffer, "%pI4", @@ -187,7 +187,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, unsigned int dataoff, * from which the server received the request. */ if (ct_sip_parse_address_param(ct, *dptr, matchend, *datalen, "received=", &poff, &plen, - &addr) > 0 && + &addr, false) > 0 && addr.ip == ct->tuplehash[dir].tuple.dst.u3.ip && addr.ip != ct->tuplehash[!dir].tuple.src.u3.ip) { buflen = sprintf(buffer, "%pI4", diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 2fb666920cc9..5c0a112aeee6 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -183,12 +183,12 @@ static int media_len(const struct nf_conn *ct, const char *dptr, return len + digits_len(ct, dptr, limit, shift); } -static int parse_addr(const struct nf_conn *ct, const char *cp, - const char **endp, union nf_inet_addr *addr, - const char *limit) +static int sip_parse_addr(const struct nf_conn *ct, const char *cp, + const char **endp, union nf_inet_addr *addr, + const char *limit, bool delim) { const char *end; - int ret = 0; + int ret; if (!ct) return 0; @@ -197,16 +197,28 @@ static int parse_addr(const struct nf_conn *ct, const char *cp, switch (nf_ct_l3num(ct)) { case AF_INET: ret = in4_pton(cp, limit - cp, (u8 *)&addr->ip, -1, &end); + if (ret == 0) + return 0; break; case AF_INET6: + if (cp < limit && *cp == '[') + cp++; + else if (delim) + return 0; + ret = in6_pton(cp, limit - cp, (u8 *)&addr->ip6, -1, &end); + if (ret == 0) + return 0; + + if (end < limit && *end == ']') + end++; + else if (delim) + return 0; break; default: BUG(); } - if (ret == 0 || end == cp) - return 0; if (endp) *endp = end; return 1; @@ -219,7 +231,7 @@ static int epaddr_len(const struct nf_conn *ct, const char *dptr, union nf_inet_addr addr; const char *aux = dptr; - if (!parse_addr(ct, dptr, &dptr, &addr, limit)) { + if (!sip_parse_addr(ct, dptr, &dptr, &addr, limit, true)) { pr_debug("ip: %s parse failed.!\n", dptr); return 0; } @@ -296,7 +308,7 @@ int ct_sip_parse_request(const struct nf_conn *ct, return 0; dptr += shift; - if (!parse_addr(ct, dptr, &end, addr, limit)) + if (!sip_parse_addr(ct, dptr, &end, addr, limit, true)) return -1; if (end < limit && *end == ':') { end++; @@ -550,7 +562,7 @@ int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, if (ret == 0) return ret; - if (!parse_addr(ct, dptr + *matchoff, &c, addr, limit)) + if (!sip_parse_addr(ct, dptr + *matchoff, &c, addr, limit, true)) return -1; if (*c == ':') { c++; @@ -599,7 +611,7 @@ int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, const char *name, unsigned int *matchoff, unsigned int *matchlen, - union nf_inet_addr *addr) + union nf_inet_addr *addr, bool delim) { const char *limit = dptr + datalen; const char *start, *end; @@ -613,7 +625,7 @@ int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr, return 0; start += strlen(name); - if (!parse_addr(ct, start, &end, addr, limit)) + if (!sip_parse_addr(ct, start, &end, addr, limit, delim)) return 0; *matchoff = start - dptr; *matchlen = end - start; @@ -675,6 +687,47 @@ static int ct_sip_parse_transport(struct nf_conn *ct, const char *dptr, return 1; } +static int sdp_parse_addr(const struct nf_conn *ct, const char *cp, + const char **endp, union nf_inet_addr *addr, + const char *limit) +{ + const char *end; + int ret; + + memset(addr, 0, sizeof(*addr)); + switch (nf_ct_l3num(ct)) { + case AF_INET: + ret = in4_pton(cp, limit - cp, (u8 *)&addr->ip, -1, &end); + break; + case AF_INET6: + ret = in6_pton(cp, limit - cp, (u8 *)&addr->ip6, -1, &end); + break; + default: + BUG(); + } + + if (ret == 0) + return 0; + if (endp) + *endp = end; + return 1; +} + +/* skip ip address. returns its length. */ +static int sdp_addr_len(const struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + union nf_inet_addr addr; + const char *aux = dptr; + + if (!sdp_parse_addr(ct, dptr, &dptr, &addr, limit)) { + pr_debug("ip: %s parse failed.!\n", dptr); + return 0; + } + + return dptr - aux; +} + /* SDP header parsing: a SDP session description contains an ordered set of * headers, starting with a section containing general session parameters, * optionally followed by multiple media descriptions. @@ -686,10 +739,10 @@ static int ct_sip_parse_transport(struct nf_conn *ct, const char *dptr, */ static const struct sip_header ct_sdp_hdrs[] = { [SDP_HDR_VERSION] = SDP_HDR("v=", NULL, digits_len), - [SDP_HDR_OWNER_IP4] = SDP_HDR("o=", "IN IP4 ", epaddr_len), - [SDP_HDR_CONNECTION_IP4] = SDP_HDR("c=", "IN IP4 ", epaddr_len), - [SDP_HDR_OWNER_IP6] = SDP_HDR("o=", "IN IP6 ", epaddr_len), - [SDP_HDR_CONNECTION_IP6] = SDP_HDR("c=", "IN IP6 ", epaddr_len), + [SDP_HDR_OWNER_IP4] = SDP_HDR("o=", "IN IP4 ", sdp_addr_len), + [SDP_HDR_CONNECTION_IP4] = SDP_HDR("c=", "IN IP4 ", sdp_addr_len), + [SDP_HDR_OWNER_IP6] = SDP_HDR("o=", "IN IP6 ", sdp_addr_len), + [SDP_HDR_CONNECTION_IP6] = SDP_HDR("c=", "IN IP6 ", sdp_addr_len), [SDP_HDR_MEDIA] = SDP_HDR("m=", NULL, media_len), }; @@ -775,8 +828,8 @@ static int ct_sip_parse_sdp_addr(const struct nf_conn *ct, const char *dptr, if (ret <= 0) return ret; - if (!parse_addr(ct, dptr + *matchoff, NULL, addr, - dptr + *matchoff + *matchlen)) + if (!sdp_parse_addr(ct, dptr + *matchoff, NULL, addr, + dptr + *matchoff + *matchlen)) return -1; return 1; } -- cgit v1.2.3 From 9d8dad742ad1c74d7e7210ee05d0b44961d5ea16 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Thu, 9 Aug 2012 19:01:26 -0700 Subject: Yama: higher restrictions should block PTRACE_TRACEME The higher ptrace restriction levels should be blocking even PTRACE_TRACEME requests. The comments in the LSM documentation are misleading about when the checks happen (the parent does not go through security_ptrace_access_check() on a PTRACE_TRACEME call). Signed-off-by: Kees Cook Cc: stable@vger.kernel.org # 3.5.x and later Signed-off-by: James Morris --- Documentation/security/Yama.txt | 14 +++++++------- include/linux/security.h | 2 -- security/yama/yama_lsm.c | 41 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 48 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/Documentation/security/Yama.txt b/Documentation/security/Yama.txt index e369de2d48cd..dd908cf64ecf 100644 --- a/Documentation/security/Yama.txt +++ b/Documentation/security/Yama.txt @@ -46,14 +46,13 @@ restrictions, it can call prctl(PR_SET_PTRACER, PR_SET_PTRACER_ANY, ...) so that any otherwise allowed process (even those in external pid namespaces) may attach. -These restrictions do not change how ptrace via PTRACE_TRACEME operates. - -The sysctl settings are: +The sysctl settings (writable only with CAP_SYS_PTRACE) are: 0 - classic ptrace permissions: a process can PTRACE_ATTACH to any other process running under the same uid, as long as it is dumpable (i.e. did not transition uids, start privileged, or have called - prctl(PR_SET_DUMPABLE...) already). + prctl(PR_SET_DUMPABLE...) already). Similarly, PTRACE_TRACEME is + unchanged. 1 - restricted ptrace: a process must have a predefined relationship with the inferior it wants to call PTRACE_ATTACH on. By default, @@ -61,12 +60,13 @@ The sysctl settings are: classic criteria is also met. To change the relationship, an inferior can call prctl(PR_SET_PTRACER, debugger, ...) to declare an allowed debugger PID to call PTRACE_ATTACH on the inferior. + Using PTRACE_TRACEME is unchanged. 