From 778b032d96909690c19d84f8d17c13be65ed6f8e Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:30:08 +0200 Subject: uprobes: Kill uprobes_srcu/uprobe_srcu_id Kill the no longer needed uprobes_srcu/uprobe_srcu_id code. It doesn't really work anyway. synchronize_srcu() can only synchronize with the code "inside" the srcu_read_lock/srcu_read_unlock section, while uprobe_pre_sstep_notifier() does srcu_read_lock() _after_ we already hit the breakpoint. I guess this probably works "in practice". synchronize_srcu() is slow and it implies synchronize_sched(), and the probed task enters the non- preemptible section at the start of exception handler. Still this is not right at least in theory, and task->uprobe_srcu_id blows task_struct. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529193008.GG8057@redhat.com Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 6029d8c54476..6bd19655c1a7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1569,7 +1569,6 @@ struct task_struct { #endif #ifdef CONFIG_UPROBES struct uprobe_task *utask; - int uprobe_srcu_id; #endif }; -- cgit v1.2.3 From c48b60538c3ba05a7a2713c4791b25405525431b Mon Sep 17 00:00:00 2001 From: Vince Weaver Date: Thu, 1 Mar 2012 17:28:14 -0500 Subject: perf/x86: Use rdpmc() rather than rdmsr() when possible in the kernel The rdpmc instruction is faster than the equivelant rdmsr call, so use it when possible in the kernel. The perfctr kernel patches did this, after extensive testing showed rdpmc to always be faster (One can look in etc/costs in the perfctr-2.6 package to see a historical list of the overhead). I have done some tests on a 3.2 kernel, the kernel module I used was included in the first posting of this patch: rdmsr rdpmc Core2 T9900: 203.9 cycles 30.9 cycles AMD fam0fh: 56.2 cycles 9.8 cycles Atom 6/28/2: 129.7 cycles 50.6 cycles The speedup of using rdpmc is large. [ It's probably possible (and desirable) to do this without requiring a new field in the hw_perf_event structure, but the fixed events make this tricky. ] Signed-off-by: Vince Weaver Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/alpine.DEB.2.00.1203011724030.26934@cl320.eecs.utk.edu Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 45db49f64bb4..1ce887abcc5c 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -677,6 +677,7 @@ struct hw_perf_event { u64 last_tag; unsigned long config_base; unsigned long event_base; + int event_base_rdpmc; int idx; int last_cpu; -- cgit v1.2.3 From 5da43bed800770906fca24deef7ae3d456823b86 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 31 May 2012 21:28:58 -0400 Subject: tracing: Add comments for the other bits of ftrace_event_call.flags TRACE_EVENT_FL_ENABLED_BIT, TRACE_EVENT_FL_FILTERED_BIT, TRACE_EVENT_FL_RECORDED_CMD_BIT, Have comments about what they are, but: TRACE_EVENT_FL_CAP_ANY_BIT, TRACE_EVENT_FL_NO_SET_FILTER_BIT, TRACE_EVENT_FL_IGNORE_ENABLE_BIT, do not, making them second class citizens. To prevent another class warfare, these bits have protested for their right to be commented. And By Golly! I'll give them what they want! Signed-off-by: Steven Rostedt --- include/linux/ftrace_event.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 176a939d1547..1aff18346c71 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -207,6 +207,9 @@ struct ftrace_event_call { * bit 1: enabled * bit 2: filter_active * bit 3: enabled cmd record + * bit 4: allow trace by non root (cap any) + * bit 5: failed to apply filter + * bit 6: ftrace internal event (do not enable) * * Changes to flags must hold the event_mutex. * -- cgit v1.2.3 From 0cda4c023132aa93f2dd94811061f812e88daf4c Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 15 Jun 2012 14:31:33 +0800 Subject: perf: Introduce perf_pmu_migrate_context() Originally from Peter Zijlstra. The helper migrates perf events from one cpu to another cpu. Signed-off-by: Zheng Yan Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1339741902-8449-5-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 1ce887abcc5c..76c5c8b724a7 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1107,6 +1107,8 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, struct task_struct *task, perf_overflow_handler_t callback, void *context); +extern void perf_pmu_migrate_context(struct pmu *pmu, + int src_cpu, int dst_cpu); extern u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running); -- cgit v1.2.3 From 7c94ee2e0917b2ea56498bff939c8aa55da27207 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 15 Jun 2012 14:31:37 +0800 Subject: perf/x86: Add Intel Nehalem and Sandy Bridge-EP uncore support The uncore subsystem in Sandy Bridge-EP consists of 8 components: Ubox, Cacheing Agent, Home Agent, Memory controller, Power Control, QPI Link Layer, R2PCIe, R3QPI. Signed-off-by: Zheng Yan Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1339741902-8449-9-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar --- include/linux/pci_ids.