2 - admin-only attach: only processes with CAP_SYS_PTRACE may use ptrace - with PTRACE_ATTACH. + with PTRACE_ATTACH, or through children calling PTRACE_TRACEME. -3 - no attach: no processes may use ptrace with PTRACE_ATTACH. Once set, - this sysctl cannot be changed to a lower value. +3 - no attach: no processes may use ptrace with PTRACE_ATTACH nor via + PTRACE_TRACEME. Once set, this sysctl value cannot be changed. The original children-only logic was based on the restrictions in grsecurity. diff --git a/include/linux/security.h b/include/linux/security.h index 4e5a73cdbbef..3dea6a9d568f 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -1242,8 +1242,6 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts) * Check that the @parent process has sufficient permission to trace the * current process before allowing the current process to present itself * to the @parent process for tracing. - * The parent process will still have to undergo the ptrace_access_check - * checks before it is allowed to trace this one. * @parent contains the task_struct structure for debugger process. * Return 0 if permission is granted. * @capget: diff --git a/security/yama/yama_lsm.c b/security/yama/yama_lsm.c index 83554ee8a587..d51b7c76c37d 100644 --- a/security/yama/yama_lsm.c +++ b/security/yama/yama_lsm.c @@ -290,10 +290,51 @@ static int yama_ptrace_access_check(struct task_struct *child, return rc; } +/** + * yama_ptrace_traceme - validate PTRACE_TRACEME calls + * @parent: task that will become the ptracer of the current task + * + * Returns 0 if following the ptrace is allowed, -ve on error. + */ +static int yama_ptrace_traceme(struct task_struct *parent) +{ + int rc; + + /* If standard caps disallows it, so does Yama. We should + * only tighten restrictions further. + */ + rc = cap_ptrace_traceme(parent); + if (rc) + return rc; + + /* Only disallow PTRACE_TRACEME on more aggressive settings. */ + switch (ptrace_scope) { + case YAMA_SCOPE_CAPABILITY: + if (!ns_capable(task_user_ns(parent), CAP_SYS_PTRACE)) + rc = -EPERM; + break; + case YAMA_SCOPE_NO_ATTACH: + rc = -EPERM; + break; + } + + if (rc) { + char name[sizeof(current->comm)]; + printk_ratelimited(KERN_NOTICE + "ptraceme of pid %d was attempted by: %s (pid %d)\n", + current->pid, + get_task_comm(name, parent), + parent->pid); + } + + return rc; +} + static struct security_operations yama_ops = { .name = "yama", .ptrace_access_check = yama_ptrace_access_check, + .ptrace_traceme = yama_ptrace_traceme, .task_prctl = yama_task_prctl, .task_free = yama_task_free, }; -- cgit v1.2.3 From f026cfa82f628db24b8cea41b9d6202af104cecb Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Tue, 14 Aug 2012 09:53:38 -0700 Subject: Revert "x86-64/efi: Use EFI to deal with platform wall clock" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit bacef661acdb634170a8faddbc1cf28e8f8b9eee. This commit has been found to cause serious regressions on a number of ASUS machines at the least. We probably need to provide a 1:1 map in addition to the EFI virtual memory map in order for this to work. Signed-off-by: H. Peter Anvin Reported-and-bisected-by: Jérôme Carretero Cc: Jan Beulich Cc: Matt Fleming Cc: Matthew Garrett Cc: Linus Torvalds Cc: Andrew Morton Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120805172903.5f8bb24c@zougloub.eu --- arch/x86/mm/pageattr.c | 10 ++++------ arch/x86/platform/efi/efi.c | 30 ++++++++++++++++++++++++++---- include/linux/efi.h | 2 ++ init/main.c | 8 ++++---- 4 files changed, 36 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 931930a96160..a718e0d23503 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -919,13 +919,11 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, /* * On success we use clflush, when the CPU supports it to - * avoid the wbindv. If the CPU does not support it, in the - * error case, and during early boot (for EFI) we fall back - * to cpa_flush_all (which uses wbinvd): + * avoid the wbindv. If the CPU does not support it and in the + * error case we fall back to cpa_flush_all (which uses + * wbindv): */ - if (early_boot_irqs_disabled) - __cpa_flush_all((void *)(long)cache); - else if (!ret && cpu_has_clflush) { + if (!ret && cpu_has_clflush) { if (cpa.flags & (CPA_PAGES_ARRAY | CPA_ARRAY)) { cpa_flush_array(addr, numpages, cache, cpa.flags, pages); diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 2dc29f51e75a..92660edaa1e7 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -234,7 +234,22 @@ static efi_status_t __init phys_efi_set_virtual_address_map( return status; } -static int efi_set_rtc_mmss(unsigned long nowtime) +static efi_status_t __init phys_efi_get_time(efi_time_t *tm, + efi_time_cap_t *tc) +{ + unsigned long flags; + efi_status_t status; + + spin_lock_irqsave(&rtc_lock, flags); + efi_call_phys_prelog(); + status = efi_call_phys2(efi_phys.get_time, virt_to_phys(tm), + virt_to_phys(tc)); + efi_call_phys_epilog(); + spin_unlock_irqrestore(&rtc_lock, flags); + return status; +} + +int efi_set_rtc_mmss(unsigned long nowtime) { int real_seconds, real_minutes; efi_status_t status; @@ -263,7 +278,7 @@ static int efi_set_rtc_mmss(unsigned long nowtime) return 0; } -static unsigned long efi_get_time(void) +unsigned long efi_get_time(void) { efi_status_t status; efi_time_t eft; @@ -606,13 +621,18 @@ static int __init efi_runtime_init(void) } /* * We will only need *early* access to the following - * EFI runtime service before set_virtual_address_map + * two EFI runtime services before set_virtual_address_map * is invoked. */ + efi_phys.get_time = (efi_get_time_t *)runtime->get_time; efi_phys.set_virtual_address_map = (efi_set_virtual_address_map_t *) runtime->set_virtual_address_map; - + /* + * Make efi_get_time can be called before entering + * virtual mode. + */ + efi.get_time = phys_efi_get_time; early_iounmap(runtime, sizeof(efi_runtime_services_t)); return 0; @@ -700,10 +720,12 @@ void __init efi_init(void) efi_enabled = 0; return; } +#ifdef CONFIG_X86_32 if (efi_native) { x86_platform.get_wallclock = efi_get_time; x86_platform.set_wallclock = efi_set_rtc_mmss; } +#endif #if EFI_DEBUG print_efi_memmap(); diff --git a/include/linux/efi.h b/include/linux/efi.h index 103adc6d7e3a..ec45ccd8708a 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -503,6 +503,8 @@ extern u64 efi_mem_attribute (unsigned long phys_addr, unsigned long size); extern int __init efi_uart_console_only (void); extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource, struct resource *bss_resource); +extern unsigned long efi_get_time(void); +extern int efi_set_rtc_mmss(unsigned long nowtime); extern void efi_reserve_boot_services(void); extern struct efi_memory_map memmap; diff --git a/init/main.c b/init/main.c index e60679de61c3..b28673087ac0 100644 --- a/init/main.c +++ b/init/main.c @@ -461,10 +461,6 @@ static void __init mm_init(void) percpu_init_late(); pgtable_cache_init(); vmalloc_init(); -#ifdef CONFIG_X86 - if (efi_enabled) - efi_enter_virtual_mode(); -#endif } asmlinkage void __init start_kernel(void) @@ -606,6 +602,10 @@ asmlinkage void __init start_kernel(void) calibrate_delay(); pidmap_init(); anon_vma_init(); +#ifdef CONFIG_X86 + if (efi_enabled) + efi_enter_virtual_mode(); +#endif thread_info_cache_init(); cred_init(); fork_init(totalram_pages); -- cgit v1.2.3 From 47be03a28cc6c80e3aa2b3e8ed6d960ff0c5c0af Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:37 +0000 Subject: netpoll: use GFP_ATOMIC in slave_enable_netpoll() and __netpoll_setup() slave_enable_netpoll() and __netpoll_setup() may be called with read_lock() held, so should use GFP_ATOMIC to allocate memory. Eric suggested to pass gfp flags to __netpoll_setup(). Cc: Eric Dumazet Cc: "David S. Miller" Reported-by: Dan Carpenter Signed-off-by: Eric Dumazet Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 6 +++--- drivers/net/team/team.c | 16 +++++++++------- include/linux/netdevice.h | 3 ++- include/linux/netpoll.h | 2 +- net/8021q/vlan_dev.c | 7 ++++--- net/bridge/br_device.c | 12 ++++++------ net/bridge/br_if.c | 2 +- net/bridge/br_private.h | 4 ++-- net/core/netpoll.c | 8 ++++---- 9 files changed, 32 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 6fae5f3ec7f6..8697136e27c0 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1235,12 +1235,12 @@ static inline int slave_enable_netpoll(struct slave *slave) struct netpoll *np; int err = 0; - np = kzalloc(sizeof(*np), GFP_KERNEL); + np = kzalloc(sizeof(*np), GFP_ATOMIC); err = -ENOMEM; if (!np) goto out; - err = __netpoll_setup(np, slave->dev); + err = __netpoll_setup(np, slave->dev, GFP_ATOMIC); if (err) { kfree(np); goto out; @@ -1292,7 +1292,7 @@ static void bond_netpoll_cleanup(struct net_device *bond_dev) read_unlock(&bond->lock); } -static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni) +static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni, gfp_t gfp) { struct bonding *bond = netdev_priv(dev); struct slave *slave; diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 87707ab39430..341b65dbbcd3 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -795,16 +795,17 @@ static void team_port_leave(struct team *team, struct team_port *port) } #ifdef CONFIG_NET_POLL_CONTROLLER -static int team_port_enable_netpoll(struct team *team, struct team_port *port) +static int team_port_enable_netpoll(struct team *team, struct team_port *port, + gfp_t gfp) { struct netpoll *np; int err; - np = kzalloc(sizeof(*np), GFP_KERNEL); + np = kzalloc(sizeof(*np), gfp); if (!np) return -ENOMEM; - err = __netpoll_setup(np, port->dev); + err = __netpoll_setup(np, port->dev, gfp); if (err) { kfree(np); return err; @@ -833,7 +834,8 @@ static struct netpoll_info *team_netpoll_info(struct team *team) } #else -static int team_port_enable_netpoll(struct team *team, struct team_port *port) +static int team_port_enable_netpoll(struct team *team, struct team_port *port, + gfp_t gfp) { return 0; } @@ -913,7 +915,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev) } if (team_netpoll_info(team)) { - err = team_port_enable_netpoll(team, port); + err = team_port_enable_netpoll(team, port, GFP_KERNEL); if (err) { netdev_err(dev, "Failed to enable netpoll on device %s\n", portname); @@ -1443,7 +1445,7 @@ static void team_netpoll_cleanup(struct net_device *dev) } static int team_netpoll_setup(struct net_device *dev, - struct netpoll_info *npifo) + struct netpoll_info *npifo, gfp_t gfp) { struct team *team = netdev_priv(dev); struct team_port *port; @@ -1451,7 +1453,7 @@ static int team_netpoll_setup(struct net_device *dev, mutex_lock(&team->lock); list_for_each_entry(port, &team->port_list, list) { - err = team_port_enable_netpoll(team, port); + err = team_port_enable_netpoll(team, port, gfp); if (err) { __team_netpoll_cleanup(team); break; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a9db4f33407f..3560d688161e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -953,7 +953,8 @@ struct net_device_ops { #ifdef CONFIG_NET_POLL_CONTROLLER void (*ndo_poll_controller)(struct net_device *dev); int (*ndo_netpoll_setup)(struct net_device *dev, - struct netpoll_info *info); + struct netpoll_info *info, + gfp_t gfp); void (*ndo_netpoll_cleanup)(struct net_device *dev); #endif int (*ndo_set_vf_mac)(struct net_device *dev, diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 28f5389c924b..bf2d51eec0f3 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -43,7 +43,7 @@ struct netpoll_info { void netpoll_send_udp(struct netpoll *np, const char *msg, int len); void netpoll_print_options(struct netpoll *np); int netpoll_parse_options(struct netpoll *np, char *opt); -int __netpoll_setup(struct netpoll *np, struct net_device *ndev); +int __netpoll_setup(struct netpoll *np, struct net_device *ndev, gfp_t gfp); int netpoll_setup(struct netpoll *np); int netpoll_trap(void); void netpoll_set_trap(int trap); diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 73a2a83ee2da..ee4ae0944cef 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -669,19 +669,20 @@ static void vlan_dev_poll_controller(struct net_device *dev) return; } -static int vlan_dev_netpoll_setup(struct net_device *dev, struct netpoll_info *npinfo) +static int vlan_dev_netpoll_setup(struct net_device *dev, struct netpoll_info *npinfo, + gfp_t gfp) { struct vlan_dev_priv *info = vlan_dev_priv(dev); struct net_device *real_dev = info->real_dev; struct netpoll *netpoll; int err = 0; - netpoll = kzalloc(sizeof(*netpoll), GFP_KERNEL); + netpoll = kzalloc(sizeof(*netpoll), gfp); err = -ENOMEM; if (!netpoll) goto out; - err = __netpoll_setup(netpoll, real_dev); + err = __netpoll_setup(netpoll, real_dev, gfp); if (err) { kfree(netpoll); goto out; diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 333484537600..ed0e0f9dc788 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -213,7 +213,8 @@ static void br_netpoll_cleanup(struct net_device *dev) } } -static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni) +static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni, + gfp_t gfp) { struct net_bridge *br = netdev_priv(dev); struct net_bridge_port *p, *n; @@ -222,8 +223,7 @@ static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni) list_for_each_entry_safe(p, n, &br->port_list, list) { if (!p->dev) continue; - - err = br_netpoll_enable(p); + err = br_netpoll_enable(p, gfp); if (err) goto fail; } @@ -236,17 +236,17 @@ fail: goto out; } -int br_netpoll_enable(struct net_bridge_port *p) +int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp) { struct netpoll *np; int err = 0; - np = kzalloc(sizeof(*p->np), GFP_KERNEL); + np = kzalloc(sizeof(*p->np), gfp); err = -ENOMEM; if (!np) goto out; - err = __netpoll_setup(np, p->dev); + err = __netpoll_setup(np, p->dev, gfp); if (err) { kfree(np); goto out; diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index e1144e1617be..171fd6b9bfe6 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -361,7 +361,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev) if (err) goto err2; - if (br_netpoll_info(br) && ((err = br_netpoll_enable(p)))) + if (br_netpoll_info(br) && ((err = br_netpoll_enable(p, GFP_KERNEL)))) goto err3; err = netdev_set_master(dev, br->dev); diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h index a768b2408edf..f507d2af9646 100644 --- a/net/bridge/br_private.h +++ b/net/bridge/br_private.h @@ -316,7 +316,7 @@ static inline void br_netpoll_send_skb(const struct net_bridge_port *p, netpoll_send_skb(np, skb); } -extern int br_netpoll_enable(struct net_bridge_port *p); +extern int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp); extern void br_netpoll_disable(struct net_bridge_port *p); #else static inline struct netpoll_info *br_netpoll_info(struct net_bridge *br) @@ -329,7 +329,7 @@ static inline void br_netpoll_send_skb(const struct net_bridge_port *p, { } -static inline int br_netpoll_enable(struct net_bridge_port *p) +static inline int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp) { return 0; } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index b4c90e42b443..37cc854774a4 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -715,7 +715,7 @@ int netpoll_parse_options(struct netpoll *np, char *opt) } EXPORT_SYMBOL(netpoll_parse_options); -int __netpoll_setup(struct netpoll *np, struct net_device *ndev) +int __netpoll_setup(struct netpoll *np, struct net_device *ndev, gfp_t gfp) { struct netpoll_info *npinfo; const struct net_device_ops *ops; @@ -734,7 +734,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) } if (!ndev->npinfo) { - npinfo = kmalloc(sizeof(*npinfo), GFP_KERNEL); + npinfo = kmalloc(sizeof(*npinfo), gfp); if (!npinfo) { err = -ENOMEM; goto out; @@ -752,7 +752,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev) ops = np->dev->netdev_ops; if (ops->ndo_netpoll_setup) { - err = ops->ndo_netpoll_setup(ndev, npinfo); + err = ops->ndo_netpoll_setup(ndev, npinfo, gfp); if (err) goto free_npinfo; } @@ -857,7 +857,7 @@ int netpoll_setup(struct netpoll *np) refill_skbs(); rtnl_lock(); - err = __netpoll_setup(np, ndev); + err = __netpoll_setup(np, ndev, GFP_KERNEL); rtnl_unlock(); if (err) -- cgit v1.2.3 From 38e6bc185d9544dfad1774b3f8902a0b061aea25 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:38 +0000 Subject: netpoll: make __netpoll_cleanup non-block Like the previous patch, slave_disable_netpoll() and __netpoll_cleanup() may be called with read_lock() held too, so we should make them non-block, by moving the cleanup and kfree() to call_rcu_bh() callbacks. Cc: "David S. Miller" Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 4 +--- include/linux/netpoll.h | 3 +++ net/8021q/vlan_dev.c | 6 +----- net/bridge/br_device.c | 6 +----- net/core/netpoll.c | 42 +++++++++++++++++++++++++++++++---------- 5 files changed, 38 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 8697136e27c0..e42891683e3b 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1257,9 +1257,7 @@ static inline void slave_disable_netpoll(struct slave *slave) return; slave->np = NULL; - synchronize_rcu_bh(); - __netpoll_cleanup(np); - kfree(np); + __netpoll_free_rcu(np); } static inline bool slave_dev_support_netpoll(struct net_device *slave_dev) { diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index bf2d51eec0f3..907812efb4d9 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -23,6 +23,7 @@ struct netpoll { u8 remote_mac[ETH_ALEN]; struct list_head rx; /* rx_np list element */ + struct rcu_head rcu; }; struct netpoll_info { @@ -38,6 +39,7 @@ struct netpoll_info { struct delayed_work tx_work; struct netpoll *netpoll; + struct rcu_head rcu; }; void netpoll_send_udp(struct netpoll *np, const char *msg, int len); @@ -48,6 +50,7 @@ int netpoll_setup(struct netpoll *np); int netpoll_trap(void); void netpoll_set_trap(int trap); void __netpoll_cleanup(struct netpoll *np); +void __netpoll_free_rcu(struct netpoll *np); void netpoll_cleanup(struct netpoll *np); int __netpoll_rx(struct sk_buff *skb); void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index ee4ae0944cef..b65623f90660 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -704,11 +704,7 @@ static void vlan_dev_netpoll_cleanup(struct net_device *dev) info->netpoll = NULL; - /* Wait for transmitting packets to finish before freeing. */ - synchronize_rcu_bh(); - - __netpoll_cleanup(netpoll); - kfree(netpoll); + __netpoll_free_rcu(netpoll); } #endif /* CONFIG_NET_POLL_CONTROLLER */ diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index ed0e0f9dc788..f41ba4048c9a 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -267,11 +267,7 @@ void br_netpoll_disable(struct net_bridge_port *p) p->np = NULL; - /* Wait for transmitting packets to finish before freeing. */ - synchronize_rcu_bh(); - - __netpoll_cleanup(np); - kfree(np); + __netpoll_free_rcu(np); } #endif diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 37cc854774a4..dc17f1db1479 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -878,6 +878,24 @@ static int __init netpoll_init(void) } core_initcall(netpoll_init); +static void rcu_cleanup_netpoll_info(struct rcu_head *rcu_head) +{ + struct netpoll_info *npinfo = + container_of(rcu_head, struct netpoll_info, rcu); + + skb_queue_purge(&npinfo->arp_tx); + skb_queue_purge(&npinfo->txq); + + /* we can't call cancel_delayed_work_sync here, as we are in softirq */ + cancel_delayed_work(&npinfo->tx_work); + + /* clean after last, unfinished work */ + __skb_queue_purge(&npinfo->txq); + /* now cancel it again */ + cancel_delayed_work(&npinfo->tx_work); + kfree(npinfo); +} + void __netpoll_cleanup(struct netpoll *np) { struct netpoll_info *npinfo; @@ -903,20 +921,24 @@ void __netpoll_cleanup(struct netpoll *np) ops->ndo_netpoll_cleanup(np->dev); RCU_INIT_POINTER(np->dev->npinfo, NULL); + call_rcu_bh(&npinfo->rcu, rcu_cleanup_netpoll_info); + } +} +EXPORT_SYMBOL_GPL(__netpoll_cleanup); - /* avoid racing with NAPI reading npinfo */ - synchronize_rcu_bh(); +static void rcu_cleanup_netpoll(struct rcu_head *rcu_head) +{ + struct netpoll *np = container_of(rcu_head, struct netpoll, rcu); - skb_queue_purge(&npinfo->arp_tx); - skb_queue_purge(&npinfo->txq); - cancel_delayed_work_sync(&npinfo->tx_work); + __netpoll_cleanup(np); + kfree(np); +} - /* clean after last, unfinished work */ - __skb_queue_purge(&npinfo->txq); - kfree(npinfo); - } +void __netpoll_free_rcu(struct netpoll *np) +{ + call_rcu_bh(&np->rcu, rcu_cleanup_netpoll); } -EXPORT_SYMBOL_GPL(__netpoll_cleanup); +EXPORT_SYMBOL_GPL(__netpoll_free_rcu); void netpoll_cleanup(struct netpoll *np) { -- cgit v1.2.3 From 57c5d46191e75312934c00eba65b13a31ca95120 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:40 +0000 Subject: netpoll: take rcu_read_lock_bh() in netpoll_rx() In __netpoll_rx(), it dereferences ->npinfo without rcu_dereference_bh(), this patch fixes it by using the 'npinfo' passed from netpoll_rx() where it is already dereferenced with rcu_dereference_bh(). Cc: "David S. Miller" Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/linux/netpoll.h | 4 ++-- net/core/netpoll.c | 3 +-- 2 files changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 907812efb4d9..5d881c388273 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -52,7 +52,7 @@ void netpoll_set_trap(int trap); void __netpoll_cleanup(struct netpoll *np); void __netpoll_free_rcu(struct netpoll *np); void netpoll_cleanup(struct netpoll *np); -int __netpoll_rx(struct sk_buff *skb); +int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo); void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, struct net_device *dev); static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) @@ -77,7 +77,7 @@ static inline bool netpoll_rx(struct sk_buff *skb) spin_lock(&npinfo->rx_lock); /* check rx_flags again with the lock held */ - if (npinfo->rx_flags && __netpoll_rx(skb)) + if (npinfo->rx_flags && __netpoll_rx(skb, npinfo)) ret = true; spin_unlock(&npinfo->rx_lock); diff --git a/net/core/netpoll.c b/net/core/netpoll.c index dc17f1db1479..d055bb01328b 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -543,13 +543,12 @@ static void arp_reply(struct sk_buff *skb) spin_unlock_irqrestore(&npinfo->rx_lock, flags); } -int __netpoll_rx(struct sk_buff *skb) +int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo) { int proto, len, ulen; int hits = 0; const struct iphdr *iph; struct udphdr *uh; - struct netpoll_info *npinfo = skb->dev->npinfo; struct netpoll *np, *tmp; if (list_empty(&npinfo->rx_np)) -- cgit v1.2.3 From 91fe4a4b9e490a24f6702dd8afe72d8afab6fcdb Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:41 +0000 Subject: netpoll: use netpoll_rx_on() in netpoll_rx() The logic of the code is same, just call netpoll_rx_on(). Cc: "David S. Miller" Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/linux/netpoll.h | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 5d881c388273..2d178baa49df 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -63,6 +63,13 @@ static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) #ifdef CONFIG_NETPOLL +static inline int netpoll_rx_on(struct sk_buff *skb) +{ + struct netpoll_info *npinfo = rcu_dereference_bh(skb->dev->npinfo); + + return npinfo && (!list_empty(&npinfo->rx_np) || npinfo->rx_flags); +} + static inline bool netpoll_rx(struct sk_buff *skb) { struct netpoll_info *npinfo; @@ -70,11 +77,11 @@ static inline bool netpoll_rx(struct sk_buff *skb) bool ret = false; local_irq_save(flags); - npinfo = rcu_dereference_bh(skb->dev->npinfo); - if (!npinfo || (list_empty(&npinfo->rx_np) && !npinfo->rx_flags)) + if (!netpoll_rx_on(skb)) goto out; + npinfo = rcu_dereference_bh(skb->dev->npinfo); spin_lock(&npinfo->rx_lock); /* check rx_flags again with the lock held */ if (npinfo->rx_flags && __netpoll_rx(skb, npinfo)) @@ -86,13 +93,6 @@ out: return ret; } -static inline int netpoll_rx_on(struct sk_buff *skb) -{ - struct netpoll_info *npinfo = rcu_dereference_bh(skb->dev->npinfo); - - return npinfo && (!list_empty(&npinfo->rx_np) || npinfo->rx_flags); -} - static inline int netpoll_receive_skb(struct sk_buff *skb) { if (!list_empty(&skb->dev->napi_list)) -- cgit v1.2.3 From 2899656b494dcd118123af1126826b115c8ea6f9 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:42 +0000 Subject: netpoll: take rcu_read_lock_bh() in netpoll_send_skb_on_dev() This patch fixes several problems in the call path of netpoll_send_skb_on_dev(): 1. Disable IRQ's before calling netpoll_send_skb_on_dev(). 2. All the callees of netpoll_send_skb_on_dev() should use rcu_dereference_bh() to dereference ->npinfo. 