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index ab741b0d0074..5f187026b812 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2755,6 +2755,17 @@ #define PCI_DEVICE_ID_INTEL_IOAT_SNB7 0x3c27 #define PCI_DEVICE_ID_INTEL_IOAT_SNB8 0x3c2e #define PCI_DEVICE_ID_INTEL_IOAT_SNB9 0x3c2f +#define PCI_DEVICE_ID_INTEL_UNC_HA 0x3c46 +#define PCI_DEVICE_ID_INTEL_UNC_IMC0 0x3cb0 +#define PCI_DEVICE_ID_INTEL_UNC_IMC1 0x3cb1 +#define PCI_DEVICE_ID_INTEL_UNC_IMC2 0x3cb4 +#define PCI_DEVICE_ID_INTEL_UNC_IMC3 0x3cb5 +#define PCI_DEVICE_ID_INTEL_UNC_QPI0 0x3c41 +#define PCI_DEVICE_ID_INTEL_UNC_QPI1 0x3c42 +#define PCI_DEVICE_ID_INTEL_UNC_R2PCIE 0x3c43 +#define PCI_DEVICE_ID_INTEL_UNC_R3QPI0 0x3c44 +#define PCI_DEVICE_ID_INTEL_UNC_R3QPI1 0x3c45 +#define PCI_DEVICE_ID_INTEL_JAKETOWN_UBOX 0x3ce0 #define PCI_DEVICE_ID_INTEL_IOAT_SNB 0x402f #define PCI_DEVICE_ID_INTEL_5100_16 0x65f0 #define PCI_DEVICE_ID_INTEL_5100_21 0x65f5 -- cgit v1.2.3 From 6d158a813efcd09661c23f16ddf7e2ff834cb20c Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 27 Jun 2012 20:46:14 -0400 Subject: tracing: Remove NR_CPUS array from trace_iterator Replace the NR_CPUS array of buffer_iter from the trace_iterator with an allocated array. This will just create an array of possible CPUS instead of the max number specified. The use of NR_CPUS in that array caused allocation failures for machines that were tight on memory. This did not cause any failures to the system itself (no crashes), but caused unnecessary failures for reading the trace files. Added a helper function called 'trace_buffer_iter()' that returns the buffer_iter item or NULL if it is not defined or the array was not allocated. Some routines do not require the array (tracing_open_pipe() for one). Reported-by: Dave Jones Signed-off-by: Steven Rostedt --- include/linux/ftrace_event.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 1aff18346c71..af961d6f7ab1 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -65,7 +65,7 @@ struct trace_iterator { void *private; int cpu_file; struct mutex mutex; - struct ring_buffer_iter *buffer_iter[NR_CPUS]; + struct ring_buffer_iter **buffer_iter; unsigned long iter_flags; /* trace_seq for __print_flags() and __print_symbolic() etc. */ -- cgit v1.2.3 From 326f418b4c74ab4a6066f38b6144bc6461f9447b Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Thu, 28 Jun 2012 15:04:57 -0400 Subject: tracepoint: Use static_key_false(), since static_branch() is deprecated Convert the last user of static_branch() -> static_key_false(). Signed-off-by: Jason Baron Cc: rostedt@goodmis.org Cc: mathieu.desnoyers@efficios.com Cc: a.p.zijlstra@chello.nl Link: http://lkml.kernel.org/r/5fffcd40a6c063769badcdd74a7d90980500dbcb.1340909155.git.jbaron@redhat.com Signed-off-by: Ingo Molnar --- include/linux/tracepoint.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index bd96ecd0e05c..802de56c41e8 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -153,7 +153,7 @@ static inline void tracepoint_synchronize_unregister(void) } \ static inline void trace_##name##_rcuidle(proto) \ { \ - if (static_branch(&__tracepoint_##name.key)) \ + if (static_key_false(&__tracepoint_##name.key)) \ __DO_TRACE(&__tracepoint_##name, \ TP_PROTO(data_proto), \ TP_ARGS(data_args), \ -- cgit v1.2.3 From 47fbc518a4b5c9a949f7cab8b14a00d3549bf138 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Thu, 28 Jun 2012 15:05:02 -0400 Subject: jump label: Remove static_branch() Remove the obsolete static_branch() interface, since the supported interface is now static_key_false()/true() - which is used by all in-tree code. See commit: c5905afb0e ("static keys: Introduce 'struct static_key', static_key_true()/false() and static_key_slow_[inc|dec]()"). Signed-off-by: Jason Baron Cc: rostedt@goodmis.org Cc: mathieu.desnoyers@efficios.com Cc: Linus Torvalds Cc: Andrew Morton Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/199332c47eef8005d5a5bf1018a80d25929a5746.1340909155.git.jbaron@redhat.com Signed-off-by: Ingo Molnar --- include/linux/jump_label.h | 17 +---------------- 1 file changed, 1 insertion(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index c513a40510f5..0976fc46d1e0 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -42,8 +42,7 @@ * allowed. * * Not initializing the key (static data is initialized to 0s anyway) is the - * same as using STATIC_KEY_INIT_FALSE and static_key_false() is - * equivalent with static_branch(). + * same as using STATIC_KEY_INIT_FALSE. * */ @@ -107,12 +106,6 @@ static __always_inline bool static_key_true(struct static_key *key) return !static_key_false(key); } -/* Deprecated. Please use 'static_key_false() instead. */ -static __always_inline bool static_branch(struct static_key *key) -{ - return arch_static_branch(key); -} - extern struct jump_entry __start___jump_table[]; extern struct jump_entry __stop___jump_table[]; @@ -166,14 +159,6 @@ static __always_inline bool static_key_true(struct static_key *key) return false; } -/* Deprecated. Please use 'static_key_false() instead. */ -static __always_inline bool static_branch(struct static_key *key) -{ - if (unlikely(atomic_read(&key->enabled)) > 0) - return true; - return false; -} - static inline void static_key_slow_inc(struct static_key *key) { atomic_inc(&key->enabled); -- cgit v1.2.3