3. Rename arp_reply() to netpoll_arp_reply(), the former is too generic. Cc: "David S. Miller" Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/linux/netpoll.h | 3 +++ net/core/netpoll.c | 31 +++++++++++++++++-------------- 2 files changed, 20 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 2d178baa49df..61aee86cf21d 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -57,7 +57,10 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, struct net_device *dev); static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) { + unsigned long flags; + local_irq_save(flags); netpoll_send_skb_on_dev(np, skb, np->dev); + local_irq_restore(flags); } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index d055bb01328b..174346ac15a0 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -54,7 +54,7 @@ static atomic_t trapped; MAX_UDP_CHUNK) static void zap_completion_queue(void); -static void arp_reply(struct sk_buff *skb); +static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo); static unsigned int carrier_timeout = 4; module_param(carrier_timeout, uint, 0644); @@ -170,7 +170,8 @@ static void poll_napi(struct net_device *dev) list_for_each_entry(napi, &dev->napi_list, dev_list) { if (napi->poll_owner != smp_processor_id() && spin_trylock(&napi->poll_lock)) { - budget = poll_one_napi(dev->npinfo, napi, budget); + budget = poll_one_napi(rcu_dereference_bh(dev->npinfo), + napi, budget); spin_unlock(&napi->poll_lock); if (!budget) @@ -185,13 +186,14 @@ static void service_arp_queue(struct netpoll_info *npi) struct sk_buff *skb; while ((skb = skb_dequeue(&npi->arp_tx))) - arp_reply(skb); + netpoll_arp_reply(skb, npi); } } static void netpoll_poll_dev(struct net_device *dev) { const struct net_device_ops *ops; + struct netpoll_info *ni = rcu_dereference_bh(dev->npinfo); if (!dev || !netif_running(dev)) return; @@ -206,17 +208,18 @@ static void netpoll_poll_dev(struct net_device *dev) poll_napi(dev); if (dev->flags & IFF_SLAVE) { - if (dev->npinfo) { + if (ni) { struct net_device *bond_dev = dev->master; struct sk_buff *skb; - while ((skb = skb_dequeue(&dev->npinfo->arp_tx))) { + struct netpoll_info *bond_ni = rcu_dereference_bh(bond_dev->npinfo); + while ((skb = skb_dequeue(&ni->arp_tx))) { skb->dev = bond_dev; - skb_queue_tail(&bond_dev->npinfo->arp_tx, skb); + skb_queue_tail(&bond_ni->arp_tx, skb); } } } - service_arp_queue(dev->npinfo); + service_arp_queue(ni); zap_completion_queue(); } @@ -302,6 +305,7 @@ static int netpoll_owner_active(struct net_device *dev) return 0; } +/* call with IRQ disabled */ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, struct net_device *dev) { @@ -309,8 +313,11 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, unsigned long tries; const struct net_device_ops *ops = dev->netdev_ops; /* It is up to the caller to keep npinfo alive. */ - struct netpoll_info *npinfo = np->dev->npinfo; + struct netpoll_info *npinfo; + + WARN_ON_ONCE(!irqs_disabled()); + npinfo = rcu_dereference_bh(np->dev->npinfo); if (!npinfo || !netif_running(dev) || !netif_device_present(dev)) { __kfree_skb(skb); return; @@ -319,11 +326,9 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, /* don't get messages out of order, and no recursion */ if (skb_queue_len(&npinfo->txq) == 0 && !netpoll_owner_active(dev)) { struct netdev_queue *txq; - unsigned long flags; txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb)); - local_irq_save(flags); /* try until next clock tick */ for (tries = jiffies_to_usecs(1)/USEC_PER_POLL; tries > 0; --tries) { @@ -347,10 +352,9 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb, } WARN_ONCE(!irqs_disabled(), - "netpoll_send_skb(): %s enabled interrupts in poll (%pF)\n", + "netpoll_send_skb_on_dev(): %s enabled interrupts in poll (%pF)\n", dev->name, ops->ndo_start_xmit); - local_irq_restore(flags); } if (status != NETDEV_TX_OK) { @@ -423,9 +427,8 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len) } EXPORT_SYMBOL(netpoll_send_udp); -static void arp_reply(struct sk_buff *skb) +static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo) { - struct netpoll_info *npinfo = skb->dev->npinfo; struct arphdr *arp; unsigned char *arp_ptr; int size, type = ARPOP_REPLY, ptype = ETH_P_ARP; -- cgit v1.2.3 From e15c3c2294605f09f9b336b2f3b97086ab4b8145 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:45 +0000 Subject: netpoll: check netpoll tx status on the right device Although this doesn't matter actually, because netpoll_tx_running() doesn't use the parameter, the code will be more readable. For team_dev_queue_xmit() we have to move it down to avoid compile errors. Cc: David Miller Signed-off-by: Jiri Pirko Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 +- include/linux/if_team.h | 30 +++++++++++++++--------------- net/bridge/br_forward.c | 2 +- 3 files changed, 17 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index e42891683e3b..d688a8af432c 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -398,7 +398,7 @@ int bond_dev_queue_xmit(struct bonding *bond, struct sk_buff *skb, sizeof(qdisc_skb_cb(skb)->slave_dev_queue_mapping)); skb->queue_mapping = qdisc_skb_cb(skb)->slave_dev_queue_mapping; - if (unlikely(netpoll_tx_running(slave_dev))) + if (unlikely(netpoll_tx_running(bond->dev))) bond_netpoll_send_skb(bond_get_slave_by_dev(bond, slave_dev), skb); else dev_queue_xmit(skb); diff --git a/include/linux/if_team.h b/include/linux/if_team.h index 6960fc1841a7..aa2e167e1ef4 100644 --- a/include/linux/if_team.h +++ b/include/linux/if_team.h @@ -96,21 +96,6 @@ static inline void team_netpoll_send_skb(struct team_port *port, } #endif -static inline int team_dev_queue_xmit(struct team *team, struct team_port *port, - struct sk_buff *skb) -{ - BUILD_BUG_ON(sizeof(skb->queue_mapping) != - sizeof(qdisc_skb_cb(skb)->slave_dev_queue_mapping)); - skb_set_queue_mapping(skb, qdisc_skb_cb(skb)->slave_dev_queue_mapping); - - skb->dev = port->dev; - if (unlikely(netpoll_tx_running(port->dev))) { - team_netpoll_send_skb(port, skb); - return 0; - } - return dev_queue_xmit(skb); -} - struct team_mode_ops { int (*init)(struct team *team); void (*exit)(struct team *team); @@ -200,6 +185,21 @@ struct team { long mode_priv[TEAM_MODE_PRIV_LONGS]; }; +static inline int team_dev_queue_xmit(struct team *team, struct team_port *port, + struct sk_buff *skb) +{ + BUILD_BUG_ON(sizeof(skb->queue_mapping) != + sizeof(qdisc_skb_cb(skb)->slave_dev_queue_mapping)); + skb_set_queue_mapping(skb, qdisc_skb_cb(skb)->slave_dev_queue_mapping); + + skb->dev = port->dev; + if (unlikely(netpoll_tx_running(team->dev))) { + team_netpoll_send_skb(port, skb); + return 0; + } + return dev_queue_xmit(skb); +} + static inline struct hlist_head *team_port_index_hash(struct team *team, int port_index) { diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index e9466d412707..02015a505d2a 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -65,7 +65,7 @@ static void __br_deliver(const struct net_bridge_port *to, struct sk_buff *skb) { skb->dev = to->dev; - if (unlikely(netpoll_tx_running(to->dev))) { + if (unlikely(netpoll_tx_running(to->br->dev))) { if (packet_length(skb) > skb->dev->mtu && !skb_is_gso(skb)) kfree_skb(skb); else { -- cgit v1.2.3 From 77ab8a54d9a8dcc4a46484a04133314f33f2aba6 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Fri, 10 Aug 2012 01:24:46 +0000 Subject: netpoll: convert several functions to bool These functions are just boolean, let them return bool instead of int. Cc: David Miller Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/linux/netpoll.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 61aee86cf21d..66d5379c305e 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -66,7 +66,7 @@ static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) #ifdef CONFIG_NETPOLL -static inline int netpoll_rx_on(struct sk_buff *skb) +static inline bool netpoll_rx_on(struct sk_buff *skb) { struct netpoll_info *npinfo = rcu_dereference_bh(skb->dev->npinfo); @@ -125,7 +125,7 @@ static inline void netpoll_poll_unlock(void *have) } } -static inline int netpoll_tx_running(struct net_device *dev) +static inline bool netpoll_tx_running(struct net_device *dev) { return irqs_disabled(); } @@ -133,11 +133,11 @@ static inline int netpoll_tx_running(struct net_device *dev) #else static inline bool netpoll_rx(struct sk_buff *skb) { - return 0; + return false; } -static inline int netpoll_rx_on(struct sk_buff *skb) +static inline bool netpoll_rx_on(struct sk_buff *skb) { - return 0; + return false; } static inline int netpoll_receive_skb(struct sk_buff *skb) { @@ -153,9 +153,9 @@ static inline void netpoll_poll_unlock(void *have) static inline void netpoll_netdev_init(struct net_device *dev) { } -static inline int netpoll_tx_running(struct net_device *dev) +static inline bool netpoll_tx_running(struct net_device *dev) { - return 0; + return false; } #endif -- cgit v1.2.3 From 8857df3aceb7a8eb7558059b7da109e41dd1fb95 Mon Sep 17 00:00:00 2001 From: Michael Hennerich Date: Fri, 20 Jul 2012 09:31:00 +0100 Subject: iio: frequency: ADF4350: Fix potential reference div factor overflow. With small channel spacing values and high reference frequencies it is possible to exceed the range of the 10-bit counter. Workaround by checking the range and widening some constrains. We don't use the REG1_PHASE value in this case the datasheet recommends to set it to 1 if not used. Signed-off-by: Michael Hennerich Signed-off-by: Jonathan Cameron --- drivers/iio/frequency/adf4350.c | 24 +++++++++++++++--------- include/linux/iio/frequency/adf4350.h | 2 ++ 2 files changed, 17 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/iio/frequency/adf4350.c b/drivers/iio/frequency/adf4350.c index 59fbb3ae40e7..e35bb8f6fe75 100644 --- a/drivers/iio/frequency/adf4350.c +++ b/drivers/iio/frequency/adf4350.c @@ -129,7 +129,7 @@ static int adf4350_set_freq(struct adf4350_state *st, unsigned long long freq) { struct adf4350_platform_data *pdata = st->pdata; u64 tmp; - u32 div_gcd, prescaler; + u32 div_gcd, prescaler, chspc; u16 mdiv, r_cnt = 0; u8 band_sel_div; @@ -158,14 +158,20 @@ static int adf4350_set_freq(struct adf4350_state *st, unsigned long long freq) if (pdata->ref_div_factor) r_cnt = pdata->ref_div_factor - 1; - do { - r_cnt = adf4350_tune_r_cnt(st, r_cnt); + chspc = st->chspc; - st->r1_mod = st->fpfd / st->chspc; - while (st->r1_mod > ADF4350_MAX_MODULUS) { - r_cnt = adf4350_tune_r_cnt(st, r_cnt); - st->r1_mod = st->fpfd / st->chspc; - } + do { + do { + do { + r_cnt = adf4350_tune_r_cnt(st, r_cnt); + st->r1_mod = st->fpfd / chspc; + if (r_cnt > ADF4350_MAX_R_CNT) { + /* try higher spacing values */ + chspc++; + r_cnt = 0; + } + } while ((st->r1_mod > ADF4350_MAX_MODULUS) && r_cnt); + } while (r_cnt == 0); tmp = freq * (u64)st->r1_mod + (st->fpfd > 1); do_div(tmp, st->fpfd); /* Div round closest (n + d/2)/d */ @@ -194,7 +200,7 @@ static int adf4350_set_freq(struct adf4350_state *st, unsigned long long freq) st->regs[ADF4350_REG0] = ADF4350_REG0_INT(st->r0_int) | ADF4350_REG0_FRACT(st->r0_fract); - st->regs[ADF4350_REG1] = ADF4350_REG1_PHASE(0) | + st->regs[ADF4350_REG1] = ADF4350_REG1_PHASE(1) | ADF4350_REG1_MOD(st->r1_mod) | prescaler; diff --git a/include/linux/iio/frequency/adf4350.h b/include/linux/iio/frequency/adf4350.h index b76b4a87065e..be91f344d5fc 100644 --- a/include/linux/iio/frequency/adf4350.h +++ b/include/linux/iio/frequency/adf4350.h @@ -87,6 +87,8 @@ #define ADF4350_MAX_BANDSEL_CLK 125000 /* Hz */ #define ADF4350_MAX_FREQ_REFIN 250000000 /* Hz */ #define ADF4350_MAX_MODULUS 4095 +#define ADF4350_MAX_R_CNT 1023 + /** * struct adf4350_platform_data - platform specific information -- cgit v1.2.3 From ac5aa7f9e0891a115ab307b4bdde9c55b9232970 Mon Sep 17 00:00:00 2001 From: Richard Genoud Date: Fri, 10 Aug 2012 16:52:58 +0200 Subject: pinctrl: header: trivial: declare struct device As struct device is used as a function argument, it should at least be declared (device.h is not included). Signed-off-by: Richard Genoud Signed-off-by: Linus Walleij --- include/linux/pinctrl/consumer.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pinctrl/consumer.h b/include/linux/pinctrl/consumer.h index 6dd96fb45482..e9b7f4350844 100644 --- a/include/linux/pinctrl/consumer.h +++ b/include/linux/pinctrl/consumer.h @@ -20,6 +20,7 @@ /* This struct is private to the core and should be regarded as a cookie */ struct pinctrl; struct pinctrl_state; +struct device; #ifdef CONFIG_PINCTRL -- cgit v1.2.3 From 3296193d1421c2d6f9e49e181cecfd917f0f5764 Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Tue, 14 Aug 2012 13:20:23 +0000 Subject: dt: introduce for_each_available_child_of_node, of_get_next_available_child Macro for_each_child_of_node() makes it easy to iterate over all of the children for a given device tree node, including those nodes that are marked as unavailable (i.e. status = "disabled"). Introduce for_each_available_child_of_node(), which is like for_each_child_of_node(), but it automatically skips unavailable nodes. This also requires the introduction of helper function of_get_next_available_child(), which returns the next available child node. Signed-off-by: Timur Tabi Signed-off-by: David S. Miller --- drivers/of/base.c | 27 +++++++++++++++++++++++++++ include/linux/of.h | 7 +++++++ 2 files changed, 34 insertions(+) (limited to 'include/linux') diff --git a/drivers/of/base.c b/drivers/of/base.c index c181b94abc36..d4a1c9a043e1 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -363,6 +363,33 @@ struct device_node *of_get_next_child(const struct device_node *node, } EXPORT_SYMBOL(of_get_next_child); +/** + * of_get_next_available_child - Find the next available child node + * @node: parent node + * @prev: previous child of the parent node, or NULL to get first + * + * This function is like of_get_next_child(), except that it + * automatically skips any disabled nodes (i.e. status = "disabled"). + */ +struct device_node *of_get_next_available_child(const struct device_node *node, + struct device_node *prev) +{ + struct device_node *next; + + read_lock(&devtree_lock); + next = prev ? prev->sibling : node->child; + for (; next; next = next->sibling) { + if (!of_device_is_available(next)) + continue; + if (of_node_get(next)) + break; + } + of_node_put(prev); + read_unlock(&devtree_lock); + return next; +} +EXPORT_SYMBOL(of_get_next_available_child); + /** * of_find_node_by_path - Find a node matching a full OF path * @path: The full path to match diff --git a/include/linux/of.h b/include/linux/of.h index 5919ee33f2b7..1b1163225f3b 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -190,10 +190,17 @@ extern struct device_node *of_get_parent(const struct device_node *node); extern struct device_node *of_get_next_parent(struct device_node *node); extern struct device_node *of_get_next_child(const struct device_node *node, struct device_node *prev); +extern struct device_node *of_get_next_available_child( + const struct device_node *node, struct device_node *prev); + #define for_each_child_of_node(parent, child) \ for (child = of_get_next_child(parent, NULL); child != NULL; \ child = of_get_next_child(parent, child)) +#define for_each_available_child_of_node(parent, child) \ + for (child = of_get_next_available_child(parent, NULL); child != NULL; \ + child = of_get_next_available_child(parent, child)) + static inline int of_get_child_count(const struct device_node *np) { struct device_node *child; -- cgit v1.2.3 From c0de08d04215031d68fa13af36f347a6cfa252ca Mon Sep 17 00:00:00 2001 From: Eric Leblond Date: Thu, 16 Aug 2012 22:02:58 +0000 Subject: af_packet: don't emit packet on orig fanout group If a packet is emitted on one socket in one group of fanout sockets, it is transmitted again. It is thus read again on one of the sockets of the fanout group. This result in a loop for software which generate packets when receiving one. This retransmission is not the intended behavior: a fanout group must behave like a single socket. The packet should not be transmitted on a socket if it originates from a socket belonging to the same fanout group. This patch fixes the issue by changing the transmission check to take fanout group info account. Reported-by: Aleksandr Kotov Signed-off-by: Eric Leblond Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ net/core/dev.c | 16 ++++++++++++++-- net/packet/af_packet.c | 9 +++++++++ 3 files changed, 25 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3560d688161e..59dc05f38247 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1522,6 +1522,8 @@ struct packet_type { struct sk_buff **(*gro_receive)(struct sk_buff **head, struct sk_buff *skb); int (*gro_complete)(struct sk_buff *skb); + bool (*id_match)(struct packet_type *ptype, + struct sock *sk); void *af_packet_priv; struct list_head list; }; diff --git a/net/core/dev.c b/net/core/dev.c index a39354ee1432..debd9372472f 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1642,6 +1642,19 @@ static inline int deliver_skb(struct sk_buff *skb, return pt_prev->func(skb, skb->dev, pt_prev, orig_dev); } +static inline bool skb_loop_sk(struct packet_type *ptype, struct sk_buff *skb) +{ + if (ptype->af_packet_priv == NULL) + return false; + + if (ptype->id_match) + return ptype->id_match(ptype, skb->sk); + else if ((struct sock *)ptype->af_packet_priv == skb->sk) + return true; + + return false; +} + /* * Support routine. Sends outgoing frames to any network * taps currently in use. @@ -1659,8 +1672,7 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) * they originated from - MvS (miquels@drinkel.ow.org) */ if ((ptype->dev == dev || !ptype->dev) && - (ptype->af_packet_priv == NULL || - (struct sock *)ptype->af_packet_priv != skb->sk)) { + (!skb_loop_sk(ptype, skb))) { if (pt_prev) { deliver_skb(skb2, pt_prev, skb->dev); pt_prev = ptype; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 8ac890a1a4c0..aee7196aac36 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1273,6 +1273,14 @@ static void __fanout_unlink(struct sock *sk, struct packet_sock *po) spin_unlock(&f->lock); } +bool match_fanout_group(struct packet_type *ptype, struct sock * sk) +{ + if (ptype->af_packet_priv == (void*)((struct packet_sock *)sk)->fanout) + return true; + + return false; +} + static int fanout_add(struct sock *sk, u16 id, u16 type_flags) { struct packet_sock *po = pkt_sk(sk); @@ -1325,6 +1333,7 @@ static int fanout_add(struct sock *sk, u16 id, u16 type_flags) match->prot_hook.dev = po->prot_hook.dev; match->prot_hook.func = packet_rcv_fanout; match->prot_hook.af_packet_priv = match; + match->prot_hook.id_match = match_fanout_group; dev_add_pack(&match->prot_hook); list_add(&match->list, &fanout_list); } -- cgit v1.2.3 From c3a5ce0416b6c172a23bc8a3760d8704d3d1535b Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Tue, 21 Aug 2012 16:16:00 -0700 Subject: string: do not export memweight() to userspace Fix the following warning: usr/include/linux/string.h:8: userspace cannot reference function or variable defined in the kernel Signed-off-by: WANG Cong Acked-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/string.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/string.h b/include/linux/string.h index ffe0442e18d2..b9178812d9df 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -144,8 +144,8 @@ static inline bool strstarts(const char *str, const char *prefix) { return strncmp(str, prefix, strlen(prefix)) == 0; } -#endif extern size_t memweight(const void *ptr, size_t bytes); +#endif /* __KERNEL__ */ #endif /* _LINUX_STRING_H_ */ -- cgit v1.2.3 From c67fe3752abe6ab47639e2f9b836900c3dc3da84 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Aug 2012 16:16:17 -0700 Subject: mm: compaction: Abort async compaction if locks are contended or taking too long Jim Schutt reported a problem that pointed at compaction contending heavily on locks. The workload is straight-forward and in his own words; The systems in question have 24 SAS drives spread across 3 HBAs, running 24 Ceph OSD instances, one per drive. FWIW these servers are dual-socket Intel 5675 Xeons w/48 GB memory. I've got ~160 Ceph Linux clients doing dd simultaneously to a Ceph file system backed by 12 of these servers. Early in the test everything looks fine procs -------------------memory------------------ ---swap-- -----io---- --system-- -----cpu------- r b swpd free buff cache si so bi bo in cs us sy id wa st 31 15 0 287216 576 38606628 0 0 2 1158 2 14 1 3 95 0 0 27 15 0 225288 576 38583384 0 0 18 2222016 203357 134876 11 56 17 15 0 28 17 0 219256 576 38544736 0 0 11 2305932 203141 146296 11 49 23 17 0 6 18 0 215596 576 38552872 0 0 7 2363207 215264 166502 12 45 22 20 0 22 18 0 226984 576 38596404 0 0 3 2445741 223114 179527 12 43 23 22 0 and then it goes to pot procs -------------------memory------------------ ---swap-- -----io---- --system-- -----cpu------- r b swpd free buff cache si so bi bo in cs us sy id wa st 163 8 0 464308 576 36791368 0 0 11 22210 866 536 3 13 79 4 0 207 14 0 917752 576 36181928 0 0 712 1345376 134598 47367 7 90 1 2 0 123 12 0 685516 576 36296148 0 0 429 1386615 158494 60077 8 84 5 3 0 123 12 0 598572 576 36333728 0 0 1107 1233281 147542 62351 7 84 5 4 0 622 7 0 660768 576 36118264 0 0 557 1345548 151394 59353 7 85 4 3 0 223 11 0 283960 576 36463868 0 0 46 1107160 121846 33006 6 93 1 1 0 Note that system CPU usage is very high blocks being written out has dropped by 42%. He analysed this with perf and found perf record -g -a sleep 10 perf report --sort symbol --call-graph fractal,5 34.63% [k] _raw_spin_lock_irqsave | |--97.30%-- isolate_freepages | compaction_alloc | unmap_and_move | migrate_pages | compact_zone | compact_zone_order | try_to_compact_pages | __alloc_pages_direct_compact | __alloc_pages_slowpath | __alloc_pages_nodemask | alloc_pages_vma | do_huge_pmd_anonymous_page | handle_mm_fault | do_page_fault | page_fault | | | |--87.39%-- skb_copy_datagram_iovec | | tcp_recvmsg | | inet_recvmsg | | sock_recvmsg | | sys_recvfrom | | system_call | | __recv | | | | | --100.00%-- (nil) | | | --12.61%-- memcpy --2.70%-- [...] There was other data but primarily it is all showing that compaction is contended heavily on the zone->lock and zone->lru_lock. commit [b2eef8c0: mm: compaction: minimise the time IRQs are disabled while isolating pages for migration] noted that it was possible for migration to hold the lru_lock for an excessive amount of time. Very broadly speaking this patch expands the concept. This patch introduces compact_checklock_irqsave() to check if a lock is contended or the process needs to be scheduled. If either condition is true then async compaction is aborted and the caller is informed. The page allocator will fail a THP allocation if compaction failed due to contention. This patch also introduces compact_trylock_irqsave() which will acquire the lock only if it is not contended and the process does not need to schedule. Reported-by: Jim Schutt Tested-by: Jim Schutt Signed-off-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compaction.h | 4 +- mm/compaction.c | 100 +++++++++++++++++++++++++++++++++++---------- mm/internal.h | 1 + mm/page_alloc.c | 17 +++++--- 4 files changed, 93 insertions(+), 29 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compaction.h b/include/linux/compaction.h index 133ddcf83397..ef658147e4e8 100644 --- a/include/linux/compaction.h +++ b/include/linux/compaction.h @@ -22,7 +22,7 @@ extern int sysctl_extfrag_handler(struct ctl_table *table, int write, extern int fragmentation_index(struct zone *zone, unsigned int order); extern unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *mask, - bool sync); + bool sync, bool *contended); extern int compact_pgdat(pg_data_t *pgdat, int order); extern unsigned long compaction_suitable(struct zone *zone, int order); @@ -64,7 +64,7 @@ static inline bool compaction_deferred(struct zone *zone, int order) #else static inline unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *nodemask, - bool sync) + bool sync, bool *contended) { return COMPACT_CONTINUE; } diff --git a/mm/compaction.c b/mm/compaction.c index bcce7897e17a..7fcd3a52e68d 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -50,6 +50,47 @@ static inline bool migrate_async_suitable(int migratetype) return is_migrate_cma(migratetype) || migratetype == MIGRATE_MOVABLE; } +/* + * Compaction requires the taking of some coarse locks that are potentially + * very heavily contended. Check if the process needs to be scheduled or + * if the lock is contended. For async compaction, back out in the event + * if contention is severe. For sync compaction, schedule. + * + * Returns true if the lock is held. + * Returns false if the lock is released and compaction should abort + */ +static bool compact_checklock_irqsave(spinlock_t *lock, unsigned long *flags, + bool locked, struct compact_control *cc) +{ + if (need_resched() || spin_is_contended(lock)) { + if (locked) { + spin_unlock_irqrestore(lock, *flags); + locked = false; + } + + /* async aborts if taking too long or contended */ + if (!cc->sync) { + if (cc->contended) + *cc->contended = true; + return false; + } + + cond_resched(); + if (fatal_signal_pending(current)) + return false; + } + + if (!locked) + spin_lock_irqsave(lock, *flags); + return true; +} + +static inline bool compact_trylock_irqsave(spinlock_t *lock, + unsigned long *flags, struct compact_control *cc) +{ + return compact_checklock_irqsave(lock, flags, false, cc); +} + /* * Isolate free pages onto a private freelist. Caller must hold zone->lock. * If @strict is true, will abort returning 0 on any invalid PFNs or non-free @@ -173,7 +214,7 @@ isolate_freepages_range(unsigned long start_pfn, unsigned long end_pfn) } /* Update the number of anon and file isolated pages in the zone */ -static void acct_isolated(struct zone *zone, struct compact_control *cc) +static void acct_isolated(struct zone *zone, bool locked, struct compact_control *cc) { struct page *page; unsigned int count[2] = { 0, }; @@ -181,8 +222,14 @@ static void acct_isolated(struct zone *zone, struct compact_control *cc) list_for_each_entry(page, &cc->migratepages, lru) count[!!page_is_file_cache(page)]++; - __mod_zone_page_state(zone, NR_ISOLATED_ANON, count[0]); - __mod_zone_page_state(zone, NR_ISOLATED_FILE, count[1]); + /* If locked we can use the interrupt unsafe versions */ + if (locked) { + __mod_zone_page_state(zone, NR_ISOLATED_ANON, count[0]); + __mod_zone_page_state(zone, NR_ISOLATED_FILE, count[1]); + } else { + mod_zone_page_state(zone, NR_ISOLATED_ANON, count[0]); + mod_zone_page_state(zone, NR_ISOLATED_FILE, count[1]); + } } /* Similar to reclaim, but different enough that they don't share logic */ @@ -228,6 +275,8 @@ isolate_migratepages_range(struct zone *zone, struct compact_control *cc, struct list_head *migratelist = &cc->migratepages; isolate_mode_t mode = 0; struct lruvec *lruvec; + unsigned long flags; + bool locked; /* * Ensure that there are not too many pages isolated from the LRU @@ -247,25 +296,22 @@ isolate_migratepages_range(struct zone *zone, struct compact_control *cc, /* Time to isolate some pages for migration */ cond_resched(); - spin_lock_irq(&zone->lru_lock); + spin_lock_irqsave(&zone->lru_lock, flags); + locked = true; for (; low_pfn < end_pfn; low_pfn++) { struct page *page; - bool locked = true; /* give a chance to irqs before checking need_resched() */ if (!((low_pfn+1) % SWAP_CLUSTER_MAX)) { - spin_unlock_irq(&zone->lru_lock); + spin_unlock_irqrestore(&zone->lru_lock, flags); locked = false; } - if (need_resched() || spin_is_contended(&zone->lru_lock)) { - if (locked) - spin_unlock_irq(&zone->lru_lock); - cond_resched(); - spin_lock_irq(&zone->lru_lock); - if (fatal_signal_pending(current)) - break; - } else if (!locked) - spin_lock_irq(&zone->lru_lock); + + /* Check if it is ok to still hold the lock */ + locked = compact_checklock_irqsave(&zone->lru_lock, &flags, + locked, cc); + if (!locked) + break; /* * migrate_pfn does not necessarily start aligned to a @@ -349,9 +395,10 @@ isolate_migratepages_range(struct zone *zone, struct compact_control *cc, } } - acct_isolated(zone, cc); + acct_isolated(zone, locked, cc); - spin_unlock_irq(&zone->lru_lock); + if (locked) + spin_unlock_irqrestore(&zone->lru_lock, flags); trace_mm_compaction_isolate_migratepages(nr_scanned, nr_isolated); @@ -461,7 +508,16 @@ static void isolate_freepages(struct zone *zone, * are disabled */ isolated = 0; - spin_lock_irqsave(&zone->lock, flags); + + /* + * The zone lock must be held to isolate freepages. This + * unfortunately this is a very coarse lock and can be + * heavily contended if there are parallel allocations + * or parallel compactions. For async compaction do not + * spin on the lock + */ + if (!compact_trylock_irqsave(&zone->lock, &flags, cc)) + break; if (suitable_migration_target(page)) { end_pfn = min(pfn + pageblock_nr_pages, zone_end_pfn); isolated = isolate_freepages_block(pfn, end_pfn, @@ -773,7 +829,7 @@ out: static unsigned long compact_zone_order(struct zone *zone, int order, gfp_t gfp_mask, - bool sync) + bool sync, bool *contended) { struct compact_control cc = { .nr_freepages = 0, @@ -782,6 +838,7 @@ static unsigned long compact_zone_order(struct zone *zone, .migratetype = allocflags_to_migratetype(gfp_mask), .zone = zone, .sync = sync, + .contended = contended, }; INIT_LIST_HEAD(&cc.freepages); INIT_LIST_HEAD(&cc.migratepages); @@ -803,7 +860,7 @@ int sysctl_extfrag_threshold = 500; */ unsigned long try_to_compact_pages(struct zonelist *zonelist, int order, gfp_t gfp_mask, nodemask_t *nodemask, - bool sync) + bool sync, bool *contended) { enum zone_type high_zoneidx = gfp_zone(gfp_mask); int may_enter_fs = gfp_mask & __GFP_FS; @@ -827,7 +884,8 @@ unsigned long try_to_compact_pages(struct zonelist *zonelist, nodemask) { int status; - status = compact_zone_order(zone, order, gfp_mask, sync); + status = compact_zone_order(zone, order, gfp_mask, sync, + contended); rc = max(status, rc); /* If a normal allocation would succeed, stop compacting */ diff --git a/mm/internal.h b/mm/internal.h index 3314f79d775a..b8c91b342e24 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -130,6 +130,7 @@ struct compact_control { int order; /* order a direct compactor needs */ int migratetype; /* MOVABLE, RECLAIMABLE etc */ struct zone *zone; + bool *contended; /* True if a lock was contended */ }; unsigned long diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 07f19248acb5..c66fb875104a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2102,7 +2102,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, int migratetype, bool sync_migration, - bool *deferred_compaction, + bool *contended_compaction, bool *deferred_compaction, unsigned long *did_some_progress) { struct page *page; @@ -2117,7 +2117,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, current->flags |= PF_MEMALLOC; *did_some_progress = try_to_compact_pages(zonelist, order, gfp_mask, - nodemask, sync_migration); + nodemask, sync_migration, + contended_compaction); current->flags &= ~PF_MEMALLOC; if (*did_some_progress != COMPACT_SKIPPED) { @@ -2163,7 +2164,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, enum zone_type high_zoneidx, nodemask_t *nodemask, int alloc_flags, struct zone *preferred_zone, int migratetype, bool sync_migration, - bool *deferred_compaction, + bool *contended_compaction, bool *deferred_compaction, unsigned long *did_some_progress) { return NULL; @@ -2336,6 +2337,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, unsigned long did_some_progress; bool sync_migration = false; bool deferred_compaction = false; + bool contended_compaction = false; /* * In the slowpath, we sanity check order to avoid ever trying to @@ -2425,6 +2427,7 @@ rebalance: nodemask, alloc_flags, preferred_zone, migratetype, sync_migration, + &contended_compaction, &deferred_compaction, &did_some_progress); if (page) @@ -2434,10 +2437,11 @@ rebalance: /* * If compaction is deferred for high-order allocations, it is because * sync compaction recently failed. In this is the case and the caller - * has requested the system not be heavily disrupted, fail the - * allocation now instead of entering direct reclaim + * requested a movable allocation that does not heavily disrupt the + * system then fail the allocation instead of entering direct reclaim. */ - if (deferred_compaction && (gfp_mask & __GFP_NO_KSWAPD)) + if ((deferred_compaction || contended_compaction) && + (gfp_mask & __GFP_NO_KSWAPD)) goto nopage; /* Try direct reclaim and then allocating */ @@ -2508,6 +2512,7 @@ rebalance: nodemask, alloc_flags, preferred_zone, migratetype, sync_migration, + &contended_compaction, &deferred_compaction, &did_some_progress); if (page) -- cgit v1.2.3